Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-1@lion>; Thu, 24 Aug 1995 07:01:02 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 23 Aug 95 16:38:41 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13874; 23 Aug 95 17:38 +0100 Received: by gw.home.vix.com id AA08605; Tue, 22 Aug 95 23:51:29 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA08597; Tue, 22 Aug 95 23:51:22 -0700 Received: by mail.Germany.EU.net with SMTP (5.51:31/EUnetD-2.5.2.a) via EUnet id IAA28032; Wed, 23 Aug 1995 08:53:11 +0200 Received: from MOIRA (EROS-MOIRA) by pr-wiesbaden.de (4.1/SMI-4.1) id AA13625; Wed, 23 Aug 95 08:59:50 +0200 Received: by MOIRA (4.1/SMI-4.1) id AA13412; Wed, 23 Aug 95 08:56:22 +0200 Date: Wed, 23 Aug 95 08:56:22 +0200 Message-Id: <9508230656.AA13412@MOIRA> From: Horst Duchene - Patzschke + Rasp GmbH To: perldb-interest@vix.com Subject: help help ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-40@lion>; Thu, 24 Aug 1995 07:05:45 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 23 Aug 95 22:27:29 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa28952; 23 Aug 95 23:27 +0100 Received: by gw.home.vix.com id AA18641; Wed, 23 Aug 95 09:55:19 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA18636; Wed, 23 Aug 95 09:55:16 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa29665; 23 Aug 95 17:22 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa07361; 23 Aug 95 17:19 +0100 Received: from ig.co.uk by lion id <13561-0@lion>; Wed, 23 Aug 1995 17:21:53 +0100 To: perldb-interest@vix.com, agw@mail.nerc-bas.ac.uk Subject: Re: DBI-0.61 Date: Wed, 23 Aug 1995 17:21:53 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508231719.aa07361@post.demon.co.uk> > From: Andrew G Wood > > Well you have not heard from me for a few days ... my compiler has been > off on holiday. But as there seemed to be a few new versions flying > around he could not wait to get back into it ..... > > cfe: Error: DBI.c, line 1411: Incompatible pointer type assignment (3.3.16.1, 3.5.4.1(21)) > AV *av = (SV*)((XRV*) (retsv)->sv_any )->xrv_rv ; > ----^ > > C'est la vie! > "Fixed in the next release." (I presume you know how to fix it in your copy - just cast to (AV*)) It's probably better for people to post these kind of compiler complaints direct to me rather than the list. I'll bounce to the list if I think it's of general interest or I need help. Thanks. > Andy. > * Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-42@lion>; Thu, 24 Aug 1995 07:05:57 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 23 Aug 95 22:48:49 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa08795; 23 Aug 95 23:48 +0100 Received: by gw.home.vix.com id AA07237; Wed, 23 Aug 95 07:10:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07229; Wed, 23 Aug 95 07:10:42 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa12139; 23 Aug 95 14:39 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id am08046; 23 Aug 95 14:36 +0100 Received: from ig.co.uk by lion id <11055-0@lion>; Wed, 23 Aug 1995 14:12:22 +0100 To: Davide.Migliavacca@inferentia.it Subject: Re: Proposed "extension": LONG RAW parameters in DBD::Oracle Cc: perldb-interest@vix.com Date: Wed, 23 Aug 1995 14:12:22 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508231436.am08046@post.demon.co.uk> > From: Shari > > What I want to do: > Insert a very big BLOB in a LONG RAW column in an Oracle table > How: > implementing support for the LONG VARRAW datatype in > bind_param/dbd_bind_ph > Why: > Because I like to be able to do it :-) > > Apparently, the most general way of doing it is by using a LONG VARRAW, > which is a buffer with 4 bytes (sb4) prepended to specify the length of > data. In fact, using the length parameter of the obdrn calls one may be > limited to 64k (according to OCI docs). I've avoided using the VAR types since the need to fiddle with the 4 byte length prefix complicates matters (and will mean that the, possibly very big, value string would have to be copied). > To use a LONG VARRAW, in the obdrn call, type is 95 (BTW, in > ~oracle/rdbms/demo/ocidfn.h there are symbolic #defines SQLT_xxx for > Oracle external datatypes) But they aren't there in earlier version of Oracle 7. > and the length parameter should be specified as -1. > > Using bind_param this should become a trivial :-) modification to > dbd_bind_ph in dbdimp.c. > Are you *desparate* for >64k blobs? > There is one additional problem (which of course may be only my scarce > knowledge of Perl). > Using a LONG VARx in C/C++ generally means using a large buffer which is > then accessed (when filling data in) starting at offset sizeof(sb4) so > that no big block moves are necessary aftewards. > Is it possible to do something like this in Perl? Not _in_ perl. Perl SV's are not organised that way. > If yes, enlighten me (please). If not, a programmer will generally do the > big block copy leading to inefficiency and memory waste (we're talking of > -maybe- megabytes). > Exactly. > Any suggestion greatly appreciated. > > Davide Migliavacca > (davide.migliavacca@inferentia.it) > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-44@lion>; Thu, 24 Aug 1995 07:06:11 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 23 Aug 95 22:58:51 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13341; 23 Aug 95 23:58 +0100 Received: by gw.home.vix.com id AA05568; Wed, 23 Aug 95 06:46:37 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05563; Wed, 23 Aug 95 06:46:32 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa12125; 23 Aug 95 14:39 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ai08046; 23 Aug 95 14:36 +0100 Received: from ig.co.uk by lion id <10357-0@lion>; Wed, 23 Aug 1995 13:02:00 +0100 To: perldb-interest@vix.com, Davide.Migliavacca@inferentia.it Subject: RE: (Little) Bug in DBD::Oracle 0.23, Oracle.xs Date: Wed, 23 Aug 1995 13:02:00 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508231436.ai08046@post.demon.co.uk> > From: Shari > > >From Tim Bunce: > > >> From: Shari > >> > >> There is a bug in _execute in file Oracle.xs (starting at line 179 in > >> the 0.23 sources) which forces all parameters to being named :pn with n > >> being its position, > > > It's not a bug. > > Sorry, I must disagree. It IS a bug in that ? placeholders are not > handled correctly. > In fact, they are transformed in :n, while _execute looks for :pn and > doesn't find them. My apologies, I missunderstood. Your original message was not as clear as this and I was reading in a hurry. It's fixed in next release. > >> I am very interested in the bind_param call since it's one of the > >> features I terribly need (working most of my time with BLOBs, that is > >> LONG RAW which are not usable with the current implementation of > >> dbd_bind_ph). > > > In what way 'not usable'? > > In that the automatic type binding in dbd_bind_ph (see my other > message to the mailing list) uses only type 1 and 8. To use LONG RAW you > need type 24, which is not easily deductible from a variable. RAW and > CHAR get a very diffrerent treatment in Oracle. so you will not be able > to bind a LONG variable to a LONG RAW column. So, I need a way to tell > DBI/DBD to bind as a LONG RAW (or a VARBINARY or whatever the name would > be). I used to work on SQL Server some time ago and if I'm not wrong, it > too required different types when using a BLOB (non-interpreted byte) > column as opposed to a very large (but still of string type) column. > The ridiculous thing is that according to the OCI manual, you SHOULD be > able to bind a LONG to a LONG RAW, the obndrv goes fine, but in execution > Oracle screams at you. > Even if the string contains only pairs of hex chars? > > If I don't know I can't fix it. (Then again, if I do know I might not > > fix it, or be able to fix it, anyway :-) > > Sorry, I broke the problem in two, but my messages to the mailing list > come back after a very long time (dunno if it happens only to me or to > others also). > Yes, the list does seem very slow at the moment. > >> I have seen it mentioned many times in the mailing list > >> messages, but didn't find it in DBI nor in the specs. Is it a proposed > >> enhancement? > >> > > Yes. > > I'll look into it again. Nag me from time to time if it doesn't show up. > > I suppose I'll have to wait for a DBI 0.7 specs announcement? > I implemented it that night. It's in the latest release: $sth->bind_param(":foo", $value, { ora_type => 24 }); Happy ? :-) It's untested since it was 7am when I released it and I knew you'd do a better job of testing it than I would :-) I look forward to you comments. NOTE: I'll take this opportunity to remind people that all interfaces outside the Oraperl emulation are very much alpha and liable to change! Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-46@lion>; Thu, 24 Aug 1995 07:06:26 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 23 Aug 95 23:05:21 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa16240; 24 Aug 95 0:05 +0100 Received: by gw.home.vix.com id AA06092; Wed, 23 Aug 95 06:54:35 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA06087; Wed, 23 Aug 95 06:54:33 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa12097; 23 Aug 95 14:39 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac08046; 23 Aug 95 14:36 +0100 Received: from ig.co.uk by lion id <09637-0@lion>; Wed, 23 Aug 1995 11:53:13 +0100 To: perldb-interest@vix.com, jliptak@coefmd3.uswc.uswest.com Subject: Error: no tab in DBD::_::st::fetchrow Date: Wed, 23 Aug 1995 11:53:13 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508231436.ac08046@post.demon.co.uk> > From: John Liptak > > I've never gotten into the perl MakeMaker stuff, what's the problem here? > > I am trying to skip from 0.58, I've been busy with other things :-( > > /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c > Error: no tab in DBD::_::st::fetchrow argument declaration ' SV *sth' Rats! The xsubpp I'm using is a more recent, and more forgiving, one than the 5.001m standard. I'll release a new DBI (and DBD if that also has this problem). Sorry for the hassle. > John Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-51@lion>; Thu, 24 Aug 1995 07:06:54 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 00:03:26 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09634; 24 Aug 95 1:03 +0100 Received: by gw.home.vix.com id AA04674; Wed, 23 Aug 95 13:36:43 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04669; Wed, 23 Aug 95 13:36:36 -0700 Received: from kukini.cs.bris.ac.uk by dira.bris.ac.uk with SMTP (PP); Wed, 23 Aug 1995 21:36:16 +0100 Received: from kiha by kukini.compsci.bristol.ac.uk id aa21422; 23 Aug 95 20:39 GMT To: perldb-interest@vix.com Subject: Compiling Oraperl for SunOS Date: Wed, 23 Aug 95 21:36:09 +0100 From: ally@cs.bris.ac.uk Message-Id: <9508232039.aa21422@uk.ac.bristol.compsci.kukini> Despite searching through the ftp.demon.co.uk archives and this interest group's postings, I still cannot persude Oraperl to compile. SYSTEM ====== uname -a gives: SunOS kiha 4.1.3_U1 3 sun4m Oracle7 Server Release 7.0.13.6.0 Perl v4.036 Oraperl v2.4 My initial error was: /home/msc/94/ally/bin/perl-4.036/perl.h", line 554: redeclaration of sprintf *** Error code 1 make: Fatal error: Command failed for target `oracle.o' So I commented out those lines in perl.h! My current makefile is based on the one from the depths of Oracle's server with relevent alterations as given in oracle7.note7.oracle71 and oracle7.note9.oracle71. # Makefile for Oraperl and Coraperl # Change these to your ORACLE installation directory and Perl # source directory # ORACLE_HOME = /home/kiha/oracle SRC = /home/msc/94/ally/bin/perl-4.036 # Oracle Definitions, copied from $(ORACLE_HOME)/c/demo/proc.mk # ALL_ORA_LIBS is the only entry that the Makefile actually uses; # change it to whatever you need to link Pro*C programs # # Prototyped from Oracle for Sun 7.1.3 release # module to be used for linking with non-deferred option NDFOPT= $(ORACLE_HOME)/rdbms/lib/kpundf.o # flag for linking with non-deferred option (default is # deferred mode) NONDEFER=false OTHERLIBS=`cat $(ORACLE_HOME)/rdbms/lib/sysliblist` $(MLSLIBS) CC=/usr/5bin/cc LDFLAGS=-Bdynamic -L$(LIBHOME) # KA added LDFLAGS LIBCORE=$(LIBHOME)/libcore.a LIBCV6=$(LIBHOME)/libcv6.a LLIBCORE=-lcore LLIBCV6=-lcv6 LLIBNLSRTL=-lnlsrtl CLIBS=$(OTHERLIBS) LIBHOME=$(ORACLE_HOME)/lib LIBOCIC= $(LIBHOME)/libocic.a LLIBORA= -lora OSNTAB= $(LIBHOME)/osntab.o SQLNETLIBS= -lsqlnet CORELIBS= $(LLIBNLSRTL) $(LLIBCV6) $(LLIBCORE) $(LLIBNLSRTL) $(LLIBCV6) \ $(LLIBCORE) TTLIBS= $(OSNTAB) $(SQLNETLIBS) $(LLIBORA) $(LIBPLSHACK) $(SQLNETLIBS) \ $(LLIBORA) $(CORELIBS) $(LDLIBS) $(CLIBS) # libraries for linking oci programs # Deferred OCILDLIBS= $(LIBOCIC) $(TTLIBS) $(LLIBCORE) # Non-deferred #OCILDLIBS= $(ORACLE_HOME)/rdbms/lib/kpundf.o $(LIBOCIC) $(TTLIBS) $(LLIBCORE) MISSEDLIBS = $(ORACLE_HOME)/lib/libnlsrtl.a \ $(ORACLE_HOME)/lib/libcore.a \ $(ORACLE_HOME)/lib/libcv6.a \ $(ORACLE_HOME)/lib/libnetv2.a \ $(ORACLE_HOME)/lib/libnlsrtl.a \ $(ORACLE_HOME)/lib/libnetwork.a \ $(ORACLE_HOME)/lib/libcore.a ALL_ORA_LIBS=$(OCILDLIBS) $(MISSEDLIBS) #OTHERLIBS = `cat $(ORACLE_HOME)/rdbms/lib/sysliblist` #CLIBS = $(OTHERLIBS) #OCILIB = $(ORACLE_HOME)/rdbms/lib/libocic.a #NETLIBS = $(ORACLE_HOME)/rdbms/lib/osntab.o \ # $(ORACLE_HOME)/rdbms/lib/libsqlnet.a #ORALIBS = $(ORACLE_HOME)/rdbms/lib/libora.a #ALL_ORA_LIBS = $(CLIBS) $(OCILIB) $(NETLIBS) $(ORALIBS) # Perl Definitions, taken from $SRC/usub/Makefile # Don't include the curses libraries here - they go in # CURSELIB # GLOBINCS = LOCINCS = LIBS = `. $(SRC)/config.sh; echo $$libs` # Oraperl Definitions # Set DEBUG to -DDEBUGGING, -DPERL_DEBUGGING or leave blank (see orafns.h) # If it is not blank, uncomment the definition of DBUG_O # DEBUG = -DDEBUGGING DBUG_O = dbug/dbug.o This gave the error: oraperl> make -f Makefile oraperl .... .... cc -target sun4 -c dbug.c (back to main directory) /usr/5bin/cc -o oraperl /home/msc/94/ally/bin/perl-4.036/uperl.o oracle.o orafns.o getcursor.o colons.o debug.o strtoul.o usersub.o dbug/dbug.o \ -lm /home/kiha/oracle/lib/libocic.a /home/kiha/oracle/lib/osntab.o -lsqlnet -lora -lsqlnet -lora -lnlsrtl -lcv6 -lcore -lnlsrtl -lcv6 -lcore `cat /home/kiha/oracle/rdbms/lib/sysliblist` -lcore /home/kiha/oracle/lib/libnlsrtl.a /home/kiha/oracle/lib/libcore.a /home/kiha/oracle/lib/libcv6.a /home/kiha/oracle/lib/libnetv2.a /home/kiha/oracle/lib/libnlsrtl.a /home/kiha/oracle/lib/libnetwork.a /home/kiha/oracle/lib/libcore.a `. /home/msc/94/ally/bin/perl-4.036/config.sh; echo $libs` ld: -lsqlnet: No such file or directory *** Error code 4 make: Fatal error: Command failed for target `oraperl' Although the end of the Makefile should not be changed, adding $(LDFLAGS) to the oraperl: definition gives a different error. oraperl: $(SRC)/uperl.o $(OOBJS) $(CC) $(LDFLAGS) -o oraperl $(SRC)/uperl.o $(OOBJS) \ -lm $(ALL_ORA_LIBS) $(LIBS) This gives: .... .... ld: -lnlsrtl: No such file or directory *** Error code 4 make: Fatal error: Command failed for target `oraperl' If anyone has made it to the end, please mail me any suggestions. I've tried our systems administrator and he's stumped as well! Thanks, Karim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-52@lion>; Thu, 24 Aug 1995 07:07:02 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 00:03:27 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09641; 24 Aug 95 1:03 +0100 Received: by gw.home.vix.com id AA13760; Wed, 23 Aug 95 08:48:51 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA13754; Wed, 23 Aug 95 08:48:49 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id KAA53791 for ; Wed, 23 Aug 1995 10:48:41 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA02788; Wed, 23 Aug 1995 10:48:35 -0500 From: mhm@austin.ibm.com Message-Id: <9508231548.AA02788@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Creating a statement handle (sth) on the C side? Date: Wed, 23 Aug 95 10:48:12 -0500 Ok, I have (I think) the latest versions of DBD and DBI and am totally confused about how to create a sth while executing C code. Quick synopsis, even when executing SQL directly, I must still prepare the statement, so I need a sth to hand my equivalent of dbd_db_prepare(sth,statement). Pointers, explicit directions, anything....please. Thanks. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-53@lion>; Thu, 24 Aug 1995 07:07:09 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 00:07:26 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa11119; 24 Aug 95 1:06 +0100 Received: by gw.home.vix.com id AA27575; Wed, 23 Aug 95 11:55:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27570; Wed, 23 Aug 95 11:55:26 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id NAA143134; Wed, 23 Aug 1995 13:55:24 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA18294; Wed, 23 Aug 1995 13:55:22 -0500 From: mhm@austin.ibm.com Message-Id: <9508231855.AA18294@dax.austin.ibm.com> To: Tim.Bunce@ig.co.uk Cc: perldb-interest@vix.com Subject: Problems with DB2 Date: Wed, 23 Aug 95 13:55:22 -0500 Ok, using DBI-0.60 and running the following script: #!/usr/local/bin/perl -w # Base DBD Driver Test print "1..$tests\n"; require DBI; print "ok 1\n"; import DBI; print "ok 2\n"; $DBI::dbi_debug=1; $switch = DBI->internal; (ref $switch eq 'DBI::dr') ? print "ok 3\n" : print "not ok 3\n"; $drh = DBI->install_driver('DB2'); (ref $drh eq 'DBI::dr') ? print "ok 4\n" : print "not ok 4\n"; print "ok 5\n" if $drh->{Version}; BEGIN { $tests = 5 } exit 0; # end. The question is did I miss a subroutine/function or implement one incorrectly, most likely. Or does this happen with DBD::Oracle? |> 1..5 |> ok 1 |> ok 2 |> ok 3 |> DBI->install_driver DBI DB2 |> at test.pl line 17 |> DBI->install_driver(DB2) loaded |> Had to create DBD::DB2::dr::imp_data_size unexpectedly at |> /usr/local/lib/perl5/DBI.pm line 287. |> DBI->install_driver(DB2) = DBI::dr=HASH(0x200767e0) |> ok 4 |> ok 5 |> DBI::END |> DBI::disconnect_all DBI |> DBI::disconnect_all for 'DB2' |> Attempt to free unreferenced scalar at /usr/local/lib/perl5/DBI.pm line 164. |> Thanks. Mike ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-54@lion>; Thu, 24 Aug 1995 07:07:15 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 00:08:02 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa11508; 24 Aug 95 1:07 +0100 Received: by gw.home.vix.com id AA20790; Wed, 23 Aug 95 10:26:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20780; Wed, 23 Aug 95 10:25:47 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa28799; 23 Aug 95 17:13 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa05932; 23 Aug 95 17:10 +0100 Received: from ig.co.uk by lion id <13292-0@lion>; Wed, 23 Aug 1995 17:13:12 +0100 To: perldb-interest@vix.com Subject: BIG NEWS Cc: kejin@empress.com Date: Wed, 23 Aug 1995 17:13:12 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508231710.aa05932@post.demon.co.uk> Just when you thought things were progressing well and we all had some idea where we are going ... things change! *** What's changed? Basically the arrival of the almost-completed SQL/CLI ISO standard and corresponding ODBC v3.0. The (new) CLI spec is rapidly gaining momentum within the industry. Some database vendors (include Infomix, IBM, Watercom, etc.) have adopted ODBC as the native CLI for their database engine(s). I see this trend growing over time. Sun and HP will soon bundle an ODBC driver manager with the O/S. *** What's the impact? A shift in the role of the DBI to more directly fulfill the role of a Driver Manager. The DBI will become a fully fledged ODBC Driver Manager, e.g. able to dynamically load third party commercial ODBC drivers which conform to the standard ODBC driver interface spec. A change in the DBI/DBD interface to bring it closer to the ODBC driver interface spec (consistent return codes etc). The application/DBI interface will probably not change much. (Mainly because I've avoided specifying much of it recently :-) *** How will it happen? Some of you may be aware that Ke Jin from Empress has very recently released the source code to a UNIX ODBC driver manager to alt.sources. The code has already been ported to SunOS, Solaris, HP/UX, UnixWare, Irix, NCR, Linux ELF, DEC Alpha, etc The code was released under the GNU General Public License. Ke has very kindly agreed that we can use the source code under the Perl Artistic Licence. For anyone curious about iODBC I've added it to the DBperl archive at ftp.demon.co.uk:/pub/perl/db/other/iODBC/iODBC-2.0b.shar. This is currently the only FTP'able copy. Ke will send updates to me so we'll keep upto date. In parallel with ongoing perl DBD driver developments I'll be studying the iODBC code to see how it can best be integrated into the DBI. I may initially integrate it as a new DBD driver and then 'absorb' the driver into the DBI. More news as it happens... (it won't be quick) Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17097-66@lion>; Thu, 24 Aug 1995 07:08:46 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 03:45:08 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa07305; 24 Aug 95 4:44 +0100 Received: by gw.home.vix.com id AA06253; Wed, 23 Aug 95 13:57:58 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA06247; Wed, 23 Aug 95 13:57:56 -0700 Received: from localhost by colossus.cse.psu.edu with SMTP id <45646>; Wed, 23 Aug 1995 16:56:45 -0400 X-Mailer: exmh version 1.6.1 5/23/95 To: perldb-interest@vix.com Subject: Re: Proposed "extension": LONG RAW parameters in DBD::Oracle Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 23 Aug 1995 16:56:58 -0400 From: John D Groenveld Message-Id: <95Aug23.165645edt.45646@colossus.cse.psu.edu> BLOB's are clearly gaining importance in our work. Perhaps, they are better manipulated in OCI/Pro*C, but I'd like to see these in perl as well. Just my biased 2 cents, John groenvel@cse.psu.edu ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21821-0@lion>; Thu, 24 Aug 1995 11:32:54 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 09:28:18 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa18983; 24 Aug 95 10:27 +0100 Received: by gw.home.vix.com id AA26025; Wed, 23 Aug 95 19:22:53 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA26020; Wed, 23 Aug 95 19:22:51 -0700 Received: from ccMail by internet_mail_gateway.tigerfund.com (IMA Internet Exchange v1.04) id 03bf0220; Wed, 23 Aug 95 22:21:06 -0500 Mime-Version: 1.0 Date: Wed, 23 Aug 1995 22:25:03 -0500 Message-Id: <03bf0220@internet_mail_gateway.tigerfund.com> From: Benjamin_Lis@tigerfund.com To: perldb-interest@vix.com Cc: Tim.Bunce@ig.co.uk Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Hi: I'm encountering difficulties building DBI on an SP2 running AIX 3.2.5. Attached is the output requested by the README file. I've built and tested Perl5.001m. any help would be greatly appreciated, Ben ------------------------------------------------------------------------------ Benjamin Lis, Jr. Tiger Management Corporation Benjamin_Lis@tigerfund.com phone: 212.984.2412 ------------------------------------------------------------------------------- Script command is started on Wed Aug 23 21:05:03 EDT 1995.tmc5_hps:/u/blis/DBI> make realclean Make: make: 1254-002 Cannot find a rule to create target realclean from dependencies. Stop. tmc5_hps:/u/blis/DBI> perl Maked file.PL PERL_SRC=/u/blis/perl5.000 1m Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker NAME => 'DBI' PERL_SRC => '/u/blis/perl5.001m' VERSION => '0.59' Looking for perl 5 by these names: miniperl perl, in these dirs: /u/blis/perl5.001m /u/blis/perl5.001m . /u/blis/bin /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus/bin/aix /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus/bin/aix/lib /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus/bin/aix /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus /usr/bin /etc /usr/sbin /usr/ucb /usr/bin/X11 /sbin /usr/bin /usr/bin/X11 /usr/local/bin /usr/local/bin/X11 /usr/lpp/ssp/bin /u/oracle/product/7.1.6/bin /usr/local/openint/bin /usr/local/openint/bin /usr/local/tex/bin/rs6000 /usr/local/tex/bin /usr/local/modulus/bin/aix/lib/res /usr/local/bin Executing /u/blis/perl5.001m/miniperl Writing Makefile for DBI Warning (non-fatal): cflags evaluation in MakeMaker differs from shell output package: DBI old: cc -c -D_ALL_SOURCE -O new: cc -c -D_NO_PROTO -O Using 'old' set. Please notify perl5-porters@nicoh.com Warning (non-fatal): cflags evaluation in MakeMaker differs from shell output package: DBI old: cc -c -D_ALL_SOURCE -O new: cc -c -D_NO_PROTO -O Using 'old' set. Please notify perl5-porters@nicoh.com tmc5_hps:/u/blis/DBI> make mkdir /u/blis/perl5.001m/lib/auto/DBI /u/blis/perl5.001m/miniperl /u/blis/perl5.001m/lib/ExtUtils/xsubpp -typemap /u/blis/perl5.001m/lib/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c cc -c -D_ALL_SOURCE -O -I/u/blis/perl5.001m DBI.c "DBI.c", line 789.43: 1506-152 (W) Operands are pointers to different types. ar cr /u/blis/perl5.001m/lib/auto/DBI/DBI.a DBI.o && : /u/blis/perl5.001m/lib/auto/DBI/DBI.a cp DBIXS.h /u/blis/perl5.001m/lib/DBI/DBIXS.h cp lib/DBD/ExampleP.pm /u/blis/perl5.001m/lib/DBD/ExampleP.pm cp lib/DBD/NullP.pm /u/blis/perl5.001m/lib/DBD/NullP.pm cp DBI.pm /u/blis/perl5.001m/lib/DBI.pm tmc5_hps:/u/blis/DBI> make test /u/blis/perl5.001m/perl -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............Identifier "DBD::Test::dr::imp_data_size" used only once: possible typo at t/min.t line 45. Can't load '/u/blis/perl5.001m/lib/auto/DBI/DBI.a' for module DBI: dlopen: /u/blis/perl5.001m/lib/auto/DBI/DBI.a: bad exec format in /u/blis/perl5.001m/lib/auto/DBI/DBI.a at /u/blis/perl5.001m/lib/DynaLoader.pm line 450. at /u/blis/perl5.001m/lib/DBI.pm line 26 FAILED on test 1 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.21 cusr 0.09 csys = 0.30 cpu) /u/blis/perl5.001m/perl -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib test.pl test.pl DBI test application $Revision: 1.8 $ Identifier "DBI::errstr" used only once: possible typo at test.pl line 55. Identifier "DBI::rows" used only once: possible typo at test.pl line 59. Can't load '/u/blis/perl5.001m/lib/auto/DBI/DBI.a' for module DBI: dlopen: /u/blis/perl5.001m/lib/auto/DBI/DBI.a: bad exec format in /u/blis/perl5.001m/lib/auto/DBI/DBI.a at /u/blis/perl5.001m/lib/DynaLoader.pm line 450. at DBI.pm line 26 make: 1254-004 The error code from the last command is 8. Make Quitting. tmc5_hps:/u/blis/DBI> make test TEST_VERV BOSE=1 /u/blis/perl5.001m/perl -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............Identifier "DBD::Test::dr::imp_data_size" used only once: possible typo at t/min.t line 45. 1..6 Can't load '/u/blis/perl5.001m/lib/auto/DBI/DBI.a' for module DBI: dlopen: /u/blis/perl5.001m/lib/auto/DBI/DBI.a: bad exec format in /u/blis/perl5.001m/lib/auto/DBI/DBI.a at /u/blis/perl5.001m/lib/DynaLoader.pm line 450. at /u/blis/perl5.001m/lib/DBI.pm line 26 FAILED on test 1 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.23 cusr 0.05 csys = 0.28 cpu) /u/blis/perl5.001m/perl -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib test.pl test.pl DBI test application $Revision: 1.8 $ Identifier "DBI::errstr" used only once: possible typo at test.pl line 55. Identifier "DBI::rows" used only once: possible typo at test.pl line 59. Can't load '/u/blis/perl5.001m/lib/auto/DBI/DBI.a' for module DBI: dlopen: /u/blis/perl5.001m/lib/auto/DBI/DBI.a: bad exec format in /u/blis/perl5.001m/lib/auto/DBI/DBI.a at /u/blis/perl5.001m/lib/DynaLoader.pm line 450. at DBI.pm line 26 make: 1254-004 The error code from the last command is 8. Make Quitting. tmc5_hps:/u/blis/DBI> make test TEST_VERBOSE=1 /u/blis/perl5.001m/perl -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............Identifier "DBD::Test::dr::imp_data_size" used only once: possible typo at t/min.t line 45. 1..6 Can't load '/u/blis/perl5.001m/lib/auto/DBI/DBI.a' for module DBI: dlopen: /u/blis/perl5.001m/lib/auto/DBI/DBI.a: bad exec format in /u/blis/perl5.001m/lib/auto/DBI/DBI.a at /u/blis/perl5.001m/lib/DynaLoader.pm line 450. at /u/blis/perl5.001m/lib/DBI.pm line 26 FAILED on test 1 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.21 cusr 0.08 csys = 0.29 cpu) /u/blis/perl5.001m/perl -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib -I/u/blis/perl5.001m/lib test.pl test.pl DBI test application $Revision: 1.8 $ Identifier "DBI::errstr" used only once: possible typo at test.pl line 55. Identifier "DBI::rows" used only once: possible typo at test.pl line 59. Can't load '/u/blis/perl5.001m/lib/auto/DBI/DBI.a' for module DBI: dlopen: /u/blis/perl5.001m/lib/auto/DBI/DBI.a: bad exec format in /u/blis/perl5.001m/lib/auto/DBI/DBI.a at /u/blis/perl5.001m/lib/DynaLoader.pm line 450. at DBI.pm line 26 make: 1254-004 The error code from the last command is 8. Make Quitting. tmc5_hps:/u/blis/DBI> perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=, osver=3.2, archname=aix uname='aix harpo 2 3 000035371400 ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-D_NO_PROTO' ccflags ='-D_NO_PROTO' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=1, d_casti32=define, d_castneg= intsize=4, alignbytes=4, usemymalloc=y, randbits=15 Libraries: so=none libpth=/lib /usr/lib /usr/ccs/lib /usr/local/lib libs=-ldbm -lld -lm -lc -lbsd -lPW libc=/usr/lib/libc.a Dynamic Linking: dlsrc=dl_none.xs, dlext=none, d_dlsymun= cccdlflags='', ccdlflags='', lddlflags='' tmc5_hps:/u/blis/DBI> exit Script command is complete on Wed Aug 23 21:09:27 EDT 1995. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04742-5@lion>; Thu, 24 Aug 1995 19:15:07 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 17:50:06 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa27618; 24 Aug 95 18:49 +0100 Received: by gw.home.vix.com id AA08050; Wed, 23 Aug 95 22:58:09 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA08045; Wed, 23 Aug 95 22:58:06 -0700 Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id HAA15499; Thu, 24 Aug 1995 07:57:37 +0200 Date: Thu, 24 Aug 1995 07:57:37 +0200 Message-Id: <199508240557.HAA15499@anna.mind.de> From: Andreas Koenig To: Benjamin_Lis@tigerfund.com Cc: perldb-interest@vix.com, Tim.Bunce@ig.co.uk In-Reply-To: <03bf0220@internet_mail_gateway.tigerfund.com> (Benjamin_Lis@tigerfund.com) Subject: DBI 0.59 problem (Was: [No subject]) Reply-To: andreas.koenig@mind.de >From: Benjamin_Lis@tigerfund.com >I'm encountering difficulties building DBI on an SP2 running AIX 3.2.5. > >Attached is the output requested by the README file. I've built and >tested Perl5.001m. [ output deleted ] From the README file: !!!!BEFORE BUILDING, TESTING AND INSTALLING this you will need to: !!!! !!!! Build, test and install Perl 5 (at least version 5.001m). !!!! It is very important to test it and install it! !!!! ^^^^^^^ BTW, the current version number is 0.61, so please get it from ftp://ftp.demon.co.uk/pub/perl/db/ or ftp://ftp.pasteur.fr/pub/Perl/CPAN/pml/byauthor/TIMB >any help would be greatly appreciated, >Ben Hope that helps, andreas ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07624-25@lion>; Fri, 25 Aug 1995 07:03:59 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 20:06:20 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa03148; 24 Aug 95 21:05 +0100 Received: by gw.home.vix.com id AA03697; Thu, 24 Aug 95 05:29:51 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03692; Thu, 24 Aug 95 05:29:49 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa23450; 24 Aug 95 12:33 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id al19404; 24 Aug 95 12:30 +0100 Received: from ig.co.uk by lion id <00540-0@lion>; Thu, 24 Aug 1995 12:16:51 +0100 To: perldb-interest@vix.com, groenvel@cse.psu.edu Subject: Re: Proposed "extension": LONG RAW parameters in DBD::Oracle Date: Thu, 24 Aug 1995 12:16:51 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508241230.al19404@post.demon.co.uk> > From: John D Groenveld > > > BLOB's are clearly gaining importance in our work. Perhaps, they are better > manipulated in OCI/Pro*C, but I'd like to see these in perl as well. > Just my biased 2 cents, > John > groenvel@cse.psu.edu > I don't follow what you're saying John. Care to rephrase it? Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08341-2@lion>; Fri, 25 Aug 1995 07:05:30 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 21:23:35 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13018; 24 Aug 95 22:22 +0100 Received: by gw.home.vix.com id AA03735; Thu, 24 Aug 95 05:30:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03714; Thu, 24 Aug 95 05:30:02 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa23447; 24 Aug 95 12:33 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ak19404; 24 Aug 95 12:30 +0100 Received: from ig.co.uk by lion id <22270-0@lion>; Thu, 24 Aug 1995 11:56:11 +0100 To: z50816@mip.lasc.lockheed.com Subject: Re: Pointers are not assignment-compatible Cc: perldb-interest@vix.com Date: Thu, 24 Aug 1995 11:56:11 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508241230.ak19404@post.demon.co.uk> > From @netserve.lasc.lockheed.com:z50816@mip.lasc.lockheed.com Thu Aug 24 10:11:50 1995 > Date: Wed, 23 Aug 1995 14:54:18 -0400 > From: Terry Greenlaw > Apparently-To: Tim.Bunce@ig.co.uk > > Tim, > I got these warnings during make: > cc -c -D_POSIX_SOURCE -D_HPUX_SOURCE -Aa +O1 +z -I/disk3/perl5/lib/perl5/hpux/CORE > DBI.c > cc: "DBI.c", line 206: warning 562: Redeclaration for "dbih_getcom"; "static" used. > cc: "DBI.c", line 1411: warning 604: Pointers are not assignment-compatible. > > The tests for DBI 0.61 all pass > > However... > when testing DBD 0.24 after a make (which was successful) > > install_driver(Oracle) failed: DBI version mismatch (DBI actual v0/s4, expected v7/s52) ^^^^^ Something's very wrong with the DBI build. Version 0, size 4 bytes! I'll look into those warnings. > I checked the DBI.pm file and it is dated today. > Any ideas? > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08341-3@lion>; Fri, 25 Aug 1995 07:05:35 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 22:35:07 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa17626; 24 Aug 95 23:34 +0100 Received: by gw.home.vix.com id AA05322; Thu, 24 Aug 95 05:49:04 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05316; Thu, 24 Aug 95 05:49:00 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa24133; 24 Aug 95 12:40 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa20648; 24 Aug 95 12:37 +0100 Received: from ig.co.uk by lion id <00697-0@lion>; Thu, 24 Aug 1995 12:37:20 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Creating a statement handle (sth) on the C side? Date: Thu, 24 Aug 1995 12:37:20 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508241238.aa20648@post.demon.co.uk> > From: mhm@austin.ibm.com > > Ok, I have (I think) the latest versions of DBD and DBI and am totally > confused about how to create a sth while executing C code. > > Quick synopsis, even when executing SQL directly, I must still > prepare the statement, so I need a sth to hand my equivalent of > dbd_db_prepare(sth,statement). > > Pointers, explicit directions, anything....please. > Just do exactly what (the latest) DBD::Oracle::Oracle.pm does. When developing a driver please copy and edit all the DBD::Oracle files rather than working from the ground up. > Ok, using DBI-0.60 and running the following script: > |> Had to create DBD::DB2::dr::imp_data_size unexpectedly at > |> /usr/local/lib/perl5/DBI.pm line 287. You're not doing what DBD/Oracle/Oracle.xs does: DBI_IMP_SIZE("DBD::Oracle::dr::imp_data_size", sizeof(imp_drh_t)); The bottom line here is that the DBI/DBD interface is undocumented. The DBD::Oracle interface must serve as both model and documentation for other drivers. Follow it *very* *very* closely. Especially the .xs file. I also urge you to upgrade to DBI-0.61 and base your DB2 driver on DBD-Oracle-0.24. > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07624-39@lion>; Fri, 25 Aug 1995 07:05:44 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 23:02:53 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa01058; 25 Aug 95 0:02 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa00193; 25 Aug 95 0:02 GMT-60:00 Received: by gw.home.vix.com id AA16090; Thu, 24 Aug 95 08:27:51 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA16084; Thu, 24 Aug 95 08:27:47 -0700 Received: from bssmlsa.nerc-bas.ac.uk.nerc-bas.ac.uk by bsfiles.nerc-bas.ac.uk (8.6.10/NERC-1.2(Solaris 2.x) id QAA17067; Thu, 24 Aug 1995 16:27:23 +0100 Date: Thu, 24 Aug 1995 16:27:17 +0100 Message-Id: <26257.199508241527@bssmlsa.nerc-bas.ac.uk.nerc-bas.ac.uk> From: Andrew G Wood To: Tim.Bunce@ig.co.uk Cc: mhm@austin.ibm.com, perldb-interest@vix.com In-Reply-To: <9508241455.aa15156@post.demon.co.uk> (message from Tim Bunce on Thu, 24 Aug 1995 14:41:15 +0100) Subject: Re: Creating a statement handle (sth) on the C side? Reply-To: agw@mail.nerc-bas.ac.uk Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII >>>>> "Tim" == Tim Bunce writes: >> From: mhm@austin.ibm.com >> >> Tim Bunce writes: |> |> I also urge you to >> upgrade to DBI-0.61 and base your DB2 driver |> on >> DBD-Oracle-0.24. >> >> Tell me where to find them as they are not at ftp.demon.uk. >> > $ ftp ftp.demon.co.uk ... ftp> dir DBI Er .. not quite ..... I think you will find that should be /pub/perl/db/DBI > ... -rw-r--r-- 1 4098 53 42791 Aug 22 06:06 DBI-0.61.tar.Z ... ftp> dir DBD/Oracle ditto. > ... -rw-r--r-- 1 4098 53 40631 Aug 22 06:03 > DBD-Oracle-0.24.tar.Z ... >> Thanks. >> >> Mike >> > Tim. Andy. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07624-40@lion>; Fri, 25 Aug 1995 07:05:50 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 23:07:02 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa03232; 25 Aug 95 0:06 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa00384; 25 Aug 95 0:06 GMT-60:00 Received: by gw.home.vix.com id AA07148; Thu, 24 Aug 95 06:10:35 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07143; Thu, 24 Aug 95 06:10:33 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id IAA68612; Thu, 24 Aug 1995 08:10:28 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA25894; Thu, 24 Aug 1995 08:10:27 -0500 From: mhm@austin.ibm.com Message-Id: <9508241310.AA25894@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Creating a statement handle (sth) on the C side? In-Reply-To: Your message of Thu, 24 Aug 95 12:37:20 +0100. <9508241238.aa20648@post.demon.co.uk> Date: Thu, 24 Aug 95 08:10:26 -0500 Tim Bunce writes: |> |> I also urge you to upgrade to DBI-0.61 and base your DB2 driver |> on DBD-Oracle-0.24. Tell me where to find them as they are not at ftp.demon.uk. Thanks. Mike ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07624-42@lion>; Fri, 25 Aug 1995 07:06:03 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 24 Aug 95 23:26:22 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa11456; 25 Aug 95 0:25 +0100 Received: by gw.home.vix.com id AA07936; Thu, 24 Aug 95 06:29:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07931; Thu, 24 Aug 95 06:29:23 -0700 Received: from localhost by colossus.cse.psu.edu with SMTP id <45619>; Thu, 24 Aug 1995 09:28:10 -0400 X-Mailer: exmh version 1.6.1 5/23/95 To: perldb-interest@vix.com Subject: Re: Proposed "extension": LONG RAW parameters in DBD::Oracle In-Reply-To: Your message of "Thu, 24 Aug 1995 07:16:51 EDT." <9508241230.al19404@post.demon.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 24 Aug 1995 09:28:28 -0400 From: John D Groenveld Message-Id: <95Aug24.092810edt.45619@colossus.cse.psu.edu> I'm sorry for being vague. I think its critical for DBPerl's success to be able to handle all types of objects including binary objects greater than 64k. I understand the difficulties with using Oracle's LONG VAR facilities, but even if it is an expensive operation, I would like to see it included. Thanks, John groenvel@cse.psu.edu ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08341-10@lion>; Fri, 25 Aug 1995 07:06:24 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 00:22:33 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa02011; 25 Aug 95 1:21 +0100 Received: by gw.home.vix.com id AA29367; Thu, 24 Aug 95 11:18:20 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29358; Thu, 24 Aug 95 11:18:16 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa21292; 24 Aug 95 16:54 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa09457; 24 Aug 95 16:51 +0100 Received: from ig.co.uk by lion id <02658-0@lion>; Thu, 24 Aug 1995 16:53:44 +0100 To: agw@mail.nerc-bas.ac.uk Subject: Re: Creating a statement handle (sth) on the C side? Cc: mhm@austin.ibm.com, perldb-interest@vix.com Date: Thu, 24 Aug 1995 16:53:44 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508241651.aa09457@post.demon.co.uk> > From: Andrew G Wood > > >>>>> "Tim" == Tim Bunce writes: > > >> From: mhm@austin.ibm.com > >> > >> Tim Bunce writes: |> |> I also urge you to > >> upgrade to DBI-0.61 and base your DB2 driver |> on > >> DBD-Oracle-0.24. > >> > >> Tell me where to find them as they are not at ftp.demon.uk. > >> > > $ ftp ftp.demon.co.uk ... > ftp> dir DBI > > Er .. not quite ..... I think you will find that should be > /pub/perl/db/DBI > Er, _you're_ telling _me_ where _I_ put my own releases! That's rich. :-) I thought it was fairly obvious that I was summarising. Mike replied to me with: > Sorry about that, seems the cache flushing mechanism on my web browser > doesn't work. Appearantly, I was looking at cached pages and not > at ftp.demon.co.uk. > > Time to go wash the egg off of my face :-) > > Mike > No harm done. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08341-11@lion>; Fri, 25 Aug 1995 07:06:32 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 00:22:35 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa02020; 25 Aug 95 1:21 +0100 Received: by gw.home.vix.com id AA07127; Thu, 24 Aug 95 12:50:12 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07107; Thu, 24 Aug 95 12:50:11 -0700 Received: from morpheus.cis.yale.edu (morpheus [130.132.143.248]) by capitoline.cis.yale.edu (8.6.12/8.6.12) with ESMTP id PAA05449 for ; Thu, 24 Aug 1995 15:50:00 -0400 Received: (from rivory@localhost) by morpheus.cis.yale.edu (8.6.12/8.6.12) id PAA28604; Thu, 24 Aug 1995 15:49:59 -0400 Date: Thu, 24 Aug 1995 15:49:59 -0400 (EDT) From: Royal Hansen X-Sender: rivory@morpheus To: interest perl Subject: unsubscribe Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII unsubsribe ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07624-49@lion>; Fri, 25 Aug 1995 07:06:53 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 00:55:32 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa12052; 25 Aug 95 1:54 +0100 Received: by gw.home.vix.com id AA14062; Thu, 24 Aug 95 07:54:22 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA14049; Thu, 24 Aug 95 07:54:18 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa08754; 24 Aug 95 14:58 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa15156; 24 Aug 95 14:55 +0100 Received: from ig.co.uk by lion id <00848-0@lion>; Thu, 24 Aug 1995 14:41:15 +0100 To: mhm@austin.ibm.com Subject: Re: Creating a statement handle (sth) on the C side? Cc: perldb-interest@vix.com Date: Thu, 24 Aug 1995 14:41:15 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508241455.aa15156@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > |> I also urge you to upgrade to DBI-0.61 and base your DB2 driver > |> on DBD-Oracle-0.24. > > Tell me where to find them as they are not at ftp.demon.uk. > $ ftp ftp.demon.co.uk ... ftp> dir DBI ... -rw-r--r-- 1 4098 53 42791 Aug 22 06:06 DBI-0.61.tar.Z ... ftp> dir DBD/Oracle ... -rw-r--r-- 1 4098 53 40631 Aug 22 06:03 DBD-Oracle-0.24.tar.Z ... > Thanks. > > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07624-50@lion>; Fri, 25 Aug 1995 07:06:59 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 00:56:20 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa12388; 25 Aug 95 1:56 +0100 Received: by gw.home.vix.com id AA10169; Thu, 24 Aug 95 13:31:05 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10163; Thu, 24 Aug 95 13:31:02 -0700 Received: from nando.net.nando.net (parsifal.nando.net) by merlin.nando.net (4.1/davel-nando/dec93) id AA07753; Thu, 24 Aug 95 16:30:33 EDT Received: by nando.net.nando.net (4.1/SMI-4.1) id AA06882; Thu, 24 Aug 95 16:30:38 EDT Date: Thu, 24 Aug 95 16:30:38 EDT From: Bhctc2 Message-Id: <9508242030.AA06882@nando.net.nando.net> To: perldb-interest@vix.com Subject: unsubscribe unsubscribe ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08341-17@lion>; Fri, 25 Aug 1995 07:07:27 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 03:15:31 GMT Received: from gw.home.vix.com by punt.demon.co.uk id ab15316; 25 Aug 95 4:14 +0100 Received: by gw.home.vix.com id AA14201; Thu, 24 Aug 95 14:16:45 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA14195; Thu, 24 Aug 95 14:16:44 -0700 Received: from centhub ([151.116.23.137]) by uswat.advtech.uswest.com (8.6.12/8.6.12) with SMTP id PAA10518; Thu, 24 Aug 1995 15:16:43 -0600 Original-Received: by centhub.mnet.uswest.com (M-Net Hub.950111) PP-warning: Illegal Received field on preceding line Received: from coefmd3 by coefmd3.uswest.com. (5.0/SMI-SVR4) id AA18474; Thu, 24 Aug 1995 15:17:01 +0700 Message-Id: <9508242117.AA18474@coefmd3.uswest.com.> X-Mailer: exmh version 1.6 4/21/95 To: perldb-interest@vix.com Cc: Tim.Bunce@ig.co.uk Subject: DBD-Oracle-0.24 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 24 Aug 1995 15:17:00 -0600 From: John Liptak Content-Length: 2389 I know that several things have been happening with bind, but I'm having problems with the current oraperl emulation. I don't seem to be able to get DBI to deal with the following script: #!/home/jliptak/bin/perl # use Oraperl; $lda = &ora_login('T:coefmd2:stage', '/', '/') || die "Error logging in\n"; $a = "select TABLE_NAME \ from ALL_TABLES \ order by TABLE_NAME"; $c = "select COLUMN_NAME, \ DATA_LENGTH, \ DATA_TYPE, \ DATA_PRECISION, \ DATA_SCALE \ from ALL_TAB_COLUMNS \ where TABLE_NAME = :1 \ order by COLUMN_ID"; $b = &ora_open($lda, $a); $d = &ora_open($lda, $c) || die "Can't open second cursor\n"; while ( @table = &ora_fetch($b) ) { print "$table[0] \n"; &ora_bind($d, $table[0]) || die "Can't bind\n"; while ( @field = &ora_fetch($d) ) { print "@field \n"; } } I get the following errors: .... IAPCOMMENT ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 IAPFLD ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 IAPMAP ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 IAPSQLTXT ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 IAPTRG ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 IAPTRIGGER ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 INFORMATION_OBJECT ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 INV_ALTERNATE_DESC ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 INV_ALTERNATE_DESC_T ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: 1 no statement executing at test.pl line 35 ..... John -- John Liptak - MTS U S WEST Communications WK (303) 624-0140 931 14TH Street 9TH FL. PG (303) 820-9284 Denver, Colorado 80202 j.liptak@ieee.org jliptak@coefmd3.mnet.uswest.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <18152-11@lion>; Fri, 25 Aug 1995 15:59:15 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 14:41:00 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa17903; 25 Aug 95 15:40 +0100 Received: by gw.home.vix.com id AA00665; Fri, 25 Aug 95 01:52:15 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00660; Fri, 25 Aug 95 01:52:11 -0700 Received: from barbera.inferentia.it (barbera.inferentia.it [151.99.220.61]) by relay.smau.it (8.6.12/8.6.12) with SMTP id KAA26425 for ; Fri, 25 Aug 1995 10:51:03 +0200 Posted-Date: Fri, 25 Aug 1995 10:51:03 +0200 Received: by barbera.inferentia.it with Microsoft Mail id <303E0F4B@barbera.inferentia.it>; Fri, 25 Aug 95 10:58:35 PDT From: Shari To: "'perldb-interest@vix.com'" Subject: Future of DBI Date: Fri, 25 Aug 95 10:58:00 PDT Message-Id: <303E0F4B@barbera.inferentia.it> Encoding: 125 TEXT X-Mailer: Microsoft Mail V3.0 I hope this is not going to arouse a flame war. With the recent announcement (see "BIG NEWS" from Tim Bunce of 1995/80/23) of the actual chance of having a single interface (ODBC 3/CLI/DBI) across platforms, maybe the focus should shift towards achieving that goal. I myself am really interested in In the meantime, the following 20 lines of perl and 20 lines of PL/SQL may substitute WOW (notice the use of :p1 insted of ? is mandated by a bug which is probably already fixed). I think this means DBI is really something real. Notice: this code may not actually work as it is, since it's an UNTESTED cut & paste from something which I am right now developing on a contract. The "real" sample, anyway, which is complicated by something which has nothing to do with the interface to the DB, it does work and you will see it soon in production (Tim: see how I care for the ALPHA warnings :-) as the english interface to the search engine of MAGELLANO on line (http://www.smau.it/magellano). Having arrays be bound would certainly speed things up. Having complete (and efficient) BLOB support would have me happy. But having a complete and "definitive" DBI spec would make me happy even more (if possible). One last word before DBI speaks: sorry for the long posting if you have a slow link. Davide Migliavacca (davide.migliavacca@inferentia.it) ======================= cut here (wow.pl) ====================== #!/usr/bin/perl # (C) 1995, Davide Migliavacca (davide.migliavacca@inferentia.it) # Use as you will but please don't let someone pay for this. # Requires DBI 0,61, DBD::Oracle 0.24 use Oraperl; use CGI; $WOW_UID = "some/user\@SOMESID"; $query = new CGI; $wow_call = substr ($query->path_info, 1); if ($query->param > 0) { # query-style call, add an argument and prepare the others $wow_call .= '('; @params = $query->param; for $i (0 .. ($query->param - 1)) { $wow_call .= ",@params[$i]=>'" . $query->param(@params[$i]). "'"; } } $wow_call .= ')'; } $lda = ora_login ('', $WOW_UID, ''); ora_do ($lda, "begin $wow_call; dbiWOW.getAllLines; end;") || die; $cursor_getnum = ora_open ($lda, 'select dbiWOW.numlines from dual'); $cursor_getline = $lda->prepare ('select dbiWOW.line(:p1) from dual'); ($num_lines) = ora_fetch ($cursor_getnum); #print "dbiWOW.numlines == $num_lines\n"; ora_close ($cursor_getnum); $i = 1; while ($i <= $num_lines) { $cursor_getline->bind_param(':p1', $i); $cursor_getline->execute; ($line) = ora_fetch ($cursor_getline); print "$line\n"; # now this is actually quite unprofessional :-) $i++; } ======================= cut here ====================== accompanied by the following PL/SQL package (very similar to wow.sql) ======================= cut here (dbiwow.sql) ============= Rem (C) 1995, Davide Migliavacca (davide.migliavacca@inferentia.it) Rem Use as you will but please don't let someone pay for this. drop package dbiWOW; create package dbiWOW as w3output dbms_output.chararr; rowcount integer := 1; procedure getAllLines; function line (i integer) return varchar2; PRAGMA RESTRICT_REFERENCES(line, WNDS, WNPS, RNDS); function numLines return integer; PRAGMA RESTRICT_REFERENCES(numlines, WNDS, WNPS, RNDS); end dbiWOW; / show errors; create package body dbiWOW is procedure getAllLines is rc integer := 1000000; i integer := 0; begin dbms_output.get_lines(w3output,rc); rowcount := rc; end getAllLines; function line (i integer) return varchar2 is begin return w3output(i); end line; function numLines return integer is begin return rowcount; end numLines; end dbiWOW; / show errors; =========================== cut here ========================= ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <00364-8@lion>; Sat, 26 Aug 1995 13:27:01 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 21:00:10 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa13886; 25 Aug 95 21:58 +0100 Received: by gw.home.vix.com id AA20896; Fri, 25 Aug 95 06:10:23 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20788; Fri, 25 Aug 95 06:10:13 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa22170; 25 Aug 95 14:01 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa17232; 25 Aug 95 13:58 +0100 Received: from ig.co.uk by lion id <14511-0@lion>; Fri, 25 Aug 1995 12:43:49 +0100 To: perldb-interest@vix.com, jliptak@coefmd3.uswc.uswest.com Subject: Re: DBD-Oracle-0.24 Date: Fri, 25 Aug 1995 12:43:49 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508251358.aa17232@post.demon.co.uk> > From: John Liptak > > I know that several things have been happening with bind, but I'm having > problems with the current oraperl emulation. I don't seem to be able > to get DBI to deal with the following script: > > ERROR EVENT: DBI::st=HASH(0xa311c)->DBI::default_handler: > 1 no statement executing Here's the fix... change code near the end of execute() in Oracle.xs from: if (error) XSRETURN_UNDEF; /* dbd_bind_ph already registered error */ to: if (error) { XSRETURN_UNDEF; /* dbd_bind_ph already registered error */ } > John Tim. p.s. While I'm here I'll quickly mention the fact that it seems that sword is defined as an int and the SWORDMAXVAL is the same as INT_MAX. The implication of this is that LONG's in the current release are _not_ limited to 64K bytes on platforms where an int is 4 bytes. Can those using longs please test this for me. Thanks. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <00363-24@lion>; Sat, 26 Aug 1995 13:28:00 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 19:58:15 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa21808; 25 Aug 95 20:57 +0100 Received: by gw.home.vix.com id AA16211; Fri, 25 Aug 95 05:09:03 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA16204; Fri, 25 Aug 95 05:09:01 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa12550; 25 Aug 95 12:27 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac29574; 25 Aug 95 12:24 +0100 Received: from ig.co.uk by lion id <14266-0@lion>; Fri, 25 Aug 1995 12:22:43 +0100 To: perldb-interest@vix.com, Davide.Migliavacca@inferentia.it Subject: Re: DBD::Oracle 0.24: ORACLE_HOME determination Date: Fri, 25 Aug 1995 12:22:43 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508251224.ac29574@post.demon.co.uk> > From: Shari > > The sub driver in package DBD::Oracle (Oracle.pm 1.19, starting at line > 24) does a good job trying to determine where @#$%! is Oracle if > ORACLE_HOME is not set. > I wonder if an additional effort would be worthwhile: getting the > "oracle" user home directory. > I had to patch (add an entry) the code because with the brain-damaged > OSF/1 3.2 and Netsite Communication Server 1.0 combination, user nobody > with shell sh would never be able to see ORACLE_HOME unless I use the > 'eval exec perl -x' hack in the CGI script emulating WOW. What use is knowing the "oracle" user home directory? > Of course this is the lowest priority issue I could devise in order to > distract you all from hard work. My deflector shields are up at the moment so I'll say: Send me some code and I'll integrate it. > Bye > Davide Migliavacca (davide.migliavacca@inferentia.it) > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <00765-3@lion>; Sat, 26 Aug 1995 13:28:05 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 20:21:41 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa02620; 25 Aug 95 21:21 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <23491-0@frigate.doc.ic.ac.uk>; Fri, 25 Aug 1995 21:21:26 +0100 Received: by gw.home.vix.com id AA13000; Fri, 25 Aug 95 04:09:54 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA12995; Fri, 25 Aug 95 04:09:52 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa10424; 25 Aug 95 12:00 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab25342; 25 Aug 95 11:58 +0100 Received: from ig.co.uk by lion id <10333-0@lion>; Fri, 25 Aug 1995 10:30:45 +0100 To: perldb-interest@vix.com, groenvel@cse.psu.edu Subject: Re: Proposed "extension": LONG RAW parameters in DBD::Oracle Date: Fri, 25 Aug 1995 10:30:45 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508251158.ab25342@post.demon.co.uk> > From: John D Groenveld > > I'm sorry for being vague. I think its critical for DBPerl's success to be able > to handle all types of objects including binary objects greater than 64k. > I understand the difficulties with using Oracle's LONG VAR facilities, but > even if it is an expensive operation, I would like to see it included. Me to. I'll look into it. > Thanks, > John > groenvel@cse.psu.edu > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <00765-13@lion>; Sat, 26 Aug 1995 13:29:00 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 22:18:28 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa25275; 25 Aug 95 23:17 +0100 Received: by gw.home.vix.com id AA00197; Fri, 25 Aug 95 01:45:01 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00184; Fri, 25 Aug 95 01:44:59 -0700 Received: from barbera.inferentia.it (barbera.inferentia.it [151.99.220.61]) by relay.smau.it (8.6.12/8.6.12) with SMTP id KAA21480; Fri, 25 Aug 1995 10:42:56 +0200 Posted-Date: Fri, 25 Aug 1995 10:42:56 +0200 Received: by barbera.inferentia.it with Microsoft Mail id <303E0D64@barbera.inferentia.it>; Fri, 25 Aug 95 10:50:28 PDT From: Shari To: "'perldb-interest@vix.com'" Cc: 'Tim Bunce' Subject: DBD::Oracle 0.24: ORACLE_HOME determination Date: Fri, 25 Aug 95 10:50:00 PDT Message-Id: <303E0D64@barbera.inferentia.it> Encoding: 14 TEXT X-Mailer: Microsoft Mail V3.0 The sub driver in package DBD::Oracle (Oracle.pm 1.19, starting at line 24) does a good job trying to determine where @#$%! is Oracle if ORACLE_HOME is not set. I wonder if an additional effort would be worthwhile: getting the "oracle" user home directory. I had to patch (add an entry) the code because with the brain-damaged OSF/1 3.2 and Netsite Communication Server 1.0 combination, user nobody with shell sh would never be able to see ORACLE_HOME unless I use the 'eval exec perl -x' hack in the CGI script emulating WOW. Of course this is the lowest priority issue I could devise in order to distract you all from hard work. Bye Davide Migliavacca (davide.migliavacca@inferentia.it) ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <00363-39@lion>; Sat, 26 Aug 1995 13:29:13 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 22:40:30 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa05815; 25 Aug 95 23:40 +0100 Received: by gw.home.vix.com id AA15464; Fri, 25 Aug 95 11:27:19 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA15458; Fri, 25 Aug 95 11:27:15 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa09972; 25 Aug 95 17:19 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa24378; 25 Aug 95 17:16 +0100 Received: from ig.co.uk by lion id <20330-0@lion>; Fri, 25 Aug 1995 17:18:47 +0100 To: mhm@austin.ibm.com Subject: Re: Ok, I'm lost... _prepare(sth,statement) Cc: perldb-interest@vix.com Date: Fri, 25 Aug 1995 17:18:47 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508251716.aa24378@post.demon.co.uk> > From: mhm@austin.ibm.com > > mhm writes: > |> > |> Ok, I'm finally to the point where I can start debugging DB2 transactions > |> and I get an error from SQLPrepare saying > |> > |> "The numeric literal 0x2007cc50" is not valid." > |> > |> So, I have an SQL error, lets check the args, first argument is corrent, > |> but the second which is supposed to be a text string contains > |> > |> "DBI::db=HASH(0x2007cc50)" > |> > > Well, I now know how I get the latter string. When I step into > > { package DBD::DB2::db; # ====== DATABASE ====== > use strict; > > sub prepare { > my($dbh, $statement)= @_; <---- @_ contains 3 arguments and my > ---- string is arg 3 > > Anyone have any ideas as to why? > Oh well, back to the debugger. > You're probably calling prepare in the wrong way. It should be: $dbh->prepare($statement); > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <00363-40@lion>; Sat, 26 Aug 1995 13:29:17 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 22:40:30 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa05670; 25 Aug 95 23:39 +0100 Received: by gw.home.vix.com id AA04690; Fri, 25 Aug 95 09:11:43 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04684; Fri, 25 Aug 95 09:11:42 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id LAA174184; Fri, 25 Aug 1995 11:11:40 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA17012; Fri, 25 Aug 1995 11:11:38 -0500 From: mhm@austin.ibm.com Message-Id: <9508251611.AA17012@dax.austin.ibm.com> To: Tim.Bunce@ig.co.uk Cc: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Ok, I'm lost... _prepare(sth,statement) In-Reply-To: Your message of Fri, 25 Aug 95 09:39:00 -0500. <9508251439.AA27524@dax.austin.ibm.com> Date: Fri, 25 Aug 95 11:11:37 -0500 mhm writes: |> |> Ok, I'm finally to the point where I can start debugging DB2 transactions |> and I get an error from SQLPrepare saying |> |> "The numeric literal 0x2007cc50" is not valid." |> |> So, I have an SQL error, lets check the args, first argument is corrent, |> but the second which is supposed to be a text string contains |> |> "DBI::db=HASH(0x2007cc50)" |> Well, I now know how I get the latter string. When I step into { package DBD::DB2::db; # ====== DATABASE ====== use strict; sub errstr { DBD::DB2::errstr(@_); } sub prepare { my($dbh, $statement)= @_; <---- @_ contains 3 arguments and my ---- string is arg 3 # create a 'blank' dbh my $sth = DBI::_new_sth($dbh, { 'Statement' => $statement, }); DBD::DB2::st::_prepare($sth, $statement) or return undef; $sth; } } Anyone have any ideas as to why? Oh well, back to the debugger. Mike ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <00765-21@lion>; Sat, 26 Aug 1995 13:30:06 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 25 Aug 95 23:57:09 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09893; 26 Aug 95 0:56 +0100 Received: by gw.home.vix.com id AA27070; Fri, 25 Aug 95 07:39:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27064; Fri, 25 Aug 95 07:39:04 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id JAA168414; Fri, 25 Aug 1995 09:39:03 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA27524; Fri, 25 Aug 1995 09:39:01 -0500 From: mhm@austin.ibm.com Message-Id: <9508251439.AA27524@dax.austin.ibm.com> To: Tim.Bunce@ig.co.uk Cc: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Ok, I'm lost... _prepare(sth,statement) Date: Fri, 25 Aug 95 09:39:00 -0500 Ok, I'm finally to the point where I can start debugging DB2 transactions and I get an error from SQLPrepare saying "The numeric literal 0x2007cc50" is not valid." So, I have an SQL error, lets check the args, first argument is corrent, but the second which is supposed to be a text string contains "DBI::db=HASH(0x2007cc50)" which appears to be some form of DBI magic which has me totally confused. Where did this magic occur? Why? Thanks. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <13150-12@lion>; Mon, 28 Aug 1995 07:03:11 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 26 Aug 95 20:04:31 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa14591; 26 Aug 95 21:04 +0100 Received: by gw.home.vix.com id AA23781; Sat, 26 Aug 95 12:05:03 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA23772; Sat, 26 Aug 95 12:04:58 -0700 Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id VAA24274; Sat, 26 Aug 1995 21:04:13 +0200 Date: Sat, 26 Aug 1995 21:04:13 +0200 Message-Id: <199508261904.VAA24274@anna.mind.de> From: Andreas Koenig To: Tim.Bunce@ig.co.uk Cc: perldb-interest@vix.com In-Reply-To: <9508261842.af01059@post.demon.co.uk> (message from Tim Bunce on Sat, 26 Aug 1995 18:41:26 +0100) Subject: Re: Announce: DBI-0.62 and DBD-Oracle-0.25 Reply-To: andreas.koenig@mind.de >Available from ftp.demon.co.uk:/pub/perl/db/... as usual. Aslo at ftp://ftp.pasteur.fr/pub/Perl/CPAN/pml/byauthor/TIMB/DBI-0.62.tar.Z ftp://ftp.pasteur.fr/pub/Perl/CPAN/pml/byauthor/TIMB/DBD-Oracle-0.25.tar.Z andreas ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <13150-18@lion>; Mon, 28 Aug 1995 07:03:50 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 26 Aug 95 21:01:44 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa06939; 26 Aug 95 22:01 +0100 Received: by gw.home.vix.com id AA20438; Sat, 26 Aug 95 11:06:38 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20433; Sat, 26 Aug 95 11:06:35 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa08760; 26 Aug 95 18:44 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id af01059; 26 Aug 95 18:42 +0100 Received: from ig.co.uk by lion id <06784-0@lion>; Sat, 26 Aug 1995 18:41:26 +0100 To: perldb-interest@vix.com Subject: Announce: DBI-0.62 and DBD-Oracle-0.25 Date: Sat, 26 Aug 1995 18:41:26 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508261842.af01059@post.demon.co.uk> Available from ftp.demon.co.uk:/pub/perl/db/... as usual. Current DBI Implementation Notes: $ary_ref = $sth->fetch; currently always returns the same array ref. This may be a feature that stays, then again it may not. Changes in DBI 0.62, 26 Aug 1995 Added method redirection method $h->func(..., $method_name). This is now the official way to call private driver methods that are not part of the DBI standard. E.g.: @ary = $sth->func('ora_types'); It can also be used to call existing methods. Has very low cost. $sth->bind_col columns now start from 1 (not 0) to match SQL. $sth->bind_columns now takes a leading attribute parameter (or undef), e.g., $sth->bind_columns($attribs, \$col1 [, \$col2 , ...]); Added handy DBD_ATTRIBS_CHECK macro to vet attribs in XS. Added handy DBD_ATTRIB_GET_SVP, DBD_ATTRIB_GET_BOOL and DBD_ATTRIB_GET_IV macros for handling attributes. Fixed STORE for NUM_OF_FIELDS and NUM_OF_PARAMS. Added FETCH for NUM_OF_FIELDS and NUM_OF_PARAMS. Dispatch no longer bothers to call _untie(). Faster startup via install_method/_add_dispatch changes. Changes in DBD::Oracle 0.25, 26 Aug 1995: User visible changes: Fixed bug in parsing '?' style placeholders. $sth->bind_param now checks for a too long LONG type string. $sth->prepare now takes attributes: ora_parse_lang => 0 | 1 | 2 (v6, auto v6/v7, v7) ora_parse_defer => 0 | 1 (control defered parsing) Other changes to be propogated to other drivers: Added usage of new DBD_ATTRIB_* macros to XS and C code. dbd_bind_ph return value inverted to make it consistent. Added braces around XSRETURN_UNDEF in execute. Checked for describe failure in FETCH. Removed NumParams since it's now called NUM_OF_PARAMS in DBI. Updated NEED_DBIXS_VERSION in Oracle.h Enjoy. Current plans (in no particular order - and with no promises :-) ... Implement DBD::Ingres. Implement a row cache for DBD::Oracle. Upgrade Oraperl emulation to beta status! Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <18444-3@lion>; Tue, 29 Aug 1995 07:01:48 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 28 Aug 95 20:42:13 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa23823; 28 Aug 95 21:41 +0100 Received: by gw.home.vix.com id AA03662; Mon, 28 Aug 95 02:23:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03641; Mon, 28 Aug 95 02:22:20 -0700 Received: from indigo.UUCP by actcom.co.il with UUCPindigo (8.6.12/actcom-0.1) id LAA00463; Mon, 28 Aug 1995 11:07:21 +0300 (rfc931-sender: uucp@localhost) Received: from sparc1a.Indigo.co.il by Indigo.co.il (4.1/SMI-4.1) id AA20919; Mon, 28 Aug 95 10:45:14 IDT Received: from swamp.indigo.co.il. by sparc1a.Indigo.co.il (4.1/SMI-4.1) id AA03402; Mon, 28 Aug 95 10:50:12 IDT Received: by swamp.indigo.co.il. (5.x/SMI-SVR4) id AA00664; Mon, 28 Aug 1995 10:46:40 +0300 Date: Mon, 28 Aug 1995 10:46:40 +0300 From: Peter Gordon Message-Id: <9508280746.AA00664@swamp.indigo.co.il.> To: distribution:;, see end of body MMDF-Warning: Parse error in original version of preceding line at punt.demon.co.uk Subject: Perl5 Places of Interest A couple of months ago I asked for information regarding where various modules could be found. I compiled the following list which I post from time to time. Could you have a quick look to see if anything is amiss, and then I will ask Randall to post it. Thanks Peter Some FTP Archives for Perl Modules: Upload to: ftp://ftp.cis.ufl.edu/pub/incoming ? Notify: spp@temples.syrinx.com (Stephen P Potter) Upload to: ftp://ftp.demon.co.uk/incoming Notify: uploads@demon.net News::NNTPClient.pm ftp://metronet.com/pub/perl/perl5/extensions/NNTPClient-0.17 or send mail at rodger@mail.boi.hp.com with the body of the message beginning with: send nntpclient PERL script that would write Lotus spreadsheets ftp://ftp.cis.ufl.edu/pub/perl/scripts/db/formats/ For any database + perl look under ftp.demon.co.uk:/pub/perl/db ftp.demon.co.uk:/pub/perl/db/mod/Sybase/sybperl-2b2.tar.gz An interface between Postgres-4.2 and Perl5 ftp.demon.co.uk:pub/perl/db/mod/pgperl5-1.1.tar.gz Fortran::NameList - still in alpha requests by email to gelato@oort.ap.sissa.it (Sergio Gelato) or list mailto:gelato@sissa.it DBZ_File - will remain alpha indefinitely for lack of interest requests by email to ian@pipex.net (Ian Phillipps) Font::AFM and URI::URL modules http://www.oslonett.no/home/aas/perl/ http://pubweb.nexor.co.uk/public/perl/contrib/lib/URI/URL.html pgperl ftp.ast.cam.ac.uk:/pub/kgb/pgperl/ (United Kingdom) linux.nrao.edu:/pub/packages/pgperl/ (United States) http://www.ast.cam.ac.uk/~kgb/pgperl.html Directory Size 1291290 Jul 5 17:37 pgplot5.0.2.tar.Z pgperl_formdemo.pg Example of pgperl Basically this is a Web Form which lets the user type in a mathematical function which is then plotted and returned as a GIF by a pgperl CGI script. http://www.ast.cam.ac.uk/cgi-bin/pgperl_formdemo.pg DBD:db2 Not ready SGMLS David Megginson A Post-Processor for SGMLS and NSGMLS ftp://aix1.uottawa.ca/pub/dmeggins/ pointers to the online docs http://www.uottawa.ca/~dmeggins/Index.html HTML::Base alpha rev 0.6 ftp://ftp.acoates.com/perl/HTML_0.6.tar.gz DBI Not ready RCS and RCE Waiting approval for release rjray@lookout.ecte.uswc.uswest.com AddINC (by Graham Barr) AlarmCall CallerItem Date::GetDate (by Graham Barr) Domainname DumpStack EventDrivenServer GL InitializeServer Installing.general IPC_SysV JACKS.html Language::Prolog (by Jack Shirazi) Mail::RFC822 (by Graham Barr) MailTools MathPari Net-FTP (by Spider Boardman) Net-ext (by Spider Boardman) NetTools Prolog-alpha SelfLoader SelfStubber Socket Statistics::Descriptive (by Jason Kastner) UNIVERSAL Usage (by Jack Shirazi) ftp://ftp.icnet.uk/icrf-public/biu/perlmods GD.pm http://www-genome.wi.mit.edu/ftp/pub/software/WWW/GD.html for creating and manipulating GIF files Sys::AlarmCall (by Jack Shirazi) ftp://ftp.icnet.uk/icrf-public/biu/perlmods/AlarmCall-1.1.tar.gz DoWhatIWant (by Water Mellon) mailto:A.Koenig@franz.ww.tu-berlin.de DoWhatIWant+ (by Water Mellon) mailto:A.Koenig@franz.ww.tu-berlin.de File::Attrib (by Tye McQueen) File::KGlob (by Tye McQueen) http://www.metronet.com/~tye/ Ioctl (by Kenneth Albanowski) Term::Control (by Kenneth Albanowski) Term::Info (by Kenneth Albanowski) Term::ReadKey (by Kenneth Albanowski) ftp://numeric.adr.com/pub/perl/modules http://numeric.adr.com/Perl/Perl5Modules.html MD5 (by Neil Winton) ftp://ftp.khoros.unm.edu/pub/perl/extensions/MD5-1.1.tar.gz Math::Pari (by Ilya Zakharevich) ExtUtils-Peek-0.3.tar.Z FreezeThaw-0.1buggy.tar.Z MathPari_a3.tar.gz Term-Gnuplot-0.2.tar.Z Term-ReadLine-0.6.tar.Z Text-Trie-0.2.tar.Z cperl-mode.el perldb-rendesvous.zip perldb-kit-0.5.tar.gz <--A _really_ good debugger for perl5 - Dumps structures and other wonders ftp://ftp.math.ohio-state.edu/pub/users/ilya/perl Msql (by Andreas Koumlnig) ftp://franz.ww.tu-berlin.de/pub/perl News::NNTPClient (by Rodger Anderson) ftp://metronet.com/pub/perl/perl5/extensions/NNTPClient-0.17 Also, send mailto:rodger@boi.hp.com to the author and ask for it. Put send nntpclient as the first thing in the body of the message. SGI::FM (by Amos Shapira) SGI::GL (by Amos Shapira) SGI::SysCalls (by Amos Shapira) ftp://ftp.huji.ac.il/users/amoss/SGI-*-0.1.tar.gz Sx (by Frederic Chauveau) ftp://ftp.pasteur.fr/pub/Perl Tcl::Tk (by Malcolm Beattie) ftp://ftp.ox.ac.uk/pub/perl/Tcl-b1.tar.gz> Tcl-b1.tar.gz Text::TeX (by Ilya Zakharevich) Text::Trie (by Ilya Zakharevich) ftp://ftp.math.ohio-state.edu/pub/users/ilya/perl WAIS (by Ulrich Pfeifer) ftp://ls6-www.informatik.uni-dortmund.de/pub/wais/waisperl*tar.gz URI::URL 3.02 lwp-0.01.tar.gz http://www.oslonett.no/home/aas/perl/www/ http://pubweb.nexor.co.uk/public/perl/contrib/lib/URI/URL.html http://www-genome.wi.mit.edu/ftp/distribution/software/WWW/ On line manual for URL netcdf http://www.unidata.ucar.edu/packages/netcdf-perl/ Perl 4 routines for converting between the various representations of Japanese ftp://ftp.ora.com/pub/examples/nutshell/ujip/perl Tk-b8 ftp.wpi.edu:perl5/Tk-b8.tar.gz PerlVision ftp://netropolis.org/pub/users/hash/perlvision.tgz Other stuff http://www.wg.omron.co.jp/~jfriedl/perl/ Interesting Perl Home Pages http://www.gov.nb.ca/hotlist/perl.htm trans Tim Freeman tsf@netcom.com Implements atomic transactions on Unix files in Perl. The package can ensure that consistent changes are made to a set of files, even in the presence of power failures, concurrent use, and errors. ftp://ftp.netcom.com/pub/ts/tsf/trans.tar.gz WebPH.pl Otis Gospodnetic otisg@cobalt.middlebury.edu ftp://cobalt.middlebury.edu/pub/WWW/WebPH/ http://www.middlebury.edu/cgi-bin/WebPH.pl HTML info http://www.ncsa.uiuc.edu/demoweb/html-primer.html http://uts.cc.utexas.edu/~churchh/htmlchek.html http://hoohoo.ncsa.uiuc.edu/webmonitor/ NT Perl 5.001l, by hip communications / Microsoft: ftp://ntperl.hip.com/ntperl NT Perl 4.036, by Intergraph: ftp://ftp.intergraph.com/pub/win32/perl ftp tools http://www.ee.ryerson.ca:8080/~elf/xapps/ftptools.shar menu.pl ftp://ftp.iastate.edu/pub/perl/perlmenu.v3.2.tar.Z Online PERL FAQ http://pubweb.nexor.co.uk/public/perl/faq/intro.html Disk Monitor I have a Perl/Tk TCP client/server app to do this, using df results. Monitored machine/filesystems are categorized into 5 severity levels and displayed as colored bar graphs, updated once/minute. Even allows some primitive load-leveling of user home directories among filesystems (no AFS at this time). ftp.Lehigh.EDU:pub/monds/monds-2.0.tar.gz Requires Perl 5.001m and Tk-b8 extension. MIME::Base64 MIME::QuotedPrintable are available from: http://www.oslonett.no/home/aas/perl/MIME.tar.gz Why Perl. ftp://mox.perl.com/pub/perl/versus/java ftp://mox.perl.com/pub/perl/versus/python ftp://mox.perl.com/pub/perl/versus/tcl-discussion ftp://mox.perl.com/pub/perl/versus/tcl-complaints SAFE Perl ftp://mox.perl.com/pub/perl/ext/README.Safe ftp://mox.perl.com/pub/perl/ext/Safe-b1.tar.gz %%% overflow headers %%% To: descarte@mcqueen.com, ajs@ajs.com, amoss@cs.huji.ac.il, awilcox@world.std.com, k@otto.ww.TU-Berlin.DE, asherman@world.std.com, P.B.Schuller@cuci.nl, bbum@friday.com, petheram@acm.org, bailey@HMIVAX.HUMGEN.UPENN.EDU, CGI-perl@webstorm.com, perldb-interest@vix.com, dmeggins@aix1.uottawa.ca, darrylo@sr.hp.com, roehrich@cray.com, eay@mincom.oz.au, Eric.Arnold@Corp.Sun.COM, fmc@pasteur.fr, aas@oslonett.no, greg@ftp.netgate.net, bossert@hardees.Rutgers.EDU, Graham.Barr@tiuk.ti.com, gsar@engin.umich.edu, gmarzot@baynetworks.com, spaf@cs.purdue.edu, ian@pipex.net, ilya@math.ohio-state.edu, js@biu.icnet.uk, Jarkko.Hietaniemi@hut.fi, jason@wagner.com, jeffrey@best.com, jv@inter.NL.net, jwatson@cnj.digex.net, kgb@ast.cam.ac.uk, kjahds@kjahds.com, karl@it.kth.se, lstein@genome.wi.mit.edu, lenst@lysator.liu.se, lwall@netlabs.com, libwww-perl@ics.uci.edu, mab@wdl.loral.com, Marc.Paquette@crim.ca, mpeppler@itf.ch, emergl@stgl.sel.alcatel.de, hedlund@best.com, mhm@austin.ibm.com, mbeattie@sable.ox.ac.uk, mrg@mame.mu.oz.au, shields@tembel.org, nik@tiuk.ti.com, nwinton@axion.bt.co.uk, ozawa@prince.pe.u-tokyo.ac.jp, pat@bronco.advance.com, peter@zditr1.arcs.ac.at, pmarquess@bfsec.bt.co.uk, perl5-porters@nicoh.com, guenther@gac.edu, pvision@netropolis.org, olson@mcs.anl.gov, rik.harris@fulcrum.com.au, rjray@lookout.ecte.uswc.uswest.com, rodger@boi.hp.com, S.E.Brenner@bioc.cam.ac.uk, support@unidata.ucar.edu, gelato@oort.ap.sissa.it, skunz@iastate.edu, spider@orb.nashua.nh.us, ptk@wpi.edu, T.L.Priest@LaRC.NASA.GOV, tchrist@mox.perl.com, tom@ssd.csd.harris.com, Tim.Bunce@ig.co.uk, tye@metronet.com, pfeifer@woodstock.informatik.uni-dortmund.de, William_Setzer@ncsu.edu, wscott@ichips.intel.com, Wei-Yuen_Tan@hip.com, ylu@mail.utexas.edu %%% end overflow headers %%% ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <18421-24@lion>; Tue, 29 Aug 1995 07:04:34 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 29 Aug 95 03:18:29 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa22896; 29 Aug 95 4:18 +0100 Received: by gw.home.vix.com id AA15498; Mon, 28 Aug 95 17:17:53 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA15492; Mon, 28 Aug 95 17:17:49 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id TAA77586; Mon, 28 Aug 1995 19:17:45 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA23606; Mon, 28 Aug 1995 19:17:43 -0500 From: mhm@austin.ibm.com Message-Id: <9508290017.AA23606@dax.austin.ibm.com> To: Tim.Bunce@ig.co.uk Cc: perldb-interest@vix.com Subject: DB2perl is minimally working..... Date: Mon, 28 Aug 95 19:17:42 -0500 Well, I finally have DB2perl working on a minimal level. :-) To be precise, it works as long as the SQL does not contain parameters. Hence, the foundation for my question, I am trying to use the DBIc_NUM_FIELDS(sth) macro without success. From within my version of int dbd_st_prepare(SV *sth, char *statement) I think I am calling DBIc_NUM_FIELDS(x) with the wrong argument, while the num_fields member of the dbih_stc_t structure gets set, it is only set in the local instance of the structure. Currently, the code does the following int dbd_st_prepare(sth, statement) SV *sth; char *statement; { D_imp_sth(sth); D_imp_dbh_from_sth; short params =0; ..... SQLNumParams(imp_sth->phstmt,¶ms); /* returns # of parameters */ ..... DBIc_NUM_PARAMS(imp_sth) = params; DBIc_IMPSET_on(imp_sth); return 1; } Yet, after I return from said function the value of $sth->{'NUMPARAMS'} is undefined. Am I missing something intuitively obvious? Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03684-1@lion>; Wed, 30 Aug 1995 10:50:17 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 29 Aug 95 16:39:53 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa28746; 29 Aug 95 17:39 +0100 Received: by gw.home.vix.com id AA15441; Tue, 29 Aug 95 00:48:41 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA15425; Tue, 29 Aug 95 00:47:32 -0700 Received: from padington.ic.uva.nl (paddington.ic.uva.nl) by mail.ic.uva.nl with SMTP id AA12278 (5.67b/IDA-1.4.4 for perldb-interest@vix.com); Tue, 29 Aug 1995 09:46:52 +0200 Message-Id: <199508290746.AA12278@mail.ic.uva.nl> Date: Tue, 29 Aug 1995 09:53:30 (MET) From: ks@ic.uva.nl, s@ic.uva.nl MMDF-Warning: Parse error in original version of preceding line at punt.demon.co.uk Subject: Happy birthday DBperl! To: perldb-interest@vix.com Cc: X-Mailer: WSBeeper 2.6 As Tim Bunce wrote on Thu, 17 Aug 1995 22:39:06 +0100: > > For those too young to remember, the DBperl effort will have its > third birthday on September 29th. It would be kind'a nice to have > something to show for it after three years :-) > ^ ^ ^ | | | -------- ========== \----------/ It's a small cake for me but a big one for mankind Tim! Regards, Karel Sprenger Informatiseringscentrum | phone: +31-20-525 2302 Universiteit van Amsterdam | +31-20-525 2741 Turfdraagsterpad 9, NL-1012 XT AMSTERDAM | fax : +31-20-525 2084 *** PGP Public Key available on servers *** | home : +31-20-675 0989 ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03895-22@lion>; Wed, 30 Aug 1995 10:53:07 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 29 Aug 95 20:18:13 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa24993; 29 Aug 95 21:17 +0100 Received: by gw.home.vix.com id AA28507; Tue, 29 Aug 95 10:02:16 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA28499; Tue, 29 Aug 95 10:02:15 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa04878; 29 Aug 95 17:11 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id af26976; 29 Aug 95 17:09 +0100 Received: from ig.co.uk by lion id <27220-0@lion>; Tue, 29 Aug 1995 13:42:09 +0100 To: Davide.Migliavacca@inferentia.it Subject: Re: (Unimportant) Patch to dbdimp.c Cc: perldb-interest@vix.com Date: Tue, 29 Aug 1995 13:42:09 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508291709.af26976@post.demon.co.uk> > From: Shari > > To avoid core dumps in AIX where the query is long: > > Davide Migliavacca (davide.migliavacca@inferentia.it) > > *** dbdimp.c Sat Aug 26 19:39:17 1995 > --- dbdimp.c.new Mon Aug 28 12:11:39 1995 > ! char msg[99]; > if (oparse_defer == 0) /* was actually parsed now */ > ! sprintf(msg,"possibly parse error at character %d of %d in '%s'", > --- 311,319 ---- > ! char msg[200]; > if (oparse_defer == 0) /* was actually parsed now */ > ! sprintf(msg,"possibly parse error at character %d of %d in '%.140s'", Yes, thanks. I'd meant to do this awhile back (I think someone else sent me a similar patch ages ago). Here's my new version which you could try out for me: if (oparse(imp_sth->cda, (text*)imp_sth->statement, (sb4)-1, (sword)oparse_defer, (ub4)oparse_lng)) { SV *msgsv; char msg[99]; sprintf(msg,"possibly parse error at character %d of %d in '", imp_sth->cda->peo+1, (int)strlen(imp_sth->statement)); msgsv = sv_2mortal(newSVpv(msg,0)); sv_catpv(msgsv, imp_sth->statement); sv_catpv(msgsv, "'"); ora_error(sth, imp_sth->cda, imp_sth->cda->rc, SvPV(msgsv,na)); oclose(imp_sth->cda); /* close the cursor */ return 0; } Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03895-32@lion>; Wed, 30 Aug 1995 10:54:04 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 29 Aug 95 22:16:12 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa25250; 29 Aug 95 23:15 +0100 Received: by gw.home.vix.com id AA28472; Tue, 29 Aug 95 10:01:37 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA28466; Tue, 29 Aug 95 10:01:36 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab04844; 29 Aug 95 17:11 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab26976; 29 Aug 95 17:08 +0100 Received: from ig.co.uk by lion id <26743-0@lion>; Tue, 29 Aug 1995 12:43:10 +0100 To: mhm@austin.ibm.com Subject: Re: DB2perl is minimally working..... Cc: perldb-interest@vix.com Date: Tue, 29 Aug 1995 12:43:10 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508291708.ab26976@post.demon.co.uk> > From: mhm@austin.ibm.com > > Well, I finally have DB2perl working on a minimal level. :-) Great. > To be precise, it works as long as the SQL does not contain parameters. > Hence, the foundation for my question, I am trying to use the > DBIc_NUM_FIELDS(sth) macro without success. > > >From within my version of int dbd_st_prepare(SV *sth, char *statement) > I think I am calling DBIc_NUM_FIELDS(x) with the wrong argument, while > the num_fields member of the dbih_stc_t structure gets set, it is only > set in the local instance of the structure. > > Currently, the code does the following > Umm, the comments above say 'DBIc_NUM_FIELDS', the code below says 'DBIc_NUM_PARAMS'. > int > dbd_st_prepare(sth, statement) > SV *sth; > char *statement; > { > D_imp_sth(sth); > D_imp_dbh_from_sth; > short params =0; > ..... > SQLNumParams(imp_sth->phstmt,¶ms); /* returns # of parameters */ > ..... > DBIc_NUM_PARAMS(imp_sth) = params; > DBIc_IMPSET_on(imp_sth); > return 1; > } > > Yet, after I return from said function the value of $sth->{'NUMPARAMS'} > is undefined. Am I missing something intuitively obvious? > I presume $sth->{'NUMPARAMS'} should be $sth->{'NUM_OF_PARAMS'} since 'NUM_OF_PARAMS' is the attribute name that the DBI recognises. I'd appreciate a snapshot of your work-so-far Mike at some point Mike. Given the closeness of the DB2 interface and SQLCLI/ODBC It'll be interesting for me. > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03684-51@lion>; Wed, 30 Aug 1995 10:54:37 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 00:06:13 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa17390; 30 Aug 95 1:06 +0100 Received: by gw.home.vix.com id AA28587; Tue, 29 Aug 95 10:03:14 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA28579; Tue, 29 Aug 95 10:03:12 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa04880; 29 Aug 95 17:11 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ag26976; 29 Aug 95 17:09 +0100 Received: from ig.co.uk by lion id <27381-0@lion>; Tue, 29 Aug 1995 13:55:06 +0100 To: Davide.Migliavacca@inferentia.it Subject: Re: Error building DBI 0.62 on AIX 3.2.5 Cc: perldb-interest@vix.com Date: Tue, 29 Aug 1995 13:55:06 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508291709.ag26976@post.demon.co.uk> > From: Shari > > uname -a > AIX sauternes 2 3 002021417600 > > Only warning during build: > > cc -c -D_ALL_SOURCE -O -I/usr/local/lib/perl5/aix/CORE DBI.c > "DBI.c", line 1049.19: 1506-152 (W) Operands are pointers to different > types. > Should be harmless (just add an (AV*) cast on the corresponding .xs line). > But: make test produces > > Can't locate DBI object method "0" via package "DBD::Switch::dr" at > t/min.t line 19. Umm, perhaps that cast problem is not so harmless on AIX. This looks like the line: $h->debug(0); is being treated as $h->0(); which would happen if the debug method had been given the IMA_FUNC_REDIRECT attribute. Add the cast and let me know if that fixes it or not. > Davide Migliavacca (davide.migliavacca@inferentia.it) > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04955-0@lion>; Wed, 30 Aug 1995 10:55:31 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 04:24:53 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa01153; 30 Aug 95 5:24 +0100 Received: by gw.home.vix.com id AA27340; Tue, 29 Aug 95 15:58:17 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27328; Tue, 29 Aug 95 15:58:14 -0700 Received: from alma.mcqueen.com (alma.mcqueen.com [193.133.160.249]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with ESMTP id XAA24381; Tue, 29 Aug 1995 23:52:34 +0100 From: descarte@mcqueen.com Received: (from descarte@localhost) by alma.mcqueen.com (8.6.12/8.6.12) id XAA13444; Tue, 29 Aug 1995 23:58:24 +0100 Message-Id: <199508292258.XAA13444@alma.mcqueen.com> Subject: DBD::mSQL driver *New release* To: perldb-interest@vix.com, msql-list@bond.edu.au Date: Tue, 29 Aug 1995 23:58:23 +0100 (BST) Cc: k@anna.mind.de, Tim.Bunce@ig.co.uk X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 606 Hi. Just a quick note to say that the newest release of the mSQL driver for perl5 has been released on: ftp://ftp.mcqueen.com/pub/languages/perl5/DBPerl at version 0.60pl0. This release has been kicked out quickly to see how well it build on machines other than mine. The upgrades have been due to upheaval in the DBI interface. Some bug fixing from previous releases has been done, and there are a few more to come by the end of the week. Not all the mSQL API has been implemented yet, but the major stuff is there! Let me know how it goes! A. -- Alligator Descartes | descarte@mcqueen.com | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-23@lion>; Thu, 31 Aug 1995 07:03:10 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 20:24:05 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa21715; 30 Aug 95 21:23 +0100 Received: by gw.home.vix.com id AA27355; Wed, 30 Aug 95 06:31:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27344; Wed, 30 Aug 95 06:31:04 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab11844; 30 Aug 95 13:11 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ag27772; 30 Aug 95 13:08 +0100 Received: from ig.co.uk by lion id <07300-0@lion>; Wed, 30 Aug 1995 12:36:55 +0100 To: perldb-interest@vix.com, ks@ic.uva.nl, s@ic.uva.nl Subject: Re: Happy birthday DBperl! Date: Wed, 30 Aug 1995 12:36:55 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508301308.ag27772@post.demon.co.uk> > From: ks@ic.uva.nl, s@ic.uva.nl > > As Tim Bunce wrote on Thu, 17 Aug 1995 22:39:06 +0100: > > > > For those too young to remember, the DBperl effort will have its > > third birthday on September 29th. It would be kind'a nice to have > > something to show for it after three years :-) > > > > ^ ^ ^ > | | | > -------- > ========== > \----------/ > > It's a small cake for me but a big one for mankind Tim! > Hey, thanks! (It'll also double as a rather good steam ship :-) Can I be gready and ask for another one on the September 29th... > Regards, > Karel Sprenger > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-25@lion>; Thu, 31 Aug 1995 07:03:22 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 20:42:06 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa01198; 30 Aug 95 21:41 +0100 Received: by gw.home.vix.com id AA20304; Wed, 30 Aug 95 04:52:31 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20299; Wed, 30 Aug 95 04:52:29 -0700 Message-Id: <9508301152.AA20299@gw.home.vix.com> Received: from mip.lasc.lockheed.com by netserve.lasc.lockheed.com with SMTP (1.37.109.4/16.2) id AA12937; Wed, 30 Aug 95 07:52:13 -0400 Received: by mip.lasc.lockheed.com (1.38.193.4/16.2) id AA10718; Wed, 30 Aug 1995 07:52:00 -0400 Date: Wed, 30 Aug 1995 07:52:00 -0400 From: Terry Greenlaw Apparently-To: perldb-interest@vix.com Tim, I'm trying to install DBI 0.62, and get the following message during "make test". The "perl Makefile.PL" and "make" went okay. I tried this before and after a "make install". Shouldn't it be trying _install_method instead of _install_me ? Any ideas? TIA, tog Mip> make test /usr/bin/perl -I./blib -I./blib -I/disk3/perl5/lib/perl5/hpux -I/disk3/perl5/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............Can't locate auto/DBI/_install_me.al in @INC at blib/DBI.pm line 100 FAILED on test 1 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.30 cusr 0.06 csys = 0.36 cpu) /usr/bin/perl -I./blib -I./blib -I/disk3/perl5/lib/perl5/hpux -I/disk3/perl5/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ Can't locate auto/DBI/_install_me.al in @INC at DBI.pm line 100 BEGIN failed--compilation aborted at test.pl line 18. *** Error code 2 Stop. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-33@lion>; Thu, 31 Aug 1995 07:04:01 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 22:11:34 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa17259; 30 Aug 95 23:10 +0100 Received: by gw.home.vix.com id AA10727; Wed, 30 Aug 95 09:37:50 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10718; Wed, 30 Aug 95 09:37:49 -0700 Message-Id: <9508301637.AA10718@gw.home.vix.com> Received: from mip.lasc.lockheed.com by netserve.lasc.lockheed.com with SMTP (1.37.109.4/16.2) id AA13908; Wed, 30 Aug 95 12:37:34 -0400 Received: by mip.lasc.lockheed.com (1.38.193.4/16.2) id AA02210; Wed, 30 Aug 1995 12:37:21 -0400 Date: Wed, 30 Aug 1995 12:37:21 -0400 From: Terry Greenlaw To: perldb-interest@vix.com Subject: DBI not compiling Oops, An addendum to the last message. DBI.bs is zero length. Hmmmmm. tog ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-35@lion>; Thu, 31 Aug 1995 07:04:13 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 22:13:49 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa18510; 30 Aug 95 23:13 +0100 Received: by gw.home.vix.com id AA27350; Wed, 30 Aug 95 06:31:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27345; Wed, 30 Aug 95 06:31:05 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa11849; 30 Aug 95 13:11 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ah27772; 30 Aug 95 13:08 +0100 Received: from ig.co.uk by lion id <07351-0@lion>; Wed, 30 Aug 1995 12:42:04 +0100 To: Davide.Migliavacca@inferentia.it Subject: RE: Error building DBI 0.62 on AIX 3.2.5 Cc: perldb-interest@vix.com Date: Wed, 30 Aug 1995 12:42:04 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508301308.ah27772@post.demon.co.uk> > From: Shari > > Nope. > What I have done: > Modified line 980 in DBI.xs > > if (atype == SVt_PVAV) { > ! ima_usage_av = (AV *)SvRV(attribs); > } > > adding the cast as (I think) you've said. That's right. > Make now doesn't complain an hitch. > Anyway, make test results in exactly the same error. > Anything I can do? Since I can't reproduce it you'll have to debug it yourself. Things to consider: Run with PERL_DBI_DEBUG=3 Is it the first method call that goes through the dispatch function? Check the value of ima->flags in the dispatch function. Is anyone else having this problem? > Davide > Tim. > ---------- > > > But: make test produces > > > > Can't locate DBI object method "0" via package "DBD::Switch::dr" at > > t/min.t line 19. > > Umm, perhaps that cast problem is not so harmless on AIX. This looks like > the line: > > $h->debug(0); > > is being treated as > > $h->0(); > > which would happen if the debug method had been given the > IMA_FUNC_REDIRECT > attribute. > > Add the cast and let me know if that fixes it or not. > ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-40@lion>; Thu, 31 Aug 1995 07:04:42 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 22:53:58 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09072; 30 Aug 95 23:52 +0100 Received: by gw.home.vix.com id AA27343; Wed, 30 Aug 95 06:31:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27334; Wed, 30 Aug 95 06:30:57 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa11859; 30 Aug 95 13:11 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ak27772; 30 Aug 95 13:08 +0100 Received: from ig.co.uk by lion id <07828-0@lion>; Wed, 30 Aug 1995 13:07:15 +0100 To: z50816@mip.lasc.lockheed.com Subject: _install_me.al Cc: perldb-interest@vix.com Date: Wed, 30 Aug 1995 13:07:15 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508301308.ak27772@post.demon.co.uk> > From: Terry Greenlaw > > Tim, I'm trying to install DBI 0.62, and get the following message during "make test". > The "perl Makefile.PL" and "make" went okay. I tried this before and after a > "make install". > Shouldn't it be trying _install_method instead of _install_me ? > Any ideas? > > Mip> make test > /usr/bin/perl -I./blib -I./blib -I/disk3/perl5/lib/perl5/hpux -I/disk3/perl5/lib/perl5 > -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t > t/min...............Can't locate auto/DBI/_install_me.al in @INC at blib/DBI.pm line 100 > FAILED on test 1 _install_method used to be defined in DBI.pm, it's now defined in DBI.xs. The auto/DBI/_install_me.al error is simply a symptom of _install_method not being found and the AutoLoader doing its best to find one (the shortened name is a last ditch attempt to deal with filesystems which don't support long names). The real question is why did perl not see the _install_method in DBI.xs? Make sure that DBI.c, DBI.o and DBI.so get built and DBI.so gets copied into ./blib. If so then try again with PERL_DL_DEBUG=2 and check that the DynaLoader is loading the right .so file. Has anyone else had this problem? Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-45@lion>; Thu, 31 Aug 1995 07:05:16 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 30 Aug 95 23:34:43 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa29604; 31 Aug 95 0:34 +0100 Received: by gw.home.vix.com id AA10464; Wed, 30 Aug 95 09:32:16 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10452; Wed, 30 Aug 95 09:32:14 -0700 Message-Id: <9508301632.AA10452@gw.home.vix.com> Received: from mip.lasc.lockheed.com by netserve.lasc.lockheed.com with SMTP (1.37.109.4/16.2) id AA13893; Wed, 30 Aug 95 12:31:59 -0400 Received: by mip.lasc.lockheed.com (1.38.193.4/16.2) id AA01670; Wed, 30 Aug 1995 12:31:47 -0400 Date: Wed, 30 Aug 1995 12:31:47 -0400 From: Terry Greenlaw To: perldb-interest@vix.com Subject: Re: _install_me.al > From <@post.demon.co.uk,@ignite.demon.co.uk:Tim.Bunce@ig.co.uk> Wed Aug 30 09:30 EDT 1995 > Received: from marble.litc.lockheed.com by mip.lasc.lockheed.com with SMTP > (1.38.193.4/16.2) id AA17088; Wed, 30 Aug 1995 09:30:23 -0400 > Return-Path: <@post.demon.co.uk,@ignite.demon.co.uk:Tim.Bunce@ig.co.uk> > Received: from disperse.demon.co.uk (disperse.demon.co.uk [158.152.1.77]) by marble.litc.lockheed.com (8.6.12/8.6.9) with SMTP id GAA04937 for ; Wed, 30 Aug 1995 06:34:37 -0700 > Received: from post.demon.co.uk by disperse.demon.co.uk id aa11859; > 30 Aug 95 13:11 +0100 > Received: from ignite.demon.co.uk by post.demon.co.uk id ak27772; > 30 Aug 95 13:08 +0100 > Received: from ig.co.uk by lion id <07828-0@lion>; > Wed, 30 Aug 1995 13:07:15 +0100 > To: z50816@mip.lasc.lockheed.com > Subject: _install_me.al > Cc: perldb-interest@vix.com > Date: Wed, 30 Aug 1995 13:07:15 +0100 > From: Tim Bunce > Sender: Tim.Bunce@ig.co.uk > Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > Message-Id: <9508301308.ak27772@post.demon.co.uk> > Status: RO > The real question is why did perl not see the _install_method in DBI.xs? > Make sure that DBI.c, DBI.o and DBI.so get built and DBI.so gets copied into > ./blib. If so then try again with PERL_DL_DEBUG=2 and check that the DynaLoader > is loading the right .so file. > Has anyone else had this problem? > Tim. I have DBI.sl and DBI.xs under ./blib/auto/DBI. SO and DLEXT are both set to sl LD_RUN_PATH="" ls -o ./blib/auto/DBI/DBI.sl -b DBI.o looks a little suspicious, though The STDERR output from make test PERL_DL_DEBUG=2: DynaLoader.pm loaded (/lib/pa1.1 /lib /usr/lib /usr/local/lib) Can't locate auto/DBI/_install_me.al in @INC at blib/DBI.pm line 100 Failed 1 test, 0.00% okay. DynaLoader.pm loaded (/lib/pa1.1 /lib /usr/lib /usr/local/lib) Can't locate auto/DBI/_install_me.al in @INC at DBI.pm line 100 BEGIN failed--compilation aborted at test.pl line 18. Stop. Help...... tog Terry Greenlaw (on-site @ Lockheed) Encompass Technologies z50816@mip.lasc.lockheed.com terry@encompass.is.net ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-49@lion>; Thu, 31 Aug 1995 07:05:47 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 00:53:21 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ad02915; 31 Aug 95 1:52 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <12526-0@frigate.doc.ic.ac.uk>; Wed, 30 Aug 1995 21:57:12 +0100 Received: by gw.home.vix.com id AA03819; Wed, 30 Aug 95 08:09:18 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03814; Wed, 30 Aug 95 08:09:17 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id KAA158161; Wed, 30 Aug 1995 10:09:16 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for mhm@austin.ibm.com at austin.ibm.com; id AA26346; Wed, 30 Aug 1995 10:09:13 -0500 From: mhm@austin.ibm.com Message-Id: <9508301509.AA26346@dax.austin.ibm.com> To: Tim Bunce Cc: mhm@austin.ibm.com, perldb-interest@vix.com Subject: Re: DB2perl is minimally working..... In-Reply-To: Your message of Wed, 30 Aug 95 14:44:29 +0100. <9508301508.ac17339@post.demon.co.uk> Date: Wed, 30 Aug 95 10:09:12 -0500 Tim Bunce writes: |> > |> 'NUM_OF_PARAMS' is the attribute name that the DBI recognises. |> > |> > Hmmmmm, could have sworn my copy of DBI.pm said "NUMPARAMS". |> |> Don't trust the text at the end of DBI.pm. It's not been updated much. |> Trust the code. Reminder posted in painfully obvious place on monitor :-) |> > I did just try it with the correct attribute "NUM_OF_PARAMS" without |> > success. It gets set locally but not globally. |> > |> Does your .xs code for FETCH look like this: |> |> void |> FETCH(sth, keysv) |> SV * sth |> SV * keysv |> CODE: |> SV *valuesv = dbd_st_FETCH(sth, keysv); |> if (!valuesv) |> valuesv = DBIS->get_attr(sth, keysv); |> ST(0) = valuesv; /* dbd_st_FETCH did sv_2mortal */ |> |> The hand-off to DBIS->get_attr is a recent addition. It's DBIS->get_attr |> that knows about NUM_* and so the hand-off is required (unless your own |> dbd_st_FETCH is willing to supply a value). |> |> You should be able to write your DB2.xs file something like this: |> |> sed -e 's/Oracle/DB2/g' < ../Oracle/Oracle.xs > DB2.xs |> |> I strongly recommend this approach since it'll make it *much* easier |> to keep up with interface changes. At this point, the only things I'm changing are the guts of dbdimp.[c,h], and a few Oracle references in Oracle/DB2.pm, nothing has changed in any DBI file or any *.xs file. Well, then it looks like I have some form of linking/interface problem as FETCH never gets called when I envoke DBIc_NUM_PARAMS(imp_sth). In fact, it behaves as if it were an assignment macro -- with out a function call. Are you sure that invoking DBIc_NUM_PARAMS(imp_sth) should cause FETCH(sth,keysv) to be called? From DBIXS.h: #define _imp2com(p,f) ((p)->com.f) #define DBIc_NUM_PARAMS(imp) _imp2com(imp, num_params) Implies to me that only an assigment is taking place, not a function call. |> > |> > P.S. Where are you announcing the release of newer versions of DBI/DBD? |> > I have not seen any announcements in a while. Yet, everytime I |> > check ftp.demon.co.uk there is a new version. |> > |> They've all been announced to perldb-interest. Are you still getting |> perldb-interest messages? I'll send copies of the announcements to you. |> Well, I was subscribed and I have not asked to be unsubscribed. Maybe I'll try resubscribing. Mike Moran mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-50@lion>; Thu, 31 Aug 1995 07:05:55 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 00:53:23 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ae02915; 31 Aug 95 1:52 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <12705-0@frigate.doc.ic.ac.uk>; Wed, 30 Aug 1995 22:04:34 +0100 Received: by gw.home.vix.com id AA21764; Wed, 30 Aug 95 05:17:50 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA21759; Wed, 30 Aug 95 05:17:48 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id HAA66600; Wed, 30 Aug 1995 07:17:46 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for mhm@austin.ibm.com at austin.ibm.com; id AA17694; Wed, 30 Aug 1995 07:17:44 -0500 From: mhm@austin.ibm.com Message-Id: <9508301217.AA17694@dax.austin.ibm.com> To: Tim Bunce Cc: mhm@austin.ibm.com, perldb-interest@vix.com Subject: Re: DB2perl is minimally working..... In-Reply-To: Your message of Tue, 29 Aug 95 12:43:10 +0100. <9508291708.ab26976@post.demon.co.uk> Date: Wed, 30 Aug 95 07:17:43 -0500 Tim Bunce writes: |> > To be precise, it works as long as the SQL does not contain |> > parameters. |> > Hence, the foundation for my question, I am trying to use the |> > DBIc_NUM_FIELDS(sth) macro without success. |> > |> > >From within my version of int dbd_st_prepare(SV *sth, char |> > *statement) |> > I think I am calling DBIc_NUM_FIELDS(x) with the wrong argument, while |> > the num_fields member of the dbih_stc_t structure gets set, it is only |> > set in the local instance of the structure. |> > |> > Currently, the code does the following |> > |> Umm, the comments above say 'DBIc_NUM_FIELDS', the code below says |> 'DBIc_NUM_PARAMS'. Well, actually either one. I tried both DBIc_NUM_FIELDS and DBIc_NUM_PARAMS, both show the same behavior. Too much cut & paste while my wife was attempting to talk to me...... |> > |> > Yet, after I return from said function the value of |> > $sth->{'NUMPARAMS'} |> > is undefined. Am I missing something intuitively obvious? |> > |> I presume $sth->{'NUMPARAMS'} should be $sth->{'NUM_OF_PARAMS'} since |> 'NUM_OF_PARAMS' is the attribute name that the DBI recognises. Hmmmmm, could have sworn my copy of DBI.pm said "NUMPARAMS". I did just try it with the correct attribute "NUM_OF_PARAMS" without success. It gets set locally but not globally. At this point, I'm assuming that I am calling DBIc_NUM_PARAMS with the wrong argument, but so far I don't have a clue as to the correct argument. dbd_st_prepare(SV *sth, char *statement, SV *attribs) { D_imp_sth(sth); D_imp_dbh_from_sth; .... DBIc_NUM_PARAMS(imp_sth) = params; } where imp_sth is derived from the incomming sth, from my current understanding is correct......but from imperical evidence, it is not correct. |> I'd appreciate a snapshot of your work-so-far Mike at some point Mike. |> Given the closeness of the DB2 interface and SQLCLI/ODBC It'll be |> interesting for me. Aside from figuring out what you are doing in the DBI/DBD::Oracle, my second largest problem is getting through all red tape and bureaucratic B*llSh*t to be able to release this on the net. Unfortunately, at this time I can't release it under penalty of my job. On the brightside, we have convinced at least one lawyer that it should be done :-) Mike P.S. Where are you announcing the release of newer versions of DBI/DBD? I have not seen any announcements in a while. Yet, everytime I check ftp.demon.co.uk there is a new version. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20080-20@lion>; Thu, 31 Aug 1995 07:07:14 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 01:21:15 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa14221; 31 Aug 95 2:20 +0100 Received: by gw.home.vix.com id AA23600; Wed, 30 Aug 95 12:03:15 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA23595; Wed, 30 Aug 95 12:03:14 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id OAA56749; Wed, 30 Aug 1995 14:03:12 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA17452; Wed, 30 Aug 1995 14:03:06 -0500 From: mhm@austin.ibm.com Message-Id: <9508301903.AA17452@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@vix.com Subject: Any problems with DBI::Dispatch Date: Wed, 30 Aug 95 14:03:06 -0500 Next problem an error message from the dispatch routine in DBI DBI DBI::db=HASH(0x200a9ef0): invalid number of parameters: handle + -1 Usage: $h->DBI::db=HASH(0x200a9ef0)() at builddb line 32. which comes from DBI.xs: if (ima->flags & IMA_HAS_USAGE) { char *err = NULL; char msg[200]; if (ima->minargs && (items < ima->minargs || (ima->maxargs>0 && items > ima->maxargs))) { /* the error reporting is a little tacky here */ sprintf(msg, "DBI %s: invalid number of parameters: handle + %ld\n", meth_name, items-1); err = msg; } /* arg type checking could be added here later */ if (err) { croak("%sUsage: %s->%s(%s)", err, "$h", meth_name, (ima->usage) ? ima->usage : "...?"); } } In the above where is items comming from? In my current case, the call which generates this error is $dbh->commit(); which is currently not being dispatched to dbd_db_commit(dbh); Thanks. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-59@lion>; Thu, 31 Aug 1995 07:08:05 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 01:58:26 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa29214; 31 Aug 95 2:57 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab24238; 31 Aug 95 0:00 GMT-60:00 Received: by gw.home.vix.com id AA03113; Wed, 30 Aug 95 07:59:39 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03108; Wed, 30 Aug 95 07:59:38 -0700 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id JAA11993; Wed, 30 Aug 1995 09:59:37 -0500 Date: Wed, 30 Aug 1995 09:59:37 -0500 (CDT) From: Ben A Lindstrom Cc: perldb-interest@vix.com Subject: New to DBI. In-Reply-To: <9508301308.ah27772@post.demon.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I'm new to DBI and I'm just in the process of setting up my machine to do developement of a QuickBase perl module. My question is how stable/finished is v06 of the specs? And how far along is Oracle? If DBI is not far enought along I might have to write my own interface for the meantime. BTW, I read part of the specs and I seem to be missing something...Is there a developer's spec for implementing the database drivers. Thanks. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19286-61@lion>; Thu, 31 Aug 1995 07:08:17 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 02:13:59 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa05452; 31 Aug 95 3:13 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab23832; 30 Aug 95 23:58 GMT-60:00 Received: by gw.home.vix.com id AA01375; Wed, 30 Aug 95 07:33:53 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01370; Wed, 30 Aug 95 07:33:49 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa24265; 30 Aug 95 15:10 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac17339; 30 Aug 95 15:08 +0100 Received: from ig.co.uk by lion id <09024-0@lion>; Wed, 30 Aug 1995 14:44:29 +0100 To: mhm@austin.ibm.com Subject: Re: DB2perl is minimally working..... Cc: perldb-interest@vix.com Date: Wed, 30 Aug 1995 14:44:29 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508301508.ac17339@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > To be precise, it works as long as the SQL does not contain > |> > parameters. > |> > Hence, the foundation for my question, I am trying to use the > |> > DBIc_NUM_FIELDS(sth) macro without success. > |> > > |> > >From within my version of int dbd_st_prepare(SV *sth, char > |> > *statement) > |> > I think I am calling DBIc_NUM_FIELDS(x) with the wrong argument, while > |> > the num_fields member of the dbih_stc_t structure gets set, it is only > |> > set in the local instance of the structure. > |> > > |> > Currently, the code does the following > |> > > |> Umm, the comments above say 'DBIc_NUM_FIELDS', the code below says > |> 'DBIc_NUM_PARAMS'. > > Well, actually either one. I tried both DBIc_NUM_FIELDS and > DBIc_NUM_PARAMS, both show the same behavior. Too much cut & paste while > my wife was attempting to talk to me...... > :-) > |> > > |> > Yet, after I return from said function the value of > |> > $sth->{'NUMPARAMS'} > |> > is undefined. Am I missing something intuitively obvious? > |> > > |> I presume $sth->{'NUMPARAMS'} should be $sth->{'NUM_OF_PARAMS'} since > |> 'NUM_OF_PARAMS' is the attribute name that the DBI recognises. > > Hmmmmm, could have sworn my copy of DBI.pm said "NUMPARAMS". Don't trust the text at the end of DBI.pm. It's not been updated much. Trust the code. > I did just try it with the correct attribute "NUM_OF_PARAMS" without > success. It gets set locally but not globally. > Does your .xs code for FETCH look like this: void FETCH(sth, keysv) SV * sth SV * keysv CODE: SV *valuesv = dbd_st_FETCH(sth, keysv); if (!valuesv) valuesv = DBIS->get_attr(sth, keysv); ST(0) = valuesv; /* dbd_st_FETCH did sv_2mortal */ The hand-off to DBIS->get_attr is a recent addition. It's DBIS->get_attr that knows about NUM_* and so the hand-off is required (unless your own dbd_st_FETCH is willing to supply a value). You should be able to write your DB2.xs file something like this: sed -e 's/Oracle/DB2/g' < ../Oracle/Oracle.xs > DB2.xs I strongly recommend this approach since it'll make it *much* easier to keep up with interface changes. > At this point, I'm assuming that I am calling DBIc_NUM_PARAMS with the > wrong argument, but so far I don't have a clue as to the correct argument. > > dbd_st_prepare(SV *sth, char *statement, SV *attribs) > { > D_imp_sth(sth); > D_imp_dbh_from_sth; > .... > DBIc_NUM_PARAMS(imp_sth) = params; > } > > where imp_sth is derived from the incomming sth, from my current > understanding is correct......but from imperical evidence, it is not > correct. > It looks just fine to me. > |> I'd appreciate a snapshot of your work-so-far Mike at some point Mike. > |> Given the closeness of the DB2 interface and SQLCLI/ODBC It'll be > |> interesting for me. > > Aside from figuring out what you are doing in the DBI/DBD::Oracle, my > second largest problem is getting through all red tape and bureaucratic > B*llSh*t to be able to release this on the net. Unfortunately, at this > time I can't release it under penalty of my job. On the brightside, we > have convinced at least one lawyer that it should be done :-) > Sigh. > Mike > > P.S. Where are you announcing the release of newer versions of DBI/DBD? > I have not seen any announcements in a while. Yet, everytime I > check ftp.demon.co.uk there is a new version. > They've all been announced to perldb-interest. Are you still getting perldb-interest messages? I'll send copies of the announcements to you. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21380-2@lion>; Thu, 31 Aug 1995 07:10:37 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 03:39:46 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa05350; 31 Aug 95 4:39 +0100 Received: by gw.home.vix.com id AA09069; Wed, 30 Aug 95 15:05:11 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA09064; Wed, 30 Aug 95 15:05:09 -0700 Received: by disperse.demon.co.uk id ak02955; 30 Aug 95 21:53 +0100 Received: from post.demon.co.uk by disperse.demon.co.uk id ab14060; 30 Aug 95 18:13 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab20582; 30 Aug 95 18:11 +0100 Received: from ig.co.uk by lion id <13599-0@lion>; Wed, 30 Aug 1995 18:06:17 +0100 To: mhm@austin.ibm.com Subject: Re: DB2perl is minimally working..... Cc: perldb-interest@vix.com Date: Wed, 30 Aug 1995 18:06:17 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508301811.ab20582@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > |> 'NUM_OF_PARAMS' is the attribute name that the DBI recognises. > |> > > |> > Hmmmmm, could have sworn my copy of DBI.pm said "NUMPARAMS". > |> > |> Don't trust the text at the end of DBI.pm. It's not been updated much. > |> Trust the code. > > Reminder posted in painfully obvious place on monitor :-) > :-) > |> > I did just try it with the correct attribute "NUM_OF_PARAMS" without > |> > success. It gets set locally but not globally. > |> > > |> Does your .xs code for FETCH look like this: > |> > |> void > |> FETCH(sth, keysv) > |> SV * sth > |> SV * keysv > |> CODE: > |> SV *valuesv = dbd_st_FETCH(sth, keysv); > |> if (!valuesv) > |> valuesv = DBIS->get_attr(sth, keysv); > |> ST(0) = valuesv; /* dbd_st_FETCH did sv_2mortal */ > |> > |> The hand-off to DBIS->get_attr is a recent addition. It's DBIS->get_attr > |> that knows about NUM_* and so the hand-off is required (unless your own > |> dbd_st_FETCH is willing to supply a value). > |> > |> You should be able to write your DB2.xs file something like this: > |> > |> sed -e 's/Oracle/DB2/g' < ../Oracle/Oracle.xs > DB2.xs > |> > |> I strongly recommend this approach since it'll make it *much* easier > |> to keep up with interface changes. > > At this point, the only things I'm changing are the guts of dbdimp.[c,h], > and a few Oracle references in Oracle/DB2.pm, nothing has changed in any > DBI file or any *.xs file. > Great. > Well, then it looks like I have some form of linking/interface problem > as FETCH never gets called when I envoke DBIc_NUM_PARAMS(imp_sth). In > fact, it behaves as if it were an assignment macro -- with out a function > call. > Er, FETCH gets called when you do $sth->{NUM_OF_PARAMS}. See below... > Are you sure that invoking DBIc_NUM_PARAMS(imp_sth) should cause > FETCH(sth,keysv) to be called? > Er, wrong way round. Look at the code and text above, FETCH calls dbd_st_FETCH and if that returns NULL it then calls DBIS->get_attr. DBIS->get_attr (in the DBI) recognises the NUM_* attribute name and calls DBIc_NUM_PARAMS(imp_sth) for you. > |> > P.S. Where are you announcing the release of newer versions of DBI/DBD? > |> > I have not seen any announcements in a while. Yet, everytime I > |> > check ftp.demon.co.uk there is a new version. > |> > > |> They've all been announced to perldb-interest. Are you still getting > |> perldb-interest messages? I'll send copies of the announcements to you. > > Well, I was subscribed and I have not asked to be unsubscribed. Maybe > I'll try resubscribing. > > Mike Moran > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <03959-0@lion>; Thu, 31 Aug 1995 19:18:38 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 18:17:42 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa24007; 31 Aug 95 19:16 +0100 Received: by gw.home.vix.com id AA07895; Wed, 30 Aug 95 22:07:19 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07890; Wed, 30 Aug 95 22:07:17 -0700 Received: by alta.COM (4.1/SMI-4.1+TAS1.10) id AA19336; Wed, 30 Aug 95 22:07:06 PDT From: Damian Martinez Message-Id: <9508310507.AA19336@alta.COM> Subject: Oracle v0.25 problems on SunOS To: perldb-interest@vix.com Date: Wed, 30 Aug 1995 22:07:05 -0700 (PDT) Cc: Tim.Bunce@ig.co.uk Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 9170 Hey all, I downloaded the latest versions of the DBI (0.62) and the Oracle DBD (0.25) and tried to make them both on a sun... The DBI made fine, tested fine, but the Oracle module had compilation problems... I successfully built both of these on a different machine (sun4m with a newer gcc: 2.6.0) **************************************************** **************************************************** INFORMATION: uname -a: SunOS basalt 4.1.3_U1 4 sun4c gcc --version: 2.5.8 oracle version: (how do I find this? The cdrom says 7.1.4 Production...) If there are newer sources/patches etc, how do I get them? Someone else purchased/installed the Oracle stuff, so I'm not sure what the procedure is for getting patches... **************************************************** perl Makefile.PL: | | Configuring DBD::Oracle version 0.25... | Remember to actually read the README file! | | $ORACLE_HOME not defined. Searching for Oracle... | Using Oracle in /usr/oracle | Oracle sysliblist: -lm | Using /usr/oracle/proc/lib/proc.mk version 1.16 | System: perl5.001 sunos basalt 4.1.3_u1 4 sun4c sun4-sunos dl_dlopen.xs | Compiler: gcc -O -DDEBUGGING | Oracle proc.mk would have used these values but we override them: | CC: /usr/5bin/cc | CFLAGS: -I. -O | LDFLAGS: -Bdynamic -L$(LIBHOME) | | Checking if your kit is complete... | Looks good | Writing Makefile for DBD::Oracle **************************************************** make: (this is as far as it got...) | mkdir ./blib | mkdir ./blib/DBD | mkdir ./blib/auto | mkdir ./blib/auto/DBD | mkdir ./blib/auto/DBD/Oracle | /usr/local/bin/perl -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c | gcc -c -DDEBUGGING -O -fpic -I/usr/local/lib/perl5/sun4-sunos/CORE -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion -Wtraditional -Wpointer-arith -Wbad-function-cast -Wcast-qual -I/usr/oracle/rdbms/demo -I/usr/local/lib/perl5/sun4-sunos/DBI Oracle.c | In file included from /usr/local/lib/perl5/sun4-sunos/CORE/perl.h:238, | from /usr/local/lib/perl5/sun4-sunos/DBI/DBIXS.h:13, | from Oracle.h:14, | from Oracle.c:19: | /usr/include/sys/time.h:64: warning: text following `#endif' violates ANSI standard | cc1: Invalid option `-Wbad-function-cast' | In file included from Oracle.h:43, | from Oracle.c:19: | dbdimp.h:20: redefinition of `imp_fbh_t' | /usr/local/lib/perl5/sun4-sunos/DBI/DBIXS.h:122: `imp_fbh_t' previously declared here | dbdimp.h:23: parse error before `dbih_drc_t' | dbdimp.h:23: warning: no semicolon at end of struct or union | dbdimp.h:24: warning: ANSI C does not allow extra `;' outside of a function | dbdimp.h:28: parse error before `dbih_dbc_t' | dbdimp.h:28: warning: no semicolon at end of struct or union | dbdimp.h:32: parse error before `}' | dbdimp.h:32: warning: ANSI C does not allow extra `;' outside of a function | dbdimp.h:37: parse error before `dbih_stc_t' | dbdimp.h:37: warning: no semicolon at end of struct or union | dbdimp.h:52: parse error before `}' | dbdimp.h:52: warning: ANSI C does not allow extra `;' outside of a function | Oracle.c: In function `XS_DBD__Oracle_errstr': | Oracle.c:38: warning: implicit declaration of function `D_imp_xxh' | Oracle.c:40: warning: implicit declaration of function `DBIc_ERRSTR' | Oracle.c:40: `imp_xxh' undeclared (first use this function) | Oracle.c:40: (Each undeclared identifier is reported only once | Oracle.c:40: for each function it appears in.) | Oracle.c:40: warning: passing arg 1 of `sv_mortalcopy' makes pointer from integer without a cast | Oracle.c: In function `XS_DBD__Oracle__dr_disconnect_all': | Oracle.c:56: `dbihcom' undeclared (first use this function) | Oracle.c:58: warning: implicit declaration of function `DBIc_ERR' | Oracle.c:58: warning: passing arg 1 of `sv_setiv' makes pointer from integer without a cast | Oracle.c:60: warning: function `DBIc_ERRSTR' was previously declared within a block | Oracle.c:62: warning: passing arg 1 of `sv_setpv' makes pointer from integer without a cast | Oracle.c:66: warning: passing arg 4 makes pointer from integer without a cast | Oracle.c:66: warning: passing arg 5 makes pointer from integer without a cast | Oracle.c: In function `XS_DBD__Oracle__db_STORE': | Oracle.c:162: structure has no member named `set_attr' | Oracle.c: In function `XS_DBD__Oracle__db_FETCH': | Oracle.c:183: structure has no member named `get_attr' | Oracle.c: In function `XS_DBD__Oracle__db_disconnect': | Oracle.c:199: `dbihcom' undeclared (first use this function) | Oracle.c:201: warning: implicit declaration of function `DBIc_ACTIVE' | Oracle.c:203: warning: implicit declaration of function `DBIc_WARN' | Oracle.c:217: warning: implicit declaration of function `DBIc_KIDS' | Oracle.c:217: warning: function `DBIc_WARN' was previously declared within a block | Oracle.c: In function `XS_DBD__Oracle__db_DESTROY': | Oracle.c:239: `dbihcom' undeclared (first use this function) | Oracle.c:243: warning: implicit declaration of function `DBIc_IMPSET' | Oracle.c:245: warning: function `DBIc_WARN' was previously declared within a block | Oracle.c:255: warning: function `DBIc_ACTIVE' was previously declared within a block | Oracle.c:257: warning: function `DBIc_WARN' was previously declared within a block | Oracle.c: In function `XS_DBD__Oracle__st__prepare': | Oracle.c:287: warning: implicit declaration of function `DBD_ATTRIBS_CHECK' | Oracle.c: In function `XS_DBD__Oracle__st_bind_param': | Oracle.c:326: warning: function `DBD_ATTRIBS_CHECK' was previously declared within a block | Oracle.c: In function `XS_DBD__Oracle__st_execute': | Oracle.c:342: `dbihcom' undeclared (first use this function) | Oracle.c:352: warning: implicit declaration of function `DBIc_NUM_PARAMS' | Oracle.c: In function `XS_DBD__Oracle__st_fetchrow': | Oracle.c:411: `dbihcom' undeclared (first use this function) | Oracle.c:415: warning: implicit declaration of function `DBIc_COMPAT' | Oracle.c:421: warning: implicit declaration of function `DBIc_NUM_FIELDS' | Oracle.c:437: warning: cast increases required alignment of target type | Oracle.c: In function `XS_DBD__Oracle__st_STORE': | Oracle.c:501: structure has no member named `set_attr' | Oracle.c: In function `XS_DBD__Oracle__st_FETCH': | Oracle.c:522: structure has no member named `get_attr' | Oracle.c: In function `XS_DBD__Oracle__st_finish': | Oracle.c:538: `dbihcom' undeclared (first use this function) | Oracle.c:540: `D_imp_dbh_from_sth' undeclared (first use this function) | Oracle.c:542: warning: function `DBIc_ACTIVE' was previously declared within a block | Oracle.c:542: `imp_dbh' undeclared (first use this function) | Oracle.c: In function `XS_DBD__Oracle__st_DESTROY': | Oracle.c:576: `dbihcom' undeclared (first use this function) | Oracle.c:580: warning: function `DBIc_IMPSET' was previously declared within a block | Oracle.c:582: warning: function `DBIc_WARN' was previously declared within a block | Oracle.c:592: warning: function `DBIc_ACTIVE' was previously declared within a block | Oracle.c:594: warning: function `DBIc_WARN' was previously declared within a block | Oracle.c: In function `boot_DBD__Oracle': | Oracle.c:645: warning: implicit declaration of function `DBI_IMP_SIZE' | Oracle.c:645: sizeof applied to an incomplete type | Oracle.c:647: sizeof applied to an incomplete type | Oracle.c:649: sizeof applied to an incomplete type | *** Error code 1 | make: Fatal error: Command failed for target `Oracle.o' | **************************************************** perl -v | This is perl, version 5.001 | | Unofficial patchlevel 1m. **************************************************** perl -e 'use Config; print Config::myconfig()' | Summary of my perl5 (patchlevel 1) configuration: | Platform: | osname=sunos, osver=4.1.3_u1, archname=sun4-sunos | uname='sunos basalt 4.1.3_u1 4 sun4c ' | hint=recommended | Compiler: | cc='gcc', optimize='-O' | cppflags='-DDEBUGGING' | ccflags ='-DDEBUGGING' | ldflags ='' | stdchar='unsigned char', d_stdstdio=define, usevfork=true | voidflags=15, castflags=0, d_casti32=define, d_castneg=define | intsize=4, alignbytes=8, usemymalloc=y, randbits=31 | Libraries: | so=so | libpth=/lib /usr/lib /usr/ucblib /usr/local/lib | libs=-ldbm -ldl -lm -lc -lposix | libc=/usr/lib/libc.so.1.9.1 | Dynamic Linking: | dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= | cccdlflags='-fpic', ccdlflags=' ', lddlflags='-assert nodefinitions' | **************************************************** On previous versions, I took out the -Wbad-function-cast compiler option and it compiled okay, but when I did that for this one, the only difference was that it didn't complain about the unknown option (!) Like I said, Both the DBI and Oracle DBD compiled fine on another machine and I don't think we have the resources to install a newer version of gcc...can anyone help me tweak these files into compiling? Thanks, Damian Martinez ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07012-5@lion>; Fri, 1 Sep 1995 07:23:54 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 22:57:55 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa25117; 31 Aug 95 23:57 +0100 Received: by gw.home.vix.com id AA01640; Thu, 31 Aug 95 03:34:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01635; Thu, 31 Aug 95 03:34:24 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab18130; 31 Aug 95 11:09 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ae26149; 31 Aug 95 11:07 +0100 Received: from ig.co.uk by lion id <24299-0@lion>; Thu, 31 Aug 1995 11:08:11 +0100 To: perldb-interest@vix.com, z50816@mip.lasc.lockheed.com Subject: Re: _install_me.al Date: Thu, 31 Aug 1995 11:08:11 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508311107.ae26149@post.demon.co.uk> > From: Terry Greenlaw > > > The real question is why did perl not see the _install_method in DBI.xs? > > > Make sure that DBI.c, DBI.o and DBI.so get built and DBI.so gets copied into > > ./blib. If so then try again with PERL_DL_DEBUG=2 and check that the DynaLoader > > is loading the right .so file. > > > Has anyone else had this problem? > > I have DBI.sl and DBI.xs under ./blib/auto/DBI. > SO and DLEXT are both set to sl > > LD_RUN_PATH="" ls -o ./blib/auto/DBI/DBI.sl -b DBI.o > looks a little suspicious, though > Why? > The STDERR output from make test PERL_DL_DEBUG=2: > > DynaLoader.pm loaded (/lib/pa1.1 /lib /usr/lib /usr/local/lib) > Can't locate auto/DBI/_install_me.al in @INC at blib/DBI.pm line 100 Why doesn't the DynaLoader log show DynaLoader::bootstrap(DBI) Umm, perhaps you've got an old DBI statically linked into your perl. Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <07015-8@lion>; Fri, 1 Sep 1995 07:25:18 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 22:59:39 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa09346; 31 Aug 95 23:58 +0100 Received: by gw.home.vix.com id AA03676; Thu, 31 Aug 95 11:01:38 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03671; Thu, 31 Aug 95 11:01:36 -0700 Received: from cst.com (sao486.cst.com [198.179.25.2]) by svcs1.digex.net (8.6.12/8.6.12) with SMTP id OAA17910 for ; Thu, 31 Aug 1995 14:01:29 -0400 Received: by sao486.cst.com id aa23287; 31 Aug 95 14:01 EDT From: "Steve A. Olson" To: perldb-interest@vix.com Subject: Help w/ Oracle Makefiles X-Mailer: ScoMail 1.0 Date: Thu, 31 Aug 1995 12:49:15 -0400 (EDT) Message-Id: <9508311249.aa23044@sao486.cst.com> I've been lurking on this list for some time now, and have a somewhat tangential question related to Oracle OCI and relinking an executable on different Operating Systems and Oracle Versions. Here's my goal: Generate a "makefile" (using perl or whatever) that links one C-Source code file which contains Oracle OCI calls and additional object and header files into an executable on, ideally, any platform (Hardware/OS combination) for any Oracle Version. The makefile should be able to link the executable without user modification of the makefile. The main issues are OS portability and Oracle Versions. I'm looking for ideas, opinions, and experiences. I'll post a follup summary if I get more that a couple of responses. Thanks! ---A-I-T--------------------------------- Applied Information Technologies, Inc. Steve A. Olson Vice President 410-917-4171 (Cellular) 410-796-5862 (FAX/VoiceMail) 410-379-5287 (HomeOffice) Steve.Olson@ait.cst.com (Internet) ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <07013-8@lion>; Fri, 1 Sep 1995 07:25:24 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 31 Aug 95 22:59:46 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa09494; 31 Aug 95 23:59 +0100 Received: by gw.home.vix.com id AA01618; Thu, 31 Aug 95 03:33:16 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01613; Thu, 31 Aug 95 03:33:15 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa18108; 31 Aug 95 11:09 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ad26149; 31 Aug 95 11:06 +0100 Received: from ig.co.uk by lion id <23727-0@lion>; Thu, 31 Aug 1995 10:59:48 +0100 To: perldb-interest@vix.com, krak@alta.com Mmdf-Warning: Unable to confirm address in preceding line at disperse.demon.co.uk Subject: Re: Oracle v0.25 problems on SunOS Date: Thu, 31 Aug 1995 10:59:48 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508311106.ad26149@post.demon.co.uk> > From: Damian Martinez > > Hey all, > > I downloaded the latest versions of the DBI (0.62) and the > Oracle DBD (0.25) and tried to make them both on a sun... > > The DBI made fine, tested fine, but the Oracle module > had compilation problems... > I successfully built both of these on a different machine > (sun4m with a newer gcc: 2.6.0) > > gcc --version: 2.5.8 GCC 2.5.8 is ancient! I recommend GCC 2.6.3 as a very stable version or 2.7.x if you're feeling adventurous. > | dbdimp.h:20: redefinition of `imp_fbh_t' > | /usr/local/lib/perl5/sun4-sunos/DBI/DBIXS.h:122: `imp_fbh_t' previously declared here This looks like you've not actually installed the DBI since 0.62 DBIXS.h doesn't have _anything_ to do with imp_fbh_t in it. > Thanks, > > Damian Martinez > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <07012-9@lion>; Fri, 1 Sep 1995 07:25:42 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 00:42:56 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa18789; 1 Sep 95 1:42 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <04194-0@frigate.doc.ic.ac.uk>; Fri, 1 Sep 1995 00:50:06 +0100 Received: by gw.home.vix.com id AA09782; Thu, 31 Aug 95 12:10:30 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA09663; Thu, 31 Aug 95 12:10:19 -0700 Received: from yquem.inferentia.it (yquem.inferentia.it [151.99.220.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id VAA22443; Thu, 31 Aug 1995 21:09:04 +0200 Posted-Date: Thu, 31 Aug 1995 21:09:04 +0200 Received: by yquem.inferentia.it with Microsoft Mail id <01BA77E7.958B6040@yquem.inferentia.it>; Thu, 31 Aug 1995 21:20:23 +-200 Message-Id: <01BA77E7.958B6040@yquem.inferentia.it> From: Davide Migliavacca To: 'Tim Bunce' Cc: "'perldb-interest@vix.com'" Subject: RE: Error building DBI 0.62 on AIX 3.2.5 Date: Thu, 31 Aug 1995 21:20:21 +-200 Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Well, it seems safemalloc is not that safe; I mean, if "safe" was what = I'd expect. It doesn't initialize memory for sure. Many many thanks to Mike that found the solution to its problem (and so = indirectly for mine also). Isn't it interesting we're both working on = IBM hardware ((:-)) Yes, do memset and everything's OK. The good thing is, I have taken a good look at DBI internals :-) The patch: (Notice, I have yet to understand why, but if you place the memset line = right after the safemalloc, where it belongs, my xlc 1.0.0.38 declares a = syntax error. Beats me - maybe too many macros around) *** DBI.xs Sat Aug 26 19:23:46 1995=0A= --- DBI.xs.new Thu Aug 31 20:56:36 1995=0A= ***************=0A= *** 977,982 ****=0A= CvXSUBANY(cv).any_ptr =3D ima;=0A= =0A= if (atype =3D=3D SVt_PVAV) {=0A= ! ima_usage_av =3D SvRV(attribs);=0A= }=0A= else if (atype =3D=3D SVt_PVHV) {=0A= --- 977,984 ----=0A= CvXSUBANY(cv).any_ptr =3D ima;=0A= =0A= + memset(ima,0,sizeof(*ima));=0A= + =0A= if (atype =3D=3D SVt_PVAV) {=0A= ! ima_usage_av =3D (AV *)SvRV(attribs);=0A= }=0A= else if (atype =3D=3D SVt_PVHV) {=0A= ---------- From: Tim Bunce[SMTP:Tim.Bunce@ig.co.uk] Sent: gioved=EC 31 agosto 1995 12.49 To: Davide.Migliavacca@relay.inferentia.it Subject: RE: Error building DBI 0.62 on AIX 3.2.5 ---------- X-Sun-Data-Type: text X-Sun-Data-Description: text X-Sun-Data-Name: text X-Sun-Content-Lines: 20 > From: Shari > =20 > install_method DBI::db::debug, DBI.pm > method: =20 > DBI::db::debug has ima->(minargs 1, maxargs 2, usage '[$debug_level]', = =20 > flags 1) > install_method DBI::dr::debug, DBI.pm > method: =20 > DBI::dr::debug has ima->(minargs 1, maxargs 2, usage '[$debug_level]', = =20 > flags 6163) Umm. I don't know why that's happening but try this patch for me. Tim. ---------- X-Sun-Data-Type: default X-Sun-Data-Name: xs.patch1 X-Sun-Content-Lines: 110 diff -c2 DBI.xs.old DBI.xs *** DBI.xs.old Thu Aug 31 11:47:21 1995 --- DBI.xs Thu Aug 31 11:45:43 1995 *************** *** 751,754 **** --- 751,757 ---- PUTBACK; --items; + if (!SvPOK(meth_name_sv) || SvNIOK(meth_name_sv)) + croak("%s->%s() invalid redirect method name '%s'", + SvPV(h,na), meth_name, SvPV(meth_name_sv, na)); meth_name =3D SvPV(meth_name_sv, na); } *************** *** 961,969 **** CODE: { ! /* install another method name/interface for the dispatcher */ ! CV *cv =3D newXS(meth_name, XS_DBI_dispatch, file); ! if (dbis->debug >=3D 3) ! fprintf(DBILOGFP,"install_method %s,\t%s\n", meth_name, file); =20 if (strnNE(meth_name, "DBI::", 5)) /* XXX m/^DBI::\w+::\w+$/ */ croak("install_method: invalid name '%s'", meth_name); --- 964,977 ---- CODE: { ! /* install another method name/interface for the DBI dispatcher */ ! int debug =3D (dbis->debug >=3D 3); ! CV *cv; ! SV **svp; ! dbi_ima_t *ima =3D NULL; ! class =3D class; /* avoid 'unused variable' warning */ =20 + if (debug) + fprintf(DBILOGFP,"install_method %s\t", meth_name); +=20 if (strnNE(meth_name, "DBI::", 5)) /* XXX m/^DBI::\w+::\w+$/ */ croak("install_method: invalid name '%s'", meth_name); *************** *** 970,992 **** =20 if (attribs && SvROK(attribs)) { ! /* store method attributes in a fast access form */ ! SV **svp; svtype atype =3D SvTYPE(SvRV(attribs)); ! dbi_ima_t *ima =3D (dbi_ima_t*)safemalloc(sizeof(*ima)); ! AV *ima_usage_av =3D Nullav; ! CvXSUBANY(cv).any_ptr =3D ima; =20 ! if (atype =3D=3D SVt_PVAV) { ! ima_usage_av =3D SvRV(attribs); ! } ! else if (atype =3D=3D SVt_PVHV) { ! DBD_ATTRIB_GET_IV(attribs, "O",1, svp, ima->flags); ! if ( (svp=3DDBD_ATTRIB_GET_SVP(attribs, "U",1)) !=3D NULL) ! ima_usage_av =3D (AV*)SvRV(*svp); ! } ! else croak("install_method %s: bad attribs", meth_name); =20 ! if (ima_usage_av) { ! AV *av =3D ima_usage_av; ima->minargs=3D SvIV(*av_fetch(av, 0, 1)); ima->maxargs=3D SvIV(*av_fetch(av, 1, 1)); --- 978,992 ---- =20 if (attribs && SvROK(attribs)) { ! /* convert and store method attributes in a fast access form */ svtype atype =3D SvTYPE(SvRV(attribs)); ! ima =3D (dbi_ima_t*)safemalloc(sizeof(*ima)); =20 ! if (atype !=3D SVt_PVHV) ! croak("install_method %s: bad attribs", meth_name); !=20 ! DBD_ATTRIB_GET_IV(attribs, "O",1, svp, ima->flags); =20 ! if ( (svp=3DDBD_ATTRIB_GET_SVP(attribs, "U",1)) !=3D NULL) { ! AV *av =3D (AV*)SvRV(*svp); ima->minargs=3D SvIV(*av_fetch(av, 0, 1)); ima->maxargs=3D SvIV(*av_fetch(av, 1, 1)); *************** *** 994,1002 **** ima->usage =3D savepv( (svp) ? SvPV(*svp,na) : ""); ima->flags |=3D IMA_HAS_USAGE; } =20 ! }else if (attribs && SvOK(attribs)) { croak("install_method %s: attributes not a ref", meth_name); } ST(0) =3D &sv_yes; } --- 994,1011 ---- ima->usage =3D savepv( (svp) ? SvPV(*svp,na) : ""); ima->flags |=3D IMA_HAS_USAGE; + if (dbis->debug >=3D 3) + fprintf(DBILOGFP," usage: min %d, max %d, '%s'", + ima->minargs, ima->maxargs, ima->usage); } + if (debug) + fprintf(DBILOGFP,", flags 0x%x", ima->flags); =20 ! } else if (attribs && SvOK(attribs)) { croak("install_method %s: attributes not a ref", meth_name); } + cv =3D newXS(meth_name, XS_DBI_dispatch, file); + CvXSUBANY(cv).any_ptr =3D ima; + if (debug) + fprintf(DBILOGFP,"\n"); ST(0) =3D &sv_yes; } ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09041-0@lion>; Fri, 1 Sep 1995 08:02:23 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 07:00:32 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa28545; 1 Sep 95 7:59 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab17230; 31 Aug 95 19:29 GMT-60:00 Received: by gw.home.vix.com id AA01118; Thu, 31 Aug 95 03:22:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01113; Thu, 31 Aug 95 03:22:46 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa18087; 31 Aug 95 11:09 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa26149; 31 Aug 95 11:06 +0100 Received: from ig.co.uk by lion id <22864-0@lion>; Thu, 31 Aug 1995 10:30:13 +0100 To: mhm@austin.ibm.com Subject: Re: Any problems with DBI::Dispatch Cc: perldb-interest@vix.com Date: Thu, 31 Aug 1995 10:30:13 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508311106.aa26149@post.demon.co.uk> > From: mhm@austin.ibm.com > > Next problem an error message from the dispatch routine in DBI > > DBI DBI::db=HASH(0x200a9ef0): invalid number of parameters: handle + -1 > Usage: $h->DBI::db=HASH(0x200a9ef0)() at builddb line 32. > > In the above where is items comming from? items is defined by the dXSARGS macro at the top of XS_DBI_dispatch. It holds the number of parameters. > In my current case, the call which generates this error is > $dbh->commit(); > which is currently not being dispatched to dbd_db_commit(dbh); Note that the error message shows 'DBI::db=HASH(0x200a9ef0)' as the name of the function and -1 as the number of parameters. This implies that the real problem is still with the if (ima->flags & IMA_FUNC_REDIRECT) test which is before the IMA_HAS_USAGE test. For some reason (ima->flags & IMA_FUNC_REDIRECT) is true when it should not be. Focus your attention on that. > Thanks. > > Mike > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <10775-0@lion>; Fri, 1 Sep 1995 10:55:42 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 07:31:10 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa05356; 1 Sep 95 8:30 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa20061; 31 Aug 95 19:51 GMT-60:00 Received: by gw.home.vix.com id AA00637; Thu, 31 Aug 95 03:15:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00632; Thu, 31 Aug 95 03:15:05 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa18095; 31 Aug 95 11:09 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab26149; 31 Aug 95 11:06 +0100 Received: from ig.co.uk by lion id <23033-0@lion>; Thu, 31 Aug 1995 10:38:14 +0100 To: mouring@netnet.net Subject: Re: New to DBI. Cc: perldb-interest@vix.com Date: Thu, 31 Aug 1995 10:38:14 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508311106.ab26149@post.demon.co.uk> > From: Ben A Lindstrom > > I'm new to DBI and I'm just in the process of setting up my machine to do > developement of a QuickBase perl module. > What is QuickBase? Does it support SQL? > My question is how stable/finished is v06 of the specs? That's actually not a very significant issue right now. Just treat the DBI code as the application interface spec and the DBD::Oracle code as the driver interface spec :-) The notes at the end of DBI.pm are useful. > And how far along is Oracle? > Far enough to be very useful. > If DBI is not far enought along I might have to write my own interface > for the meantime. > Please don't do that. I'll help you with any problems you may have. > BTW, I read part of the specs and I seem to be missing something...Is there > a developer's spec for implementing the database drivers. > Yes, the DBD::Oracle module :-). Basically study the DBD::Oracle .xs file and hack the dbdimp.c file as needed. > Thanks. > Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <10775-1@lion>; Fri, 1 Sep 1995 10:56:00 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 07:33:53 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa06828; 1 Sep 95 8:33 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa20376; 31 Aug 95 19:53 GMT-60:00 Received: by gw.home.vix.com id AA04047; Thu, 31 Aug 95 04:19:10 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04042; Thu, 31 Aug 95 04:19:09 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa18568; 31 Aug 95 11:12 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa26784; 31 Aug 95 11:09 +0100 Received: from ig.co.uk by lion id <24991-0@lion>; Thu, 31 Aug 1995 11:12:16 +0100 To: buzz@bear.com Subject: Re: Happy birthday DBperl! Cc: perldb-interest@vix.com Date: Thu, 31 Aug 1995 11:12:16 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9508311109.aa26784@post.demon.co.uk> > From: Buzz Moschetti > > Tim Bunce writes: > > > > For those too young to remember, the DBperl effort will have its > > > > third birthday on September 29th. It would be kind'a nice to have > > > > something to show for it after three years :-) > > Close. The original mail went out Mon, 28 Sep 92 9:08:31 > :-) Duly noted. (I was actually going by the first 'Happy Birthday!' message you posted on 29 Sep 93 saying 'The DBperl effort is one year old today.' Anyway, what's a day either way between old friends :-) Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <01550-4@lion>; Mon, 4 Sep 1995 07:25:14 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 21:57:26 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa06311; 1 Sep 95 22:56 +0100 Received: by gw.home.vix.com id AA01773; Fri, 1 Sep 95 07:36:15 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01768; Fri, 1 Sep 95 07:36:13 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa19523; 1 Sep 95 14:22 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa06160; 1 Sep 95 14:20 +0100 Received: from ig.co.uk by lion id <14894-0@lion>; Fri, 1 Sep 1995 14:22:49 +0100 To: perldb-interest@vix.com Subject: Announce: DBI-0.63 Date: Fri, 1 Sep 1995 14:22:49 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509011420.aa06160@post.demon.co.uk> Available from the usual place. Current DBI Implementation Notes: $ary_ref = $sth->fetch; currently always returns the same array ref. This may be a feature that stays, then again it may not. $sth->bind_param currently takes a variable value not a reference. That may change later. Changes in DBI 0.63, 1 Sep 1995 Minor update. Fixed uninitialised memory bug in method attribute handling and streamlined processing and debugging. Revised usage definitions for bind_* methods and readblob. Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <01550-5@lion>; Mon, 4 Sep 1995 07:25:34 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 22:16:19 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa17498; 1 Sep 95 23:15 +0100 Received: by gw.home.vix.com id AA00295; Fri, 1 Sep 95 07:12:37 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00290; Fri, 1 Sep 95 07:12:36 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa19301; 1 Sep 95 14:20 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab05531; 1 Sep 95 14:17 +0100 Received: from ig.co.uk by lion id <13617-0@lion>; Fri, 1 Sep 1995 13:12:16 +0100 To: gbalfanz@netcom.com Subject: Re: Perl5 and Informix? Cc: perldb-interest@vix.com, descarte@mcqueen.com Date: Fri, 1 Sep 1995 13:12:16 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509011417.ab05531@post.demon.co.uk> > From: Greg Balfanz > > Tim, > > I am despretly looking for a perl5 informix interface. I assume the > usub isqlperl will only work w/perl4? > Yes. > It does not have to be DBI:: or odbc compatible, good old > propietary informix would suffice. > > Is the perl5 XS conceivable for a moderatly skilled programmer to > interface to informix? > > Any pointers? > Anyone with enough skill to do it has enough to do it as a DBI driver. Alligator Descartes has expressed an interest in doing an Informix interface. He's currently doing the DBD::mSQL driver. Hopefully he could either do it for you or give you a kick start and help you along. I hope you're on the perldb-interest mailing list. If not join straigh away by sending mail to perldb-interest-request@vix.com. Catchup with current work by browsing ftp.demon.co.uk:/pub/perl/db/ in the DBI and DBD/Oracle directories. > Thanks in advance for your time. > > Greg Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01517-10@lion>; Mon, 4 Sep 1995 07:25:58 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 22:03:57 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09387; 1 Sep 95 23:03 +0100 Received: by gw.home.vix.com id AA19732; Fri, 1 Sep 95 11:00:18 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA19709; Fri, 1 Sep 95 11:00:16 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id NAA122511; Fri, 1 Sep 1995 13:00:11 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for davide.migliavacca@relay.inferentia.it at austin.ibm.com; id AA17982; Fri, 1 Sep 1995 13:00:04 -0500 From: mhm@austin.ibm.com Message-Id: <9509011800.AA17982@dax.austin.ibm.com> To: Tim Bunce Cc: davide.migliavacca@relay.inferentia.it, perldb-interest@vix.com Subject: Re: Error building DBI 0.62 on AIX 3.2.5 In-Reply-To: Your message of Fri, 01 Sep 95 10:50:48 +0100. <9509011053.aa26832@post.demon.co.uk> Date: Fri, 01 Sep 95 13:00:03 -0500 Tim Bunce writes: > > > Isn't it interesting we're both working on IBM hardware ((:-)) > > You have my sympathies ;-) Better than working on Sun hardware ;-) > > Yes, do memset and everything's OK. > > I'll probably use Larry's Newz macro. > > > The good thing is, I have taken a good look at DBI internals :-) > > > Great, maybe you can tell me how it works! :-)) Give a few more passes through the DBI in the debugger and maybe I'll be able to answer your questions...... Mike Moran Commercial Performance IBM Austin mhm@austin.ibm.com ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <01550-9@lion>; Mon, 4 Sep 1995 07:26:32 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 02 Sep 95 20:07:55 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa29859; 2 Sep 95 21:07 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa08121; 31 Aug 95 21:58 GMT-60:00 Received: by gw.home.vix.com id AA07768; Thu, 31 Aug 95 05:26:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07762; Thu, 31 Aug 95 05:26:25 -0700 Message-Id: <9508311226.AA07762@gw.home.vix.com> Received: from mip.lasc.lockheed.com by netserve.lasc.lockheed.com with SMTP (1.37.109.4/16.2) id AA16542; Thu, 31 Aug 95 08:26:08 -0400 Received: by mip.lasc.lockheed.com (1.38.193.4/16.2) id AA09452; Thu, 31 Aug 1995 08:26:00 -0400 Date: Thu, 31 Aug 1995 08:26:00 -0400 From: Terry Greenlaw To: perldb-interest@vix.com Subject: Re: _install_me.al > From <@post.demon.co.uk,@ignite.demon.co.uk:Tim.Bunce@ig.co.uk> Wed Aug 30 09:30 EDT 1995 > Received: from marble.litc.lockheed.com by mip.lasc.lockheed.com with SMTP > (1.38.193.4/16.2) id AA17088; Wed, 30 Aug 1995 09:30:23 -0400 > Return-Path: <@post.demon.co.uk,@ignite.demon.co.uk:Tim.Bunce@ig.co.uk> > Received: from disperse.demon.co.uk (disperse.demon.co.uk [158.152.1.77]) by marble.litc.lockheed.com (8.6.12/8.6.9) with SMTP id GAA04937 for ; Wed, 30 Aug 1995 06:34:37 -0700 > Received: from post.demon.co.uk by disperse.demon.co.uk id aa11859; > 30 Aug 95 13:11 +0100 > Received: from ignite.demon.co.uk by post.demon.co.uk id ak27772; > 30 Aug 95 13:08 +0100 > Received: from ig.co.uk by lion id <07828-0@lion>; > Wed, 30 Aug 1995 13:07:15 +0100 > To: z50816@mip.lasc.lockheed.com > Subject: _install_me.al > Cc: perldb-interest@vix.com > Date: Wed, 30 Aug 1995 13:07:15 +0100 > From: Tim Bunce > Sender: Tim.Bunce@ig.co.uk > Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > Message-Id: <9508301308.ak27772@post.demon.co.uk> > Status: RO > > From: Terry Greenlaw > > > > Tim, I'm trying to install DBI 0.62, and get the following message during "make test". > > The "perl Makefile.PL" and "make" went okay. I tried this before and after a > > "make install". > > Shouldn't it be trying _install_method instead of _install_me ? > > Any ideas? > > > > Mip> make test > > /usr/bin/perl -I./blib -I./blib -I/disk3/perl5/lib/perl5/hpux -I/disk3/perl5/lib/perl5 > > -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t > > t/min...............Can't locate auto/DBI/_install_me.al in @INC at blib/DBI.pm line 100 > > FAILED on test 1 > _install_method used to be defined in DBI.pm, it's now defined in DBI.xs. > The auto/DBI/_install_me.al error is simply a symptom of _install_method not > being found and the AutoLoader doing its best to find one (the shortened name > is a last ditch attempt to deal with filesystems which don't support long names). > The real question is why did perl not see the _install_method in DBI.xs? > Make sure that DBI.c, DBI.o and DBI.so get built and DBI.so gets copied into > ./blib. If so then try again with PERL_DL_DEBUG=2 and check that the DynaLoader > is loading the right .so file. > Has anyone else had this problem? > Tim. Tim, You were right about having a static version of DBI already linked into Perl. When I rebuilt and reinstalled perl the DBI build was easy. I built DBI dynamically at first, but a statically built DBD/Oracle doesn't get along well with a dynamic DBI. Building them both statically fixed the problem. There is (supposedly) a patch from HP for ld that will solve the "illegal loader fixup needed" problem, allowing Oracle to finally load dynamically on HP/UX 9. If anyone knows the patch number, could they please post it? Thanks for the info... tog Terry Greenlaw (on-site @ Lockheed) Encompass Technologies z50816@mip.lasc.lockheed.com terry@encompass.is.net ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01517-14@lion>; Mon, 4 Sep 1995 07:26:47 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 01 Sep 95 23:17:27 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa23636; 2 Sep 95 0:17 +0100 Received: by gw.home.vix.com id AA25826; Fri, 1 Sep 95 06:07:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA25821; Fri, 1 Sep 95 06:07:23 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id IAA52731; Fri, 1 Sep 1995 08:07:21 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA23994; Fri, 1 Sep 1995 08:07:18 -0500 From: mhm@austin.ibm.com Message-Id: <9509011307.AA23994@dax.austin.ibm.com> To: Tim.Bounce@ig.co.uk Cc: perldb-interest@vix.com Subject: Building the %attribs for bind_param ? Date: Fri, 01 Sep 95 08:07:18 -0500 The question boils down to what is the best way to build the %attribs hash-ref for bind_param and bind-col. Given the number of possible things that a DB2 user might want to bind to a parameter or in binding the incoming data to a host variable, I think I need to supply a function which builds and returns the %attribs hash-ref. Would it be better to define it as an XSUB routine or in the pm file under DBD::DB2::st? I currently leaning towards doing it in perl in the pm file. Any comments, anything else I need to consider? Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <01550-17@lion>; Mon, 4 Sep 1995 07:29:16 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 17:56:41 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa29087; 3 Sep 95 18:54 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab12399; 31 Aug 95 22:35 GMT-60:00 Received: by gw.home.vix.com id AA15122; Thu, 31 Aug 95 07:10:53 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA15117; Thu, 31 Aug 95 07:10:51 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id JAA34735; Thu, 31 Aug 1995 09:10:49 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for mhm@austin.ibm.com at austin.ibm.com; id AA22494; Thu, 31 Aug 1995 09:10:47 -0500 From: mhm@austin.ibm.com Message-Id: <9508311410.AA22494@dax.austin.ibm.com> To: Tim Bunce Cc: mhm@austin.ibm.com, perldb-interest@vix.com Subject: Re: Any problems with DBI::Dispatch In-Reply-To: Your message of Thu, 31 Aug 95 10:30:13 +0100. <9508311106.aa26149@post.demon.co.uk> Date: Thu, 31 Aug 95 09:10:46 -0500 Tim Bunce writes: |> |> Note that the error message shows 'DBI::db=HASH(0x200a9ef0)' as the |> name of the function and -1 as the number of parameters. This |> implies that the real problem is still with the |> if (ima->flags & IMA_FUNC_REDIRECT) test which is before the |> IMA_HAS_USAGE test. |> |> For some reason (ima->flags & IMA_FUNC_REDIRECT) is true when it |> should not be. |> Focus your attention on that. Found the problem, in DBI.xs - install_method asssumes at time of creation that newly created ima structure will contain all zeros or nulls. In my case, this is not true. On the installation of the "db::commit" method everything in the structure is zero/null except the last field -- flags, so I end up with both IMA_HAS_USAGE and IMA_FUNC_REDIRECT, plus quite a few others we haven't even thought of. :-) Quick solution is to memset the allocated structure or call calloc. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01517-20@lion>; Mon, 4 Sep 1995 07:29:26 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 02 Sep 95 00:53:06 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13533; 2 Sep 95 1:52 +0100 Received: by gw.home.vix.com id AA00284; Fri, 1 Sep 95 07:12:03 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00279; Fri, 1 Sep 95 07:12:02 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa19304; 1 Sep 95 14:19 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac05531; 1 Sep 95 14:17 +0100 Received: from ig.co.uk by lion id <13626-0@lion>; Fri, 1 Sep 1995 13:13:48 +0100 To: mhm@austin.ibm.com Subject: Re: Problem with definition of %DBI_IF {bind_param} Cc: perldb-interest@vix.com Date: Fri, 1 Sep 1995 13:13:48 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509011417.ac05531@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim, > > In DBI.pm, > > my %DBI_IF = ( # Define the DBI Interface: > bind_param => { U =>[2,3,'\\$var [, \%attribs]'] }, > } > > Shouldn't it be > > bind_param => { U =>[3,4,'\\$parameter,\\$var [, \%attribs]'] }, > > Since you must specify which parameter, the variable to use and > possibly the attributes...that makes argmin = 3 and argmax = 4 once > you include the statement handle. > > or am I totally lost and confused? > Nope, you're right. Thanks. Actually $var is curently not a reference. I'm not sure about about that yet. Umm, bind_col and bind_columns are also wrong! > Mike Moran > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <01569-1@lion>; Mon, 4 Sep 1995 07:29:26 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 18:28:10 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa16492; 3 Sep 95 19:26 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa22973; 1 Sep 95 0:07 GMT-60:00 Received: by gw.home.vix.com id AA07675; Thu, 31 Aug 95 11:47:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07670; Thu, 31 Aug 95 11:47:23 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id NAA71630; Thu, 31 Aug 1995 13:47:22 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA20932; Thu, 31 Aug 1995 13:47:21 -0500 From: mhm@austin.ibm.com Message-Id: <9508311847.AA20932@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@vix.com Subject: Problem with definition of %DBI_IF {bind_param} Date: Thu, 31 Aug 95 13:47:20 -0500 Tim, In DBI.pm, my %DBI_IF = ( # Define the DBI Interface: bind_param => { U =>[2,3,'\\$var [, \%attribs]'] }, } Shouldn't it be bind_param => { U =>[3,4,'\\$parameter,\\$var [, \%attribs]'] }, Since you must specify which parameter, the variable to use and possibly the attributes...that makes argmin = 3 and argmax = 4 once you include the statement handle. or am I totally lost and confused? Mike Moran mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01549-13@lion>; Mon, 4 Sep 1995 07:30:13 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 02 Sep 95 01:59:21 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09396; 2 Sep 95 2:58 +0100 Received: by gw.home.vix.com id AA11590; Fri, 1 Sep 95 15:10:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA11585; Fri, 1 Sep 95 15:10:44 -0700 Received: from alma.mcqueen.com (alma.mcqueen.com [193.133.160.249]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with ESMTP id XAA12241; Fri, 1 Sep 1995 23:04:59 +0100 From: descarte@mcqueen.com Received: (from descarte@localhost) by alma.mcqueen.com (8.6.12/8.6.12) id XAA20814; Fri, 1 Sep 1995 23:10:55 +0100 Message-Id: <199509012210.XAA20814@alma.mcqueen.com> Subject: Re: perl5 and informix? To: Greg Balfanz Date: Fri, 1 Sep 1995 23:10:54 +0100 (BST) Cc: perldb-interest@vix.com, Tim.Bunce@ig.co.uk In-Reply-To: <199509011701.KAA11095@netcom20.netcom.com> from "Greg Balfanz" at Sep 1, 95 10:01:07 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 443 > Tim Bounce said you might have a perl5 informix or even DBI? > > I'm trying to find one, any help would be appreciated. Well, it's under rapid development at the moment. I'll have more details on it on Monday or Tuesday, since I'm trying to get about 6 different projects completed all at the same time just now! It's rudimentarily working. I just need to tidy things a little. > Greg -- Alligator Descartes | descarte@mcqueen.com | ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <01516-24@lion>; Mon, 4 Sep 1995 07:30:25 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 20:02:30 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa02785; 3 Sep 95 20:59 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab20396; 1 Sep 95 20:47 GMT-60:00 Received: by gw.home.vix.com id AA25766; Fri, 1 Sep 95 06:05:19 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA25761; Fri, 1 Sep 95 06:05:17 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa06976; 1 Sep 95 12:04 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa06723; 1 Sep 95 12:02 +0100 Received: from ig.co.uk by lion id <12535-0@lion>; Fri, 1 Sep 1995 11:47:06 +0100 To: yp@gr.osf.org Subject: Re: perl5 interface for postgres95 beta 0.003 Cc: perldb-interest@vix.com Date: Fri, 1 Sep 1995 11:47:06 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509011202.aa06723@post.demon.co.uk> > From: Yves Paindaveine > > Tim, > > >Yves, your Pg95 module is not a Perl 5 DBI module compatible driver. > >Have you considered writing a DBD module from Postgres95? > > I thought you were using acronyms (DBI for DB interface) but not > talking about projects. It was so fuzzy about what's going on or still > in activity that I decided to make the simplest. But I may be > interested in writing more portable code, can you possibly explain me > the goals and differences between DBI and DBD (or attach a document)? > Please have a browse through ftp.demon.co.uk:/pub/perl/db/. Especially the DBI directory (grab README, dbispec.v06.Z, DBI-0.62.tar.Z and perldb-interest/perldb-i.v06[gh]) and grab DBD-Oracle-0.25.tar.Z from the DBD/Oracle directory. In overview: Read (the out of date) dbispec.v06 to get the flavor of things. Unpack, build, test and install the DBI module. Unpack the DBD::Oracle module and s/Oracle/Pg95/g in Oracle.xs Hack out the guts of dbdimp.c and replace with pg95 code. This is a gross simplification but should get you going. > >Are you on the perldb-interest mailing list? > > No. I started a week ago with perl and database, along with httpd > management and cgi writing. I went to perl due to cgi... Please send > me the address where to register. Are there other groups/news > which may be of interest? > Send a request to perldb-interest-request@vix.com. If you have any queries before you're subscription gets processed just email your message to the list: perldb-interest@vix.com Good luck. > I've seen in a later mail that you put the package in perl/mod. Do > you know that my previously uploaded package (before you told me not > to do so) is in another place (/perl I think). > Thanks. I'll look into it. > -yp > Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <01550-24@lion>; Mon, 4 Sep 1995 07:31:00 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 22:40:50 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa22374; 3 Sep 95 23:39 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa16271; 2 Sep 95 0:21 GMT-60:00 Received: by gw.home.vix.com id AA20943; Fri, 1 Sep 95 05:00:44 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20938; Fri, 1 Sep 95 05:00:42 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab00700; 1 Sep 95 10:56 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab26832; 1 Sep 95 10:53 +0100 Received: from ig.co.uk by lion id <10691-0@lion>; Fri, 1 Sep 1995 10:52:25 +0100 To: mhm@austin.ibm.com Subject: Re: Any problems with DBI::Dispatch Cc: perldb-interest@vix.com Date: Fri, 1 Sep 1995 10:52:25 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509011053.ab26832@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > |> Note that the error message shows 'DBI::db=HASH(0x200a9ef0)' as the > |> name of the function and -1 as the number of parameters. This > |> implies that the real problem is still with the > |> if (ima->flags & IMA_FUNC_REDIRECT) test which is before the > |> IMA_HAS_USAGE test. > |> > |> For some reason (ima->flags & IMA_FUNC_REDIRECT) is true when it > |> should not be. > |> Focus your attention on that. > > Found the problem, in DBI.xs - install_method asssumes at time of > creation that newly created ima structure will contain all zeros or nulls. > In my case, this is not true. > > On the installation of the "db::commit" method everything in the > structure is zero/null except the last field -- flags, so I end up with > both IMA_HAS_USAGE and IMA_FUNC_REDIRECT, plus quite a few others > we haven't even thought of. :-) > > Quick solution is to memset the allocated structure or call calloc. > Thanks Mike. I'm sorry you've had to spend time finding my bug. Expect a new release shortly. > Mike > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01549-22@lion>; Mon, 4 Sep 1995 07:31:59 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 02 Sep 95 04:01:16 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id am15491; 2 Sep 95 5:00 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <23093-0@frigate.doc.ic.ac.uk>; Fri, 1 Sep 1995 21:16:35 +0100 Received: by gw.home.vix.com id AA20577; Fri, 1 Sep 95 04:59:40 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20572; Fri, 1 Sep 95 04:59:38 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa00661; 1 Sep 95 10:55 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa26832; 1 Sep 95 10:53 +0100 Received: from ig.co.uk by lion id <10664-0@lion>; Fri, 1 Sep 1995 10:50:48 +0100 To: davide.migliavacca@relay.inferentia.it Subject: RE: Error building DBI 0.62 on AIX 3.2.5 Cc: perldb-interest@vix.com Date: Fri, 1 Sep 1995 10:50:48 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509011053.aa26832@post.demon.co.uk> > From: Davide Migliavacca > > Well, it seems safemalloc is not that safe; I mean, if "safe" was what > I'd expect. It doesn't initialize memory for sure. The safe in safemalloc means that it'll never return a NULL pointer. > Many many thanks to Mike that found the solution to its problem (and so > indirectly for mine also). Yes, thanks Mike. I'll check through the code for any similar problems. > Isn't it interesting we're both working on IBM hardware ((:-)) You have my sympathies ;-) > Yes, do memset and everything's OK. I'll probably use Larry's Newz macro. > The good thing is, I have taken a good look at DBI internals :-) > Great, maybe you can tell me how it works! :-)) Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01568-7@lion>; Mon, 4 Sep 1995 07:32:04 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 02 Sep 95 04:01:17 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id an15491; 2 Sep 95 5:00 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <24359-0@frigate.doc.ic.ac.uk>; Fri, 1 Sep 1995 22:17:49 +0100 Received: by gw.home.vix.com id AA00429; Fri, 1 Sep 95 07:16:43 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00421; Fri, 1 Sep 95 07:16:41 -0700 Received: from alma.mcqueen.com (alma.mcqueen.com [193.133.160.249]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with ESMTP id PAA09914; Fri, 1 Sep 1995 15:13:05 +0100 From: descarte@mcqueen.com Received: (from descarte@localhost) by alma.mcqueen.com (8.6.12/8.6.12) id PAA17338; Fri, 1 Sep 1995 15:18:59 +0100 Message-Id: <199509011418.PAA17338@alma.mcqueen.com> Subject: Re: Perl5 and Informix? To: Tim Bunce Date: Fri, 1 Sep 1995 15:18:59 +0100 (BST) Cc: perldb-interest@vix.com In-Reply-To: <9509011417.ab05531@post.demon.co.uk> from "Tim Bunce" at Sep 1, 95 01:12:16 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 784 > > It does not have to be DBI:: or odbc compatible, good old > > propietary informix would suffice. > > > > Is the perl5 XS conceivable for a moderatly skilled programmer to > > interface to informix? > > > > Any pointers? > > > Anyone with enough skill to do it has enough to do it as a DBI driver. > > Alligator Descartes has expressed an interest > in doing an Informix interface. He's currently doing the DBD::mSQL driver. > > Hopefully he could either do it for you or give you a kick start and > help you along. I'm actually on this right this minute. I've got no estimation of the time it'll take, but hopefully there'll be an alpha mid next week as long as I get some more diskspace! > Tim. -- Alligator Descartes | descarte@mcqueen.com | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01549-23@lion>; Mon, 4 Sep 1995 07:32:04 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 02 Sep 95 04:01:17 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id al15491; 2 Sep 95 5:00 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <22916-0@frigate.doc.ic.ac.uk>; Fri, 1 Sep 1995 21:13:12 +0100 Received: by gw.home.vix.com id AA01802; Fri, 1 Sep 95 07:36:55 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01795; Fri, 1 Sep 95 07:36:52 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id JAA124617; Fri, 1 Sep 1995 09:36:50 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for mhm@austin.ibm.com at austin.ibm.com; id AA23992; Fri, 1 Sep 1995 09:35:33 -0500 From: mhm@austin.ibm.com Message-Id: <9509011435.AA23992@dax.austin.ibm.com> To: Tim Bunce Cc: mhm@austin.ibm.com, perldb-interest@vix.com Subject: Re: Problem with definition of %DBI_IF {bind_param} In-Reply-To: Your message of Fri, 01 Sep 95 13:13:48 +0100. <9509011417.ac05531@post.demon.co.uk> Date: Fri, 01 Sep 95 09:35:32 -0500 Tim Bunce writes: > > Umm, bind_col and bind_columns are also wrong! > I hadn't gotten that far....yet, still struggling with DB2 parameter binding. :-( Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01568-19@lion>; Mon, 4 Sep 1995 07:33:49 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 01:49:19 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa29509; 3 Sep 95 2:49 +0100 Received: by gw.home.vix.com id AA10298; Sat, 2 Sep 95 15:31:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10293; Sat, 2 Sep 95 15:31:00 -0700 Received: by deathstar.kersur.net (Linux Smail3.1.29.1 #16) id m0sp15B-0001aVC; Sat, 2 Sep 95 18:30 EDT Message-Id: Date: Sat, 2 Sep 95 18:30 EDT From: root To: perldb-interest@vix.com Subject: DBI build failure cc: Tim.Bunce@ig.co.uk Tried installing and testing DBI on my Linux system. The test suite failed. Enclosed is the debugging information requested: --------------------------------------------------------------------------- deathstar:/usr/src/DBI# perl Makefile.PL Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker NAME => 'DBI' VERSION => '0.62' Using header files found in /usr/local/lib/perl5/i586-linux/CORE Looking for perl 5 by these names: miniperl perl, in these dirs: /usr/local/bin /usr/bin /bin /usr/local/sbin /usr/sbin /sbin /usr/X11R6/bin /usr/X11R5/bin /usr/games /usr/local/adm/bin . /usr/bin Executing /usr/local/bin/perl Writing Makefile for DBI deathstar:/usr/src/DBI# make mkdir /usr/src/DBI/blib mkdir /usr/src/DBI/blib/auto mkdir /usr/src/DBI/blib/auto/DBI /usr/local/bin/perl /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c cc -c -Dbool=char -DHAS_BOOL -O2 -I/usr/local/lib/perl5/i586-linux/CORE DBI.c DBI.c: In function `XS_DBI__install_method': DBI.c:1016: warning: assignment from incompatible pointer type Running mkbootstrap for DBI () Writing DBI.bs ld -o /usr/src/DBI/blib/auto/DBI/DBI.o -r DBI.o cp DBI.bs /usr/src/DBI/blib/auto/DBI/DBI.bs mkdir /usr/src/DBI/blib/DBI cp DBIXS.h /usr/src/DBI/blib/DBI/DBIXS.h mkdir /usr/src/DBI/blib/DBD cp lib/DBD/ExampleP.pm /usr/src/DBI/blib/DBD/ExampleP.pm cp lib/DBD/NullP.pm /usr/src/DBI/blib/DBD/NullP.pm cp DBI.pm /usr/src/DBI/blib/DBI.pm deathstar:/usr/src/DBI# make test /usr/local/bin/perl -I/usr/src/DBI/blib -I/usr/src/DBI/blib -I/usr/local/lib/perl5/i586-linux -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............ok All tests successful. Files=1, Tests=6, 1 secs ( 0.74 cusr 0.17 csys = 0.91 cpu) /usr/local/bin/perl -I/usr/src/DBI/blib -I/usr/src/DBI/blib -I/usr/local/lib/perl5/i586-linux -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ Switch: DBI Switch by Tim Bunce, 1.44 DebugDispatch: 0 Available Drivers: ExampleP, NullP Read DBI special variables (expect 0, 99, 99): err: 0 err: 99 errstr: 99 Attempt to modify DBI special variables. Expect a 'Can't modify' error message: Can't modify $DBI::&rows special variable at test.pl line 59 Driver installed as DBI::dr=HASH(0xe4e88) Connected as DBI::db=HASH(0x102e94) Test error handling: prepare invalid query. Expect an ERROR EVENT message: ERROR EVENT: DBI::db=HASH(0x102e70)->DBI::default_handler: Unknown field names: unknown_field_name at /usr/src/DBI/blib/DBD/ExampleP.pm line 63 prepare returned UNDEF. $DBI::err=0 Cursor prepare'd as DBI::st=HASH(0x102fd8) Prepared as DBI::st=HASH(0x6d0c0) Number of fields: 3 Data type of first field: 1 Driver name: ExampleP Fetching data from both cursors. Expect several rows of data: fetchrow scalar context problem at test.pl line 124. make: *** [test] Error 2 deathstar:/usr/src/DBI# make test TEST_VERBOSE=1 /usr/local/bin/perl -I/usr/src/DBI/blib -I/usr/src/DBI/blib -I/usr/local/lib/perl5/i586-linux -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............1..6 ok 1 ok 2 ok 3 ok 4 ok 4 ok 5 ok 6 ok All tests successful. Files=1, Tests=6, 1 secs ( 0.75 cusr 0.16 csys = 0.91 cpu) /usr/local/bin/perl -I/usr/src/DBI/blib -I/usr/src/DBI/blib -I/usr/local/lib/perl5/i586-linux -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ Switch: DBI Switch by Tim Bunce, 1.44 DebugDispatch: 0 Available Drivers: ExampleP, NullP Read DBI special variables (expect 0, 99, 99): err: 0 err: 99 errstr: 99 Attempt to modify DBI special variables. Expect a 'Can't modify' error message: Can't modify $DBI::&rows special variable at test.pl line 59 Driver installed as DBI::dr=HASH(0xe4e94) Connected as DBI::db=HASH(0x102ea0) Test error handling: prepare invalid query. Expect an ERROR EVENT message: ERROR EVENT: DBI::db=HASH(0x102e7c)->DBI::default_handler: Unknown field names: unknown_field_name at /usr/src/DBI/blib/DBD/ExampleP.pm line 63 prepare returned UNDEF. $DBI::err=0 Cursor prepare'd as DBI::st=HASH(0x102fe4) Prepared as DBI::st=HASH(0x6d0cc) Number of fields: 3 Data type of first field: 1 Driver name: ExampleP Fetching data from both cursors. Expect several rows of data: fetchrow scalar context problem at test.pl line 124. make: *** [test] Error 2 deathstar:/usr/src/DBI# perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=linux, osver=1, archname=i586-linux uname='linux deathstar 1.2.9 #18 tue jul 11 19:35:44 edt 1995 i586 ' hint=recommended Compiler: cc='cc', optimize='-O2' cppflags='-Dbool=char -DHAS_BOOL' ccflags ='-Dbool=char -DHAS_BOOL' ldflags ='' stdchar='char', d_stdstdio=, usevfork=false voidflags=15, castflags=0, d_casti32=, d_castneg=define intsize=4, alignbytes=4, usemymalloc=n, randbits=31 Libraries: so=sa libpth=/lib /usr/lib /usr/local/lib libs=-ldbm -ldld -lm -lc -lbsd libc=/usr/lib/libc.sa Dynamic Linking: dlsrc=dl_dld.xs, dlext=o, d_dlsymun= cccdlflags='', ccdlflags='', lddlflags='-r' deathstar:/usr/src/DBI# ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01618-9@lion>; Mon, 4 Sep 1995 07:35:38 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 18:27:50 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa03046; 3 Sep 95 19:27 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab23337; 1 Sep 95 0:09 GMT-60:00 Received: by gw.home.vix.com id AA00420; Thu, 31 Aug 95 10:23:21 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00415; Thu, 31 Aug 95 10:23:20 -0700 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id MAA29720; Thu, 31 Aug 1995 12:23:18 -0500 Date: Thu, 31 Aug 1995 12:23:18 -0500 (CDT) From: Ben A Lindstrom Cc: perldb-interest@vix.com Subject: Re: New to DBI. In-Reply-To: <9508311106.ab26149@post.demon.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 31 Aug 1995, Tim Bunce wrote: > > > From: Ben A Lindstrom > > > > I'm new to DBI and I'm just in the process of setting up my machine to do > > developement of a QuickBase perl module. > > > What is QuickBase? Does it support SQL? > Yes, QuickBaseis a NeXTStep database written by the people at Object World. Currently it's not a 100% ansi-sql, but close enought. =-) > > If DBI is not far enought along I might have to write my own interface > > for the meantime. > > > Please don't do that. I'll help you with any problems you may have. > I really did not perfer to do this anyways. =-) ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01616-16@lion>; Mon, 4 Sep 1995 07:36:00 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 20:18:06 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa00526; 3 Sep 95 21:17 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab26866; 1 Sep 95 21:40 GMT-60:00 Received: by gw.home.vix.com id AA21088; Fri, 1 Sep 95 05:07:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA21081; Fri, 1 Sep 95 05:07:00 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa01022; 1 Sep 95 10:58 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa27162; 1 Sep 95 10:55 +0100 Received: from ig.co.uk by lion id <10977-0@lion>; Fri, 1 Sep 1995 10:57:12 +0100 To: perldb-interest@vix.com, solson@cst.com Subject: Re: Help w/ Oracle Makefiles Date: Fri, 1 Sep 1995 10:57:12 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509011055.aa27162@post.demon.co.uk> > From: "Steve A. Olson" > > I've been lurking on this list for some time now, and have a somewhat > tangential question related to Oracle OCI and relinking an executable > on different Operating Systems and Oracle Versions. > > Here's my goal: > Generate a "makefile" (using perl or whatever) that links > one C-Source code file which contains Oracle OCI calls and > additional object and header files into an executable on, ideally, > any platform (Hardware/OS combination) for any Oracle Version. > > The makefile should be able to link the executable without user > modification of the makefile. > > The main issues are OS portability and Oracle Versions. > > I'm looking for ideas, opinions, and experiences. I'll post a follup > summary if I get more that a couple of responses. Thanks! > I don't think it's possible to have one makefile for Oracle on many platforms. The best approach seems to be to supply a script which write a makefile using a chunk of the Oracle proc.mk firl for that platform. You're welcome to use the DBD::Oracle module Makefile.PL code for this. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <01618-13@lion>; Mon, 4 Sep 1995 07:36:05 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 03 Sep 95 22:29:45 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa08892; 3 Sep 95 23:29 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab09874; 1 Sep 95 23:25 GMT-60:00 Received: by gw.home.vix.com id AA21802; Fri, 1 Sep 95 11:23:35 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA21797; Fri, 1 Sep 95 11:23:34 -0700 Received: from isye.isye.gatech.edu by gatech.edu with SMTP id AA02162 (5.65c/Gatech-10.0-IDA for ); Fri, 1 Sep 1995 14:26:28 -0400 Received: from gemini.gatech.edu by isye.isye.gatech.edu (4.1/SMI-4.1) id AA00385; Fri, 1 Sep 95 14:23:31 EDT Received: by gemini.gatech.edu (4.1/SMI-4.1) id AA05368; Fri, 1 Sep 95 14:23:29 EDT Date: Fri, 1 Sep 1995 14:23:28 -0400 (EDT) From: Jennifer Anne Turns X-Sender: jennifer@gemini To: perldb-interest@vix.com Subject: Subscribe Request Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII If this is the right place, please subscribe me. SUBSCRIBE Jennifer Turns jennifer@isye.isye.gatech.edu Thanks. ------------------------------------------------------------------------- Jennifer Turns Center for Human-Machine Systems Research jennifer@chmsr.gatech.edu School of Industrial and Systems Engineering voice: (404) 894-4318 Georgia Institute of Technology fax: (404) 894-2301 765 Ferst Drive, Atlanta, GA 30332-0205 "As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality" -- Albert Einstein ------------------------------------------------------------------------- ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-7@lion>; Tue, 5 Sep 1995 07:22:20 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 04 Sep 95 19:46:39 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa24845; 4 Sep 95 20:45 +0100 Received: by gw.home.vix.com id AA03102; Mon, 4 Sep 95 07:25:07 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03097; Mon, 4 Sep 95 07:25:05 -0700 Received: from alma.mcqueen.com (alma.mcqueen.com [193.133.160.249]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with ESMTP id PAA20799; Mon, 4 Sep 1995 15:16:46 +0100 From: descarte@mcqueen.com Received: (from descarte@localhost) by alma.mcqueen.com (8.6.12/8.6.12) id PAA13240; Mon, 4 Sep 1995 15:22:47 +0100 Message-Id: <199509041422.PAA13240@alma.mcqueen.com> Subject: ANNOUNCE: DBD::mSQL-0.60pl1 released To: perldb-interest@vix.com, msql-list@bond.edu.au Date: Mon, 4 Sep 1995 15:22:47 +0100 (BST) Cc: Tim.Bunce@ig.co.uk, k@anna.mind.de X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 186 Available from: ftp://ftp.mcqueen.com/pub/languages/perl5/DBPerl/ Let me know how things go. Many bug fixes. Now supports DELETE! -- Alligator Descartes | descarte@mcqueen.com | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-18@lion>; Tue, 5 Sep 1995 07:25:20 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 04 Sep 95 21:31:19 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa15479; 4 Sep 95 22:30 +0100 Received: by gw.home.vix.com id AA12178; Mon, 4 Sep 95 09:41:51 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA12171; Mon, 4 Sep 95 09:41:48 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa29059; 4 Sep 95 17:20 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa11340; 4 Sep 95 17:18 +0100 Received: from ig.co.uk by lion id <09840-0@lion>; Mon, 4 Sep 1995 14:38:05 +0100 To: gbalfanz@netcom.com, descarte@mcqueen.com Subject: Re: perl5 and informix? Cc: perldb-interest@vix.com Date: Mon, 4 Sep 1995 14:38:05 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509041718.aa11340@post.demon.co.uk> > From: descarte@mcqueen.com > > > Tim Bounce Boing! > > said you might have a perl5 informix or even DBI? > > I'm trying to find one, any help would be appreciated. > > Well, it's under rapid development at the moment. I'll have more details on > it on Monday or Tuesday, since I'm trying to get about 6 different projects > completed all at the same time just now! > > It's rudimentarily working. I just need to tidy things a little. > Wonderful! > Alligator Descartes | > descarte@mcqueen.com | > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-26@lion>; Tue, 5 Sep 1995 07:27:03 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 04 Sep 95 23:07:12 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa00692; 5 Sep 95 0:06 +0100 Received: by gw.home.vix.com id AA20506; Mon, 4 Sep 95 11:48:30 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20498; Mon, 4 Sep 95 11:48:26 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa10796; 4 Sep 95 19:44 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac06907; 4 Sep 95 19:42 +0100 Received: from ig.co.uk by lion id <20306-0@lion>; Mon, 4 Sep 1995 18:25:52 +0100 To: root@kersur.net Subject: Re: DBI 0.62 on Linux Cc: perldb-interest@vix.com Date: Mon, 4 Sep 1995 18:25:52 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509041942.ac06907@post.demon.co.uk> > From: root > > > Tried installing and testing DBI on my Linux system. The test suite > failed. Enclosed is the debugging information requested: > > --------------------------------------------------------------------------- > > deathstar:/usr/src/DBI# make test > [...] > Expect several rows of data: > fetchrow scalar context problem at test.pl line 124. > make: *** [test] Error 2 > deathstar:/usr/src/DBI# perl -e 'use Config; print Config::myconfig()' > Summary of my perl5 (patchlevel 1) configuration: > Platform: > osname=linux, osver=1, archname=i586-linux > uname='linux deathstar 1.2.9 #18 tue jul 11 19:35:44 edt 1995 i586 ' > Compiler: > cc='cc', optimize='-O2' > cppflags='-Dbool=char -DHAS_BOOL' This is a problem I've only seen with Perl's built with -DDEBUGGING. I've not got to the bottom of it yet (I've not actually spent much time looking). Does 'perl -Dx -e 0' say 'Recompile perl with -DDEBUGGING ...' for you? Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-37@lion>; Tue, 5 Sep 1995 07:28:22 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 00:43:57 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09900; 5 Sep 95 1:43 +0100 Received: by gw.home.vix.com id AA27421; Mon, 4 Sep 95 13:33:22 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27416; Mon, 4 Sep 95 13:33:19 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa29108; 4 Sep 95 17:21 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ah11340; 4 Sep 95 17:18 +0100 Received: from ig.co.uk by lion id <11035-0@lion>; Mon, 4 Sep 1995 15:46:03 +0100 To: fon@iinet.com.au Subject: Re: URGENT! I need a database Cc: perldb-interest@vix.com Date: Mon, 4 Sep 1995 15:46:03 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509041718.ah11340@post.demon.co.uk> > From: Surmani > [...] > > I've been using mSqlPerl on Linux, but now find it too limited for > my application (a web search engine). > > Please give me some advice on which database I should use. Commercial is > OK, but I would prefer to run it on Linux with a perl interface (DBI ?). > If it does SQL then even better. > > Please send me an URGENT response on this, > to the email address below: > > Adrian Smith > fon@iinet.com.au > I've heard that Oracle 7 runs on Linux (but it's unsupported). Since I'm not familiar with Linux I'll copy to the list for others to make suggestions (and no doubt quiz you about why mSQL is too limited for your application :-) Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-39@lion>; Tue, 5 Sep 1995 07:28:41 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 01:34:59 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa24371; 5 Sep 95 2:34 +0100 Received: by gw.home.vix.com id AA05018; Mon, 4 Sep 95 15:35:29 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05005; Mon, 4 Sep 95 15:35:26 -0700 Received: from eur.ericsson.se by rome (8.6.12/EUR) id PAA10705; Mon, 4 Sep 1995 15:35:12 -0700 Received: from ops by eur.ericsson.se (8.6.12/EUR-Dublin) id PAA07507; Mon, 4 Sep 1995 15:35:11 -0700 Received: from navajo.raynet.com by ops (4.1/SMI-4.1) id AA26696; Sun, 3 Sep 95 15:41:07 PDT Received: from localhost by navajo.raynet.com (4.1/SMI-4.1) id AA08554; Sun, 3 Sep 95 15:41:05 PDT Message-Id: <9509032241.AA08554@navajo.raynet.com> From: Yary Hluchan To: perldb-interest@vix.com Subject: new guy ignores labor day, goes to work and gripes X-Mailer: exmh version 1.4.1 7/21/94 Date: Sun, 03 Sep 1995 15:41:04 -0700 I hate it when my machine crashes with unsaved work. That's what happens when your xterm is a Mac... or is that what happens when you use computers... Let's try again d) Multiple SQL statements may not be combined in a single statement handle, e.g., a single $sh. This restriction has most impact upon Sybase users. An alternative approach may be possible later. e) Non-sequential record reads are not supported in this version of the DBperl API. E.g., records can only be fetched in the order that the database returned them and once fetched they are forgotten. f) Positioned updates and deletes are not currently supported by DBperl. E.g., it is not possible to execute a select query, fetch a number of rows and then update or delete the 'current row' before fetching more rows. All cursors are assumed to be read-only and unnamed. This may change in the future. Restrictions are made to keep people from doing what they want to do, ususally for the greater good. In this case I suppose you're saving work for the DBD implementors- I guess the tradeoff is DBI spec complexity vs. module usefulness... anyway I understand not all databases make it easy to implement all these features, but most allow at least one of the above. The DBI spec should specify a syntax for these, so that an application built to take advantage of one engine's select-and-update is portable to a different engine that also supports select-and-update. You've already identified that people do these things... it can be in the portability section as something no engine actually supports, but when people decide they need it, do you really want a different interface each time? e) Do we need functions to control: autocommit -- on / off locking -- level=row/page/table, readlocks, timeouts etc f) Do we need functions to: call database procedures return space left in/for the database I'd guess these are like the features denied above- different for each DB- well autocommit should be pretty easy (if implemented, is there going to be a default enforced?)- as for locking you could reasonably request "this level or stricter" as part of the spec... Standard DATA_TYPE values: 'CHARACTER', 'CHARACTER VARYING', 'DECIMAL', 'FLOAT', 'DOUBLE PRECISION', 'INTEGER', 'NUMERIC', 'REAL', 'SMALLINT', 'BIT', 'BIT VARYING', 'DATE', 'TIME', 'TIMESTAMP'. Are there any assumptions as to the lengths of the numerics? I guess there's always info_col in case of emergency... NUMERIC_SCALE isn't explicitly defined in the doc but I assume that what it's for. c) How to handle date and time intervals in a distinct manner from absolute dates and times is yet to be defined. It may be ignored. have native-time-interval functions been proposed? Something like ntint2sec/ sec2ntint that convert a number of seconds into a time interval and back. well that's all for now. -yary (yary@ops.raynet.com) ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-41@lion>; Tue, 5 Sep 1995 07:29:02 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 01:35:00 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa24422; 5 Sep 95 2:34 +0100 Received: by gw.home.vix.com id AA05023; Mon, 4 Sep 95 15:35:32 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05012; Mon, 4 Sep 95 15:35:28 -0700 Received: from eur.ericsson.se by rome (8.6.12/EUR) id PAA10711; Mon, 4 Sep 1995 15:35:13 -0700 Received: from ops by eur.ericsson.se (8.6.12/EUR-Dublin) id PAA07511; Mon, 4 Sep 1995 15:35:12 -0700 Received: from navajo.raynet.com by ops (4.1/SMI-4.1) id AA26719; Sun, 3 Sep 95 15:46:30 PDT Received: from localhost by navajo.raynet.com (4.1/SMI-4.1) id AA08571; Sun, 3 Sep 95 15:46:28 PDT Message-Id: <9509032246.AA08571@navajo.raynet.com> From: Yary Hluchan To: perldb-interest@vix.com Subject: new guy ignores labor day, goes to work and gripes X-Mailer: exmh version 1.4.1 7/21/94 Date: Sun, 03 Sep 1995 15:46:28 -0700 I hate it when my machine crashes with unsaved work. That's what happens when your xterm is a Mac... or is that what happens when you use computers... Let's try again d) Multiple SQL statements may not be combined in a single statement handle, e.g., a single $sh. This restriction has most impact upon Sybase users. An alternative approach may be possible later. e) Non-sequential record reads are not supported in this version of the DBperl API. E.g., records can only be fetched in the order that the database returned them and once fetched they are forgotten. f) Positioned updates and deletes are not currently supported by DBperl. E.g., it is not possible to execute a select query, fetch a number of rows and then update or delete the 'current row' before fetching more rows. All cursors are assumed to be read-only and unnamed. This may change in the future. Restrictions are made to keep people from doing what they want to do, ususally for the greater good. In this case I suppose you're saving work for the DBD implementors- I guess the tradeoff is DBI spec complexity vs. module usefulness... anyway I understand not all databases make it easy to implement all these features, but most allow at least one of the above. The DBI spec should specify a syntax for these, so that an application built to take advantage of one engine's select-and-update is portable to a different engine that also supports select-and-update. You've already identified that people do these things... it can be in the portability section as something no engine actually supports, but when people decide they need it, do you really want a different interface each time? e) Do we need functions to control: autocommit -- on / off locking -- level=row/page/table, readlocks, timeouts etc f) Do we need functions to: call database procedures return space left in/for the database I'd guess these are like the features denied above- different for each DB- well autocommit should be pretty easy (if implemented, is there going to be a default enforced?)- as for locking you could reasonably request "this level or stricter" as part of the spec... Standard DATA_TYPE values: 'CHARACTER', 'CHARACTER VARYING', 'DECIMAL', 'FLOAT', 'DOUBLE PRECISION', 'INTEGER', 'NUMERIC', 'REAL', 'SMALLINT', 'BIT', 'BIT VARYING', 'DATE', 'TIME', 'TIMESTAMP'. Are there any assumptions as to the lengths of the numerics? I guess there's always info_col in case of emergency... NUMERIC_SCALE isn't explicitly defined in the doc but I assume that what it's for. c) How to handle date and time intervals in a distinct manner from absolute dates and times is yet to be defined. It may be ignored. have native-time-interval functions been proposed? Something like ntint2sec/ sec2ntint that convert a number of seconds into a time interval and back. well that's all for now. -yary ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-43@lion>; Tue, 5 Sep 1995 07:29:22 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 01:38:45 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa25307; 5 Sep 95 2:38 +0100 Received: by gw.home.vix.com id AA20792; Mon, 4 Sep 95 11:50:19 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20774; Mon, 4 Sep 95 11:50:17 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa10817; 4 Sep 95 19:45 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ag06907; 4 Sep 95 19:42 +0100 Received: from ig.co.uk by lion id <20577-0@lion>; Mon, 4 Sep 1995 19:37:13 +0100 To: perldb-interest@vix.com, z50816@mip.lasc.lockheed.com Subject: Re: _install_me.al Date: Mon, 4 Sep 1995 19:37:13 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509041942.ag06907@post.demon.co.uk> > From: Terry Greenlaw > > > The real question is why did perl not see the _install_method in DBI.xs? > > > Make sure that DBI.c, DBI.o and DBI.so get built and DBI.so gets copied into > > ./blib. If so then try again with PERL_DL_DEBUG=2 and check that the > > DynaLoader is loading the right .so file. > > Tim, > You were right about having a static version of DBI already linked into Perl. > When I rebuilt and reinstalled perl the DBI build was easy. I built DBI > dynamically at first, but a statically built DBD/Oracle doesn't get along > well with a dynamic DBI. Building them both statically fixed the problem. > What do you mean by 'doesn't get along well with a dynamic DBI'? It ought to work. > There is (supposedly) a patch from HP for ld that will solve the "illegal > loader fixup needed" problem, allowing Oracle to finally load dynamically > on HP/UX 9. That would be great. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-45@lion>; Tue, 5 Sep 1995 07:29:35 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 02:08:56 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa03839; 5 Sep 95 3:08 +0100 Received: by gw.home.vix.com id AA07081; Mon, 4 Sep 95 16:11:50 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07074; Mon, 4 Sep 95 16:11:46 -0700 Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id BAA29541; Tue, 5 Sep 1995 01:09:48 +0200 Date: Tue, 5 Sep 1995 01:09:48 +0200 Message-Id: <199509042309.BAA29541@anna.mind.de> From: Andreas Koenig To: yhluchan@ops.raynet.com Cc: perldb-interest@vix.com In-Reply-To: <9509012351.AA02775@shawnee.raynet.com> (message from Yary Hluchan on Fri, 01 Sep 1995 16:51:14 -0700) Subject: Re: Welcome noise Reply-To: andreas.koenig@mind.de > Just one more hurdle to pass: where does one get the DBI >sources? I found out about this list via http://www.metronet.com/perlinfo/perl5 >/p5.modules.list, which has lots of contact addresses but no ftp sites. Which >is understandable, ideas don't sit in directories. Welcome to the world of idearectories, first address is ftp://ftp.demon.co.uk/pub/perl/db completely mirrored at ftp://ftp.funet.fi/pub/unix/languages/perl/ports/db also good: ftp://ftp.pasteur.fr/pub/Perl/CPAN/pml/byauthor/TIMB ftp://ftp.cs.ruu.nl/pub/PERL/perl5.0/CPAN/pml/byauthor/TIMB andreas ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-47@lion>; Tue, 5 Sep 1995 07:29:48 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 02:18:02 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa06128; 5 Sep 95 3:17 +0100 Received: by gw.home.vix.com id AA05011; Mon, 4 Sep 95 15:35:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04997; Mon, 4 Sep 95 15:35:25 -0700 Received: from eur.ericsson.se by rome (8.6.12/EUR) id PAA10719; Mon, 4 Sep 1995 15:35:15 -0700 Received: from ops by eur.ericsson.se (8.6.12/EUR-Dublin) id PAA07522; Mon, 4 Sep 1995 15:35:14 -0700 Received: from shawnee.raynet.com by ops (4.1/SMI-4.1) id AA27671; Fri, 1 Sep 95 16:51:16 PDT Received: from localhost by shawnee.raynet.com (4.1/SMI-4.1) id AA02775; Fri, 1 Sep 95 16:51:15 PDT Message-Id: <9509012351.AA02775@shawnee.raynet.com> From: Yary Hluchan To: perldb-interest@vix.com Subject: Welcome noise X-Mailer: exmh version 1.4.1 7/21/94 Date: Fri, 01 Sep 1995 16:51:14 -0700 Hello! I managed to find the mailing list and join without posting to the general group. Just one more hurdle to pass: where does one get the DBI sources? I found out about this list via http://www.metronet.com/perlinfo/perl5 /p5.modules.list, which has lots of contact addresses but no ftp sites. Which is understandable, ideas don't sit in directories. -y! ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-51@lion>; Tue, 5 Sep 1995 07:30:19 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 03:08:11 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa18444; 5 Sep 95 4:07 +0100 Received: by gw.home.vix.com id AA01348; Mon, 4 Sep 95 14:33:18 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01342; Mon, 4 Sep 95 14:33:16 -0700 Received: from alma.mcqueen.com (alma.mcqueen.com [193.133.160.249]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with ESMTP id WAA23630; Mon, 4 Sep 1995 22:29:20 +0100 From: descarte@mcqueen.com Received: (from descarte@localhost) by alma.mcqueen.com (8.6.12/8.6.12) id WAA01591; Mon, 4 Sep 1995 22:35:21 +0100 Message-Id: <199509042135.WAA01591@alma.mcqueen.com> Subject: Re: URGENT! I need a database (fwd) To: perldb-interest@vix.com Date: Mon, 4 Sep 1995 22:35:21 +0100 (BST) Cc: fon@iinet.com.au X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1449 > > I've been using mSqlPerl on Linux, but now find it too limited for > > my application (a web search engine). > > > > Please give me some advice on which database I should use. Commercial is > > OK, but I would prefer to run it on Linux with a perl interface (DBI ?). > > If it does SQL then even better. > > > > Please send me an URGENT response on this, > > to the email address below: > > > > Adrian Smith > > fon@iinet.com.au > > > I've heard that Oracle 7 runs on Linux (but it's unsupported). Yup, completely unsupported! It does run though. I've seen it with my own eyes. Took me about 3 days to get the bastard thing to work at all, let alone properly. I wouldn't really recommend it. Besides, the binaries you get are for the SCO platform ( Satan incarnate ) and you need to run iBCS ( iCBS? ) emulation under the Linux kernel to get it to run at all. At least, I did. > Since I'm not familiar with Linux I'll copy to the list for others to > make suggestions (and no doubt quiz you about why mSQL is too limited > for your application :-) Yes, I must say, it sounds a trifle bizarre. Informix may be a better bet on a Linux platform since it's memory grunginess isn't quite as disgusting as Oracle. Of course, if your web engine's serious then you need something with a bit more welly than a PC, I'd say. Infoseek runs an 8 CPU Sparc1000 last I heard, for instance. > Tim. -- Alligator Descartes | descarte@mcqueen.com | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-69@lion>; Tue, 5 Sep 1995 07:35:26 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 04:44:43 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa11479; 5 Sep 95 5:44 +0100 Received: by gw.home.vix.com id AA05004; Mon, 4 Sep 95 15:35:26 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04995; Mon, 4 Sep 95 15:35:24 -0700 Received: from eur.ericsson.se by rome (8.6.12/EUR) id PAA10701; Mon, 4 Sep 1995 15:35:11 -0700 Received: from ops by eur.ericsson.se (8.6.12/EUR-Dublin) id PAA07503; Mon, 4 Sep 1995 15:35:10 -0700 Received: from navajo.raynet.com by ops (4.1/SMI-4.1) id AA26722; Sun, 3 Sep 95 15:46:54 PDT Received: from localhost by navajo.raynet.com (4.1/SMI-4.1) id AA08577; Sun, 3 Sep 95 15:46:52 PDT Message-Id: <9509032246.AA08577@navajo.raynet.com> From: Yary Hluchan To: perldb-interest@vix.com Subject: erratum X-Mailer: exmh version 1.4.1 7/21/94 Date: Sun, 03 Sep 1995 15:46:52 -0700 ] well that's all for now. ] -yary (yary@ops.raynet.com) and I signed with the wrong address, sheesh. That's yhluchan@ops.raynet.com while I'm here, it'd be nice if the perlmod list posted at http:// www.metronet.com/perlinfo/perl5/p5.modules.list had ftp addresses as well as mailing list addresses... a sentiment I already sent but it hasn't bounced back to me yet ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <23491-71@lion>; Tue, 5 Sep 1995 07:35:45 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 04:45:20 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa11493; 5 Sep 95 5:44 +0100 Received: by gw.home.vix.com id AA07098; Mon, 4 Sep 95 16:12:37 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07093; Mon, 4 Sep 95 16:12:36 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa10781; 4 Sep 95 19:44 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab06907; 4 Sep 95 19:41 +0100 Received: from ig.co.uk by lion id <20269-0@lion>; Mon, 4 Sep 1995 18:19:51 +0100 To: mhm@austin.ibm.com Subject: Re: Building the %attribs for bind_param ? Cc: perldb-interest@vix.com, Jacqui.Caren@ig.co.uk Date: Mon, 4 Sep 1995 18:19:51 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509041941.ab06907@post.demon.co.uk> > From: mhm@austin.ibm.com > To: Tim.Bounce@ig.co.uk > > The question boils down to what is the best way to build the %attribs > hash-ref for bind_param and bind-col. Given the number of possible > things that a DB2 user might want to bind to a parameter or in binding > the incoming data to a host variable, I think I need to supply a > function which builds and returns the %attribs hash-ref. > > Would it be better to define it as an XSUB routine or in the pm file > under DBD::DB2::st? I currently leaning towards doing it in perl in > the pm file. > > Any comments, anything else I need to consider? > I don't see any need for this. Could you give me some examples of the 'number of possible things that a DB2 user might want to bind to ...' > Mike > Tim. p.s. Mail to 'Tim.Bounce@ig.co.uk' only reaches me because my mail admin has a sense of humor! I'm now going to shoot her. Further mail will bounce! ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <12667-8@lion>; Wed, 6 Sep 1995 07:34:19 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 04:24:21 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa14316; 6 Sep 95 5:23 +0100 Received: from frigate.doc.ic.ac.uk by gate.demon.co.uk id ab24724; 6 Sep 95 0:00 GMT-60:00 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <07848-0@frigate.doc.ic.ac.uk>; Tue, 5 Sep 1995 23:41:07 +0100 Received: by gw.home.vix.com id AA29294; Tue, 5 Sep 95 05:41:10 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29286; Tue, 5 Sep 95 05:40:59 -0700 Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id OAA00941; Tue, 5 Sep 1995 14:33:25 +0200 Date: Tue, 5 Sep 1995 14:33:25 +0200 Message-Id: <199509051233.OAA00941@anna.mind.de> From: Andreas Koenig To: descarte@mcqueen.com Cc: perldb-interest@vix.com, Tim.Bunce@ig.co.uk In-Reply-To: <199509041422.PAA13240@alma.mcqueen.com> (descarte@mcqueen.com) Subject: Re: ANNOUNCE: DBD::mSQL-0.60pl1 released Reply-To: andreas.koenig@mind.de > > ftp://ftp.mcqueen.com/pub/languages/perl5/DBPerl/ > >Let me know how things go. That's how I patched it: --- mSQL.xs.601 Tue Sep 5 14:16:46 1995 +++ mSQL.xs Tue Sep 5 14:24:51 1995 @@ -99,7 +99,7 @@ SV * dbh SV * keysv CODE: - SV *valuesv = dbd_db_FETCH(dbh, keysv); + SV *valuesv = (SV*)dbd_db_FETCH(dbh, keysv); if (!valuesv) { /* XXX hand-off to DBI for possible processing */ croak("Can't get %s->{%s}: unrecognised attribute", @@ -264,7 +264,7 @@ long len SV * destsv CODE: - ST(0) = dbd_st_readblob(sth, field, offset, len, destsv); + ST(0) = (SV*)dbd_st_readblob(sth, field, offset, len, destsv); void @@ -286,7 +286,7 @@ SV * sth SV * keysv CODE: - SV *valuesv = dbd_st_FETCH(sth, keysv); + SV *valuesv = (SV*)dbd_st_FETCH(sth, keysv); if (!valuesv) { /* XXX hand-off to DBI for possible processing */ croak("Can't get %s->{%s}: unrecognised attribute", And here is the output of the test script: Statement handle DBI::st=HASH(0x10014b44) DESTROY ignored - never set up at test2.pl line 96. Cannot prepare sth (0): ** Connecting to first instance.... ------------------------------------------------------- MINERVA_DEBUG found. libmsql started with the following:- ------------------------------------------------------- ** Disconnecting from first instance.... ** Connecting to second instance..... **** Trying cursor returning rows as list..... FUNC!: SELECT ReturnRow: 1 Test Guy1 ReturnRow: 2 Test Guy2 ReturnRow: 3 Rubbish Guy1 ReturnRow: 3 Rubbish Guy1 ReturnRow: 3 Rubbish Guy1 ReturnRow: 3 Rubbish Guy1 **** Re-doing the cursor but fetching fields separately..... FUNC!: SELECT Id: 1 Name: Test Guy1 Id: 2 Name: Test Guy2 Id: 3 Name: Rubbish Guy1 Id: 3 Name: Rubbish Guy1 Id: 3 Name: Rubbish Guy1 Id: 3 Name: Rubbish Guy1 **** Trying an INSERT now....... FUNC!: INSERT INSERT present in statement **** Re-doing the cursor but fetching fields separately..... FUNC!: SELECT Id: 1 Name: Test Guy1 Id: 2 Name: Test Guy2 Id: 3 Name: Rubbish Guy1 Id: 3 Name: Rubbish Guy1 Id: 3 Name: Rubbish Guy1 Id: 3 Name: Rubbish Guy1 Id: 3 Name: Rubbish Guy1 **** Trying a CREATE statement...... FUNC!: CREATE CREATE present in statement rc: 0 what: Table "pants3" exists Hope, that helps, andreas ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12666-34@lion>; Wed, 6 Sep 1995 07:39:03 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 02:49:49 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ab29495; 6 Sep 95 3:48 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <10298-0@frigate.doc.ic.ac.uk>; Wed, 6 Sep 1995 01:40:31 +0100 Received: by gw.home.vix.com id AA03486; Tue, 5 Sep 95 13:08:14 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03479; Tue, 5 Sep 95 13:08:12 -0700 Received: (thorn@localhost) by amaretto.cc.mcgill.ca (8.6.10/8.6.5) id QAA07596; Tue, 5 Sep 1995 16:08:10 -0400 Date: Tue, 5 Sep 1995 16:08:10 -0400 (EDT) From: Ron Hall To: perldb-interest@vix.com Subject: How does one do this....? Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Using oraperl2.4 Oracle 7.blah blah blah how does one execute a stired procedure? Just Curious? Ron Hall | DISCLAIMER: I said it. I must've meant it. There is Analyst | nobody else to blame. | Clown:n: A man who acts too natural. +1 514 398 3718 | "Whenever doubt can be called into question, do so." ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12666-39@lion>; Wed, 6 Sep 1995 07:39:50 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 05 Sep 95 22:19:19 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa11794; 5 Sep 95 23:19 +0100 Received: by gw.home.vix.com id AA08067; Tue, 5 Sep 95 07:47:32 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA08062; Tue, 5 Sep 95 07:47:31 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id JAA157585 for ; Tue, 5 Sep 1995 09:47:30 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA19524; Tue, 5 Sep 1995 09:47:27 -0500 From: mhm@austin.ibm.com Message-Id: <9509051447.AA19524@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: hash refs as parameters to bind_param Date: Tue, 05 Sep 95 09:47:25 -0500 What is the correct way to create the %attribs hash ref for bind_param? $sth->bind_param($param_index_or_name,\$value[,\%attribs]) Whatever combination of things I try I get the following from dbd_bind_ph: DBI::st=HASH(0x200d5fc8)->bind_param(:p1): invalid attribute parameter at genreq.pl line 15. which comes from dbd_bind_ph: if (!SvROK(attribs) || SvTYPE(SvRV(attribs)) == SVt_PVHV) croak("%s->bind_param(%s): invalid attribute parameter", SvPV(sth,na), name); It is the "SvTYPE(SvRV(attribs)) == SVt_PVHV" which is not true. Anybody have some examples? Thanks. Mike Moran ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12666-56@lion>; Wed, 6 Sep 1995 07:44:28 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 00:30:50 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa11867; 6 Sep 95 1:30 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa18509; 5 Sep 95 23:16 GMT-60:00 Received: by gw.home.vix.com id AA26853; Tue, 5 Sep 95 05:13:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA26848; Tue, 5 Sep 95 05:13:46 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id HAA126634 for ; Tue, 5 Sep 1995 07:13:44 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA20052; Tue, 5 Sep 1995 07:13:43 -0500 From: mhm@austin.ibm.com Message-Id: <9509051213.AA20052@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: Your message of Mon, 04 Sep 95 18:19:51 +0100. <9509041941.ab06907@post.demon.co.uk> Date: Tue, 05 Sep 95 07:13:42 -0500 Tim Bunce writes: > > > From: mhm@austin.ibm.com > > To: Tim.Bounce@ig.co.uk > > > > The question boils down to what is the best way to build the %attribs > > hash-ref for bind_param and bind-col. Given the number of possible > > things that a DB2 user might want to bind to a parameter or in binding > > the incoming data to a host variable, I think I need to supply a > > function which builds and returns the %attribs hash-ref. > > > > Would it be better to define it as an XSUB routine or in the pm file > > under DBD::DB2::st? I currently leaning towards doing it in perl in > > the pm file. > > > > Any comments, anything else I need to consider? > > > I don't see any need for this. Could you give me some examples of the > 'number of possible things that a DB2 user might want to bind to ...' > Well, let us see: 1. Specify the incoming host variable type (16 options). 2. Specify the SQL variable type (25 options) 3. Specify precision for numeric, character, and blob types. 4. Specify scale for numeric and time stamp types 5. Specify whether the statement gets evaluated as an input, input deferred (stored proceedure), output or output deferred. 6. Specify truncation levels for character and binary types 7. Specify either a host variable to bound or special value to a parameter. (DB2 allows the posting of several special "variables" which cause the database to perform a special function, yet typically the only result is a positive return code) Of the above list, I can guarantee that I need to support the first 4 on every statement executed. Item 5, I can defer until I want/need to support stored procedures. As for items 6&7, what I am doing currently has no need, but I would guess that folks attempting to use DB2perl on existing databases will probably want them fairly quickly. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12666-62@lion>; Wed, 6 Sep 1995 07:45:28 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 00:48:09 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa18416; 6 Sep 95 1:47 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa05929; 5 Sep 95 21:47 GMT-60:00 Received: by gw.home.vix.com id AA29351; Tue, 5 Sep 95 05:43:44 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29346; Tue, 5 Sep 95 05:43:41 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab20500; 5 Sep 95 13:37 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa17629; 5 Sep 95 13:35 +0100 Received: from ig.co.uk by lion id <01337-0@lion>; Tue, 5 Sep 1995 13:37:28 +0100 To: perldb-interest@vix.com, yhluchan@ops.raynet.com Subject: Re: new guy ignores labor day, goes to work and gripes Date: Tue, 5 Sep 1995 13:37:28 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509051335.aa17629@post.demon.co.uk> > From: Yary Hluchan > > d) Multiple SQL statements may not be combined in a single statement > handle, e.g., a single $sh. This restriction has most impact upon > Sybase users. An alternative approach may be possible later. > Something may be possible for non-selects. It's not a priority. > e) Non-sequential record reads are not supported in this version of the > DBperl API. E.g., records can only be fetched in the order that the > database returned them and once fetched they are forgotten. > A FetchScroll function (ala X/Open SQL CLI) is looking more probable now. > f) Positioned updates and deletes are not currently supported by > DBperl. E.g., it is not possible to execute a select query, fetch a > number of rows and then update or delete the 'current row' before > fetching more rows. All cursors are assumed to be read-only and > unnamed. This may change in the future. > This is also looking more probable. > Restrictions are made to keep people from doing what they want to do, ususally > for the greater good. In this case I suppose you're saving work for the DBD > implementors- I guess the tradeoff is DBI spec complexity vs. module > usefulness... anyway I understand not all databases make it easy to implement > all these features, but most allow at least one of the above. The DBI spec > should specify a syntax for these, so that an application built to take > advantage of one engine's select-and-update is portable to a different engine > that also supports select-and-update. You've already identified that people do > these things... it can be in the portability section as something no engine > actually supports, but when people decide they need it, do you really want a > different interface each time? > Once more than one driver supports a concept via a private method then we can talk about a common interface. Till then it's not worth it. The primary focus remains on the real standards (X/Open SQL CLI). > e) Do we need functions to control: > autocommit -- on / off > locking -- level=row/page/table, readlocks, timeouts etc > > f) Do we need functions to: > call database procedures > return space left in/for the database > > I'd guess these are like the features denied above- different for each DB- > well autocommit should be pretty easy (if implemented, is there going to be a > default enforced?)- as for locking you could reasonably request "this level or > stricter" as part of the spec... > 'autocommit -- on / off' has arrived quietly as $dbh->{AutoCommit}. 'call database procedures' will probably get addressed when I do DBD::Ingres. > Standard DATA_TYPE values: > 'CHARACTER', 'CHARACTER VARYING', 'DECIMAL', 'FLOAT', > 'DOUBLE PRECISION', 'INTEGER', 'NUMERIC', 'REAL', 'SMALLINT', > 'BIT', 'BIT VARYING', 'DATE', 'TIME', 'TIMESTAMP'. > > Are there any assumptions as to the lengths of the numerics? I guess there's > always info_col in case of emergency... NUMERIC_SCALE isn't explicitly defined > in the doc but I assume that what it's for. > > c) How to handle date and time intervals in a distinct manner from > absolute dates and times is yet to be defined. It may be ignored. > > have native-time-interval functions been proposed? Something like ntint2sec/ > sec2ntint that convert a number of seconds into a time interval and back. > Discussed at great length in the archives. Data type issues will now follow the standards (the earlier version of (X/Open SQL CLI didn't include DATE / TIME etc). The bottom line is that the DBI will grow over time and follow whatever standards are appropriate. (Where no standard is applicable we'll make something up!) The 0.6 spec is very dated now. I have recently received the latest X/Open SQL specs (CLI and SQL) and I'll be talking about the implications of those once I've had time to consider them. Expect a period of heavy discussion and rapid change in the not-too-distant future. > well that's all for now. > -yary (yary@ops.raynet.com) > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12666-70@lion>; Wed, 6 Sep 1995 07:46:46 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 02:30:24 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa23446; 6 Sep 95 3:29 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa21183; 5 Sep 95 20:08 GMT-60:00 Received: by gw.home.vix.com id AA10292; Tue, 5 Sep 95 01:39:37 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10287; Tue, 5 Sep 95 01:39:34 -0700 Received: from yquem.inferentia.it (yquem.inferentia.it [151.99.220.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id KAA21545; Tue, 5 Sep 1995 10:38:30 +0200 Posted-Date: Tue, 5 Sep 1995 10:38:30 +0200 Received: by yquem.inferentia.it with Microsoft Mail id <01BA7B7C.E11F5C80@yquem.inferentia.it>; Tue, 5 Sep 1995 10:46:38 +-200 Message-Id: <01BA7B7C.E11F5C80@yquem.inferentia.it> From: Davide Migliavacca To: "'perldb-interest@vix.com'" Cc: 'Tim Bunce' Subject: DBI at work Date: Tue, 5 Sep 1995 10:46:37 +-200 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable DBI 0.62 and DBD::Oracle 0.25, with a wowperl-like wrapper, are now at = REAL work on http://www.smau.it You can access the search engine using http://www.smau.it/magellano/ricerche/gsearch.shtml (italian) http://www.smau.it/magellano/english/ricerche/gsearch.shtml (english) The search is implemented using a wowperl-derivative (the CGI onlinefe = you can see mentioned in the source of those forms) which calls an = Oracle 7.1 PL/SQL package and then outputs the results. BTW, the translation of the page elements is done automagically by the = wowperl derivative which acts as a front-end to the stored procedures. A number of batch-built pages (all lists) are also built calling the = same thing but in background (since building those lists at runtime is = too costly). Unfortunately the real data is not very multi-language, so very few = things will be understandable by a non-italian speaking guy. That's not = the fault of DBI :-) The system runs on a DEC Alphaserver 1000 4/200 with 128 Mb of RAM, = Oracle 7.1 in a very large configuration, OSF/1 3.2, perl 5.001m, DBI = 0.62, DBD::Oracle 0.25, Oraperl and CGI.pm. DBI will be even more at work during the fair the server is referring to = (SMAU 95), in which it will take care of a moltitude of features, = comprising a BLOB replicator for Oracle 7.1. Yes, it is already working! Davide Migliavacca (a proud user of DBI) ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12666-93@lion>; Wed, 6 Sep 1995 07:51:03 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 06:06:42 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa04964; 6 Sep 95 7:06 +0100 Received: by gw.home.vix.com id AB07081; Tue, 5 Sep 95 13:47:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07072; Tue, 5 Sep 95 13:47:39 -0700 Received: from eur.ericsson.se by rome (8.6.12/EUR) id NAA19369; Tue, 5 Sep 1995 13:46:58 -0700 Received: from ops by eur.ericsson.se (8.6.12/EUR-Dublin) id NAA16402; Tue, 5 Sep 1995 13:46:58 -0700 Received: from shawnee.raynet.com by ops (4.1/SMI-4.1) id AA21079; Tue, 5 Sep 95 13:48:38 PDT Received: from localhost by shawnee.raynet.com (4.1/SMI-4.1) id AA07878; Tue, 5 Sep 95 13:48:36 PDT Message-Id: <9509052048.AA07878@shawnee.raynet.com> From: Yary Hluchan To: perldb-interest@vix.com Subject: Re: new guy ignores labor day, goes to work and gripes In-Reply-To: <9509051335.aa17629@post.demon.co.uk> X-Mailer: exmh version 1.4.1 7/21/94 Date: Tue, 05 Sep 1995 13:48:36 -0700 ] The 0.6 spec is very dated now. I have recently received the latest ] X/Open SQL specs (CLI and SQL) and I'll be talking about the implications ] of those once I've had time to consider them. Expect a period of heavy ] discussion and rapid change in the not-too-distant future. oooh, I can't wait! There was some talk about how to represent booleans in the package- my first pass at a solution would be to have a hash that's populated by the DBD (and possibly client program?), like %Bool=(TRUE => 1, 1 => 1, FALSE => 0, 0 => 0) I suppose I should dig through the archives sometime soon... -y! (Eagerly awaiting a chance to crash & dissect the Ingres mod) ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <18800-1@lion>; Wed, 6 Sep 1995 14:59:05 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 08:22:11 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa03208; 6 Sep 95 9:20 +0100 Received: by gw.home.vix.com id AA29851; Tue, 5 Sep 95 18:34:34 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29846; Tue, 5 Sep 95 18:34:33 -0700 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id UAA14485; Tue, 5 Sep 1995 20:34:30 -0500 Date: Tue, 5 Sep 1995 20:34:29 -0500 (CDT) From: Ben A Lindstrom Cc: perldb-interest@vix.com Subject: DBI/QuickBase/Objective-C In-Reply-To: <9508311106.ab26149@post.demon.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII After skim through the dbdimp.h and I think I have a good idea on how to quickly implement the Quickbase engine, but I was wonder if anyone knew if Perl can handle Objective-C or should I write a 2nd library that changes Objective-C into plan C. Thanks. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <25042-7@lion>; Wed, 6 Sep 1995 22:14:52 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 21:11:08 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ab11190; 6 Sep 95 22:10 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <28292-0@frigate.doc.ic.ac.uk>; Wed, 6 Sep 1995 21:26:40 +0100 Received: by gw.home.vix.com id AA24524; Wed, 6 Sep 95 01:05:54 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA24518; Wed, 6 Sep 95 01:05:45 -0700 Received: from bssmlsa.nerc-bas.ac.uk.nerc-bas.ac.uk by bsfiles.nerc-bas.ac.uk (8.6.10/NERC-1.2(Solaris 2.x) id JAA16416; Wed, 6 Sep 1995 09:07:13 +0100 Date: Wed, 6 Sep 1995 09:05:39 +0100 Message-Id: <6054.199509060805@bssmlsa.nerc-bas.ac.uk.nerc-bas.ac.uk> From: Andrew G Wood To: perldb-interest@vix.com In-Reply-To: (message from Ron Hall on Tue, 5 Sep 1995 16:08:10 -0400 (EDT)) Subject: Re: How does one do this....? Reply-To: agw@mail.nerc-bas.ac.uk Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII >>>>> "Ron" == Ron Hall writes: > Using oraperl2.4 Oracle 7.blah blah blah how does one execute a > stired procedure? Just Curious? I think you will need to get hold of the two new extensions that I have been developing over the past months .... SpoOn and WooDen .... soo your Perl will look somethings like: use WooDen; use SpoOn; Obtained from the unusual place .... beware they are still under development! :-) Andy. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <27189-1@lion>; Thu, 7 Sep 1995 07:21:13 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 23:41:17 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa29860; 7 Sep 95 0:41 +0100 Received: by gw.home.vix.com id AA10768; Wed, 6 Sep 95 11:09:03 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10760; Wed, 6 Sep 95 11:09:01 -0700 Received: from isye.isye.gatech.edu by gatech.edu with SMTP id AA02503 (5.65c/Gatech-10.0-IDA for ); Wed, 6 Sep 1995 14:11:56 -0400 Received: from chmsr.gatech.edu by isye.isye.gatech.edu (4.1/SMI-4.1) id AA10132; Wed, 6 Sep 95 14:08:55 EDT Received: by chmsr.gatech.edu (4.1/SMI-4.1) id AA07547; Wed, 6 Sep 95 14:08:52 EDT Date: Wed, 6 Sep 1995 14:08:52 -0400 (EDT) From: Jennifer Anne Turns X-Sender: jennifer@chmsr To: perldb-interest@vix.com, Tim.Bunce@ig.co.uk Cc: Benjamin Jung , Jennifer Anne Turns Subject: Problems "make"ing the DBD::Oracle driver Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I am in the process of trying to get the Perl5 to communicate with Oracle7. I have downloaded the most recent versions of the DBI (0.63) and DBD (0.25) and have successfully installed DBI and tested it. I am having problems when "make"ing DBD. The information reulting from the first two steps of this process are given below: --------------------------------------------------------------------- THIS IS THE SUCCESSFUL OUTPUT FROM THE COMMAND: perl Makefile.PL root > perl Makefile.PL Configuring DBD::Oracle version 0.25... Remember to actually read the README file! Using Oracle in /usr/ld/oracle Oracle sysliblist: -lm Using /usr/ld/oracle/proc/lib/proc.mk version 7.27 System: perl5.001 sunos srl 5.2 generic_100999-40 sun4m sparc sun4-solaris dl_dlopen.xs Compiler: cc -O Oracle proc.mk would have used these values but we override them: CC: /usr/5bin/cc CFLAGS: -I. -O LDFLAGS: -Bdynamic -L$(LIBHOME) Checking if your kit is complete... Looks good Writing Makefile for DBD::Oracle --------------------------------------------------------------------- THIS IS THE UNSUCCESSFUL OUTPUT FROM THE COMMAND: make root > make cc -c -O -Kpic -I/usr/local/lib/perl5/sun4-solaris/CORE -I/usr/ld/oracle/rdbms/demo -I/usr/local/lib/perl5/sun4-solaris/DBI Oracle.c cc -c -O -Kpic -I/usr/local/lib/perl5/sun4-solaris/CORE -I/usr/ld/oracle/rdbms/demo -I/usr/local/lib/perl5/sun4-solaris/DBI dbdimp.c Running Mkbootstrap for DBD::Oracle () chmod 644 Oracle.bs LD_RUN_PATH="" cc -o blib/auto/DBD/Oracle/Oracle.so -G Oracle.o dbdimp.o -L/usr/ld/oracle/lib /usr/ld/oracle/lib/libocic.a /usr/ld/oracle/lib/osntab.o -lsqlnet -lora /usr/ld/oracle/lib/libpls.a -lsqlnet -lnlsrtl -lcv6 -lcore -lnlsrtl -lcv6 -lcore `cat /usr/ld/oracle/rdbms/lib/sysliblist` -lm ld: fatal: file /usr/ld/oracle/lib/libocic.a: unknown type, unable to process using elf(3E) libraries *** Error code 1 make: Fatal error: Command failed for target `blib/auto/DBD/Oracle/Oracle.so' ------------------------------------------------------------------- Someone here suggested that perhaps the problem is with the C compiler that is being used - which is the following: Sparc C Compiler 2.0.1 Any suggestions? jennifer ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-5@lion>; Thu, 7 Sep 1995 07:21:29 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 06 Sep 95 23:20:26 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa22470; 7 Sep 95 0:20 +0100 Received: by gw.home.vix.com id AA14903; Wed, 6 Sep 95 05:38:34 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA14898; Wed, 6 Sep 95 05:38:33 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab22600; 6 Sep 95 10:22 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa03435; 6 Sep 95 10:20 +0100 Received: from ig.co.uk by lion id <16076-0@lion>; Wed, 6 Sep 1995 10:22:37 +0100 To: perldb-interest@vix.com, thorn@cc.mcgill.ca Subject: Re: How does one do this....? Date: Wed, 6 Sep 1995 10:22:37 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509061020.aa03435@post.demon.co.uk> > From: Ron Hall > > Using oraperl2.4 Oracle 7.blah blah blah how does one execute > a stired procedure? Just Curious? > Take a look in ftp.demon.co.uk:/pub/perl/db/perl4/oraperl/* Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <27189-3@lion>; Thu, 7 Sep 1995 07:21:29 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 00:21:47 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa16105; 7 Sep 95 1:19 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa18073; 6 Sep 95 23:34 GMT-60:00 Received: by gw.home.vix.com id AA01038; Wed, 6 Sep 95 09:06:18 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01033; Wed, 6 Sep 95 09:06:16 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa11864; 6 Sep 95 15:29 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa26609; 6 Sep 95 15:26 +0100 Received: from ig.co.uk by lion id <19608-0@lion>; Wed, 6 Sep 1995 15:10:28 +0100 To: mouring@netnet.net Subject: Re: DBI/QuickBase/Objective-C Cc: perldb-interest@vix.com Date: Wed, 6 Sep 1995 15:10:28 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509061527.aa26609@post.demon.co.uk> > From: Ben A Lindstrom > > After skim through the dbdimp.h and I think I have a good idea on how > to quickly implement the Quickbase engine, I hope you've looked at more than just dbdimp.h! > but I was wonder if anyone > knew if Perl can handle Objective-C or should I write a 2nd library that > changes Objective-C into plan C. > Try asking in perl5-porters@nicoh.com and comp.lang.perl.misc. Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <27189-6@lion>; Thu, 7 Sep 1995 07:21:57 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 01:22:22 GMT Received: from gate.demon.co.uk by punt2.demon.co.uk id aa12263; 7 Sep 95 2:17 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa22729; 7 Sep 95 0:17 GMT-60:00 Received: by gw.home.vix.com id AA20854; Wed, 6 Sep 95 07:03:26 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20844; Wed, 6 Sep 95 07:03:20 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aj29113; 6 Sep 95 12:02 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab16785; 6 Sep 95 11:46 +0100 Received: from ig.co.uk by lion id <16290-0@lion>; Wed, 6 Sep 1995 10:48:43 +0100 To: perldb-interest@vix.com, yhluchan@ops.raynet.com Subject: Re: new guy ignores labor day, goes to work and gripes Date: Wed, 6 Sep 1995 10:48:43 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509061146.ab16785@post.demon.co.uk> > From: Yary Hluchan > > ] The 0.6 spec is very dated now. I have recently received the latest > ] X/Open SQL specs (CLI and SQL) and I'll be talking about the implications > ] of those once I've had time to consider them. Expect a period of heavy > ] discussion and rapid change in the not-too-distant future. > > oooh, I can't wait! > > There was some talk about how to represent booleans in the package- my first > pass at a solution would be to have a hash that's populated by the DBD (and > possibly client program?), like %Bool=(TRUE => 1, 1 => 1, FALSE => 0, 0 => 0) > I doubt that'll ever make it into reality. I'll probably stick to perl's concept of truth and just allow a TRUE and FALSE function to be exported/imported for those who feel the need to be verbose. > I suppose I should dig through the archives sometime soon... > > -y! > (Eagerly awaiting a chance to crash & dissect the Ingres mod) > I'm bogged down in real work at the moment. Maybe next week. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-13@lion>; Thu, 7 Sep 1995 07:22:43 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 00:54:58 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa08049; 7 Sep 95 1:54 +0100 Received: by gw.home.vix.com id AA03562; Wed, 6 Sep 95 09:36:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03557; Wed, 6 Sep 95 09:36:23 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa16637; 6 Sep 95 16:30 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa07771; 6 Sep 95 16:28 +0100 Received: from ig.co.uk by lion id <20204-0@lion>; Wed, 6 Sep 1995 16:13:56 +0100 To: sante@iasi.com Subject: Re: DBperl for perl5 Cc: perldb-interest@vix.com Date: Wed, 6 Sep 1995 16:13:56 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509061628.aa07771@post.demon.co.uk> > From: Stephanie Sante > > Hello Tim. I am consulting here at IAS and they have given me a DBperl > API spec version 0.6 Draft. So far I have been unable to locate the > DBperl libs for perl5. I have logged onto demon.co.uk and read the > README file. However I do not see DBperl for perl5. I do see sybperl > for perl5 though. This client has been doing reports for a client using > perl4 and engperl. I am doing a conversion to perl5 and for a client which > will be using sybase. It was the hope that the port to DBperl and perl5 > would allow the DB connections to be generic in the future and the $ENV > or $Args would simply drive the DB application reports. Could you please > shed some light on where the current status of things are and where I can > get the approbriate libraries. Thank you very very much. > The spec is very out of date. DBperl got renamed to 'the DBI module' (with 'drivers' having DBD::* module names) quite a while ago. Take a look at ftp.demon.co.uk:/pub/perl/db/... DBI/DBI-0.63.tar.Z (read the notes at the end of DBI.pm) DBI/perldb-interest/perldb-i.v06* (at least the last couple) DBD/Oracle/DBD-Oracle-0.25.tar.Z (as an example of a driver) No 'DBD::Sybase' driver exists yet but it should be easy to implement one as an emulation layer over the existing Sybase module. Volunteers are very welcome. The status is that everything is alpha but many people are using it (DBI + DBD::Oracle) for real work. If Sybase is your only target database then I'd recommend sticking with the Sybase module for now. To stay in touch subscribe to the mailing list by sending a message to perldb-interest-request@vix.com. > - Stephanie - > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-15@lion>; Thu, 7 Sep 1995 07:23:20 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 01:11:03 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa15851; 7 Sep 95 2:10 +0100 Received: by gw.home.vix.com id AA27875; Wed, 6 Sep 95 14:24:07 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27135; Wed, 6 Sep 95 14:16:35 -0700 Received: from ganef.cs.ucla.edu by maui.cs.ucla.edu (Sendmail 4.1/3.27) id AA20591; Tue, 5 Sep 95 18:13:42 PDT Received: by ganef.cs.ucla.edu (4.1/SMI-4.1) id AA11245; Tue, 5 Sep 95 18:11:03 PDT From: Scott Michel Message-Id: <9509060111.AA11245@ganef.cs.ucla.edu> Subject: DBD::mSQL bought the farm... ;-( To: perldb-interest@vix.com Date: Tue, 5 Sep 1995 18:11:03 -0700 (PDT) Reply-To: scottm@intime.com X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 29387 Al: It's lava. It's molten. It sorta works, but I don't have the current technical expertise at the moment (ok, gimme 15 seconds!?!) to figure out exactly what DBD::mSQL's complaint it. Seems to be internal. I'm hoping the supplied is adequate? -scottm perl -e 'use Config; print Config::myconfig()' ---------------------------------------------- Summary of my perl5 (patchlevel 1) configuration: Platform: osname=sunos, osver=4.1.3_u1, archname=sun4-sunos uname='sunos ganef 4.1.3_u1 1 sun4m ' hint=recommended Compiler: cc='gcc', optimize='-O2' cppflags='-DHIDEMYMALLOC' ccflags ='-DHIDEMYMALLOC' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=n, randbits=31 Libraries: so=so libpth=/lib /usr/lib /usr/ucblib /usr/local/lib libs=-lnsl -ldbm -ldl -lm -lc -lposix libc=/usr/lib/libc.so.1.9 Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-fpic', ccdlflags=' ', lddlflags='-assert nodefinitions' Ok, so far. Make the test database, create the pants2 table, make the necessary mods to test2.pl... and: -------------------------------------------------------------------------- ** Connecting to first instance.... ------------------------------------------------------- MINERVA_DEBUG found. libmsql started with the following:- ------------------------------------------------------- ** Disconnecting from first instance.... ** Connecting to second instance..... **** Trying cursor returning rows as list..... ReturnRow: 1 Test Guy1 ReturnRow: 2 Test Guy2 **** Re-doing the cursor but fetching fields separately..... Id: 1 Name: Test Guy1 Id: 2 Name: Test Guy2 **** Trying an INSERT now....... INSERT present in statement **** Re-doing the cursor but fetching fields separately..... Id: 1 Name: Test Guy1 Id: 2 Name: Test Guy2 Id: 3 Name: Rubbish Guy1 **** Trying a CREATE statement...... CREATE present in statement **** Inserting some rows..... INSERT present in statement **** Re-doing the cursor but fetching fields separately..... Id: 1 Name: Rubbish ** Disconnecting from second instance.... disconnect(DBI::db=HASH(0xe8b88)) invalidates 1 associated cursor(s) at ./test2.pl line 131. -------------------------------------------------------------------------- We clean up a little (drop tables pants2, pants3, and recreate pants2), and with debugging level 2, we find (stderr and stdout seem to be a little mixed, but we expect that): -------------------------------------------------------------------------- DBI dispatch debug level set to 2 >> connect DISPATCH (DBI::dr=HASH(0xb9758) @3 g0 ae5aa0 r1) -> connect for DBD::mSQL::dr (DBI::dr=HASH(0xb9758)~0xe8a14 'ganef' 'test') New DBI::db (for DBD::mSQL::db, parent=DBI::dr=HASH(0xe8a14), id=) dbih_setup_handle(DBI::db=HASH(0xe8954)=>DBI::db=HASH(0xe8af8), DBD::mSQL::db, NULL) New DBI::db => DBI::db=HASH(0xe8954) (inner=DBI::db=HASH(0xe8af8)) for DBD::mSQL::db <- connect= DBI::db=HASH(0xe8954) >> disconnect DISPATCH (DBI::db=HASH(0xe8954) @1 g0 ae4480 r1) -> disconnect for DBD::mSQL::db (DBI::db=HASH(0xe8954)~0xe8af8) <- disconnect= 1 >> connect DISPATCH (DBI::dr=HASH(0xb9758) @3 g0 ae5aa0 r1) -> connect for DBD::mSQL::dr (DBI::dr=HASH(0xb9758)~0xe8a14 'ganef' 'test') New DBI::db (for DBD::mSQL::db, parent=DBI::dr=HASH(0xe8a14), id=) dbih_setup_handle(DBI::db=HASH(0xe8ae0)=>DBI::db=HASH(0xf4e48), DBD::mSQL::db, NULL) New DBI::db => DBI::db=HASH(0xe8ae0) (inner=DBI::db=HASH(0xf4e48)) for DBD::mSQL::db <- connect= DBI::db=HASH(0xe8ae0) >> DESTROY DISPATCH (DBI::db=HASH(0xe8954) @1 g0 a0 r1) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::db=HASH(0xe8af8) @1 g0 a0 r1) -> DESTROY for DBD::mSQL::db (DBI::db=HASH(0xe8af8)) <- DESTROY= 1 dbih_clearcom 0x8ad30 done >> prepare DISPATCH (DBI::db=HASH(0xe8ae0) @2 g0 ae4a28 r1) -> prepare for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48 'SELECT * FROM pants2') New DBI::st (for DBD::mSQL::st, parent=DBI::db=HASH(0xf4e48), id=) dbih_setup_handle(DBI::st=HASH(0xe8ba0)=>DBI::st=HASH(0xe8af8), DBD::mSQL::st, NULL) New DBI::st => DBI::st=HASH(0xe8ba0) (inner=DBI::st=HASH(0xe8af8)) for DBD::mSQL::st <- prepare= DBI::st=HASH(0xe8ba0) >> execute DISPATCH (DBI::st=HASH(0xe8ba0) @1 g0 ae8108 r1) -> execute for DBD::mSQL::st (DBI::st=HASH(0xe8ba0)~0xe8af8) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy1 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX <- execute= 1 >> fetchrow DISPATCH (DBI::st=HASH(0xe8ba0) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xe8ba0)~0xe8af8) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy1 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 0 <- fetchrow= ( '1' 'Test Guy1' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xe8ba0) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xe8ba0)~0xe8af8) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 2 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy2 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 1 <- fetchrow= ( '2' 'Test Guy2' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xe8ba0) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xe8ba0)~0xe8af8) In: DBD::mSQL::dbd_describe() Number of fields: 0 Current ROWID: 2 <- fetchrow= ( ) [0 items] >> finish DISPATCH (DBI::st=HASH(0xe8ba0) @1 g0 ae6cf0 r1) -> finish for DBD::mSQL::st (DBI::st=HASH(0xe8ba0)~0xe8af8) <- finish= 1 >> prepare DISPATCH (DBI::db=HASH(0xe8ae0) @2 g0 ae4a28 r1) -> prepare for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48 'SELECT * from pants2') New DBI::st (for DBD::mSQL::st, parent=DBI::db=HASH(0xf4e48), id=) dbih_setup_handle(DBI::st=HASH(0xe8b04)=>DBI::st=HASH(0xf4f50), DBD::mSQL::st, NULL) New DBI::st => DBI::st=HASH(0xe8b04) (inner=DBI::st=HASH(0xf4f50)) for DBD::mSQL::st <- prepare= DBI::st=HASH(0xe8b04) >> DESTROY DISPATCH (DBI::st=HASH(0xe8ba0) @1 g0 a0 r1) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::st=HASH(0xe8af8) @1 g0 a0 r1) -> DESTROY for DBD::mSQL::st (DBI::st=HASH(0xe8af8)) <- DESTROY= 1 dbih_clearcom 0x8ab10 done >> execute DISPATCH (DBI::st=HASH(0xe8b04) @1 g0 ae8108 r1) -> execute for DBD::mSQL::st (DBI::st=HASH(0xe8b04)~0xf4f50) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy1 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX <- execute= 1 >> fetchrow DISPATCH (DBI::st=HASH(0xe8b04) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xe8b04)~0xf4f50) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy1 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 0 <- fetchrow= ( '1' 'Test Guy1' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xe8b04) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xe8b04)~0xf4f50) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 2 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy2 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 1 <- fetchrow= ( '2' 'Test Guy2' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xe8b04) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xe8b04)~0xf4f50) In: DBD::mSQL::dbd_describe() Number of fields: 0 Current ROWID: 2 <- fetchrow= ( ) [0 items] >> finish DISPATCH (DBI::st=HASH(0xe8b04) @1 g0 ae6cf0 r1) -> finish for DBD::mSQL::st (DBI::st=HASH(0xe8b04)~0xf4f50) <- finish= 1 >> prepare DISPATCH (DBI::db=HASH(0xe8ae0) @2 g0 ae4a28 r1) -> prepare for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48 'INSERT INTO pants2 VALUES ( 3, 'Rubbish Guy1' )') New DBI::st (for DBD::mSQL::st, parent=DBI::db=HASH(0xf4e48), id=) dbih_setup_handle(DBI::st=HASH(0xf4f8c)=>DBI::st=HASH(0xb6cb4), DBD::mSQL::st, NULL) New DBI::st => DBI::st=HASH(0xf4f8c) (inner=DBI::st=HASH(0xb6cb4)) for DBD::mSQL::st <- prepare= DBI::st=HASH(0xf4f8c) >> DESTROY DISPATCH (DBI::st=HASH(0xe8b04) @1 g0 a0 r1) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::st=HASH(0xf4f50) @1 g0 a0 r1) -> DESTROY for DBD::mSQL::st (DBI::st=HASH(0xf4f50)) <- DESTROY= 1 dbih_clearcom 0xf2738 done >> execute DISPATCH (DBI::st=HASH(0xf4f8c) @1 g0 ae8108 r1) -> execute for DBD::mSQL::st (DBI::st=HASH(0xf4f8c)~0xb6cb4) In: DBD::mSQL::dbd_describe() <- execute= 1 >> finish DISPATCH (DBI::st=HASH(0xf4f8c) @1 g0 ae6cf0 r1) -> finish for DBD::mSQL::st (DBI::st=HASH(0xf4f8c)~0xb6cb4) <- finish= 1 >> prepare DISPATCH (DBI::db=HASH(0xe8ae0) @2 g0 ae4a28 r1) -> prepare for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48 'SELECT * from pants2') New DBI::st (for DBD::mSQL::st, parent=DBI::db=HASH(0xf4e48), id=) dbih_setup_handle(DBI::st=HASH(0xf5010)=>DBI::st=HASH(0xf4fc8), DBD::mSQL::st, NULL) New DBI::st => DBI::st=HASH(0xf5010) (inner=DBI::st=HASH(0xf4fc8)) for DBD::mSQL::st <- prepare= DBI::st=HASH(0xf5010) >> DESTROY DISPATCH (DBI::st=HASH(0xf4f8c) @1 g0 a0 r1) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::st=HASH(0xb6cb4) @1 g0 a0 r1) -> DESTROY for DBD::mSQL::st (DBI::st=HASH(0xb6cb4)) <- DESTROY= 1 dbih_clearcom 0xf2e48 done >> execute DISPATCH (DBI::st=HASH(0xf5010) @1 g0 ae8108 r1) -> execute for DBD::mSQL::st (DBI::st=HASH(0xf5010)~0xf4fc8) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy1 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX <- execute= 1 >> fetchrow DISPATCH (DBI::st=HASH(0xf5010) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xf5010)~0xf4fc8) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy1 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 0 <- fetchrow= ( '1' 'Test Guy1' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xf5010) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xf5010)~0xf4fc8) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 2 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Test Guy2 9 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 1 <- fetchrow= ( '2' 'Test Guy2' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xf5010) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xf5010)~0xf4fc8) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 3 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Rubbish Guy1 12 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 2 <- fetchrow= ( '3' 'Rubbish Guy1' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xf5010) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xf5010)~0xf4fc8) In: DBD::mSQL::dbd_describe() Number of fields: 0 Current ROWID: 3 <- fetchrow= ( ) [0 items] >> finish DISPATCH (DBI::st=HASH(0xf5010) @1 g0 ae6cf0 r1) -> finish for DBD::mSQL::st (DBI::st=HASH(0xf5010)~0xf4fc8) <- finish= 1 >> prepare DISPATCH (DBI::db=HASH(0xe8ae0) @2 g0 ae4a28 r1) -> prepare for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48 'CREATE TABLE pants3 ( id int, name char(64) )') New DBI::st (for DBD::mSQL::st, parent=DBI::db=HASH(0xf4e48), id=) dbih_setup_handle(DBI::st=HASH(0xf4f74)=>DBI::st=HASH(0xf5028), DBD::mSQL::st, NULL) New DBI::st => DBI::st=HASH(0xf4f74) (inner=DBI::st=HASH(0xf5028)) for DBD::mSQL::st <- prepare= DBI::st=HASH(0xf4f74) >> DESTROY DISPATCH (DBI::st=HASH(0xf5010) @1 g0 a0 r1) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::st=HASH(0xf4fc8) @1 g0 a0 r1) -> DESTROY for DBD::mSQL::st (DBI::st=HASH(0xf4fc8)) <- DESTROY= 1 dbih_clearcom 0xf24e8 done >> execute DISPATCH (DBI::st=HASH(0xf4f74) @1 g0 ae8108 r1) -> execute for DBD::mSQL::st (DBI::st=HASH(0xf4f74)~0xf5028) In: DBD::mSQL::dbd_describe() <- execute= 1 >> finish DISPATCH (DBI::st=HASH(0xf4f74) @1 g0 ae6cf0 r1) -> finish for DBD::mSQL::st (DBI::st=HASH(0xf4f74)~0xf5028) <- finish= 1 >> prepare DISPATCH (DBI::db=HASH(0xe8ae0) @2 g0 ae4a28 r1) -> prepare for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48 'INSERT INTO pants3 VALUES ( 1, 'Rubbish' )') New DBI::st (for DBD::mSQL::st, parent=DBI::db=HASH(0xf4e48), id=) dbih_setup_handle(DBI::st=HASH(0xf5088)=>DBI::st=HASH(0xf4fbc), DBD::mSQL::st, NULL) New DBI::st => DBI::st=HASH(0xf5088) (inner=DBI::st=HASH(0xf4fbc)) for DBD::mSQL::st <- prepare= DBI::st=HASH(0xf5088) >> DESTROY DISPATCH (DBI::st=HASH(0xf4f74) @1 g0 a0 r1) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::st=HASH(0xf5028) @1 g0 a0 r1) -> DESTROY for DBD::mSQL::st (DBI::st=HASH(0xf5028)) <- DESTROY= 1 dbih_clearcom 0xf3440 done >> execute DISPATCH (DBI::st=HASH(0xf5088) @1 g0 ae8108 r1) -> execute for DBD::mSQL::st (DBI::st=HASH(0xf5088)~0xf4fbc) In: DBD::mSQL::dbd_describe() <- execute= 1 >> finish DISPATCH (DBI::st=HASH(0xf5088) @1 g0 ae6cf0 r1) -> finish for DBD::mSQL::st (DBI::st=HASH(0xf5088)~0xf4fbc) <- finish= 1 >> prepare DISPATCH (DBI::db=HASH(0xe8ae0) @2 g0 ae4a28 r1) -> prepare for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48 'SELECT * from pants3') New DBI::st (for DBD::mSQL::st, parent=DBI::db=HASH(0xf4e48), id=) dbih_setup_handle(DBI::st=HASH(0xf4f80)=>DBI::st=HASH(0xf50c4), DBD::mSQL::st, NULL) New DBI::st => DBI::st=HASH(0xf4f80) (inner=DBI::st=HASH(0xf50c4)) for DBD::mSQL::st <- prepare= DBI::st=HASH(0xf4f80) >> DESTROY DISPATCH (DBI::st=HASH(0xf5088) @1 g0 a0 r1) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::st=HASH(0xf4fbc) @1 g0 a0 r1) -> DESTROY for DBD::mSQL::st (DBI::st=HASH(0xf4fbc)) <- DESTROY= 1 dbih_clearcom 0xf25a8 done >> execute DISPATCH (DBI::st=HASH(0xf4f80) @1 g0 ae8108 r1) -> execute for DBD::mSQL::st (DBI::st=HASH(0xf4f80)~0xf50c4) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Rubbish 7 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX <- execute= 1 >> fetchrow DISPATCH (DBI::st=HASH(0xf4f80) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xf4f80)~0xf50c4) In: DBD::mSQL::dbd_describe() In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'Fetching Field In: DBD::mSQL::dbd_describe'LinkRow: 0 Name: 1 1 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX In: DBD::mSQL::dbd_describe'LinkRow: 1 Name: Rubbish 7 In: DBD::mSQL::dbd_describe'newSV In: DBD::mSQL::dbd_describe'SvUPGRADE In: DBD::mSQL::dbd_describe'SvREADONLY_ON In: DBD::mSQL::dbd_describe'SvPOK_only In: DBD::mSQL::dbd_describe'SvPVX Number of fields: 2 Current ROWID: 0 <- fetchrow= ( '1' 'Rubbish' ) [2 items] >> fetchrow DISPATCH (DBI::st=HASH(0xf4f80) @1 g1 a0 r1) -> fetchrow for DBD::mSQL::st (DBI::st=HASH(0xf4f80)~0xf50c4) In: DBD::mSQL::dbd_describe() Number of fields: 0 Current ROWID: 1 <- fetchrow= ( ) [0 items] >> finish DISPATCH (DBI::st=HASH(0xf4f80) @1 g0 ae6cf0 r1) -> finish for DBD::mSQL::st (DBI::st=HASH(0xf4f80)~0xf50c4) <- finish= 1 >> disconnect DISPATCH (DBI::db=HASH(0xe8ae0) @1 g0 ae4480 r1) -> disconnect for DBD::mSQL::db (DBI::db=HASH(0xe8ae0)~0xf4e48) disconnect(DBI::db=HASH(0xf4e48)) invalidates 1 associated cursor(s) at ./test2.pl line 131. <- disconnect= 1 DBI::END DBI::disconnect_all DBI DBI::disconnect_all for 'mSQL' >> disconnect_all DISPATCH (DBI::dr=HASH(0xb9758) @1 g0 ad02f8 r2) -> disconnect_all for DBD::mSQL::dr (DBI::dr=HASH(0xb9758)~0xe8a14) <- disconnect_all= 1 DBI::END complete >> DESTROY DISPATCH (DBI::st=HASH(0xf50c4) @1 g0 a0 r0) -> DESTROY for DBD::mSQL::st (DBI::st=HASH(0xf50c4)) <- DESTROY= 1 dbih_clearcom 0xf3580 done >> DESTROY DISPATCH (DBI::dr=HASH(0xe89d8) @1 g0 a0 r0) dbih_clearcom 0x93b00 done >> DESTROY DISPATCH (DBI::db=HASH(0xf4e48) @1 g0 a0 r0) -> DESTROY for DBD::mSQL::db (DBI::db=HASH(0xf4e48)) <- DESTROY= 1 >> DESTROY DISPATCH (DBI::dr=HASH(0xe8a14) @1 g0 a0 r0) dbih_clearcom 0x8a970 done dbih_clearcom 0xf5480 done >> DESTROY DISPATCH (DBI::db=HASH(0xe8ae0) @1 g0 a0 r0) (inner handle already deleted) >> DESTROY DISPATCH (DBI::dr=HASH(0xe89b4) @1 g0 a0 r0) (inner handle already deleted) >> DESTROY DISPATCH (DBI::st=HASH(0xf4f80) @1 g0 a0 r0) (inner handle already deleted) >> DESTROY DISPATCH (DBI::dr=HASH(0xb9758) @1 g0 a0 r0) (inner handle already deleted) Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. ** Connecting to first instance.... ------------------------------------------------------- MINERVA_DEBUG found. libmsql started with the following:- ------------------------------------------------------- ** Disconnecting from first instance.... imp_dbh->sock: 3 ** Connecting to second instance..... **** Trying cursor returning rows as list..... 2 rows matched DBD::mSQL::dbd_db_prepare'imp_sth->fbh_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy1 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 0 In: DBD::mSQL::fetchrow'imp_sth->row_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy1 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Test Guy1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 9 ReturnRow: 1 Test Guy1 In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 1 In: DBD::mSQL::fetchrow'imp_sth->row_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 2 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy2 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 2 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Test Guy2 In: DBD::mSQL::execute'FieldBufferDump->rlen: 9 ReturnRow: 2 Test Guy2 In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 2 In: DBD::mSQL::fetchrow'imp_sth->row_num: 2 **** Re-doing the cursor but fetching fields separately..... 2 rows matched DBD::mSQL::dbd_db_prepare'imp_sth->fbh_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy1 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 0 In: DBD::mSQL::fetchrow'imp_sth->row_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy1 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Test Guy1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 9 Id: 1 Name: Test Guy1 In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 1 In: DBD::mSQL::fetchrow'imp_sth->row_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 2 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy2 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 2 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Test Guy2 In: DBD::mSQL::execute'FieldBufferDump->rlen: 9 Id: 2 Name: Test Guy2 In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 2 In: DBD::mSQL::fetchrow'imp_sth->row_num: 2 **** Trying an INSERT now....... INSERT present in statement **** Re-doing the cursor but fetching fields separately..... 3 rows matched DBD::mSQL::dbd_db_prepare'imp_sth->fbh_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy1 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 0 In: DBD::mSQL::fetchrow'imp_sth->row_num: 3 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy1 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Test Guy1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 9 Id: 1 Name: Test Guy1 In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 1 In: DBD::mSQL::fetchrow'imp_sth->row_num: 3 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 2 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Test Guy2 9 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 2 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Test Guy2 In: DBD::mSQL::execute'FieldBufferDump->rlen: 9 Id: 2 Name: Test Guy2 In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 2 In: DBD::mSQL::fetchrow'imp_sth->row_num: 3 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 3 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Rubbish Guy1 12 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 3 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Rubbish Guy1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 12 Id: 3 Name: Rubbish Guy1 In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 3 In: DBD::mSQL::fetchrow'imp_sth->row_num: 3 **** Trying a CREATE statement...... CREATE present in statement **** Inserting some rows..... INSERT present in statement **** Re-doing the cursor but fetching fields separately..... 1 rows matched DBD::mSQL::dbd_db_prepare'imp_sth->fbh_num: 2 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Rubbish 7 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 0 In: DBD::mSQL::fetchrow'imp_sth->row_num: 1 Entering imp_sth->fbh test cycle In: DBD::mSQL::dbd_describe'FBHDump[0]: 1 1 In: DBD::mSQL::dbd_describe'FBHDump[1]: Rubbish 7 Out: DBD::mSQL::dbd_describe() In: DBD::mSQL::execute'FieldBufferDump: 0 In: DBD::mSQL::execute'FieldBufferDump->cbuf: 1 In: DBD::mSQL::execute'FieldBufferDump->rlen: 1 In: DBD::mSQL::execute'FieldBufferDump: 1 In: DBD::mSQL::execute'FieldBufferDump->cbuf: Rubbish In: DBD::mSQL::execute'FieldBufferDump->rlen: 7 Id: 1 Name: Rubbish In: DBD::mSQL::fetchrow In: DBD::mSQL::fetchrow'imp_sth->currow: 1 In: DBD::mSQL::fetchrow'imp_sth->row_num: 1 ** Disconnecting from second instance.... imp_dbh->sock: 3 -------------------------------------------------------------------------- ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-17@lion>; Thu, 7 Sep 1995 07:24:53 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 01:32:22 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa26891; 7 Sep 95 2:31 +0100 Received: by gw.home.vix.com id AA20988; Wed, 6 Sep 95 13:08:15 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20982; Wed, 6 Sep 95 13:08:14 -0700 Message-Id: <9509062008.AA20982@gw.home.vix.com> Received: from nicoh (burundi.nicoh.com) by africa.nicoh.com with SMTP (1.37.109.15/16.2) id AA245958483; Wed, 6 Sep 1995 14:14:43 -0600 Date: Wed, 6 Sep 1995 14:14:43 -0600 X-Sender: mcclmatt@nicoh.com X-Mailer: Windows Eudora Version 2.0.3 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: perldb-interest@vix.com From: Matt McClung Subject: unsubscribe -------- Matt McClung electronic mail : mcclmatt@nicoh.com NICOH Net voice mail : (208) 233-5802 Internet Access in Idaho! snail mail : 200 S. Main, Suite O, Pocatello ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-22@lion>; Thu, 7 Sep 1995 07:34:57 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 02:14:21 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa15714; 7 Sep 95 3:13 +0100 Received: by gw.home.vix.com id AA17189; Wed, 6 Sep 95 12:24:59 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA17183; Wed, 6 Sep 95 12:24:57 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa01581; 6 Sep 95 19:55 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa24100; 6 Sep 95 19:52 +0100 Received: from ig.co.uk by lion id <24162-0@lion>; Wed, 6 Sep 1995 19:55:20 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: hash refs as parameters to bind_param Date: Wed, 6 Sep 1995 19:55:20 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509061952.aa24100@post.demon.co.uk> > From: mhm@austin.ibm.com > > What is the correct way to create the %attribs hash ref for bind_param? > > $sth->bind_param($param_index_or_name,\$value[,\%attribs]) > > Whatever combination of things I try I get the following from > dbd_bind_ph: > > DBI::st=HASH(0x200d5fc8)->bind_param(:p1): invalid attribute parameter at genreq.pl line 15. > $sth->bind_param(1, $foo, { attribute_name => $value }); If used in often (eg in a loop) do: $attribs1 = { attribute_name => $value }; while( ... ) { $sth->bind_param(1, $foo, $attribs1); } > Thanks. > Mike Moran Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-36@lion>; Thu, 7 Sep 1995 07:37:00 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 02:54:22 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa02999; 7 Sep 95 3:53 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <29699-0@frigate.doc.ic.ac.uk>; Wed, 6 Sep 1995 22:23:40 +0100 Received: by gw.home.vix.com id AA12814; Wed, 6 Sep 95 05:01:14 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA12794; Wed, 6 Sep 95 05:00:29 -0700 Received: by ki1.chemie.fu-berlin.de (Smail3.1.28.1) from mail.hanse.de (134.100.239.2) with smtp id ; Wed, 6 Sep 95 10:35 MEST Received: from by mail.hanse.de with bsmtp for id ; Wed, 6 Sep 95 09:58 MET DST Received: by mwhh.Hanse.DE (Smail3.1.28.1 #20) id m0sqFew-000HyhC; Wed, 6 Sep 95 09:16 GMT+0100 Received: (from georg@localhost) by deerwood.hanse.de (8.6.11/8.6.11) id HAA01233; Wed, 6 Sep 1995 07:24:22 +0200 Date: Wed, 6 Sep 1995 07:24:22 +0200 (MET DST) From: Georg Rehfeld To: Tim Bunce Cc: fon@iinet.com.au, perldb-interest@vix.com Subject: Re: URGENT! I need a database In-Reply-To: <9509041718.ah11340@post.demon.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi Tim, Adrian, Surmani, Alligator and all others, oops, I did'nt forget to note you (the perldb-interest-list), did I? > > From: Surmani > > [...] > > > > I've been using mSqlPerl on Linux, but now find it too limited for > > my application (a web search engine). > > > > Please give me some advice on which database I should use. Commercial is > > OK, but I would prefer to run it on Linux with a perl interface (DBI ?). > > If it does SQL then even better. > > > > Please send me an URGENT response on this, > > to the email address below: > > > > Adrian Smith > > fon@iinet.com.au > > > I've heard that Oracle 7 runs on Linux (but it's unsupported). > > Since I'm not familiar with Linux I'll copy to the list for others to > make suggestions (and no doubt quiz you about why mSQL is too limited > for your application :-) > > Tim. > Yes, the SCO version of Oracle 7 runs nicely, fine, fast and stable on Linux! If you can do without Pro*C on the Linux platform (and I know, most of us can't)! It runs, using the iBCS emulator stuff, as Alligator already mentioned, and that is a runtime translation extension to Linux, recognising the foreign origin of the executables at load time and then translating each foreign kernel call to the corresponding Linux kernel call. And the iBCS guys have done a good job (as have the Oracle programmers). All utilities work, sql*plus, sql*dba, exp, imp etc. Networking via sql*net is possible including distributed databasing, I accessed the database from other UNIX systems (ISC, AIX), and from ms-dos/windows via native sql*net and via ODBC. Performance is good (about 80 inserts/second with an oraperl test program on a DX2/66 MHz cpu with only 16MB main mem). Remaining problems are: - the totally lacking support from Oracle - the unability to use pro*c because of the SCO format Oracle libs that can't be used to link Linux objects. - the price of approx. $1500 for a 5 user license for the SCO Workgroup Server software (after the free 90 days evaluation period; have a look at http//www.oracle.com for download or order of a cdrom). If any of you cares: I'll mail an installation 'cookbook' for the SCO Oracle full product cdrom on request; it seems to work with the Workgroup server cdrom too. Please don't forget to reply only to me, not to the list, if you use this mail to get the installation notes. Best regards Georg ___ ___ | + | |__ ' Georg Rehfeld, D-20535 Hamburg, Jordanstr. 8, [49] (40) 2518356 |_|_\ |___, georg@deerwood.hanse.de ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-37@lion>; Thu, 7 Sep 1995 07:37:09 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 02:54:23 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ab02999; 7 Sep 95 3:53 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <29726-0@frigate.doc.ic.ac.uk>; Wed, 6 Sep 1995 22:25:11 +0100 Received: by gw.home.vix.com id AA22461; Wed, 6 Sep 95 07:20:59 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA22452; Wed, 6 Sep 95 07:20:55 -0700 Message-Id: <9509061420.AA22452@gw.home.vix.com> Received: by hpunix.gsfc.state.ga.us (1.38.193.5/16.2) id AA21633; Wed, 6 Sep 1995 10:23:23 -0400 From: "Marc A. Mnich" Subject: subscribe To: perldb-interest@vix.com Date: Wed, 6 Sep 95 10:23:23 EDT Mailer: Elm [revision: 70.85.2.1] subscribe -- /\ /\ _ _ _ Marc A. Mnich (404) 248-1266 / / / (/ / ( Georgia Student Finance Commission marcm@hpunix.gsfc.state.ga.us ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-53@lion>; Thu, 7 Sep 1995 07:39:12 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 03:14:49 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa11356; 7 Sep 95 4:14 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <05311-0@frigate.doc.ic.ac.uk>; Thu, 7 Sep 1995 03:56:32 +0100 Received: by gw.home.vix.com id AA17249; Wed, 6 Sep 95 12:25:49 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA17244; Wed, 6 Sep 95 12:25:48 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa01299; 6 Sep 95 19:51 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa23158; 6 Sep 95 19:48 +0100 Received: from ig.co.uk by lion id <23989-0@lion>; Wed, 6 Sep 1995 19:51:06 +0100 To: descarte@mcqueen.com, andreas.koenig@mind.de Subject: Re: ANNOUNCE: DBD::mSQL-0.60pl1 released Cc: perldb-interest@vix.com Date: Wed, 6 Sep 1995 19:51:06 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509061948.aa23158@post.demon.co.uk> > From: Andreas Koenig > > > ftp://ftp.mcqueen.com/pub/languages/perl5/DBPerl/ > > > >Let me know how things go. > > That's how I patched it: > > - SV *valuesv = dbd_db_FETCH(dbh, keysv); > + SV *valuesv = (SV*)dbd_db_FETCH(dbh, keysv); Casting is an evil which the DBI and DBDs go to some lengths to avoid. The dbd_db_FETCH function (etc) should have been declared to return the correct type (see *.h in DBD::Oracle). Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-56@lion>; Thu, 7 Sep 1995 07:39:40 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 03:33:53 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa19287; 7 Sep 95 4:33 +0100 Received: by gw.home.vix.com id AA20499; Wed, 6 Sep 95 13:00:49 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20494; Wed, 6 Sep 95 13:00:48 -0700 Received: by intfw.bear.com (4.1/SMI-4.1) id AA01971; Wed, 6 Sep 95 16:00:03 EDT Received: from fastbear(165.168.74.3) by intfw via smap (V1.3) id sma001903; Wed Sep 6 15:57:05 1995 Received: from ursa2.bear.com by fastbear.bear.com (4.1/SMI-4.1/1.0 AMR 12/15/94) id AA08753; Wed, 6 Sep 95 15:58:58 EDT Received: from bear.com (ursa) by ursa2.bear.com (4.1/SMI-4.1/JMD+AMR+DJS) id AA07991; Wed, 6 Sep 95 15:56:52 EDT Received: from jaguar.bears by bear.com (4.1/SMI-4.0/JMD+AR+DJS) id AA03502; Wed, 6 Sep 95 15:56:51 EDT Received: by jaguar.bears (4.1/SMI-4.1) id AA13588; Wed, 6 Sep 95 15:55:49 EDT Date: Wed, 6 Sep 95 15:55:49 EDT From: Buzz Moschetti Message-Id: <9509061955.AA13588@jaguar.bears> To: perldb-interest@vix.com Subject: Interperl users speak out! Reply-To: Buzz Moschetti Is anyone using Interperl? Is there any real benefit to creating DBI:Interbase? Users at my organization need not reply. -- ******************************************************************************* Bear Stearns is not responsible for any recommendation, solicitation, offer or agreement or any information about any transaction, customer account or account activity contained in this communication. ******************************************************************************* ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-86@lion>; Thu, 7 Sep 1995 07:43:34 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 05:43:27 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa18947; 7 Sep 95 6:42 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa00861; 6 Sep 95 20:35 GMT-60:00 Received: by gw.home.vix.com id AA02089; Wed, 6 Sep 95 02:19:42 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA02084; Wed, 6 Sep 95 02:19:35 -0700 Received: from timetable.brunel.ac.uk by sirius.brunel.ac.uk with SMTP (PP) id <07045-0@sirius.brunel.ac.uk>; Wed, 6 Sep 1995 10:19:19 +0100 From: Timetabling@brunel.ac.uk Message-Id: <3645.9509060919@timetable.brunel.ac.uk> Subject: Oracle7.1.6 + Perl5.001m + Solaris2.4 To: perldb-interest@vix.com Date: Wed, 6 Sep 1995 10:19:09 +0100 (BST) Cc: Tim.Bunce@ig.co.uk X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Having problems with compiling the Oracle.c file... get the following errors: ---- gcc -c -O -fpic -I/timetable/timetable1/usr/lib/perl5/sun4-solaris/CORE -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion -Wtraditional -Wpointer-arith -Wbad-function-cast -Wcast-qual -I/timetable/timetable2/oracle7.1.6/rdbms/demo -I/timetable/timetable1/usr/lib/perl5/sun4-solaris/DBI Oracle.c Oracle.c: In function `XS_DBD__Oracle__db_disconnect': Oracle.c:206: warning: int format, I32 arg (arg 3) Oracle.c: In function `XS_DBD__Oracle__db_DESTROY': Oracle.c:244: warning: implicit declaration of function `DBIc_ENDING' Oracle.c: In function `XS_DBD__Oracle__st__prepare': Oracle.c:271: too few arguments to function `dbd_st_prepare' Oracle.c: In function `XS_DBD__Oracle__st_execute': Oracle.c:323: warning: passing arg 2 of `dbd_bind_ph' from incompatible pointer type Oracle.c:323: warning: passing arg 3 of `dbd_bind_ph' from incompatible pointer type Oracle.c: In function `XS_DBD__Oracle__st_fetchrow': Oracle.c:392: structure has no member named `fbh_num' Oracle.c:426: structure has no member named `fbh_num' Oracle.c:426: structure has no member named `fbh_num' Oracle.c:428: structure has no member named `fbh_num' Oracle.c:435: warning: implicit declaration of function `dbtype_is_long' Oracle.c: In function `XS_DBD__Oracle__st_readblob': Oracle.c:503: too few arguments to function `dbd_st_readblob' Oracle.c:503: warning: assignment makes pointer from integer without a cast *** Error code 1 make: Fatal error: Command failed for target `Oracle.o' --- Any ideas? Oracle 7.1.6, Perl 5.001m and Solaris 2.4 using gcc-2.6.3. [ Brunel University Timetable Unit ] [ timetabling@brunel.ac.uk ] [Telephone: 01895 274000 extension 2104] ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <27188-88@lion>; Thu, 7 Sep 1995 07:43:47 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 05:45:47 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa20070; 7 Sep 95 6:45 +0100 Received: by gw.home.vix.com id AA10777; Wed, 6 Sep 95 17:20:38 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10735; Wed, 6 Sep 95 17:20:31 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa10929; 6 Sep 95 22:03 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa19229; 6 Sep 95 22:01 +0100 Received: from ig.co.uk by lion id <24560-0@lion>; Wed, 6 Sep 1995 21:23:03 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Building the %attribs for bind_param ? Date: Wed, 6 Sep 1995 21:23:03 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509062201.aa19229@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > > > > > From: mhm@austin.ibm.com > > > To: Tim.Bounce@ig.co.uk > > > > > > The question boils down to what is the best way to build the %attribs > > > hash-ref for bind_param and bind-col. Given the number of possible > > > things that a DB2 user might want to bind to a parameter or in binding > > > the incoming data to a host variable, I think I need to supply a > > > function which builds and returns the %attribs hash-ref. > > > > > > Would it be better to define it as an XSUB routine or in the pm file > > > under DBD::DB2::st? I currently leaning towards doing it in perl in > > > the pm file. > > > > > > Any comments, anything else I need to consider? > > > > > I don't see any need for this. Could you give me some examples of the > > 'number of possible things that a DB2 user might want to bind to ...' > > > > Well, let us see: > Umm, > 1. Specify the incoming host variable type (16 options). > 2. Specify the SQL variable type (25 options) Over 90% of these should be automatic and transparent. Generally the driver asks the database interface what the type of the variable is and then acts accordingly. Basically perl scalars are strings (PV), numbers (NV) or ints (IV). NVs and IV are easy, everything else gets bound as a string and you let the database interface do any required conversions (eg to dates). > 3. Specify precision for numeric, Just use the precision of the variable itself (the database interface should tell you this). > character, and blob types. (Precision for character and blob types? Eh?) > 4. Specify scale for numeric and time stamp types Same as precision. > 5. Specify whether the statement gets evaluated as an input, > input deferred (stored proceedure), output or output deferred. Assuming I understand you 'deferred' is not safe since the perl variable you bind to may mutate between the bind and the execute. Input vs Output is an unresolved topic right now. > 6. Specify truncation levels for character and binary types As for precision, I think (if I understand you correctly). > 7. Specify either a host variable to bound or special value to > a parameter. (DB2 allows the posting of several special "variables" > which cause the database to perform a special function, yet typically > the only result is a positive return code) > _This_ sounds like it would need an attribute. > Of the above list, I can guarantee that I need to support the first > 4 on every statement executed. Item 5, I can defer until I want/need > to support stored procedures. As for items 6&7, what I am doing > currently has no need, but I would guess that folks attempting to > use DB2perl on existing databases will probably want them fairly quickly. > You should not need to specify 1..4 or 6 most of the time. Is this related to DB2 not being able to describe variables for you? If so you have my sympathies and, yes, you would need to manually specify, somewhere, any required information which DB2 is unable to tell you itself. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <29082-11@lion>; Thu, 7 Sep 1995 12:07:09 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 08:22:48 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa12600; 7 Sep 95 9:22 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab19334; 6 Sep 95 23:44 GMT-60:00 Received: by gw.home.vix.com id AA28375; Wed, 6 Sep 95 08:38:41 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA28367; Wed, 6 Sep 95 08:38:38 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa11869; 6 Sep 95 15:29 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab26609; 6 Sep 95 15:27 +0100 Received: from ig.co.uk by lion id <19620-0@lion>; Wed, 6 Sep 1995 15:12:57 +0100 To: perldb-interest@vix.com, Timetabling@brunel.ac.uk Subject: Re: Oracle7.1.6 + Perl5.001m + Solaris2.4 Date: Wed, 6 Sep 1995 15:12:57 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509061527.ab26609@post.demon.co.uk> > From: Timetabling@brunel.ac.uk > > Having problems with compiling the Oracle.c file... get the following errors: > > Oracle.c: In function `XS_DBD__Oracle__db_DESTROY': > Oracle.c:244: warning: implicit declaration of function `DBIc_ENDING' > Oracle.c: In function `XS_DBD__Oracle__st__prepare': > Oracle.c:271: too few arguments to function `dbd_st_prepare' > Oracle.c: In function `XS_DBD__Oracle__st_fetchrow': > Oracle.c:392: structure has no member named `fbh_num' > > Any ideas? > > Oracle 7.1.6, Perl 5.001m and Solaris 2.4 using gcc-2.6.3. > What you haven't listed are the version numbers of the DBI and DBD::Oracle modules. It looks like your DBI module is out of date. Get the latest of both from ftp.demon.co.uk:/pub/perl/db/... Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03328-0@lion>; Thu, 7 Sep 1995 15:08:26 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 11:34:41 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa09382; 7 Sep 95 12:34 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <11702-0@frigate.doc.ic.ac.uk>; Thu, 7 Sep 1995 12:03:15 +0100 Received: by gw.home.vix.com id AA19243; Wed, 6 Sep 95 19:11:45 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA19237; Wed, 6 Sep 95 19:11:44 -0700 Received: by thor.nh.aspectdv.com (4.1/SMI-4.1) id AA19151; Wed, 6 Sep 95 22:12:26 EDT Date: Wed, 6 Sep 95 22:12:26 EDT From: Michael L'Italien Message-Id: <9509070212.AA19151@thor.nh.aspectdv.com> To: perldb-interest@vix.com Subject: unsubscribe ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03328-8@lion>; Thu, 7 Sep 1995 15:09:10 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 12:20:28 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa01176; 7 Sep 95 13:19 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id aa22037; 7 Sep 95 0:11 GMT-60:00 Received: by gw.home.vix.com id AA12035; Wed, 6 Sep 95 11:23:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA12030; Wed, 6 Sep 95 11:23:23 -0700 Received: (thorn@localhost) by amaretto.cc.mcgill.ca (8.6.10/8.6.5) id OAA09936; Wed, 6 Sep 1995 14:23:12 -0400 Date: Wed, 6 Sep 1995 14:23:11 -0400 (EDT) From: Ron Hall To: Andrew G Wood Cc: perldb-interest@vix.com Subject: Re: How does one do this....? In-Reply-To: <6054.199509060805@bssmlsa.nerc-bas.ac.uk.nerc-bas.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Wed, 6 Sep 1995, Andrew G Wood wrote: > >>>>> "Ron" == Ron Hall writes: > > > Using oraperl2.4 Oracle 7.blah blah blah how does one execute a > > stired procedure? Just Curious? > > I think you will need to get hold of the two new extensions that I have > been developing over the past months .... SpoOn and WooDen .... soo your > Perl will look somethings like: > > use WooDen; > use SpoOn; Oh you wags.....I meant stored and besides stirred has 2 rs :) > Ron Hall | DISCLAIMER: I said it. I must've meant it. There is Analyst | nobody else to blame. | Clown:n: A man who acts too natural. +1 514 398 3718 | "Whenever doubt can be called into question, do so." ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09812-11@lion>; Fri, 8 Sep 1995 07:25:04 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 21:26:16 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa26528; 7 Sep 95 22:25 +0100 Received: by gw.home.vix.com id AA17590; Thu, 7 Sep 95 09:28:48 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA17580; Thu, 7 Sep 95 09:28:42 -0700 Received: from lazarus.nrtc.northrop.com by nrtc.nrtc.northrop.com id aa28583; 7 Sep 95 7:57 PDT Received: from world.nad.northrop.com by lazarus.nrtc.northrop.com (15.4a/15.6.b) id AA26586; Thu, 7 Sep 95 09:23:13 pdt Received: by world.nad.northrop.com (4.1/SMI-4.1.1) id AA26967; Thu, 7 Sep 95 09:20:01 PDT Date: Thu, 7 Sep 1995 09:20:00 -0700 (PDT) From: "Charles G. Margolin" To: mhm@austin.ibm.com Cc: Tim Bunce , perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: <9509071234.AA06790@dax.austin.ibm.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 7 Sep 1995 mhm@austin.ibm.com wrote: > Exactly, none of the information about the what the database thinks > the parameters should be is available from the database. The enduser > is required to specify all of following things to bind a variable to > a parameter. > > 1. Column type - INPUT/OUTPUT, deferred/nondeferred > 2. Incoming C type for the variable. > 3. Expected SQL type. > 4. Precision of the variable > 5. Scale of the variable > 6. Pointer to variable - containing parameter value > 7. Max size of parameter variable > 8. Pointer to variable containing actual size of parameter value > Maybe this is way off base, since I have never worked with the Unix version of DB2, but in ancient times I used to work with mainframe DB2 and I did some dynamic SQL coding in various languages. With mainframe DB2 I could do a PREPARE and a DESCRIBE. Then I would use the info that DESCRIBE stores into the SQLDA block about data requirements to complete the SQLDA by supplying the addresses. I could even cause some type conversions by changing the data type. Does DB2/Unix lack this feature, or am I missing something important? Charles Margolin email: cmargoli@world.nad.northrop.com Northrop Grumman Corporation phone: (310)332-6011 One Northrop Ave. Orgn. 624/23 Hawthorne, CA 90250-3277 ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09549-21@lion>; Fri, 8 Sep 1995 07:25:15 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 21:26:19 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa26139; 7 Sep 95 22:24 +0100 Received: by gw.home.vix.com id AA23025; Thu, 7 Sep 95 10:35:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA23018; Thu, 7 Sep 95 10:35:44 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id MAA38942; Thu, 7 Sep 1995 12:35:43 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA18276; Thu, 7 Sep 1995 12:35:38 -0500 From: mhm@austin.ibm.com Message-Id: <9509071735.AA18276@dax.austin.ibm.com> To: "Charles G. Margolin" Cc: perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: Your message of Thu, 07 Sep 95 09:50:53 -0700. Date: Thu, 07 Sep 95 12:35:38 -0500 "Charles G. Margolin" writes: > > Here's a kludge: Perhaps you could code a routine using the DB2 > precompiler which would receive an arbitrary SQL statement, > PREPARE and DESCRIBE it, and return the resulting SQLDA structure. > Then use that SQLDA for the info you need in your CLI-based routines. > Thought about it but there are two major problems: 1) The big problem is that it requires each machine to have the precompiler installed and available to the enduser. Not always the case in production shop. 2) While CLI can deal with SQLDA, it is not the normal method. Hence, significantly more work for me and thus, more places for things to go wrong. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09812-36@lion>; Fri, 8 Sep 1995 07:30:35 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 23:57:17 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ax15493; 8 Sep 95 0:56 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <22397-0@frigate.doc.ic.ac.uk>; Thu, 7 Sep 1995 20:14:47 +0100 Received: by gw.home.vix.com id AA01360; Thu, 7 Sep 95 05:34:07 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01355; Thu, 7 Sep 95 05:34:05 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id HAA37866; Thu, 7 Sep 1995 07:34:05 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA06790; Thu, 7 Sep 1995 07:34:03 -0500 From: mhm@austin.ibm.com Message-Id: <9509071234.AA06790@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: Your message of Wed, 06 Sep 95 21:23:03 +0100. <9509062201.aa19229@post.demon.co.uk> Date: Thu, 07 Sep 95 07:34:03 -0500 Tim Bunce writes: > [ all of the foundation discussion about binding parameter in DB2 deleted ] > > Is this related to DB2 not being able to describe variables for you? > If so you have my sympathies and, yes, you would need to manually > specify, somewhere, any required information which DB2 is unable > to tell you itself. Exactly, none of the information about the what the database thinks the parameters should be is available from the database. The enduser is required to specify all of following things to bind a variable to a parameter. 1. Column type - INPUT/OUTPUT, deferred/nondeferred 2. Incoming C type for the variable. 3. Expected SQL type. 4. Precision of the variable 5. Scale of the variable 6. Pointer to variable - containing parameter value 7. Max size of parameter variable 8. Pointer to variable containing actual size of parameter value Thus, I was asking is it better to create a function in the xs file or the pm file? Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <10631-15@lion>; Fri, 8 Sep 1995 07:30:46 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 23:58:29 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id az15493; 8 Sep 95 0:57 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <23821-0@frigate.doc.ic.ac.uk>; Thu, 7 Sep 1995 21:00:37 +0100 Received: by gw.home.vix.com id AA04158; Wed, 6 Sep 95 23:03:43 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04152; Wed, 6 Sep 95 23:03:39 -0700 Received: by fstgds01.tu-graz.ac.at (5.57/Ultrix3.0-C) id AA08877; Thu, 7 Sep 95 08:03:31 +0200 Received: from psrini (fiicm1pc33) by iicm.tu-graz.ac.at with SMTP id AA18221 (5.67b/IDA-1.5t for ); Thu, 7 Sep 1995 08:03:26 +0200 Date: Thu, 7 Sep 1995 08:03:26 +0200 Message-Id: <199509070603.AA18221@iicm.tu-graz.ac.at> X-Sender: psrini@iicm.tu-graz.ac.at X-Mailer: Windows Eudora Version 1.4.3 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: perldb-interest@vix.com From: "P. Srinivasan" Subject: unsubscribe --------------------------------------------+-------------------------------- P. Srinivasan |email: psrini@iicm.tu-graz.ac.at Institute for Information Processing | and Computer Supported New Media (IICM) | Technical University Graz |Tel: ++43 316 832551-33 Schieszstattgasse 4A, A-8010 Graz, Austria |Fax: ++43 316 824394 --------------------------------------------+-------------------------------- ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09549-46@lion>; Fri, 8 Sep 1995 07:31:11 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 23:58:33 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id bb15493; 8 Sep 95 0:57 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <29210-0@frigate.doc.ic.ac.uk>; Thu, 7 Sep 1995 23:55:20 +0100 Received: by gw.home.vix.com id AA20494; Thu, 7 Sep 95 10:01:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20485; Thu, 7 Sep 95 10:01:00 -0700 Received: from lazarus.nrtc.northrop.com by nrtc.nrtc.northrop.com id aa29202; 7 Sep 95 8:34 PDT Received: from world.nad.northrop.com by lazarus.nrtc.northrop.com (15.4a/15.6.b) id AA26701; Thu, 7 Sep 95 09:54:04 pdt Received: by world.nad.northrop.com (4.1/SMI-4.1.1) id AA00954; Thu, 7 Sep 95 09:50:53 PDT Date: Thu, 7 Sep 1995 09:50:53 -0700 (PDT) From: "Charles G. Margolin" To: mhm@austin.ibm.com Cc: perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: <9509071638.AA19858@dax.austin.ibm.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 7 Sep 1995 mhm@austin.ibm.com wrote: > > "Charles G. Margolin" writes: > > > > Maybe this is way off base, since I have never worked with the Unix > > version of DB2, but in ancient times I used to work with mainframe > > DB2 and I did some dynamic SQL coding in various languages. > > > > With mainframe DB2 I could do a PREPARE and a DESCRIBE. Then I would use > > the info that DESCRIBE stores into the SQLDA block about data > > requirements to complete the SQLDA by supplying the addresses. I could > > even cause some type conversions by changing the data type. > > > > Does DB2/Unix lack this feature, or am I missing something important? > > > > Not too far off base. Basically, what you are talking about is > embeded SQL within DB2 and here the term dynamic basically means that > you know all the sql that needs to be performed, but not all of the > values to be used in the SQL. Not possible in an interpreted > environment. > > What I am dealing with is the Call Level Interface (CLI) which is > atleast one level of abstraction (maybe two, :-) ) lower than > embeded SQL. Samething holds for DBD/Oracle and I don't know about > Informix, Sybase, and Ingres. > > > Unfortunately for me, the current version of DB2 CLI does not > support any form of describe for a parameterized SQL statement, which > makes it quite difficult to make DBD/DB2 look and work just like > DBD/Oracle as I must require the user to provide more information > prior to binding a statement. > Here's a kludge: Perhaps you could code a routine using the DB2 precompiler which would receive an arbitrary SQL statement, PREPARE and DESCRIBE it, and return the resulting SQLDA structure. Then use that SQLDA for the info you need in your CLI-based routines. Just a thought. Charles Margolin email: cmargoli@world.nad.northrop.com Northrop Grumman Corporation phone: (310)332-6011 One Northrop Ave. Orgn. 624/23 Hawthorne, CA 90250-3277 ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09812-39@lion>; Fri, 8 Sep 1995 07:31:15 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 07 Sep 95 23:58:34 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id bd15493; 8 Sep 95 0:57 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <00197-0@frigate.doc.ic.ac.uk>; Fri, 8 Sep 1995 00:28:48 +0100 Received: by gw.home.vix.com id AA18475; Thu, 7 Sep 95 09:38:12 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA18470; Thu, 7 Sep 95 09:38:11 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id LAA55967; Thu, 7 Sep 1995 11:38:10 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA19858; Thu, 7 Sep 1995 11:38:04 -0500 From: mhm@austin.ibm.com Message-Id: <9509071638.AA19858@dax.austin.ibm.com> To: "Charles G. Margolin" Cc: perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: Your message of Thu, 07 Sep 95 09:20:00 -0700. Date: Thu, 07 Sep 95 11:38:04 -0500 "Charles G. Margolin" writes: > > Maybe this is way off base, since I have never worked with the Unix > version of DB2, but in ancient times I used to work with mainframe > DB2 and I did some dynamic SQL coding in various languages. > > With mainframe DB2 I could do a PREPARE and a DESCRIBE. Then I would use > the info that DESCRIBE stores into the SQLDA block about data > requirements to complete the SQLDA by supplying the addresses. I could > even cause some type conversions by changing the data type. > > Does DB2/Unix lack this feature, or am I missing something important? > Not too far off base. Basically, what you are talking about is embeded SQL within DB2 and here the term dynamic basically means that you know all the sql that needs to be performed, but not all of the values to be used in the SQL. Not possible in an interpreted environment. What I am dealing with is the Call Level Interface (CLI) which is atleast one level of abstraction (maybe two, :-) ) lower than embeded SQL. Samething holds for DBD/Oracle and I don't know about Informix, Sybase, and Ingres. Unfortunately for me, the current version of DB2 CLI does not support any form of describe for a parameterized SQL statement, which makes it quite difficult to make DBD/DB2 look and work just like DBD/Oracle as I must require the user to provide more information prior to binding a statement. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09812-41@lion>; Fri, 8 Sep 1995 07:31:54 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 00:56:28 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13466; 8 Sep 95 1:55 +0100 Received: by gw.home.vix.com id AA16977; Thu, 7 Sep 95 15:01:45 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA16972; Thu, 7 Sep 95 15:01:41 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa08473; 7 Sep 95 20:31 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab20784; 7 Sep 95 20:28 +0100 Received: from ig.co.uk by lion id <06926-0@lion>; Thu, 7 Sep 1995 20:29:02 +0100 To: mhm@austin.ibm.com Subject: Re: Building the %attribs for bind_param ? Cc: perldb-interest@vix.com Date: Thu, 7 Sep 1995 20:29:02 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509072028.ab20784@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > > > [ all of the foundation discussion about binding parameter in > DB2 deleted ] > > > Is this related to DB2 not being able to describe variables for you? > > If so you have my sympathies and, yes, you would need to manually > > specify, somewhere, any required information which DB2 is unable > > to tell you itself. > > Exactly, none of the information about the what the database thinks > the parameters should be is available from the database. The enduser > is required to specify all of following things to bind a variable to > a parameter. > > 1. Column type - INPUT/OUTPUT, deferred/nondeferred I thought DB2 could describe columns but not parameters. If so then stick to variables being INPUT and nondeferred only for now (and like I said before, deferred is probably to dangerous to use). > 2. Incoming C type for the variable. Are there any types which DB2 can't auto-convert to a string buffer? > 3. Expected SQL type. > 4. Precision of the variable > 5. Scale of the variable > 6. Pointer to variable - containing parameter value > 7. Max size of parameter variable > 8. Pointer to variable containing actual size of parameter value > Umm, this talk of pointers worries me. It's not something the user should have to think about. I recommend you strive to find a selection of values and automatic typ conversions which enable you to do something like $sth->bind_param(1, $var1, $bpa_int); $sth->bind_param(2, $var2, $bpa_string); $sth->bind_param(3, $var3, $bpa_float); $sth->bind_param(4, $var4, $bpa_date); E.g., predefine a selection of bind param attribute hashes which are set to liberal values. If this is not possible for DB2 get yourself a new database! > Thus, I was asking is it better to create a function in the xs file > or the pm file? > Do whatever it is you have in mind in the pm file first if possible. Migrate to xs only if needed. What do you envisage your function interface looking like? Do IBM plan to fix the DB2 api? > Mike > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09812-42@lion>; Fri, 8 Sep 1995 07:32:08 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 01:07:30 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ab18508; 8 Sep 95 2:06 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <01095-0@frigate.doc.ic.ac.uk>; Fri, 8 Sep 1995 01:10:35 +0100 Received: by gw.home.vix.com id AA24169; Thu, 7 Sep 95 10:50:04 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA24144; Thu, 7 Sep 95 10:50:02 -0700 Received: from lazarus.nrtc.northrop.com by nrtc.nrtc.northrop.com id aa00117; 7 Sep 95 9:23 PDT Received: from world.nad.northrop.com by lazarus.nrtc.northrop.com (15.4a/15.6.b) id AA26995; Thu, 7 Sep 95 10:49:29 pdt Received: by world.nad.northrop.com (4.1/SMI-4.1.1) id AA07448; Thu, 7 Sep 95 10:46:18 PDT Date: Thu, 7 Sep 1995 10:46:17 -0700 (PDT) From: "Charles G. Margolin" To: mhm@austin.ibm.com Cc: perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: <9509071735.AA18276@dax.austin.ibm.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 7 Sep 1995 mhm@austin.ibm.com wrote: > > "Charles G. Margolin" writes: > > > > Here's a kludge: Perhaps you could code a routine using the DB2 > > precompiler which would receive an arbitrary SQL statement, > > PREPARE and DESCRIBE it, and return the resulting SQLDA structure. > > Then use that SQLDA for the info you need in your CLI-based routines. > > > > Thought about it but there are two major problems: > > 1) The big problem is that it requires each machine to have the > precompiler installed and available to the enduser. Not always the > case in production shop. > > 2) While CLI can deal with SQLDA, it is not the normal method. Hence, > significantly more work for me and thus, more places for things > to go wrong. > Actually, 1 is less of a problem because the only the installer needs to precompile. The precompilable module does not specify any tables and does not have to be precompiled by endusers. A longer shot would be to try to supply the C program spit out by the precompiler. Item 2 is indeed a problem because you would have to decode it and use the contents piecemeal. Charles Margolin email: cmargoli@world.nad.northrop.com Northrop Grumman Corporation phone: (310)332-6011 One Northrop Ave. Orgn. 624/23 Hawthorne, CA 90250-3277 ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <09549-53@lion>; Fri, 8 Sep 1995 07:32:48 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 01:50:43 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa06130; 8 Sep 95 2:49 +0100 Received: by gw.home.vix.com id AA21606; Thu, 7 Sep 95 16:07:53 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA21600; Thu, 7 Sep 95 16:07:51 -0700 Received: from nas.edu (chariot.nas.edu [144.171.1.14]) by darius.nas.edu (8.6.11/8.6.10) with SMTP id TAA11544 for ; Thu, 7 Sep 1995 19:07:49 -0400 Received: from cc:Mail by nas.edu id AA810526045; Thu, 07 Sep 95 18:57:00 EST Date: Thu, 07 Sep 95 18:57:00 EST From: Susan Kim Encoding: 2 Text Message-Id: <9508078105.AA810526045@nas.edu> To: perldb-interest@vix.com Subject: unsubscribe unsubscribe ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <13990-0@lion>; Fri, 8 Sep 1995 11:36:14 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 07:50:33 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa12356; 8 Sep 95 7:26 +0100 Received: by gw.home.vix.com id AA07676; Thu, 7 Sep 95 19:48:52 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07671; Thu, 7 Sep 95 19:48:51 -0700 X400-Received: by mta bnr.ca in /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Thu, 7 Sep 1995 20:27:52 -0400 X400-Received: by /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Thu, 7 Sep 1995 20:27:47 -0400 X400-Received: by /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Tue, 5 Sep 1995 20:26:00 -0400 Date: Thu, 7 Sep 1995 20:26:00 -0400 X400-Originator: /dd.id=1679065/g=andrew/i=a/s=markham/@bnr.ca X400-Mts-Identifier: [/PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/;bcars735.b.663:08.08.95.00.27.47] X400-Content-Type: P2-1984 (2) Content-Identifier: Anyone have D... From: "andrew (a.) markham" Message-Id: <"4664 Thu Sep 7 20:27:48 1995"@bnr.ca> To: perldb-interest@vix.com Subject: Anyone have DBD::mSQL compiling for HP-UX? Before I spend a few days trying to port DBD::mSQL, does anyone have it at compilation state for HP-UX? Any info appreciated! -- Andy Markham BNR, Inc. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03404-6@lion>; Mon, 11 Sep 1995 07:22:02 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 21:23:53 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa14505; 8 Sep 95 22:23 +0100 Received: by gw.home.vix.com id AA16748; Fri, 8 Sep 95 05:39:22 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA16742; Fri, 8 Sep 95 05:39:21 -0700 Received: by tymix.Tymnet.COM (4.1/SMI-4.1) id AA02013; Fri, 8 Sep 95 05:39:24 PDT Received: from 131.146.22.157 by tymix.Tymnet.COM (in.smtpd); 8 Sep 95 5:39:23 PDT Received: by ferengi.tymnet.com (4.1/SMI-4.1) id AA28182; Fri, 8 Sep 95 12:39:19 GMT Date: Fri, 8 Sep 95 12:39:19 GMT From: Terry Kane Message-Id: <9509081239.AA28182@ferengi.tymnet.com> To: perldb-interest@vix.com Subject: unsubscribe unsubscribe ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03404-10@lion>; Mon, 11 Sep 1995 07:22:58 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 22:23:25 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13568; 8 Sep 95 23:22 +0100 Received: by gw.home.vix.com id AA11339; Fri, 8 Sep 95 04:04:39 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA11334; Fri, 8 Sep 95 04:04:38 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa02511; 8 Sep 95 11:36 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac05587; 8 Sep 95 11:33 +0100 Received: from ig.co.uk by lion id <13968-0@lion>; Fri, 8 Sep 1995 11:35:31 +0100 To: robertig@x500.bt.co.uk Subject: Re: Calling PL/SQL blocks from Oraperl Cc: perldb-interest@vix.com Date: Fri, 8 Sep 1995 11:35:31 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509081133.ac05587@post.demon.co.uk> > From: "Robertson, Ian" > > Tim > > I'm having problems binding in oraperl to a PL/SQL procedure parameter > list. I can call stored procedures with no parameters, and I can bind SQL. > > Here is the offending code: > > $csr = &ora_open($lda, "DECLARE BEGIN itest.proc1(:1); END;") > || die "ora_open: $ora_errno: $ora_errstr\n"; > > &ora_bind($csr, '166544' ) > || warn "ora_bind($csr): $ora_errno: $ora_errstr\n"; > > @fields = &ora_fetch($csr) > || warn "ora_fetch($csr): $ora_errno: $ora_errstr\n"; > > I get the following complaint back from Oraperl > > ERROR EVENT: DBI::st=HASH(0x98e24)->DBI::default_handler: > 6550 ORA-06550: line 1, column 15: > PLS-00306: wrong number or types of arguments in call to 'PROC1' > ORA-06550: line 1, column 15: > PL/SQL: Statement ignored (DBD: odescr failed) > at /usr/local/lib/perl5/Oraperl.pm line 93 > Describe failed during DBI::st=HASH(0x98e24)->FETCH(NUM_OF_PARAMS) at > /usr/local/lib/perl5/Oraperl.pm line 93. > > I have tried combinations of CHAR/VARCHAR2/NUMBER without success. > > If you could point me to some documnetation or sample code for this I > would be most grateful. Ideally I would like to > know what types can be translated between oraperl and PL/SQL, and > particularly how this is achieved for vectors. > Umm, this is an interesting one. ora_open does: sub ora_open { my($lda, $stmt, $cache) = @_; my $csr = $lda->prepare($stmt) or return undef; # only execute here if no bind vars specified $csr->execute or return undef unless $csr->{NUM_OF_PARAMS}; $csr; } The $csr->{NUM_OF_PARAMS} triggers a describe call (odescr) which fails since the parameters have not been bound. The problem is that NUM_OF_PARAMS should not trigger an odescr since the value is already known. To be fixed in the next release. Meanwhile try adding these lines: if (kl==13 && strEQ(key, "NUM_OF_PARAMS")) /* handled by DBI */ return Nullsv; just above the line: if (!imp_sth->done_desc && !dbd_describe(sth, imp_sth)) { in dbd_st_FETCH(). > Thanks Ian > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03397-15@lion>; Mon, 11 Sep 1995 07:24:17 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 23:26:45 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa11953; 9 Sep 95 0:26 +0100 Received: by gw.home.vix.com id AA10620; Fri, 8 Sep 95 03:51:16 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA10615; Fri, 8 Sep 95 03:51:15 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ac02484; 8 Sep 95 11:36 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa05587; 8 Sep 95 11:33 +0100 Received: from ig.co.uk by lion id <13643-0@lion>; Fri, 8 Sep 1995 10:05:39 +0100 To: perldb-interest@vix.com Subject: How to unsubscribe Date: Fri, 8 Sep 1995 10:05:39 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509081133.aa05587@post.demon.co.uk> Anyone wishing to unsubscribe should post to perldb-interest-request@vix.com rather than the normal list address. Thanks. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03397-16@lion>; Mon, 11 Sep 1995 07:24:38 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 23:31:36 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa14383; 9 Sep 95 0:30 +0100 Received: by gw.home.vix.com id AA21254; Fri, 8 Sep 95 06:31:19 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA21249; Fri, 8 Sep 95 06:31:17 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id IAA143660 for ; Fri, 8 Sep 1995 08:31:17 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA15936; Fri, 8 Sep 1995 08:31:15 -0500 From: mhm@austin.ibm.com Message-Id: <9509081331.AA15936@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: Your message of Thu, 07 Sep 95 20:29:02 +0100. <9509072028.ab20784@post.demon.co.uk> Date: Fri, 08 Sep 95 08:31:15 -0500 Tim Bunce writes: > > Exactly, none of the information about the what the database thinks > > the parameters should be is available from the database. The enduser > > is required to specify all of following things to bind a variable to > > a parameter. > > > > 1. Column type - INPUT/OUTPUT, deferred/nondeferred > > I thought DB2 could describe columns but not parameters. If so then > stick to variables being INPUT and nondeferred only for now (and like > I said before, deferred is probably to dangerous to use). This field referes to the method by which DB2 will attempt to use the incoming parameter. Each type (INPUT/OUTPUT, deferred/nondeferred) inplies a slight different behavior. And yes, initially, everything will be type INPUT, but if I ever want to support stored precedures I must make allowances for OUTPUT and deferred. > > 2. Incoming C type for the variable. > > Are there any types which DB2 can't auto-convert to a string buffer? No, but depending on how you set up the SQL the optimize might guess and inappropriate conversion. Doesn't happen very often, but if it does happen I want the enduser to be able to override the defaults. > > 3. Expected SQL type. > > 4. Precision of the variable > > 5. Scale of the variable > > 6. Pointer to variable - containing parameter value > > 7. Max size of parameter variable > > 8. Pointer to variable containing actual size of parameter value > > > Umm, this talk of pointers worries me. It's not something the user > should have to think about. If I do my job right, the enduser will never know he is dealing with pointers. So far hiding that fact has been very easy. The one place pointers are causing me problems is getting the addres of the $var in the $sth->bind_param(1, $var, \%attribs), here a ref to $var would make thing a bit easier. > I recommend you strive to find a selection of values and automatic > typ conversions which enable you to do something like > > $sth->bind_param(1, $var1, $bpa_int); > $sth->bind_param(2, $var2, $bpa_string); > $sth->bind_param(3, $var3, $bpa_float); > $sth->bind_param(4, $var4, $bpa_date); > > E.g., predefine a selection of bind param attribute hashes which are > set to liberal values. > > If this is not possible for DB2 get yourself a new database! Quit possible and already done for the main four cases. Unfortunately, I already know that data belonging to some of the potential users, contains things other than the main four types, as such I was trying to come up with a generic method of allowing the end users to mutate an existing default to match the particular combination of things they want to use. > > Thus, I was asking is it better to create a function in the xs file > > or the pm file? > > > Do whatever it is you have in mind in the pm file first if possible. > Migrate to xs only if needed. > > What do you envisage your function interface looking like? At this point I'm not quite sure which is the best approach. Thus far, I have predefined attributes for int, string, float and date and was thinking about something like $new_attribs = build_attrib($existing_attrib, $new_type[, \%changes] ) and probably, show_attribs($existing_attrib); so in essence, if called with two args it would simply change the ctype to sqltype mapping. Or, if called with three args would change some/all of the other underlying attributes. And probably, show_attribs($existing_attrib); which would dump the contents of an existing attribute hash ref. Right now I think, I can do everything I need to do in the DB2.pm file, but it will be a few days before I know for sure. > > Do IBM plan to fix the DB2 api? Don't know. The CLI folks have been trying for two years, the engine folks claim to have heard nothing about it until just recently. I am sure that if X/Open mandates GetStmtAttribute, the we will get both functions. Mike ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03397-19@lion>; Mon, 11 Sep 1995 07:25:48 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 08 Sep 95 23:53:33 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa24838; 9 Sep 95 0:53 +0100 Received: by gw.home.vix.com id AA12449; Fri, 8 Sep 95 11:10:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA12413; Fri, 8 Sep 95 11:10:03 -0700 Received: by netcon.smc.edu (5.59/25-eef, Santa Monica College) id AA09672; Fri, 8 Sep 95 13:11:51 EDT Received: by intime.intime.com (Smail3.1.29.1 #2) id m0sr7WM-00040pC; Fri, 8 Sep 95 10:47 PDT Message-Id: Subject: Re: Building the %attribs for bind_param ? To: perldb-interest@vix.com Date: Fri, 8 Sep 1995 10:47:22 -0700 (PDT) From: Scott Michel In-Reply-To: <9509072028.ab20784@post.demon.co.uk> from "Tim Bunce" at Sep 7, 95 08:29:02 pm X-Mailer: ELM [version 2.4 PL22] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 666 Organization: IntelliMED Corporation, Santa Monica, CA, USA Tim Bunce says: > > 8. Pointer to variable containing actual size of parameter value > > > Umm, this talk of pointers worries me. It's not something the user > should have to think about. This sounds like a holdover from embedded SQL, and isn't that important for Perl. There are so many ways around it because Perl is a dynamic environment unlike the "special" arrangements one has to make in a static environment like C. > If this is not possible for DB2 get yourself a new database! Let's stop being religious. Remember, IBM may be Inferior But Marketable, but they've been doing this for many more years, and likely many more to come in the future. -scottm ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03397-20@lion>; Mon, 11 Sep 1995 07:26:09 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 09 Sep 95 00:19:26 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa06851; 9 Sep 95 1:18 +0100 Received: by gw.home.vix.com id AA12125; Fri, 8 Sep 95 11:03:56 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA12116; Fri, 8 Sep 95 11:03:55 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id NAA176804 for ; Fri, 8 Sep 1995 13:03:54 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA28914; Fri, 8 Sep 1995 13:03:52 -0500 From: mhm@austin.ibm.com Message-Id: <9509081803.AA28914@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Problem with XSUB or DBI magic Date: Fri, 08 Sep 95 13:03:52 -0500 I am trying to use: $sth->bind_param($param_number,$value, $attrib_type) yet when I get into dbd_bind_ph(sth,param,value,attributes) I only have 3 valid values on the stack as attributes contains a Nullsv. The other interesting (or is it just my stupidity :-) ) thing is that value returns after the following sv_setsv(phs->sv, value); value_ptr = SvPV(phs->sv, value_len); is SCALAR(0x20080558) I'm lost at this point. Any ideas. Mike Moran mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03397-21@lion>; Mon, 11 Sep 1995 07:26:29 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 09 Sep 95 00:24:35 GMT Received: from gate.demon.co.uk by punt.demon.co.uk id aa09239; 9 Sep 95 1:24 +0100 Received: from gw.home.vix.com by gate.demon.co.uk id ab18004; 9 Sep 95 1:24 GMT-60:00 Received: by gw.home.vix.com id AA12742; Fri, 8 Sep 95 11:10:36 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA12737; Fri, 8 Sep 95 11:10:36 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id NAA154270 for ; Fri, 8 Sep 1995 13:10:35 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA03862; Fri, 8 Sep 1995 13:10:34 -0500 From: mhm@austin.ibm.com Message-Id: <9509081810.AA03862@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Anyone else notice .... Date: Fri, 08 Sep 95 13:10:34 -0500 that messages to the list seem to be taking longer and longer to be propogated. The last few I have seen, have taken in excess of 12 hours from orgination to delivery at my system....not including time zone differences. Just curious. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03404-20@lion>; Mon, 11 Sep 1995 07:26:53 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 09 Sep 95 01:18:15 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa03287; 9 Sep 95 2:18 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <21344-0@frigate.doc.ic.ac.uk>; Sat, 9 Sep 1995 02:14:50 +0100 Received: by gw.home.vix.com id AA01146; Fri, 8 Sep 95 14:33:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01141; Fri, 8 Sep 95 14:33:05 -0700 Received: from lesol1.dseg.ti.com ([128.247.231.86]) by gate.ti.com (8.6.12/) with ESMTP id QAA13810 for ; Fri, 8 Sep 1995 16:33:02 -0500 Received: from ms3.dseg.ti.com (ms3.dseg.ti.com [128.247.215.78]) by lesol1.dseg.ti.com (8.6.9/8.6.6) with SMTP id QAA29466 for ; Fri, 8 Sep 1995 16:31:44 -0500 Received: from ms19.nbserv by ms3.dseg.ti.com (4.1/SMI-4.1) id AA23158; Fri, 8 Sep 95 16:31:42 CDT Date: Fri, 8 Sep 95 16:31:42 CDT From: John Cerney 0182636 Message-Id: <9509082131.AA23158@ms3.dseg.ti.com> To: perldb-interest@vix.com Subject: unsubscribe unsubscribe ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03397-23@lion>; Mon, 11 Sep 1995 07:27:11 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 09 Sep 95 01:18:34 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa03287; 9 Sep 95 2:18 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <21369-0@frigate.doc.ic.ac.uk>; Sat, 9 Sep 1995 02:15:22 +0100 Received: by gw.home.vix.com id AA18585; Fri, 8 Sep 95 12:07:56 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA18571; Fri, 8 Sep 95 12:07:55 -0700 Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA13016; Fri, 8 Sep 1995 12:05:07 -0700 Message-Id: <199509081905.MAA13016@toccata.fugue.com> To: mhm@austin.ibm.com Cc: perldb-interest@vix.com Subject: Re: Anyone else notice .... In-Reply-To: Your message of "Fri, 08 Sep 1995 13:10:34 CDT." <9509081810.AA03862@dax.austin.ibm.com> Date: Fri, 08 Sep 1995 12:05:06 -0700 From: Ted Lemon Your message was delivered to me about six minutes after it was received at vix.com. However, there is definitely a lag on the mailing list, because it's quite large these days. Sendmail runs the list sequentially, and if a lot of hosts above you on the list are unreachable, it will take a long time to time out on them, and consequently delivery to you will be delayed. Twelve hours does seem like a pretty long time, though - it probably implies that delivery to you is being deferred because your host isn't reliably reachable. This could be the result of a routing problem on the Internet, too heavy a load on your mail server, or a wide variety of less obvious things. A copy of the headers on a message that you feel was delayed a long time would help us to determine where in the delivery process your message is being delayed. _MelloN_ ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04346-1@lion>; Mon, 11 Sep 1995 07:27:20 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 09 Sep 95 02:07:38 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa21404; 9 Sep 95 3:07 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <22686-0@frigate.doc.ic.ac.uk>; Sat, 9 Sep 1995 03:06:45 +0100 Received: by gw.home.vix.com id AA21869; Fri, 8 Sep 95 12:42:07 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA21861; Fri, 8 Sep 95 12:42:04 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.11/8.6.11) with SMTP id OAA165692 for ; Fri, 8 Sep 1995 14:42:03 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA26978; Fri, 8 Sep 1995 14:41:57 -0500 From: mhm@austin.ibm.com Message-Id: <9509081941.AA26978@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: defining Global hash refs in a pm file Date: Fri, 08 Sep 95 14:41:56 -0500 Found my problem, I am not defining my default parameter attribute hash refs correctly in the DB2.pm file. Question is how do I define them in the global name space from the DB2.pm file ? Thanks. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03397-41@lion>; Mon, 11 Sep 1995 07:33:26 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 10 Sep 95 05:23:15 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa21614; 10 Sep 95 6:23 +0100 Received: by gw.home.vix.com id AA15752; Sat, 9 Sep 95 18:03:30 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA15747; Sat, 9 Sep 95 18:03:29 -0700 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id UAA27165; Sat, 9 Sep 1995 20:03:28 -0500 Date: Sat, 9 Sep 1995 20:03:27 -0500 (CDT) From: Ben A Lindstrom To: perldb-interest@vix.com Subject: Finally compiling DBI Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I finally was able to get access to my Target NeXTStep 3.1 machine. It compiles almost clean...Just one warning (at 1049 Incompatable Type) but I don't think that is too major. It seems to bomb at test 5 with the following. Can't locate DBI object method "0" via package "DBD::Switch::dr" at t/min.t line 19. FAILED on test 5 Then everything works until the end. Test error handling: prepare invalid query. Expect an ERROR EVENT message: Can't locate DBI object method "Unknown field name:: unknown_field_name" via package "DBD::ExampleP::db" at blib/DBD/ExampleP.pm line 63. *** Exit 2 I don't know what is to be expected or not. I'm sure the *** Exit 2 is not to be expected. =) I don't know about the "Test error handling" message. I'll still install it so I have something to work from. But anyone have any ideas? ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <03404-38@lion>; Mon, 11 Sep 1995 07:33:40 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 10 Sep 95 06:50:40 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa19002; 10 Sep 95 7:50 +0100 Received: by gw.home.vix.com id AA19753; Sat, 9 Sep 95 19:11:25 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA19748; Sat, 9 Sep 95 19:11:24 -0700 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id VAA29379; Sat, 9 Sep 1995 21:11:23 -0500 Date: Sat, 9 Sep 1995 21:11:22 -0500 (CDT) From: Ben A Lindstrom To: perldb-interest@vix.com Subject: XS/PM/C question. Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Ok, I'm in the process of trying to compile the partly modified Oracle modules (that now contains QuickBase calls). But I can't seem simplely compile it a shared library by hand. xsubpp deals with the .xs nicely, but I am unsure what exactly is done with the dbdimp.c/h file. Any information would be greatly helpful. Thanks. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04346-20@lion>; Mon, 11 Sep 1995 07:34:11 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sun, 10 Sep 95 15:16:20 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa00918; 10 Sep 95 16:16 +0100 Received: by gw.home.vix.com id AA25786; Sun, 10 Sep 95 04:56:17 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA25780; Sun, 10 Sep 95 04:56:15 -0700 Received: from kukini.cs.bris.ac.uk by dira.bris.ac.uk with SMTP (PP); Sun, 10 Sep 1995 12:56:12 +0100 Received: from kiha by kukini.compsci.bristol.ac.uk id aa05824; 10 Sep 95 11:59 GMT To: perldb-interest@vix.com Subject: Compiling Perl for System V Date: Sun, 10 Sep 95 11:55:51 GMT From: ally@cs.bris.ac.uk Message-Id: <9509101155.aa02603@uk.ac.bristol.compsci.kiha> I've finally got Perl v4.036 and Oraperl v2.4 to compile for Sun System V and it seems to be working fine with Oracle v7.0.13.6.0. Many thanks to all those that replied to my earlier cry for help. Only one problem remains. How do I compile Perl without getting a "FAILED on test 1" for op/groups? Karim Ally Dept. of Computer Science University of Bristol ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20831-7@lion>; Tue, 12 Sep 1995 07:21:53 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 11 Sep 95 21:31:00 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13205; 11 Sep 95 22:30 +0100 Received: by gw.home.vix.com id AA04165; Mon, 11 Sep 95 07:34:09 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04153; Mon, 11 Sep 95 07:34:08 -0700 Received: from uucp6.UU.NET by relay3.UU.NET with SMTP id QQzgsw11410; Mon, 11 Sep 1995 10:34:05 -0400 Received: from cambex.UUCP by uucp6.UU.NET with UUCP/RMAIL ; Mon, 11 Sep 1995 10:34:07 -0400 Received: by cambex.cambex.com (AIX 3.2/UCB 5.64/4.03) id AA31489; Mon, 11 Sep 1995 10:20:33 -0400 Date: Mon, 11 Sep 1995 10:20:33 -0400 From: Dave Ford Message-Id: <9509111420.AA31489@cambex.cambex.com> To: perldb-interest@vix.com, skim@nas.edu Subject: Re: unsubscribe > unsubscribe Could someone please post how you unsubscribe so we can all stop getting these messages? Dave Ford Cambex ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20831-8@lion>; Tue, 12 Sep 1995 07:22:02 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 11 Sep 95 21:33:56 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id aa15492; 11 Sep 95 22:33 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <07158-0@frigate.doc.ic.ac.uk>; Mon, 11 Sep 1995 22:33:33 +0100 Received: by gw.home.vix.com id AA02944; Mon, 11 Sep 95 07:12:13 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA02935; Mon, 11 Sep 95 07:12:08 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id JAA123152 for ; Mon, 11 Sep 1995 09:12:07 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA26812; Mon, 11 Sep 1995 09:12:03 -0500 From: mhm@austin.ibm.com Message-Id: <9509111412.AA26812@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: $sth->execute(@bind_values) question Date: Mon, 11 Sep 95 09:12:02 -0500 Under DBD::Oracle does $sth->execute() ... not no args... work if $sth->bind_param($index_or_name,$var,$attribs) called first. At this point, DB2perl will work correctly if I use bind_param and provide a list of bind values to execute. If I don't provide the list of bind values then the data is corrupted going in. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20831-21@lion>; Tue, 12 Sep 1995 07:25:13 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 11 Sep 95 23:13:55 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09077; 12 Sep 95 0:13 +0100 Received: by gw.home.vix.com id AA00496; Mon, 11 Sep 95 06:30:50 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00490; Mon, 11 Sep 95 06:30:48 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.11/8.6.11) with SMTP id IAA113295; Mon, 11 Sep 1995 08:30:43 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA25222; Mon, 11 Sep 1995 08:30:41 -0500 From: mhm@austin.ibm.com Message-Id: <9509111330.AA25222@dax.austin.ibm.com> To: mhm@austin.ibm.com Cc: perldb-interest@vix.com Subject: Re: defining Global hash refs in a pm file -- still !correct In-Reply-To: Your message of Fri, 08 Sep 95 14:41:56 -0500. <9509081941.AA26978@dax.austin.ibm.com> Date: Mon, 11 Sep 95 08:30:40 -0500 mhm@austin.ibm.com writes: > > Found my problem, I am not defining my default parameter attribute > hash refs correctly in the DB2.pm file. Question is how do I > define them in the global name space from the DB2.pm file ? Well, I have tried the obvious things. Either defining the hash refs as $main::attrib_char = { 'name' => value} or referencing them as $DBD::DB2::attrib_char, neither of which succeeds when I attempt to an svp = hv_fetch((HV*)SvRV(attribs), "Stype",0, 0); I get back a NULL svp. So where have I run amuck? Mike ------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20831-38@lion>; Tue, 12 Sep 1995 07:27:38 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 12 Sep 95 02:30:34 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa15187; 12 Sep 95 3:30 +0100 Received: by gw.home.vix.com id AA14182; Mon, 11 Sep 95 15:35:13 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA14172; Mon, 11 Sep 95 15:35:11 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa14473; 11 Sep 95 15:35 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab09988; 11 Sep 95 15:33 +0100 Received: from ig.co.uk by lion id <07387-0@lion>; Mon, 11 Sep 1995 10:28:13 +0100 To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: Finally compiling DBI Date: Mon, 11 Sep 1995 10:28:13 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509111533.ab09988@post.demon.co.uk> > From: Ben A Lindstrom > > I finally was able to get access to my Target NeXTStep 3.1 machine. > > It compiles almost clean...Just one warning (at 1049 Incompatable Type) > but I don't think that is too major. > > It seems to bomb at test 5 with the following. > > Can't locate DBI object method "0" via package "DBD::Switch::dr" at > t/min.t line 19. > FAILED on test 5 > Upgrade the DBI. Here's the change note: Changes in DBI 0.63, 1 Sep 1995 Minor update. Fixed uninitialised memory bug in method attribute handling and streamlined processing and debugging. Revised usage definitions for bind_* methods and readblob. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20831-40@lion>; Tue, 12 Sep 1995 07:27:55 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 12 Sep 95 02:58:30 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa28398; 12 Sep 95 3:57 +0100 Received: by gw.home.vix.com id AA14177; Mon, 11 Sep 95 15:35:11 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA14170; Mon, 11 Sep 95 15:35:09 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa14480; 11 Sep 95 15:35 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac09988; 11 Sep 95 15:33 +0100 Received: from ig.co.uk by lion id <07394-0@lion>; Mon, 11 Sep 1995 10:29:58 +0100 To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: XS/PM/C question. Date: Mon, 11 Sep 1995 10:29:58 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509111533.ac09988@post.demon.co.uk> > From: Ben A Lindstrom > > Ok, I'm in the process of trying to compile the partly modified Oracle > modules (that now contains QuickBase calls). > > But I can't seem simplely compile it a shared library by hand. xsubpp > deals with the .xs nicely, but I am unsure what exactly is done with > the dbdimp.c/h file. > > Any information would be greatly helpful. > You'll need to give me more to go on than that. (Also see the notes in the README about the kind of information I'll typically need/want.) Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20831-42@lion>; Tue, 12 Sep 1995 07:28:13 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 12 Sep 95 03:37:28 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa14457; 12 Sep 95 4:36 +0100 Received: by gw.home.vix.com id AA14289; Mon, 11 Sep 95 15:38:05 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AB14283; Mon, 11 Sep 95 15:38:03 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa14568; 11 Sep 95 15:36 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ad10062; 11 Sep 95 15:33 +0100 Received: from ig.co.uk by lion id <08126-0@lion>; Mon, 11 Sep 1995 11:18:23 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: defining Global hash refs in a pm file Date: Mon, 11 Sep 1995 11:18:23 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509111534.ad10062@post.demon.co.uk> > From: mhm@austin.ibm.com > > Found my problem, I am not defining my default parameter attribute > hash refs correctly in the DB2.pm file. Question is how do I define them > in the global name space from the DB2.pm file ? > The short answer is that you don't do it via the DBI. I think it would be reasonable for the application to say: use DBI; use DBD::DB2 qw($bpa_foo $bpa_bar); since this clearly marks the application as non portable, which it is. I'll add a build_attrib method (in C for speed) to the next DBI release. > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20831-45@lion>; Tue, 12 Sep 1995 07:28:41 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 12 Sep 95 04:29:49 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09236; 12 Sep 95 5:28 +0100 Received: by gw.home.vix.com id AA05107; Mon, 11 Sep 95 13:44:44 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05097; Mon, 11 Sep 95 13:44:39 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab14662; 11 Sep 95 20:47 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa15368; 11 Sep 95 20:44 +0100 Received: from ig.co.uk by lion id <18406-0@lion>; Mon, 11 Sep 1995 20:43:10 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Building the %attribs for bind_param ? Date: Mon, 11 Sep 1995 20:43:10 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509112045.aa15368@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > > > Exactly, none of the information about the what the database thinks > > > the parameters should be is available from the database. The enduser > > > is required to specify all of following things to bind a variable to > > > a parameter. > > > > > > 1. Column type - INPUT/OUTPUT, deferred/nondeferred > > > > I thought DB2 could describe columns but not parameters. If so then > > stick to variables being INPUT and nondeferred only for now (and like > > I said before, deferred is probably to dangerous to use). > > This field referes to the method by which DB2 will attempt to use the > incoming parameter. Each type (INPUT/OUTPUT, deferred/nondeferred) > inplies a slight different behavior. And yes, initially, everything > will be type INPUT, but if I ever want to support stored precedures > I must make allowances for OUTPUT and deferred. > Having just had a quick read of the SQL CLI spec I understand your usage of deferred now. It comes under the 'not urgent' category and is unlikely to have direct DBI support until the DBI starts absorbing the SQL CLI spec. Mapping PutData etc into Perl needs some thought. > > > 2. Incoming C type for the variable. > > > > Are there any types which DB2 can't auto-convert to a string buffer? > > No, but depending on how you set up the SQL the optimize might guess > and inappropriate conversion. Doesn't happen very often, but if it > does happen I want the enduser to be able to override the defaults. > But it does simplify the usual case, and that's important. > > > 3. Expected SQL type. > > > 4. Precision of the variable > > > 5. Scale of the variable > > > 6. Pointer to variable - containing parameter value > > > 7. Max size of parameter variable > > > 8. Pointer to variable containing actual size of parameter value > > > > > Umm, this talk of pointers worries me. It's not something the user > > should have to think about. > > If I do my job right, the enduser will never know he is dealing with > pointers. So far hiding that fact has been very easy. The one > place pointers are causing me problems is getting the addres of > the $var in the $sth->bind_param(1, $var, \%attribs), here a ref > to $var would make thing a bit easier. > I was never very sure about this one. It started out being a reference but I changed it while pondering the danger of deferred writes to perl variable memory addresses. I've also mentioned some time back having a bind_param() for simple common usage and bind_param_inout() for fancy stuff (which would generate an error if unsupported). I think this might be best since it seems rather counterintuitive to have to say things like: $sth->bind_param(1, \42); and the creation, dereferencing and freeing of the reference is wasteful. Umm, I think I'm going to leave bind_param as-is for now and formally state that the DBI spec (such as it is) does not currently support output parameters. Feel free to use the $h->func(..., 'method_name') mechanism to add your own bind_param_inout style function. I'll come back to this issue when addressing the changes arising from the SQL CLI spec. > > I recommend you strive to find a selection of values and automatic > > typ conversions which enable you to do something like > > > > $sth->bind_param(1, $var1, $bpa_int); > > $sth->bind_param(2, $var2, $bpa_string); > > $sth->bind_param(3, $var3, $bpa_float); > > $sth->bind_param(4, $var4, $bpa_date); > > > > E.g., predefine a selection of bind param attribute hashes which are > > set to liberal values. > > > > If this is not possible for DB2 get yourself a new database! > > Quit possible and already done for the main four cases. Unfortunately, > I already know that data belonging to some of the potential users, > contains things other than the main four types, as such I was trying > to come up with a generic method of allowing the end users to mutate > an existing default to match the particular combination of things > they want to use. > > > > Thus, I was asking is it better to create a function in the xs file > > > or the pm file? > > > > > Do whatever it is you have in mind in the pm file first if possible. > > Migrate to xs only if needed. > > > > What do you envisage your function interface looking like? > > At this point I'm not quite sure which is the best approach. Thus far, > I have predefined attributes for int, string, float and date and was > thinking about something like > > $new_attribs = build_attrib($existing_attrib, $new_type[, \%changes] ) > so in essence, if called with two args it would simply change the > ctype to sqltype mapping. Or, if called with three args would change > some/all of the other underlying attributes. How does that add significant value over just saying something like: $new_attribs = { %$existing_attribs, db2_type => $new_type }; ? > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <25533-0@lion>; Tue, 12 Sep 1995 15:14:16 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 12 Sep 95 13:06:32 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa10679; 12 Sep 95 14:06 +0100 Received: by gw.home.vix.com id AA08234; Mon, 11 Sep 95 21:16:22 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA08229; Mon, 11 Sep 95 21:16:20 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id XAA156987 for ; Mon, 11 Sep 1995 23:16:19 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA22576; Mon, 11 Sep 1995 23:16:19 -0500 From: mhm@austin.ibm.com Message-Id: <9509120416.AA22576@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Re: Building the %attribs for bind_param ? In-Reply-To: Your message of Mon, 11 Sep 95 20:43:10 +0100. <9509112045.aa15368@post.demon.co.uk> Date: Mon, 11 Sep 95 23:16:18 -0500 Tim Bunce writes: |> > If I do my job right, the enduser will never know he is dealing with |> > pointers. So far hiding that fact has been very easy. The one |> > place pointers are causing me problems is getting the addres of |> > the $var in the $sth->bind_param(1, $var, \%attribs), here a ref |> > to $var would make thing a bit easier. |> > |> I was never very sure about this one. It started out being a reference |> but I changed it while pondering the danger of deferred writes to perl |> variable memory addresses. |> |> I've also mentioned some time back having a bind_param() for simple |> common usage and bind_param_inout() for fancy stuff (which would |> generate an error if unsupported). I think this might be best since |> it seems rather counterintuitive to have to say things like: |> |> $sth->bind_param(1, \42); |> |> and the creation, dereferencing and freeing of the reference is wasteful. |> |> Umm, I think I'm going to leave bind_param as-is for now and formally |> state that the DBI spec (such as it is) does not currently support |> output parameters. Feel free to use the $h->func(..., 'method_name') |> mechanism to add your own bind_param_inout style function. Ok, my perl ignorance is showing....If I don't have a reference how do I tell DB2 which memory location to get its values from? DB2 CLI does everything by reference and nothing by value. Thus, I need the address of the perl variable containing the value of the parameter. |> > > What do you envisage your function interface looking like? |> > |> > At this point I'm not quite sure which is the best approach. Thus far, |> > I have predefined attributes for int, string, float and date and was |> > thinking about something like |> > |> > $new_attribs = build_attrib($existing_attrib, $new_type[, \%changes] ) |> |> > so in essence, if called with two args it would simply change the |> > ctype to sqltype mapping. Or, if called with three args would change |> > some/all of the other underlying attributes. |> |> How does that add significant value over just saying something like: |> |> $new_attribs = { %$existing_attribs, db2_type => $new_type }; Once again, I'm totally lost with that syntax. Would you explain what that says to you? Thanks. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <15626-3@lion>; Wed, 13 Sep 1995 07:21:02 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 13 Sep 95 01:17:00 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa01869; 13 Sep 95 2:16 +0100 Received: by gw.home.vix.com id AA05153; Tue, 12 Sep 95 03:35:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05147; Tue, 12 Sep 95 03:35:43 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa08351; 12 Sep 95 7:21 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa10657; 12 Sep 95 7:18 +0100 Received: from ig.co.uk by lion id <18588-0@lion>; Mon, 11 Sep 1995 21:05:44 +0100 To: perldb-interest@vix.com, jennifer@chmsr.isye.gatech.edu Subject: Re: Problems "make"ing the DBD::Oracle driver Cc: benjamin@mime1.marc.gatech.edu Date: Mon, 11 Sep 1995 21:05:44 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509120718.aa10657@post.demon.co.uk> > From: Jennifer Anne Turns > Sorry for the delay in getting back to you. > I am in the process of trying to get the Perl5 to communicate with > Oracle7. I have downloaded the most recent versions of the DBI (0.63) > and DBD (0.25) and have successfully installed DBI and tested it. I am > having problems when "make"ing DBD. The information reulting from the > first two steps of this process are given below: > > > --------------------------------------------------------------------- > THIS IS THE SUCCESSFUL OUTPUT FROM THE COMMAND: perl Makefile.PL > > Configuring DBD::Oracle version 0.25... > Remember to actually read the README file! > > Using Oracle in /usr/ld/oracle > Oracle sysliblist: -lm > Using /usr/ld/oracle/proc/lib/proc.mk version 7.27 > System: perl5.001 sunos srl 5.2 generic_100999-40 sun4m sparc sun4-solaris > dl_dlopen.xs Solaris 2.2 is very old and buggy. > Compiler: cc -O > Oracle proc.mk would have used these values but we override them: > CC: /usr/5bin/cc > CFLAGS: -I. -O > LDFLAGS: -Bdynamic -L$(LIBHOME) > > --------------------------------------------------------------------- > THIS IS THE UNSUCCESSFUL OUTPUT FROM THE COMMAND: make > > root > make > cc -c -O -Kpic -I/usr/local/lib/perl5/sun4-solaris/CORE > -I/usr/ld/oracle/rdbms/demo -I/usr/local/lib/perl5/sun4-solaris/DBI Oracle.c > cc -c -O -Kpic -I/usr/local/lib/perl5/sun4-solaris/CORE > -I/usr/ld/oracle/rdbms/demo -I/usr/local/lib/perl5/sun4-solaris/DBI dbdimp.c > Running Mkbootstrap for DBD::Oracle () > chmod 644 Oracle.bs > LD_RUN_PATH="" cc -o blib/auto/DBD/Oracle/Oracle.so -G Oracle.o dbdimp.o > -L/usr/ld/oracle/lib /usr/ld/oracle/lib/libocic.a /usr/ld/oracle/lib/osntab.o > -lsqlnet -lora /usr/ld/oracle/lib/libpls.a -lsqlnet -lnlsrtl -lcv6 -lcore > -lnlsrtl -lcv6 -lcore `cat /usr/ld/oracle/rdbms/lib/sysliblist` -lm > ld: fatal: file /usr/ld/oracle/lib/libocic.a: unknown type, unable to process > using elf(3E) libraries > *** Error code 1 > make: Fatal error: Command failed for target `blib/auto/DBD/Oracle/Oracle.so' > Is there any chance that the /usr/ld/oracle tree is not a Solaris 2 release of Oracle (e.g., it's a SunOS 4.1.3 release)? Try this: $ ar x /usr/ld/oracle/lib/libocic.a oci.o $ file oci.o oci.o: ELF 32-bit MSB relocatable SPARC Version 1 if it doesn't say 'ELF 32-bit ...' then either it's not a Solaris 2 release or your oracle files have been damaged. > jennifer > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <15626-4@lion>; Wed, 13 Sep 1995 07:21:09 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 13 Sep 95 01:20:25 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa03310; 13 Sep 95 2:19 +0100 Received: by gw.home.vix.com id AA13790; Tue, 12 Sep 95 12:15:51 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA13785; Tue, 12 Sep 95 12:15:50 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id OAA40317; Tue, 12 Sep 1995 14:15:50 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA18596; Tue, 12 Sep 1995 14:15:44 -0500 From: mhm@austin.ibm.com Message-Id: <9509121915.AA18596@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@vix.com Subject: Re: Getting the address of a variable. In-Reply-To: Your message of Tue, 12 Sep 95 16:58:44 +0100. <9509121656.aa05312@post.demon.co.uk> Date: Tue, 12 Sep 95 14:15:43 -0500 Tim Bunce writes: > > > Mike, did you mean to post to perl5-porters instead of perldb-interest? Well, sort of... currently my position on the perldb-interest list is so far down and there are so many "unreliable" delivery sites ahead of me, message turn around is somewhere between 8-24 hours. This is just to have my message returned to me. perl5-porters on the other hand is 1-2 hours. While the example was perldb the question actually related to perlguts so ... > Anyway, the answer is... it depends on the type of the variable. > Take a look at sv.h. For example, for SVt_PV (a typical string) > sv_any in struct sv points to a struct xpv. The struct xpv xpv_pv > field is the pointer to the string. > > But *don't* use it directly! sv.h defines many macros which should > be used instead of refering to the struct fields directly. > > SvPVX(sv) will get you the xpv_pv value - or maybe a core dump if the > sv is not of the right type. SvPV(sv) will also get you the xpv_pv > value if the sv is a SVt_PV type (simplification) but if it's not > it'll call sv_2pv() to convert the value to a string. > > For what you want to do you should *not* pass the address of the > bind_param $var_in_question to SQLBindParameter, if you do you run the > risk of memory corruption if $var_in_question is changed in any way. > > I suggest that you use sv_setsv() to make a copy of $var_in_question > and then check the type (using the SvTYPE macro) and, maybe using a > switch statement, pass appropriate parameters to SQLBindParameter() > using SvIVX(), SvNVX(), SvPVX() etc after checking SvIOK(), SvNOK(), > SvPOK() etc. > Which would work if you always wanted to call $sth->execute() with arguments. It fails if you attempt to prebind as you don't have the appropriate value in the variable at the time the copy is made. Or am I misinterpreting what you are attempting to say? What I want to be able to do is the following 1) $sth->bind_param(...) a bunch of other things $sth->execute() and 2) a bunch of things $sth->execute(arg1...argN) Both should work exactly the same, should use the same code and should NOT require that the value in argX be copied to bind_varX. Given those two cases and the three assumptions, I don't see how to do it without know the real address of the perl variable. Do you? Thanks. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <15627-5@lion>; Wed, 13 Sep 1995 07:21:27 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 13 Sep 95 02:18:32 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa28415; 13 Sep 95 3:18 +0100 Received: by gw.home.vix.com id AA04095; Tue, 12 Sep 95 10:23:51 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04086; Tue, 12 Sep 95 10:23:47 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa29946; 12 Sep 95 16:09 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac23439; 12 Sep 95 16:07 +0100 Received: from ig.co.uk by lion id <25762-0@lion>; Tue, 12 Sep 1995 15:31:47 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: $sth->execute(@bind_values) question Date: Tue, 12 Sep 1995 15:31:47 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509121607.ac23439@post.demon.co.uk> > From: mhm@austin.ibm.com > > Under DBD::Oracle does $sth->execute() ... not no args... work if > $sth->bind_param($index_or_name,$var,$attribs) called first. > > At this point, DB2perl will work correctly if I use bind_param and > provide a list of bind values to execute. If I don't provide the > list of bind values then the data is corrupted going in. > You'll need to keep track of what's been bound so you can complain if any of the parameters have not been bound at execute time. DBD::Oracle doesn't do this since Oracle does it for you (oexec generates an error if anything has not been bound yet). Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <26338-39@lion>; Thu, 14 Sep 1995 07:27:37 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 14 Sep 95 02:24:46 GMT Received: from relay-1.mail.demon.net by punt2.demon.co.uk id aa20993; 14 Sep 95 2:32 +0100 Received: from gate.demon.co.uk by relay-1.mail.demon.net id aa23072; 12 Sep 95 18:47 +0100 Received: by gw.home.vix.com id AA17511; Mon, 11 Sep 95 23:45:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA17505; Mon, 11 Sep 95 23:45:26 -0700 Received: (from total@localhost) by xmission.xmission.com (8.6.12/8.6.12) id AAA24741 for perldb-interest@vix.com; Tue, 12 Sep 1995 00:45:23 -0600 From: "Total Health Mktg." Message-Id: <199509120645.AAA24741@xmission.xmission.com> Subject: mSQL & DBM info request To: perldb-interest@vix.com Date: Tue, 12 Sep 1995 00:45:20 -0600 (MDT) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 465 Are any readers on this list using mSQL or DBM for CGI database applications? I am new to this and am trying to decide which, if any, database will be best. If there is a FAQ with this information would someone kindly direct me to it. Any readers out there using either of these for your database application I would appreciate your evalutions. Are there any sources on the net that I can learn more about these programs? Thank you, Jim clements@xmission.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <19528-7@lion>; Fri, 15 Sep 1995 00:32:51 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 14 Sep 95 23:15:17 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa01865; 15 Sep 95 0:15 +0100 Received: by gw.home.vix.com id AA29890; Tue, 12 Sep 95 20:42:43 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29885; Tue, 12 Sep 95 20:42:29 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa29920; 12 Sep 95 16:09 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa23439; 12 Sep 95 16:07 +0100 Received: from ig.co.uk by lion id <25718-0@lion>; Tue, 12 Sep 1995 15:24:11 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Building the %attribs for bind_param ? Date: Tue, 12 Sep 1995 15:24:11 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509121607.aa23439@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > If I do my job right, the enduser will never know he is dealing with > |> > pointers. So far hiding that fact has been very easy. The one > |> > place pointers are causing me problems is getting the addres of > |> > the $var in the $sth->bind_param(1, $var, \%attribs), here a ref > |> > to $var would make thing a bit easier. > |> > > |> I was never very sure about this one. It started out being a reference > |> but I changed it while pondering the danger of deferred writes to perl > |> variable memory addresses. > |> > |> I've also mentioned some time back having a bind_param() for simple > |> common usage and bind_param_inout() for fancy stuff (which would > |> generate an error if unsupported). I think this might be best since > |> it seems rather counterintuitive to have to say things like: > |> > |> $sth->bind_param(1, \42); > |> > |> and the creation, dereferencing and freeing of the reference is wasteful. > |> > |> Umm, I think I'm going to leave bind_param as-is for now and formally > |> state that the DBI spec (such as it is) does not currently support > |> output parameters. Feel free to use the $h->func(..., 'method_name') > |> mechanism to add your own bind_param_inout style function. > > Ok, my perl ignorance is showing....If I don't have a reference how > do I tell DB2 which memory location to get its values from? DB2 CLI > does everything by reference and nothing by value. Thus, I need the > address of the perl variable containing the value of the parameter. > Take a look at DBD::Oracle (again). Note that the bind_param function takes a local copy of the value and passes the address of that to Oracle. > |> > > What do you envisage your function interface looking like? > |> > > |> > At this point I'm not quite sure which is the best approach. Thus far, > |> > I have predefined attributes for int, string, float and date and was > |> > thinking about something like > |> > > |> > $new_attribs = build_attrib($existing_attrib, $new_type[, \%changes] ) > |> > |> > so in essence, if called with two args it would simply change the > |> > ctype to sqltype mapping. Or, if called with three args would change > |> > some/all of the other underlying attributes. > |> > |> How does that add significant value over just saying something like: > |> > |> $new_attribs = { %$existing_attribs, db2_type => $new_type }; > > Once again, I'm totally lost with that syntax. Would you explain what > that says to you? Hashes in a list context expand to a list of key/value pairs. %hash1 = ( Key,Value ); $hash1_ref = \%hash1; @list1 = %$hash1_ref; # @list1 is now ( Key,Value ) @list2 = ( A,1, %$hash1_ref, B,2 ); # @list2 is now ( A,1, Key,Value, B,2) %hash2 = ( A,1, %$hash1_ref, B,2 ); # @hash2 is a hash with 3 key/value pairs %hash2_ref = { A,1, %$hash1_ref, B,2 }; # %hash2_ref is ref to a hash with 3 key/value pairs Does that help? > Thanks. > Mike > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <20191-5@lion>; Fri, 15 Sep 1995 01:21:14 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 00:19:50 GMT Received: from relay-1.mail.demon.net by punt.demon.co.uk id aa29182; 15 Sep 95 1:19 +0100 Received: from gw.home.vix.com by relay-1.mail.demon.net id aa18966; 14 Sep 95 22:48 +0100 Received: by gw.home.vix.com id AA07143; Thu, 14 Sep 95 04:12:29 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07137; Thu, 14 Sep 95 04:12:27 -0700 Received: from venus.geoquest.slb.com (venus.montrouge.geoquest.slb.com) by gourou.montrouge.geoquest.slb.com (4.1/Montrouge-February94 - MTB) id AA13107; Thu, 14 Sep 95 13:20:42 +0200 Date: Thu, 14 Sep 95 13:20:42 +0200 From: Hayet MAKHLOUF Message-Id: <9509141120.AA13107@gourou.montrouge.geoquest.slb.com> To: perldb-interest@vix.com Subject: unsubscribe unsubscribe ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21959-1@lion>; Fri, 15 Sep 1995 07:21:28 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 00:52:01 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa12516; 15 Sep 95 1:51 +0100 Received: by gw.home.vix.com id AA13931; Thu, 14 Sep 95 06:12:57 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA13919; Thu, 14 Sep 95 06:12:53 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id IAA171584 for ; Thu, 14 Sep 1995 08:12:53 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA16154; Thu, 14 Sep 1995 08:12:53 -0500 From: mhm@austin.ibm.com Message-Id: <9509141312.AA16154@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Adding sqlstate to DBI...how? Date: Thu, 14 Sep 95 08:12:52 -0500 Ok, I give up. I haven't found the correct magically incantation to add sqlstate in with err and errstate. So, how do you do it? Thanks. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <21958-7@lion>; Fri, 15 Sep 1995 07:21:53 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 01:29:34 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id ab27655; 15 Sep 95 2:06 +0100 Received: by gw.home.vix.com id AA11157; Wed, 13 Sep 95 19:16:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA11152; Wed, 13 Sep 95 19:16:46 -0700 Received: by alta.COM (4.1/SMI-4.1+TAS1.10) id AA11867; Wed, 13 Sep 95 19:16:27 PDT From: Damian Martinez Message-Id: <9509140216.AA11867@alta.COM> Subject: ld.so: _stderr undefined symbol?! To: perldb-interest@vix.com Date: Wed, 13 Sep 1995 19:16:26 -0700 (PDT) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 6919 Hello to all, I have a feeling this is more of a compiler type problem, but this is the only place I can think of to come, esp since I only see this when doing the DBI tests... I recently sent a message detailing the fact that I couldn't get the DBI to compile on a Sun 4.1.3_U1 with gcc 2.5.8. Response seemed to point at my ancient compiler so I built and installed gcc 2.6.3 and libg and libg++ (whatever the latest versions were)... OK, so DBI-0.63 compiles allright (or seems to) but when i run the tests, ld.so complains about not finding the _stderr symbol (!?) D'oH! this sounds really bad and I'm no expert on anything to do with C or linking or whatever this might point to...Can anyone help? Here's some info on the make/make test/Configure... ____________________ basalt ~/perldb/DBI-0.63 % perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=sunos, osver=4.1.3_u1, archname=sun4-sunos uname='sunos basalt 4.1.3_u1 4 sun4c ' hint=recommended Compiler: cc='gcc', optimize='-O' cppflags='-DDEBUGGING' ccflags ='-DDEBUGGING' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=true voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=31 Libraries: so=so libpth=/lib /usr/lib /usr/ucblib /usr/local/lib libs=-ldbm -ldl -lm -lc -lposix libc=/usr/lib/libc.so.1.9.1 Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-fpic', ccdlflags=' ', lddlflags='-assert nodefinitions' basalt ~/perldb/DBI-0.63 % perl Makefile.PL Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker (v4.16) Checking if your kit is complete... Looks good DEFINE => '-Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion' NAME => 'DBI' VERSION => '0.63' Using PERL=/usr/local/bin/perl Writing Makefile for DBI basalt ~/perldb/DBI-0.63 % make mkdir ./blib mkdir ./blib/auto mkdir ./blib/auto/DBI /usr/local/bin/perl -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c gcc -c -DDEBUGGING -O -fpic -I/usr/local/lib/perl5/sun4-sunos/CORE -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion DBI.c In file included from /usr/local/include/netinet/in.h:23, from /usr/local/lib/perl5/sun4-sunos/CORE/perl.h:205, from DBIXS.h:13, from DBI.c:19: /usr/local/include/sys/socket.h:41: warning: comma at end of enumerator list /usr/local/include/sys/socket.h:180: warning: comma at end of enumerator list /usr/local/include/sys/socket.h:260: warning: comma at end of enumerator list In file included from /usr/local/lib/perl5/sun4-sunos/CORE/perl.h:205, from DBIXS.h:13, from DBI.c:19: /usr/local/include/netinet/in.h:40: warning: comma at end of enumerator list /usr/local/include/netinet/in.h:80: warning: comma at end of enumerator list DBI.c: In function `dbih_getcom': DBI.c:228: warning: cast increases required alignment of target type DBI.c:236: warning: cast increases required alignment of target type DBI.c: In function `dbih_setup_handle': DBI.c:314: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD_____st_fetchrow': DBI.c:1532: warning: cast increases required alignment of target type DBI.c:1532: warning: cast increases required alignment of target type DBI.c:1540: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD_____st_fetch': DBI.c:1593: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD___mem__common_DESTROY': DBI.c:1695: warning: cast increases required alignment of target type Running Mkbootstrap for DBI () chmod 644 DBI.bs LD_RUN_PATH="" ld -o blib/auto/DBI/DBI.so -assert nodefinitions DBI.o chmod 755 blib/auto/DBI/DBI.so cp DBI.bs ./blib/auto/DBI/DBI.bs chmod 644 blib/auto/DBI/DBI.bs mkdir ./blib/DBI cp DBIXS.h blib/DBI/DBIXS.h chmod 644 blib/DBI/DBIXS.h mkdir ./blib/DBD cp lib/DBD/ExampleP.pm blib/DBD/ExampleP.pm chmod 644 blib/DBD/ExampleP.pm cp lib/DBD/NullP.pm blib/DBD/NullP.pm chmod 644 blib/DBD/NullP.pm cp DBI.pm blib/DBI.pm chmod 644 blib/DBI.pm basalt ~/perldb/DBI-0.63 % make test /usr/local/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............ld.so: Undefined symbol: _stderr FAILED on test 1 Failed 1 test, 0.00% okay. Files=1, Tests=6, 3 secs ( 1.03 cusr 0.55 csys = 1.58 cpu) /usr/local/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ ld.so: Undefined symbol: _stderr make: *** [test_dynamic] Error 127 basalt ~/perldb/DBI-0.63 % make test TEST_VERBOSE=1 /usr/local/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............1..6 ld.so: Undefined symbol: _stderr FAILED on test 1 Failed 1 test, 0.00% okay. Files=1, Tests=6, 2 secs ( 1.05 cusr 0.53 csys = 1.58 cpu) /usr/local/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ ld.so: Undefined symbol: _stderr make: *** [test_dynamic] Error 127 Somone said my LD_LIBRARY_PATH might be wrong...here it is too: basalt ~/perldb/DBI-0.63 % echo $LD_LIBRARY_PATH /usr/local/lib:/usr/lib:/usr/X11R5/lib:/u/moc/lib:/usr/openwin/lib:/usr/X11R5/lib/X11/lib When I tried to rebuild the perl dist with the newer gcc, it died on this: basalt /src/packages/perl5.001m % make ... `sh cflags libperl.a pp_sys.o` pp_sys.c CCCMD = gcc -c -DDEBUGGING -O pp_sys.c: In function `pp_sysread': pp_sys.c:954: warning: passing arg 6 of `recvfrom' from incompatible pointer type pp_sys.c: In function `pp_accept': pp_sys.c:1484: warning: passing arg 3 of `accept' from incompatible pointer type pp_sys.c: In function `pp_ssockopt': pp_sys.c:1579: warning: passing arg 5 of `getsockopt' from incompatible pointer type pp_sys.c: In function `pp_getpeername': pp_sys.c:1646: warning: passing arg 3 of `getsockname' from incompatible pointer type pp_sys.c:1650: warning: passing arg 3 of `getpeername' from incompatible pointer type pp_sys.c: In function `pp_gpwent': pp_sys.c:3620: structure has no member named `pw_age' pp_sys.c:3629: structure has no member named `pw_comment' make: *** [pp_sys.o] Error 1 Anyone? Sorry for the long message... -Damian Martinez ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21959-16@lion>; Fri, 15 Sep 1995 07:24:48 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 02:10:25 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa12547; 15 Sep 95 3:09 +0100 Received: by gw.home.vix.com id AA27835; Thu, 14 Sep 95 15:26:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA27811; Thu, 14 Sep 95 15:25:55 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa27763; 14 Sep 95 23:19 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ab23294; 14 Sep 95 23:16 +0100 Received: from ig.co.uk by lion id <18386-0@lion>; Thu, 14 Sep 1995 23:08:34 +0100 To: mhm@austin.ibm.com Subject: Re: $sth->execute(@bind_values) question Cc: perldb-interest@vix.com Date: Thu, 14 Sep 1995 23:08:34 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509142316.ab23294@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > > > > > From: mhm@austin.ibm.com > > > > > > Under DBD::Oracle does $sth->execute() ... not no args... work if > > > $sth->bind_param($index_or_name,$var,$attribs) called first. > > > > > > At this point, DB2perl will work correctly if I use bind_param and > > > provide a list of bind values to execute. If I don't provide the > > > list of bind values then the data is corrupted going in. > > > > > You'll need to keep track of what's been bound so you can complain if > > any of the parameters have not been bound at execute time. > > > > DBD::Oracle doesn't do this since Oracle does it for you (oexec generates > > an error if anything has not been bound yet). > > Turns out that I have not been providing the correct address to > SQLBindParameter. It is appearantly of a local scope and disappears > between the call to $sth->bind_param and $sth->execute(). > > In the case of $sth->execute(arg1...argN) the memory address have been > used recently enough to still contain valid data when SQLExecute runs. > > Next question, how do I get the memory address of $var from within > my $sth->bind_param function when I don't know what the enduser named $var? > Sigh. Mike, we could go on like this for weeks. I have already said that the way to do this is to copy the value using sv_setsv() and then use the appropriate SvPVX/SvIVX/SvNVX etc macro to get the address to pass to SQLBindParameter. Use the code in DBD::Oracle to get it working for strings. Please try to learn from the code, both in DBD::Oracle and perl itself. Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <21958-19@lion>; Fri, 15 Sep 1995 07:25:12 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 03:33:39 GMT Received: from relay-1.mail.demon.net by punt2.demon.co.uk id aa20340; 15 Sep 95 1:48 +0100 Received: from gw.home.vix.com by relay-1.mail.demon.net id aa29023; 15 Sep 95 1:08 +0100 Received: by gw.home.vix.com id AA11384; Thu, 14 Sep 95 05:37:18 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA11379; Thu, 14 Sep 95 05:37:17 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA177478 for ; Thu, 14 Sep 1995 07:37:12 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA22926; Thu, 14 Sep 1995 07:37:12 -0500 From: mhm@austin.ibm.com Message-Id: <9509141237.AA22926@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Error Handling..... Date: Thu, 14 Sep 95 07:37:12 -0500 Ok, I finally made it to the point where I have to rework dbd_st_fetch and as I worked my way through I encountered all the places where DBD::Oracle is checking for specific return codes. The question is are they Oracle return codes or are they sql_state codes? For portability across databases, I would like to see it be the standardized SQLSTATE codes. But more of an impact on me is that DB2 only has 7 cli return codes, all of the diagnostic info is returned in the SQLSTATE codes. Any chance that DBDs could go to SQLSTATEs as we move to incorporating more of the CLI spec? Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21959-66@lion>; Fri, 15 Sep 1995 07:34:34 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 03:35:26 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ba10126; 15 Sep 95 4:34 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <16351-0@frigate.doc.ic.ac.uk>; Thu, 14 Sep 1995 22:43:59 +0100 Received: by gw.home.vix.com id AA03288; Tue, 12 Sep 95 10:14:20 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA03282; Tue, 12 Sep 95 10:14:17 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id MAA151238; Tue, 12 Sep 1995 12:14:16 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA15788; Tue, 12 Sep 1995 12:14:14 -0500 From: mhm@austin.ibm.com Message-Id: <9509121714.AA15788@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@vix.com Subject: Re: $sth->execute(@bind_values) question In-Reply-To: Your message of Tue, 12 Sep 95 15:31:47 +0100. <9509121607.ac23439@post.demon.co.uk> Date: Tue, 12 Sep 95 12:14:13 -0500 Tim Bunce writes: > > > From: mhm@austin.ibm.com > > > > Under DBD::Oracle does $sth->execute() ... not no args... work if > > $sth->bind_param($index_or_name,$var,$attribs) called first. > > > > At this point, DB2perl will work correctly if I use bind_param and > > provide a list of bind values to execute. If I don't provide the > > list of bind values then the data is corrupted going in. > > > You'll need to keep track of what's been bound so you can complain if > any of the parameters have not been bound at execute time. > > DBD::Oracle doesn't do this since Oracle does it for you (oexec generates > an error if anything has not been bound yet). > Turns out that I have not been providing the correct address to SQLBindParameter. It is appearantly of a local scope and disappears between the call to $sth->bind_param and $sth->execute(). In the case of $sth->execute(arg1...argN) the memory address have been used recently enough to still contain valid data when SQLExecute runs. Next question, how do I get the memory address of $var from within my $sth->bind_param function when I don't know what the enduser named $var? Thanks. Mike ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21959-87@lion>; Fri, 15 Sep 1995 07:37:24 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 04:55:10 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa07406; 15 Sep 95 5:54 +0100 Received: by gw.home.vix.com id AA05188; Tue, 12 Sep 95 22:31:09 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05174; Tue, 12 Sep 95 22:31:01 -0700 Received: from dokoka (dokoka.ucsb.edu) by hub.ucsb.edu; id AA12202 sendmail 4.1/UCSB-2.1-sun Tue, 12 Sep 95 22:25:22 PDT for perldb-interest@vix.com Received: from dokoka.ucsb.edu by dokoka via ESMTP (940816.SGI.8.6.9/940406.SGIv2) id WAA10073; Tue, 12 Sep 1995 22:25:17 -0700 Message-Id: <199509130525.WAA10073@dokoka> To: PerlDB Interest List , MiniSQL Interest List Subject: DBI connection management Reply-To: Alan Stebbens Date: Tue, 12 Sep 1995 22:25:16 -0700 From: Alan Stebbens Am I clueless, or is there no way to discover what databases exist for a given engine using DBI? In Msql (the Perl5 extension separate from DBI/DBD), you can do: $dbh = Connect Msql $host; @tables = ListDBs $dbh; With DBI, there doesn't seem to be any way to do this, even if I supported it in the DBD/mSQL, except, horribly, through the func_call mechanism. On a quasi-related note, I've been working with the perl Msql extension (0.63) , and there seems to be some bugs in it. I've fixed a couple, am not if I did it the correct way. * If you invoke "$dbh = Connect Msql;" (that is, with a default host), then there was a segvio caused by a strlen on the null "host" variable. * I also noticed that it didn't support the LENGTH column in the ListFields subroutine, so I added that to the hash list. * There wasn't a Makefile.PL file, so I kind of made one up. I'm not sure if the Makefile.SH is necessary or relevant still. * I added the IDENT_TYPE (but, what is it :^) Anyway, here're the patches, and I'd appreciate feedback. Thanks for msql, DBI/DBD/mSQL, and Msql. _______________________________________ Alan K. Stebbens College of Engineering University of California, Santa Barbara diff -rcw ext/Msql/Makefile.PL ext.aks/Msql/Makefile.PL *** ext/Msql/Makefile.PL Fri Sep 1 19:33:53 1995 --- ext.aks/Msql/Makefile.PL Tue Sep 12 02:29:59 1995 *************** *** 0 **** --- 1,10 ---- + use ExtUtils::MakeMaker; + # See lib/ExtUtils/MakeMaker.pm for details of how to influence + # the contents of the Makefile that is written. + WriteMakefile( + 'NAME' => 'Msql', + 'VERSION' => '1.0.8', + 'LIBS' => ['-L/eci/msql/lib -lmsql'], # e.g., '-lm' + 'DEFINE' => '', # e.g., '-DHAVE_SOMETHING' + 'INC' => '-I/eci/msql/include', # e.g., '-I/usr/include/other' + ); diff -rcw ext/Msql/Makefile.SH ext.aks/Msql/Makefile.SH *** ext/Msql/Makefile.SH Mon Sep 5 10:41:20 1994 --- ext.aks/Msql/Makefile.SH Fri Sep 1 18:46:19 1995 *************** *** 1,5 **** ! potential_libs="-L/usr/local/Minerva/lib -lmsql" ! msqlinc="/usr/local/Minerva/include" msqlversion=a1 --- 1,5 ---- ! potential_libs="-L/eci/msql/lib -lmsql" ! msqlinc="/eci/msql/include" msqlversion=a1 Only in ext.aks/Msql: Makefile.old Only in ext.aks/Msql: Msql.bs Only in ext.aks/Msql: Msql.c Only in ext.aks/Msql: Msql.o diff -rcw ext/Msql/Msql.pm ext.aks/Msql/Msql.pm *** ext/Msql/Msql.pm Sun Sep 4 08:04:37 1994 --- ext.aks/Msql/Msql.pm Tue Sep 12 02:39:46 1995 *************** *** 9,14 **** --- 9,15 ---- CHAR_TYPE INT_TYPE REAL_TYPE + IDENT_TYPE ); sub AUTOLOAD { diff -rcw ext/Msql/Msql.xs ext.aks/Msql/Msql.xs *** ext/Msql/Msql.xs Mon Sep 5 12:02:13 1994 --- ext.aks/Msql/Msql.xs Tue Sep 12 16:02:29 1995 *************** *** 52,57 **** --- 52,58 ---- dMSQL; \ AV * avkey; \ AV * avnam; \ + AV * avlen; \ AV * avnnl; \ AV * avtab; \ AV * avtyp; \ *************** *** 115,120 **** --- 116,122 ---- hv_store(hv,"NUMFIELDS",9,(SV *)newSViv((IV)msqlNumFields(result)),0); \ avnam = (AV*)(SV*)newAV(); \ avtab = (AV*)(SV*)newAV(); \ + avlen = (AV*)(SV*)newAV(); \ avtyp = (AV*)(SV*)newAV(); \ avkey = (AV*)(SV*)newAV(); \ avnnl = (AV*)(SV*)newAV(); \ *************** *** 123,128 **** --- 125,131 ---- curField = msqlFetchField(result); \ av_push(avnam,(SV*)newSVpv(curField->name,strlen(curField->name))); \ av_push(avtab,(SV*)newSVpv(curField->table,strlen(curField->table))); \ + av_push(avlen,(SV*)newSViv(curField->length)); \ av_push(avtyp,(SV*)newSViv(curField->type)); \ av_push(avkey,(SV*)newSViv(IS_PRI_KEY(curField->flags))); \ av_push(avnnl,(SV*)newSViv(IS_NOT_NULL(curField->flags))); \ *************** *** 130,135 **** --- 133,139 ---- } \ rv = newRV((SV*)avnam); hv_store(hv,"NAME",4,rv,0); \ rv = newRV((SV*)avtab); hv_store(hv,"TABLE",5,rv,0); \ + rv = newRV((SV*)avlen); hv_store(hv,"LENGTH",6,rv,0); \ rv = newRV((SV*)avtyp); hv_store(hv,"TYPE",4,rv,0); \ rv = newRV((SV*)avkey); hv_store(hv,"IS_PRI_KEY",10,rv,0); \ rv = newRV((SV*)avnnl); hv_store(hv,"IS_NOT_NULL",11,rv,0) *************** *** 179,184 **** --- 183,194 ---- #else goto not_there; #endif + if (strEQ(name, "IDENT_TYPE")) + #ifdef IDENT_TYPE + return IDENT_TYPE; + #else + goto not_there; + #endif break; case 'J': break; *************** *** 350,357 **** } else { my_iniHV; svsock = (SV*)newSViv(sock); ! svdb = (SV*)newSVpv(db,na); ! svhost = (SV*)newSVpv(host,strlen(host)); retMSQLSOCK; } } --- 360,367 ---- } else { my_iniHV; svsock = (SV*)newSViv(sock); ! svdb = db ? (SV*)newSVpv(db,na) : 0; ! svhost = host ? (SV*)newSVpv(host,strlen(host)) : 0; retMSQLSOCK; } } *************** *** 395,401 **** position-pointer with DataSeek(). We do a lot here: we investigate some info about the fields, that ! will be returned by this handle: NAME, TABLE, TYPE, IS_PRI_KEY, and IS_NOT_NULL */ dMSQLSTATE; --- 405,411 ---- position-pointer with DataSeek(). We do a lot here: we investigate some info about the fields, that ! will be returned by this handle: NAME, TABLE, LENGTH, TYPE, IS_PRI_KEY, and IS_NOT_NULL */ dMSQLSTATE; ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21959-89@lion>; Fri, 15 Sep 1995 07:38:05 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 04:58:03 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa08320; 15 Sep 95 5:57 +0100 Received: by gw.home.vix.com id AA05307; Tue, 12 Sep 95 22:39:08 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05302; Tue, 12 Sep 95 22:39:06 -0700 Received: from dokoka (dokoka.ucsb.edu) by hub.ucsb.edu; id AA12330 sendmail 4.1/UCSB-2.1-sun Tue, 12 Sep 95 22:38:46 PDT for perldb-interest@vix.com Received: from dokoka.ucsb.edu by dokoka via ESMTP (940816.SGI.8.6.9/940406.SGIv2) for id WAA10119; Tue, 12 Sep 1995 22:38:47 -0700 Message-Id: <199509130538.WAA10119@dokoka> To: PerlDB Interest List Subject: Another bug.. Reply-To: Alan Stebbens Date: Tue, 12 Sep 1995 22:38:45 -0700 From: Alan Stebbens In the code fragment below use Msql; ... defined($sdbh) or carp "\$sdbh not set!\n"; my (@databases) = ListDBs $sdbh; I'm getting the following error: "Use of uninitialized value at dbs line 640." The error is occuring on the "ListDBs $sdbh" line. I'm guessing that there is some internal variable not being initialized properly, possibly the "db" variable in the connection handle? Is there a way I can determine where, exactly, the error is being generated? AdThanksVance. _______________________________________ Alan K. Stebbens College of Engineering University of California, Santa Barbara ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24147-1@lion>; Fri, 15 Sep 1995 07:42:41 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 06:36:46 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa13716; 15 Sep 95 7:36 +0100 Received: by gw.home.vix.com id AA28395; Tue, 12 Sep 95 20:11:14 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA25170; Tue, 12 Sep 95 19:02:07 -0700 Received: from [205.197.236.11] (fozzie.cortex.net [205.197.236.11]) by dino.cortex.net (8.6.10/Cortex 1.1) with SMTP id WAA22755; Tue, 12 Sep 1995 22:00:48 -0400 Posted-Date: Tue, 12 Sep 1995 22:00:48 -0400 X-Sender: brent@205.197.236.2 Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 12 Sep 1995 22:08:06 -0400 To: perldb-interest@vix.com From: Brent Halliburton Cc: Tim.Bunce@ig.co.uk I am attempting to get DBI installed on my Sparc 20 running Solaris 2.4. It is none to happy when it hits test 5 but it doesn't core dump until a bit later, which makes the whole process a bit more interesting. I would really like to get this working pronto, so a quick diagnosis would be much appreciated. Brent Halliburton brent@cortex.net > make test TEST_VERBOSE=1 gcc -c -O -fpic -I/usr/local/lib/perl5/sun4-solaris/CORE -Wall -pedantic -Wno-c omment -Wpointer-arith -Wcast-align -Wconversion DBI.c DBI.c: In function `dbih_getcom': DBI.c:220: warning: cast increases required alignment of target type DBI.c:228: warning: cast increases required alignment of target type DBI.c: In function `dbih_setup_handle': DBI.c:306: warning: cast increases required alignment of target type DBI.c:318: warning: value computed is not used DBI.c:331: warning: value computed is not used DBI.c:337: warning: value computed is not used DBI.c:338: warning: value computed is not used DBI.c:339: warning: value computed is not used DBI.c: In function `dbih_sth_bind_col': DBI.c:518: warning: value computed is not used DBI.c: In function `dbih_set_attr': DBI.c:569: warning: value computed is not used DBI.c: In function `dbih_get_attr': DBI.c:604: warning: value computed is not used DBI.c: In function `XS_DBD_____st_fetchrow': DBI.c:1330: warning: cast increases required alignment of target type DBI.c:1330: warning: cast increases required alignment of target type DBI.c:1334: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD_____st_fetch': DBI.c:1367: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD___mem__common_DESTROY': DBI.c:1457: warning: cast increases required alignment of target type Running mkbootstrap for DBI () ld -o /import/dino/home/brent/dbi/DBI/blib/auto/DBI/DBI.so -G DBI.o cp DBI.bs /import/dino/home/brent/dbi/DBI/blib/auto/DBI/DBI.bs cp DBIXS.h /import/dino/home/brent/dbi/DBI/blib/DBI/DBIXS.h cp lib/DBD/ExampleP.pm /import/dino/home/brent/dbi/DBI/blib/DBD/ExampleP.pm cp lib/DBD/NullP.pm /import/dino/home/brent/dbi/DBI/blib/DBD/NullP.pm cp DBI.pm /import/dino/home/brent/dbi/DBI/blib/DBI.pm /usr/local/bin/perl -I/import/dino/home/brent/dbi/DBI/blib -I/import/dino/home/b rent/dbi/DBI/blib -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............1..6 ok 1 ok 2 ok 3 ok 4 FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=6, 5 secs ( 0.21 cusr 0.20 csys = 0.41 cpu) /usr/local/bin/perl -I/import/dino/home/brent/dbi/DBI/blib -I/import/dino/home/b rent/dbi/DBI/blib -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/lib/perl5 tes t.pl test.pl DBI test application $Revision: 1.12 $ *** Signal 11 - core dumped make: Fatal error: Command failed for target `test' ========================================== The stack trace on the generated core file: (gdb) where #0 0x49124 in pp_entersub () #1 0x1ccdc in perl_call_sv () #2 0x1c9b0 in perl_call_method () #3 0x40e90 in magic_methpack () #4 0x40efc in magic_getpack () #5 0x3fa7c in mg_get () #6 0x4b50c in sv_2pv () #7 0x450a0 in pp_concat () #8 0x44730 in run () #9 0x1c598 in perl_run () #10 0x1b4e8 in main () (gdb) ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24906-3@lion>; Fri, 15 Sep 1995 10:16:17 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 08:36:20 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa01085; 15 Sep 95 9:36 +0100 Received: by gw.home.vix.com id AA15694; Thu, 14 Sep 95 19:07:38 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA15689; Thu, 14 Sep 95 19:07:35 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa23880; 14 Sep 95 14:48 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa24332; 14 Sep 95 14:46 +0100 Received: from ig.co.uk by lion id <03244-0@lion>; Thu, 14 Sep 1995 14:09:03 +0100 To: mhm@austin.ibm.com, roehrich@cray.com Subject: Re: Building the %attribs for bind_param ? Cc: perldb-interest@vix.com Date: Thu, 14 Sep 1995 14:09:03 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509141446.aa24332@post.demon.co.uk> > From: Dean Roehrich > > >|> %hash2_ref = { A,1, %$hash1_ref, B,2 }; > >|> # %hash2_ref is ref to a hash with 3 key/value pairs > >|> Does that help? > > > >I think so, but I'll need to sleep on it before I fully understand it. > > > >Mike > > That last one should be: > > $hash2_ref = { A,1, %$hash1_ref, B,2 }; > # $hash2_ref is ref to a hash with 3 key/value pairs > Well spotted, thanks Dean. > Dean > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-7@lion>; Mon, 18 Sep 1995 07:21:17 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Fri, 15 Sep 95 20:08:47 GMT Received: from relay-1.mail.demon.net by punt.demon.co.uk id aa12053; 15 Sep 95 20:27 +0100 Received: from gw.home.vix.com by relay-1.mail.demon.net id aa21480; 15 Sep 95 19:17 +0100 Received: by gw.home.vix.com id AA02471; Fri, 15 Sep 95 00:04:18 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA02466; Fri, 15 Sep 95 00:04:12 -0700 Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id JAA14448; Fri, 15 Sep 1995 09:03:04 +0200 Date: Fri, 15 Sep 1995 09:03:04 +0200 Message-Id: <199509150703.JAA14448@anna.mind.de> From: Andreas Koenig To: aks@hub.ucsb.edu Cc: PerlDB Interest List In-Reply-To: <199509130538.WAA10119@dokoka> (message from Alan Stebbens on Tue, 12 Sep 1995 22:38:45 -0700) Subject: Re: Another bug.. Reply-To: andreas.koenig@mind.de >>>>> "Alan" == Alan Stebbens writes: Alan, I think there are very few people on perdb-interest using MsqlPerl. It's not completely off topic here, but better suited is msqlperl@franz.ww.tu-berlin.de, IMO. BTW, sending to me and CC-ing whoever you want is *much* better. Alan> In the code fragment below use Msql; Alan> ... Alan> defined($sdbh) or carp "\$sdbh not set!\n"; Alan> my(@databases) = ListDBs $sdbh; Alan> I'm getting the following error: Alan> "Use of uninitialized value at dbs line 640." Alan> The error is occuring on the "ListDBs $sdbh" line. Sorry, I can't reproduce that here, neither on IRIX 5.3 nor on NeXT 3.3. Try this program: % perl -lwe ' use Msql; my @foo = Msql->Connect()->ListDBs; ' I'm using MsqlPerl 1.02, perl 5.001m and msql 1.0.8. And you? Alan> Is there a way I can determine where, exactly, the error is Alan> being generated? Good question, I don't know. Alan> AdThanksVance. andreas ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-44@lion>; Mon, 18 Sep 1995 07:28:21 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 16 Sep 95 01:14:43 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa25618; 16 Sep 95 2:14 +0100 Received: by gw.home.vix.com id AA06958; Fri, 15 Sep 95 08:50:11 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA06940; Fri, 15 Sep 95 08:50:10 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id KAA43306; Fri, 15 Sep 1995 10:49:59 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for roehrich@cray.com at austin.ibm.com; id AA27474; Fri, 15 Sep 1995 10:49:57 -0500 From: mhm@austin.ibm.com Message-Id: <9509151549.AA27474@dax.austin.ibm.com> To: perldb-interest@vix.com Cc: roehrich@cray.com Subject: Another hash ref question... Date: Fri, 15 Sep 95 10:49:56 -0500 Tim.Bunce@ig.co.uk suggested: $new_attrib = { %$existing_attrib, db2_type => $new_type }; So, I attempted $attrib_dec = { %$attrib_char, 'Stype' => SQL_DECIMAL } ; which returns the perl error "Odd number of elements in hash list". Additionally, I substituted "," for "=>", same error. I don't see why it doesn't work. Have I found a perl bug? Or, more likely just being stupid? Thanks. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-51@lion>; Mon, 18 Sep 1995 07:29:11 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 16 Sep 95 02:10:50 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa18286; 16 Sep 95 3:09 +0100 Received: by gw.home.vix.com id AA01212; Fri, 15 Sep 95 14:09:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01206; Fri, 15 Sep 95 14:08:59 -0700 Received: by lafcol.lafayette.edu (4.1/SMI-4.1) id AA12948; Fri, 15 Sep 95 17:07:17 EDT Date: Fri, 15 Sep 1995 17:07:16 -0400 (EDT) From: Andy Dougherty Subject: Re: ld.so: _stderr undefined symbol?! To: Tim Bunce Cc: perldb-interest@vix.com, krak@alta.com, perl5-porters@africa.nicoh.com In-Reply-To: <9509151230.ae21515@post.demon.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 15 Sep 1995, Tim Bunce wrote: > > From: Damian Martinez > > > ld.so: Undefined symbol: _stderr > > > Somone said my LD_LIBRARY_PATH might be wrong...here it is too: > > basalt ~/perldb/DBI-0.63 % echo $LD_LIBRARY_PATH > > /usr/local/lib:/usr/lib:/usr/X11R5/lib:/u/moc/lib:/usr/openwin/lib:/usr/X11R5/lib/X11/lib Hmm. I thought including /usr/lib in LD_LIBRARY_PATH was the usual culprit for this sort of thing. Andy Dougherty doughera@lafcol.lafayette.edu Dept. of Physics Lafayette College, Easton PA 18042 ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-54@lion>; Mon, 18 Sep 1995 07:29:38 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 16 Sep 95 02:35:32 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa28590; 16 Sep 95 3:35 +0100 Received: by gw.home.vix.com id AA24332; Fri, 15 Sep 95 12:41:58 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA24327; Fri, 15 Sep 95 12:41:56 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id OAA176928; Fri, 15 Sep 1995 14:41:55 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA02882; Fri, 15 Sep 1995 14:41:53 -0500 From: mhm@austin.ibm.com Message-Id: <9509151941.AA02882@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Adding sqlstate to DBI...how? In-Reply-To: Your message of Fri, 15 Sep 95 12:25:35 +0100. <9509151230.af21515@post.demon.co.uk> Date: Fri, 15 Sep 95 14:41:52 -0500 Tim Bunce writes: > > > Subject: Adding sqlstate to DBI...how? > > > > Ok, I give up. I haven't found the correct magically incantation to > > add sqlstate in with err and errstate. So, how do you do it? > > > Do what? > make a variable called sqlstate coexist with err and errstate such that sqlstate is available to the perl side, and the C side of things. So far I have not been able to get sqlstate defined such that it is know to both sides without generating some form of either perl error or C compiler error. Mike ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-56@lion>; Mon, 18 Sep 1995 07:29:57 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 16 Sep 95 02:41:38 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa01047; 16 Sep 95 3:41 +0100 Received: by gw.home.vix.com id AA21558; Fri, 15 Sep 95 12:05:16 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA21551; Fri, 15 Sep 95 12:05:05 -0700 Received: by disperse.demon.co.uk id aa12269; 15 Sep 95 19:46 +0100 Received: from [158.152.1.72] by disperse.demon.co.uk id al01989; 15 Sep 95 15:24 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ae21515; 15 Sep 95 12:30 +0100 Received: from ig.co.uk by lion id <26271-0@lion>; Fri, 15 Sep 1995 12:21:14 +0100 To: perldb-interest@vix.com, krak@alta.com Subject: Re: ld.so: _stderr undefined symbol?! Cc: perl5-porters@nicoh.com Date: Fri, 15 Sep 1995 12:21:14 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509151230.ae21515@post.demon.co.uk> > From: Damian Martinez > > Hello to all, > > I have a feeling this is more of a compiler type problem, > but this is the only place I can think of to come, esp > since I only see this when doing the DBI tests... > I've CC'd this to perl5-porters since it includes perl build problems which may be related. > I recently sent a message detailing the fact that I couldn't > get the DBI to compile on a Sun 4.1.3_U1 with gcc 2.5.8. > Response seemed to point at my ancient compiler so I built > and installed gcc 2.6.3 and libg and libg++ (whatever the latest > versions were)... > libg may well be the problem here. > OK, so DBI-0.63 compiles allright (or seems to) but when i run the > tests, ld.so complains about not finding the _stderr symbol (!?) > D'oH! this sounds really bad and I'm no expert on anything to do > with C or linking or whatever this might point to...Can anyone help? > > Here's some info on the make/make test/Configure... > > ____________________ > > basalt ~/perldb/DBI-0.63 % perl -e 'use Config; print Config::myconfig()' > Summary of my perl5 (patchlevel 1) configuration: > Platform: > osname=sunos, osver=4.1.3_u1, archname=sun4-sunos > uname='sunos basalt 4.1.3_u1 4 sun4c ' > hint=recommended > Compiler: > cc='gcc', optimize='-O' > cppflags='-DDEBUGGING' > ccflags ='-DDEBUGGING' > ldflags ='' > stdchar='unsigned char', d_stdstdio=define, usevfork=true > voidflags=15, castflags=0, d_casti32=define, d_castneg=define > intsize=4, alignbytes=8, usemymalloc=y, randbits=31 > Libraries: > so=so > libpth=/lib /usr/lib /usr/ucblib /usr/local/lib > libs=-ldbm -ldl -lm -lc -lposix > libc=/usr/lib/libc.so.1.9.1 > Dynamic Linking: > dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= > cccdlflags='-fpic', ccdlflags=' ', lddlflags='-assert nodefinitions' > gcc -c -DDEBUGGING -O -fpic -I/usr/local/lib/perl5/sun4-sunos/CORE -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion DBI.c > In file included from /usr/local/include/netinet/in.h:23, > from /usr/local/lib/perl5/sun4-sunos/CORE/perl.h:205, > from DBIXS.h:13, > from DBI.c:19: > /usr/local/include/sys/socket.h:41: warning: comma at end of enumerator list > [...more similar errors...] > ld.so: Undefined symbol: _stderr > Somone said my LD_LIBRARY_PATH might be wrong...here it is too: > basalt ~/perldb/DBI-0.63 % echo $LD_LIBRARY_PATH > /usr/local/lib:/usr/lib:/usr/X11R5/lib:/u/moc/lib:/usr/openwin/lib:/usr/X11R5/lib/X11/lib > > When I tried to rebuild the perl dist with the newer gcc, it died on this: > basalt /src/packages/perl5.001m % make > ... > `sh cflags libperl.a pp_sys.o` pp_sys.c > CCCMD = gcc -c -DDEBUGGING -O > pp_sys.c: In function `pp_sysread': > pp_sys.c:954: warning: passing arg 6 of `recvfrom' from incompatible pointer type > pp_sys.c: In function `pp_gpwent': > pp_sys.c:3620: structure has no member named `pw_age' > pp_sys.c:3629: structure has no member named `pw_comment' > [...more similar errors...] My guess is that libg is causing your problems. Libg offers little on suns and is probably best avoided. I don't know how difficult it is to stop gcc trying to use it. You should get perl built and installed cleanly using the new compiler before trying to build extensions with it. It's probable that an extension built using libg is not compatible with a perl built using libc. Don't forget to re-Configure (probably best to delete config.sh and just do a Configure -des). Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-58@lion>; Mon, 18 Sep 1995 07:30:26 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 16 Sep 95 02:59:46 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa07529; 16 Sep 95 3:59 +0100 Received: by gw.home.vix.com id AA20878; Fri, 15 Sep 95 11:56:57 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA20867; Fri, 15 Sep 95 11:56:55 -0700 Received: by disperse.demon.co.uk id ab12186; 15 Sep 95 19:44 +0100 Received: from [158.152.1.72] by disperse.demon.co.uk id av01328; 15 Sep 95 15:24 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id af21515; 15 Sep 95 12:30 +0100 Received: from ig.co.uk by lion id <26306-0@lion>; Fri, 15 Sep 1995 12:25:35 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Adding sqlstate to DBI...how? Date: Fri, 15 Sep 1995 12:25:35 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509151230.af21515@post.demon.co.uk> > Subject: Adding sqlstate to DBI...how? > > Ok, I give up. I haven't found the correct magically incantation to > add sqlstate in with err and errstate. So, how do you do it? > Do what? Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-66@lion>; Mon, 18 Sep 1995 07:32:16 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 16 Sep 95 07:34:09 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa10929; 16 Sep 95 8:33 +0100 Received: by gw.home.vix.com id AA26087; Fri, 15 Sep 95 21:08:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA26082; Fri, 15 Sep 95 21:08:22 -0700 Received: by disperse.demon.co.uk id ak11891; 15 Sep 95 19:44 +0100 Received: from [158.152.1.72] by disperse.demon.co.uk id ak01918; 15 Sep 95 15:24 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ac21515; 15 Sep 95 12:30 +0100 Received: from ig.co.uk by lion id <26084-0@lion>; Fri, 15 Sep 1995 11:55:28 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Error Handling..... Date: Fri, 15 Sep 1995 11:55:28 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509151230.ac21515@post.demon.co.uk> > From: mhm@austin.ibm.com > > Ok, I finally made it to the point where I have to rework dbd_st_fetch > and as I worked my way through I encountered all the places where > DBD::Oracle is checking for specific return codes. The question is > are they Oracle return codes or are they sql_state codes? > They are Oracle specific. > For portability across databases, I would like to see it be the > standardized SQLSTATE codes. So would I. > But more of an impact on me is that DB2 > only has 7 cli return codes, all of the diagnostic info is returned > in the SQLSTATE codes. > > Any chance that DBDs could go to SQLSTATEs as we move to incorporating > more of the CLI spec? > Yes, but it's not easy. Oracle has an entire book listing thousands of error code and what they mean. It's unrealistic to map all of those into SQLSTATE code. Some compromise will have to be worked out. The *entire* error handling scheme in the DBI is incompletely designed. It'll change to align with the CLI spec. > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <11330-68@lion>; Mon, 18 Sep 1995 07:32:32 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Sat, 16 Sep 95 08:35:37 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa07103; 16 Sep 95 9:35 +0100 Received: by gw.home.vix.com id AA26112; Fri, 15 Sep 95 21:08:56 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA26096; Fri, 15 Sep 95 21:08:47 -0700 Received: by disperse.demon.co.uk id ad12057; 15 Sep 95 19:45 +0100 Received: from [158.152.1.72] by disperse.demon.co.uk id at01437; 15 Sep 95 15:24 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa21515; 15 Sep 95 12:30 +0100 Received: from ig.co.uk by lion id <25717-0@lion>; Fri, 15 Sep 1995 11:30:30 +0100 To: perldb-interest@vix.com, msql-list@bond.edu.au, aks@hub.ucsb.edu Subject: Re: DBI connection management Date: Fri, 15 Sep 1995 11:30:30 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509151230.aa21515@post.demon.co.uk> > From: Alan Stebbens > > Am I clueless, or is there no way to discover what databases exist > for a given engine using DBI? > Not yet. SQL CLI defines DataSources() so DBI will also define that at some point. Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <20799-0@lion>; Mon, 18 Sep 1995 21:08:57 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 18 Sep 95 19:42:25 GMT Received: from gw.home.vix.com by punt2.demon.co.uk id aa28319; 18 Sep 95 20:41 +0100 Received: by gw.home.vix.com id AA16559; Mon, 18 Sep 95 05:21:45 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA16554; Mon, 18 Sep 95 05:21:43 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA115123; Mon, 18 Sep 1995 07:21:42 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA27070; Mon, 18 Sep 1995 07:21:37 -0500 From: mhm@austin.ibm.com Message-Id: <9509181221.AA27070@dax.austin.ibm.com> To: perldb-interest@vix.com, roehrich@cray.com Subject: Re: Another hash ref question... In-Reply-To: Your message of Mon, 18 Sep 95 12:21:16 +0100. <9509181237.ae05163@post.demon.co.uk> Date: Mon, 18 Sep 95 07:21:37 -0500 Tim Bunce writes: > > > From: mhm@austin.ibm.com > > > > Tim.Bunce@ig.co.uk suggested: > > > > $new_attrib = { %$existing_attrib, db2_type => $new_type }; > > > > So, I attempted > > > > $attrib_dec = { %$attrib_char, 'Stype' => SQL_DECIMAL } ; > > > > which returns the perl error "Odd number of elements in hash list". > > Additionally, I substituted "," for "=>", same error. > > > > I don't see why it doesn't work. Have I found a perl bug? Or, more > > likely just being stupid? > > > Run the code with perl -w, it'll probably say 'Use of uninitialized > value'. > (The number of times I've been tempted to make 'use DBI' set $^W=1 ...) I am running with perl -w and I DON'T get the 'Use of uninitialized value' message. So perl atleast beleives it it initialized. > The "Odd number of elements in hash list" is a bug in perl related to > the expansion of @$ref and %$ref where $ref is undefined. I think it's > been reported but I couldn't see it in a quick scan of the bug list so > I've CC'd this to perlbug@perl.com. > > Your next problem is why $attrib_char is undefined ... It appears that it is uninitialized because it is imported from a module. If I move the defininition into the current script everything works as expected, but if I do it in a seperate module which is loaded by my script it is not recognized as a hash ref. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <21761-9@lion>; Mon, 18 Sep 1995 23:32:05 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 18 Sep 95 21:12:48 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa29087; 18 Sep 95 22:12 +0100 Received: by gw.home.vix.com id AA13958; Mon, 18 Sep 95 04:49:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA13949; Mon, 18 Sep 95 04:49:00 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa09849; 18 Sep 95 12:40 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ae05163; 18 Sep 95 12:37 +0100 Received: from ig.co.uk by lion id <16736-0@lion>; Mon, 18 Sep 1995 12:21:16 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Another hash ref question... Cc: roehrich@cray.com, perlbug@perl.com Date: Mon, 18 Sep 1995 12:21:16 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509181237.ae05163@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim.Bunce@ig.co.uk suggested: > > $new_attrib = { %$existing_attrib, db2_type => $new_type }; > > So, I attempted > > $attrib_dec = { %$attrib_char, 'Stype' => SQL_DECIMAL } ; > > which returns the perl error "Odd number of elements in hash list". > Additionally, I substituted "," for "=>", same error. > > I don't see why it doesn't work. Have I found a perl bug? Or, more > likely just being stupid? > Run the code with perl -w, it'll probably say 'Use of uninitialized value'. (The number of times I've been tempted to make 'use DBI' set $^W=1 ...) The "Odd number of elements in hash list" is a bug in perl related to the expansion of @$ref and %$ref where $ref is undefined. I think it's been reported but I couldn't see it in a quick scan of the bug list so I've CC'd this to perlbug@perl.com. Your next problem is why $attrib_char is undefined ... Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <22284-0@lion>; Mon, 18 Sep 1995 23:36:08 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 18 Sep 95 22:34:42 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa09207; 18 Sep 95 23:34 +0100 Received: by gw.home.vix.com id AA05702; Mon, 18 Sep 95 10:33:28 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA05697; Mon, 18 Sep 95 10:33:26 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa09663; 18 Sep 95 17:39 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa18848; 18 Sep 95 17:37 +0100 Received: from ig.co.uk by lion id <19350-0@lion>; Mon, 18 Sep 1995 17:39:27 +0100 To: lwall@scalpel.netlabs.com, mhm@austin.ibm.com Subject: Re: Another hash ref question... Cc: perldb-interest@vix.com Date: Mon, 18 Sep 1995 17:39:27 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509181737.aa18848@post.demon.co.uk> > From: mhm@austin.ibm.com > > > Larry Wall writes: > > I fixed this one back in May. > > > > NETaa14160: @$undef should produce null list in list context > > Not exactly the same problem. My problem is that the hash ref is > defined and exported from a module that is being used by the current > script. > Well you actually have two problems. Larry's given you the fix for the second. > use Some_package qw( $hash_ref ); > > $attrib_dec = { %$hash_ref, 'Stype' => SQL_DECIMAL } ; > > fails. The question/problem is why is the hash_ref undefined when > it has been defined in Some_package and exported with EXPORT_OK ? > Your code: package DBD::DB2; require DBI; require DynaLoader; @ISA = qw(DynaLoader); @EXPORT_OK = qw($attrib_int ); you should have said: @ISA = qw(Exporter DynaLoader); > Mike > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-4@lion>; Tue, 19 Sep 1995 07:21:04 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 18 Sep 95 23:29:16 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ad05407; 19 Sep 95 0:28 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <21708-0@frigate.doc.ic.ac.uk>; Mon, 18 Sep 1995 23:12:53 +0100 Received: by gw.home.vix.com id AA23776; Mon, 18 Sep 95 07:26:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA23770; Mon, 18 Sep 95 07:26:42 -0700 Received: from WMX.WMDATA.SE (wmx.wmdata.se [164.9.179.100]) by linda.fdata.se (8.6.12/8.6.9) with SMTP id QAA12661 for ; Mon, 18 Sep 1995 16:23:16 +0200 X400-Received: by /PRMD=WMDATAWMX/ADMD=WMDATA/C=SE/; Relayed; Mon, 18 Sep 1995 16:13:24 +0100 Date: Mon, 18 Sep 1995 16:13:24 +0100 X400-Originator: Johan.Widingsjo@STO4.wmdata.se X400-Recipients: perldb-interest@vix.com X400-Mts-Identifier: [/PRMD=WMDATA/ADMD=WMDATA/C=SE/;0012400001541748000002] X400-Content-Type: P2-1988 (22) Content-Identifier: CSI NC V3.0 From: "Widingsjo, Johan" Message-Id: <0004BC61.MAI*/S=WMJOWID/OU=STO4/OU=WMDATA/O=MSMAIL/PRMD=WMDATA/ADMD=WMDATA/C=SE/@MHS> To: 'PerlDB-Interest' Subject: Select from Oracle Hello, I'm very new to this and have only a little experience of perl prior to this, (no oop in perl, only c++, smalltalk and simula). It seems to me when reading the dbispec v0.6 that things have changed around quite a bit. So I'm wondering if there's any examples or helpfiles. Or if someone could take a quick look at my test code and see what I'm doing wrong. I'm connecting to an Oracle 7.1.6 database from a solaris 2.4 system. The execute statement returns 1 and the fetch I don't know what it does :-) The array comes out empty even tho I have a couple of records in the table and the do function can insert more records. Any help would be appreciated /Johan #!/usr/local/bin/perl -w use DBI; use DBD::Oracle; $dbh = DBI->connect( "xxxx", "xxxx", "xxxx", "Oracle"); print "Databas: $dbh\n"; # This line actually inserts a record in the database # So it seems to be connecting ok $status = $dbh->do("insert into test (ID, TEXT) values (2,'insert from perl') "); $st=$dbh->prepare("select * from test"); print "ST: $st\n"; $rc=$st->execute(); print "Execute returns: $rc\n"; @datat = $st->fetch(); print "#: $#datat\n"; print "Data: @datat\n"; $st->finish(); $dbh->disconnect(); ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-7@lion>; Tue, 19 Sep 1995 07:21:20 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Mon, 18 Sep 95 23:29:21 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ae05407; 19 Sep 95 0:28 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <21804-0@frigate.doc.ic.ac.uk>; Mon, 18 Sep 1995 23:16:49 +0100 Received: by gw.home.vix.com id AA13926; Mon, 18 Sep 95 04:46:38 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA13920; Mon, 18 Sep 95 04:46:36 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa09851; 18 Sep 95 12:40 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id af05163; 18 Sep 95 12:37 +0100 Received: from ig.co.uk by lion id <16821-0@lion>; Mon, 18 Sep 1995 12:36:30 +0100 To: mhm@austin.ibm.com Subject: Re: Adding sqlstate to DBI...how? Cc: perldb-interest@vix.com, roehrich@cray.com Date: Mon, 18 Sep 1995 12:36:30 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509181237.af05163@post.demon.co.uk> > From: mhm@austin.ibm.com > > Tim Bunce writes: > > > > > Subject: Adding sqlstate to DBI...how? > > > > > > Ok, I give up. I haven't found the correct magically incantation to > > > add sqlstate in with err and errstate. So, how do you do it? > > > > > Do what? > > > > make a variable called sqlstate coexist with err and errstate such that > sqlstate is available to the perl side, and the C side of things. So > far I have not been able to get sqlstate defined such that it is know > to both sides without generating some form of either perl error or > C compiler error. > Take a look at how the DBI implements DBI::err and especially DBI::errstr (lines 35 thru 40 in DBI.pm and DBI::var::FETCH in DBI.xs). The DBI has a special mechanism for this since the big problem is how to relate the variable to the correct driver. The methods suggested by Dean are great for normal situations but not appropriate for the DBI. To cut a long story short here's what you need to do to implement $DBI::sqlstate: tie $DBI::sqlstate, 'DBI::var', '&sqlstate'; # call &sqlstate in last used pkg and then write an sqlstate method for your DBD::DB2::db and DBD::DB2::st classes. > Mike > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-47@lion>; Tue, 19 Sep 1995 07:26:56 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 01:52:08 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa03652; 19 Sep 95 2:51 +0100 Received: by gw.home.vix.com id AA07474; Mon, 18 Sep 95 11:01:29 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA07468; Mon, 18 Sep 95 11:01:27 -0700 Received: from rijken.inter.NL.net by altrade.nijmegen.inter.nl.net via root@ah99-11.Arnhem.NL.net [193.79.228.236] with ESMTP id TAA22333 (8.6.10/1.5); Mon, 18 Sep 1995 19:58:55 +0200 Received: (from mrijken@localhost) by rijken.inter.NL.net (8.6.11/8.6.11) id TAA00164; Mon, 18 Sep 1995 19:52:56 +0200 Date: Mon, 18 Sep 1995 19:52:55 +0200 (MET DST) From: Marti Rijken To: perldb-interest@vix.com Cc: descarte@mcqueen.com Subject: DBD-mSQL-0.60pl2 test failed Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I picked up DBD-mSQL-0.60pl2 and DBI-0.63 from ftp.mcqueen.com/pub/languages/perl5/DBPerl/ and I'm trying to get them running with perl5.001m on linux-1.2.13. Perl was installed successfully, failing only on the test related to a known bash bug. DBI compiled and tested OK, and was subsequently installed. But DBD-mSQL is giving me trouble. At first some compiler warnings: mSQL.c: In function `XS_DBD__mSQL__db_FETCH': mSQL.c:159: warning: initialization makes pointer from integer without a cast mSQL.c: In function `XS_DBD__mSQL__st_readblob': mSQL.c:476: warning: assignment makes pointer from integer without a cast mSQL.c: In function `XS_DBD__mSQL__st_FETCH': mSQL.c:517: warning: initialization makes pointer from integer without a cast After I applied Andreas Koenig's patch from 5 september to mSQL.xs these warnings were gone and compilation completed sucsessfully. The test however failed in the very beginning: Can't locate Msql.pm in @INC at t/msql.t line 6. BEGIN failed--compilation aborted at t/msql.t line 6 That's where it writes `use Msql'. Msql..?? I found a file Msql.pm in the MsqlPerl-1.02 package, but not in DBD-mSQL. I tried changing it to `use mSQL', but this didn't get me much further: Undefined symbols present after loading ./blib/auto/DBD/mSQL/mSQL.o: DBIc_ENDING dbd_st_readblob at mSQL.pm line 18 Anyone has an idea if I'm doing something wrong? I appended a complete typescript of compilation and test at the and of this message. BTW. My ultimate goal is writing scripts that with only minor modifications will work with both DBD-mSQL (at home) and DBD-Oracle (at work). I hope this is the right assumption? (I know it's still Alpha). I have some working oraperl scripts that I hope will continue to work with the oraperl emulation of DBD-Oracle. Will this oraperl emulation work with DBD-mSLQ after some modification? Anyone tried? Thanks for your help, Marti ___ Marti Rijken ____ mrijken@inter.nl.net ____ Arnhem, Netherlands ___ _______ Province of Gelderland, nature conservation department ______ typescript: Script started on Sun Sep 17 21:39:38 1995 # perl Makefile.PS Configuring DBD::mSQL version 0.60pl1... Remember to actually read the README file! $MSQL_HOME not defined. Searching for mSQL... Using mSQL in /usr/local/Minerva mSQL sysliblist: -L/usr/local/Minerva/lib -lmsql -lm System: perl5.001 linux rijken 1.2.13 #2 sat aug 19 12:07:17 met dst 1995 i486 i486-linux dl_dld.xs Compiler: cc -O2 -D__USE_BSD_SIGNAL -Dbool=char -DHAS_BOOL Writing Makefile for DBD::mSQL # make mkdir ./blib mkdir ./blib/DBD mkdir ./blib/auto mkdir ./blib/auto/DBD mkdir ./blib/auto/DBD/mSQL cc -c -D__USE_BSD_SIGNAL -Dbool=char -DHAS_BOOL -O2 -I/usr/local/lib/perl5/i486-linux/CORE -I/usr/local/Minerva/include -I/usr/local/lib/perl5/i486-linux/DBI dbdimp.c /usr/bin/perl -I/usr/local/lib/perl5/i486-linux -I/usr/local/lib/perl5 /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap mSQL.xs >mSQL.tc && mv mSQL.tc mSQL.c cc -c -D__USE_BSD_SIGNAL -Dbool=char -DHAS_BOOL -O2 -I/usr/local/lib/perl5/i486-linux/CORE -I/usr/local/Minerva/include -I/usr/local/lib/perl5/i486-linux/DBI mSQL.c Running Mkbootstrap for DBD::mSQL (/usr/local/Minerva/lib/libmsql.a /usr/lib/libm.sa) Writing mSQL.bs chmod 644 mSQL.bs LD_RUN_PATH="/usr/local/Minerva/lib" ld -o blib/auto/DBD/mSQL/mSQL.o -r dbdimp.o mSQL.o -L/lib -L/usr/local/Minerva/lib -L/usr/local/Minerva/lib chmod 755 blib/auto/DBD/mSQL/mSQL.o cp mSQL.bs ./blib/auto/DBD/mSQL/mSQL.bs chmod 644 blib/auto/DBD/mSQL/mSQL.bs cp mSQL.pm blib/DBD/mSQL.pm chmod 644 blib/DBD/mSQL.pm cp test2.pl blib/DBD/test2.pl chmod 644 blib/DBD/test2.pl # make test /usr/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/i486-linux -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/msql..............Can't locate Msql.pm in @INC at t/msql.t line 6. BEGIN failed--compilation aborted at t/msql.t line 6. FAILED on test 0 Failed 1 test, 0.00% okay. Files=0, Tests=0, 0 secs ( 0.02 cusr 0.06 csys = 0.08 cpu) Script started on Sun Sep 17 21:42:11 1995 # make test /usr/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/i486-linux -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/msql..............Undefined symbols present after loading ./blib/auto/DBD/mSQL/mSQL.o: DBIc_ENDING dbd_st_readblob at mSQL.pm line 18 FAILED on test 0 Failed 1 test, 0.00% okay. Files=0, Tests=0, 2 secs ( 1.26 cusr 0.27 csys = 1.53 cpu) # perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=linux, osver=1, archname=i486-linux uname='linux rijken 1.2.13 #2 sat aug 19 12:07:17 met dst 1995 i486' hint=recommended Compiler: cc='cc', optimize='-O2' cppflags='-D__USE_BSD_SIGNAL -Dbool=char -DHAS_BOOL' ccflags ='-D__USE_BSD_SIGNAL -Dbool=char -DHAS_BOOL' ldflags ='' stdchar='char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=, d_castneg=define intsize=4, alignbytes=4, usemymalloc=n, randbits=31 Libraries: so=sa libpth=/lib /usr/lib /usr/local/lib libs=-lgdbm -ldbm -ldb -ldld -lm -lc -lbsd libc=/usr/lib/libc.sa Dynamic Linking: dlsrc=dl_dld.xs, dlext=o, d_dlsymun= cccdlflags=' ', ccdlflags=' ', lddlflags='-r ' ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-63@lion>; Tue, 19 Sep 1995 07:28:38 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 02:26:08 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa16275; 19 Sep 95 3:25 +0100 Received: by gw.home.vix.com id AA01024; Mon, 18 Sep 95 09:22:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA01019; Mon, 18 Sep 95 09:22:04 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id LAA111883; Mon, 18 Sep 1995 11:21:06 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for Tim.Bunce@ig.co.uk at austin.ibm.com; id AA27286; Mon, 18 Sep 1995 11:20:52 -0500 From: mhm@austin.ibm.com Message-Id: <9509181620.AA27286@dax.austin.ibm.com> To: Larry Wall Cc: Tim Bunce , perldb-interest@vix.com, mhm@austin.ibm.com, roehrich@cray.com, perlbug@perl.com Subject: Re: Another hash ref question... In-Reply-To: Your message of Mon, 18 Sep 95 08:56:30 -0700. <9509181556.AA28166@scalpel.netlabs.com> Date: Mon, 18 Sep 95 11:20:51 -0500 Larry Wall writes: > I fixed this one back in May. > > NETaa14160: @$undef should produce null list in list context > > The resolution was > > It didn't check for list context before returning undef. > > Here's the "unofficial" patch: > > *** pp_hot.c@@/main/25 Sat May 27 22:42:22 1995 > --- pp_hot.c Sun May 28 21:32:04 1995 > *************** > *** 360,365 **** > --- 360,367 ---- > if (op->op_flags & OPf_REF || > op->op_private & HINT_STRICT_REFS) > DIE(no_usym, "an ARRAY"); > + if (GIMME == G_ARRAY) > + RETURN; > RETPUSHUNDEF; > } > sym = SvPV(sv,na); > *************** > *** 431,436 **** > --- 433,442 ---- > if (op->op_flags & OPf_REF || > op->op_private & HINT_STRICT_REFS) > DIE(no_usym, "a HASH"); > + if (GIMME == G_ARRAY) { > + SP--; > + RETURN; > + } > RETSETUNDEF; > } > sym = SvPV(sv,na); > Not exactly the same problem. My problem is that the hash ref is defined and exported from a module that is being used by the current script. use Some_package qw( $hash_ref ); $attrib_dec = { %$hash_ref, 'Stype' => SQL_DECIMAL } ; fails. The question/problem is why is the hash_ref undefined when it has been defined in Some_package and exported with EXPORT_OK ? Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-65@lion>; Tue, 19 Sep 1995 07:28:51 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 02:28:43 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa17337; 19 Sep 95 3:28 +0100 Received: by gw.home.vix.com id AA13665; Mon, 18 Sep 95 12:23:08 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA13660; Mon, 18 Sep 95 12:23:07 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id OAA154240; Mon, 18 Sep 1995 14:23:05 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for lwall@scalpel.netlabs.com at austin.ibm.com; id AA03818; Mon, 18 Sep 1995 14:23:04 -0500 From: mhm@austin.ibm.com Message-Id: <9509181923.AA03818@dax.austin.ibm.com> To: perldb-interest@vix.com Cc: lwall@scalpel.netlabs.com Subject: Re: Another hash ref question... In-Reply-To: Your message of Mon, 18 Sep 95 17:39:27 +0100. <9509181737.aa18848@post.demon.co.uk> Date: Mon, 18 Sep 95 14:23:04 -0500 Tim Bunce writes: |> Your code: |> |> package DBD::DB2; |> require DBI; |> require DynaLoader; |> @ISA = qw(DynaLoader); |> @EXPORT_OK = qw($attrib_int ); |> |> you should have said: |> |> @ISA = qw(Exporter DynaLoader); Well, I did ask if it was a bug or something stupid on my part :-) Thanks. Mike ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-67@lion>; Tue, 19 Sep 1995 07:29:03 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 02:34:11 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa19178; 19 Sep 95 3:33 +0100 Received: by gw.home.vix.com id AA04527; Mon, 18 Sep 95 10:15:09 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA04522; Mon, 18 Sep 95 10:15:08 -0700 Received: from dokoka (dokoka.ucsb.edu) by hub.ucsb.edu; id AA03491 sendmail 4.1/UCSB-2.1-sun Mon, 18 Sep 95 10:15:05 PDT for perldb-interest@vix.com Received: from dokoka.ucsb.edu by dokoka via ESMTP (940816.SGI.8.6.9/940406.SGIv2) id KAA20127; Mon, 18 Sep 1995 10:15:03 -0700 Message-Id: <199509181715.KAA20127@dokoka> To: andreas.koenig@mind.de Cc: PerlDB Interest List Subject: Re: Another bug.. Reply-To: "Alan K. Stebbens" In-Reply-To: Your message of Fri, 15 Sep 1995 09:03:04 PDT. <199509150703.JAA14448@anna.mind.de> Date: Mon, 18 Sep 1995 10:14:54 -0700 From: Alan Stebbens > I'm using MsqlPerl 1.02, perl 5.001m and msql 1.0.8. And you? Ok. Where is MsqlPerl 1.02? As you noticed before, I've got a much older version. I do have perl5.001m, and msql-1.0.8. > Alan> Is there a way I can determine where, exactly, the error is > Alan> being generated? > > Good question, I don't know. Too bad. Thanks for the pointers. Alan ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-69@lion>; Tue, 19 Sep 1995 07:29:15 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 02:41:52 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa21234; 19 Sep 95 3:40 +0100 Received: by gw.home.vix.com id AA29302; Mon, 18 Sep 95 08:58:44 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29297; Mon, 18 Sep 95 08:58:43 -0700 Received: by proxy.netlabs.com (4.1/SMI-4.1) id AA08137; Mon, 18 Sep 95 09:00:38 PDT Received: from vaccine-bb.netlabs.com(192.190.199.1) by proxy.netlabs.com via smap (V1.3) id sma008135; Mon Sep 18 09:00:23 1995 Received: from scalpel.netlabs.com by netlabs.com (4.1/SMI-4.1) id AA01786; Mon, 18 Sep 95 08:57:20 PDT Received: from localhost.netlabs.com by scalpel.netlabs.com (4.1/SMI-4.1) id AA28166; Mon, 18 Sep 95 08:56:30 PDT Message-Id: <9509181556.AA28166@scalpel.netlabs.com> To: Tim Bunce Cc: perldb-interest@vix.com, mhm@austin.ibm.com, roehrich@cray.com, perlbug@perl.com Subject: Re: Another hash ref question... In-Reply-To: Your message of "Mon, 18 Sep 95 12:21:16 BST." <9509181237.ae05163@post.demon.co.uk> Date: Mon, 18 Sep 95 08:56:30 -0700 From: Larry Wall : > From: mhm@austin.ibm.com : > : > Tim.Bunce@ig.co.uk suggested: : > : > $new_attrib = { %$existing_attrib, db2_type => $new_type }; : > : > So, I attempted : > : > $attrib_dec = { %$attrib_char, 'Stype' => SQL_DECIMAL } ; : > : > which returns the perl error "Odd number of elements in hash list". : > Additionally, I substituted "," for "=>", same error. : > : > I don't see why it doesn't work. Have I found a perl bug? Or, more : > likely just being stupid? : > : Run the code with perl -w, it'll probably say 'Use of uninitialized value'. : (The number of times I've been tempted to make 'use DBI' set $^W=1 ...) : : The "Odd number of elements in hash list" is a bug in perl related to : the expansion of @$ref and %$ref where $ref is undefined. I think it's : been reported but I couldn't see it in a quick scan of the bug list so : I've CC'd this to perlbug@perl.com. I fixed this one back in May. NETaa14160: @$undef should produce null list in list context The resolution was It didn't check for list context before returning undef. Here's the "unofficial" patch: *** pp_hot.c@@/main/25 Sat May 27 22:42:22 1995 --- pp_hot.c Sun May 28 21:32:04 1995 *************** *** 360,365 **** --- 360,367 ---- if (op->op_flags & OPf_REF || op->op_private & HINT_STRICT_REFS) DIE(no_usym, "an ARRAY"); + if (GIMME == G_ARRAY) + RETURN; RETPUSHUNDEF; } sym = SvPV(sv,na); *************** *** 431,436 **** --- 433,442 ---- if (op->op_flags & OPf_REF || op->op_private & HINT_STRICT_REFS) DIE(no_usym, "a HASH"); + if (GIMME == G_ARRAY) { + SP--; + RETURN; + } RETSETUNDEF; } sym = SvPV(sv,na); Larry ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24244-95@lion>; Tue, 19 Sep 1995 07:32:04 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 04:36:37 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa29073; 19 Sep 95 5:35 +0100 Received: by gw.home.vix.com id AA06379; Mon, 18 Sep 95 10:44:48 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA06371; Mon, 18 Sep 95 10:44:41 -0700 Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id TAA23985; Mon, 18 Sep 1995 19:43:39 +0200 Date: Mon, 18 Sep 1995 19:43:39 +0200 Message-Id: <199509181743.TAA23985@anna.mind.de> From: Andreas Koenig To: aks@hub.ucsb.edu Cc: Tim Bunce , perldb-interest@vix.com In-Reply-To: <199509181715.KAA20127@dokoka> (message from Alan Stebbens on Mon, 18 Sep 1995 10:14:54 -0700) Subject: Re: Another bug.. Reply-To: andreas.koenig@mind.de >> I'm using MsqlPerl 1.02, perl 5.001m and msql 1.0.8. And you? Alan> Ok. Where is MsqlPerl 1.02? As you noticed before, I've Alan> got a much older version. Check the directory ANDK in one of these places ftp://ftp.funet.fi/pub/languages/perl/CPAN/mod/pml/byauthor/ ftp://ftp.pasteur.fr/pub/Perl/CPAN/pml/byauthor/ ftp://ftp.cs.ruu.nl/pub/PERL/perl5.0/CPAN/pml/byauthor/ ftp://janus.sedl.org/pub/mirrors/CPAN/mod/pml/byauthor/ ftp://franz.ww.TU-Berlin.DE/pub/pml/byauthor/ Alan> I do have perl5.001m, and msql-1.0.8. Alan> Is there a way I can determine where, exactly, the error is Alan> being generated? >> Good question, I don't know. Alan> Too bad. Thanks for the pointers. One very valuable option is always to cut the problem down to less than 10 lines and send me the script, so I can reproduce the problem. Alan> Alan andreas ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <26330-9@lion>; Tue, 19 Sep 1995 12:01:45 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 08:05:39 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa20841; 19 Sep 95 9:04 +0100 Received: by gw.home.vix.com id AA09059; Mon, 18 Sep 95 19:31:04 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA09051; Mon, 18 Sep 95 19:30:57 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa14646; 18 Sep 95 23:54 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa14967; 18 Sep 95 23:52 +0100 Received: from ig.co.uk by lion id <22407-0@lion>; Mon, 18 Sep 1995 23:53:30 +0100 To: lwall@scalpel.netlabs.com Subject: Re: Another hash ref question... Cc: perldb-interest@vix.com Date: Mon, 18 Sep 1995 23:53:30 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509182352.aa14967@post.demon.co.uk> > From: Tim Bunce > > > From: mhm@austin.ibm.com > > > > use Some_package qw( $hash_ref ); > > > > $attrib_dec = { %$hash_ref, 'Stype' => SQL_DECIMAL } ; > > > > fails. The question/problem is why is the hash_ref undefined when > > it has been defined in Some_package and exported with EXPORT_OK ? > > > Your code: > > package DBD::DB2; > require DBI; > require DynaLoader; > @ISA = qw(DynaLoader); > @EXPORT_OK = qw($attrib_int ); > > you should have said: > > @ISA = qw(Exporter DynaLoader); > Thinking about this some more... this is actually a rather nasty problem. The code use FOO qw(bar baz); will 'fail' silently if no import method is found. This is bad enough but probably quickly spotted if the imports are needed (as in this case). Consider the case where the user writes use FOO; without actually needing to import anything (this is sadly very common). If FOO is not an Exporter then the import will 'fail' silently. So far, not so bad. One day the module may be 'enhanced' and may now be an Exporter. The user of the module may now, unknowingly, have all sorts of symbols imported into the users namespace potentially causing obscure problems. I'm probably trying to close the stable door after the horse has bolted but I think it would be a good idea for the import an unimport methods to _not_ be treaded specially (gv_fetchmethod() for those interested). This would mean that modules which were not exporters could not be loaded with 'use' but would have to be loaded with 'require'. Taking this 'action at a distance through interface changes' to it's logical conclusion would also mean forcing people to explicitly list the imports they wanted or use :DEFAULT (but I know I've got no chance with that argument now :-) Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <26330-14@lion>; Tue, 19 Sep 1995 12:02:06 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 08:33:20 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa03437; 19 Sep 95 9:32 +0100 Received: by gw.home.vix.com id AA16674; Mon, 18 Sep 95 21:47:21 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA16668; Mon, 18 Sep 95 21:47:18 -0700 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id XAA29619; Mon, 18 Sep 1995 23:47:17 -0500 Date: Mon, 18 Sep 1995 23:47:16 -0500 (CDT) From: Mouring Cat Cc: perldb-interest@vix.com Subject: Oracle Driver note. In-Reply-To: <9509182352.aa14967@post.demon.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I'm finally able to start to implement the beginning of the QuickBase drivers for DBI and noticed a few differences between the headers and implementation in the Oracle Drivers. 1) ora_error _((..... int rc .....)); is how it's defined in the dbdimp.h and ora_error( ... sb2 rc ....)); is how it's defined in the .c file. Which is the correct version? Hmm...Not much else...From what I see..Since I don't have my Oracle database right now. =) ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04638-12@lion>; Wed, 20 Sep 1995 07:23:01 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 21:22:59 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa02980; 19 Sep 95 22:22 +0100 Received: by gw.home.vix.com id AA19168; Tue, 19 Sep 95 06:06:10 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA19163; Tue, 19 Sep 95 06:06:08 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab11884; 19 Sep 95 13:24 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa12249; 19 Sep 95 13:21 +0100 Received: from ig.co.uk by lion id <25702-0@lion>; Tue, 19 Sep 1995 10:40:31 +0100 To: perldb-interest@vix.com, Johan.Widingsjo@sto4.wmdata.se Subject: Re: Select from Oracle Date: Tue, 19 Sep 1995 10:40:31 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509191321.aa12249@post.demon.co.uk> > From: "Widingsjo, Johan" > > Hello, > > I'm very new to this and have only a little experience of perl prior to > this, (no oop in perl, only c++, smalltalk and simula). > It seems to me when reading the dbispec v0.6 that things have > changed around quite a bit. Yes. The spec is very old now. > So I'm wondering if there's any > examples or helpfiles. Or if someone could take a quick > look at my test code and see what I'm doing wrong. > Technically the DBI interface is *unsupported* the only supported interface at the moment is the Oraperl emulation interface (documented by the original oraperl manuals). > I'm connecting to an Oracle 7.1.6 database from > a solaris 2.4 system. > > The execute statement returns 1 and the fetch I don't know what it does > :-) The array comes out empty even tho I have a couple of records in the > table and the do function can insert more records. > You don't say what versions of DBI and DBD::Oracle you're using. Please check that you're using the latest (DBI 0.63, DBD::Oracle 0.25). > use DBI; > use DBD::Oracle; > You should not say 'use DBD::Oracle' here (but it doesn't relate to this problem). The DBI->connect loads DBD::Oracle for you. > # This line actually inserts a record in the database > # So it seems to be connecting ok > $status = $dbh->do("insert into test (ID, TEXT) values (2,'insert from > perl') "); > Perhaps you need to commit here. > $st=$dbh->prepare("select * from test"); > print "ST: $st\n"; > > $rc=$st->execute(); > print "Execute returns: $rc\n"; > > @datat = $st->fetch(); > print "#: $#datat\n"; > print "Data: @datat\n"; Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04638-22@lion>; Wed, 20 Sep 1995 07:25:13 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Tue, 19 Sep 95 23:41:28 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa08635; 20 Sep 95 0:40 +0100 Received: by gw.home.vix.com id AA06746; Tue, 19 Sep 95 10:33:12 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA06739; Tue, 19 Sep 95 10:33:11 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id MAA119284 for ; Tue, 19 Sep 1995 12:33:10 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA17310; Tue, 19 Sep 1995 12:33:09 -0500 From: mhm@austin.ibm.com Message-Id: <9509191733.AA17310@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: disconnect is no working.... Date: Tue, 19 Sep 95 12:33:08 -0500 What is the DBI error message after DIB::END complete attempting to tell me? Thanks Mike mhm@austin.ibm.com Log of debug info: DBI->connect(inv, , , DB2, ) DBI->install_driver DBI DB2 at genreq.pl line 11 DBI->install_driver(DB2) loaded install_driver(DB2): setup @ISA for DBD::DB2::dr at genreq.pl line 11 install_driver(DB2): setup @ISA for DBD::DB2::db at genreq.pl line 11 install_driver(DB2): setup @ISA for DBD::DB2::st at genreq.pl line 11 DB2_HOME set to /usr/lpp/db2_02_01 New DBI::dr (for DBD::DB2::dr, parent=, id=) New DBI::dr => DBI::dr=HASH(0x2006860c) (inner=DBI::dr=HASH(0x200c5e18)) for DBD::DB2::dr DBI->install_driver(DB2) = DBI::dr=HASH(0x2006860c) DBI->connect using DB2 driver DBI::dr=HASH(0x2006860c) New DBI::db (for DBD::DB2::db, parent=DBI::dr=HASH(0x200c5e18), id=) New DBI::db => DBI::db=HASH(0x200c5d94) (inner=DBI::db=HASH(0x200c5d1c)) for DBD::DB2::db DBI->connect = DBI::db=HASH(0x200c5d94) New DBI::st (for DBD::DB2::st, parent=DBI::db=HASH(0x200c5d1c), id=) New DBI::st => DBI::st=HASH(0x200c5dd0) (inner=DBI::st=HASH(0x200c5cec)) for DBD::DB2::st 766336 Entering parsing phase disconnect(DBI::db=HASH(0x200c5d1c)) invalidates 1 associated cursor(s) at genreq.pl line 69. DBI::END DBI::disconnect_all DBI DBI::disconnect_all for 'DB2' DBI::END complete SQLSTATE = ERROR EVENT: DBI::st=HASH(0x200c5cec)->DBI::default_handler: -2 at genreq.pl line 0 ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04638-29@lion>; Wed, 20 Sep 1995 07:26:20 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 20 Sep 95 00:43:20 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa05847; 20 Sep 95 1:43 +0100 Received: by gw.home.vix.com id AA22069; Tue, 19 Sep 95 06:58:40 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA22064; Tue, 19 Sep 95 06:58:34 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id aa12168; 19 Sep 95 13:27 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa12929; 19 Sep 95 13:25 +0100 Received: from ig.co.uk by lion id <29295-0@lion>; Tue, 19 Sep 1995 13:27:32 +0100 To: mouring@netnet.net Subject: Re: Oracle Driver note. Cc: perldb-interest@vix.com Date: Tue, 19 Sep 1995 13:27:32 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509191325.aa12929@post.demon.co.uk> > From: Mouring Cat > > I'm finally able to start to implement the beginning of the QuickBase > drivers for DBI and noticed a few differences between the headers and > implementation in the Oracle Drivers. > > 1) > ora_error _((..... int rc .....)); is how it's defined in the dbdimp.h > and > ora_error( ... sb2 rc ....)); is how it's defined in the .c file. > Which is the correct version? > The second, but it doesn't really matter for you since ora_error is an internal DBD::Oracle specific function which you sould replace with something more suited to your database engine. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <04638-71@lion>; Wed, 20 Sep 1995 07:31:57 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 20 Sep 95 03:10:59 GMT Received: from gw.home.vix.com by punt.demon.co.uk id aa27542; 20 Sep 95 4:10 +0100 Received: by gw.home.vix.com id AA00542; Tue, 19 Sep 95 15:39:07 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA00536; Tue, 19 Sep 95 15:39:05 -0700 Received: from alma.mcqueen.com (alma.mcqueen.com [193.133.160.249]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with ESMTP id XAA08336; Tue, 19 Sep 1995 23:34:58 +0100 From: descarte@mcqueen.com Received: (from descarte@localhost) by alma.mcqueen.com (8.6.12/8.6.12) id XAA01515; Tue, 19 Sep 1995 23:41:29 +0100 Message-Id: <199509192241.XAA01515@alma.mcqueen.com> Subject: DBD::mSQL pl3 available To: perldb-interest@vix.com, msql-list@bond.edu.au Date: Tue, 19 Sep 1995 23:41:28 +0100 (BST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 664 A couple of minor bug-fixes in this patchlevel including support for not specifying the database hostname ( it defaults to localhost ) and handling of NULLs ( fairly importanto! ) Also, a mirror of ftp.demon.co.uk has been taken on a nightly basis to hopefully circumvent the problems people seem to have in connecting to that machine to download any perl db stuff. The location for any perldb stuff ( especially new DBD::mMSQL releases ) is: ftp://ftp.mcqueen.com/pub/dbperl NOTE!: This address has changed today, so the old one wont work! A. -- Alligator Descartes | PGP Key available on request. descarte@mcqueen.com | http://www.mcqueen.com/hermetica ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <14095-3@lion>; Thu, 21 Sep 1995 07:21:00 +0100 Received: from punt2.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Wed, 20 Sep 95 18:40:51 GMT Received: from [192.5.5.1] by punt2.demon.co.uk id aa12349; 20 Sep 95 19:38 +0100 Received: by gw.home.vix.com id AA28386; Tue, 19 Sep 95 23:20:59 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA28381; Tue, 19 Sep 95 23:20:56 -0700 Received: by disperse.demon.co.uk id ac17525; 20 Sep 95 0:08 +0100 Received: from post.demon.co.uk by disperse.demon.co.uk id ab07837; 19 Sep 95 17:27 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id aa07394; 19 Sep 95 17:24 +0100 Received: from ig.co.uk by lion id <01538-0@lion>; Tue, 19 Sep 1995 17:26:57 +0100 To: descarte@mcqueen.com Subject: Re: $dbh->do behaviour Cc: perldb-interest@vix.com Date: Tue, 19 Sep 1995 17:26:57 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509191724.aa07394@post.demon.co.uk> > From: descarte@mcqueen.com > > Hi Tim. > > Spotted another quirk with $dbh->do in DBD::Oracle ( haven't tested against > DBD::mSQL yet ). > > I get: > > Statement handle DBI::st=HASH(0xa145c) destroyed without finish() at ... > > After every $dbh->do call. Of course, $dbh->finish won't work..... > > Any ideas? Or is still just a hangover from the DBI glitch I asked about > before? > > I'm doing an 'undef $csr' after each call, incidentally. That cured the > problem on cursors with the prepare, execute, finish cycle, but hasn't made > any difference over do. > On the one hand I could change do to call finish. On the other hand I think the warning is probably over-picky. I don't want to have to call finish before destroying a cursor. The next release of DBD::Oracle will not have the warning. Feel free to delete it from your copy now. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <14096-39@lion>; Thu, 21 Sep 1995 07:27:16 +0100 Received: from punt.demon.co.uk via puntmail for dbperl@ignite.demon.co.uk; Thu, 21 Sep 95 06:19:29 GMT Received: from frigate.doc.ic.ac.uk by punt.demon.co.uk id ad08028; 21 Sep 95 7:19 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <21838-0@frigate.doc.ic.ac.uk>; Thu, 21 Sep 1995 03:41:55 +0100 Received: by gw.home.vix.com id AA24863; Wed, 20 Sep 95 13:21:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA24858; Wed, 20 Sep 95 13:21:45 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id PAA23663 for ; Wed, 20 Sep 1995 15:21:43 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA18372; Wed, 20 Sep 1995 15:21:42 -0500 From: mhm@austin.ibm.com Message-Id: <9509202021.AA18372@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Lost in the flow ... Date: Wed, 20 Sep 95 15:21:42 -0500 Starting to work on fetching data from DB2 and I don't understand the flow through DBI. Normally, things would look something like this: $sth = $dbh->prepare("select f1, f2, f3 from f_table"); $sth->execute(); fine except that in dbdimp.c, dbd_st_repare calls dbd_preparse which insists that the statement contain parameters. So, how does one call a nonparameterized select and later add the parameters to fetch to the same $sth? And to finish the example: $sth->bind_param(1,$f1,$attrib); $sth->bind_param(2,$f2,$attrib); $sth->bind_param(3,$f3,$attrib); $sth->fetch(); Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <25246-12@lion>; Fri, 22 Sep 1995 02:22:12 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 22 Sep 95 01:16:04 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id al16509; 22 Sep 95 2:14 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <07445-0@frigate.doc.ic.ac.uk>; Thu, 21 Sep 1995 21:17:15 +0100 Received: by gw.home.vix.com id AA26514; Thu, 21 Sep 95 04:57:54 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA26477; Thu, 21 Sep 95 04:57:50 -0700 Received: (from skimo@localhost) by dns.ufsia.ac.be (8.6.12/8.6.9) id OAA24209 for perldb-interest@vix.com; Thu, 21 Sep 1995 14:01:34 +0200 From: Sven Verdoolaege Message-Id: <199509211201.OAA24209@dns.ufsia.ac.be> Subject: DBI on NT, possible ? To: perldb-interest@vix.com Date: Thu, 21 Sep 1995 14:01:34 +0200 (MET DST) X-Stardate: [-31]6322.50 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 134 Hi, I know there's an ntperl, but is it possible to run DBI and DBD::Oracle on it ? Has anyone had any experience with this ? Sven ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <26845-15@lion>; Fri, 22 Sep 1995 07:24:00 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 22 Sep 95 05:48:54 GMT Received: from gw.home.vix.com by punt-1.mail.demon.net id aa02135; 22 Sep 95 6:47 +0100 Received: by gw.home.vix.com id AA29545; Thu, 21 Sep 95 20:37:11 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29540; Thu, 21 Sep 95 20:37:07 -0700 Received: from post.demon.co.uk by disperse.mail.demon.net id ac22085; 22 Sep 95 2:24 +0100 Received: from ignite.demon.co.uk by post.mail.demon.net id aa05807; 22 Sep 95 2:21 +0100 Received: from ig.co.uk by lion id <25733-0@lion>; Fri, 22 Sep 1995 02:23:33 +0100 To: perldb-interest@vix.com, skimo@dns.ufsia.ac.be Subject: Re: DBI on NT, possible ? Date: Fri, 22 Sep 1995 02:23:33 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509220221.aa05807@post.mail.demon.net> > From: Sven Verdoolaege > > Hi, > > I know there's an ntperl, > but is it possible to run DBI and DBD::Oracle on it ? > Probably. > Has anyone had any experience with this ? > Not that I'm aware of. Seems like you've just volunteered :-) > Sven > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <28562-1@lion>; Fri, 22 Sep 1995 11:32:40 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 22 Sep 95 07:04:52 GMT Received: from gw.home.vix.com by punt-1.mail.demon.net id aa01759; 22 Sep 95 8:04 +0100 Received: by gw.home.vix.com id AA24931; Thu, 21 Sep 95 19:17:35 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA24926; Thu, 21 Sep 95 19:17:31 -0700 Received: from post.demon.co.uk by disperse.mail.demon.net id aa21920; 22 Sep 95 2:21 +0100 Received: from ignite.demon.co.uk by post.mail.demon.net id ab05459; 22 Sep 95 2:18 +0100 Received: from ig.co.uk by lion id <25103-0@lion>; Fri, 22 Sep 1995 01:51:20 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: disconnect is no working.... Date: Fri, 22 Sep 1995 01:51:20 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509220218.ab05459@post.mail.demon.net> > From: mhm@austin.ibm.com > > What is the DBI error message after DIB::END complete attempting to tell me? > Entering parsing phase > disconnect(DBI::db=HASH(0x200c5d1c)) invalidates 1 associated cursor(s) > at genreq.pl line 69. > DBI::END > DBI::disconnect_all DBI > DBI::disconnect_all for 'DB2' > DBI::END complete > SQLSTATE = > ERROR EVENT: DBI::st=HASH(0x200c5cec)->DBI::default_handler: > -2 > at genreq.pl line 0 > I don't know. It might be the invalidated cursor (note the ::st) being DESTROY'd during global destruction and the DESTROY code having problems because it's parent has been disconnect'd (and possibly DESTROY'd). Try running it with perl -Do and/or PERL_DBI_DEBUG=2 in the environment. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <28562-9@lion>; Fri, 22 Sep 1995 11:33:34 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 22 Sep 95 10:28:08 GMT Received: from gw.home.vix.com by punt-1.mail.demon.net id aa03605; 22 Sep 95 11:27 +0100 Received: by gw.home.vix.com id AA29412; Thu, 21 Sep 95 20:31:35 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29399; Thu, 21 Sep 95 20:31:30 -0700 Received: from post.demon.co.uk by disperse.demon.co.uk id ab20085; 21 Sep 95 12:22 +0100 Received: from ignite.demon.co.uk by post.demon.co.uk id ad24797; 21 Sep 95 12:20 +0100 Received: from ig.co.uk by lion id <16687-0@lion>; Thu, 21 Sep 1995 11:14:38 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Lost in the flow ... Date: Thu, 21 Sep 1995 11:14:38 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509211220.ad24797@post.demon.co.uk> > From: mhm@austin.ibm.com > > Starting to work on fetching data from DB2 and I don't understand the flow > through DBI. > > Normally, things would look something like this: > > $sth = $dbh->prepare("select f1, f2, f3 from f_table"); > > $sth->execute(); > > fine except that in dbdimp.c, dbd_st_repare calls dbd_preparse which insists > that the statement contain parameters. > Why? What makes you think it "insists that the statement contain parameters"? It works for DBD::Oracle. > So, how does one call a nonparameterized select and later add the parameters > to fetch to the same $sth? > I don't understand that sentance. Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <12132-3@lion>; Mon, 25 Sep 1995 07:21:12 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 22 Sep 95 22:58:09 GMT Received: from gw.home.vix.com by punt-2.mail.demon.net id aa03763; 22 Sep 95 23:57 +0100 Received: by gw.home.vix.com id AA11104; Fri, 22 Sep 95 07:25:20 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA11099; Fri, 22 Sep 95 07:25:18 -0700 Received: from alma.mcqueen.com (alma.mcqueen.com [193.133.160.249]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with ESMTP id PAA03103 for ; Fri, 22 Sep 1995 15:21:26 +0100 From: descarte@mcqueen.com Received: (from descarte@localhost) by alma.mcqueen.com (8.6.12/8.6.12) id PAA23164 for perldb-interest@vix.com; Fri, 22 Sep 1995 15:28:02 +0100 Message-Id: <199509221428.PAA23164@alma.mcqueen.com> Subject: DBperl WWW pages To: perldb-interest@vix.com Date: Fri, 22 Sep 1995 15:28:01 +0100 (BST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 252 For information on DBperl, DBI and DBD::Oracle, DBD::mSQL, please check out http://www.hermetica.com Suggestions on content are most welcome! -- Alligator Descartes | PGP Key available on request. descarte@mcqueen.com | http://www.hermetica.com/ ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12133-4@lion>; Mon, 25 Sep 1995 07:21:22 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 22 Sep 95 20:02:35 GMT Received: from gw.home.vix.com by punt-1.mail.demon.net id aa02510; 22 Sep 95 21:01 +0100 Received: by gw.home.vix.com id AA02439; Fri, 22 Sep 95 05:09:32 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA02434; Fri, 22 Sep 95 05:09:30 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA59010 for ; Fri, 22 Sep 1995 07:09:29 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA19790; Fri, 22 Sep 1995 07:08:05 -0500 From: mhm@austin.ibm.com Message-Id: <9509221208.AA19790@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Re: Lost in the flow ... In-Reply-To: Your message of Thu, 21 Sep 95 11:14:38 +0100. <9509211220.ad24797@post.demon.co.uk> Date: Fri, 22 Sep 95 07:08:05 -0500 Tim Bunce writes: |> > From: mhm@austin.ibm.com |> > |> > Starting to work on fetching data from DB2 and I don't understand the flow |> > through DBI. |> > |> > Normally, things would look something like this: |> > |> > $sth = $dbh->prepare("select f1, f2, f3 from f_table"); |> > |> > $sth->execute(); |> > |> > fine except that in dbdimp.c, dbd_st_repare calls dbd_preparse which insists |> > that the statement contain parameters. |> > |> Why? What makes you think it "insists that the statement contain parameters"? |> It works for DBD::Oracle. Well, I think most of it was related to a) being frustrated and b) not having the total flow figured out. Any rate, some how I managed to get dbd_bind_ph called on a "select * from tab" with "Parameter pX unknown". I have since, determined the correct sequences of thing and am proud to say DB2perl is working ... database control, inserts, updates, and deletions. All that is left is dealing with BLOBs, CLOBs, DLOBs, adding parameter strings to fetch/fetchrow, adding a global SQLSTATE variable and dealing with the closing database errors. |> > So, how does one call a nonparameterized select and later add the |> > parameters to fetch to the same $sth? |> > |> I don't understand that sentance. It took me while to figure out how to do something like $sth = $dbh->prepare("select f1,f2,f3,f4 from tab;"); $sth->execute(); $sth->fetchrow("using $sth into :field1, :field2, :field3, :junk;"); Which took me about three days to realize that all I really needed to do was ($field1,$field2, $field3, $junk) = $sth->fetchrow(); Sometimes I can be so dense :-( Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12133-6@lion>; Mon, 25 Sep 1995 07:21:32 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 22 Sep 95 21:46:02 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa19603; 22 Sep 95 22:45 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <25472-0@frigate.doc.ic.ac.uk>; Fri, 22 Sep 1995 22:28:23 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA01996 for perldb-interest-real; Fri, 22 Sep 1995 12:24:35 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA01992 for ; Fri, 22 Sep 1995 12:24:34 -0700 Message-Id: <199509221924.MAA01992@toccata.fugue.com> To: perldb-interest@toccata.fugue.com Subject: Mailing list move... Date: Fri, 22 Sep 1995 12:24:34 -0700 From: Ted Lemon I've moved the mailing list from vix.com to fugue.com so that it'll be a little easier for me to maintain. I've also added a bouncer to the perldb-interest-request mailing list so that it confirms receipt of all messages, which will hopefully reduce the number of subscribe messages that go to the main list because somebody didn't get an acknowledgement and got impatient (this is an amazingly common occurrance). I'm thinking of adding a filter to the main list that notices administrivia messages and sends them only to me, but I don't know when I'll have time to do that. Meanwhile, if you're telling your friends about the mailing list, please tell them the new addresses: Administrivia: perldb-interest-request@fugue.com Content: perldb-interest@fugue.com Thanks! _MelloN_ ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12133-10@lion>; Mon, 25 Sep 1995 07:22:11 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 23 Sep 95 00:33:50 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ab07285; 23 Sep 95 1:33 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <27446-0@frigate.doc.ic.ac.uk>; Sat, 23 Sep 1995 00:30:27 +0100 Received: by gw.home.vix.com id AA28569; Fri, 22 Sep 95 11:08:21 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA28564; Fri, 22 Sep 95 11:08:19 -0700 Received: from dino.sikorsky.com by rotor.sikorsky.com (4.1/SMI-4.1) id AA20058; Fri, 22 Sep 95 14:04:23 EDT Received: from tss0.bpt.sikorsky.com by dino.sikorsky.com (4.1/SMI-4.1) id AA13918; Fri, 22 Sep 95 14:08:41 EDT Received: by tss0.bpt.sikorsky.com (5.65/DEC-Ultrix/4.3) id AA16461; Fri, 22 Sep 1995 14:07:55 -0400 Received: by sun38.sikorsky (5.x/SMI-SVR4) id AA07048; Fri, 22 Sep 1995 14:08:06 -0400 Date: Fri, 22 Sep 1995 14:08:06 -0400 From: Dave Rattanni Message-Id: <9509221808.AA07048@sun38.sikorsky> To: perldb-interest@vix.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: pxGAlCdpYRIs+BIsZuqvHw== subscribe perldb-interest ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12133-12@lion>; Mon, 25 Sep 1995 07:22:45 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 23 Sep 95 00:33:52 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ad07285; 23 Sep 95 1:33 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <27482-0@frigate.doc.ic.ac.uk>; Sat, 23 Sep 1995 00:32:37 +0100 Received: by gw.home.vix.com id AA11843; Fri, 22 Sep 95 07:38:57 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA11838; Fri, 22 Sep 95 07:38:55 -0700 Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id JAA45430 for ; Fri, 22 Sep 1995 09:38:54 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA03256; Fri, 22 Sep 1995 09:38:53 -0500 From: mhm@austin.ibm.com Message-Id: <9509221438.AA03256@dax.austin.ibm.com> To: perldb-interest@vix.com Subject: Re: disconnect is not working.... In-Reply-To: Your message of Fri, 22 Sep 95 01:51:20 +0100. <9509220218.ab05459@post.mail.demon.net> Date: Fri, 22 Sep 95 09:38:52 -0500 Tim Bunce writes: |> > From: mhm@austin.ibm.com |> > Entering parsing phase |> > disconnect(DBI::db=HASH(0x200c5d1c)) invalidates 1 associated cursor(s) |> > at genreq.pl line 69. |> > DBI::END |> > DBI::disconnect_all DBI |> > DBI::disconnect_all for 'DB2' |> > DBI::END complete |> > SQLSTATE = |> > ERROR EVENT: DBI::st=HASH(0x200c5cec)->DBI::default_handler: |> > -2 |> > at genreq.pl line 0 |> > |> I don't know. It might be the invalidated cursor (note the ::st) being |> DESTROY'd during global destruction and the DESTROY code having problems |> because it's parent has been disconnect'd (and possibly DESTROY'd). |> You're right, user error on my part. I was explicitly calling $dbh->disconnect() which was being implicitly called by the END block. Gosh, I really have a case of the stupids this week..... Thanks for the pointer. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12133-14@lion>; Mon, 25 Sep 1995 07:23:31 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 23 Sep 95 00:33:54 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ac07285; 23 Sep 95 1:33 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <27455-0@frigate.doc.ic.ac.uk>; Sat, 23 Sep 1995 00:30:39 +0100 Received: by gw.home.vix.com id AA13238; Fri, 22 Sep 95 08:00:03 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA13214; Fri, 22 Sep 95 08:00:00 -0700 Received: from dino.sikorsky.com by rotor.sikorsky.com (4.1/SMI-4.1) id AA19001; Fri, 22 Sep 95 10:56:04 EDT Received: from tss0.bpt.sikorsky.com by dino.sikorsky.com (4.1/SMI-4.1) id AA13190; Fri, 22 Sep 95 11:00:22 EDT Received: by tss0.bpt.sikorsky.com (5.65/DEC-Ultrix/4.3) id AA10787; Fri, 22 Sep 1995 10:59:37 -0400 Received: by sun38.sikorsky (5.x/SMI-SVR4) id AA06936; Fri, 22 Sep 1995 10:59:46 -0400 Date: Fri, 22 Sep 1995 10:59:46 -0400 From: Dave Rattanni Message-Id: <9509221459.AA06936@sun38.sikorsky> To: perldb-interest@vix.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: tfE1lYbBkkMInrepSHAAGA== subscribe ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12133-16@lion>; Mon, 25 Sep 1995 07:24:00 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 23 Sep 95 01:31:42 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa00239; 23 Sep 95 2:31 +0100 Received: from gw.home.vix.com by frigate.doc.ic.ac.uk with SMTP (PP) id <28824-0@frigate.doc.ic.ac.uk>; Sat, 23 Sep 1995 02:11:45 +0100 Received: by gw.home.vix.com id AA29908; Fri, 22 Sep 95 11:29:39 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Precedence: junk Sender: perldb-interest-request@vix.com Received: by gw.home.vix.com id AA29898; Fri, 22 Sep 95 11:29:36 -0700 Received: from dino.sikorsky.com by rotor.sikorsky.com (4.1/SMI-4.1) id AA20144; Fri, 22 Sep 95 14:25:32 EDT Received: from tss0.bpt.sikorsky.com by dino.sikorsky.com (4.1/SMI-4.1) id AA14000; Fri, 22 Sep 95 14:29:50 EDT Received: by tss0.bpt.sikorsky.com (5.65/DEC-Ultrix/4.3) id AA16576; Fri, 22 Sep 1995 14:29:04 -0400 Received: by sun38.sikorsky (5.x/SMI-SVR4) id AA07059; Fri, 22 Sep 1995 14:29:15 -0400 Date: Fri, 22 Sep 1995 14:29:15 -0400 From: Dave Rattanni Message-Id: <9509221829.AA07059@sun38.sikorsky> To: perldb-interest@vix.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: HRdq5g0y2I+ZoHHXHWPTyQ== subscribe perldb-interest-REQUEST@vix.com ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <12133-29@lion>; Mon, 25 Sep 1995 07:28:01 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 23 Sep 95 08:23:54 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa12480; 23 Sep 95 9:23 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <02713-0@frigate.doc.ic.ac.uk>; Sat, 23 Sep 1995 09:22:54 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA03315 for perldb-interest-real; Sat, 23 Sep 1995 00:03:24 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA03311 for ; Sat, 23 Sep 1995 00:03:22 -0700 Received: by gw.home.vix.com id AA17334; Sat, 23 Sep 95 00:05:28 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id JAA28109; Sat, 23 Sep 1995 09:02:57 +0200 Date: Sat, 23 Sep 1995 09:02:57 +0200 Message-Id: <199509230702.JAA28109@anna.mind.de> From: Andreas Koenig To: descarte@mcqueen.com Cc: perldb-interest@vix.com In-Reply-To: <199509221428.PAA23164@alma.mcqueen.com> (descarte@mcqueen.com) Subject: Re: DBperl WWW pages Reply-To: andreas.koenig@mind.de > For information on DBperl, DBI and DBD::Oracle, DBD::mSQL, > please check out > http://www.hermetica.com > Suggestions on content are most welcome! Very nice page! Suggestions? I'd recommend it as it is. > -- Alligator Descartes Regards, andreas ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <15055-5@lion>; Thu, 28 Sep 1995 15:04:41 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 28 Sep 95 11:41:28 GMT Received: from disperse.demon.co.uk by punt-1.mail.demon.net id aa29969; 28 Sep 95 12:40 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id cx23394; 28 Sep 95 6:28 +0100 Received: from relay-1.mail.demon.net by punt-2.mail.demon.net id aa08257; 28 Sep 95 2:58 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id ab04324; 27 Sep 95 6:09 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA01386 for perldb-interest-real; Tue, 26 Sep 1995 20:38:55 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA01382 for ; Tue, 26 Sep 1995 20:38:53 -0700 Received: by gw.home.vix.com id AA21814; Tue, 26 Sep 95 20:40:45 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-2.mail.demon.net id ah03704; 26 Sep 95 19:38 GMT Received: from ignite.demon.co.uk by post.mail.demon.net id ab08704; 25 Sep 95 13:55 +0100 Received: from ig.co.uk by lion id <16196-0@lion>; Mon, 25 Sep 1995 12:44:07 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: Lost in the flow ... Date: Mon, 25 Sep 1995 12:44:07 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509251355.ab08704@post.mail.demon.net> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > From: mhm@austin.ibm.com > > I have since, determined the correct sequences of thing and am proud to > say DB2perl is working ... database control, inserts, updates, and > deletions. Great. > |> > So, how does one call a nonparameterized select and later add the > |> > parameters to fetch to the same $sth? > |> > > |> I don't understand that sentance. > > It took me while to figure out how to do something like > > $sth = $dbh->prepare("select f1,f2,f3,f4 from tab;"); > $sth->execute(); > $sth->fetchrow("using $sth into :field1, :field2, :field3, :junk;"); > > Which took me about three days to realize that all I really needed to > do was > > ($field1,$field2, $field3, $junk) = $sth->fetchrow(); > > Sometimes I can be so dense :-( > Sometimes the DBI can be so undocumented ... :-( If you are using av = DBIS->get_fbav (which you should be) then you can also do this: $sth->bind_columns(undef, \$field1, \$field2, \$field3, \$junk); while($sth->fetch) { ... } which will be faster. You must also be using sv_set*() for this to work. Since bind_col and bind_columns will be part of the DBI spec you'll need to support them. > Mike > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <16160-0@lion>; Thu, 28 Sep 1995 16:13:17 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 28 Sep 95 15:10:40 GMT Received: from relay-1.mail.demon.net by punt-2.mail.demon.net id aa23917; 28 Sep 95 16:06 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa29223; 28 Sep 95 15:49 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA07123 for perldb-interest-real; Thu, 28 Sep 1995 04:57:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA07119 for ; Thu, 28 Sep 1995 04:57:24 -0700 Received: by gw.home.vix.com id AA26724; Thu, 28 Sep 95 04:59:31 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by hatch.sonalysts.com (1.37.109.16/16.2) id AA020729539; Thu, 28 Sep 1995 07:58:59 -0400 Date: Thu, 28 Sep 1995 07:58:58 -0400 (EDT) From: Matt Stahl To: perldb-interest@vix.com Subject: DBI/DBD and perlNT Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I'm a perl newbie, and would like to be able to use DBI to access Oracle databases under winNT3.51. Unfortunately, the latest (?) version of NTperl (5.001l) doesn't support dynamic loading. Anyone know when the next release of NTperl is due (and whether or not it will support dynamic loading)? Any other ideas? --Matt ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <17620-1@lion>; Thu, 28 Sep 1995 18:27:45 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 28 Sep 95 17:23:32 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa06716; 28 Sep 95 18:21 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA07292 for perldb-interest-real; Thu, 28 Sep 1995 06:34:32 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA07288 for ; Thu, 28 Sep 1995 06:34:30 -0700 Received: by gw.home.vix.com id AA02989; Thu, 28 Sep 95 06:36:39 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from mogate (mogate.sps.mot.com) by spsgate.sps.mot.com (4.1/SMI-4.1/Email 2.1 10/25/93) id AA29912 for perldb-interest@vix.com; Thu, 28 Sep 95 06:36:34 MST Received: from email56 by mogate (4.1/SMI-4.1/Email-2.0) id AA08604; Thu, 28 Sep 95 06:33:12 MST Received: (from r28874@localhost) by ismuc.sps.mot.com (8.6.10/8.6.6) id OAA13992; Thu, 28 Sep 1995 14:30:36 +0200 From: Christian Murphy Message-Id: <199509281230.OAA13992@ismuc.sps.mot.com> Subject: Re: DBI/DBD and perlNT To: Matt Stahl Date: Thu, 28 Sep 1995 14:30:35 +0200 (MES) Cc: perldb-interest@vix.com In-Reply-To: from "Matt Stahl" at Sep 28, 95 07:58:58 am X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 969 Hi Matt, > I'm a perl newbie, and would like to be able to use DBI to access Oracle > databases under winNT3.51. Unfortunately, the latest (?) version of > NTperl (5.001l) doesn't support dynamic loading. I have used the DBI on UNIX SVR3, which does not support dynamic loading either (aside: there is a shared library kludge that you wouldn't want to know about). It just means that you have to link it all statically. That makes your executable larger. It's been a while since I did it, but I just used "make LINKTYPE=static". You may want to look at the Makefile and Makefile.PL to see if you can do something like that. > Anyone know when the next release of NTperl is due (and whether or not it > will support dynamic loading)? Sorry, no ideas on this ... I've never seen NT in the flesh. Hope this helps, -christian -- Christian Murphy [ I've done this with cc because there seem to be long delays on the mailing list.] ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <17755-0@lion>; Thu, 28 Sep 1995 18:37:45 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 28 Sep 95 17:36:35 GMT Received: from relay-1.mail.demon.net by punt-2.mail.demon.net id aa14537; 28 Sep 95 18:35 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa07021; 28 Sep 95 18:33 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA07343 for perldb-interest-real; Thu, 28 Sep 1995 07:02:59 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA07339 for ; Thu, 28 Sep 1995 07:02:54 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ae23888; 28 Sep 95 15:04 +0100 Received: from ig.co.uk by lion id <14204-0@lion>; Thu, 28 Sep 1995 12:02:48 +0100 To: perldb-interest@fugue.com Subject: Ping Date: Thu, 28 Sep 1995 12:02:48 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9509281504.ae23888@relay-3.mail.demon.net> Just a test to see if the new address is working okay and to let you know that I'm still alive - but rather snowed under with real work. Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <17834-0@lion>; Thu, 28 Sep 1995 18:41:23 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 28 Sep 95 17:39:29 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa01528; 28 Sep 95 18:38 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa07217; 28 Sep 95 18:36 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA07433 for perldb-interest-real; Thu, 28 Sep 1995 07:33:48 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA07429 for ; Thu, 28 Sep 1995 07:33:46 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id PAA08805 for perldb-interest@fugue.com; Thu, 28 Sep 1995 15:32:02 +0100 From: Alligator Descartes Message-Id: <199509281432.PAA08805@fruitbat.mcqueen.com> Subject: DBI WWW pages available To: perldb-interest@fugue.com Date: Thu, 28 Sep 1995 15:32:02 +0100 (BST) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 473 Just a quick reminder. WWW pages on DBI are available at: http://www.hermetica.com Please take a look if you haven't already. Thanks to the load of people who have stopped by and made suitably encouraging comments! I'm not here until the 9th October, so don't expect any more stuff until then! All suggestions and other sundry conversation welcomed. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.mcqueen.com/hermetica ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <20490-1@lion>; Fri, 29 Sep 1995 07:09:59 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 28 Sep 95 17:51:42 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa22351; 28 Sep 95 18:49 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA07561 for perldb-interest-real; Thu, 28 Sep 1995 08:21:28 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA07557 for ; Thu, 28 Sep 1995 08:21:26 -0700 Received: by gw.home.vix.com id AA09396; Thu, 28 Sep 95 08:23:28 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ab23854; 28 Sep 95 16:19 +0100 Received: from ig.co.uk by lion id <16257-0@lion>; Thu, 28 Sep 1995 16:18:42 +0100 To: perldb-interest@vix.com, mgstahl@sonalysts.com Subject: Re: DBI/DBD and perlNT Date: Thu, 28 Sep 1995 16:18:42 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509281619.ab23854@relay-3.mail.demon.net> > From: Matt Stahl > > I'm a perl newbie, and would like to be able to use DBI to access Oracle > databases under winNT3.51. Unfortunately, the latest (?) version of > NTperl (5.001l) doesn't support dynamic loading. > > Anyone know when the next release of NTperl is due (and whether or not it > will support dynamic loading)? > > Any other ideas? > Er, if dynamic loading doesn't work then try static loading. Beyond that I can't help with NT issues. Maybe others can. > --Matt > Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <20490-13@lion>; Fri, 29 Sep 1995 07:13:13 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 28 Sep 95 19:40:26 GMT Received: from disperse.demon.co.uk by punt-2.mail.demon.net id aa17861; 28 Sep 95 20:38 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id ae14729; 27 Sep 95 20:31 +0100 Received: from relay-1.mail.demon.net by punt-2.mail.demon.net id aa12324; 27 Sep 95 20:26 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa20152; 25 Sep 95 22:41 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA10944 for perldb-interest-real; Mon, 25 Sep 1995 11:39:12 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from mail-relay1.cis.yale.edu (mail-relay1.cis.yale.edu [130.132.21.199]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA10940 for ; Mon, 25 Sep 1995 11:39:09 -0700 Received: from capitoline.cis.yale.edu by mail-relay1.cis.yale.edu with SMTP id AA17129 (5.67a/IDA-1.5 for ); Mon, 25 Sep 1995 14:35:25 -0400 Received: from morpheus.cis.yale.edu (morpheus [130.132.143.248]) by capitoline.cis.yale.edu (8.6.12/8.6.12) with ESMTP id OAA29850 for ; Mon, 25 Sep 1995 14:41:13 -0400 Received: (from rivory@localhost) by morpheus.cis.yale.edu (8.6.12/8.6.12) id OAA09465; Mon, 25 Sep 1995 14:41:10 -0400 Date: Mon, 25 Sep 1995 14:41:10 -0400 (EDT) From: Royal Hansen X-Sender: rivory@morpheus To: interest perl Subject: mSQLperl Message-Id: ; Fri, 29 Sep 1995 12:27:53 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 29 Sep 95 09:40:31 GMT Received: from disperse.demon.co.uk by punt-2.mail.demon.net id ab26420; 29 Sep 95 10:38 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id ak11976; 29 Sep 95 10:38 +0100 Received: from disperse.demon.co.uk by punt-2.mail.demon.net id ab24990; 28 Sep 95 11:46 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id au22915; 28 Sep 95 4:45 +0100 Received: from disperse.demon.co.uk by punt-2.mail.demon.net id ab13654; 27 Sep 95 21:24 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id ad16186; 27 Sep 95 20:56 +0100 Received: from toccata.fugue.com by punt-2.mail.demon.net id aa19562; 27 Sep 95 20:40 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA03812 for perldb-interest-real; Wed, 27 Sep 1995 10:49:11 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA03807; Wed, 27 Sep 1995 10:49:10 -0700 Message-Id: <199509271749.KAA03807@toccata.fugue.com> To: mhm@austin.ibm.com cc: perldb-interest@fugue.com Subject: Re: was the move successful ? In-reply-to: Your message of "Tue, 26 Sep 1995 13:33:06 CDT." <9509261833.AA19360@dax.austin.ibm.com> Date: Wed, 27 Sep 1995 10:49:10 -0700 From: Ted Lemon > Did the move work? I haven't gotten any mail/news since receiving the > above message. Things slow or did I get dropped from the list. As far as I know, it's working fine. The fact that you got the announcement is certainly indicative of that. Hm. It's possible that the link from vix.com isn't working - I'll check on that... _MelloN_ ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <24761-3@lion>; Fri, 29 Sep 1995 12:27:55 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 29 Sep 95 09:37:11 GMT Received: from disperse.demon.co.uk by punt-1.mail.demon.net id aa14920; 29 Sep 95 10:36 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id ac11976; 29 Sep 95 10:36 +0100 Received: from disperse.demon.co.uk by punt-2.mail.demon.net id aa15997; 28 Sep 95 11:22 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id bp23973; 28 Sep 95 6:27 +0100 Received: from relay-1.mail.demon.net by punt-2.mail.demon.net id aa07720; 28 Sep 95 2:57 +0100 Received: from disperse by relay-1.mail.demon.net id aa03971; 27 Sep 95 5:59 +0100 Received: from punt.demon.co.uk by relay-2.mail.demon.net id ad15446; 26 Sep 95 22:04 GMT Received: from disperse.demon.co.uk by punt-1.mail.demon.net id aa24772; 27 Sep 95 5:02 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id fx03836; 26 Sep 95 19:47 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa19982; 25 Sep 95 10:22 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA08748 for perldb-interest-real; Mon, 25 Sep 1995 00:42:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from ppsw3.cam.ac.uk (pp@mauve.csi.cam.ac.uk [131.111.8.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA08744 for ; Mon, 25 Sep 1995 00:42:52 -0700 From: aml2@mrc-lmb.cam.ac.uk Received: from server1.mrc-lmb.cam.ac.uk by mauve.csi.cam.ac.uk with SMTP-CAM (XTPP8.1) as ppsw.cam.ac.uk; Mon, 25 Sep 1995 08:42:52 +0100 Received: from alf1.mrc-lmb.cam.ac.uk by server1.mrc-lmb.cam.ac.uk (5.65) id AA25571; Mon, 25 Sep 1995 08:44:11 +0100 Received: by alf1.mrc-lmb.cam.ac.uk (5.65) id AA21849; Mon, 25 Sep 1995 08:44:11 +0100 Message-Id: <9509250744.AA21849@alf1.mrc-lmb.cam.ac.uk> To: perldb-interest@fugue.com Date: Mon, 25 Sep 1995 08:44:11 +0100 (BST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 27 subscribe perldb-interest ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <24759-5@lion>; Fri, 29 Sep 1995 12:28:08 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 29 Sep 95 09:45:40 GMT Received: from disperse.demon.co.uk by punt-2.mail.demon.net id aa29423; 29 Sep 95 10:45 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id ci11976; 29 Sep 95 10:44 +0100 Received: from disperse.demon.co.uk by punt-2.mail.demon.net id aa11268; 28 Sep 95 12:25 +0100 Received: from punt2.demon.co.uk by relay-2.mail.demon.net id bb23764; 28 Sep 95 5:47 +0100 Received: from relay-1.mail.demon.net by punt-2.mail.demon.net id aa26187; 28 Sep 95 0:57 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa17121; 26 Sep 95 21:41 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA00466 for perldb-interest-real; Tue, 26 Sep 1995 11:31:09 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA00462; Tue, 26 Sep 1995 11:31:03 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id NAA88043; Tue, 26 Sep 1995 13:33:08 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for mellon@fugue.com at austin.ibm.com; id AA19360; Tue, 26 Sep 1995 13:33:07 -0500 Message-Id: <9509261833.AA19360@dax.austin.ibm.com> To: perldb-interest@fugue.com Cc: mellon@fugue.com Subject: was the move successful ? Date: Tue, 26 Sep 95 13:33:06 -0500 _MelloN_ writes: |> I've moved the mailing list from vix.com to fugue.com so that it'll be |> a little easier for me to maintain. I've also added a bouncer to the |> perldb-interest-request mailing list so that it confirms receipt of |> all messages, which will hopefully reduce the number of subscribe |> messages that go to the main list because somebody didn't get an |> acknowledgement and got impatient (this is an amazingly common |> occurrance). I'm thinking of adding a filter to the main list that |> notices administrivia messages and sends them only to me, but I don't |> know when I'll have time to do that. |> |> Meanwhile, if you're telling your friends about the mailing list, |> please tell them the new addresses: |> |> Administrivia: perldb-interest-request@fugue.com |> Content: perldb-interest@fugue.com |> Did the move work? I haven't gotten any mail/news since receiving the above message. Things slow or did I get dropped from the list. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <25812-0@lion>; Fri, 29 Sep 1995 14:24:52 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 29 Sep 95 13:22:06 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa05260; 29 Sep 95 14:19 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA11425 for perldb-interest-real; Fri, 29 Sep 1995 04:30:19 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA11421 for ; Mon, 2 Oct 1995 07:08:05 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 30 Sep 95 02:09:11 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa11745; 30 Sep 95 3:08 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa19404; 30 Sep 95 2:52 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA12652 for perldb-interest-real; Fri, 29 Sep 1995 17:40:22 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA12648 for ; Fri, 29 Sep 1995 17:40:21 -0700 Received: by gw.home.vix.com id AA24063; Fri, 29 Sep 95 17:42:26 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from HQPACAF-Message_Server by wpgate.hqpacaf.af.mil with Novell_GroupWise; Fri, 29 Sep 1995 14:33:36 -1000 Message-Id: X-Mailer: Novell GroupWise 4.1 Date: Fri, 29 Sep 1995 14:32:40 -1000 From: Damon LaCaille To: perldb-interest@vix.com Subject: WOW Errors Sorry to bug everyone again, but I think I've pretty much narrowed my problem down to a few things, and hopefully somebody out on this mailing list can help me out here... I'm finally calling the procedure properly, I'm getting the following error though: dbLogon(): orlon() returns 1034d In my last post I stated what the Oracle errors book says it could be. We've relinked oracle, it was initialized, and the ORACLE_SID environment variable is pointing to the correct instance of Oracle (TRNG). So I don't know what's wrong here... maybe somebody can help? Listed below is the "wow" shell that kicks off everything else. Oh, and before I forget,the 5 programs the WOW docs tell you to run, I ran. BUt only the first two (ht.sql, and wow.sql) compiled correctly, the others (emp.sql, showemp.sql, hanoi.sql and math.sql) said "Warning: package body created with compilation errors". So perhaps that could be the problem as well... well anyhow, here is the current WOW shell we're using to kick off wowstub, etc. ========================================================= #!/bin/csh setenv ORACLE_HOME /ora set path = ( ${ORACLE_HOME}/bin /bin /usr/bin /usr/ucb /usr/local/bin . ) setenv DEBUG NONE setenv WOW_UID scott/tiger setenv ORACLE_SID TRNG if ( $DEBUG == "NONE" ) then ./wowstub $1 else if ( $DEBUG == "POST" ) then setenv REQUEST_METHOD POST setenv PATH_INFO "/math.decimal2nbase" setenv CONTENT_LENGTH 11 echo "a=1234&b=22" | ./wowstub else if ( $DEBUG == "GET" ) then setenv REQUEST_METHOD GET setenv PATH_INFO "/math.decimal2nbase" setenv QUERY_STRING "a=1234&b=22" ./wowstub "a=1234&b=22" else /usr/5bin/echo "Content-type: text/html\n\n" echo "

Error

" echo "The WOW gateway is improperlyconfigured." echo "The DEBUG variable must be set to either NONE," echo "POST, or GET." endif exit 0 ========================================================= So please, if somebody can help, please respond to this e-mail. Thanks a bunch. Damon LaCaille ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08285-44@lion>; Mon, 2 Oct 1995 07:21:18 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 30 Sep 95 09:17:04 GMT Received: from [204.254.239.2] by punt-1.mail.demon.net id aa16378; 30 Sep 95 10:16 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA13150 for perldb-interest-real; Sat, 30 Sep 1995 01:16:55 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA13146 for ; Sat, 30 Sep 1995 01:16:52 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id JAA14788 for perldb-interest@fugue.com; Sat, 30 Sep 1995 09:15:08 +0100 From: Alligator Descartes Message-Id: <199509300815.JAA14788@fruitbat.mcqueen.com> Subject: DBD::Sybase anyone? To: perldb-interest@fugue.com Date: Sat, 30 Sep 1995 09:15:07 +0100 (BST) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 335 Just a thought, does anyone want to pick up developing a Sybase DBD for DBI? If so, let me know. If anyone's willing to give it a crack, and needs some help, I'll put together a "How-to" document for next week. Ta. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.mcqueen.com/hermetica ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08285-50@lion>; Mon, 2 Oct 1995 07:23:15 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 30 Sep 95 12:14:32 GMT Received: from disperse.demon.co.uk by punt-1.mail.demon.net id aa25574; 30 Sep 95 13:14 +0100 Received: from punt.demon.co.uk by relay-2.mail.demon.net id cn03630; 30 Sep 95 13:08 +0100 Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa03816; 27 Sep 95 5:35 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <15625-0@frigate.doc.ic.ac.uk>; Wed, 27 Sep 1995 05:35:22 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA01316 for perldb-interest-real; Tue, 26 Sep 1995 20:09:05 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA01312 for ; Tue, 26 Sep 1995 20:09:03 -0700 Received: by gw.home.vix.com id AA19950; Tue, 26 Sep 95 20:11:09 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-2.mail.demon.net id nh00252; 26 Sep 95 19:36 GMT Received: from ignite.demon.co.uk by post.mail.demon.net id ab19406; 25 Sep 95 12:06 +0100 Received: from ig.co.uk by lion id <15464-0@lion>; Mon, 25 Sep 1995 11:37:32 +0100 To: mellon@fugue.com Subject: Re: Mailing list move... Cc: perldb-interest@vix.com Date: Mon, 25 Sep 1995 11:37:32 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509251206.ab19406@post.mail.demon.net> > From: Ted Lemon > > > I've moved the mailing list from vix.com to fugue.com so that it'll be > a little easier for me to maintain. I've also added a bouncer to the > perldb-interest-request mailing list so that it confirms receipt of > all messages, which will hopefully reduce the number of subscribe > messages that go to the main list because somebody didn't get an > acknowledgement and got impatient (this is an amazingly common > occurrance). I'm thinking of adding a filter to the main list that > notices administrivia messages and sends them only to me, but I don't > know when I'll have time to do that. > > Meanwhile, if you're telling your friends about the mailing list, > please tell them the new addresses: > > Administrivia: perldb-interest-request@fugue.com > Content: perldb-interest@fugue.com > > Thanks! > > _MelloN_ > Thanks Ted. I presume perldb-interest*@vix.com will continue as aliases for a while (they're mentioned in many places). Tim. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <08283-79@lion>; Mon, 2 Oct 1995 07:28:32 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 30 Sep 95 17:06:36 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa02645; 30 Sep 95 18:05 +0100 Received: from punt.demon.co.uk by relay-4.mail.demon.net id cc03122; 30 Sep 95 14:49 +0100 Received: from [204.254.239.2] by punt-1.mail.demon.net id aa24768; 28 Sep 95 11:14 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA05370 for perldb-interest-real; Thu, 28 Sep 1995 01:08:03 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA05366 for ; Thu, 28 Sep 1995 01:08:01 -0700 Received: by gw.home.vix.com id AA10479; Thu, 28 Sep 95 01:09:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by disperse.mail.demon.net id bo12154; 26 Sep 95 14:12 +0100 Received: from [158.152.1.72] by disperse.mail.demon.net id be26985; 26 Sep 95 9:39 +0100 Received: from ignite.demon.co.uk by post.mail.demon.net id aa08704; 25 Sep 95 13:54 +0100 Received: from ig.co.uk by lion id <16132-0@lion>; Mon, 25 Sep 1995 12:35:46 +0100 To: perldb-interest@vix.com, mhm@austin.ibm.com Subject: Re: disconnect is not working.... Date: Mon, 25 Sep 1995 12:35:46 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9509251354.aa08704@post.mail.demon.net> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > From: mhm@austin.ibm.com > |> > Entering parsing phase > |> > disconnect(DBI::db=HASH(0x200c5d1c)) invalidates 1 associated cursor(s) > |> > at genreq.pl line 69. > |> > DBI::END > |> > DBI::disconnect_all DBI > |> > DBI::disconnect_all for 'DB2' > |> > DBI::END complete > |> > SQLSTATE = > |> > ERROR EVENT: DBI::st=HASH(0x200c5cec)->DBI::default_handler: > |> > -2 > |> > at genreq.pl line 0 > |> > > |> I don't know. It might be the invalidated cursor (note the ::st) being > |> DESTROY'd during global destruction and the DESTROY code having problems > |> because it's parent has been disconnect'd (and possibly DESTROY'd). > |> > You're right, user error on my part. I was explicitly calling > $dbh->disconnect() which was being implicitly called by the END block. > Well, actually the DBI only calls $drh->disconnect_all(); for each driver. None of the drivers that I'm aware of actually implement it! Even so calling $dbh->disconnect() multiple times should never be a problem because the DBIcf_COMSET, DBIcf_IMPSET and DBIcf_ACTIVE flags should be tested/reset as appropriate (and, as ever DBD::Oracle is the reference implementation for this). All the above also applies to $sth->finish(). Testing/resetting those flags and understanding that a DESTROY of a parent handle does NOT actually free the parents dbihcom struct while any child dbihcom structs still have references to it are the key points. > Thanks for the pointer. > Mike > mhm@austin.ibm.com > Tim. ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08285-127@lion>; Mon, 2 Oct 1995 07:42:04 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 30 Sep 95 16:39:29 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa04119; 30 Sep 95 17:39 +0100 Received: from punt.demon.co.uk by relay-4.mail.demon.net id dj02880; 30 Sep 95 14:30 +0100 Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ad10369; 28 Sep 95 4:42 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <01493-0@frigate.doc.ic.ac.uk>; Thu, 28 Sep 1995 02:37:39 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA04674 for perldb-interest-real; Wed, 27 Sep 1995 16:38:11 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA04670 for ; Wed, 27 Sep 1995 16:38:06 -0700 Received: by gw.home.vix.com id AA11694; Wed, 27 Sep 95 16:40:03 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by eve.mscs.mu.edu (Smail3.1.28.1 #9) id m0sy61O-0002DiC; Wed, 27 Sep 95 18:36 CDT Message-Id: From: Saima Zobair Subject: PERL4.036 HELP!!! installation stuff To: perldb-interest@vix.com Date: Wed, 27 Sep 1995 18:36:13 -0500 (CDT) X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 12057 Hello, I am in a REAL big mess!! PLEASE HELP!!! I am trying to install perl-4.036 on a solaris2.4. I made changes that were suggested in README.solaris2 (from http://www.hermetica.com). After following the instructions I ran the Configure and makedepend. the function: dbm_clearerr() in hash.c given in the perl4.036 installation packet is giving me the problem!! Also I dont understand the output from the 'make' very well. Please send me an answer !! Thanks, Saima Zobair zobair@mscs.mu.edu On running the 'make' I got the following output: Expect 27 shift/reduce and 57 reduce/reduce conflicts /usr/ccs/bin/yacc -d perly.y conflicts: 27 shift/reduce, 57 reduce/reduce sh ./perly.fixer y.tab.c perly.c mv y.tab.h perly.h echo 'extern YYSTYPE yylval;' >>perly.h `sh cflags perly.o` perly.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from perly.y:41: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags array.o` array.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from array.c:25: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags cmd.o` cmd.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from cmd.c:35: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags cons.o` cons.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from cons.c:36: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags consarg.o` consarg.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from consarg.c:35: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags doarg.o` doarg.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from doarg.c:61: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags doio.o` doio.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from doio.c:52: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition In file included from /usr/include/fcntl.h:18, from doio.c:97: /usr/include/sys/fcntl.h:137: warning: `O_ACCMODE' redefined /usr/local/include/compat.h:153: warning: this is the location of the previous definition `sh cflags dolist.o` dolist.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from dolist.c:52: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags dump.o` dump.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from dump.c:22: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags eval.o` eval.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from eval.c:48: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition In file included from /usr/include/fcntl.h:18, from eval.c:55: /usr/include/sys/fcntl.h:137: warning: `O_ACCMODE' redefined /usr/local/include/compat.h:153: warning: this is the location of the previous definition `sh cflags form.o` form.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from form.c:32: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags perl.o` perl.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from perl.c:60: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags regcomp.o` regcomp.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from regcomp.c:75: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags regexec.o` regexec.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from regexec.c:66: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags stab.o` stab.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from stab.c:45: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags str.o` str.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from str.c:48: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition `sh cflags toke.o` toke.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from toke.c:62: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition In file included from /usr/include/fcntl.h:18, from toke.c:68: /usr/include/sys/fcntl.h:137: warning: `O_ACCMODE' redefined /usr/local/include/compat.h:153: warning: this is the location of the previous definition `sh cflags util.o` util.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from util.c:46: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition In file included from /usr/include/fcntl.h:18, from util.c:61: /usr/include/sys/fcntl.h:137: warning: `O_ACCMODE' redefined /usr/local/include/compat.h:153: warning: this is the location of the previous definition `sh cflags hash.o` hash.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from hash.c:26: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition In file included from /usr/include/fcntl.h:18, from hash.c:588: /usr/include/sys/fcntl.h:137: warning: `O_ACCMODE' redefined /usr/local/include/compat.h:153: warning: this is the location of the previous definition `sh cflags usersub.o` usersub.c CCCMD = gcc -c -fpcc-struct-return -traditional -Dvolatile=__volatile__ -I/opt/gnu/lib/gcc-lib/sparc-sun-solaris2/2.5.6/include -O In file included from /usr/local/include/db.h:40, from /usr/local/include/ndbm.h:42, from perl.h:332, from usersub.c:20: /usr/local/include/sys/cdefs.h:68: warning: `volatile' redefined *Initialization*:1: warning: this is the location of the previous definition /opt/gnu/bin/gcc array.o cmd.o cons.o consarg.o doarg.o doio.o dolist.o dump.o eval.o form.o perl.o regcomp.o regexec.o stab.o str.o toke.o util.o hash.o perly.o usersub.o -lsocket -lnsl -lm -o perl Undefined first referenced symbol in file dbm_clearerr hash.o ld: fatal: Symbol referencing errors. No output written to perl make: *** [perl] Error 1 ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08285-149@lion>; Mon, 2 Oct 1995 07:46:45 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 30 Sep 95 17:05:03 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa15973; 30 Sep 95 18:04 +0100 Received: from punt.demon.co.uk by relay-4.mail.demon.net id be03122; 30 Sep 95 14:47 +0100 Received: from [204.254.239.2] by punt-1.mail.demon.net id aa28703; 28 Sep 95 10:12 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA05271 for perldb-interest-real; Thu, 28 Sep 1995 00:17:59 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA05267 for ; Thu, 28 Sep 1995 00:17:57 -0700 Received: by gw.home.vix.com id AA08215; Thu, 28 Sep 95 00:20:04 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by solair1.inter.NL.net (5.65b/solair1.Inter.NL.net-1.31) id AA16275; Thu, 28 Sep 1995 08:19:53 +0100 Date: Thu, 28 Sep 1995 08:19:52 +0100 (MET) From: Marti Rijken To: perldb-interest@vix.com Subject: Re: DBD-mSQL-0.60pl2 test failed In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In followup to my mail from 18 Sep, that has just arrived in perldb-interest, I can say that my question has been solved for some time (Thanks Alligator Descartes!). Somehow a mail relay had blocked my message for 10 days.... -Marti Rijken ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <08285-174@lion>; Mon, 2 Oct 1995 07:52:27 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 01 Oct 95 03:27:04 GMT Received: from [204.254.239.2] by punt-1.mail.demon.net id aa09474; 1 Oct 95 4:26 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA15646 for perldb-interest-real; Sat, 30 Sep 1995 19:09:53 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id TAA15642 for ; Sat, 30 Sep 1995 19:09:50 -0700 Received: by gw.home.vix.com id AA20315; Sat, 30 Sep 95 19:12:00 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from rodan.syr.edu (spoliset@rodan.syr.EDU [128.230.1.55]) by mailsender.syr.edu (8.7/8.7) with SMTP id WAA16659 for ; Sat, 30 Sep 1995 22:23:22 -0400 (EDT) Received: by rodan.syr.edu (4.1/Spike-2.0) id AA05887; Sat, 30 Sep 95 22:11:24 EDT Date: Sat, 30 Sep 1995 22:11:24 -0400 (EDT) From: Srinivas Polisetty X-Sender: spoliset@rodan.syr.edu To: perldb-interest@vix.com Subject: oraperl installation errors... please help... In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi All, Could someone help me with my installation of 'oraperl' for which I had been struggling for a while now, and in vain. My machine specs: Sun4 Sparc, running SunOS UNIX My makefile reads: --------------------------------------------------------- ORACLE_HOME = /usr/oracle SRC = /home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 OTHERLIBS = `cat $(ORACLE_HOME)/rdbms/lib/sysliblist` CLIBS = $(OTHERLIBS) OCILIB = $(ORACLE_HOME)/lib/libocic.a NETLIBS = $(ORACLE_HOME)/lib/osntab.o \ $(ORACLE_HOME)/lib/libsqlnet.a ORALIBS = $(ORACLE_HOME)/lib/libora.a ALL_ORA_LIBS = $(CLIBS) $(OCILIB) $(NETLIBS) $(ORALIBS) GLOBINCS = LOCINCS = LIBS = `. $(SRC)/config.sh; echo $$libs` DEBUG = -DPERL_DEBUGGING DBUG_O = dbug/dbug.o CURSELIB = -lcurses TESTDATA = nodb furma TianpfWS SRCS = oracle.mus orafns.c getcursor.c colons.c usersub.c \ debug.c strtoul.c OBJS = oracle.o orafns.o getcursor.o colons.o debug.o $(STRTOUL) OOBJS = $(OBJS) usersub.o $(DBUG_O) COBJS = $(OBJS) cusersub.o $(DBUG_O) HDRS = patchlevel.h orafns.h DEFS = $(STRTOL) $(PUTENV) $(STR_2MORTAL) $(DEBUG) $(CACHE) $(BIND) CFLAGS = -Idbug -I$(SRC) $(GLOBINCS) $(LOCINCS) $(DEFS) oraperl: $(SRC)/uperl.o $(OOBJS) $(CC) -o oraperl $(SRC)/uperl.o $(OOBJS) \ -lm $(ALL_ORA_LIBS) $(LIBS) coraperl: $(SRC)/uperl.o $(COBJS) $(SRC)/usub/curses.o $(CC) -o coraperl $(SRC)/uperl.o $(COBJS) $(SRC)/usub/curses.o \ -lm $(ALL_ORA_LIBS) $(LIBS) $(CURSELIB) all: oraperl coraperl test: oraperl @oraperl -e '&ora_version' @(cd testdir ; \ rm -f My-Results ; \ echo "Testing oraperl, please wait ..." ; \ for i in *.pl ; do ../oraperl $$i $(TESTDATA) ; done > My-Results ; \ if cmp -s Standard-Results My-Results ; \ then echo "Test successful" ; \ else echo "Test failed - compare My-Results with Standard-Results" ;\ fi; echo) install: oraperl install.pl @./oraperl ./install.pl ${SRC} cusersub.c: usersub.c @rm -f cusersub.c ln usersub.c cusersub.c cusersub.o: cusersub.c $(CC) -c $(CFLAGS) -DCURSES cusersub.c oracle.c: $(SRC)/usub/mus oracle.mus $(SRC)/usub/mus oracle.mus >oracle.c $(OOBJS) $(COBJS): $(HDRS) dbug/dbug.o: (cd dbug ; $(MAKE) dbug.o) @echo " (back to main directory)" clean: (cd dbug ; $(MAKE) clean) @echo " (back to main directory)" rm -f nohup.out *.o oracle.c cusersub.c rm -f testdir/My-Results listing tags core realclean clobber: clean (cd dbug ; $(MAKE) clobber) (cd doc ; $(MAKE) clobber) @echo " (back to main directory)" rm -f oraperl coraperl listing: pr -fn Makefile $(HDRS) $(SRCS) >listing docs: (cd doc ; $(MAKE) docs) @echo " (back to main directory)" shar: clean shar -n oraperl-v2 -a -s kstock@encore.com -F -o :Part -l 64 \ Readme [C-Q]* Row* [S-z]* ------------------------------------------------------------------ Also a transcript of the session that followed: pacman:/home/T6D/HPFI/users/sairam/installed/oraperl/oraperl-v2.4% make oraperl cc -Idbug -I/home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 -DPERL_DE BUGGING -target sun4 -c oracle.c cc -Idbug -I/home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 -DPERL_DE BUGGING -target sun4 -c orafns.c cc -Idbug -I/home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 -DPERL_DE BUGGING -target sun4 -c getcursor.c cc -Idbug -I/home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 -DPERL_DE BUGGING -target sun4 -c colons.c cc -Idbug -I/home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 -DPERL_DE BUGGING -target sun4 -c debug.c cc -Idbug -I/home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 -DPERL_DE BUGGING -target sun4 -c strtoul.c cc -Idbug -I/home/T6D/HPFI/users/sairam/installed/perl/perl-4.036 -DPERL_DE BUGGING -target sun4 -c usersub.c (cd dbug ; make dbug.o) cc -target sun4 -c dbug.c (back to main directory) cc -o oraperl /home/T6D/HPFI/users/sairam/installed/perl/perl-4.036/uperl.o orac le.o orafns.o getcursor.o colons.o debug.o strtoul.o usersub.o dbug/dbug.o \ -lm `cat /usr/oracle/rdbms/lib/sysliblist` /usr/oracle/rdbms/lib/libocic.a /usr/oracle/rdbms/lib/osntab.o /usr/oracle/rdbms/lib/libsqlnet.a /usr/oracle/r dbms/lib/libora.a `. /home/T6D/HPFI/users/sairam/installed/perl/perl-4.036/confi g.sh; echo $libs` ld: /usr/oracle/rdbms/lib/libocic.a: No such file or directory *** Error code 4 make: Fatal error: Command failed for target `oraperl' ------------------------------- Then I saw that the required files were in /usr/oracle/lib directory and changed the variables as they appear on the Makefile reproduced above. And when I said 'make oraperl' again, I had the following error. ------------------------------- pacman:/home/T6D/HPFI/users/sairam/installed/oraperl/oraperl-v2.4% make oraperl cc -o oraperl /home/T6D/HPFI/users/sairam/installed/perl/perl-4.036/uperl.o orac le.o orafns.o getcursor.o colons.o debug.o strtoul.o usersub.o dbug/dbug.o \ -lm `cat /usr/oracle/rdbms/lib/sysliblist` /usr/oracle/lib/libocic.a /usr/ oracle/lib/osntab.o /usr/oracle/lib/libsqlnet.a /usr/oracle/lib/libora.a `. /ho me/T6D/HPFI/users/sairam/installed/perl/perl-4.036/config.sh; echo $libs` ld: Undefined symbol _ligraw _liclas _licbas _lircpu _lmemupr _sslsig _lmfeecb _limopen _lintrn _ligcnv _ldtsto _liwalpx _limspax _limcpcx _limpuc _lmemup _lihcurrlangid _lmsrcbn _lstrlcomp _limnceq _ldtini _ss_memmove _ligstm _sldtgd _ldtdts _lmemlcomp _lnmscn _lnmren _lmemlcmp _lmfegks _lmsrtrm _limc2wx _lihmupr _lmsrgbf _liwc2ux _lincml _liisch _sltbev _lmemlo _lmsrcin _lmsrip _sslsleep _ligpth _liiskip _limr2w _ligratio _limdigx _liwlowx _lircop _ora_sprintf _btoi _lmfehtl _lmfepwe _ora_vsprintf _limfwdx _liwdigx _osnqrn _liwc2mx _liwctex _limalpx _lnmnur _limcpen _lihci2h _lihcrenv _ligncc _liicml _liclas _debug _lihcsn _lihlinfo _init_oracle _licbas *** Error code 2 make: Fatal error: Command failed for target `oraperl' ----------------------------------------- Some thing is terribly wrong, and I can't figure out what!! Could anyone please help me with this. Thank you very much, Srinvas. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <19055-0@lion>; Mon, 2 Oct 1995 17:33:43 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 02 Oct 95 16:30:48 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa08617; 2 Oct 95 17:29 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA19836 for perldb-interest-real; Mon, 2 Oct 1995 06:21:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA19832 for ; Mon, 2 Oct 1995 06:21:22 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id IAA25055 for ; Mon, 2 Oct 1995 08:23:34 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA24494; Mon, 2 Oct 1995 08:23:29 -0500 Message-Id: <9510021323.AA24494@dax.austin.ibm.com> To: Chet Murthy Cc: perldb-interest@fugue.com Subject: Re: DB2perl In-Reply-To: Your message of Sun, 01 Oct 95 00:47:56 -0400. <9510010447.AA35053@lusitania.watson.ibm.com> Date: Mon, 02 Oct 95 08:23:29 -0500 Tim, What sort of optimizations were thinking about for dbd_st_fetch? DBD::Oracle, as well as DBD::DB2 both carry your comment about the ability to optimize later. One of my alpha testers is asking if the following change is a valid fix for null variables returned from the database. Given the availability of null indicators I see no reason to make the change unless it can be worked into the optimizations you have been thinking about. Mike mhm@austin.ibm.com Chet Murthy writes: > > > Hi, Mike, > > I found two problems. I don't know what to do about either, but I > figure I'll throw them your way, and see what you think: > > (1) When a field is NULL, the "rlen" of a fbh_t ends up being set to > -1. This isn't handled, and I change the following in dbd_st_fetch. > > #ifdef ISTHISAFIX > if (fbh->rlen < 0) { > sv_setsv (sv,Nullsv); > } else { > SvCUR(fbh->sv) = fbh->rlen; > sv_setsv(sv, fbh->sv); /* XXX can be optimised later */ > } > #else > SvCUR(fbh->sv) = fbh->rlen; > sv_setsv(sv, fbh->sv); /* XXX can be optimised later */ > #endif > > The problem is that if you project on a single field, and that field > happens come up NULL sometimes ... Perhaps the right thing to do is > to return instead the empty string. But then how to differentiate > that from an empty string ... > > If you get the IBMSAMPL database, then you can tickle this problem with > the perl script I enclose in this file, and: > That's the entire purpose of null indicators. Do your select using a null indicator variable, then either if or unless to skip your code on a null. This also gets rid of the differentiation problem. ---+++--- Received: from punt2.demon.co.uk by lion with SMTP (PP) id <19153-0@lion>; Mon, 2 Oct 1995 17:43:47 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 02 Oct 95 16:43:05 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa14577; 2 Oct 95 17:41 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA20003 for perldb-interest-real; Mon, 2 Oct 1995 07:55:04 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA19999 for ; Mon, 2 Oct 1995 07:55:02 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id JAA44311 for ; Mon, 2 Oct 1995 09:57:10 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA26246; Mon, 2 Oct 1995 09:57:09 -0500 Message-Id: <9510021457.AA26246@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: Dealing with null indicator variables Date: Mon, 02 Oct 95 09:57:09 -0500 How does one deal with null indicator variables specified by the enduser? From what I can tell it doesn't appear to be done at the present time. Am I correct or just missing the appropriate piece of magic? Mike -------------------------------------------------------------------------- Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <25744-15@lion>; Tue, 3 Oct 1995 02:30:29 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 03 Oct 95 01:24:56 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ai10955; 3 Oct 95 2:23 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <25398-0@frigate.doc.ic.ac.uk>; Tue, 3 Oct 1995 00:01:56 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA20744 for perldb-interest-real; Mon, 2 Oct 1995 13:29:45 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA20740 for ; Mon, 2 Oct 1995 13:29:38 -0700 Received: by gw.home.vix.com id AA29890; Mon, 2 Oct 95 13:31:48 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from euterpe.cas.american.edu by beach.cas.american.edu (8.6.10/) id QAA25148; Mon, 2 Oct 1995 16:40:45 -0400 Received: by euterpe.cas.american.edu (8.6.10) id QAA21963; Mon, 2 Oct 1995 16:25:07 -0400 Date: Mon, 2 Oct 1995 16:25:07 -0400 (EDT) From: Abinash Tripathy X-Sender: abinash@euterpe To: perldb-interest@vix.com Subject: Make Errors in DBI on SunOS 4.1.3 and Perl 5.0 Message-Id: Mime-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="1430353583-216079660-812665507=:21959" This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. --1430353583-216079660-812665507=:21959 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I have followed the steps needed to Make DBI. 1. >perl Makefile.PL (works fine - no errors) 2. >make (error - listed in the attachment) Any help will be greatly appreciated. Thanks, Abinash Abinash Tripathy 4201 Massachusetts Ave, N.W Apt 3030C Ph:(202) 363 0604(home) Washington D.C 20016 (202) 885 1479 (FAX) (202) 885 2767(work) email:abinash@email.cas.american.edu at0509a@american.edu WWW :http://www.cas.american.edu/~abinash/ -- --1430353583-216079660-812665507=:21959 Content-Type: TEXT/PLAIN; charset=US-ASCII; name="makerror.txt" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Bummer bWFrZQ0KbWtkaXIgLi9ibGliDQpta2RpciAuL2JsaWIvYXV0bw0KbWtkaXIg Li9ibGliL2F1dG8vREJJDQovdXNyL2xvY2FsL2Jpbi9wZXJsIC1JL3Vzci9s b2NhbC9saWIvcGVybDUvc3VuNC1zdW5vcyAtSS91c3IvbG9jYWwvbGliL3Bl cmw1IC91cw0Kci9sb2NhbC9saWIvcGVybDUvRXh0VXRpbHMveHN1YnBwIC10 eXBlbWFwIC91c3IvbG9jYWwvbGliL3Blcmw1L0V4dFV0aWxzL3R5cGVtYXAN CiBEQkkueHMgPkRCSS50YyAmJiBtdiBEQkkudGMgREJJLmMNCmNjIC1jIC1P IC1waWMgLUkvdXNyL2xvY2FsL2xpYi9wZXJsNS9zdW40LXN1bm9zL0NPUkUg ICBEQkkuYw0KLi9EQklYUy5oOiAzNTogdW5kZWZpbmVkIGNvbnRyb2wNCm1h a2U6ICoqKiBbREJJLm9dIEVycm9yIDI= --1430353583-216079660-812665507=:21959-- ---+++--- Received: from punt.demon.co.uk by lion with SMTP (PP) id <25744-17@lion>; Tue, 3 Oct 1995 02:30:37 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 03 Oct 95 01:24:59 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aj10955; 3 Oct 95 2:24 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <25491-0@frigate.doc.ic.ac.uk>; Tue, 3 Oct 1995 00:06:12 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA20820 for perldb-interest-real; Mon, 2 Oct 1995 13:49:56 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from studsys.mscs.mu.edu (studsys.mscs.mu.edu [134.48.4.25]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA20816 for ; Mon, 2 Oct 1995 13:49:52 -0700 Received: from studsys by studsys.mscs.mu.edu with smtp (Smail3.1.28.1 #9) id m0szrqD-0001NdC; Mon, 2 Oct 95 15:52 CDT Message-Id: Date: Mon, 02 Oct 95 15:52:01 -0500 From: Malathi Gade X-Mailer: Mozilla 1.1N (X11; I; SunOS 5.4 sun4m) MIME-Version: 1.0 To: perldb-interest@fugue.com Subject: Installation of DBI!! HELP!! X-URL: http://www.hermetica.com/technologia/DBI/faq.html#q13 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Hello, I am trying to install DBI. I am getting an error message which seems like I am missing a config.sh file in perl directory... Can you please help me with this. thanks, Malathi Gade gade@mscs.mu.edu here it goes: Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker NAME = 'DBI' VERSION = '0.63' Use of uninitialized value at /usr/local/lib/perl5/ExtUtils/MakeMaker.pm line 198. Can't find config.sh at /usr/local/lib/perl5/ExtUtils/MakeMaker.pm line 198. -- ********************************************************************************Malathi Gade Department of Mathematics, Statistics and Computerscience Marquette University Milwaukee WI 53233 Voice-mail: (414)342-2288 E-mail: gade@studsys.mscs.mu.edu ******************************************************************************** ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06079-2@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:05:53 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 03 Oct 95 21:01:10 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa14873; 3 Oct 95 22:00 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa10839; 3 Oct 95 20:57 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA23870 for perldb-interest-real; Tue, 3 Oct 1995 10:19:51 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from hub.ucsb.edu (hub.ucsb.edu [128.111.24.40]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA23866 for ; Tue, 3 Oct 1995 10:19:45 -0700 Received: from dokoka (dokoka.ucsb.edu) by hub.ucsb.edu; id AA16472 sendmail 4.1/UCSB-2.1-sun Tue, 3 Oct 95 10:21:31 PDT for perldb-interest@fugue.com Received: from dokoka.ucsb.edu by dokoka via ESMTP (940816.SGI.8.6.9/940406.SGIv2) id KAA10281; Tue, 3 Oct 1995 10:21:25 -0700 Message-Id: <199510031721.KAA10281@dokoka> To: mhm@austin.ibm.com Cc: perldb-interest@fugue.com Subject: Re: Fetching null columns Reply-To: "Alan K. Stebbens" In-Reply-To: Your message of Tue, 03 Oct 1995 10:40:51 PDT. <9510031540.AA21984@dax.austin.ibm.com> Date: Tue, 03 Oct 1995 10:21:24 -0700 From: Alan Stebbens > Ok, question time. How do we expect the DBI::DBD interfaces to work > when it comes to columns returning null? > > Suppose we have a table which contains a mixture of null and nonnull > columns. As it stands now, the order of the columns is determined by > either the order in the requesting sql statement or by the order > returned by "DescribeColumn" (name determined by whether you DB vendor > if more closely following X/Open CLI or MicroSoft ODBC standards). As > such we have a small problem. > > Take the following: > > > #!/usr/local/bin/perl > use DBI; > > DBI->_debug_dispatch(5); > $dbh = DBI->connect('sample'); > > $stmt = $ARGV[0]; > $sth = $dbh->prepare($stmt); > print "NUM Params: ",$sth->{'NUM_OF_PARAMS'},"\n"; > $sth->execute(); > while (@row = $sth->fetchrow) { > $" = "||"; print "Row: @row\n"; > } > $sth->finish(); > > Depending on the number and order of requested columns the while loop > may not execute. If the $stmt requests only one column from the table > and the column can contain NULLs then it is possible to return only > part of the data. If $stmt requests more than one column and the first > column requested can contain NULLs, we break again. > > I guess, my point is that we need some method of returning something > other than a Nullsv or sv_undef and giving the enduser a method of > checking the null indicator from the fbh struct. Why not have "fetchrow" return "undef" for `no more data'? Then, the loop looks like this: while (defined(@row = $sth->fetchrow)) { ... } Each column (array element) can then be "undef" (for null), or a defined value, including the empty string. A special "isnull" function is unnecessary. > perl -de 0 Loading DB routines from $RCSfile: perl5db.pl,v $$Revision: 4.1 $$Date: 92/08/07 18:24:07 $ Emacs support available. Enter h for help. main::(-e:1): 0 DB<1> $x[0] = 1 DB<2> $x[1] = '' DB<3> $x[2] = 2 DB<4> $x[3] = undef DB<5> $x[4] = 'foo' DB<6> p join(', ', @x) 1, , 2, , foo DB<7> p join(', ', map(defined, @x)) 1, 1, 1, , 1 DB<8> ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06079-7@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:06:49 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 02:19:40 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa08290; 4 Oct 95 3:18 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa10141; 4 Oct 95 3:11 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA25119 for perldb-interest-real; Tue, 3 Oct 1995 17:48:30 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA25115 for ; Tue, 3 Oct 1995 17:48:28 -0700 Received: by gw.home.vix.com id AA28419; Tue, 3 Oct 95 17:50:39 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id bb26504; 3 Oct 95 12:45 +0100 Received: from ig.co.uk by lion id <26421-0@lion>; Tue, 3 Oct 1995 02:36:16 +0100 To: perldb-interest@vix.com, abinash@cage.home.vix.com Subject: Re: Make Errors in DBI on SunOS 4.1.3 and Perl 5.0 Date: Tue, 3 Oct 1995 02:36:16 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510031245.bb26504@relay-3.mail.demon.net> > From: Abinash Tripathy > Content-Type> : > MULTIPART/MIXED> ; > BOUNDARY="1430353583-216079660-812665507=:21959"> > > I have followed the steps needed to Make DBI. > > 1. >perl Makefile.PL (works fine - no errors) > > 2. >make (error - listed in the attachment) > > Any help will be greatly appreciated. Thanks, Abinash > > Abinash Tripathy > 4201 Massachusetts Ave, N.W Apt 3030C Ph:(202) 363 0604(home) > Washington D.C 20016 (202) 885 1479 (FAX) (202) 885 2767(work) > email:abinash@email.cas.american.edu at0509a@american.edu > WWW :http://www.cas.american.edu/~abinash/ > -- > > --1430353583-216079660-812665507=:21959 > Content-Type: TEXT/PLAIN; charset=US-ASCII; name="makerror.txt" > Content-Transfer-Encoding: BASE64 > Content-ID: > Content-Description: Bummer > > bWFrZQ0KbWtkaXIgLi9ibGliDQpta2RpciAuL2JsaWIvYXV0bw0KbWtkaXIg > Li9ibGliL2F1dG8vREJJDQovdXNyL2xvY2FsL2Jpbi9wZXJsIC1JL3Vzci9s > b2NhbC9saWIvcGVybDUvc3VuNC1zdW5vcyAtSS91c3IvbG9jYWwvbGliL3Bl > cmw1IC91cw0Kci9sb2NhbC9saWIvcGVybDUvRXh0VXRpbHMveHN1YnBwIC10 > eXBlbWFwIC91c3IvbG9jYWwvbGliL3Blcmw1L0V4dFV0aWxzL3R5cGVtYXAN > CiBEQkkueHMgPkRCSS50YyAmJiBtdiBEQkkudGMgREJJLmMNCmNjIC1jIC1P > IC1waWMgLUkvdXNyL2xvY2FsL2xpYi9wZXJsNS9zdW40LXN1bm9zL0NPUkUg > ICBEQkkuYw0KLi9EQklYUy5oOiAzNTogdW5kZWZpbmVkIGNvbnRyb2wNCm1h > a2U6ICoqKiBbREJJLm9dIEVycm9yIDI= > --1430353583-216079660-812665507=:21959-- > You'll find me rather more helpful if you avoid encodings I can't read as plain text since I'm not using a mime reader and I've not time to fiddle. Given the size of the attachment I presume you've not read the part of the README which lists what to info to send. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06079-11@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:08:06 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 02:36:06 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa13235; 4 Oct 95 3:34 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa06468; 3 Oct 95 19:50 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA23664 for perldb-interest-real; Tue, 3 Oct 1995 08:38:49 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA23660 for ; Tue, 3 Oct 1995 08:38:47 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id KAA120593 for ; Tue, 3 Oct 1995 10:40:53 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA21984; Tue, 3 Oct 1995 10:40:51 -0500 Message-Id: <9510031540.AA21984@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: Fetching null columns Date: Tue, 03 Oct 95 10:40:51 -0500 Ok, question time. How do we expect the DBI::DBD interfaces to work when it comes to columns returning null? Suppose we have a table which contains a mixture of null and nonnull columns. As it stands now, the order of the columns is determined by either the order in the requesting sql statement or by the order returned by "DescribeColumn" (name determined by whether you DB vendor if more closely following X/Open CLI or MicroSoft ODBC standards). As such we have a small problem. Take the following: #!/usr/local/bin/perl use DBI; DBI->_debug_dispatch(5); $dbh = DBI->connect('sample'); $stmt = $ARGV[0]; $sth = $dbh->prepare($stmt); print "NUM Params: ",$sth->{'NUM_OF_PARAMS'},"\n"; $sth->execute(); while (@row = $sth->fetchrow) { $" = "||"; print "Row: @row\n"; } $sth->finish(); Depending on the number and order of requested columns the while loop may not execute. If the $stmt requests only one column from the table and the column can contain NULLs then it is possible to return only part of the data. If $stmt requests more than one column and the first column requested can contain NULLs, we break again. I guess, my point is that we need some method of returning something other than a Nullsv or sv_undef and giving the enduser a method of checking the null indicator from the fbh struct. We already maintain the information, so it is just a matter of making it available to the enduser. In essence, an $sth->isnull($index) function or $sth->{'Column Name'} Problems: For $sth->isnull($index): 1) requires the enduser to be able to determine the order of the columns requested. Not much of a problem if the enduser specified all the columns in the SQL, but a major problem is the enduser did a "select * from...." For $sth->{'column name'}: 1) the memory overhead. 2) Do we create a hash for every column or only those that can contain NULLs. What do we return if the enduser decides to test a column which can not contain NULLs. Any thoughts? Comments? Mike mhm@austin.ibm.com 1) provide --------------------------------------------------------------------------- Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06079-14@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:10:45 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 04:49:37 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa20545; 4 Oct 95 5:49 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa04567; 4 Oct 95 1:53 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA24802 for perldb-interest-real; Tue, 3 Oct 1995 16:34:07 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from eve.mscs.mu.edu (eve.mscs.mu.edu [134.48.4.166]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA24798 for ; Tue, 3 Oct 1995 16:34:05 -0700 Received: by eve.mscs.mu.edu (Smail3.1.28.1 #9) id m0t0Gp6-0002DjC; Tue, 3 Oct 95 18:32 CDT Message-Id: From: Saima Zobair Subject: subscribe To: perldb-interest@fugue.com Date: Tue, 3 Oct 1995 18:32:32 -0500 (CDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 138 Hello, I would like to subscribe to the perldb-interest mailing list. Please include me in!! Thanks, Saima Zobair zobair@mscs.mu.edu ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06079-17@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:11:21 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 05:50:35 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa04078; 4 Oct 95 6:49 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa16352; 4 Oct 95 4:37 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA25469 for perldb-interest-real; Tue, 3 Oct 1995 19:08:05 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id TAA25465 for ; Tue, 3 Oct 1995 19:07:59 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id be26504; 3 Oct 95 12:46 +0100 Received: from ig.co.uk by lion id <26473-0@lion>; Tue, 3 Oct 1995 02:47:56 +0100 To: skimo@dns.ufsia.ac.be Subject: Re: Multiple DBD's Cc: perldb-interest@fugue.com Date: Tue, 3 Oct 1995 02:47:56 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510031246.be26504@relay-3.mail.demon.net> > From: Sven Verdoolaege > > > > From: Sven Verdoolaege > > > > > > I hope my email troubles are over; if not and if you > > > get this message but can't reach me here, try SVEN.VERDOOLAEGE@student.kuleuven.ac.be; though not until tomorrow. > > > > > > I sent you a long version of my question earlier (don't know whether you got it) > > > > > I got it and replied. My reply may be heldup/lost. > > > > > Here's the short version: > > > Can i connect to two databases using different DBD's simultaneously > > > > Yes. > > > > > If it is possible, could you give me a small example ? > > > (something with an Oracle and an mSQL db is the object) > > > > > $dbh1 = DBI->connect(..., 'Oracle') or die "..."; > > $dbh2 = DBI->connect(..., 'mSQL') or die "..."; > > > doesn't seem to work for me; maybe this is a bug in DBD::mSQL then ? Maybe. CC'd to perldb-interest. > my script: > #!/u01/home/verdools/bin/perl > use DBI; > $DB='ufpers'; > $ENV{'MSQL_HOME'} = "/usr/local/Minerva"; > $ENV{'MSQL_TCP_PORT'} = 22532; > $dbhD = DBI->connect( 'dns',$DB,'','mSQL') || die("geen database"); > $dbhD->disconnect; > $ENV{'ORACLE_HOME'} = "/u01/home/oracle/product/7.1.4"; > $ENV{'ORAHOME'} = "/u01/home/oracle/product/7.1.4"; > print "before\n"; > $dbhS = DBI->connect( $DB, 'job_verdool_s', 'my password:-)','Oracle' ) || die("geen ; > print "after\n"; > $dbhS->disconnect; > > my output: > before > rc: 0 what: Unknown MSQL Server Host (ufpers) > Database handle DBI::db=HASH(0x1400fff18) DESTROY ignored - never set up at /u0. > geen database at ./cp2 line 11. > > if i comment the connect and disconnect lines from mSQL, i get > before > after > > if i put the oracle stuff before the mSQL stuff, i get > before > after > Memory fault - core dumped > A stack trace from the core dump may be very helpful. A log trace from PERL_DBI_DEBUG=3 may also be very helpful. > Sven > Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06080-21@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:12:48 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 03 Oct 95 22:53:24 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id aj20995; 3 Oct 95 23:52 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <07872-0@frigate.doc.ic.ac.uk>; Tue, 3 Oct 1995 22:26:01 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA24204 for perldb-interest-real; Tue, 3 Oct 1995 12:58:58 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA24200 for ; Tue, 3 Oct 1995 12:58:56 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id PAA72943; Tue, 3 Oct 1995 15:01:09 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for mhm@austin.ibm.com at austin.ibm.com; id AA25860; Tue, 3 Oct 1995 15:01:03 -0500 Message-Id: <9510032001.AA25860@dax.austin.ibm.com> To: "Alan K. Stebbens" Cc: mhm@austin.ibm.com, perldb-interest@fugue.com Subject: Re: Fetching null columns In-Reply-To: Your message of Tue, 03 Oct 95 10:21:24 -0700. <199510031721.KAA10281@dokoka> Date: Tue, 03 Oct 95 15:01:02 -0500 Alan Stebbens writes: > > Why not have "fetchrow" return "undef" for `no more data'? Then, the > loop looks like this: > > while (defined(@row = $sth->fetchrow)) { > ... > } > > Each column (array element) can then be "undef" (for null), or a defined > value, including the empty string. A special "isnull" function is > unnecessary. If the first item in @row is undef then the loop is terminated. Other than defining a new method of detmining a NULL column, the only thing left is to do something along the lines of "($status,@row) = $sth->fetchrow();" or assuming that $row[0] is always the status. Mike mhm@austin.ibm.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06080-25@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:13:43 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 03 Oct 95 22:53:22 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ak20995; 3 Oct 95 23:52 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <07959-0@frigate.doc.ic.ac.uk>; Tue, 3 Oct 1995 22:29:32 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA24198 for perldb-interest-real; Tue, 3 Oct 1995 12:57:33 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA24194 for ; Tue, 3 Oct 1995 12:57:32 -0700 From: lm_truong@ccmail.pnl.gov Received: by gw.home.vix.com id AA07360; Tue, 3 Oct 95 12:59:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ccmail.pnl.gov by pnl.gov (PMDF V4.3-13 #6012) id <01HW0651XB3494DNM0@pnl.gov>; Tue, 03 Oct 1995 12:59:42 -0700 (PDT) Date: Tue, 03 Oct 1995 13:10 -0700 (PDT) Subject: oraperl &ora_open and &ora_close problem To: perldb-interest@vix.com Message-Id: <01HW0651YCR694DNM0@pnl.gov> Mime-Version: 1.0 Content-Transfer-Encoding: 7BIT I have a problem with oraperl &ora_open and &ora_close. Whenever I execute a sql statement using ora_open, I have to close it. This degrades my program's performance. Is there a way to tell ORACLE to reuse the cursor in oraperl? Thanks in advance for any help! Lien ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06080-34@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:15:34 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 03 Oct 95 23:53:52 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ag17202; 4 Oct 95 0:52 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <11401-0@frigate.doc.ic.ac.uk>; Wed, 4 Oct 1995 00:47:19 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA24624 for perldb-interest-real; Tue, 3 Oct 1995 15:07:07 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from merlin.nando.net (root@merlin.nando.net [152.52.2.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA24620 for ; Tue, 3 Oct 1995 15:06:59 -0700 Received: from nando.net.nando.net (parsifal.nando.net) by merlin.nando.net (4.1/davel-nando/dec93) id AA14039; Tue, 3 Oct 95 18:09:00 EDT Received: by nando.net.nando.net (4.1/SMI-4.1) id AB07391; Tue, 3 Oct 95 18:09:07 EDT From: Chip Salzenberg Message-Id: <9510032209.AB07391@nando.net.nando.net> Subject: Re: Fetching null columns To: mhm@austin.ibm.com Date: Tue, 3 Oct 1995 18:09:02 -0400 (EDT) Cc: aks@hub.ucsb.edu, perldb-interest@fugue.com In-Reply-To: <9510032001.AA25860@dax.austin.ibm.com> from "mhm@austin.ibm.com" at Oct 3, 95 03:01:02 pm X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 623 According to mhm@austin.ibm.com: > Alan Stebbens writes: > > Each column (array element) can then be "undef" (for null), or a defined > > value, including the empty string. A special "isnull" function is > > unnecessary. > > If the first item in @row is undef then the loop is terminated. If that's true in 5.001m, then there's a bug in 5.001m. It's still a good interface to use undef "NULL column" and () for "no more rows". -- Chip Salzenberg, aka "Hey, it's the Miss Alternate Universe Pageant!" -- Crow T. Robot, MST3K: "Stranded In Space" ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06080-40@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:16:39 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 02:10:07 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ac04946; 4 Oct 95 3:09 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <12948-0@frigate.doc.ic.ac.uk>; Wed, 4 Oct 1995 02:14:42 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA24855 for perldb-interest-real; Tue, 3 Oct 1995 16:40:35 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from eve.mscs.mu.edu (eve.mscs.mu.edu [134.48.4.166]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA24851 for ; Tue, 3 Oct 1995 16:40:34 -0700 Received: by eve.mscs.mu.edu (Smail3.1.28.1 #9) id m0t0GvJ-0002DjC; Tue, 3 Oct 95 18:38 CDT Message-Id: From: Saima Zobair Subject: subscription To: perldb-interest@fugue.com Date: Tue, 3 Oct 1995 18:38:56 -0500 (CDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 261 Hello, I am a little confused about the 2 mailing lists. I want to be on both the mailing lists for perldb-interest... I am not sure which one I am on right now. Please put me on both of them if they are not the same. Thanks, Saima Zobair zobair@mscs.mu.edu ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06080-46@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:17:46 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 02:15:37 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa06994; 4 Oct 95 3:14 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA25125 for perldb-interest-real; Tue, 3 Oct 1995 17:48:49 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA25121 for ; Tue, 3 Oct 1995 17:48:40 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id bk26504; 3 Oct 95 12:46 +0100 Received: from ig.co.uk by lion id <26793-0@lion>; Tue, 3 Oct 1995 03:44:22 +0100 To: chet@watson.ibm.com Subject: Re: FYI -- progress on problem Cc: mhm@austin.ibm.com, perldb-interest@fugue.com Date: Tue, 3 Oct 1995 03:44:22 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510031246.bk26504@relay-3.mail.demon.net> > From: Chet Murthy > > Hi, Tim, Mike, > > I finally have a relatively precise profile of the problem I'm seeing. > > It is that the parent handle of a database implementation > datastructure is being freed before the kid. In particular, in this > case, it is the root handle (the driver handle, I think?) which is > being freed. > Remember that the dbihcom structure will hang around until all the references to it (in its kids) are freed. At least that's what's supposed to happen. Things get a little wild during global destruction! > I will investigate further, and I think that I can fix this, but I > figured I'd let you know. Also, there's a one-line typo in: > > if (parent) { > imp_xxh_t *parent_com = DBIh_COM(parent); > DBIc_PARENT_H(imp) = SvREFCNT_inc(parent); /* ensure it lives */ > DBIc_PARENT_COM(imp) = parent_com; /* shortcut for speed */ > DBIc_TYPE(imp) = DBIc_TYPE(parent_com) + 1; /* XXX */ > DBIc_FLAGS(imp) = DBIc_FLAGS(parent_com) & DBIcf_INHERITMASK; > ++DBIc_KIDS(parent_com); > > } else { /* only a driver (drh) has no parent */ > DBIc_PARENT_H(imp) = &sv_undef; > DBIc_PARENT_COM(imp) = NULL; > ----------------^^^---------------------------------- > > This was _DBIc_PARENT_H in the sources that I FTP'd. > > DBIc_TYPE(imp) = DBIt_DR; > DBIc_FLAGS(imp) = 0; > DBIc_WARN_on(imp); /* only set here, childern inherit */ > } > > > I assume that this shoudl DBIc_PARENT_COM, as I have written it above. Thanks. Fixed in my copy now. The best way to show code changes is a context diff. Failing that something like this works well: DBIc_PARENT_H(imp) = &sv_undef; - DBIc_PARENT_H(imp) = NULL; + DBIc_PARENT_COM(imp) = NULL; DBIc_TYPE(imp) = DBIt_DR; It's often important that I can read and understand these things without any ambiguity. > But that didn't fix my problem (unfortunately). > Off-hand I can't think of any nasty effects of that one. > From: Chet Murthy > > It seems that there is a problem in the dbih_clearcom function. > Here's the scenario: if we're doing global destruction, and the > handles get destroyed in the wrong order, then during that very same > destruction the memory for those handles could get re-used. So if: > > (1) handle H2 points to handle H1 > > (2) H1 is freed *first* > > (3) H1's memory is re-used > > (4) H2 is then freed, and decrements the "kid" count of H1 > > (5) This corrupts the memory of H1, which is being used for something > completely different. > You're not making it clear if H1 etc is a handle, a dbihcom struct or both. > There is definitely one problem, > > if (DBIc_PARENT_COM(imp_xxh) && !gd) { > --------------------------------^^^^^^^---------------HERE > --DBIc_KIDS(DBIc_PARENT_COM(imp_xxh)); > } I can't tell without looking at the code what it is you're saying... > Does this sound right? And if so, is the fix I propose the right one? > And if not (or even if yes) are there other places where I might look > for such things? [rummage] Ah, you're suggesting adding the "&& !gd" test. Umm, okay. Done. Note that drivers which test DBIc_KIDS should not bother if dirty is true (DBD::Oracle checks the dirty flag already). > Thanks, > --chet-- Thanks Chet - welcome to the team! :-) Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06080-51@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:18:42 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 03:53:35 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa18689; 4 Oct 95 4:51 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA25508 for perldb-interest-real; Tue, 3 Oct 1995 19:19:04 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id TAA25504 for ; Tue, 3 Oct 1995 19:19:01 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id bc26504; 3 Oct 95 12:45 +0100 Received: from ig.co.uk by lion id <26440-0@lion>; Tue, 3 Oct 1995 02:40:23 +0100 To: perldb-interest@fugue.com, gade@studsys.mscs.mu.edu Subject: Re: Installation of DBI!! HELP!! Date: Tue, 3 Oct 1995 02:40:23 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510031245.bc26504@relay-3.mail.demon.net> > From: Malathi Gade > > Hello, > > I am trying to install DBI. I am getting an error message which > seems like I am missing a config.sh file in perl directory... > > Can you please help me with this. > > thanks, > Malathi Gade > gade@mscs.mu.edu > > here it goes: > > Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. > DO NOT build the DBI under the Perl source tree. > Use 'make test' to execute self tests. > MakeMaker > NAME = 'DBI' > VERSION = '0.63' > > Use of uninitialized value at /usr/local/lib/perl5/ExtUtils/MakeMaker.pm line > 198. > Can't find config.sh at /usr/local/lib/perl5/ExtUtils/MakeMaker.pm line 198. > Looks like a perl installation problem. Make sure you have Perl5.001*m* built, tested and installed cleanly. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06080-59@ignite.demon.co.uk>; Wed, 4 Oct 1995 07:20:06 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 04:39:23 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa09067; 4 Oct 95 5:35 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA25661 for perldb-interest-real; Tue, 3 Oct 1995 20:06:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from hub.ucsb.edu (hub.ucsb.edu [128.111.24.40]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA25657 for ; Tue, 3 Oct 1995 20:06:00 -0700 Received: from dokoka (dokoka.ucsb.edu) by hub.ucsb.edu; id AA03201 sendmail 4.1/UCSB-2.1-sun Tue, 3 Oct 95 20:08:11 PDT for perldb-interest@fugue.com Received: from dokoka.ucsb.edu by dokoka via ESMTP (940816.SGI.8.6.9/940406.SGIv2) id UAA12642; Tue, 3 Oct 1995 20:08:09 -0700 Message-Id: <199510040308.UAA12642@dokoka> To: Chip Salzenberg Cc: mhm@austin.ibm.com, perldb-interest@fugue.com Subject: Re: Fetching null columns Reply-To: "Alan K. Stebbens" In-Reply-To: Your message of Tue, 03 Oct 1995 18:09:02 PDT. <9510032209.AB07391@nando.net.nando.net> Date: Tue, 03 Oct 1995 20:08:03 -0700 From: Alan Stebbens > According to mhm@austin.ibm.com: > > Alan Stebbens writes: > > > Each column (array element) can then be "undef" (for null), or a defined > > > value, including the empty string. A special "isnull" function is > > > unnecessary. > > > > If the first item in @row is undef then the loop is terminated. > > If that's true in 5.001m, then there's a bug in 5.001m. It's still a > good interface to use undef "NULL column" and () for "no more rows". This is not true in 5.001m. It is safe to use the construct: while (defined(@row = dbfetchrow $dbh)) { ... } Here's a piece of Perl 5.001m: > perl -de 0 Loading DB routines from $RCSfile: perl5db.pl,v $$Revision: 4.1 $$Date: 92/08/07 18:24:07 $ Emacs support available. Enter h for help. main::(-e:1): 0 DB<1> $x[0] = undef; DB<2> $x[1] = 'foo'; DB<3> p join(',',@x); ,foo DB<4> p join(',',map(defined,@x)) ,1 --> DB<5> if(defined(@x)){print "ok\n";}else{print"undef\n";} ok DB<6> p $] 5.001 DB<7> ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <08661-0@ignite.demon.co.uk>; Wed, 4 Oct 1995 11:21:50 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 07:45:18 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id ab02921; 4 Oct 95 8:44 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa27491; 4 Oct 95 8:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA26118 for perldb-interest-real; Tue, 3 Oct 1995 23:32:29 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id XAA26112 for ; Tue, 3 Oct 1995 23:32:26 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id az26504; 3 Oct 95 12:45 +0100 Received: from ig.co.uk by lion id <25722-0@lion>; Tue, 3 Oct 1995 02:26:27 +0100 To: mhm@austin.ibm.com, chet@watson.ibm.com Subject: Re: DB2perl Cc: perldb-interest@fugue.com Date: Tue, 3 Oct 1995 02:26:27 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510031245.az26504@relay-3.mail.demon.net> > From: mhm@austin.ibm.com > > Tim, > What sort of optimizations were thinking about for dbd_st_fetch? > DBD::Oracle, as well as DBD::DB2 both carry your comment about > the ability to optimize later. > Mainly using native types where possible. For Oracle we currently use strings for everything which is not optimal. > One of my alpha testers is asking if the following change is a valid > fix for null variables returned from the database. Given the availability > of null indicators I see no reason to make the change unless it can > be worked into the optimizations you have been thinking about. > From: mhm@austin.ibm.com > Subject: Dealing with null indicator variables > > How does one deal with null indicator variables specified by the enduser? > > >From what I can tell it doesn't appear to be done at the present time. > Am I correct or just missing the appropriate piece of magic? To be a DBI conforming driver for a database which supports NULLS you must return NULLS as undefs and accept undefs as meaning NULL for bind variables. Here's the DBD::Oracle code: In dbd_bind_ph: if (SvOK(newvalue)) { ... } else { value_ptr = ""; value_len = 0; phs->indp = -1; } In dbd_st_fetch: } else if (rc == 1405) { /* field is null - return undef */ (void)SvOK_off(sv); Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <08660-5@ignite.demon.co.uk>; Wed, 4 Oct 1995 11:23:55 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 07:44:36 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa26487; 4 Oct 95 8:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA26095 for perldb-interest-real; Tue, 3 Oct 1995 23:32:14 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA26091 for ; Tue, 3 Oct 1995 23:32:12 -0700 Received: by gw.home.vix.com id AA25382; Tue, 3 Oct 95 23:34:21 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id au26504; 3 Oct 95 12:45 +0100 Received: from ig.co.uk by lion id <25524-0@lion>; Tue, 3 Oct 1995 01:22:24 +0100 To: perldb-interest@vix.com, LACAILLD@wpgate.hqpacaf.af.mil Subject: Re: WOW Errors Date: Tue, 3 Oct 1995 01:22:24 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510031245.au26504@relay-3.mail.demon.net> > From: Damon LaCaille > > Sorry to bug everyone again, but I think I've pretty much narrowed my > problem down to a few things, and hopefully somebody out on this > mailing list can help me out here... > > I'm finally calling the procedure properly, I'm getting the following error > though: > > dbLogon(): orlon() returns 1034d > > In my last post I stated what the Oracle errors book says it could be. > We've relinked oracle, it was initialized, and the ORACLE_SID > environment variable is pointing to the correct instance of Oracle (TRNG). > So I don't know what's wrong here... maybe somebody can help? I've got no experience with WOW and this list is not really the place for supporting it. Sorry. Try comp.database.oracle and comp.infosystems.www.authoring.cgi. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <08660-7@ignite.demon.co.uk>; Wed, 4 Oct 1995 11:24:17 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 07:45:42 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa26962; 4 Oct 95 8:44 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA26114 for perldb-interest-real; Tue, 3 Oct 1995 23:32:27 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA26109 for ; Tue, 3 Oct 1995 23:32:25 -0700 Received: by gw.home.vix.com id AA25399; Tue, 3 Oct 95 23:34:37 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ay26504; 3 Oct 95 12:45 +0100 Received: from ig.co.uk by lion id <25680-0@lion>; Tue, 3 Oct 1995 02:14:00 +0100 To: perldb-interest@vix.com, spoliset@mailbox.syr.edu Subject: Re: oraperl installation errors... please help... Date: Tue, 3 Oct 1995 02:14:00 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510031245.ay26504@relay-3.mail.demon.net> > From: Srinivas Polisetty > > Hi All, > Could someone help me with my installation of 'oraperl' for which > I had been struggling for a while now, and in vain. > > My machine specs: Sun4 Sparc, running SunOS UNIX > > [...] > > Some thing is terribly wrong, and I can't figure out what!! Could anyone > please help me with this. > > Thank you very much, > Srinvas. > This list is not really the place toget support for Perl4 oraperl. Take a look in ftp.demon.co.uk:/pub/perl/db/perl4/oraperl/ and if you still have problems post a message in comp.lang.perl.misc and comp.databases.oracle. Sorry I can't be more help. No time. Home to bed. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <08660-9@ignite.demon.co.uk>; Wed, 4 Oct 1995 11:24:26 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 07:50:32 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa29024; 4 Oct 95 8:49 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA26089 for perldb-interest-real; Tue, 3 Oct 1995 23:31:45 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA26085 for ; Tue, 3 Oct 1995 23:31:44 -0700 Received: by gw.home.vix.com id AA25370; Tue, 3 Oct 95 23:33:56 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ar26504; 3 Oct 95 12:44 +0100 Received: from ig.co.uk by lion id <25439-0@lion>; Tue, 3 Oct 1995 00:56:38 +0100 To: perldb-interest@vix.com, ks@ic.uva.nl Subject: Re: Happy birthday DBperl! Date: Tue, 3 Oct 1995 00:56:38 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510031244.ar26504@relay-3.mail.demon.net> Thanks Karel. And thanks to everyone who has put up with my alternate ranting and silences over the last three years. I'm busy again at the moment but maybe not for too much longer. Tim. ----- Begin Included Message ----- From: "ks@ic.uva.nl" As Tim Bunce wrote on Thu, 17 Aug 1995 22:39:06 +0100: > > For those too young to remember, the DBperl effort will have its > third birthday on September 29th. It would be kind'a nice to have > something to show for it after three years :-) > ^ ^ ^ | | | -------- ========== \----------/ It's a small cake for me but a big one for mankind Tim! Regards, Karel Sprenger ----- End Included Message ----- ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <12965-4@ignite.demon.co.uk>; Wed, 4 Oct 1995 16:01:41 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 13:51:31 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa06696; 4 Oct 95 14:50 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA28243 for perldb-interest-real; Wed, 4 Oct 1995 04:56:58 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id EAA28239 for ; Wed, 4 Oct 1995 04:56:55 -0700 Received: from post.demon.co.uk by relay-1.mail.demon.net id ad12625; 4 Oct 95 12:18 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <10170-0@ignite.demon.co.uk>; Wed, 4 Oct 1995 11:59:13 +0100 To: perldb-interest@fugue.com, zobair@marque.mscs.mu.edu Subject: Re: subscription Date: Wed, 4 Oct 1995 11:59:13 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510041218.ad12625@relay-1.mail.demon.net> > From: Saima Zobair > > Hello, > > I am a little confused about the 2 mailing lists. I want to > be on both the mailing lists for perldb-interest... > I am not sure which one I am on right now. Please put > me on both of them if they are not the same. > There is only one list. The perldb-interest-request address is for administration messages such as subscription requests. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <14653-0@ignite.demon.co.uk>; Wed, 4 Oct 1995 17:53:48 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 16:50:08 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa24326; 4 Oct 95 17:49 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA28443 for perldb-interest-real; Wed, 4 Oct 1995 05:38:33 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA28439 for ; Wed, 4 Oct 1995 05:38:29 -0700 Received: by gw.home.vix.com id AA22744; Wed, 4 Oct 95 05:40:07 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from skimo@localhost) by dns.ufsia.ac.be (8.6.12/8.6.9) id OAA11350; Wed, 4 Oct 1995 14:41:55 +0100 From: Sven Verdoolaege Message-Id: <199510041341.OAA11350@dns.ufsia.ac.be> Subject: Re: Multiple DBD's To: Tim Bunce Date: Wed, 4 Oct 1995 14:41:55 +0100 (MET) Cc: perldb-interest@vix.com In-Reply-To: <9510031246.be26504@relay-3.mail.demon.net> from "Tim Bunce" at Oct 3, 95 02:47:56 am X-Stardate: [-31]6387.85 X-Url: http://dns.ufsia.ac.be/~skimo X-Mailer: ELM [version 2.4 PL24 ME8a] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 10748 For those on the list: i have problems connecting to both an mSQL and an Oracle database here's the trace from PERL_DBI_DEBUG=3 IN DBI.pm DBI dispatch debug level set to 3 install_method DBI::db::EXISTS install_method DBI::db::event usage: min 2, max 3, '$message, $retvalue', flags 0x1 install_method DBI::db::errstr install_method DBI::db::debug usage: min 1, max 2, '[$debug_level]', flags 0x1 install_method DBI::db::CLEAR install_method DBI::db::NEXTKEY install_method DBI::db::commit usage: min 1, max 1, '', flags 0x1 install_method DBI::db::DESTROY install_method DBI::db::FETCH install_method DBI::db::rollback usage: min 1, max 1, '', flags 0x1 install_method DBI::db::errstate usage: min 1, max 1, '', flags 0x1 install_method DBI::db::disconnect usage: min 1, max 1, '', flags 0x1 install_method DBI::db::rows install_method DBI::db::func , flags 0x2 install_method DBI::db::STORE install_method DBI::db::prepare usage: min 2, max 3, '$statement [, \%attribs]', flags 0x1 install_method DBI::db::errmsg usage: min 1, max 1, '', flags 0x1 install_method DBI::db::FIRSTKEY install_method DBI::db::do usage: min 2, max 3, '$statement [, \%attribs]', flags 0x1 install_method DBI::db::private_data usage: min 1, max 1, '', flags 0x1 install_method DBI::db::handler usage: min 2, max 2, '\&handler', flags 0x1 install_method DBI::dr::EXISTS install_method DBI::dr::disconnect_all usage: min 1, max 1, '', flags 0x1 install_method DBI::dr::event usage: min 2, max 3, '$message, $retvalue', flags 0x1 install_method DBI::dr::errstr install_method DBI::dr::debug usage: min 1, max 2, '[$debug_level]', flags 0x1 install_method DBI::dr::CLEAR install_method DBI::dr::NEXTKEY install_method DBI::dr::connect usage: min 1, max 5, '[$db [,$user [,$passwd [,\%attr]]]]', flags 0x1 install_method DBI::dr::FETCH install_method DBI::dr::DESTROY install_method DBI::dr::rows install_method DBI::dr::func , flags 0x2 install_method DBI::dr::STORE install_method DBI::dr::FIRSTKEY install_method DBI::dr::private_data usage: min 1, max 1, '', flags 0x1 install_method DBI::st::EXISTS install_method DBI::st::finish usage: min 1, max 1, '', flags 0x1 install_method DBI::st::fetchrow install_method DBI::st::event usage: min 2, max 3, '$message, $retvalue', flags 0x1 install_method DBI::st::errstr install_method DBI::st::debug usage: min 1, max 2, '[$debug_level]', flags 0x1 install_method DBI::st::CLEAR install_method DBI::st::NEXTKEY install_method DBI::st::fetch install_method DBI::st::FETCH install_method DBI::st::DESTROY install_method DBI::st::rows install_method DBI::st::func , flags 0x2 install_method DBI::st::bind_param usage: min 3, max 4, '$parameter, $var [, \%attribs]', flags 0x1 install_method DBI::st::STORE install_method DBI::st::FIRSTKEY install_method DBI::st::execute usage: min 1, max 0, '[@args]', flags 0x1 install_method DBI::st::private_data usage: min 1, max 1, '', flags 0x1 install_method DBI::st::bind_col usage: min 3, max 4, '$column, \$var [, \%attribs]', flags 0x1 install_method DBI::st::readblob usage: min 4, max 5, '$field, $offset, $len [, \$buf [, $bufoffset]]', flags 0x1 install_method DBI::st::bind_columns usage: min 3, max 0, '\%attribs, \$var1 [, \$var2, ...]', flags 0x1 install_driver(Switch): setup @ISA for DBD::Switch::dr at /u01/home/verdools/lib/perl5/DBI.pm line 346 install_driver(Switch): setup @ISA for DBD::Switch::db at /u01/home/verdools/lib/perl5/DBI.pm line 346 install_driver(Switch): setup @ISA for DBD::Switch::st at /u01/home/verdools/lib/perl5/DBI.pm line 346 DBI->connect(dns, ufpers, , mSQL, ) DBI->install_driver DBI mSQL at ./cp2 line 6 DBI->install_driver(mSQL) loaded install_driver(mSQL): setup @ISA for DBD::mSQL::dr at ./cp2 line 6 install_driver(mSQL): setup @ISA for DBD::mSQL::db at ./cp2 line 6 install_driver(mSQL): setup @ISA for DBD::mSQL::st at ./cp2 line 6 New DBI::dr (for DBD::mSQL::dr, parent=, id=) dbih_setup_handle(DBI::dr=HASH(0x140085538)=>DBI::dr=HASH(0x1400eb098), DBD::mSQL::dr, NULL) New DBI::dr => DBI::dr=HASH(0x140085538) (inner=DBI::dr=HASH(0x1400eb098)) for DBD::mSQL::dr DBI->install_driver(mSQL) = DBI::dr=HASH(0x140085538) DBI->connect using mSQL driver DBI::dr=HASH(0x140085538) >> connect DISPATCH (DBI::dr=HASH(0x140085538) @4 g0 a1400e8828 r1) -> connect for DBD::mSQL::dr (DBI::dr=HASH(0x140085538)~0x1400eb098 'dns' 'ufpers' '') New DBI::db (for DBD::mSQL::db, parent=DBI::dr=HASH(0x1400eb098), id=) dbih_setup_handle(DBI::db=HASH(0x1400e0718)=>DBI::db=HASH(0x1400e0228), DBD::mSQL::db, NULL) New DBI::db => DBI::db=HASH(0x1400e0718) (inner=DBI::db=HASH(0x1400e0228)) for DBD::mSQL::db <- connect= DBI::db=HASH(0x1400e0718) DBI->connect = DBI::db=HASH(0x1400e0718) >> disconnect DISPATCH (DBI::db=HASH(0x1400e0718) @1 g0 a1400e1588 r1) -> disconnect for DBD::mSQL::db (DBI::db=HASH(0x1400e0718)~0x1400e0228) <- disconnect= 1 DBI->connect(ufpers, job_verdool_s, h7mq24ll, Oracle, ) DBI->install_driver DBI Oracle at ./cp2 line 10 DBI->install_driver(Oracle) loaded install_driver(Oracle): setup @ISA for DBD::Oracle::dr at ./cp2 line 10 install_driver(Oracle): setup @ISA for DBD::Oracle::db at ./cp2 line 10 install_driver(Oracle): setup @ISA for DBD::Oracle::st at ./cp2 line 10 New DBI::dr (for DBD::Oracle::dr, parent=, id=) dbih_setup_handle(DBI::dr=HASH(0x1400e0728)=>DBI::dr=HASH(0x1400eb468), DBD::Oracle::dr, NULL) New DBI::dr => DBI::dr=HASH(0x1400e0728) (inner=DBI::dr=HASH(0x1400eb468)) for DBD::Oracle::dr DBI->install_driver(Oracle) = DBI::dr=HASH(0x1400e0728) DBI->connect using Oracle driver DBI::dr=HASH(0x1400e0728) >> connect DISPATCH (DBI::dr=HASH(0x1400e0728) @4 g0 a1400e8828 r1) -> connect for DBD::Oracle::dr (DBI::dr=HASH(0x1400e0728)~0x1400eb468 'ufpers' 'job_verdool_s' 'h7mq24ll') New DBI::db (for DBD::Oracle::db, parent=DBI::dr=HASH(0x1400eb468), id=) dbih_setup_handle(DBI::db=HASH(0x140100fa8)=>DBI::db=HASH(0x1400eb408), DBD::Oracle::db, NULL) New DBI::db => DBI::db=HASH(0x140100fa8) (inner=DBI::db=HASH(0x1400eb408)) for DBD::Oracle::db >> DESTROY DISPATCH (DBI::db=HASH(0x140100fa8) @1 g0 a0 r2) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::db=HASH(0x1400eb408) @1 g0 a0 r2) -> DESTROY for DBD::Oracle::db (DBI::db=HASH(0x1400eb408)) Database handle DBI::db=HASH(0x1400eb408) DESTROY ignored - never set up at /u01/home/verdools/lib/perl5/DBI.pm line 140. <- DESTROY= 1 dbih_clearcom 0x140100f48 (com 0x140025c08) FLAGS 11: COMSET WARN IMP_DATA 'DBI::db=HASH(0x1400eb408)' PARENT DBI::dr=HASH(0x1400eb468) dbih_clearcom 0x140025c08 done <- connect= undef DBI->connect = geen database at ./cp2 line 10. DBI::END DBI::disconnect_all DBI DBI::disconnect_all for 'mSQL' >> disconnect_all DISPATCH (DBI::dr=HASH(0x140085538) @1 g0 a1400e1248 r2) -> disconnect_all for DBD::mSQL::dr (DBI::dr=HASH(0x140085538)~0x1400eb098) <- disconnect_all= 1 DBI::disconnect_all for 'Oracle' >> disconnect_all DISPATCH (DBI::dr=HASH(0x1400e0728) @1 g0 a1400e1248 r2) -> disconnect_all for DBD::Oracle::dr (DBI::dr=HASH(0x1400e0728)~0x1400eb468) <- disconnect_all= 1 DBI::END complete >> DESTROY DISPATCH (DBI::dr=HASH(0x1400eb468) @1 g0 a0 r0) dbih_clearcom 0x140100f88 (com 0x140107188) FLAGS 11: COMSET WARN IMP_DATA 'DBI::dr=HASH(0x1400eb468)' PARENT NULL dbih_clearcom 0x140107188 done >> DESTROY DISPATCH (DBI::dr=HASH(0x1400e0728) @1 g0 a0 r0) (inner handle already deleted) >> DESTROY DISPATCH (DBI::db=HASH(0x1400e0228) @1 g0 a0 r0) -> DESTROY for DBD::mSQL::db (DBI::db=HASH(0x1400e0228)) <- DESTROY= 1 dbih_clearcom 0x1400e0218 (com 0x140025808) FLAGS 11: COMSET WARN IMP_DATA 'DBI::db=HASH(0x1400e0228)' PARENT DBI::dr=HASH(0x1400eb098) >> DESTROY DISPATCH (DBI::dr=HASH(0x1400eb098) @1 g0 a0 r0) dbih_clearcom 0x1400e06f8 (com 0x1400f4688) FLAGS 11: COMSET WARN IMP_DATA 'DBI::dr=HASH(0x1400eb098)' PARENT NULL dbih_clearcom 0x1400f4688 done dbih_clearcom 0x140025808 done >> DESTROY DISPATCH (DBI::dr=HASH(0x140085538) @1 g0 a0 r0) (inner handle already deleted) >> DESTROY DISPATCH (DBI::db=HASH(0x1400e0718) @1 g0 a0 r0) (inner handle already deleted) Attempt to free unreferenced scalar during global destruction. Bad free() ignored during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. here's my config again: chulu.ufsia.ac.be> perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=dec_osf, osver=3.2, archname=alpha-dec_osf uname='osf1 chulu.ufsia.ac.be v3.2 214 alpha ' hint=previous Compiler: cc='cc', optimize='-O2 -Olimit 2900' cppflags='-DSTANDARD_C' ccflags ='-DSTANDARD_C' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/usr/shlib /lib /usr/lib /usr/ccs/lib libs=-ldbm -lm -lc -lbsd -lPW libc=/usr/shlib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags=' ', ccdlflags=' ', lddlflags='-shared -expect_unresolved "*" -s' > > my script: > > #!/u01/home/verdools/bin/perl > > use DBI; > > $DB='ufpers'; > > $ENV{'MSQL_HOME'} = "/usr/local/Minerva"; > > $ENV{'MSQL_TCP_PORT'} = 22532; > > $dbhD = DBI->connect( 'dns',$DB,'','mSQL') || die("geen database"); > > $dbhD->disconnect; > > $ENV{'ORACLE_HOME'} = "/u01/home/oracle/product/7.1.4"; > > $ENV{'ORAHOME'} = "/u01/home/oracle/product/7.1.4"; > > print "before\n"; > > $dbhS = DBI->connect( $DB, 'job_verdool_s', 'my password:-)','Oracle' ) || die("geen ; > > print "after\n"; > > $dbhS->disconnect; > > > > my output: > > before > > rc: 0 what: Unknown MSQL Server Host (ufpers) > > Database handle DBI::db=HASH(0x1400fff18) DESTROY ignored - never set up at /u0. > > geen database at ./cp2 line 11. > > > > if i comment the connect and disconnect lines from mSQL, i get > > before > > after > > > > if i put the oracle stuff before the mSQL stuff, i get > > before > > after > > Memory fault - core dumped > > ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <17498-4@ignite.demon.co.uk>; Thu, 5 Oct 1995 07:08:34 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 04 Oct 95 18:01:35 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa28435; 4 Oct 95 19:01 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA29199 for perldb-interest-real; Wed, 4 Oct 1995 08:14:37 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA29195 for ; Wed, 4 Oct 1995 08:14:36 -0700 Received: by gw.home.vix.com id AA03449; Wed, 4 Oct 95 08:16:48 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-1.mail.demon.net id ac12625; 4 Oct 95 12:18 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <10159-0@ignite.demon.co.uk>; Wed, 4 Oct 1995 11:56:55 +0100 To: perldb-interest@vix.com, lm_truong@ccmail.pnl.gov Subject: Re: oraperl &ora_open and &ora_close problem Date: Wed, 4 Oct 1995 11:56:55 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510041218.ac12625@relay-1.mail.demon.net> > From: lm_truong@ccmail.pnl.gov > > I have a problem with oraperl &ora_open and &ora_close. Whenever I > execute a sql statement using ora_open, I have to close it. This > degrades my program's performance. Is there a way to tell ORACLE to > reuse the cursor in oraperl? > Yes, in a limited way. You can re-bind parameters to repeat a query or insert etc using the same cursor. Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <17500-16@ignite.demon.co.uk>; Thu, 5 Oct 1995 07:11:21 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 00:55:15 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ae02572; 5 Oct 95 1:54 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <23124-0@frigate.doc.ic.ac.uk>; Wed, 4 Oct 1995 22:46:23 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA29993 for perldb-interest-real; Wed, 4 Oct 1995 12:01:32 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from proxy.netlabs.com (ns.sems.com [192.94.48.47]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA29989 for ; Wed, 4 Oct 1995 12:01:30 -0700 Received: by proxy.netlabs.com (4.1/SMI-4.1) id AA21412; Wed, 4 Oct 95 12:02:49 PDT Received: from vaccine-bb.sems.com(192.190.199.1) by proxy.netlabs.com via smap (V1.3) id sma021409; Wed Oct 4 12:02:23 1995 Received: from scalpel.netlabs.com by netlabs.com (4.1/SMI-4.1) id AA21532; Wed, 4 Oct 95 11:59:25 PDT Received: from localhost.netlabs.com by scalpel.netlabs.com (4.1/SMI-4.1) id AA26792; Wed, 4 Oct 95 11:58:24 PDT Message-Id: <9510041858.AA26792@scalpel.netlabs.com> To: Tim Bunce Cc: perldb-interest@fugue.com, mhm@austin.ibm.com, perlbug@perl.com, Mark Jefferys Subject: Re: perl bug: array of undefs is falsely false (was: Fetching null columns) In-Reply-To: Your message of "Wed, 04 Oct 95 12:29:37 BST." <9510041229.aa14202@relay-1.mail.demon.net> Date: Wed, 04 Oct 95 11:58:23 -0700 From: Larry Wall I've added Mark Jefferys to this discussion, since he was involved in the change to list assignment, I believe. Larry : > From: mhm@austin.ibm.com : > : > Ok, question time. How do we expect the DBI::DBD interfaces to work : > when it comes to columns returning null? : > : Instances of NULLS are represented on input and output as undefs. Period. : : : > $sth->execute(); : > while (@row = $sth->fetchrow) { : > ... : > } : > : > If the $stmt requests only one column from the table : > and the column can contain NULLs then it is possible to return only : > part of the data. : : This is a perl bug discovered recently by Michael Peppler working on Sybase. : I CC'd a reply to perlbug@perl.com but I think it bounced. I'll CC this. : : > If $stmt requests more than one column and the first : > column requested can contain NULLs, we break again. : > : Not quite. It breaks if *all* the columns are nulls. Here's a test case: : : sub undef_a { (undef, undef, 1, undef) } : sub undef_b { (undef, undef, undef, undef) } : sub undef_c { (undef) } : : (@ary = undef_a()) ? print "ok\n" : print "not ok\n"; : (@ary = undef_b()) ? print "ok\n" : print "not ok\n"; : (@ary = undef_c()) ? print "ok\n" : print "not ok\n"; : : It prints: : : ok : not ok : not ok : : This *is* a serious problem for database work! : : > I guess, my point is that we need some method of returning something : > other than a Nullsv or sv_undef and giving the enduser a method of : > checking the null indicator from the fbh struct. : > : Nope. Perl needs fixing! This really needs to be fixed before 5.002. : (Note: _never_ return Nullsv on the stack, always use &sv_undef.) : : In the meantime you can use the alternative (and faster) $sth->fetch : method upon which fetchrow is implemented: : : while($aryref = $sth->fetch) { : .... code using @$aryref or $aryref->[n] ... : } : : but this won't help people who need to stick to the Oraperl emulation interface. : : : Tim. : ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <17498-22@ignite.demon.co.uk>; Thu, 5 Oct 1995 07:13:26 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 01:57:57 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa19334; 5 Oct 95 2:57 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA01008 for perldb-interest-real; Wed, 4 Oct 1995 17:35:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA01004 for ; Wed, 4 Oct 1995 17:35:23 -0700 Received: by gw.home.vix.com id AA15366; Wed, 4 Oct 95 17:37:37 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id TAA16211; Wed, 4 Oct 1995 19:35:23 -0500 Date: Wed, 4 Oct 1995 19:35:23 -0500 (CDT) From: Ben Lindstrom To: DBI perl Subject: I need a little help here.. Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I've been very quiet for too long. =) I'm trying to understand the different between dbd_db_* commands and dbd_st_* commands in the dbdimp.c file. Could someone enlight me on mainly the differents between the FETCH and STORE in both areas? I'm sorta lost. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <17500-27@ignite.demon.co.uk>; Thu, 5 Oct 1995 07:15:09 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 02:39:38 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa09078; 5 Oct 95 3:38 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <24252-0@frigate.doc.ic.ac.uk>; Thu, 5 Oct 1995 00:00:07 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA00443 for perldb-interest-real; Wed, 4 Oct 1995 13:48:05 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA00439 for ; Wed, 4 Oct 1995 13:48:02 -0700 From: rantapaa@math.umn.edu Received: by gw.home.vix.com id AA27334; Wed, 4 Oct 95 13:50:14 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from s6.math.umn.edu by s5.math.umn.edu; Wed, 4 Oct 1995 15:50:06 -0500 Received: by s6.math.umn.edu (8.6.12) id UAA24055; Wed, 4 Oct 1995 20:49:51 GMT Date: Wed, 4 Oct 1995 20:49:51 GMT Message-Id: <199510042049.UAA24055@s6.math.umn.edu> To: perldb-interest@vix.com Subject: Can I get meta-information without having to create a oraperl handle? A few revisions ago I was able to do the following: $drh = DBI->install_driver('Oracle'); $dbh = $drh->connect($Database, $User, $Auth); $c = $dbh->prepare("select * from $Table"); die "Prepare failed ($DBI::err): $DBI::errstr\n" unless ($c); @name = @{$c->{NAME}}; # @nullable = @{$c->{NULLABLE}}; @type = @{$c->{ora_types}}; @length = @{$c->{ora_lengths}}; With the most recent version of the Oracle DBD, this no longer works. Examining the code reveals that "ora_types" and "ora_lengths" are only available if the handle is "oraperl compatible" (tm) ;-). I'd like to get at this meta-information. Is there some way to do this or is there a way being planned to get at it? Thanks, Erik Rantapaa rantapaa@math.umn.edu ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <21666-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 12:38:54 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 07:23:10 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa20198; 5 Oct 95 8:22 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id ab02677; 4 Oct 95 14:35 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA28170 for perldb-interest-real; Wed, 4 Oct 1995 04:33:24 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA28166 for ; Wed, 4 Oct 1995 04:33:22 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id GAA65548; Wed, 4 Oct 1995 06:35:36 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA25130; Wed, 4 Oct 1995 06:35:31 -0500 Message-Id: <9510041135.AA25130@dax.austin.ibm.com> To: Chip Salzenberg Cc: perldb-interest@fugue.com Subject: Re: Fetching null columns In-Reply-To: Your message of Tue, 03 Oct 95 18:09:02 -0400. <9510032209.AB07391@nando.net.nando.net> Date: Wed, 04 Oct 95 06:35:31 -0500 chs@nando.net (Chip Salzenberg) writes: |> According to mhm@austin.ibm.com: |> > Alan Stebbens writes: |> > > Each column (array element) can then be "undef" (for null), or a defined |> > > value, including the empty string. A special "isnull" function is |> > > unnecessary. |> > |> > If the first item in @row is undef then the loop is terminated. |> |> If that's true in 5.001m, then there's a bug in 5.001m. It's still a |> good interface to use undef "NULL column" and () for "no more rows". Most definitely true in 5.001m :-( In the mean time, I still need to find away to support NULL columns that doesn't break loops. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <21666-2@ignite.demon.co.uk>; Thu, 5 Oct 1995 12:39:07 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 07:44:12 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa25750; 5 Oct 95 8:43 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa03649; 4 Oct 95 14:40 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA28212 for perldb-interest-real; Wed, 4 Oct 1995 04:49:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id EAA28208 for ; Wed, 4 Oct 1995 04:49:03 -0700 Received: from post.demon.co.uk by relay-1.mail.demon.net id aa14202; 4 Oct 95 12:29 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <10534-0@ignite.demon.co.uk>; Wed, 4 Oct 1995 12:29:37 +0100 To: perldb-interest@fugue.com, mhm@austin.ibm.com Subject: perl bug: array of undefs is falsely false (was: Fetching null columns) Cc: perlbug@perl.com, lwall@sems.com Date: Wed, 4 Oct 1995 12:29:37 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510041229.aa14202@relay-1.mail.demon.net> > From: mhm@austin.ibm.com > > Ok, question time. How do we expect the DBI::DBD interfaces to work > when it comes to columns returning null? > Instances of NULLS are represented on input and output as undefs. Period. > $sth->execute(); > while (@row = $sth->fetchrow) { > ... > } > > If the $stmt requests only one column from the table > and the column can contain NULLs then it is possible to return only > part of the data. This is a perl bug discovered recently by Michael Peppler working on Sybase. I CC'd a reply to perlbug@perl.com but I think it bounced. I'll CC this. > If $stmt requests more than one column and the first > column requested can contain NULLs, we break again. > Not quite. It breaks if *all* the columns are nulls. Here's a test case: sub undef_a { (undef, undef, 1, undef) } sub undef_b { (undef, undef, undef, undef) } sub undef_c { (undef) } (@ary = undef_a()) ? print "ok\n" : print "not ok\n"; (@ary = undef_b()) ? print "ok\n" : print "not ok\n"; (@ary = undef_c()) ? print "ok\n" : print "not ok\n"; It prints: ok not ok not ok This *is* a serious problem for database work! > I guess, my point is that we need some method of returning something > other than a Nullsv or sv_undef and giving the enduser a method of > checking the null indicator from the fbh struct. > Nope. Perl needs fixing! This really needs to be fixed before 5.002. (Note: _never_ return Nullsv on the stack, always use &sv_undef.) In the meantime you can use the alternative (and faster) $sth->fetch method upon which fetchrow is implemented: while($aryref = $sth->fetch) { .... code using @$aryref or $aryref->[n] ... } but this won't help people who need to stick to the Oraperl emulation interface. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <21666-4@ignite.demon.co.uk>; Thu, 5 Oct 1995 12:39:24 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 08:32:10 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa12105; 5 Oct 95 9:32 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id ab06420; 4 Oct 95 14:58 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA28257 for perldb-interest-real; Wed, 4 Oct 1995 04:57:22 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id EAA28253 for ; Wed, 4 Oct 1995 04:57:19 -0700 Received: from post.demon.co.uk by relay-1.mail.demon.net id ae12625; 4 Oct 95 12:18 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <10235-0@ignite.demon.co.uk>; Wed, 4 Oct 1995 12:11:55 +0100 To: zobair@marque.mscs.mu.edu Subject: Re: DBI Installation HELP!! Cc: perldb-interest@fugue.com Date: Wed, 4 Oct 1995 12:11:55 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510041218.ae12625@relay-1.mail.demon.net> > From: Saima Zobair > > Hello, > > I am trying to install DBI on a solaris2.4/oracle7 . > $ perl Makefile.PL (works perfectly) > $ make (works with some warnings) > $ make test - There is an error in test #5 with a core dump. > > Someone please help me with this!! > Any help will be greatly appreciated. > > ----------------perl Makefile.PL output------------------------ > > $ perl Makefile.PL > Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. > DO NOT build the DBI under the Perl source tree. > Use 'make test' to execute self tests. > > MakeMaker > DEFINE => '-Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion' > NAME => 'DBI' > VERSION => '0.63' > Using header files found in /usr/local/lib/perl5.0/lib/sun4-solaris/CORE > Looking for perl 5 by these names: > miniperl perl, > in these dirs: > /usr/openwin/bin /usr/openwin/bin/xview /usr/openwin/demo /usr/X11/bin /usr/openwin/bin /usr/openwin/bin/xview /usr/openwin/demo /usr/X11/bin /usr/openwin/bin /usr/openwin/bin/xview /usr/openwin/demo /usr/X11/bin /usr/local/www//bin /usr/local/bin /usr/openwin/bin /usr/bin /usr/ccs/bin /usr/sbin /usr/local/www/HOTJAVA/bin /serva_usr_local . /usr/sbin /usr/bin /usr/local/bin /usr/local/etc /usr/ccs/bin /usr/lib/lp/postscript /usr/games /usr/openwin/demo /usr/ucb /usr/local/www//bin /usr/etc /etc /usr/local/www/tools/GATEWAY/bin /opt/SUNWits/Graphics-sw/xil/bin /opt/SUNWguide/bin /opt/SUNWsunsol/bin /opt/SUNWdiag/bin /opt/SUNWwabi/bin /opt/SUNWdxlib/bin /opt/SUNWspro/bin /rsch_oracle/home/dba/roracle/product/7.0.16/bin /usr/local/bin > Executing /usr/local/bin/perl > Writing Makefile for DBI Are you using Perl5.001m ? > --------------------------------make output------------------------------- > gcc -c -O -fpic -I/usr/local/lib/perl5.0/lib/sun4-solaris/CORE -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion DBI.c > In file included from /usr/local/lib/gcc-lib/sparc-sun-solaris2.2/2.5.0/include/sys/param.h:37, Your GCC is not only *very* old, it's also an x.y.0 release. I strongly suggest you upgrade. I recommend 2.6.3 as a solid release. > -------------------------------make test output------------------------- > /usr/local/bin/perl -I/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib -I/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib -I/usr/local/lib/perl5.0/lib/sun4-solaris -I/usr/local/lib/perl5.0/lib test.pl > test.pl > DBI test application $Revision: 1.12 $ > make: *** [test] Segmentation Fault (core dumped) To summarise: make sure you're using Perl5.001*m* and upgrade your compiler. Out of idle curiosity I'd also be interested in a stack trace from the core dump (after building with -g instead of -O) and the last part of the output of running make test with PERL_DBI_DEBUG environment variable set to 2. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22307-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 13:25:34 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 11:50:57 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa20440; 5 Oct 95 12:50 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa16539; 4 Oct 95 16:11 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA28548 for perldb-interest-real; Wed, 4 Oct 1995 06:02:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA28544 for ; Wed, 4 Oct 1995 06:02:22 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id IAA89256 for ; Wed, 4 Oct 1995 08:04:36 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA21960; Wed, 4 Oct 1995 08:04:34 -0500 Message-Id: <9510041304.AA21960@dax.austin.ibm.com> To: perldb-interest@fugue.com Cc: chet@watson.ibm.com Subject: Re: DB2perl In-Reply-To: Your message of Tue, 03 Oct 95 02:26:27 +0100. <9510031245.az26504@relay-3.mail.demon.net> Date: Wed, 04 Oct 95 08:04:34 -0500 Tim Bunce writes: |> To be a DBI conforming driver for a database which supports NULLS you |> must return NULLS as undefs and accept undefs as meaning NULL for bind |> variables. Here's the DBD::Oracle code: |> |> In dbd_bind_ph: |> |> if (SvOK(newvalue)) { |> ... |> } else { |> value_ptr = ""; |> value_len = 0; |> phs->indp = -1; |> } Well, this parts is easily covered....I didn't have to change anything :-) |> |> In dbd_st_fetch: |> |> } else if (rc == 1405) { /* field is null - return undef */ |> (void)SvOK_off(sv); This is the hard part as DB2 does not provide a return code to indicate a NULL column. I get a successful rc and an indicator variable set to SQL_NULL_DATA if the column is null. Fortunately, the indicator is carried in the fbh structure.....time to restructure the code. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <24013-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 16:27:00 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 14:54:17 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa12127; 5 Oct 95 15:53 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA03226 for perldb-interest-real; Thu, 5 Oct 1995 04:39:49 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA03222 for ; Thu, 5 Oct 1995 04:39:46 -0700 From: Kenneth.Holmlund@physics.umu.se Received: by gw.home.vix.com id AA00509; Thu, 5 Oct 95 04:42:00 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from minerva.tp.umu.se (minerva [192.36.171.34]) by duessa.tp.umu.se (8.6.9/8.6.9) with ESMTP id MAA20206 for ; Thu, 5 Oct 1995 12:29:15 +0100 Received: (holmlund@localhost) by minerva.tp.umu.se (8.6.9/8.6.9) id MAA20353 for perldb-interest@vix.com; Thu, 5 Oct 1995 12:31:52 +0100 Date: Thu, 5 Oct 1995 12:31:52 +0100 Message-Id: <199510051131.MAA20353@minerva.tp.umu.se> To: perldb-interest@vix.com Subject: subscribe subscribe Kenneth Holmlund Hi, I've tried to contact the listserv to add me to this list but I didn't succeed so I guess I have to bother you all :) How do I subscribe ? Kenneth Holmlund Kenneth.Holmlund@Physics.UmU.SE ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <24013-2@ignite.demon.co.uk>; Thu, 5 Oct 1995 16:27:08 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 15:07:35 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa17570; 5 Oct 95 16:06 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA03379 for perldb-interest-real; Thu, 5 Oct 1995 06:12:12 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA03375 for ; Thu, 5 Oct 1995 06:12:09 -0700 Received: by gw.home.vix.com id AA05939; Thu, 5 Oct 95 06:14:18 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ab12538; 5 Oct 95 12:39 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <20253-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 10:56:32 +0100 To: perldb-interest@vix.com, rantapaa@math.umn.edu Subject: Re: Can I get meta-information without having to create a oraperl handle? Date: Thu, 5 Oct 1995 10:56:32 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510051239.ab12538@relay-3.mail.demon.net> > From: rantapaa@math.umn.edu > > A few revisions ago I was able to do the following: > > $drh = DBI->install_driver('Oracle'); > $dbh = $drh->connect($Database, $User, $Auth); > > $c = $dbh->prepare("select * from $Table"); > die "Prepare failed ($DBI::err): $DBI::errstr\n" unless ($c); > > @name = @{$c->{NAME}}; > # @nullable = @{$c->{NULLABLE}}; > @type = @{$c->{ora_types}}; > @length = @{$c->{ora_lengths}}; > > With the most recent version of the Oracle DBD, this no longer works. > Examining the code reveals that "ora_types" and "ora_lengths" are only > available if the handle is "oraperl compatible" (tm) ;-). > Ooops :) I guess I was being a little over zelous in disabling those attributes before providing standard alternatives. > I'd like to get at this meta-information. Is there some way to do this > or is there a way being planned to get at it? > Please don't forget that the Oraperl emulation is the only supported interface. All else is changeable. Still, you have two alternatives right now. Either do what Oraperl.pm does: $drh->{CompatMode} = 1; Or do what I've just done to the source: *** tmp.c Thu Oct 5 10:54:43 1995 --- dbdimp.c Thu Oct 5 10:55:00 1995 *************** *** 878,880 **** ! if (oraperl && kl==11 && strEQ(key, "ora_lengths")) { AV *av = newAV(); --- 878,880 ---- ! if (kl==11 && strEQ(key, "ora_lengths")) { AV *av = newAV(); *************** *** 884,886 **** ! } else if (oraperl && kl==9 && strEQ(key, "ora_types")) { AV *av = newAV(); --- 884,886 ---- ! } else if (kl==9 && strEQ(key, "ora_types")) { AV *av = newAV(); > Thanks, > > Erik Rantapaa > rantapaa@math.umn.edu > Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <24206-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 16:31:43 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 15:28:11 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa26402; 5 Oct 95 16:27 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA03410 for perldb-interest-real; Thu, 5 Oct 1995 06:20:15 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA03406 for ; Thu, 5 Oct 1995 06:20:12 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id aa12538; 5 Oct 95 12:38 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <20212-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 10:49:34 +0100 To: cmullen@cs.oberlin.edu Subject: Re: perl and databases Cc: perldb-interest@fugue.com Date: Thu, 5 Oct 1995 10:49:34 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510051238.aa12538@relay-3.mail.demon.net> > From: Spencer Mullen > > are there any perl routines for talking to a database that lives on > a different machine and architecture than the machine on which > the perl program is running over a tcp/ip connection? > Not yet, but it's always been part of my plans that a proxy driver should be developed which would implement this. Don't expect it any time soon - unless you want to start implementing it yourself. Meanwhile you have to rely on database vendor specific networking (Oracle SQL*Net, Ingres/Net etc etc) Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <25367-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 17:50:37 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 05 Oct 95 16:24:30 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa05540; 5 Oct 95 17:23 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa25083; 4 Oct 95 17:11 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA28326 for perldb-interest-real; Wed, 4 Oct 1995 05:15:48 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA28322 for ; Wed, 4 Oct 1995 05:15:46 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA86003; Wed, 4 Oct 1995 07:17:57 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for chs@nando.net at austin.ibm.com; id AA21836; Wed, 4 Oct 1995 07:17:56 -0500 Message-Id: <9510041217.AA21836@dax.austin.ibm.com> To: "Alan K. Stebbens" Cc: Chip Salzenberg , mhm@austin.ibm.com, perldb-interest@fugue.com Subject: Re: Fetching null columns In-Reply-To: Your message of Tue, 03 Oct 95 20:08:03 -0700. <199510040308.UAA12642@dokoka> Date: Wed, 04 Oct 95 07:17:55 -0500 Alan Stebbens writes: |> > According to mhm@austin.ibm.com: |> > > Alan Stebbens writes: |> > > > Each column (array element) can then be "undef" (for null), or a |> > > >defined value, including the empty string. A special "isnull" |> > > >function is unnecessary. |> > > |> > > If the first item in @row is undef then the loop is terminated. |> > |> > If that's true in 5.001m, then there's a bug in 5.001m. It's still a |> > good interface to use undef "NULL column" and () for "no more rows". |> |> This is not true in 5.001m. It is safe to use the construct: |> |> while (defined(@row = dbfetchrow $dbh)) { |> ... |> } [ working perl5.001 sample removed ] This fragment fails when a null column is returned as undef. #!/usr/local/bin/perl use DBI; DBI->_debug_dispatch(5); $dbh = DBI->connect('sample'); $stmt = $ARGV[0]; $sth = $dbh->prepare($stmt); print "NUM Params: ",$sth->{'NUM_OF_PARAMS'},"\n"; $sth->execute(); while (defined(@row = $sth->fetchrow)) { $" = "||"; print "Row: @row\n"; } $sth->finish(); Excerpts from the debug log: # column description details fbh 0: 'REMARKS' NULLable, type 12, dbsize 254, dsize 254, p254 s0 out: ftype 1, indp 0, bufl 255, rlen 255, rcode 0 <- prepare= DBI::st=HASH(0x2003fa60) >> FETCH DISPATCH (DBI::st=HASH(0x2003fa9c) @2 g0 a0 r1) -> FETCH for DBD::DB2::st (DBI::st=HASH(0x2003fa9c) 'NUM_OF_PARAMS') FETCH DBI::st=HASH(0x2003fa9c) 'NUM_OF_PARAMS' <- FETCH= 0 >> execute DISPATCH (DBI::st=HASH(0x2003fa60) @1 g0 a20021f48 r1) -> execute for DBD::DB2::st (DBI::st=HASH(0x2003fa60)~0x2003fa9c) <- execute= 1 >> fetchrow DISPATCH (DBI::st=HASH(0x2003fa60) @1 g1 a0 r1) -> fetchrow for DBD::DB2::st (DBI::st=HASH(0x2003fa60)~0x2003fa9c) dbih_get_fbav 1/1 => 200b2194 #fetch occurring dbd_st_fetch 1 fields 0: rc=0 '' #undef returned as single item fetched in @row <- fetchrow= ( undef ) [1 items] #notice that the contents contents of the while loop have not be #executed >> finish DISPATCH (DBI::st=HASH(0x2003fa60) @1 g0 a20021ea8 r1) -> finish for DBD::DB2::st (DBI::st=HASH(0x2003fa60)~0x2003fa9c) <- finish= 1 DBI::END DBI::disconnect_all DBI DBI::disconnect_all for 'DB2' So now we have a test case provided by "Alan K. Stebbens" which proves it works in 5.001 and I have a test case which proves it is broken in 5.001 :-( Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <29231-0@ignite.demon.co.uk>; Fri, 6 Oct 1995 07:05:40 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 03:45:36 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ak28285; 6 Oct 95 4:44 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <09999-0@frigate.doc.ic.ac.uk>; Fri, 6 Oct 1995 00:09:16 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA04680 for perldb-interest-real; Thu, 5 Oct 1995 14:43:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA04676 for ; Thu, 5 Oct 1995 14:43:24 -0700 Received: by gw.home.vix.com id AA13080; Thu, 5 Oct 95 14:45:38 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id QAA01083; Thu, 5 Oct 1995 16:45:36 -0500 Date: Thu, 5 Oct 1995 16:45:36 -0500 (CDT) From: Ben Lindstrom Cc: perldb-interest@vix.com Subject: Re: I need a little help here.. In-Reply-To: <9510051239.ac12538@relay-3.mail.demon.net> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 5 Oct 1995, Tim Bunce wrote: > Does that help? > > Are you looking at the code in order to develop another driver? > I think so...I'm going to be doing more work on it this weekend. I've stopped looking at the oracle code while I was trying to get a connect/disconnect version of the QuickBase drivers. Hopefully the Sparc with Oracle on it will help me better understand the code more. Just figuring out what the QuickBase Objective-C libraries support internally is a neat trick. =) ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <29232-11@ignite.demon.co.uk>; Fri, 6 Oct 1995 07:07:41 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 03:17:23 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa21305; 6 Oct 95 4:16 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA05067 for perldb-interest-real; Thu, 5 Oct 1995 18:30:35 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from skye.as.utexas.edu (skye.as.utexas.edu [128.83.129.233]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id SAA05063 for ; Thu, 5 Oct 1995 18:30:32 -0700 Received: by skye.as.utexas.edu (5.x/SMI-SVR4) id AA04364; Thu, 5 Oct 1995 20:30:02 -0500 Date: Thu, 5 Oct 1995 20:30:00 -0500 (CDT) From: Mark Jefferys Cc: Larry Wall , Tim Bunce , perldb-interest@fugue.com, mhm@austin.ibm.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false (was: Fetching null columns) In-Reply-To: <9510041858.AA26792@scalpel.netlabs.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Someone (Tim?) wrote: % > From: mhm@austin.ibm.com % > % > Ok, question time. How do we expect the DBI::DBD interfaces to work % > when it comes to columns returning null? % > % Instances of NULLS are represented on input and output as undefs. Period. % % % > $sth->execute(); % > while (@row = $sth->fetchrow) { % > ... % > } % > % > If the $stmt requests only one column from the table % > and the column can contain NULLs then it is possible to return only % > part of the data. % % This is a perl bug discovered recently by Michael Peppler working on Sybase. % I CC'd a reply to perlbug@perl.com but I think it bounced. I'll CC this. Well, this can't be considered a bug, since it's intentional (maybe a documentation bug---I was really lax on that obligation). It might be a bad idea, although I don't really like any of the solutions: This is there so that loops like while (($user, $uid) = (getpwent)[0,2]) { ... } doesn't run forever. (getpwent returns () at the end, which turns into (undef) x 2, which still ends up false on assignment.) There are several ways I can think of doing this, offhand: 1. Document the real behavior. Suggest using @arr = func(); if (@arr) ... or if ((@arr = func()), @arr) ... when a list of undefs should be true. I suspect there are a some other cases where this is what the programmer wants, but I have no idea how often. 2. Make list-assignments return undef for zero objects assigned, and SV_NO for undef-only assignments. This allows: if (defined(@arr = func())) ... 3. Make list-assignments normal, but have list-slices cut off trailing undefs, so (getpwent)[0,2] returns an empty list. (This method was implemented in one of the perl alphas.) This would fix the immediate problem, but would add others: A. This will cause code which relies on list-slices returning a fixed number of items to break. Sub calls are a likely source of problems: func1((func2())[1,3], ++$cnt); # Expects $cnt in 3rd place, not last. It gets even worse if someone tries to write a DB module in Perl: sub row_subset { my($self, $row, @cols) = @_; ($self->get_row($row))[$self->col_nums(@cols)]; # col_nums converts column names to numerical indices } if (@arr = row_subset($i, X, Y)) { # oops: X,Y are undefs # Same problem, different cause B. If any of the elements sliced is magic, we either have to get its value, which wastes time and can cause problems, or *not* get its value, which can fail. 4. Same as #3, except only eliminate undefs if you can eliminate all of them. Slightly reduces the problem with #A, does nothing for #B. 5. Same as #3, except cut off trailing *out-of-bound* references in list-slices. More likely to do what you want with #A, and fixes #B. Brings into sharp relief the question, "Should array-slices act like list-slices here." I don't like making them different, but it makes it that much harder to get tied arrays working. (And if you find this question hard, should hash slices cut off trailing non-existent elements? 1/2 :-)) 6. Combine #4 and #5. Same comments as #5. 7. Make a list-slice of an empty list be empty. This is an example of minimal intervention: make sure empty lists stay empty, and thus false, but otherwise acts normally. This would be somewhat easier to implement for tied arrays (and hashes). 8. Make array assignments of slices magical, returning false if the original list was empty (or something). A really do-what-I-want solution. 9. Use awk. :-) I kinda like #2 or #7 at the moment; #8 has a quirky, Perlish feel to it that attracts me, although I suspect that it's a bad idea. Actually, #9 is looking pretty good at the moment... Mark ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <00376-0@ignite.demon.co.uk>; Fri, 6 Oct 1995 07:10:31 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 06:09:25 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa19845; 6 Oct 95 7:08 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA05576 for perldb-interest-real; Thu, 5 Oct 1995 21:26:43 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from metronet.com (tye@fohnix.metronet.com [192.245.137.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id VAA05572 for ; Thu, 5 Oct 1995 21:26:40 -0700 Received: by metronet.com id AA20463 (5.67a/IDA1.5hp for perldb-interest@fugue.com); Thu, 5 Oct 1995 23:28:26 -0500 From: Tye McQueen Message-Id: <199510060428.AA20463@metronet.com> Subject: Re: perl bug: array of undefs is falsely false To: Mark Jefferys Date: Thu, 5 Oct 1995 23:28:24 -0500 (CDT) Cc: lwall@scalpel.netlabs.com, Tim.Bunce@ig.co.uk, perldb-interest@fugue.com, mhm@austin.ibm.com, perlbug@perl.com In-Reply-To: from "Mark Jefferys" at Oct 5, 95 08:30:00 pm Content-Transfer-Encoding: 8bit X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 6394 Excerpts from the mail message of Mark Jefferys: ) ) ["@x= (undef,undef)" is false.] ) ) Well, this can't be considered a bug, since it's intentional (maybe a ) documentation bug---I was really lax on that obligation). It might be ) a bad idea, although I don't really like any of the solutions: Indeed, a complicated issue. I've rolled this around at many angles... I think it is clear that often you don't want undefs created as place holders and often you don't want trailing undefs stripped. Perl puts practicality before rigid consistancy so creating and stripping undefs is common, but not universal. I think the best solution is to allow the user to select from both types of behavior, so long as it can be done reasonably. For example, using "-1" in "split(/re/,$str,-1)" lets us prevent the default stripping of trailing nulls. I'm sure we don't want to change @arr[@sub] and @hash{@sub} so that they no longer return ("create") undefs for out-of-bounds subscripts and non-existant keys. But I think being able to request exactly that behavior could be generally useful. If we allow this behavior for anonymous array slices as well, then that can be used for the if( @arr= (getpwent)[0,2] ) case and thus removes any motivation for having "@x= (undef,undef)" considered false (at least I don't see any other motivation). How about: @arr[@sub] for map { $arr[$_] } @sub @arr([@sub]) for map { $_ < $[ || $#arr < $_ ? () : $arr[$_] } @sub @hash{@sub} for map { $arr{$_} } @sub @hash({@sub}) for map { $_ in keys(%hash) ? $hash{$_} : () } @sub where "$_ in keys(%hash)" is pseudo Perl for whether hv_fetch() returned 0. [I don't think this adds ambiguity to the parsing, but I certainly could be wrong.] That is, @arr[@sub] and @hash{@sub}, stay the same. @arr([@sub]) does not "create" undefs as place holders for out-of-bounds elements. @hash({@sub}) does not "create" undefs as place holders for non-existant keys. So we would now have to do: if( @arr= (getpwent)([0,2]) ) ^ ^ since "if( @arr= (getpwent)[0,2] )" never worked in perl4 anyway. I'd also like a way to tell if a key is present in a hash, ie. distinquish between hv_fetch() returning 0 vs. &(&sv_undef). For most tied hashes, it would be the same as defined($hash{$key}), but defining a non-default EXISTS() would allow a difference. So using @tied_hash({@sub}) would cause EXISTS() to be called if FETCH() returned undef [I probably wouldn't change hv_fetch() but instead change a few callers of hv_fetch() and add a caller]. Perhaps a new boolean: "$key in %hash" or "in(%hash,$key)". Perhaps "keys(%hash,@keys)" could return the elements of @keys that are keys of %hash: "map { $_ in %hash ? $_ : () } @keys". But, in case you hate @arr([@sub]) and can't suggest a better way to suppress undef place holders, here are my thoughts on the previously proposed solutions: ) 1. Document the real behavior. Suggest using ) if ((@arr = func()), @arr) ... ) when a list of undefs should be true. Kinda yuck but no worse than a few other rare cases in Perl. Probably not possible in the face of the powerful special-interest lobby from the DB users. :) ) 2. Make list-assignments return undef for zero objects assigned, ) and SV_NO for undef-only assignments. This allows: ) if (defined(@arr = func())) ... This doesn't allow you, in (func())[0,2], to distinguish whether func() returned () or (undef,"foo",undef,"bar"). It also means defined() is required above but if you try to make a habit of using defined() you'll get bit because while( defined( @arr= (getpwent)[0,2] ) ) { will never end. ) 3. Make list-assignments normal, but have list-slices cut off ) trailing undefs, so ) (getpwent)[0,2] ) returns an empty list. ) A. This will cause code which relies on list-slices ) returning a fixed number of items to break. Sub calls ) are a likely source of problems: ) func1((func2())[1,3], ++$cnt); ) # Expects $cnt in 3rd place, not last. ) B. If any of the elements sliced is magic, we either have to ) get its value, which wastes time and can cause problems, ) or *not* get its value, which can fail. I don't understand this last one (B). ) 4. Same as #3, except only eliminate undefs if you can eliminate ) all of them. Slightly reduces the problem with #A, does nothing ) for #B. ) 5. Same as #3, except cut off trailing *out-of-bound* references in ) list-slices. More likely to do what you want with #A, and fixes ) #B. ) 6. Combine #4 and #5. Same comments as #5. The problem of documenting these properly makes me very wary of them. ) 7. Make a list-slice of an empty list be empty. This is an example ) of minimal intervention: make sure empty lists stay empty, and ) thus false, but otherwise acts normally. This would be somewhat ) easier to implement for tied arrays (and hashes). This raises the problem of (A) again. ) 8. Make array assignments of slices magical, returning false if the ) original list was empty (or something). A really do-what-I-want ) solution. But will still surprise people with while( @ids= (getpwent)[2,3] ) { if( $ids[0] == $uid && $ids[1] == $gid ) { last; } } if( @ids ) { # We found a match! } because we'll always "find a match". If it can be implemented relatively easilly, I'd vote for this. [Why do I keep thinking we need a new value, NoneSuch, which is used for things that have never been set (such as out-of-bounds elements)? exists() would be to NoneSuch as defined() is to sv_undef. There would be no _direct_ way to set anything to NoneSuch ("$#arr=5" indirectly sets "$arr[6]" to NoneSuch, "delete($hash{key})" indirectly sets "$hash{key}" to NoneSuch). In an assignment, NoneSuch is always transformed to sv_undef. Would this help the problem of sv_undef causing too many "Use of uninitialized value" warnings?] -- Tye McQueen tye@metronet.com || tye@doober.usu.edu Nothing is obvious unless you are overlooking something http://www.metronet.com/~tye/ (scripts, links, nothing fancy) ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03653-2@ignite.demon.co.uk>; Fri, 6 Oct 1995 12:12:47 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 11:11:03 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa01674; 6 Oct 95 12:09 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa12336; 4 Oct 95 19:17 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA29468 for perldb-interest-real; Wed, 4 Oct 1995 09:12:27 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA29464 for ; Wed, 4 Oct 1995 09:12:23 -0700 Received: by gw.home.vix.com id AA07431; Wed, 4 Oct 95 09:14:35 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from hermes.demon.co.uk by relay-3.mail.demon.net id aa10090; 4 Oct 95 10:09 +0100 Date: Tue, 03 Oct 1995 10:32:13 +0100 Message-Id: <743@hermes.demon.co.uk> From: Adrian Lee Organization: None whatsoever Reply-To: Adrian@hermes.demon.co.uk To: perldb-interest@vix.com Subject: Subscribe X-Mailer: Newswin Alpha 0.9 Lines: 9 subscribe perldb-interest -- Adrian Lee No connection whatsoever with any other demon.co.uk users. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03769-0@ignite.demon.co.uk>; Fri, 6 Oct 1995 12:17:39 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 11:16:37 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa03797; 6 Oct 95 12:15 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa12945; 4 Oct 95 19:20 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA29111 for perldb-interest-real; Wed, 4 Oct 1995 07:47:05 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from black.csi.cam.ac.uk (root@black.csi.cam.ac.uk [131.111.8.33]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA29090 for ; Wed, 4 Oct 1995 07:43:47 -0700 Received: from peewit [131.111.10.37] by black.csi.cam.ac.uk with smtp (Smail-3.1.29.0 #36) id m0t0V2N-000CDNC; Wed, 4 Oct 95 15:43 BST Date: Wed, 4 Oct 1995 15:42:26 BST From: Charles Jardine Subject: Re: perl bug: array of undefs is falsely false (was: Fetching null columns) To: Tim Bunce cc: perldb-interest@fugue.com, mhm@austin.ibm.com, perlbug@perl.com, lwall@sems.com Message-ID: Priority: Normal MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII On Wed, 4 Oct 1995 12:29:37 +0100 Tim Bunce wrote: > Not quite. It breaks if *all* the columns are nulls. Here's a test case: The following test with perl 5.001 patchlevel 1l indicates that it is quite subtle. It is the evaluation in scalar context of the assignment of a list of undefs which is broken. Evaluation of the list itself in scalar context works. This would not be a good reason to change the DBI spec;-) Sample program: print "$]\n"; @u = (undef); # a list of one undef @x = ('x'); # a list of one defined value $un1 = @u; $un2 = (@v = @u); $un3 = @v; $xn1 = @x; $xn2 = (@y = @x); $xn3 = @y; (defined $un1) ? print "un1=$un1\n" : print "un1 undefined\n"; (defined $un2) ? print "un2=$un2\n" : print "un2 undefined\n"; (defined $un3) ? print "un3=$un3\n" : print "un3 undefined\n"; (defined $xn1) ? print "xn1=$xn1\n" : print "xn1 undefined\n"; (defined $xn2) ? print "xn2=$xn2\n" : print "xn2 undefined\n"; (defined $xn3) ? print "xn3=$xn3\n" : print "xn3 undefined\n"; Output: 5.001 un1=1 un2 undefined un3=1 xn1=1 xn2=1 xn3=1 For the record, perl4 does not show this bug, so existing oraperl users need not worry. Charles Jardine, Univ of Cambridge (UK). ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <17801-0@ignite.demon.co.uk>; Fri, 6 Oct 1995 21:00:18 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 19:51:59 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa17975; 6 Oct 95 20:51 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA08131 for perldb-interest-real; Fri, 6 Oct 1995 11:13:11 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from eve.mscs.mu.edu (eve.mscs.mu.edu [134.48.4.166]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA08127; Fri, 6 Oct 1995 11:13:09 -0700 Received: by eve.mscs.mu.edu (Smail3.1.28.1 #9) id m0t1HFD-0002DjC; Fri, 6 Oct 95 13:11 CDT Message-Id: From: Saima Zobair Subject: Help! oraperl: oracle error on ora_login() To: perldb-interest@fugue.com Date: Fri, 6 Oct 1995 13:11:38 -0500 (CDT) Cc: perldb-interest-request@fugue.com X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 501 Hi! I am trying to connect oracle and the webserver using oraperl2.4 I am using oracle7. When I run the oraperl script on the command line it executes fine and the database query runs perfectly. But when I try to run the same oraperl file from /cgi through netscape... it gives a an error on the &oralogin(). On printing the $ora_errstr I get: Error while trying to retrieve text for error ORA-02701 Can someone please help me with this ? Thanks, Saima Zobair zobair@mscs.mu.edu ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:05:50 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 23:03:55 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa17285; 9 Oct 95 0:02 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA04136 for perldb-interest-real; Sun, 8 Oct 1995 14:49:34 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA04132 for ; Sun, 8 Oct 1995 14:49:32 -0700 Received: by gw.home.vix.com id AA27676; Sun, 8 Oct 95 14:51:49 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from 128.230.1.225 (npac-dial-225.syr.edu [128.230.1.225]) by newstand.syr.edu (8.6.12/8.6.12) with SMTP id RAA21204; Sun, 8 Oct 1995 17:47:58 -0400 Message-Id: <199510082147.RAA21204@newstand.syr.edu> Date: Sun, 08 Oct 95 17:54:06 0000 From: Gang Cheng Organization: NPAC X-Mailer: Mozilla 1.1N (Macintosh; I; 68K) Mime-Version: 1.0 To: www-vrml@wired.com, sp-discussion@mcs.anl.gov, perldb-interest@vix.com Subject: (no subject) X-Url: http://kayak.npac.syr.edu:1963/cgi-bin/news/showbody?bid=24000661&word=~1 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii We at Northeast Parallel Architecture Center (NPAC), Syracue University has developed an archive system for searching/reading USENET newsgrou ps, mailing lists and personal mailbox from WWW. An Oracle database server is used to store/manage mails and two search/navigation interfaces accesible by any WWW browser to the archive are provided: one is an advanced search interface allowing queries with various options such as query by mail header, by date, by subject (keywords), by sender. The other is a Hypermail-like navigation interface for users familiar with Hypermail. We offer free access to this archive for the Internet Internet community. Check the URL . The whole archive is automatically updated in every 30 minutes. Currently the following newsgroups (total 40) are archived in the database: comp.arch comp.databases comp.databases.oracle comp.parallel comp.parallel.mpi comp.parallel.pvm comp.lang.java alt.www.hotjava comp.infosystems comp.infosystems.www.authoring.cgi comp.infosystems.www.announce comp.infosystems.www.authoring.html comp.infosystems.www.authoring.images comp.infosystems.www.authoring.misc comp.infosystems.www.misc comp.infosystems.www.advocacy comp.infosystems.www.browsers.mac comp.infosystems.www.browsers.ms-windows comp.infosystems.www.browsers.x comp.infosystems.www.browsers.misc comp.infosystems.www.servers.ms-windows comp.infosystems.www.servers.unix comp.infosystems.www.servers.mac comp.infosystems.www.servers.misc comp.graphics comp.graphics.visualization comp.lang.c++ comp.multimedia comp.security.misc comp.sources.unix comp.sources.x comp.unix.admin comp.unix.aix comp.unix.misc comp.unix.programmer comp.unix.solaris alt.comp.compression comp.databases.object comp.lang.perl.misc syr.general The following mailing lists are archived: java-interest java-porting java-announce hotjava-interest sp-discussion www-vrml perldb-interest-request@fugue.com This system is still under development and more functions are planned to add to this system. Please send your comments to Gang Cheng (gcheng@npac.syr.edu . Thanks. --- Gang ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28310-1@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:06:13 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 22:46:21 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa05449; 6 Oct 95 23:45 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id ab21777; 5 Oct 95 18:21 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA03844 for perldb-interest-real; Thu, 5 Oct 1995 08:33:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA03839 for ; Thu, 5 Oct 1995 08:33:01 -0700 Received: by gw.home.vix.com id AA15159; Thu, 5 Oct 95 08:35:15 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ab03093; 5 Oct 95 16:30 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <24126-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 16:30:00 +0100 To: perldb-interest@vix.com, Kenneth.Holmlund@physics.umu.se Subject: Re: subscribe Date: Thu, 5 Oct 1995 16:30:00 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510051630.ab03093@relay-3.mail.demon.net> > From: Kenneth.Holmlund@physics.umu.se > > subscribe Kenneth Holmlund > Hi, > I've tried to contact the listserv to add me to this > list but I didn't succeed so I guess I have to > bother you all :) What 'listserv' and who do you know that you didn't succeed? > How do I subscribe ? By mailing to perldb-interest-request@vix.com (a human. Hi Ted). > Kenneth Holmlund > Kenneth.Holmlund@Physics.UmU.SE > Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-2@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:06:23 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 06 Oct 95 22:38:35 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa14448; 6 Oct 95 23:38 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA08832 for perldb-interest-real; Fri, 6 Oct 1995 14:17:20 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from skye.as.utexas.edu (skye.as.utexas.edu [128.83.129.233]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id OAA08828 for ; Fri, 6 Oct 1995 14:17:18 -0700 Received: by skye.as.utexas.edu (5.x/SMI-SVR4) id AA05206; Fri, 6 Oct 1995 16:16:53 -0500 Date: Fri, 6 Oct 1995 16:16:52 -0500 (CDT) From: Mark Jefferys To: Tim Bunce Cc: tye@metronet.com, lwall@scalpel.netlabs.com, perldb-interest@fugue.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: <9510062128.ab20640@relay-3.mail.demon.net> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 6 Oct 1995, Tim Bunce wrote: % This topic generated quite a discussion! % % I'd like to narrow it down and check if we've reached a consensus: % %> From: Mark Jefferys %> %>% since "if( @arr= (getpwent)[0,2] )" never worked in perl4 anyway. %> %> Note that this has worked in perl5 for a little while, so there may be %> a few scripts that this would break. % % Since it was never well documented I doubt there are that many scripts % that use this other than those you've written yourself. Then again % Larry may take the view that the possibility makes the risk of undoing % the change to high. Actually, I'm not too worried about my own scripts. I'm concerned because I remember this coming up as a "Why doesn't this work?", "Because, but it would be better if it did," situation. Some people may have written "while (@arr = (getpwent)[0,2]) ..." or an equivalent, never realizing they were using a feature. Changing it now would break their scripts (if they exist at all). %>% Excerpts from the mail message of Mark Jefferys: [ #1 is current behavior: list-assign in scalar context returns ] [ SV_UNDEF if only undefs assigned, otherwise count is returned. ] %>%) 2. Make list-assignments return undef for zero objects assigned, %>%) and SV_NO for undef-only assignments. This allows: %>%) %>%) if (defined(@arr = func())) ... % %>%) 3. Make list-assignments normal, but have list-slices cut off %>%) trailing undefs, so %>%) (getpwent)[0,2] %>%) returns an empty list. %>%) 7. Make a list-slice of an empty list be empty. This is an example %>%) of minimal intervention: make sure empty lists stay empty, and %>%) thus false, but otherwise acts normally. This would be somewhat %>%) easier to implement for tied arrays (and hashes). %>%) I kinda like #2 or #7 at the moment; [...] % % So, to summarise the situation (and check that _I_ understand it :) ... % % - Currently evaluation in scalar context of the assignment of a list % of undefs is regarded as false. % % - It seems widely agreed that this is not ideal. % % - This was added to cater for: while (($user, $uid) = (getpwent)[0,2]) { } Technically, #3 was added to cater to this. I didn't think much of the solution, and offered #1 as an alternative (with source). I don't remember any other ideas being discussed. % - It seems widely agreed that this is useful. % % I also like #7 "Make a list-slice of an empty list be empty". % Feels kind'a like null-propagation in SQL expressions. % % #2 seems too subtle for people to remember to use in their code. % Especially where they don't expect to get nulls but might in error % conditions. Perhaps. I offered #2 as a variation on #1 which is easier to work around. Mark ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28310-11@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:08:29 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 00:29:41 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa18337; 7 Oct 95 1:28 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id ac21520; 5 Oct 95 18:19 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA03774 for perldb-interest-real; Thu, 5 Oct 1995 08:14:34 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA03770 for ; Thu, 5 Oct 1995 08:14:33 -0700 Received: by gw.home.vix.com id AA13887; Thu, 5 Oct 95 08:16:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ac12538; 5 Oct 95 12:39 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <20332-0@ignite.demon.co.uk>; Thu, 5 Oct 1995 11:12:10 +0100 To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: I need a little help here.. Date: Thu, 5 Oct 1995 11:12:10 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510051239.ac12538@relay-3.mail.demon.net> > From: Ben Lindstrom > > I've been very quiet for too long. =) > > I'm trying to understand the different between dbd_db_* commands and dbd_st_* > commands in the dbdimp.c file. > The dbd_db_* functions implement methods for database handles and the dbd_st_* functions implement methods for statement handles. Both are called from the .xs file. > Could someone enlight me on mainly the differents between the > FETCH and STORE in both areas? I'm sorta lost. > dbd_db_FETCH is called when the application does $value = $dbh->{attribute}; dbd_st_FETCH is called when the application does $value = $sth->{attribute}; ^^ ^^ Similarly the dbd_*_STORE functions are called when setting attribute values. Database and statement handles have different attributes and possibly may have common attribute names with different meanings. It's natural to implement then as separate functions in the driver. dbd_*_STORE functions return FALSE if they don't recognise the attribute name. dbd_*_FETCH functions return NULL if they don't recognise the attribute name. In both these cases the code in the .xs file hands off to some DBI code which checks to see if it recognises the attribute name (e.g., NUM_OF_PARAMS). Does that help? Are you looking at the code in order to develop another driver? Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-14@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:09:55 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 02:15:07 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa23838; 7 Oct 95 3:13 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA10059 for perldb-interest-real; Fri, 6 Oct 1995 18:11:48 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from proxy.netlabs.com (ns.netlabs.com [192.94.48.47]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id SAA10055 for ; Fri, 6 Oct 1995 18:11:46 -0700 Received: by proxy.netlabs.com (4.1/SMI-4.1) id AA08173; Fri, 6 Oct 95 18:14:09 PDT Received: from vaccine-bb.netlabs.com(192.190.199.1) by proxy.netlabs.com via smap (V1.3) id sma008084; Fri Oct 6 18:13:28 1995 Received: from scalpel.netlabs.com by netlabs.com (4.1/SMI-4.1) id AA07576; Fri, 6 Oct 95 15:50:31 PDT Received: from localhost.netlabs.com by scalpel.netlabs.com (4.1/SMI-4.1) id AA09086; Fri, 6 Oct 95 15:49:27 PDT Message-Id: <9510062249.AA09086@scalpel.netlabs.com> To: Tim Bunce Cc: tye@metronet.com, mark@skye.as.utexas.edu, perldb-interest@fugue.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: Your message of "Fri, 06 Oct 95 21:27:26 BST." <9510062128.ab20640@relay-3.mail.demon.net> Date: Fri, 06 Oct 95 15:49:27 -0700 From: Larry Wall : > % Excerpts from the mail message of Mark Jefferys: : : > % ) 2. Make list-assignments return undef for zero objects assigned, : > % ) and SV_NO for undef-only assignments. This allows: : > % ) : > % ) if (defined(@arr = func())) ... : : > % ) 3. Make list-assignments normal, but have list-slices cut off : > % ) trailing undefs, so : > % ) (getpwent)[0,2] : > % ) returns an empty list. : > % ) A. This will cause code which relies on list-slices : > % ) returning a fixed number of items to break. Sub calls : > % ) are a likely source of problems: : > % ) func1((func2())[1,3], ++$cnt); : > % ) # Expects $cnt in 3rd place, not last. : : > % ) 7. Make a list-slice of an empty list be empty. This is an example : > % ) of minimal intervention: make sure empty lists stay empty, and : > % ) thus false, but otherwise acts normally. This would be somewhat : > % ) easier to implement for tied arrays (and hashes). : > % : > % This raises the problem of (A) again. : > : > Granted, although far less often than the others. I like this one : > mostly because the exceptions I can think of are more contrived than : > for the others, and pretty easy to work around: : > : > @arr ? @arr[@idx] : (undef) x @idx : > : > % ) I kinda like #2 or #7 at the moment; [...] I think that as we move into using prototypes, func1() above would probably be declared as sub func1 ($$$); so you wouldn't be able to say func1((func2())[1,3], ++$cnt); in any event. (Unless I did something Really Strange with argument processing to make it look inside a list slice.) : So, to summarise the situation (and check that _I_ understand it :) ... : : - Currently evaluation in scalar context of the assignment of a list : of undefs is regarded as false. : : - It seems widely agreed that this is not ideal. : : - This was added to cater for: while (($user, $uid) = (getpwent)[0,2]) { } : : - It seems widely agreed that this is useful. : : I also like #7 "Make a list-slice of an empty list be empty". : Feels kind'a like null-propagation in SQL expressions. A little, though I think Perl's notion is a little more fail-soft. : #2 seems too subtle for people to remember to use in their code. : Especially where they don't expect to get nulls but might in error : conditions. : : Does anyone have and hard objections or better alternatives? : : Larry, do you concur? Hail the concurring hero, eh? [sorry] I'm leaning toward #7 too. Especially since it makes array assignments a little faster... On the other hand, doing something a bit #8ish wouldn't be too difficult. It would be relatively trivial to inform the OP_LSLICE that it was on the right side of an OP_AASSIGN. I don't think we'd even have to worry about whether the assignment was in a scalar context, though even that wouldn't be difficult to do. Larry ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-16@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:10:59 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 02:35:44 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa02556; 7 Oct 95 3:33 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA10136 for perldb-interest-real; Fri, 6 Oct 1995 18:35:42 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from kjahds.com (root@ringotty6.jvnc.net [192.67.236.206]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id SAA10132 for ; Fri, 6 Oct 1995 18:35:37 -0700 Received: by kjahds.com (Linux Smail3.1.28.1 #1) id m0t1OBN-0000XMC; Fri, 6 Oct 95 21:36 EDT Date: Fri, 6 Oct 1995 21:36:09 -0400 (EDT) From: Kenneth Albanowski To: Larry Wall cc: Tim Bunce , tye@metronet.com, mark@skye.as.utexas.edu, perldb-interest@fugue.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: <9510062249.AA09086@scalpel.netlabs.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 6 Oct 1995, Larry Wall wrote: > I think that as we move into using prototypes, func1() above would > probably be declared as > > sub func1 ($$$); > > so you wouldn't be able to say > > func1((func2())[1,3], ++$cnt); > > in any event. (Unless I did something Really Strange with argument > processing to make it look inside a list slice.) I think this is a case of knowing just enough LISP to be dangerous. The thing is, some people (me included, I'm afraid) tend to write code like that. Talk about breaking current scripts... > I'm leaning toward #7 too. Especially since it makes array assignments > a little faster... It certainly seems a decent solution. Easy to explain, and minimal surprise. > On the other hand, doing something a bit #8ish wouldn't be too difficult. > It would be relatively trivial to inform the OP_LSLICE that it was on the > right side of an OP_AASSIGN. I don't think we'd even have to worry about > whether the assignment was in a scalar context, though even that wouldn't > be difficult to do. Non-minimal surprise and context sensistivity... Then again, this _is_ perl. ;-) > Larry -- Kenneth Albanowski (kjahds@kjahds.com, CIS: 70705,126) ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28310-16@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:10:59 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 01:18:16 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa04643; 7 Oct 95 2:17 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa12526; 7 Oct 95 2:02 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA09587 for perldb-interest-real; Fri, 6 Oct 1995 16:59:00 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from watson.ibm.com (watson.ibm.com [129.34.139.4]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA09583 for ; Fri, 6 Oct 1995 16:58:58 -0700 Received: from WATSON by watson.ibm.com (IBM VM SMTP V2R3) with BSMTP id 8747; Fri, 06 Oct 95 20:01:01 EDT Received: from YKTVMV by watson.vnet.ibm.com with "VAGENT.V1.02 on VAGENT2" id 6210; Fri, 6 Oct 1995 20:01:01 EDT Received: from lusitania.watson.ibm.com by yktvmv.watson.ibm.com (IBM VM SMTP V2Rx) with TCP; Fri, 06 Oct 95 20:00:46 EDT Received: from localhost by lusitania.watson.ibm.com (AIX 3.2/UCB 5.64/950830) id AA28953; Fri, 6 Oct 1995 20:00:26 -0400 Message-Id: <9510070000.AA28953@lusitania.watson.ibm.com> X-Mailer: exmh version 1.6.2 7/18/95 To: perldb-interest@fugue.com Subject: Bug (?) in fetch() Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 06 Oct 1995 20:00:25 -0400 From: Chet Murthy [I'm new to this list, so if this is not the way to present bugs with DBI, or the drivers, please let me know. --chet--] There seems to be a bug in the fetch() function in DB2.xs, which, I think, is also present in the Oracle version. I enclose the diffs for the DB2 version, which, should be transportable to the Oracle code. I have checked that this code exists in Oracle.xs as below: $Id: Oracle.xs,v 1.33 1995/08/26 17:39:01 timbo Rel $ (I think it's DBD::Oracle v0.25) but have fixed the problem in DB2.xs. Synopsis: If I call $st->fetch(), which calls dbd_st_fetch, it returns an AV, which is cast to an SV, and then returned immediately. But it needs to be wrapped up in a reference before being returned. This results in the code below: ---------------------------------------------------------------- #!/u/chet/contrib/bin/perl use DBI; $dbh = DBI->connect('sample'); $stmt = $ARGV[0]; $sth = $dbh->prepare($stmt); print "NUM Params: ",$sth->{'NUM_OF_PARAMS'},"\n"; $sth->execute(); while ($ary = $sth->fetch) { $" = "||"; print "Row: @{$ary}\n"; } $sth->finish(); ---------------------------------------------------------------- failing instantly with an "undef" -- that is, the first fetch fails, even if the query returns something, because the return value of dbd_st_fetch is lost. If I turn on debugging, I see that DBI::dispatch thinks that it's been returned an undef (since an AV is not an SV). The modification: ---------------------------------------------------------------- --- DB2.xs.~1~ Fri Sep 22 14:46:02 1995 +++ DB2.xs Fri Oct 6 19:53:05 1995 @@ -221,7 +221,12 @@ fetch(sth) SV * sth CODE: - ST(0) = (SV*)dbd_st_fetch(sth); + AV * av = dbd_st_fetch(sth); + if (av == Nullav) { + ST (0) = &sv_undef; + } else { + ST(0) = newRV ((SV *)av); + } ---------------------------------------------------------------- (1) wraps non-NULL AV's in a reference, which is what is returned (2) returns &sv_undef if the array is empty -- i.e., if the fetch failed. --chet-- ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-27@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:15:25 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 04:52:30 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa02296; 7 Oct 95 5:50 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA10483 for perldb-interest-real; Fri, 6 Oct 1995 20:48:57 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA10479 for ; Fri, 6 Oct 1995 20:48:55 -0700 Received: by gw.home.vix.com id AA12409; Fri, 6 Oct 95 20:51:05 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ad16968; 6 Oct 95 19:27 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <16239-0@ignite.demon.co.uk>; Fri, 6 Oct 1995 19:16:14 +0100 To: skimo@dns.ufsia.ac.be Subject: Re: Multiple DBD's Cc: perldb-interest@vix.com Date: Fri, 6 Oct 1995 19:16:14 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510061927.ad16968@relay-3.mail.demon.net> > From: Sven Verdoolaege > > For those on the list: i have problems connecting to > both an mSQL and an Oracle database > > > > my script: > > > #!/u01/home/verdools/bin/perl > > > use DBI; > > > $DB='ufpers'; > > > $ENV{'MSQL_HOME'} = "/usr/local/Minerva"; > > > $ENV{'MSQL_TCP_PORT'} = 22532; > > > $dbhD = DBI->connect( 'dns',$DB,'','mSQL') || die("geen database"); > > > $dbhD->disconnect; > > > $ENV{'ORACLE_HOME'} = "/u01/home/oracle/product/7.1.4"; > > > $ENV{'ORAHOME'} = "/u01/home/oracle/product/7.1.4"; > > > print "before\n"; > > > $dbhS = DBI->connect( $DB, 'job_verdool_s', 'my password:-)','Oracle' ) || die("geen ; > > > print "after\n"; > > > $dbhS->disconnect; > > > > > > my output: > > > before > > > rc: 0 what: Unknown MSQL Server Host (ufpers) > > > Database handle DBI::db=HASH(0x1400fff18) DESTROY ignored - never set up at /u0. > > > geen database at ./cp2 line 11. I think the "DESTROY ignored - never set up" warning may be a bug in the mSQL driver due to clearing DBIc_IMPSET too early. It's only a warning. The connect failed for the reason shown. > > > if i comment the connect and disconnect lines from mSQL, i get > > > before > > > after > > > > > > if i put the oracle stuff before the mSQL stuff, i get > > > before > > > after > > > Memory fault - core dumped I'd need to see the core dump but only if you still get it after upgrading to the latest DBI and DBD::Oracle. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28310-31@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:18:31 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 08:19:09 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa27424; 7 Oct 95 9:18 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa21994; 6 Oct 95 22:46 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA08642 for perldb-interest-real; Fri, 6 Oct 1995 13:35:28 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA08638 for ; Fri, 6 Oct 1995 13:35:24 -0700 Received: by gw.home.vix.com id AA23208; Fri, 6 Oct 95 13:37:38 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id QAA19986; Fri, 6 Oct 1995 16:36:18 -0400 Message-Id: <199510062036.QAA19986@pilot.njin.net> To: perldb-interest@fugue.com Cc: phj@pilot.njin.net Subject: How to query for primary keys ? Date: Fri, 06 Oct 1995 16:36:17 -0400 From: "Patricia H. Ju" Hi, How does one query 1. if a certain field is a primary key 2. if a table has a primary key 3. what the primary key of a table is, if one exists using DBD-Oracle? Patricia ........... "Men have forgotten this truth," said the fox. "But you must not forget it. You become responsible, forever, for what you have tamed . . ." ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28310-33@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:19:03 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 08:21:12 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa28014; 7 Oct 95 9:20 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa22300; 6 Oct 95 22:49 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA08655 for perldb-interest-real; Fri, 6 Oct 1995 13:36:16 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA08651 for ; Fri, 6 Oct 1995 13:36:02 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ab20640; 6 Oct 95 21:28 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <18052-0@ignite.demon.co.uk>; Fri, 6 Oct 1995 21:27:26 +0100 To: tye@metronet.com, mark@skye.as.utexas.edu Subject: Re: perl bug: array of undefs is falsely false Cc: lwall@scalpel.netlabs.com, perldb-interest@fugue.com, perlbug@perl.com Date: Fri, 6 Oct 1995 21:27:26 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510062128.ab20640@relay-3.mail.demon.net> This topic generated quite a discussion! I'd like to narrow it down and check if we've reached a consensus: > From: Mark Jefferys > > % since "if( @arr= (getpwent)[0,2] )" never worked in perl4 anyway. > > Note that this has worked in perl5 for a little while, so there may be > a few scripts that this would break. Since it was never well documented I doubt there are that many scripts that use this other than those you've written yourself. Then again Larry may take the view that the possibility makes the risk of undoing the change to high. > % Excerpts from the mail message of Mark Jefferys: > % ) 2. Make list-assignments return undef for zero objects assigned, > % ) and SV_NO for undef-only assignments. This allows: > % ) > % ) if (defined(@arr = func())) ... > % ) 3. Make list-assignments normal, but have list-slices cut off > % ) trailing undefs, so > % ) (getpwent)[0,2] > % ) returns an empty list. > % ) A. This will cause code which relies on list-slices > % ) returning a fixed number of items to break. Sub calls > % ) are a likely source of problems: > % ) func1((func2())[1,3], ++$cnt); > % ) # Expects $cnt in 3rd place, not last. > % ) 7. Make a list-slice of an empty list be empty. This is an example > % ) of minimal intervention: make sure empty lists stay empty, and > % ) thus false, but otherwise acts normally. This would be somewhat > % ) easier to implement for tied arrays (and hashes). > % > % This raises the problem of (A) again. > > Granted, although far less often than the others. I like this one > mostly because the exceptions I can think of are more contrived than > for the others, and pretty easy to work around: > > @arr ? @arr[@idx] : (undef) x @idx > > % ) I kinda like #2 or #7 at the moment; [...] So, to summarise the situation (and check that _I_ understand it :) ... - Currently evaluation in scalar context of the assignment of a list of undefs is regarded as false. - It seems widely agreed that this is not ideal. - This was added to cater for: while (($user, $uid) = (getpwent)[0,2]) { } - It seems widely agreed that this is useful. I also like #7 "Make a list-slice of an empty list be empty". Feels kind'a like null-propagation in SQL expressions. #2 seems too subtle for people to remember to use in their code. Especially where they don't expect to get nulls but might in error conditions. Does anyone have and hard objections or better alternatives? Larry, do you concur? Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-43@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:20:05 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 17:04:06 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa24412; 7 Oct 95 18:03 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA00200 for perldb-interest-real; Sat, 7 Oct 1995 08:48:26 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from adrianp.powertech.no (root@port-58.ppp.powertech.no [194.19.5.58]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA00196 for ; Sat, 7 Oct 1995 08:48:22 -0700 Received: by adrianp.powertech.no id (Debian /\oo/\ Smail3.1.29.1 #29.33); Sat, 7 Oct 95 16:41 MET Message-Id: Date: Sat, 7 Oct 95 16:41 MET From: Adrian Phillips To: perldb-interest@fugue.com help ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-47@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:21:04 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 07:10:04 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id ab00682; 7 Oct 95 8:09 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA10738 for perldb-interest-real; Fri, 6 Oct 1995 23:11:36 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from metronet.com (tye@fohnix.metronet.com [192.245.137.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id XAA10734 for ; Fri, 6 Oct 1995 23:11:33 -0700 Received: by metronet.com id AA05516 (5.67a/IDA1.5hp for perldb-interest@fugue.com); Sat, 7 Oct 1995 01:13:16 -0500 From: Tye McQueen Message-Id: <199510070613.AA05516@metronet.com> Subject: Re: perl bug: array of undefs is falsely false To: Mark Jefferys Date: Sat, 7 Oct 1995 01:13:14 -0500 (CDT) Cc: tye@metronet.com, lwall@scalpel.netlabs.com, Tim.Bunce@ig.co.uk, perldb-interest@fugue.com, mhm@austin.ibm.com, perlbug@perl.com In-Reply-To: from "Mark Jefferys" at Oct 6, 95 01:49:44 pm Content-Transfer-Encoding: 8bit X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 3108 Excerpts from the mail message of Mark Jefferys: ) % I'd also like a way to tell if a key is present in a hash, ie. [...] ) Or how about "exists $hash{$key}"? I don't think it would take too ) much effort to add that. :-) No fair adding it my copy while I'm not looking! :) ) % It also means defined() is required above but if you try to ) % make a habit of using defined() you'll get bit because ) % while( defined( @arr= (getpwent)[0,2] ) ) { ) % will never end. ) ) I don't follow this. If Perl cannot deduce what the user wants, then ) it will have to provide two different ways of writing this: one for ) each interpretation. *All* of the solutions require rewriting if Perl ) gets it wrong (including yours: @arr[] vs. @arr([]))---it's only a ) question of how often it does, and how confusing the exceptions are. You're right, that disadvantage isn't specific to this case. I still like the idea of having the option to slice without creating undefs for out-of-bounds subscripts. ) % [Why do I keep thinking we need a new value, NoneSuch, which is ) % used for things that have never been set (such as out-of-bounds ) % elements)? [...] ) There's also a good reason for having "exists $hash{$key}" not simply ) get $hash{$key} and test it: I wasn't suggesting that. Since a tied hash would use a Perl "sub EXISTS", it would have _no way_ of returning NoneSuch. Putting NoneSuch on the Perl stack to be returned would transform it into undef. ) Maybe... It would be nice to fix the spurious warnings. But there ) will always be potential problems with it: I'm sure someone will ask ) why ) $save = oper(); ) if (exists $save) ... ) doesn't act like ) if (exists oper()) ... Any such people already can't handle why sub notdef { undef; } print "TRUE\n" if defined(¬def); prints TRUE. Now that exists() exists, we've become a bit inconsistant here. Ignoring backward compatibility: defined() should only check an expression value so defined(&func) should be the same as defined($x=&func). To see if a function exists, you should use exists(&func). exists() does not take an expression or value as an argument so saving some value in $save=oper() doesn't effect what exists($save) returns (except when doing so creates $save). Now silly stuff on how exists() and delete() should be extended: delete(&func) would be used to make the next &func re-autoload. undef(&func) would keep its current behavior, to make &func() an error (leave &func cached as if autoloading failed). $x= "global"; { my $x= "my"; print "($x) "; delete($x); # Remove lexical from scope, a compile-time op print "<$x> "; delete($x); # Delete $x from symbol table print "[$x]\n"; } # This would print "(my) []". Notice how my mind likes to deny that it is "x" in the symbol table and not "$x"? But NoneSuch would allow us to hide that fact. ;> -- Tye McQueen tye@metronet.com || tye@doober.usu.edu Nothing is obvious unless you are overlooking something http://www.metronet.com/~tye/ (scripts, links, nothing fancy) ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28310-47@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:22:12 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 12:44:47 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa29273; 7 Oct 95 13:44 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id ab12616; 7 Oct 95 2:02 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA09629 for perldb-interest-real; Fri, 6 Oct 1995 17:03:29 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from watson.ibm.com (watson.ibm.com [129.34.139.4]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA09625 for ; Fri, 6 Oct 1995 17:03:28 -0700 Received: from WATSON by watson.ibm.com (IBM VM SMTP V2R3) with BSMTP id 8773; Fri, 06 Oct 95 20:05:32 EDT Received: from YKTVMV by watson.vnet.ibm.com with "VAGENT.V1.02 on VAGENT2" id 6234; Fri, 6 Oct 1995 20:05:32 EDT Received: from lusitania.watson.ibm.com by yktvmv.watson.ibm.com (IBM VM SMTP V2Rx) with TCP; Fri, 06 Oct 95 20:05:31 EDT Received: from localhost by lusitania.watson.ibm.com (AIX 3.2/UCB 5.64/950830) id AA01390; Fri, 6 Oct 1995 20:05:11 -0400 Message-Id: <9510070005.AA01390@lusitania.watson.ibm.com> X-Mailer: exmh version 1.6.2 7/18/95 To: perldb-interest@fugue.com Subject: Protocol for Nullsv? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 06 Oct 1995 20:05:11 -0400 From: Chet Murthy I have read in the Perl documentation that it is bad style to return Nullsv -- that one should instead return &sv_undef. I have also noticed that in the Oracle DBD driver (dbdimp.c) to be exact, Nullav is returned all over the place. Is this intentional, or merely accidental? That is, can one in fact use Nullsv (or Nullav) freely? Thanks, --chet-- ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-55@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:23:10 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 09:39:27 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa10847; 7 Oct 95 10:38 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA10956 for perldb-interest-real; Sat, 7 Oct 1995 01:40:32 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from mox.perl.com (Warm-Fuzzies@perl.com [199.45.129.30]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA10952 for ; Sat, 7 Oct 1995 01:40:28 -0700 Received: from localhost.perl.com (Warm-Fuzzies@localhost.perl.com [127.0.0.1]) by mox.perl.com (8.6.12/8.6.12) with SMTP id CAA07863; Sat, 7 Oct 1995 02:42:05 -0600 To: Tye McQueen cc: Mark Jefferys , lwall@scalpel.netlabs.com, Tim.Bunce@ig.co.uk, perldb-interest@fugue.com, mhm@austin.ibm.com, perlbug@perl.com, tchrist@mox.perl.com Subject: Re: perl bug: array of undefs is falsely false In-reply-to: Tye McQueen's message <199510070613.AA05516@metronet.com> of Sat, 07 Oct 95 01:13:14 CDT. References: <199510070613.AA05516@metronet.com> Date: Sat, 07 Oct 95 02:42:04 MDT Message-ID: <7860.813055324@mox> From: Tom Christiansen > defined() should only check an expression value so defined(&func) > should be the same as defined($x=&func). You'd break a lot of code. I and surely many others are used to defined &f # function existence (modulo AUTOLOAD || inheritance) vs defined f() # first call function, then check retval I have noticed that all this delete() vs splice() vs undef() stuff is a tad inconsistent or hard to follow for some folks, and I guess I don't blame them too much.. --tom ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-61@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:24:24 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 20:31:56 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ai27174; 7 Oct 95 21:31 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <03464-0@frigate.doc.ic.ac.uk>; Fri, 6 Oct 1995 21:30:29 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA08272 for perldb-interest-real; Fri, 6 Oct 1995 11:50:26 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from skye.as.utexas.edu (skye.as.utexas.edu [128.83.129.233]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA08268 for ; Fri, 6 Oct 1995 11:50:22 -0700 Received: by skye.as.utexas.edu (5.x/SMI-SVR4) id AA05069; Fri, 6 Oct 1995 13:49:45 -0500 Date: Fri, 6 Oct 1995 13:49:44 -0500 (CDT) From: Mark Jefferys To: Tye McQueen Cc: lwall@scalpel.netlabs.com, Tim.Bunce@ig.co.uk, perldb-interest@fugue.com, mhm@austin.ibm.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: <199510060428.AA20463@metronet.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 5 Oct 1995, Tye McQueen wrote: % How about: % % @arr[@sub] for map { $arr[$_] } @sub % @arr([@sub]) for map { $_ < $[ || $#arr < $_ ? () : $arr[$_] } @sub % @hash{@sub} for map { $arr{$_} } @sub % @hash({@sub}) for map { $_ in keys(%hash) ? $hash{$_} : () } @sub % % where "$_ in keys(%hash)" is pseudo Perl for whether hv_fetch() % returned 0. [I don't think this adds ambiguity to the parsing, % but I certainly could be wrong.] % % That is, @arr[@sub] and @hash{@sub}, stay the same. @arr([@sub]) % does not "create" undefs as place holders for out-of-bounds % elements. @hash({@sub}) does not "create" undefs as place holders % for non-existant keys. % % So we would now have to do: % % if( @arr= (getpwent)([0,2]) ) % ^ ^ % since "if( @arr= (getpwent)[0,2] )" never worked in perl4 anyway. Note that this has worked in perl5 for a little while, so there may be a few scripts that this would break. % I'd also like a way to tell if a key is present in a hash, ie. ... % Perhaps a new boolean: "$key in %hash" or "in(%hash,$key)". % Perhaps "keys(%hash,@keys)" could return the elements of @keys % that are keys of %hash: "map { $_ in %hash ? $_ : () } @keys". Or how about "exists $hash{$key}"? I don't think it would take too much effort to add that. :-) % ) 2. Make list-assignments return undef for zero objects assigned, % ) and SV_NO for undef-only assignments. This allows: % ) if (defined(@arr = func())) ... % % This doesn't allow you, in (func())[0,2], to distinguish whether % func() returned () or (undef,"foo",undef,"bar"). Very true. % It also means defined() is required above but if you try to % make a habit of using defined() you'll get bit because % while( defined( @arr= (getpwent)[0,2] ) ) { % will never end. I don't follow this. If Perl cannot deduce what the user wants, then it will have to provide two different ways of writing this: one for each interpretation. *All* of the solutions require rewriting if Perl gets it wrong (including yours: @arr[] vs. @arr([]))---it's only a question of how often it does, and how confusing the exceptions are. % ) 3. Make list-assignments normal, but have list-slices cut off % ) trailing undefs, so % ) (getpwent)[0,2] % ) returns an empty list. % ) A. This will cause code which relies on list-slices % ) returning a fixed number of items to break. Sub calls % ) are a likely source of problems: % ) func1((func2())[1,3], ++$cnt); % ) # Expects $cnt in 3rd place, not last. % ) B. If any of the elements sliced is magic, we either have to % ) get its value, which wastes time and can cause problems, % ) or *not* get its value, which can fail. % % I don't understand this last one (B). Annoying technical issue. GMagical scalars cannot simply be tested for undefs---they don't have a meaningful value unless mg_get is called on them first. Doing so will waste time, since mg_get will probably be called again when the value is used. Also, if I remember correctly, I tried to fix this in the alpha, and ran into some sort of difficulty with getting the value correctly. % ) 7. Make a list-slice of an empty list be empty. This is an example % ) of minimal intervention: make sure empty lists stay empty, and % ) thus false, but otherwise acts normally. This would be somewhat % ) easier to implement for tied arrays (and hashes). % % This raises the problem of (A) again. Granted, although far less often than the others. I like this one mostly because the exceptions I can think of are more contrived than for the others, and pretty easy to work around: @arr ? @arr[@idx] : (undef) x @idx % ) 8. Make array assignments of slices magical, returning false if the % ) original list was empty (or something). A really do-what-I-want % ) solution. % % But will still surprise people with % % while( @ids= (getpwent)[2,3] ) { % if( $ids[0] == $uid && $ids[1] == $gid ) { % last; % } % } % if( @ids ) { % # We found a match! % } % % because we'll always "find a match". Also, this won't work if there is a sub-call separating the slice from the assignment: sub zoop { (getpwent)[2,3] } while (@ids = zoop()) ... % If it can be implemented relatively easilly, I'd vote for this. % % [Why do I keep thinking we need a new value, NoneSuch, which is % used for things that have never been set (such as out-of-bounds % elements)? exists() would be to NoneSuch as defined() is to % sv_undef. There would be no _direct_ way to set anything to % NoneSuch ("$#arr=5" indirectly sets "$arr[6]" to NoneSuch, % "delete($hash{key})" indirectly sets "$hash{key}" to NoneSuch). % In an assignment, NoneSuch is always transformed to sv_undef. % Would this help the problem of sv_undef causing too many "Use % of uninitialized value" warnings?] Maybe... It would be nice to fix the spurious warnings. But there will always be potential problems with it: I'm sure someone will ask why $save = oper(); ... if (exists $save) ... doesn't act like if (exists oper()) ... There's also a good reason for having "exists $hash{$key}" not simply get $hash{$key} and test it: For some tied arrays, checking exists may be far more efficient than getting the value. I suppose this use could be optimized, though. Mark ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-68@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:26:12 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 16:02:01 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa25008; 7 Oct 95 17:00 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA12723 for perldb-interest-real; Sat, 7 Oct 1995 07:42:43 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA12719 for ; Sat, 7 Oct 1995 07:42:41 -0700 Received: by gw.home.vix.com id AA26779; Sat, 7 Oct 95 07:44:55 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from skimo@localhost) by dns.ufsia.ac.be (8.6.12/8.6.9) id QAA26842; Sat, 7 Oct 1995 16:47:22 +0100 From: Sven Verdoolaege Message-Id: <199510071547.QAA26842@dns.ufsia.ac.be> Subject: Re: Multiple DBD's To: Tim Bunce Date: Sat, 7 Oct 1995 16:47:21 +0100 (MET) Cc: perldb-interest@vix.com In-Reply-To: <9510061927.ad16968@relay-3.mail.demon.net> from "Tim Bunce" at Oct 6, 95 07:16:14 pm X-Stardate: [-31]6403.28 X-Url: http://dns.ufsia.ac.be/~skimo X-Mailer: ELM [version 2.4 PL24 ME8a] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1971 I think you may have missed the point > > > > From: Sven Verdoolaege > > > > For those on the list: i have problems connecting to > > both an mSQL and an Oracle database > > > > > > my script: > > > > #!/u01/home/verdools/bin/perl > > > > use DBI; > > > > $DB='ufpers'; > > > > $ENV{'MSQL_HOME'} = "/usr/local/Minerva"; > > > > $ENV{'MSQL_TCP_PORT'} = 22532; > > > > $dbhD = DBI->connect( 'dns',$DB,'','mSQL') || die("geen database"); > > > > $dbhD->disconnect; > > > > $ENV{'ORACLE_HOME'} = "/u01/home/oracle/product/7.1.4"; > > > > $ENV{'ORAHOME'} = "/u01/home/oracle/product/7.1.4"; > > > > print "before\n"; > > > > $dbhS = DBI->connect( $DB, 'job_verdool_s', 'my password:-)','Oracle' ) || die("geen ; ^^^^^^ > > > > print "after\n"; > > > > $dbhS->disconnect; > > > > > > > > my output: > > > > before > > > > rc: 0 what: Unknown MSQL Server Host (ufpers) ^^^^ > > > > Database handle DBI::db=HASH(0x1400fff18) DESTROY ignored - never set up at /u0. > > > > geen database at ./cp2 line 11. > > I think the "DESTROY ignored - never set up" warning may be a bug in > the mSQL driver due to clearing DBIc_IMPSET too early. It's only > a warning. The connect failed for the reason shown. > > > > > if i comment the connect and disconnect lines from mSQL, i get > > > > before > > > > after > > > > > > > > if i put the oracle stuff before the mSQL stuff, i get > > > > before > > > > after > > > > Memory fault - core dumped > > I'd need to see the core dump but only if you still get it after > upgrading to the latest DBI and DBD::Oracle. i guess i forgot to mention i use DBI-0.63 DBD-Oracle-0.25 DBD-mSQL-0.60pl5 I believe these are the most recent ones (apart from pl6 which didn't change much) In what way would you like to see the core dump ? Note that i'm not very familiar with OSF1 alpha. I'm more of a linux user. Sven ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-70@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:26:44 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 17:04:01 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa28073; 7 Oct 95 18:03 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA00265 for perldb-interest-real; Sat, 7 Oct 1995 09:05:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from anna.mind.de (anna.mind.de [193.101.72.227]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA00259 for ; Sat, 7 Oct 1995 09:05:43 -0700 Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id RAA10882; Sat, 7 Oct 1995 17:06:31 +0100 Date: Sat, 7 Oct 1995 17:06:31 +0100 Message-Id: <199510071606.RAA10882@anna.mind.de> From: Andreas Koenig To: adrianp@powertech.no CC: perldb-interest@fugue.com In-reply-to: Reply-to: andreas.koenig@mind.de >>>>> "Adrian" == Adrian Phillips writes: Adrian> help perldb-interest in a human-driven mailing list. You subscribe and unsubscribe by sending requests to perldb-interest-request@fugue.com and let Ted _Mellon_ (sp?) some time to proceed. andreas ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-72@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:27:07 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 17:32:00 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa12055; 7 Oct 95 18:30 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA00345 for perldb-interest-real; Sat, 7 Oct 1995 09:29:19 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from adrianp.powertech.no (root@port-23.ppp.powertech.no [194.19.5.23]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA00341 for ; Sat, 7 Oct 1995 09:29:14 -0700 Received: by adrianp.powertech.no id (Debian /\oo/\ Smail3.1.29.1 #29.33); Sat, 7 Oct 95 17:26 MET Message-Id: Date: Sat, 7 Oct 95 17:26 MET From: Adrian Phillips To: perldb-interest@fugue.com In-reply-to: <199510071606.RAA10882@anna.mind.de> (message from Andreas Koenig on Sat, 7 Oct 1995 17:06:31 +0100) >>>>> "Andreas" == Andreas Koenig writes: >>>>> "Adrian" == Adrian Phillips writes: Adrian> help Andreas> perldb-interest in a human-driven mailing list. You Andreas> subscribe and unsubscribe by sending requests to Andreas> perldb-interest-request@fugue.com and let Ted _Mellon_ Andreas> (sp?) some time to proceed. Andreas> andreas Yes ? - (brain in gear this time). Is this a perldb developers only discussion group or for any and all interested in the development of perl=>db interfaces ? Thanks, Adrian ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-78@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:28:43 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 20:33:07 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ar27174; 7 Oct 95 21:32 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <05537-0@frigate.doc.ic.ac.uk>; Fri, 6 Oct 1995 22:30:58 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA08544 for perldb-interest-real; Fri, 6 Oct 1995 13:10:49 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA08540 for ; Fri, 6 Oct 1995 13:10:45 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ab17288; 6 Oct 95 21:09 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <17955-0@ignite.demon.co.uk>; Fri, 6 Oct 1995 21:07:28 +0100 To: perldb-interest@fugue.com, zobair@marque.mscs.mu.edu Subject: Re: Help! oraperl: oracle error on ora_login() Date: Fri, 6 Oct 1995 21:07:28 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510062109.ab17288@relay-3.mail.demon.net> > From: Saima Zobair > > Hi! > > I am trying to connect oracle and the webserver using oraperl2.4 > I am using oracle7. When I run the oraperl script on the > command line it executes fine and the database query runs perfectly. > But when I try to run the same oraperl file from /cgi through > netscape... it gives a an error on the &oralogin(). On printing > the $ora_errstr I get: > > Error while trying to retrieve text for error ORA-02701 > > Can someone please help me with this ? > This is strictly off-topic for this mailing list. Remember that CGI scripts typically execute as user 'nobody' and have very few environment variables set up. You will need to explicilty setup environment variables such as ORACLE_HOME. For any further problems I suggest you post messages to comp.lang.perl.misc and comp.infosystems.www.authoring.cgi. Tim. p.s. Please don't post non-admin messages to perldb-interest-request. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-83@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:29:43 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 07:51:25 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa15356; 8 Oct 95 8:51 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA01731 for perldb-interest-real; Sun, 8 Oct 1995 00:07:50 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA01727 for ; Sun, 8 Oct 1995 00:07:47 -0700 Received: by gw.home.vix.com id AA26415; Sun, 8 Oct 95 00:10:04 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id CAA23614; Sun, 8 Oct 1995 02:09:55 -0500 Date: Sun, 8 Oct 1995 02:09:55 -0500 (CDT) From: Ben Lindstrom To: andreas.koenig@mind.de Cc: DBI perl Subject: Re: (fwd) C + ObjC + Perl == Problems. (fwd) In-Reply-To: <199510080704.IAA13730@anna.mind.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Sun, 8 Oct 1995, Andreas Koenig wrote: > >>>>> "Ben" == Ben Lindstrom writes: > [..] > Ben> This is what I'm doing... > > Ben> cc -c -DUSE_NEXT_CTYPE -O dbdimp.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE > Ben> cc -c -DUSE_NEXT_CTYPE -O QBase.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE > Ben> ld -o QBase.so -lNeXT_s -lMedia_s QBase.o dbdimp.o > [..] > You most definitely have to link statically. I have done something > similar quite a long time ago, and I'm not sure anymore, what did the > trick: either build the extension statically or link the libNeXT into > perl. Or both :) > I thought that was what I was doing with: ld -o QBase.so -lNeXT_s -lMedia_s [...] Linking in the NeXT development libraries. Maybe I have something slightly wrong. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-86@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:30:32 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 20:33:10 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aw27174; 7 Oct 95 21:32 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <07748-0@frigate.doc.ic.ac.uk>; Fri, 6 Oct 1995 23:29:35 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA08861 for perldb-interest-real; Fri, 6 Oct 1995 14:19:16 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gwa.ericsson.com (gwa.ericsson.com [198.215.127.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA08857 for ; Fri, 6 Oct 1995 14:19:13 -0700 Received: from mr2.exu.ericsson.se (mr2.exu.ericsson.com [138.85.147.12]) by gwa.ericsson.com (8.6.10/8.6.10) with ESMTP id QAA28845 for ; Fri, 6 Oct 1995 16:20:49 -0500 Received: from eur.ericsson.se (dublin.eur.ericsson.se [153.88.1.19]) by mr2.exu.ericsson.se (8.6.8/NAHUB-MR1.1) with ESMTP id QAA06738 for ; Fri, 6 Oct 1995 16:20:48 -0500 Received: from ops by eur.ericsson.se (8.6.12/EUR-Dublin) id OAA13798; Fri, 6 Oct 1995 14:20:49 -0700 Received: from shawnee.raynet.com by ops (4.1/SMI-4.1) id AA15031; Fri, 6 Oct 95 14:22:24 PDT Received: from localhost by shawnee.raynet.com (4.1/SMI-4.1) id AA21450; Fri, 6 Oct 95 14:22:22 PDT Message-Id: <9510062122.AA21450@shawnee.raynet.com> From: Yary Hluchan To: perldb-interest@fugue.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: X-Mailer: exmh version 1.4.1 7/21/94 Date: Fri, 06 Oct 1995 14:22:21 -0700 On Fri, 6 Oct 1995 13:49:44 -0500 (CDT) Mark Jefferys writ: ] On Thu, 5 Oct 1995, Tye McQueen wrote: ] % ] % [Why do I keep thinking we need a new value, NoneSuch, which is ] % used for things that have never been set (such as out-of-bounds ] % elements)? exists() would be to NoneSuch as defined() is to ] % sv_undef. There would be no _direct_ way to set anything to ] % NoneSuch ("$#arr=5" indirectly sets "$arr[6]" to NoneSuch, ] % "delete($hash{key})" indirectly sets "$hash{key}" to NoneSuch). ] % In an assignment, NoneSuch is always transformed to sv_undef. ] % Would this help the problem of sv_undef causing too many "Use ] % of uninitialized value" warnings?] I like this solution best, or a modification of it, though I'm sure it'll make it'll make some implementors nauseous... I'd like to see another "null" value akin to "undef" for "not exists" ] Maybe... It would be nice to fix the spurious warnings. But there ] will always be potential problems with it: I'm sure someone will ask ] why ] ] $save = oper(); ] ... ] if (exists $save) ... ] ] doesn't act like ] ] if (exists oper()) ... If "not exists" isn't cast to "undef" on assignment, then these work the same. ] There's also a good reason for having "exists $hash{$key}" not simply ] get $hash{$key} and test it: For some tied arrays, checking exists ] may be far more efficient than getting the value. I suppose this use ] could be optimized, though. Tied arrays have to impelent an "EXISTS" method, so that's where the optimization would be, right? There's no reason then why "exists $hash{$key}" must get $hash{$key} and test it. Four valued logic, let's do it. yary "constistency is a virtue" hluchan ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-90@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:31:09 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 23:03:55 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa17278; 9 Oct 95 0:02 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA04124 for perldb-interest-real; Sun, 8 Oct 1995 14:42:50 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA04120 for ; Sun, 8 Oct 1995 14:42:49 -0700 Received: by gw.home.vix.com id AA27230; Sun, 8 Oct 95 14:45:05 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from root@localhost) by tom.compulink.co.uk (8.6.9/8.6.9) id WAA22717 for perldb-interest@vix.com; Sun, 8 Oct 1995 22:45:03 +0100 Date: Sun, 8 Oct 95 22:44 BST-1 From: Huseyin Ramazanoglu Subject: Unsubscribe To: perldb-interest@vix.com Cc: huseyinr@cix.compulink.co.uk Reply-To: huseyinr@cix.compulink.co.uk Message-Id: unsubscribe huseyinr@cix.compulink.co.uk ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-92@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:31:49 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 20:33:13 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id bc27174; 7 Oct 95 21:32 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <09761-0@frigate.doc.ic.ac.uk>; Sat, 7 Oct 1995 00:19:09 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA09130 for perldb-interest-real; Fri, 6 Oct 1995 15:04:57 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA09126 for ; Fri, 6 Oct 1995 15:04:55 -0700 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id SAA24778 for ; Fri, 6 Oct 1995 18:07:03 -0400 Message-Id: <199510062207.SAA24778@pilot.njin.net> To: perldb-interest@fugue.com Subject: Apologies Date: Fri, 06 Oct 1995 18:07:00 -0400 From: "Patricia H. Ju" Sorry for the perl question. I did not realize this list was mainly adminnish. Patricia ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-93@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:31:51 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 23:14:45 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa22035; 9 Oct 95 0:13 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA04169 for perldb-interest-real; Sun, 8 Oct 1995 14:57:26 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA04165 for ; Sun, 8 Oct 1995 14:57:24 -0700 Received: by gw.home.vix.com id AA28491; Sun, 8 Oct 95 14:59:41 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from 128.230.1.225 (npac-dial-225.syr.edu [128.230.1.225]) by newstand.syr.edu (8.6.12/8.6.12) with SMTP id RAA21233; Sun, 8 Oct 1995 17:55:57 -0400 Message-Id: <199510082155.RAA21233@newstand.syr.edu> Date: Sun, 08 Oct 95 18:02:05 0000 From: Gang Cheng Organization: NPAC X-Mailer: Mozilla 1.1N (Macintosh; I; 68K) Mime-Version: 1.0 Newsgroups: syr.general,comp.databases,comp.databases.oracle,comp.arch To: perldb-interest@vix.com, sp-discussion@mcs.anl.gov, www-vrml@wired.com Subject: A WWW archive system for mailing-list and USENET newsgroups X-Url: http://kayak.npac.syr.edu:1963/cgi-bin/news/showbody?bid=42000095&word=~1 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii We at Northeast Parallel Architecture Center (NPAC), Syracue University has developed an archive system for searching/reading USENET newsgrou ps, mailing lists and personal mailbox from WWW. An Oracle database server is used to store/manage mails and two search/navigation interfaces accesible by any WWW browser to the archive are provided: one is an advanced search interface allowing queries with various options such as query by mail header, by date, by subject (keywords), by sender. The other is a Hypermail-like navigation interface for users familiar with Hypermail. We offer free access to this archive for the Internet Internet community. Check the URL . The whole archive is automatically updated in every 30 minutes. Currently the following newsgroups (total 40) are archived in the database: comp.arch comp.databases comp.databases.oracle comp.parallel comp.parallel.mpi comp.parallel.pvm comp.lang.java alt.www.hotjava comp.infosystems comp.infosystems.www.authoring.cgi comp.infosystems.www.announce comp.infosystems.www.authoring.html comp.infosystems.www.authoring.images comp.infosystems.www.authoring.misc comp.infosystems.www.misc comp.infosystems.www.advocacy comp.infosystems.www.browsers.mac comp.infosystems.www.browsers.ms-windows comp.infosystems.www.browsers.x comp.infosystems.www.browsers.misc comp.infosystems.www.servers.ms-windows comp.infosystems.www.servers.unix comp.infosystems.www.servers.mac comp.infosystems.www.servers.misc comp.graphics comp.graphics.visualization comp.lang.c++ comp.multimedia comp.security.misc comp.sources.unix comp.sources.x comp.unix.admin comp.unix.aix comp.unix.misc comp.unix.programmer comp.unix.solaris alt.comp.compression comp.databases.object comp.lang.perl.misc syr.general The following mailing lists are archived: java-interest java-porting java-announce hotjava-interest sp-discussion www-vrml perldb-interest-request@fugue.com This system is still under development and more functions are planned to add to this system. Please send your comments to Gang Cheng (gcheng@npac.syr.edu . Thanks. --- Gang ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-99@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:33:38 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 23:54:10 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa06770; 9 Oct 95 0:53 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA04357 for perldb-interest-real; Sun, 8 Oct 1995 15:44:10 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from skye.as.utexas.edu (skye.as.utexas.edu [128.83.129.233]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA04353 for ; Sun, 8 Oct 1995 15:44:07 -0700 Received: by skye.as.utexas.edu (5.x/SMI-SVR4) id AA06137; Sun, 8 Oct 1995 17:43:41 -0500 Date: Sun, 8 Oct 1995 17:43:40 -0500 (CDT) From: Mark Jefferys To: Tye McQueen Cc: perldb-interest@fugue.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: <199510070613.AA05516@metronet.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Sat, 7 Oct 1995, Tye McQueen wrote: % Excerpts from the mail message of Mark Jefferys: % ) % I'd also like a way to tell if a key is present in a hash, ie. % [...] % ) Or how about "exists $hash{$key}"? I don't think it would take too % ) much effort to add that. :-) % % No fair adding it my copy while I'm not looking! :) Oops. % defined() should only check an expression value so defined(&func) % should be the same as defined($x=&func). This is mostly the confusion between &func as a way of referring to &func, and as a way of calling it. A good reason to deprecate the latter, IMHO. % To see if a function exists, you should use exists(&func). % exists() does not take an expression or value as an argument % so saving some value in $save=oper() doesn't effect what % exists($save) returns (except when doing so creates $save). OK. I like that, actually. % Now silly stuff on how exists() and delete() should be extended: % % delete(&func) would be used to make the next &func re-autoload. % undef(&func) would keep its current behavior, to make &func() an % error (leave &func cached as if autoloading failed). % % $x= "global"; % { my $x= "my"; % print "($x) "; % delete($x); # Remove lexical from scope, a compile-time op % print "<$x> "; % delete($x); # Delete $x from symbol table % print "[$x]\n"; % } % # This would print "(my) []". That's either really clever or really gross---I can't quite decide. I think I like it. % Notice how my mind likes to deny that it is "x" in the symbol % table and not "$x"? But NoneSuch would allow us to hide that % fact. ;> Perhaps. I don't see any reason why we need to add an actual NoneSuch scalar to make this work. IDEA: One thing that could be done---and this would cut down on spurious warnings a lot---is to create the SVs, AVs, and HVs when first accessed, instead of when their containing GVs are created. A warning could be produced if this is done in a non-modification context. I don't know precisely how this would affect performance, though. For example, there would need to be an extra check in pp_sv to see that the accessed scalar was non-zero. On the other hand, Perl wouldn't waste time (and space) creating unused AVs and HVs when someone uses $x without the others. And the current warning checks could be thrown out. Mark ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28311-102@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:34:15 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 01:57:34 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa09225; 9 Oct 95 2:56 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA04892 for perldb-interest-real; Sun, 8 Oct 1995 17:48:05 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA04888 for ; Sun, 8 Oct 1995 17:48:03 -0700 Received: by gw.home.vix.com id AA10168; Sun, 8 Oct 95 17:50:20 -0700 Date: Sun, 8 Oct 95 17:50:20 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from song.iuol.cn.net by iui.iuol.cn.net id aa11795; 9 Oct 95 8:41 PST Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit From: lhsong@iuol.cn.net Subject: Unsubscribe To: perldb-interest@vix.com X-Mailer: SPRY Mail Version: 04.00.06.17 Message-Id: <9510090841.aa11795@iui.iuol.cn.net> unsubscribe lhsong@iuol.cn.net ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-103@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:34:21 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 20:33:17 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id bg27174; 7 Oct 95 21:32 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <11763-0@frigate.doc.ic.ac.uk>; Sat, 7 Oct 1995 01:36:47 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA09521 for perldb-interest-real; Fri, 6 Oct 1995 16:43:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gsb.crimson.com (gsb.crimson.com [128.83.193.48]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA09517 for ; Fri, 6 Oct 1995 16:43:04 -0700 Received: from rhine.ece.utexas.edu by gsb.crimson.com (AIX 3.2/UCB 5.64/4.03) id AA35167; Fri, 6 Oct 1995 18:45:16 -0500 Date: Fri, 6 Oct 1995 18:42:13 -0500 (CDT) From: Stephen Cheung X-Sender: stephenc@rhine.ece.utexas.edu To: perldb-interest@fugue.com Subject: unsubscribe Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-108@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:35:13 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 07 Oct 95 20:33:19 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id bh27174; 7 Oct 95 21:32 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <11857-0@frigate.doc.ic.ac.uk>; Sat, 7 Oct 1995 01:38:04 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA09444 for perldb-interest-real; Fri, 6 Oct 1995 16:30:01 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from mox.perl.com (Warm-Fuzzies@perl.com [199.45.129.30]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA09437 for ; Fri, 6 Oct 1995 16:29:53 -0700 Received: from localhost.perl.com (Warm-Fuzzies@localhost.perl.com [127.0.0.1]) by mox.perl.com (8.6.12/8.6.12) with SMTP id RAA00259; Fri, 6 Oct 1995 17:30:35 -0600 To: Mark Jefferys cc: Tim Bunce , tye@metronet.com, lwall@scalpel.netlabs.com, perldb-interest@fugue.com, perlbug@perl.com, tchrist@mox.perl.com Subject: Re: perl bug: array of undefs is falsely false In-reply-to: Mark Jefferys's message of Fri, 06 Oct 95 16:16:52 CDT. References: Date: Fri, 06 Oct 95 17:30:34 MDT Message-ID: <256.813022234@mox> From: Tom Christiansen > %> From: Mark Jefferys > %> > %>% since "if( @arr= (getpwent)[0,2] )" never worked in perl4 anyway. > %> > %> Note that this has worked in perl5 for a little while, so there may be > %> a few scripts that this would break. Yes, I remember getting Larry to fix it, as I'm the one who submitted that bug. I do see why you're upset by it, though. --tom ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-145@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:41:40 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 02:53:19 GMT Received: from [204.254.239.2] by punt-1.mail.demon.net id aa24954; 8 Oct 95 3:51 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA01280 for perldb-interest-real; Sat, 7 Oct 1995 18:39:02 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from mailgw.claircom.com (mailgw.claircom.com [199.5.241.51]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id SAA01276 for ; Sat, 7 Oct 1995 18:39:00 -0700 Received: from nimo.claircom.com by mailgw.claircom.com with smtp (Smail3.1.26.7 #2) id m0t1kjr-0004tjC; Sat, 7 Oct 95 18:41 PDT Received: from airone.claircom.com by nimo.claircom.com with smtp (Smail3.1.26.7 #2) id m0t1kjq-0005V1C; Sat, 7 Oct 95 18:41 PDT Received: from soc4.airone.claircom.com by airone.claircom.com (5.0/SMI-SVR4) id AA13033; Sat, 7 Oct 1995 18:39:27 -0700 Received: by soc4.airone.claircom.com (4.1/SMI-4.1) id AA02763; Sat, 7 Oct 95 18:37:23 PDT Date: Sat, 7 Oct 95 18:37:23 PDT From: William Warner Message-Id: <9510080137.AA02763@soc4.airone.claircom.com> To: perldb-interest@fugue.com Subject: subscribe me content-length: 13 subscribe me ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-170@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:46:53 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 07:38:39 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa20469; 8 Oct 95 8:38 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA01655 for perldb-interest-real; Sat, 7 Oct 1995 23:42:12 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA01651 for ; Sat, 7 Oct 1995 23:42:11 -0700 Received: by gw.home.vix.com id AA25041; Sat, 7 Oct 95 23:44:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id BAA23083; Sun, 8 Oct 1995 01:44:25 -0500 Date: Sun, 8 Oct 1995 01:44:24 -0500 (CDT) From: Ben Lindstrom To: DBI perl Cc: quickbase@gun.com Subject: (fwd) C + ObjC + Perl == Problems. (fwd) Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NOTE: I also posted this to next.programmers and perl.misc newsgroups I'm writing a loadable module for Perl DBI for QuickBase...And I'm having problems resolve some of the Objective-Cisms. This is what I'm doing... cc -c -DUSE_NEXT_CTYPE -O dbdimp.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE cc -c -DUSE_NEXT_CTYPE -O QBase.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE ld -o QBase.so -lNeXT_s -lMedia_s QBase.o dbdimp.o The ld returns: ld: Undefined symbols:ld: _stack_sp _markstack_ptr _stack_base _croak _sv_mortalcopy _dirty _perl_get_sv _Xpv _sv_2bool _sv_setiv _sv_setpv _newSViv _sv_2mortal _na _sv_2pv _sv_no _sv_yes _newSVpv _sv_undef _warn _sv_newmortal _op _dowantarray _stack_max _stack_grow _sv_2iv _newSV _newRV _newXS _printf _objc_msgSend _hv_store _sprintf _fprintf _hv_fetch _sv_setsv .objc_class_name_QuickBase Now, there are some Perl stuff in there that I don't care about because perl5 deals with it when it loads it...But .objc_class_name_QuickBase and _objc_msgSend are ObjCism that I need to get rid of... Is there ANY way of doing this in NeXT? Or what are my options? ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-172@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:47:18 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 07:52:01 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa24814; 8 Oct 95 8:51 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA01699 for perldb-interest-real; Sun, 8 Oct 1995 00:03:29 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA01695 for ; Sun, 8 Oct 1995 00:03:27 -0700 Received: by gw.home.vix.com id AA26211; Sun, 8 Oct 95 00:05:40 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id IAA13730; Sun, 8 Oct 1995 08:04:23 +0100 Date: Sun, 8 Oct 1995 08:04:23 +0100 Message-Id: <199510080704.IAA13730@anna.mind.de> From: Andreas Koenig To: mouring@netnet.net Cc: perldb-interest@vix.com, quickbase@gun.com In-Reply-To: (message from Ben Lindstrom on Sun, 8 Oct 1995 01:44:24 -0500 (CDT)) Subject: Re: (fwd) C + ObjC + Perl == Problems. (fwd) Reply-To: andreas.koenig@mind.de >>>>> "Ben" == Ben Lindstrom writes: Ben> NOTE: I also posted this to next.programmers and perl.misc newsgroups Ben> I'm writing a loadable module for Perl DBI for QuickBase...And I'm having Ben> problems resolve some of the Objective-Cisms. Ben> This is what I'm doing... Ben> cc -c -DUSE_NEXT_CTYPE -O dbdimp.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE Ben> cc -c -DUSE_NEXT_CTYPE -O QBase.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE Ben> ld -o QBase.so -lNeXT_s -lMedia_s QBase.o dbdimp.o Ben> The ld returns: [...] Ben> .objc_class_name_QuickBase Ben> Now, there are some Perl stuff in there that I don't care about because perl5 Ben> deals with it when it loads it...But Ben> .objc_class_name_QuickBase and _objc_msgSend are ObjCism that I need to Ben> get rid of... Ben> Is there ANY way of doing this in NeXT? Or what are my options? You most definitely have to link statically. I have done something similar quite a long time ago, and I'm not sure anymore, what did the trick: either build the extension statically or link the libNeXT into perl. Or both :) Hope that helps, andreas ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-179@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:48:37 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 08 Oct 95 09:42:41 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa05516; 8 Oct 95 10:42 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA02043 for perldb-interest-real; Sun, 8 Oct 1995 01:29:41 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA02039 for ; Sun, 8 Oct 1995 01:29:40 -0700 Received: by gw.home.vix.com id AA01671; Sun, 8 Oct 95 01:31:54 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id JAA14024; Sun, 8 Oct 1995 09:30:56 +0100 Date: Sun, 8 Oct 1995 09:30:56 +0100 Message-Id: <199510080830.JAA14024@anna.mind.de> From: Andreas Koenig To: mouring@netnet.net Cc: perldb-interest@vix.com In-Reply-To: (message from Ben Lindstrom on Sun, 8 Oct 1995 02:09:55 -0500 (CDT)) Subject: Re: (fwd) C + ObjC + Perl == Problems. (fwd) Reply-To: andreas.koenig@mind.de >>>>> "Ben" == Ben Lindstrom writes: Ben> On Sun, 8 Oct 1995, Andreas Koenig wrote: >> >>>>> "Ben" == Ben Lindstrom writes: >> Ben> [..] Ben> This is what I'm doing... >> Ben> cc -c -DUSE_NEXT_CTYPE -O dbdimp.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE Ben> cc -c -DUSE_NEXT_CTYPE -O QBase.c -ObjC -I/usr/local/lib/perl5/i386-next/CORE Ben> ld -o QBase.so -lNeXT_s -lMedia_s QBase.o dbdimp.o >> Ben> [..] >> You most definitely have to link statically. I have done something >> similar quite a long time ago, and I'm not sure anymore, what did the >> trick: either build the extension statically or link the libNeXT into >> perl. Or both :) >> Ben> I thought that was what I was doing with: Ben> ld -o QBase.so -lNeXT_s -lMedia_s [...] Nope, you are trying to link libNeXT_s into the .so file which is a dynamically loaded file which refuses to take the library contents because libNeXT_s is a shared library. Try 'make perl' and see what happens with this newly created perl. Ben> Linking in the NeXT development libraries. Ben> Maybe I have something slightly wrong. As I said, I'm not quite firm in what I say, but I know, I did see NeXT objects called from a perl script on my monitor. andreas ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <28312-194@ignite.demon.co.uk>; Mon, 9 Oct 1995 07:51:13 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 00:09:39 GMT Received: from [204.254.239.2] by punt-1.mail.demon.net id aa16519; 9 Oct 95 1:09 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA04344 for perldb-interest-real; Sun, 8 Oct 1995 15:41:24 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from skye.as.utexas.edu (skye.as.utexas.edu [128.83.129.233]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA04340 for ; Sun, 8 Oct 1995 15:41:22 -0700 Received: by skye.as.utexas.edu (5.x/SMI-SVR4) id AA06096; Sun, 8 Oct 1995 17:40:50 -0500 Date: Sun, 8 Oct 1995 17:40:49 -0500 (CDT) From: Mark Jefferys To: Larry Wall Cc: Tim Bunce , tye@metronet.com, perldb-interest@fugue.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: <9510062249.AA09086@scalpel.netlabs.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 6 Oct 1995, Larry Wall wrote: % I think that as we move into using prototypes, func1() above would % probably be declared as % % sub func1 ($$$); % % so you wouldn't be able to say % % func1((func2())[1,3], ++$cnt); % % in any event. (Unless I did something Really Strange with argument % processing to make it look inside a list slice.) Assuming someone used prototypes. Also, the same limitation applies to the arguments of a format. Mark ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07679-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 13:24:32 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 11:09:17 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa03862; 9 Oct 95 12:08 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA06994 for perldb-interest-real; Mon, 9 Oct 1995 02:14:39 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from cvrelay.cv.com (cvrelay.CV.COM [130.21.162.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA06990 for ; Mon, 9 Oct 1995 02:14:37 -0700 Received: from dm-server.cv.com. (dm-server.CV.COM [130.21.231.252]) by cvrelay.cv.com (8.6.9/8.6.9) with SMTP id FAA04368 for ; Mon, 9 Oct 1995 05:20:30 -0400 Received: from cobra.cv.com by dm-server.cv.com. (5.0/SMI-SVR4) id AA10807; Mon, 9 Oct 1995 10:13:44 +0100 Errors-To: mkr@cobra Received: by cobra.cv.com (4.1/SMI-4.1) id AA20319; Mon, 9 Oct 95 10:15:50 +0100 Date: Mon, 9 Oct 95 10:15:50 +0100 From: Martin {martlbub} Kraegeloh Message-Id: <9510090915.AA20319@cobra.cv.com> Errors-To: mkr@dm-server.cv.com Reply-To: mkr@dm-server.cv.com Errors-To: mkr@dm-server.cv.com To: perldb-interest@fugue.com content-length: 10 subscribe ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09708-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 16:44:17 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 15:09:56 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa21249; 9 Oct 95 16:09 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa13208; 9 Oct 95 16:08 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA07510 for perldb-interest-real; Mon, 9 Oct 1995 06:26:56 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA07506 for ; Mon, 9 Oct 1995 06:26:54 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ae03980; 9 Oct 95 13:25 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <02772-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 10:59:02 +0100 To: perldb-interest@fugue.com, adrianp@powertech.no Date: Mon, 9 Oct 1995 10:59:02 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510091325.ae03980@relay-3.mail.demon.net> > From: Adrian Phillips > > help > I suspect you may have meant to post that to perldb-interest-request@vix.com thinking that it's an automated server. Actually Ted Lemon runs it by hand. Send a subscription message to perldb-interest-request@vix.com and Ted'll process it for you. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09706-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 16:44:17 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 15:17:39 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa21214; 9 Oct 95 16:16 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA07462 for perldb-interest-real; Mon, 9 Oct 1995 06:07:17 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA07458 for ; Mon, 9 Oct 1995 06:07:13 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id af03980; 9 Oct 95 13:25 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <02910-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 11:20:47 +0100 To: perldb-interest@fugue.com, chet@watson.ibm.com Subject: Re: Protocol for Nullsv? Date: Mon, 9 Oct 1995 11:20:47 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510091325.af03980@relay-3.mail.demon.net> > From: Chet Murthy > > I have read in the Perl documentation that it is bad style to return > Nullsv -- that one should instead return &sv_undef. > In general yes. Either that or put nothing on the stack at all. > I have also noticed that in the Oracle DBD driver (dbdimp.c) to be > exact, Nullav is returned all over the place. > If dbd_st_fetch counts as 'all over the place' then yes. > Is this intentional, or merely accidental? That is, can one in fact > use Nullsv (or Nullav) freely? > Yes (mostly). You can't put NULLs on the perl stack. Generally the XS code does the right thing with NULLs being returned from dbdimp.c. > Thanks, > --chet-- > From: Chet Murthy > > [I'm new to this list, so if this is not the way to present bugs with > DBI, or the drivers, please let me know. --chet--] > This is the way to do it. For bugs mail me and CC to perldb-interest. > There seems to be a bug in the fetch() function in DB2.xs, which, I > think, is also present in the Oracle version. I enclose the diffs for > the DB2 version, which, should be transportable to the Oracle code. Yes, it's a bug. It may also have been carried over to mSQL etc. Thanks for spotting it. > Synopsis: If I call $st->fetch(), which calls dbd_st_fetch, it returns > an AV, which is cast to an SV, and then returned immediately. But it > needs to be wrapped up in a reference before being returned. > > This results in the code below: > > ---------------------------------------------------------------- > #!/u/chet/contrib/bin/perl > > use DBI; > $dbh = DBI->connect('sample'); > $stmt = $ARGV[0]; > $sth = $dbh->prepare($stmt); > print "NUM Params: ",$sth->{'NUM_OF_PARAMS'},"\n"; > $sth->execute(); > while ($ary = $sth->fetch) { > $" = "||"; print "Row: @{$ary}\n"; > } > $sth->finish(); > ---------------------------------------------------------------- > > failing instantly with an "undef" -- that is, the first fetch fails, even > if the query returns something, because the return value of dbd_st_fetch > is lost. If I turn on debugging, I see that DBI::dispatch thinks that > it's been returned an undef (since an AV is not an SV). > > The modification: > > ---------------------------------------------------------------- > --- DB2.xs.~1~ Fri Sep 22 14:46:02 1995 > +++ DB2.xs Fri Oct 6 19:53:05 1995 > @@ -221,7 +221,12 @@ > fetch(sth) > SV * sth > CODE: > - ST(0) = (SV*)dbd_st_fetch(sth); > + AV * av = dbd_st_fetch(sth); > + if (av == Nullav) { > + ST (0) = &sv_undef; > + } else { > + ST(0) = newRV ((SV *)av); > + } > ---------------------------------------------------------------- > > (1) wraps non-NULL AV's in a reference, which is what is returned > > (2) returns &sv_undef if the array is empty -- i.e., if the fetch > failed. > That looks fine. I think the newRV line needs to be: ST(0) = sv_2mortal(newRV((SV *)av)); to avoid leaks. Thanks again. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <10725-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 17:03:04 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 16:02:36 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa13742; 9 Oct 95 17:02 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa17941; 9 Oct 95 16:59 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA07684 for perldb-interest-real; Mon, 9 Oct 1995 07:16:40 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA07680 for ; Mon, 9 Oct 1995 07:16:37 -0700 Received: from fruitbat (fruitbat.mcqueen.com [193.133.160.246]) by fruitbat.mcqueen.com (8.6.12/8.6.12) with SMTP id PAA18670; Mon, 9 Oct 1995 15:14:50 +0100 Message-ID: <30792E59.1833@hermetica.com> Date: Mon, 09 Oct 1995 15:14:49 +0100 From: Alligator Descartes X-Mailer: Mozilla 2.0b1 (X11; I; SunOS 5.3 sun4m) MIME-Version: 1.0 To: Tim Bunce CC: perldb-interest@fugue.com Subject: Re: OraPerl on NT References: <9510091448.aa21879@relay-3.mail.demon.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Tim Bunce wrote: > > If you like, I can ftp my source files to you. Just give me an > > address as to were to put them. > > > If it's not that complicated perhaps you could just write a step-by-step > README.NT (in the same vein as the existing README.oracle7 & README.Solaris2) > That's generally better then just sending me a whole stack of source files. That sounds like a plan. If you want to send me a copy of the text, I'll put it up on the DBI site for other's perusal.... > Either way since this relates to Perl4 and oraperl-2.4 all I'll do is > drop them into the archive. Likewise, you can send the stuff across to me at: ftp://ftp.mcqueen.com/incoming and let me know. > Any chance you could have a go at getting NTperl *5* and DBI + DBD::Oracle > ported? Perversely enough, I'm having a look at this just now.........Completely horrible it is too..... 8-) > Tim. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.mcqueen.com/hermetica ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <11105-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 17:14:30 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 16:08:56 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa20873; 9 Oct 95 17:08 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <04901-0@frigate.doc.ic.ac.uk>; Mon, 9 Oct 1995 16:48:29 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA07631 for perldb-interest-real; Mon, 9 Oct 1995 07:03:23 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA07627 for ; Mon, 9 Oct 1995 07:03:16 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id aa21879; 9 Oct 95 14:48 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <08252-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 14:48:39 +0100 To: Ed_Bornman@ccmail.orl.mmc.com Subject: Re: OraPerl on NT Cc: perldb-interest@fugue.com Date: Mon, 9 Oct 1995 14:48:39 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510091448.aa21879@relay-3.mail.demon.net> > From: Ed_Bornman@ccmail.orl.mmc.com > > Dear Tim; > > > I have Oraperl running at this time on NT. I'm using NT 3.5 and > accessing Oracle on my NT box. Seems to be running great! > > Here is what I did. I used a copy of the NTPerl source and a copy of > the OraPerl 2.4 source. I added the need commands to the ntperl > makefile to add the oracle files. Combined the userstub from ntperl > with the needed changes from the userstub from oraperl. I'm using > Microsoft Visual C++ 2.0 as my compiler. The biggest trick was the > linking. You must have ProC version 1.6 for NT to get the 32 bit > libs. > > If you like, I can ftp my source files to you. Just give me an > address as to were to put them. > If it's not that complicated perhaps you could just write a step-by-step README.NT (in the same vein as the existing README.oracle7 & README.Solaris2) That's generally better then just sending me a whole stack of source files. Either way since this relates to Perl4 and oraperl-2.4 all I'll do is drop them into the archive. Any chance you could have a go at getting NTperl *5* and DBI + DBD::Oracle ported? Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <12564-2@ignite.demon.co.uk>; Mon, 9 Oct 1995 18:50:16 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 17:19:59 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa19447; 9 Oct 95 18:19 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa24982; 9 Oct 95 18:16 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA08044 for perldb-interest-real; Mon, 9 Oct 1995 08:34:29 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA08040 for ; Mon, 9 Oct 1995 08:34:27 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id KAA104512; Mon, 9 Oct 1995 10:36:44 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA24438; Mon, 9 Oct 1995 10:36:44 -0500 Message-Id: <9510091536.AA24438@dax.austin.ibm.com> To: perldb-interest@fugue.com, adrianp@powertech.no In-Reply-To: Your message of Mon, 09 Oct 95 10:59:02 +0100. <9510091325.ae03980@relay-3.mail.demon.net> Date: Mon, 09 Oct 95 10:36:43 -0500 Tim Bunce writes: > > > From: Adrian Phillips > > > > help > > > I suspect you may have meant to post that to perldb-interest-request@vix.com > thinking that it's an automated server. > > Actually Ted Lemon runs it by hand. > > Send a subscription message to perldb-interest-request@vix.com > and Ted'll process it for you. > > Tim. Make that perldb-interest-request@fugue.com ^^^^^ New host. Mike mhm@austin.ibm.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <14169-22@ignite.demon.co.uk>; Mon, 9 Oct 1995 20:57:32 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 19:33:57 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ar03080; 9 Oct 95 20:33 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <09253-0@frigate.doc.ic.ac.uk>; Mon, 9 Oct 1995 20:06:23 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA08388 for perldb-interest-real; Mon, 9 Oct 1995 10:06:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from proxy.netlabs.com (ns.sems.com [192.94.48.47]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA08384 for ; Mon, 9 Oct 1995 10:06:23 -0700 Received: by proxy.netlabs.com (4.1/SMI-4.1) id AA17585; Mon, 9 Oct 95 10:13:30 PDT Received: from vaccine-bb.sems.com(192.190.199.1) by proxy.netlabs.com via smap (V1.3) id sma017583; Mon Oct 9 10:13:00 1995 Received: from scalpel.netlabs.com by netlabs.com (4.1/SMI-4.1) id AA28212; Mon, 9 Oct 95 10:10:01 PDT Received: from localhost.netlabs.com by scalpel.netlabs.com (4.1/SMI-4.1) id AA27196; Mon, 9 Oct 95 10:08:56 PDT Message-Id: <9510091708.AA27196@scalpel.netlabs.com> To: Mark Jefferys Cc: Tye McQueen , perldb-interest@fugue.com, perlbug@perl.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: Your message of "Sun, 08 Oct 95 17:43:40 CDT." Date: Mon, 09 Oct 95 10:08:55 -0700 From: Larry Wall : On Sat, 7 Oct 1995, Tye McQueen wrote: : : % Excerpts from the mail message of Mark Jefferys: : % ) % I'd also like a way to tell if a key is present in a hash, ie. : % [...] : % ) Or how about "exists $hash{$key}"? I don't think it would take too : % ) much effort to add that. :-) : % : % No fair adding it my copy while I'm not looking! :) : : Oops. : : % defined() should only check an expression value so defined(&func) : % should be the same as defined($x=&func). Um, defined() is also magical for arrays and hashes, guys. : This is mostly the confusion between &func as a way of referring to : &func, and as a way of calling it. A good reason to deprecate the : latter, IMHO. Gee, it's context dependent. How awful... :-) : % To see if a function exists, you should use exists(&func). : % exists() does not take an expression or value as an argument : % so saving some value in $save=oper() doesn't effect what : % exists($save) returns (except when doing so creates $save). : : OK. I like that, actually. But I think exists(&func) should determine whether the function has been *declared*, not whether the function has been defined. I would like to be able to distinguish the two. : % Now silly stuff on how exists() and delete() should be extended: : % : % delete(&func) would be used to make the next &func re-autoload. : % undef(&func) would keep its current behavior, to make &func() an : % error (leave &func cached as if autoloading failed). : % : % $x= "global"; : % { my $x= "my"; : % print "($x) "; : % delete($x); # Remove lexical from scope, a compile-time op : % print "<$x> "; : % delete($x); # Delete $x from symbol table : % print "[$x]\n"; : % } : % # This would print "(my) []". : : That's either really clever or really gross---I can't quite decide. I : think I like it. I'd rather not say what I think of it. : % Notice how my mind likes to deny that it is "x" in the symbol : % table and not "$x"? But NoneSuch would allow us to hide that : % fact. ;> : : Perhaps. I don't see any reason why we need to add an actual NoneSuch : scalar to make this work. : : IDEA: One thing that could be done---and this would cut down on spurious : warnings a lot---is to create the SVs, AVs, and HVs when first : accessed, instead of when their containing GVs are created. A warning : could be produced if this is done in a non-modification context. : : I don't know precisely how this would affect performance, though. : For example, there would need to be an extra check in pp_sv to see : that the accessed scalar was non-zero. On the other hand, Perl : wouldn't waste time (and space) creating unused AVs and HVs when : someone uses $x without the others. And the current warning checks : could be thrown out. As far as I know, the only time Perl creates AVs and HVs on spec is when sv_setsv() creates a fake *. And I found last week that I could comment out that Perl4-derived code now because in Perl 5, fake globs are true aliases rather than temporary namespaces, as in Perl 4. (Perl 4 actually copied the pointers, while Perl 5 keeps a reference counted GP ("glob pointer") structure.) So it looks like we've already reached a state where the Perl 5 compiler always knows when it has to create values, at least for arrays and hashes. I believe it still creates SVs on spec though. You'll note the code that pushes SV values doesn't bother to check whether it's an lvalue. Larry ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <15683-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 22:31:46 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 21:03:52 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa10002; 9 Oct 95 22:03 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA08854 for perldb-interest-real; Mon, 9 Oct 1995 12:24:34 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA08850 for ; Mon, 9 Oct 1995 12:24:31 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id ac03980; 9 Oct 95 13:24 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <02735-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 10:47:11 +0100 To: phj@pilot.njin.net Subject: Re: How to query for primary keys ? Cc: perldb-interest@fugue.com Date: Mon, 9 Oct 1995 10:47:11 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510091324.ac03980@relay-3.mail.demon.net> > From: "Patricia H. Ju" > > Hi, > > How does one query > 1. if a certain field is a primary key > 2. if a table has a primary key > 3. what the primary key of a table is, if one exists > using DBD-Oracle? > You currently have to rummage in the system catalogs. X/Open has defined an interface for this info so I'll appear in the DBI/DBD's in time. Tim. ---+++--- Received: from punt-4.mail.demon.net by ignite.demon.co.uk with SMTP (PP) id <15681-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 22:31:47 +0100 Received: from punt-4.mail.demon.net by mailstore for dbperl@ignite.demon.co.uk id <813273604:27633>; Mon, 9 Oct 95 21:20:04 GMT Received: from relay-1.mail.demon.net by punt-4.mail.demon.net id aa27536; 9 Oct 95 22:19 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id ab15296; 9 Oct 95 22:15 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA08904 for perldb-interest-real; Mon, 9 Oct 1995 12:40:36 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA08900 for ; Mon, 9 Oct 1995 12:40:33 -0700 Received: by gw.home.vix.com id AA22492; Mon, 9 Oct 95 12:42:48 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from ignite.demon.co.uk by relay-3.mail.demon.net id aa03579; 9 Oct 95 19:16 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <13084-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 19:16:20 +0100 To: perldb-interest@vix.com, gcheng@npac.syr.edu Subject: Re: (no subject) Date: Mon, 9 Oct 1995 19:16:20 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510091916.aa03579@relay-3.mail.demon.net> > From: Gang Cheng > > We at Northeast Parallel Architecture Center (NPAC), Syracue > University has developed an archive system for searching/reading > USENET newsgrou > ps, > mailing lists and personal mailbox from WWW. An Oracle database > server is used > to store/manage mails and two search/navigation interfaces > accesible by any WWW browser > to the archive are provided: one is an advanced search interface > allowing queries with various options such as query by mail > header, > by date, by subject (keywords), by sender. The other is a > Hypermail-like > navigation interface for users familiar with Hypermail. > > We offer free access to this archive for the Internet > Internet community. Check the URL > > . > > The whole archive is automatically updated in > every 30 minutes. Currently the following newsgroups (total 40) > are archived > in the database: > > comp.arch > comp.databases > comp.databases.oracle > comp.parallel > comp.parallel.mpi > comp.parallel.pvm > comp.lang.java > alt.www.hotjava > comp.infosystems > comp.infosystems.www.authoring.cgi > comp.infosystems.www.announce > comp.infosystems.www.authoring.html > comp.infosystems.www.authoring.images > comp.infosystems.www.authoring.misc > comp.infosystems.www.misc > comp.infosystems.www.advocacy > comp.infosystems.www.browsers.mac > comp.infosystems.www.browsers.ms-windows > comp.infosystems.www.browsers.x > comp.infosystems.www.browsers.misc > comp.infosystems.www.servers.ms-windows > comp.infosystems.www.servers.unix > comp.infosystems.www.servers.mac > comp.infosystems.www.servers.misc > comp.graphics > comp.graphics.visualization > comp.lang.c++ > comp.multimedia > comp.security.misc > comp.sources.unix > comp.sources.x > comp.unix.admin > comp.unix.aix > comp.unix.misc > comp.unix.programmer > comp.unix.solaris > alt.comp.compression > comp.databases.object > comp.lang.perl.misc > syr.general > > The following mailing lists are archived: > > java-interest > java-porting > java-announce > hotjava-interest > sp-discussion > www-vrml > perldb-interest-request@fugue.com > > This system is still under development and more functions are > planned to > add to this system. Please send your comments to Gang Cheng > (gcheng@npac.syr.edu > . Thanks. > > --- Gang > I'd suggest adding comp.lang.perl.announce, comp.lang.perl.tk (recently passed voting) and the perl5-porters@nicoh.com mailing list. The perldb-interest-request@fugue.com should be perldb-interest@fugue.com. Thanks for archiving all this. Tim. ---+++--- Received: from punt-4.mail.demon.net by ignite.demon.co.uk with SMTP (PP) id <15681-2@ignite.demon.co.uk>; Mon, 9 Oct 1995 22:32:18 +0100 Received: from punt-4.mail.demon.net by mailstore for dbperl@ignite.demon.co.uk id <813274103:01817>; Mon, 9 Oct 95 21:28:24 GMT Received: from relay-1.mail.demon.net by punt-4.mail.demon.net id aa01573; 9 Oct 95 22:27 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa15748; 9 Oct 95 22:20 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA08957 for perldb-interest-real; Mon, 9 Oct 1995 12:56:44 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA08953 for ; Mon, 9 Oct 1995 12:56:41 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id aa02942; 9 Oct 95 19:14 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <12984-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 19:11:01 +0100 To: perldb-interest@fugue.com, phj@pilot.njin.net Subject: Re: Apologies Date: Mon, 9 Oct 1995 19:11:01 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510091914.aa02942@relay-3.mail.demon.net> > From: "Patricia H. Ju" > > Sorry for the perl question. I did not realize this list was mainly > adminnish. > > Patricia No need to apologise Patricia your question about determining primary keys was perfectly on topic. I certainly didn't mean to give the impression that your message was out of place. Perhaps I sent you a message aimed at someone else. If so then I apologise. Tim. p.s. Sometimes I can get very terse when I'm rushed and that can come across in a bad way. If that's the cause in this case then I'm sorry. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <17911-2@ignite.demon.co.uk>; Tue, 10 Oct 1995 07:06:08 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 09 Oct 95 22:30:54 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa07484; 9 Oct 95 23:30 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA09176 for perldb-interest-real; Mon, 9 Oct 1995 13:50:29 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA09170 for ; Mon, 9 Oct 1995 13:50:21 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id aa19851; 9 Oct 95 16:44 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <08951-0@ignite.demon.co.uk>; Mon, 9 Oct 1995 16:09:42 +0100 To: descarte@hermetica.com Subject: Re: OraPerl on NT Cc: perldb-interest@fugue.com Date: Mon, 9 Oct 1995 16:09:42 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510091644.aa19851@relay-3.mail.demon.net> > From: Alligator Descartes > > Tim Bunce wrote: > > > > If you like, I can ftp my source files to you. Just give me an > > > address as to were to put them. > > > > > If it's not that complicated perhaps you could just write a step-by-step > > README.NT (in the same vein as the existing README.oracle7 & README.Solaris2) > > That's generally better then just sending me a whole stack of source files. > > That sounds like a plan. If you want to send me a copy of the text, I'll put > it up on the DBI site for other's perusal.... Ideally it should go onto my ftp archive from which it gets mirrored to CPAN sites around the world. Your DBI WWW site should have lots of pointers into archives rather than copying chunks of archives locally otherwise we'll end up duplicating much material with all the problems that leads to. The CPAN work is the answer. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <19356-0@ignite.demon.co.uk>; Tue, 10 Oct 1995 07:25:30 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 10 Oct 95 06:24:34 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id ab21434; 10 Oct 95 7:23 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id WAA10237 for perldb-interest-real; Mon, 9 Oct 1995 22:02:36 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id WAA10233 for ; Mon, 9 Oct 1995 22:02:34 -0700 Received: by gw.home.vix.com id AA28844; Mon, 9 Oct 95 22:04:49 -0700 Date: Mon, 9 Oct 95 22:04:49 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from song.iuol.cn.net by iui.iuol.cn.net id aa15671; 10 Oct 95 12:55 PST Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit From: lhsong@iuol.cn.net Subject: Unsubscribe To: perldb-interest@vix.com X-Mailer: SPRY Mail Version: 04.00.06.17 Message-Id: <9510101255.aa15671@iui.iuol.cn.net> unsubscribe lhsong@iuol.cn.net ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <27518-33@ignite.demon.co.uk>; Wed, 11 Oct 1995 07:11:25 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 05:15:09 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa09141; 11 Oct 95 6:14 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA13669 for perldb-interest-real; Tue, 10 Oct 1995 21:04:12 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id VAA13665 for ; Tue, 10 Oct 1995 21:04:10 -0700 Received: by gw.home.vix.com id AA02322; Tue, 10 Oct 95 21:06:26 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id XAA01915; Tue, 10 Oct 1995 23:06:21 -0500 Date: Tue, 10 Oct 1995 23:06:20 -0500 (CDT) From: Ben Lindstrom To: DBI perl Subject: Partial Success with DBD::QBase (QuickBase) Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII After going through all the headaches of getting things to resolve for creating a shared library...I now come to the following. (I've not had too much time to look into it.) ----- ./test.pl -d 5 QBase ---- DBI dispatch debug level set to 5 >> FETCH DISPATCH (DBI::dr=HASH(0xc2698) @2 g0 a0 r1) -DebugDispatch: 5 Available Drivers: ExampleP, NullP, QBase Read DBI special variable <> $DBI::err (*) FETCH fro0 err: 99 errstr: 99 Attempt to modify DBI special variables. Expect a 'Can't modify' error message: Driver installed as DBI::dr=HASH(0x12d780) m lasth=DBI::dr=HASH(0xc2698) DBD::Switch::dr::errstr <> $DBI::err (*) FETCH from lasth=DBI::dr=HASH(0xc2698) err = ''99'' Can't modify $DBI::&rows special variable at ./test.pl line 59 DBI->install_driver DBI QBase at ./test.pl line 67 DBI->install_driver(QBase) loaded install_driver(QBase): setup @ISA for DBD::QBase::dr at ./test.pl line 67 install_driver(QBase): setup @ISA for DBD::QBase::db at ./test.pl line 67 install_driver(QBase): setup @ISA for DBD::QBase::st at ./test.pl line 67 QBase_HOME not set! New DBI::dr (for DBD::QBase::dr, parent=, id=) dbih_setup_handle(DBI::dr=HASH(0x12d780)=>DBI::dr=HASH(0x13ddbc), DBD::QBase::dr, NULL) New DBI::dr => DBI::dr=HASH(0x12d780) (inner=DBI::dr=HASH(0x13ddbc)) for DBD::QBase::dr DBI->install_driver(QBase) = DBI::dr=HASH(0x12d780) >> connect DISPATCH (DBI::dr=HASH(0x12d780) @4 g0 a11c13c r1) -> connect for DBD::QBase::dr (DBI::dr=HASH(0x12d780)~0x13ddbc '' '' '') New DBI::db (for DBD::QBase::db, parent=DBI::dr=HASH(0x13ddbc), id=) dbih_setup_handle(DBI::db=HASH(0x13dd38)=>DBI::db=HASH(0x13dd74), DBD::QBase::db, NULL) New DBI::db => DBI::db=HASH(0x13dd38) (inner=DBI::db=HASH(0x13dd74)) for DBD::QBase::db segmentation fault -------- I'm happy I've successfully made it this far.. If my understanding is coorrect.. It's having a problem with my connect to database routines. Currect? ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <01282-0@ignite.demon.co.uk>; Wed, 11 Oct 1995 11:54:02 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 10:44:53 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa28087; 10 Oct 95 19:19 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa23699; 10 Oct 95 18:47 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA12256 for perldb-interest-real; Tue, 10 Oct 1995 07:47:55 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-3.mail.demon.net (post.demon.co.uk [158.152.1.72]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA12252 for ; Tue, 10 Oct 1995 07:47:52 -0700 Received: from ignite.demon.co.uk by relay-3.mail.demon.net id aa11204; 10 Oct 95 14:49 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <22592-0@ignite.demon.co.uk>; Tue, 10 Oct 1995 14:49:11 +0100 To: perldb-interest@fugue.com Subject: Any thoughts on the Oraperl emulation? Date: Tue, 10 Oct 1995 14:49:11 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510101449.aa11204@relay-3.mail.demon.net> My current plans are 1. Make minor release soon to tidy up some loose ends (such as the $sth->fetch bug) 2. Implement DBD::Oracle row cache and release new version 3. Aim to complete oraperl emulation and announce it as beta (that's just the emulation that's going beta At that point I hope to start addressing SQL CLI standards issues in ernest. Meanwhile does anyone have any comments on the current Oraperl emulation that I can incorporate into the next release? Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <02128-0@ignite.demon.co.uk>; Wed, 11 Oct 1995 13:30:34 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 11:43:35 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa09370; 11 Oct 95 12:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA15484 for perldb-interest-real; Wed, 11 Oct 1995 03:10:03 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA15480 for ; Wed, 11 Oct 1995 03:10:02 -0700 Received: by gw.home.vix.com id AA26977; Wed, 11 Oct 95 03:12:14 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from pccomtech.bgl.lu (cs0-a5.dial.eo.lu [193.93.48.249]) by master.flower.it (8.6.11/8.6.9) with SMTP id MAA05103 for ; Wed, 11 Oct 1995 12:17:11 +0100 Message-Id: <199510111117.MAA05103@master.flower.it> X-Sender: mbalestra@mail.flower.it X-Mailer: Windows Eudora Version 1.4.4 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 11 Oct 1995 12:05:50 +0100 To: perldb-interest@vix.com From: Massimo Balestra Subject: Stored functions Hi everybody, I use DBI 0.63 with the DBD::Oracle 0.25 with the Oraperl emulation. The platform is an IBM AIX (R6000) RISC machine. A simple question: there is a way to access directly oracle stored functions (stand-alone and package ones) ? I tried with this artifice: local($SELECT) = "SELECT () FROM dual "; local ($SH ) = &ora_open($ORA_LDA, $SELECT); @INP = &ora_fetch($SH); &ora_close($SH); And it works but it is only a trick. I would like to know if there is something more "clean". Thank You Massimo Balestra ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <02128-2@ignite.demon.co.uk>; Wed, 11 Oct 1995 13:30:47 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 11:44:35 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id ab09557; 11 Oct 95 12:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA15430 for perldb-interest-real; Wed, 11 Oct 1995 02:47:33 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA15426 for ; Wed, 11 Oct 1995 02:47:26 -0700 Received: by gw.home.vix.com id AA23537; Wed, 11 Oct 95 02:49:40 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-1.mail.demon.net id ac06993; 11 Oct 95 10:47 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <00150-0@ignite.demon.co.uk>; Wed, 11 Oct 1995 10:33:35 +0100 To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: Partial Success with DBD::QBase (QuickBase) Date: Wed, 11 Oct 1995 10:33:35 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510111047.ac06993@relay-1.mail.demon.net> > From: Ben Lindstrom > > After going through all the headaches of getting things to resolve for > creating a shared library...I now come to the following. (I've not had > too much time to look into it.) > > >> connect DISPATCH (DBI::dr=HASH(0x12d780) @4 g0 a11c13c r1) > -> connect for DBD::QBase::dr (DBI::dr=HASH(0x12d780)~0x13ddbc '' '' '') > New DBI::db (for DBD::QBase::db, parent=DBI::dr=HASH(0x13ddbc), id=) > dbih_setup_handle(DBI::db=HASH(0x13dd38)=>DBI::db=HASH(0x13dd74), DBD::QBase::db, NULL) > New DBI::db => DBI::db=HASH(0x13dd38) (inner=DBI::db=HASH(0x13dd74)) for DBD::QBase::db > segmentation fault > > I'm happy I've successfully made it this far.. > > If my understanding is coorrect.. It's having a problem with my connect > to database routines. Currect? You'll need to post a stack trace from the core dump before I can help very much. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06715-1@ignite.demon.co.uk>; Wed, 11 Oct 1995 17:57:24 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 16:29:43 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa27510; 11 Oct 95 17:29 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA16002 for perldb-interest-real; Wed, 11 Oct 1995 06:52:35 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA15998 for ; Wed, 11 Oct 1995 06:52:33 -0700 Received: from post.demon.co.uk by relay-1.mail.demon.net id ab09058; 11 Oct 95 14:45 +0100 Received: from toad by oink with SMTP (PP) id <03189-0@oink>; Wed, 11 Oct 1995 14:03:42 +0100 Received: by toad (5.0/SMI-SVR4) id AA04343; Wed, 11 Oct 1995 14:03:22 +0000 Date: Wed, 11 Oct 1995 14:03:22 +0000 From: Tim Bunce Message-Id: <9510111303.AA04343@toad> To: perldb-interest@fugue.com Subject: Test mail from new mail configuration - please ignore X-Sun-Charset: US-ASCII content-length: 322 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Sorry to bore you with this but I need to be sure that all is well. The machine on my desk is about to upgrade from Solaris 1 to 2. Because of this our Solaris 1 based PP mail system is moving to a different box and I'm originating this from a third box. Hopefully all this will be transparent to everyone outside. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <06943-0@ignite.demon.co.uk>; Wed, 11 Oct 1995 18:25:00 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 16:57:53 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id ab12509; 11 Oct 95 13:28 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa25599; 11 Oct 95 13:02 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA15479 for perldb-interest-real; Wed, 11 Oct 1995 03:10:02 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA15474 for ; Wed, 11 Oct 1995 03:10:00 -0700 Received: by gw.home.vix.com id AA26976; Wed, 11 Oct 95 03:12:11 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from pccomtech.bgl.lu (cs0-a5.dial.eo.lu [193.93.48.249]) by master.flower.it (8.6.11/8.6.9) with SMTP id MAA05101; Wed, 11 Oct 1995 12:16:53 +0100 Message-Id: <199510111116.MAA05101@master.flower.it> X-Sender: mbalestra@mail.flower.it X-Mailer: Windows Eudora Version 1.4.4 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 11 Oct 1995 12:05:41 +0100 To: perldb-interest@vix.com From: Massimo Balestra Subject: Coredump on syntax errors Cc: Tim.Bunce@ig.co.uk Hi everybody, I use DBI 0.63 with the DBD::Oracle 0.25 with the Oraperl emulation. The platform is an IBM AIX (R6000) RISC machine. The result of perl -e 'use Config; print Config::myconfig()' is: Summary of my perl5 (patchlevel 1) configuration: Platform: osname=aix, osver=3.2.5, archname=aix uname='aix bgspectrum2 2 3 002016317100 ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-D_ALL_SOURCE' ccflags ='-D_ALL_SOURCE' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=1, d_casti32=define, d_castneg= intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/lib /usr/lib /usr/ccs/lib /usr/local/lib libs=-ldbm -lld -lm -lc -lbsd -lPW libc=/usr/lib/libc.a Dynamic Linking: dlsrc=dl_aix.xs, dlext=so, d_dlsymun= cccdlflags=' ', ccdlflags='-bE:perl.exp', lddlflags='-H512 -T512 -bhalt:4 -bM:SRE -bI:$(PERL_INC)/perl.exp -bE:$(BASEEXT).exp -e _nostart -lc' My problem is: when I am developing the functions to access the oracle database and, in the first testing, there is a syntax error in my SQL statement, I get a coredump. It is not a big problem because it happens only during the first testing but I would like to receive an error. It happens always with "insert" and "update" and does not happens with "select", with "select" I get a parse error. An example: #!/usr/local/bin/perl $ORA_LDA =LOGON; # Personal Function to access the database in a standard way $SELECT = "UPDATE friends, SET name = 'Pasquale' WHERE id = 2078 "; # There is a syntax error: the comma after the table name # With this statement there is the same behavior # $SELECT = "INSERT INTO friends, values (2079, 'Pasquale') "; local ($SH ) = &ora_open($ORA_LDA, $SELECT); ######### IN THIS MOMENT I GET A CORE DUMP ######### I have a coredump if I use &ora_do instead of &ora_open. The installation seems not to have problem, everything is fine. Any suggestion? Thank You Massimo Balestra ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09648-4@ignite.demon.co.uk>; Thu, 12 Oct 1995 07:06:26 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 20:33:05 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa23362; 11 Oct 95 21:32 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id ac08656; 11 Oct 95 21:17 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA16528 for perldb-interest-real; Wed, 11 Oct 1995 10:54:05 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA16524 for ; Wed, 11 Oct 1995 10:54:00 -0700 Received: by gw.home.vix.com id AA26604; Wed, 11 Oct 95 10:56:02 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-1.mail.demon.net id aa12003; 11 Oct 95 18:25 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <06909-0@ignite.demon.co.uk>; Wed, 11 Oct 1995 18:21:34 +0100 To: perldb-interest@vix.com, mbalestra@flower.it Subject: Re: Stored functions Date: Wed, 11 Oct 1995 18:21:34 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510111825.aa12003@relay-1.mail.demon.net> > From: Massimo Balestra > > Hi everybody, > > I use DBI 0.63 with the DBD::Oracle 0.25 with the Oraperl emulation. The > platform is an IBM AIX (R6000) RISC machine. > > A simple question: there is a way to access directly oracle stored functions > (stand-alone and package ones) ? I tried with this artifice: > local($SELECT) = "SELECT () FROM dual "; > local ($SH ) = &ora_open($ORA_LDA, $SELECT); > @INP = &ora_fetch($SH); > &ora_close($SH); > > And it works but it is only a trick. I would like to know if there is > something more "clean". > Use "BEGIN ... END". See your PL/SQL manual and the notes on ftp.demon.co.uk:/pub/perl/db/perl4/oraperl/ Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09647-11@ignite.demon.co.uk>; Thu, 12 Oct 1995 07:07:45 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 20:29:20 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa10527; 11 Oct 95 21:28 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA16806 for perldb-interest-real; Wed, 11 Oct 1995 12:05:38 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA16802 for ; Wed, 11 Oct 1995 12:05:34 -0700 Received: from post.demon.co.uk by relay-1.mail.demon.net id ab12003; 11 Oct 95 18:25 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <06922-0@ignite.demon.co.uk>; Wed, 11 Oct 1995 18:24:37 +0100 To: skimo@dns.ufsia.ac.be Subject: Re: Multiple DBD's --- problem found Cc: perldb-interest@fugue.com Date: Wed, 11 Oct 1995 18:24:37 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510111825.ab12003@relay-1.mail.demon.net> > From: Sven Verdoolaege > > The problem seems to be: identical names (eg dbd_db_login) > in both mSQL en Oracle. > > I wonder how you managed to get that to work on other machines, > but on this machine it doesn't. > > doing a > perl -pie 's/dbd/odb/g' * > seems to have removed the problem (or the symptoms :-) > of course this isn't a ``solution'' > (it was the first thing i could think of) > > did i do something wrong that made this not work on my system or > should those names be prefixed or something ? Dynamic linking should avoid the problem. Static linking should have generated an error at link time. (Yes, static linking of multiple DBD's will require #define's to rename the functions). What platform are you on? (myconfig output would be handy) Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09646-10@ignite.demon.co.uk>; Thu, 12 Oct 1995 07:07:51 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 21:03:07 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa21646; 11 Oct 95 22:02 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA16719 for perldb-interest-real; Wed, 11 Oct 1995 11:53:08 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA16715 for ; Wed, 11 Oct 1995 11:53:06 -0700 Received: by gw.home.vix.com id AA00979; Wed, 11 Oct 95 11:55:19 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id NAA06447; Wed, 11 Oct 1995 13:47:47 -0500 Date: Wed, 11 Oct 1995 13:47:47 -0500 (CDT) From: Ben Lindstrom To: Tim Bunce Cc: perldb-interest@vix.com Subject: Re: Partial Success with DBD::QBase (QuickBase) In-Reply-To: <9510111047.ac06993@relay-1.mail.demon.net> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII > You'll need to post a stack trace from the core dump before I can > help very much. > Maybe I need to review GDB, but how do get the stack from the coredump? (After locating it on my NeXT) I'm only use to debuging ObjC code via GDB. Thanks. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09648-12@ignite.demon.co.uk>; Thu, 12 Oct 1995 07:08:17 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 21:47:14 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa00518; 11 Oct 95 22:46 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id ab18823; 11 Oct 95 22:25 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA17077 for perldb-interest-real; Wed, 11 Oct 1995 12:58:31 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA17073 for ; Wed, 11 Oct 1995 12:58:29 -0700 Received: by gw.home.vix.com id AA06917; Wed, 11 Oct 95 13:00:42 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by fahrenheit-451.media.mit.edu; id AA11000; Wed, 11 Oct 1995 16:00:09 -0400 Date: Wed, 11 Oct 1995 16:00:09 -0400 From: Jon Orwant Message-Id: <9510112000.AA11000@fahrenheit-451.media.mit.edu> To: mouring@netnet.net Cc: Tim.Bunce@ig.co.uk, perldb-interest@vix.com In-Reply-To: (message from Ben Lindstrom on Wed, 11 Oct 1995 13:47:47 -0500 (CDT)) Subject: Re: Partial Success with DBD::QBase (QuickBase) Reply-To: orwant@media.mit.edu > Maybe I need to review GDB, but how do get the stack from the coredump? > (After locating it on my NeXT) a "help stack" on NeXT GDB yields: bt -- Print backtrace of all stack frames backtrace -- Print backtrace of all stack frames frame -- Select and print a stack frame down -- Select and print stack frame called by this one up -- Select and print stack frame that called this one return -- Make selected stack frame return to its caller tl -- List all of threads thread-list -- List all of threads ts -- Select and print a thread thread-select -- Select and print a thread Jon ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09647-22@ignite.demon.co.uk>; Thu, 12 Oct 1995 07:10:12 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 11 Oct 95 21:14:29 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa01266; 11 Oct 95 22:13 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA16786 for perldb-interest-real; Wed, 11 Oct 1995 12:01:41 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from dns.ufsia.ac.be (skimo@dns.ufsia.ac.be [146.175.16.12]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA16782 for ; Wed, 11 Oct 1995 12:01:34 -0700 Received: (from skimo@localhost) by dns.ufsia.ac.be (8.6.12/8.6.9) id VAA03414; Wed, 11 Oct 1995 21:05:36 +0100 From: Sven Verdoolaege Message-Id: <199510112005.VAA03414@dns.ufsia.ac.be> Subject: multiDBD To: Tim Bunce Date: Wed, 11 Oct 1995 21:05:35 +0100 (MET) Cc: perldb X-Stardate: [-31]6424.18 X-URL: http://dns.ufsia.ac.be/~skimo X-Mailer: ELM [version 2.4 PL24 ME8a] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 324 The problem seems to be: identical names (eg dbd_db_login) in both mSQL and Oracle I don't know how you got that to work on your system, but it doesn't seem to work on mine. I temporarily removed the problem by changing every occurance of dbd by odb (just to make unique names) Tim, did you get my other messages ? Sven ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <09648-61@ignite.demon.co.uk>; Thu, 12 Oct 1995 07:19:51 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 12 Oct 95 02:17:21 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa10093; 12 Oct 95 3:16 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id ab23749; 12 Oct 95 3:11 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA17827 for perldb-interest-real; Wed, 11 Oct 1995 17:52:27 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA17823 for ; Wed, 11 Oct 1995 17:52:24 -0700 Received: by gw.home.vix.com id AA26690; Wed, 11 Oct 95 17:40:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id TAA27534; Wed, 11 Oct 1995 19:37:28 -0500 Date: Wed, 11 Oct 1995 19:37:28 -0500 (CDT) From: Ben Lindstrom To: Tim Bunce Cc: perldb-interest@vix.com Subject: Re: Partial Success with DBD::QBase (QuickBase) In-Reply-To: <9510111047.ac06993@relay-1.mail.demon.net> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Wed, 11 Oct 1995, Tim Bunce wrote: > > > From: Ben Lindstrom > > > > After going through all the headaches of getting things to resolve for > > creating a shared library...I now come to the following. (I've not had > > too much time to look into it.) > > You'll need to post a stack trace from the core dump before I can > help very much. After looking at the core dump I release my ObjC went from resolving functsion to failure in the runtime system. objc_msgSend() fails because of Memory access exception on address (invalid address). Yes, Objectice C is losing it's appeal. =) Anyone have ideas? I linked in -ObjC and -lMedia_s -lNeXT_s into perl and -lQBSQL5_1 into my driver also with -ObjC on the linker arguments. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <12864-0@ignite.demon.co.uk>; Thu, 12 Oct 1995 07:20:29 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 12 Oct 95 06:13:19 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa26605; 12 Oct 95 7:13 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id WAA18452 for perldb-interest-real; Wed, 11 Oct 1995 22:07:50 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id WAA18448 for ; Wed, 11 Oct 1995 22:07:47 -0700 Received: by gw.home.vix.com id AA13013; Wed, 11 Oct 95 22:10:01 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id GAA17739; Thu, 12 Oct 1995 06:08:50 +0100 Date: Thu, 12 Oct 1995 06:08:50 +0100 Message-Id: <199510120508.GAA17739@anna.mind.de> From: Andreas Koenig To: mouring@netnet.net Cc: Tim.Bunce@ig.co.uk, perldb-interest@vix.com In-Reply-To: (message from Ben Lindstrom on Wed, 11 Oct 1995 19:37:28 -0500 (CDT)) Subject: Re: Partial Success with DBD::QBase (QuickBase) Reply-To: andreas.koenig@mind.de >>>>> "Ben" == Ben Lindstrom writes: Ben> On Wed, 11 Oct 1995, Tim Bunce wrote: >> >> > From: Ben Lindstrom >> > >> > After going through all the headaches of getting things to resolve for >> > creating a shared library...I now come to the following. (I've not had >> > too much time to look into it.) >> >> You'll need to post a stack trace from the core dump before I can >> help very much. Ben> After looking at the core dump I release my ObjC went from resolving Ben> functsion to failure in the runtime system. Ben> objc_msgSend() fails because of Memory access exception on address Ben> (invalid address). Ben> Yes, Objectice C is losing it's appeal. =) Anyone have ideas? Ben> I linked in -ObjC and -lMedia_s -lNeXT_s into perl and -lQBSQL5_1 into my Ben> driver also with -ObjC on the linker arguments. No doubt it is desirable, that you can use dynamically linked ObjC extensions in the same manner as stock C extensions. _I_ can't tell you how to do that. I suggested you build a satic version of the extension, but you misunderstood me probably. It's pointless, I think, if you link libNeXT_s into perl, it just won't be used at all if you do not ALSO link in your libQBSQL, and I think the order has to be -lQBSQL5_1 -lNeXT_s and not the other way round. Once this solution works you can at least test your application and worry about the dynamic linking later. Try make perl in your extension directory. This might do. andreas ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <13230-0@ignite.demon.co.uk>; Thu, 12 Oct 1995 08:08:12 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 12 Oct 95 06:46:57 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa11229; 12 Oct 95 7:46 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id WAA18536 for perldb-interest-real; Wed, 11 Oct 1995 22:27:59 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id WAA18531 for ; Wed, 11 Oct 1995 22:27:56 -0700 Received: by gw.home.vix.com id AA14219; Wed, 11 Oct 95 22:30:10 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id AAA11014; Thu, 12 Oct 1995 00:30:05 -0500 Date: Thu, 12 Oct 1995 00:30:04 -0500 (CDT) From: Ben Lindstrom To: andreas.koenig@mind.de Cc: Tim.Bunce@ig.co.uk, perldb-interest@vix.com Subject: Re: Partial Success with DBD::QBase (QuickBase) In-Reply-To: <199510120508.GAA17739@anna.mind.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 12 Oct 1995, Andreas Koenig wrote: > >>>>> "Ben" == Ben Lindstrom writes: > No doubt it is desirable, that you can use dynamically linked ObjC > extensions in the same manner as stock C extensions. _I_ can't tell > you how to do that. I suggested you build a satic version of the > extension, but you misunderstood me probably. > Still don't understand what you mean by static version. NeXT really has no Shared libraries. I linked QBSQL into perl and I still recieved the error...If you can explain a little better...If you need I can post my current Makefile I'm using. > > Try make perl in your extension directory. This might do. > Staticly link DBI and my QuickBase driver right into perl? Is that what we are talking about....Hmm...Never thought about that. IT would work for a temporary solution if DBI can be static compiled into perl. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <13230-2@ignite.demon.co.uk>; Thu, 12 Oct 1995 08:08:22 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 12 Oct 95 06:48:15 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa11686; 12 Oct 95 7:47 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id WAA18645 for perldb-interest-real; Wed, 11 Oct 1995 22:44:42 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id WAA18641 for ; Wed, 11 Oct 1995 22:44:41 -0700 Received: by gw.home.vix.com id AA15119; Wed, 11 Oct 95 22:46:55 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id GAA17879; Thu, 12 Oct 1995 06:45:45 +0100 Date: Thu, 12 Oct 1995 06:45:45 +0100 Message-Id: <199510120545.GAA17879@anna.mind.de> From: Andreas Koenig To: mouring@netnet.net Cc: Tim.Bunce@ig.co.uk, perldb-interest@vix.com In-Reply-To: (message from Ben Lindstrom on Thu, 12 Oct 1995 00:30:04 -0500 (CDT)) Subject: Re: Partial Success with DBD::QBase (QuickBase) Reply-To: andreas.koenig@mind.de >>>>> "Ben" == Ben Lindstrom writes: Ben> On Thu, 12 Oct 1995, Andreas Koenig wrote: >> >>>>> "Ben" == Ben Lindstrom writes: >> No doubt it is desirable, that you can use dynamically linked ObjC >> extensions in the same manner as stock C extensions. _I_ can't tell >> you how to do that. I suggested you build a satic version of the >> extension, but you misunderstood me probably. >> Ben> Still don't understand what you mean by static version. NeXT really Ben> has no Shared libraries. I linked QBSQL into perl and I still Ben> recieved the error...If you can explain a little better...If you need Ben> I can post my current Makefile I'm using. >> >> Try make perl in your extension directory. This might do. >> Ben> Staticly link DBI and my QuickBase driver right into Ben> perl? Is that what we are talking about....Hmm...Never thought Ben> about that. IT would work for a temporary solution if Ben> DBI can be static compiled into perl. Right, right, right! ANY extension can be linked staticly into perl. Right, I forgot DBI. So go into your DBI directory and say make install LINKTYPE=static # will make DBI.a available as /usr/local....DBI/DBI.a cd make perl # will find and use DBI.a make install LINKTYPE=static # optional -- for future perl builds with # additional static extensions I frankly concede, that's not optimal but a valid temorary solution until your extension is stable. In fact it's a very handy way to make debugging easier for any extension builder, not only in your ObjC case. andreas ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <25453-10@ignite.demon.co.uk>; Fri, 13 Oct 1995 07:07:32 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 12 Oct 95 18:28:51 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa18171; 12 Oct 95 19:27 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA21052 for perldb-interest-real; Thu, 12 Oct 1995 09:27:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from dns.ufsia.ac.be (skimo@dns.ufsia.ac.be [146.175.16.12]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA21048 for ; Thu, 12 Oct 1995 09:26:58 -0700 Received: (from skimo@localhost) by dns.ufsia.ac.be (8.6.12/8.6.9) id SAA04679; Thu, 12 Oct 1995 18:26:40 +0100 From: Sven Verdoolaege Message-Id: <199510121726.SAA04679@dns.ufsia.ac.be> Subject: Re: Multiple DBD's --- problem found To: Tim Bunce Date: Thu, 12 Oct 1995 18:26:39 +0100 (MET) Cc: perldb In-Reply-To: <9510111825.ab12003@relay-1.mail.demon.net> from "Tim Bunce" at Oct 11, 95 06:24:37 pm X-Stardate: [-31]6428.63 X-URL: http://dns.ufsia.ac.be/~skimo X-Mailer: ELM [version 2.4 PL24 ME8a] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2003 > > The problem seems to be: identical names (eg dbd_db_login) > > in both mSQL en Oracle. > > > > I wonder how you managed to get that to work on other machines, > > but on this machine it doesn't. > > > > doing a > > perl -pie 's/dbd/odb/g' * > > seems to have removed the problem (or the symptoms :-) > > of course this isn't a ``solution'' > > (it was the first thing i could think of) > > > > did i do something wrong that made this not work on my system or > > should those names be prefixed or something ? > > Dynamic linking should avoid the problem. As you can see i am using dynamic linking. I'm not an expert on this, but i can't see how dynamic linking could avoid the problem Where do you tell the dynamic loader that you need the functions in Oracle.so and not the ones already loaded from mSQL.so ? > > Static linking should have generated an error at link time. > (Yes, static linking of multiple DBD's will require #define's > to rename the functions). > > What platform are you on? (myconfig output would be handy) here it is (again) (hopefully noone else remembered all this when i mailed a trace that contained my password :-) chulu.ufsia.ac.be> perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=dec_osf, osver=3.2, archname=alpha-dec_osf uname='osf1 chulu.ufsia.ac.be v3.2 214 alpha ' hint=previous Compiler: cc='cc', optimize='-O2 -Olimit 2900' cppflags='-DSTANDARD_C' ccflags ='-DSTANDARD_C' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/usr/shlib /lib /usr/lib /usr/ccs/lib libs=-ldbm -lm -lc -lbsd -lPW libc=/usr/shlib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags=' ', ccdlflags=' ', lddlflags='-shared -expect_unresolved "*" -s' Sven ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07313-4@ignite.demon.co.uk>; Fri, 13 Oct 1995 23:40:52 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 13 Oct 95 19:24:06 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa07785; 13 Oct 95 20:22 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA24666 for perldb-interest-real; Fri, 13 Oct 1995 11:11:45 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from beach.cas.american.edu (beach.cas.american.edu [147.9.100.60]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA24662 for ; Fri, 13 Oct 1995 11:11:43 -0700 Received: from cage.cas.american.edu by beach.cas.american.edu (8.6.10/) id OAA27848; Fri, 13 Oct 1995 14:23:13 -0400 Received: by cage.cas.american.edu (8.6.10) id OAA11056; Fri, 13 Oct 1995 14:14:54 -0400 Date: Fri, 13 Oct 1995 14:14:54 -0400 (EDT) From: Abinash Tripathy To: perldb-interest@fugue.com Subject: unsubscribe Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Abinash Tripathy 4201 Massachusetts Ave, N.W Apt 3030C Ph:(202) 363 0604(home) Washington D.C 20016 (202) 885 1479 (FAX) (202) 885 2767(work) email:abinash@email.cas.american.edu at0509a@american.edu WWW :http://www.cas.american.edu/~abinash/ -- ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07315-8@ignite.demon.co.uk>; Fri, 13 Oct 1995 23:41:18 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 13 Oct 95 19:48:01 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa03858; 13 Oct 95 20:47 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa22748; 13 Oct 95 20:20 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA24610 for perldb-interest-real; Fri, 13 Oct 1995 10:58:52 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gate.ti.com (news.ti.com [192.94.94.33]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA24606 for ; Fri, 13 Oct 1995 10:58:48 -0700 Received: from kadet.asic.sc.ti.com ([156.117.203.118]) by gate.ti.com (8.6.12/) with ESMTP id NAA14038 for ; Fri, 13 Oct 1995 13:00:29 -0500 Received: from horta.asic.sc.ti.com (horta.asic.sc.ti.com [156.117.206.77]) by kadet.asic.sc.ti.com (8.6.10/8.6.6) with ESMTP id MAA22718 for ; Fri, 13 Oct 1995 12:59:34 -0500 Received: from localhost (jsharp@localhost) by horta.asic.sc.ti.com (8.6.4/8.6.4) id MAA02214 for perldb-interest@fugue.com; Fri, 13 Oct 1995 12:59:23 -0500 Date: Fri, 13 Oct 1995 12:59:23 -0500 From: John Sharp Message-Id: <199510131759.MAA02214@horta.asic.sc.ti.com> To: perldb-interest@fugue.com Subject: subscribe subscribe ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <15660-11@ignite.demon.co.uk>; Sun, 15 Oct 1995 11:52:49 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 14 Oct 95 11:54:02 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id ab01657; 14 Oct 95 12:53 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA27321 for perldb-interest-real; Sat, 14 Oct 1995 03:49:09 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id DAA27317 for ; Sat, 14 Oct 1995 03:49:06 -0700 Received: from post.demon.co.uk by relay-4.mail.demon.net id g.ab18399; 14 Oct 95 0:47 +0100 Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id ae21423; 14 Oct 95 0:46 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <08868-0@ignite.demon.co.uk>; Sat, 14 Oct 1995 00:23:55 +0100 To: jon.perry@msd_hsv1.hv.boeing.com Subject: Re: Oraperl for Windows-NT Cc: perldb-interest@fugue.com, fox@austin.asc.slb.com Date: Sat, 14 Oct 1995 00:23:55 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510140047.ae21423@relay-3.mail.demon.net> > From: Jon Perry > > Tim, > > I understand you might have some insight to when Oraperl that > has been ported to NT, will be available on ftp.demon.co.uk. > My group at Boeing is interested in this tool for generating > some dynamic html. Do you know any dates when Oraperl for NT > might be at this site? Thanks for any help... or any other > recommendation on where to look. > I've CC'd this to the perldb-interest mailing list and Jim Fox (who I think is doing something) in the hope someone can update you (and me) about an NT port. > Jon Perry > Information Systems > Boeing > Huntsville, AL USA Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <15660-32@ignite.demon.co.uk>; Sun, 15 Oct 1995 12:01:08 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 15 Oct 95 03:35:43 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa01839; 15 Oct 95 4:34 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA28017 for perldb-interest-real; Sat, 14 Oct 1995 18:31:36 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from watson.ibm.com (watson.ibm.com [129.34.139.4]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id SAA28013 for ; Sat, 14 Oct 1995 18:31:34 -0700 Received: from WATSON by watson.ibm.com (IBM VM SMTP V2R3) with BSMTP id 3181; Sat, 14 Oct 95 21:33:37 EDT Received: from YKTVMV by watson.vnet.ibm.com with "VAGENT.V1.02 on VAGENT2" id 1428; Sat, 14 Oct 1995 21:33:37 EDT Received: from lusitania.watson.ibm.com by yktvmv.watson.ibm.com (IBM VM SMTP V2Rx) with TCP; Sat, 14 Oct 95 21:33:36 EDT Received: from localhost by lusitania.watson.ibm.com (AIX 3.2/UCB 5.64/950830) id AA38632; Sat, 14 Oct 1995 21:33:16 -0400 Message-Id: <9510150133.AA38632@lusitania.watson.ibm.com> X-Mailer: exmh version 1.6.2 7/18/95 To: perldb-interest@fugue.com Subject: Overriding malloc/free on AIX 3.2.5? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 14 Oct 1995 21:33:16 -0400 From: Chet Murthy Hum ... OK -- a very ugly problem here. I was running a program using DBI::DB2, and had a bug which manifested itself as a segfault in free. I couldn't figure it out, but I noticed that free() in DB2.so didn't seem to be coming from perl -- that is, it wasn't being overridden. I tried various things, and finally concluded that, given that ld uses its import-list *last*, after all object-files and archive-files, it isn't possible to override free() so that all .so files will use Perl's implementation, unless free() is named something else. So I turned on the "HIDEMYMALLOC" flag, and put Myfree, Mymalloc, Myrealloc into perl.exp, recompiled everything, and all of a sudden, everything worked. I don't know what to do about this in the long run. It seems clear to me that Perl's standard method of overriding malloc with a private version (i.e., by just supplying malloc(), etc) isn't working for shared-library extensions (at least, not on AIX 3.2.5). Anybody got any suggestions? --chet-- ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <18634-0@ignite.demon.co.uk>; Sun, 15 Oct 1995 14:16:19 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 15 Oct 95 12:43:07 GMT Received: from [204.254.239.2] by punt-1.mail.demon.net id aa14038; 15 Oct 95 13:42 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA29976 for perldb-interest-real; Sun, 15 Oct 1995 04:25:08 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id EAA29972 for ; Sun, 15 Oct 1995 04:25:05 -0700 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.bl01378; 13 Oct 95 9:59 +0100 Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id ab10474; 13 Oct 95 7:05 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <22703-0@ignite.demon.co.uk>; Thu, 12 Oct 1995 17:46:12 +0100 To: skimo@dns.ufsia.ac.be Subject: Re: Multiple DBD's --- problem found Cc: perldb-interest@fugue.com, perl5-porters@nicoh.com Date: Thu, 12 Oct 1995 17:46:12 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510130705.ab10474@relay-3.mail.demon.net> > From: Sven Verdoolaege > > > > The problem seems to be: identical function names (eg dbd_db_login) > > > in both DBD::mSQL and DBD::Oracle. > > > > > > I wonder how you managed to get that to work on other machines, > > > but on this machine it doesn't. > > > > > > doing a perl -pie 's/dbd/odb/g' * > > > seems to have removed the problem (or the symptoms :-) > > > of course this isn't a ``solution'' > > > (it was the first thing i could think of) > > > > > > did i do something wrong that made this not work on my system or > > > should those names be prefixed or something ? > > > > Dynamic linking should avoid the problem. > > As you can see i am using dynamic linking. > I'm not an expert on this, but i can't see how dynamic linking could avoid > the problem > Where do you tell the dynamic loader that you need the functions in > Oracle.so and not the ones already loaded from mSQL.so ? The dbd_db_login function is called from the XS_DBD__Oracle__db__login function in Oracle.c that is generated from Oracle.xs. When Oracle.so is created the linker knows that it needs to resolve that link when it's loaded. Firstly, when dynamically loading a shared object the runtime linker should resolve it using the symbol in the shared object that it's loading. Secondly, the symbols in the dynamically loaded object should not be visible for later runtime linking by default (Solaris 2 has a specific flag for this which dl_dlopen.xs does not use). > > Static linking should have generated an error at link time. > > (Yes, static linking of multiple DBD's will require #define's > > to rename the functions). > > > > What platform are you on? (myconfig output would be handy) > Summary of my perl5 (patchlevel 1) configuration: > Platform: > osname=dec_osf, osver=3.2, archname=alpha-dec_osf > uname='osf1 chulu.ufsia.ac.be v3.2 214 alpha ' > hint=previous > Compiler: > cc='cc', optimize='-O2 -Olimit 2900' > cppflags='-DSTANDARD_C' > ccflags ='-DSTANDARD_C' > ldflags ='' > stdchar='unsigned char', d_stdstdio=define, usevfork=false > voidflags=15, castflags=0, d_casti32=define, d_castneg=define > intsize=4, alignbytes=8, usemymalloc=y, randbits=15 > Libraries: > so=so > libpth=/usr/shlib /lib /usr/lib /usr/ccs/lib > libs=-ldbm -lm -lc -lbsd -lPW > libc=/usr/shlib/libc.so > Dynamic Linking: > dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= > cccdlflags=' ', ccdlflags=' ', lddlflags='-shared -expect_unresolved "*" -s' I've not actually tried this yet. Perhaps others could try the mSQL and Oracle combination on other platforms. Right now I suspect it's an OSF specific issue so I've CC'd this to the perl5-porters in the hope that any OSF users can clarify symbol visibility and resolution behaviour for me. > Sven > Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <19413-0@ignite.demon.co.uk>; Sun, 15 Oct 1995 15:59:38 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 15 Oct 95 14:46:41 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa03315; 15 Oct 95 15:46 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA00293 for perldb-interest-real; Sun, 15 Oct 1995 06:35:00 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA00289 for ; Sun, 15 Oct 1995 06:34:57 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id OAA03300; Sun, 15 Oct 1995 14:33:02 +0100 From: Alligator Descartes Message-Id: <199510151333.OAA03300@fruitbat.mcqueen.com> Subject: Re: Multiple DBD's --- problem found To: Tim Bunce Date: Sun, 15 Oct 1995 14:33:01 +0100 (BST) Cc: perldb-interest@fugue.com, skimo@dns.ufsia.ac.be In-Reply-To: <9510130705.ab10474@relay-3.mail.demon.net> from "Tim Bunce" at Oct 12, 95 05:46:12 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 559 > I've not actually tried this yet. Perhaps others could try the mSQL and > Oracle combination on other platforms. Works perfectly for me on Solaris 2.3/4. I'm just about to try it on Linux and HP-UX9 as well, so I'll let you know what happens there. My scenario is working with simultaneous connections to 2 Oracle, 1 mSQL and 1 Informix database, passing data between all three in a data-warehousing situation..... > > Sven > > > Tim. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.mcqueen.com/hermetica ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <20350-2@ignite.demon.co.uk>; Sun, 15 Oct 1995 19:32:47 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 15 Oct 95 18:06:51 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa00845; 15 Oct 95 19:05 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA00584 for perldb-interest-real; Sun, 15 Oct 1995 09:48:44 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA00580 for ; Sun, 15 Oct 1995 09:48:36 -0700 Received: from post.demon.co.uk by relay-4.mail.demon.net id g.aj22876; 15 Oct 95 13:07 +0100 Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id ab12572; 15 Oct 95 13:03 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <18084-0@ignite.demon.co.uk>; Sun, 15 Oct 1995 13:02:49 +0100 To: perldb-interest@fugue.com, chet@watson.ibm.com Subject: Re: Overriding malloc/free on AIX 3.2.5? Cc: perlbug@perl.com, wscott@ichips.intel.com Date: Sun, 15 Oct 1995 13:02:49 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510151303.ab12572@relay-3.mail.demon.net> > From: Chet Murthy > > Hum ... OK -- a very ugly problem here. > > I was running a program using DBI::DB2, and had a bug which manifested > itself as a segfault in free. I couldn't figure it out, but I noticed > that free() in DB2.so didn't seem to be coming from perl -- that is, > it wasn't being overridden. > > I tried various things, and finally concluded that, given that ld uses > its import-list *last*, after all object-files and archive-files, it > isn't possible to override free() so that all .so files will use > Perl's implementation, unless free() is named something else. I'll pass this on to the perl5-porters and Wayne Scott (who wrote dl_aix.xs) since I don't know enough about the obscure dynamic loading mechanism used by AIX 3. > So I turned on the "HIDEMYMALLOC" flag, and put Myfree, Mymalloc, > Myrealloc into perl.exp, recompiled everything, and all of a sudden, > everything worked. Interesting. perl_exp.SH should put the My* names into perl.exp for you. That's a bug - but not a very high priority one. > I don't know what to do about this in the long run. It seems clear to > me that Perl's standard method of overriding malloc with a private > version (i.e., by just supplying malloc(), etc) isn't working for > shared-library extensions (at least, not on AIX 3.2.5). I'm fairly confident that if this problem affected other platforms it would have been spotted before now - I hope. > Anybody got any suggestions? > > --chet-- Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <23015-10@ignite.demon.co.uk>; Mon, 16 Oct 1995 07:07:16 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 16 Oct 95 00:32:26 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id ab28298; 16 Oct 95 1:32 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA01003 for perldb-interest-real; Sun, 15 Oct 1995 15:47:38 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA00999 for ; Sun, 15 Oct 1995 15:47:30 -0700 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.am00359; 15 Oct 95 17:34 +0100 Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id aa19959; 15 Oct 95 17:13 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <19772-0@ignite.demon.co.uk>; Sun, 15 Oct 1995 17:12:59 +0100 To: descarte@hermetica.com Subject: Re: Multiple DBD's --- problem found Cc: perldb-interest@fugue.com, skimo@dns.ufsia.ac.be Date: Sun, 15 Oct 1995 17:12:59 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510151713.aa19959@relay-3.mail.demon.net> > From: Alligator Descartes > > > I've not actually tried this yet. Perhaps others could try the mSQL and > > Oracle combination on other platforms. > > Works perfectly for me on Solaris 2.3/4. I'm just about to try it on Linux > and HP-UX9 as well, so I'll let you know what happens there. > > My scenario is working with simultaneous connections to 2 Oracle, 1 mSQL and > 1 Informix database, passing data between all three in a data-warehousing > situation..... Wow! Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <23016-34@ignite.demon.co.uk>; Mon, 16 Oct 1995 07:11:43 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 16 Oct 95 01:24:35 GMT Received: from [204.254.239.2] by punt-1.mail.demon.net id aa00546; 16 Oct 95 2:24 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA01196 for perldb-interest-real; Sun, 15 Oct 1995 17:26:50 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from mailgw.claircom.com (mailgw.claircom.com [199.5.241.51]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA01192 for ; Sun, 15 Oct 1995 17:26:45 -0700 Received: from nimo.claircom.com by mailgw.claircom.com with smtp (Smail3.1.26.7 #2) id m0t4dQM-0004toC; Sun, 15 Oct 95 17:29 PDT Received: from airone.claircom.com by nimo.claircom.com with smtp (Smail3.1.26.7 #2) id m0t4dQL-0005VkC; Sun, 15 Oct 95 17:29 PDT Received: from soc4.airone.claircom.com by airone.claircom.com (5.0/SMI-SVR4) id AA18708; Sun, 15 Oct 1995 17:27:08 -0700 Received: by soc4.airone.claircom.com (4.1/SMI-4.1) id AA22083; Sun, 15 Oct 95 17:19:08 PDT Date: Sun, 15 Oct 95 17:19:08 PDT From: William Warner Message-Id: <9510160019.AA22083@soc4.airone.claircom.com> To: perldb-interest@fugue.com Subject: Re: Multiple DBD's --- problem found Cc: skimo@dns.ufsia.ac.be content-length: 1044 > From owner-perldb-interest@fugue.com Sun Oct 15 17:13 PDT 1995 > Sender: perldb-interest-request@fugue.com > To: descarte@hermetica.com > Subject: Re: Multiple DBD's --- problem found > Cc: perldb-interest@fugue.com, skimo@dns.ufsia.ac.be > Date: Sun, 15 Oct 1995 17:12:59 +0100 > From: Tim Bunce > Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > Content-Type> : > text> > Content-Length: 507 > > > > From: Alligator Descartes > > > > > I've not actually tried this yet. Perhaps others could try the mSQL and > > > Oracle combination on other platforms. > > > > Works perfectly for me on Solaris 2.3/4. I'm just about to try it on Linux > > and HP-UX9 as well, so I'll let you know what happens there. > > > > My scenario is working with simultaneous connections to 2 Oracle, 1 mSQL and > > 1 Informix database, passing data between all three in a data-warehousing > > situation..... > > Wow! > > Tim. > If this true, it's time for me to try perl-db now. ww ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <26984-0@ignite.demon.co.uk>; Mon, 16 Oct 1995 11:47:49 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 16 Oct 95 10:09:30 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa08692; 16 Oct 95 11:08 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA01629 for perldb-interest-real; Mon, 16 Oct 1995 01:10:13 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA01625 for ; Mon, 16 Oct 1995 01:10:06 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id JAA04737; Mon, 16 Oct 1995 09:07:54 +0100 From: Alligator Descartes Message-Id: <199510160807.JAA04737@fruitbat.mcqueen.com> Subject: Re: DBD::mSQL driver expectancies. To: tim Date: Mon, 16 Oct 1995 09:07:53 +0100 (BST) Cc: Tim.Bunce@ig.co.uk, perldb-interest@fugue.com In-Reply-To: <199510140411.EAA07562@www.cas.unt.edu> from "tim" at Oct 14, 95 04:11:39 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1456 > I'm trying to check out various interfaces to mSQL (for a WWW cgi > implementation) and am having some problems with the current > patch levels of DBI/DBD::mSQL. This is what I am working with: > > -rw-r--r-- 1 tim wheel 24775 Sep 20 21:11 /tmp/DBD-mSQL-0.60pl6.tar.gz > -rw-r--r-- 1 tim wheel 45973 Sep 1 12:22 /tmp/DBI-0.63.tar.Z > > I receive this error when running ./test2.pl: > > DynaLoader.pm loaded (/lib /usr/lib /usr/local/lib) > DynaLoader::bootstrap(DBI) > DynaLoader::bootstrap(DBD::mSQL) > install_driver(mSQL) failed: DBI version mismatch (DBI actual v3407879/s7, expec > ted v4/s44) at /usr/local/lib/perl5/DynaLoader.pm line 463. > at /usr/local/lib/perl5/DBI.pm line 167 > DBI::install_driver called at ./test2.pl line 12 > > Which leads me to believe that a newer version of DBI exists > someplace (or I need to drop back a patch level or two on DBD::mSQL). > Any clues? That combination of DBI/DBD::mSQL is the current working configuration ( bar a couple of bugs in each piece! ), so you should be fine with it. What platform are you on? Also, what patchlevel of perl are you on. You need to be at patchlevel 5.001m at least. If you could get that info to us, we can take a look from there. > Thanks in advance, > Tim. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <01222-5@ignite.demon.co.uk>; Mon, 16 Oct 1995 17:01:28 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 16 Oct 95 12:49:35 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa17656; 16 Oct 95 13:48 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA03292 for perldb-interest-real; Mon, 16 Oct 1995 04:02:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from hustle.rahul.net (hustle.rahul.net [192.160.13.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id EAA03288 for ; Mon, 16 Oct 1995 04:02:52 -0700 Received: from postoffice.UUCP by hustle.rahul.net with UUCP id AA20159 (5.67b8/IDA-1.5 for perldb-interest@fugue.com); Mon, 16 Oct 1995 04:05:08 -0700 Received: by ttmc.com (5.x/SMI-SVR4) id AA11577; Mon, 16 Oct 1995 08:00:02 -0300 Date: Mon, 16 Oct 1995 08:00:02 -0300 From: Dan Wisehart Message-Id: <9510161100.AA11577@ttmc.com> To: perldb-interest@fugue.com Subject: subscribe X-Sun-Charset: US-ASCII subscribe ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07557-10@ignite.demon.co.uk>; Tue, 17 Oct 1995 08:09:47 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 05:59:46 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa08786; 17 Oct 95 6:59 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA05576 for perldb-interest-real; Mon, 16 Oct 1995 21:58:51 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id VAA05572 for ; Mon, 16 Oct 1995 21:58:50 -0700 Received: by gw.home.vix.com id AA27328; Mon, 16 Oct 95 22:01:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id AAA05836; Tue, 17 Oct 1995 00:01:04 -0500 Date: Tue, 17 Oct 1995 00:01:04 -0500 (CDT) From: Ben Lindstrom To: DBI perl Subject: Question on SV/AV/etc Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I'm to the point that I'm attempting to implement dbd_st_fetch() Now, I need to have someone point me to some nice documention (besides man perl* =) that describe how to populate and deal with AV* and SV* since I am totally lost!! =) Good news, as long as I don't have to return anything I can access the database. =) Just fetchs and destruction problems seem to be the major things I need to deal with (No, *blob* routines are not going to be played with until I know at least what I'm doing with fetch. =) Any ideas? Hints? Or nice little padded cell? =) ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07558-28@ignite.demon.co.uk>; Tue, 17 Oct 1995 08:12:17 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 16 Oct 95 21:53:44 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ap16757; 16 Oct 95 22:53 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <00848-0@frigate.doc.ic.ac.uk>; Mon, 16 Oct 1995 20:46:27 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA03855 for perldb-interest-real; Mon, 16 Oct 1995 11:08:20 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA03851 for ; Mon, 16 Oct 1995 11:08:18 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id NAA104734 for ; Mon, 16 Oct 1995 13:10:35 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA24488; Mon, 16 Oct 1995 13:10:30 -0500 Message-Id: <9510161810.AA24488@dax.austin.ibm.com> To: Chet Murthy Cc: perldb-interest@fugue.com Subject: Re: Overriding malloc/free on AIX 3.2.5? In-Reply-To: Your message of Sat, 14 Oct 95 21:33:16 -0400. <9510150133.AA38632@lusitania.watson.ibm.com> Date: Mon, 16 Oct 95 13:10:29 -0500 Chet Murthy writes: |> |>Hum ... OK -- a very ugly problem here. |> |>I was running a program using DBI::DB2, and had a bug which manifested |>itself as a segfault in free. I couldn't figure it out, but I noticed |>that free() in DB2.so didn't seem to be coming from perl -- that is, |>it wasn't being overridden. |> |>I tried various things, and finally concluded that, given that ld uses |>its import-list *last*, after all object-files and archive-files, it |>isn't possible to override free() so that all .so files will use |>Perl's implementation, unless free() is named something else. |> |>So I turned on the "HIDEMYMALLOC" flag, and put Myfree, Mymalloc, |>Myrealloc into perl.exp, recompiled everything, and all of a sudden, |>everything worked. |> |>I don't know what to do about this in the long run. It seems clear to |>me that Perl's standard method of overriding malloc with a private |>version (i.e., by just supplying malloc(), etc) isn't working for |>shared-library extensions (at least, not on AIX 3.2.5). |> Well, the problem is not AIX it is in either DBI or DB2. I haven't been far enough in the debugger yet to isolate which. It seems to be caused by a bad value in DBIc_NUM_FIELDS(imp_sth). More when I isolate it. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07558-88@ignite.demon.co.uk>; Tue, 17 Oct 1995 08:22:07 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 01:03:27 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id bc00618; 17 Oct 95 2:02 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <07120-0@frigate.doc.ic.ac.uk>; Tue, 17 Oct 1995 00:23:23 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA04447 for perldb-interest-real; Mon, 16 Oct 1995 14:41:17 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA04441 for ; Mon, 16 Oct 1995 14:41:14 -0700 From: james.taylor@srs.gov Received: by gw.home.vix.com id AA26203; Mon, 16 Oct 95 14:43:08 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by srs.gov id AA03068 (InterLock SMTP Gateway 1.1 for perldb-interest@vix.com); Mon, 16 Oct 1995 17:41:05 -0400 Received: by srs.gov (Internal Mail Agent-1); Mon, 16 Oct 1995 17:41:05 -0400 Date: Mon, 16 Oct 1995 17:40:56 -0400 Message-Id: <199510162140.RAA12099@dark-knight.srs.gov> To: perldb-interest@vix.com Subject: DBD for Oracle Cc: Tim.Bunce@ig.co.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: sWS2RV4ihYodu4Ln2GQ9Bg== perldb-interest, I admit that I've haven't had time to test the code for quite some time... Is anyone else doing anything? What is the current status of any improvements? James Taylor ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07558-95@ignite.demon.co.uk>; Tue, 17 Oct 1995 08:23:22 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 01:03:27 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id bf00618; 17 Oct 95 2:02 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <07727-0@frigate.doc.ic.ac.uk>; Tue, 17 Oct 1995 00:44:32 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA04579 for perldb-interest-real; Mon, 16 Oct 1995 15:13:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from hpunix.gsfc.state.ga.us (hpunix.GSFC.STATE.GA.US [168.29.236.106]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA04575 for ; Mon, 16 Oct 1995 15:13:51 -0700 Message-Id: <199510162213.PAA04575@toccata.fugue.com> Received: from localhost by hpunix.gsfc.state.ga.us with SMTP (1.38.193.5/16.2) id AA23822; Mon, 16 Oct 1995 18:18:06 -0400 Date: Mon, 16 Oct 95 18:18:07 -0400 From: "Marc M. from Tactics" X-Mailer: Mozilla 1.1N (X11; I; HP-UX A.09.04 9000/867) Mime-Version: 1.0 To: perldb-interest@fugue.com Subject: Stored procedures and bind variables X-Url: http://www.hermetica.com/technologia/DBI/faq.html Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Hi everybody! I am using DBI 0.63 and DBD:Oracle 0.25 on an Hp. Does anyone know how to run stored procedures and pass values (from OUT, IN OUT parameters) back to perl? I cannot find any documentation on how to bind variables. Is the only alternative to shell out to sqlplus? There must be some way to do this. HELP!!!! Marc. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07558-98@ignite.demon.co.uk>; Tue, 17 Oct 1995 08:23:52 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 01:03:28 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id bj00618; 17 Oct 95 2:02 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <09119-0@frigate.doc.ic.ac.uk>; Tue, 17 Oct 1995 01:31:04 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA04804 for perldb-interest-real; Mon, 16 Oct 1995 16:15:37 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from zonk.geko.com.au (garyaj@zonk.geko.com.au [203.2.239.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA04799 for ; Mon, 16 Oct 1995 16:15:22 -0700 Received: (from garyaj@localhost) by zonk.geko.com.au (8.6.12/8.6.12) id JAA25369 for perldb-interest@fugue.com; Tue, 17 Oct 1995 09:14:29 +1000 From: Gary Ashton-Jones Message-Id: <199510162314.JAA25369@zonk.geko.com.au> To: perldb-interest@fugue.com Date: Tue, 17 Oct 1995 09:14:29 +1000 (EST) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Content-Length: 12 unsubscribe ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07558-121@ignite.demon.co.uk>; Tue, 17 Oct 1995 08:27:35 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 03:40:30 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa20943; 17 Oct 95 4:39 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA05229 for perldb-interest-real; Mon, 16 Oct 1995 19:16:08 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id TAA05225 for ; Mon, 16 Oct 1995 19:16:07 -0700 Received: by gw.home.vix.com id AA15833; Mon, 16 Oct 95 19:18:23 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id VAA10207; Mon, 16 Oct 1995 21:18:22 -0500 Date: Mon, 16 Oct 1995 21:18:22 -0500 (CDT) From: Ben Lindstrom To: DBI perl Subject: Staticly linking DBI + QuickBase driver Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I keep getting _boot_QBase unresolved in linking while doing static linking. I just made a blank function called that. Anyone know where that would be made legaly? Thanks. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <11136-0@ignite.demon.co.uk>; Tue, 17 Oct 1995 10:00:06 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 08:20:00 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa01016; 17 Oct 95 9:18 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA05878 for perldb-interest-real; Tue, 17 Oct 1995 00:11:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA05874 for ; Tue, 17 Oct 1995 00:11:52 -0700 Received: by gw.home.vix.com id AA05738; Tue, 17 Oct 95 00:14:06 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id IAA17336; Tue, 17 Oct 1995 08:12:57 +0100 Date: Tue, 17 Oct 1995 08:12:57 +0100 Message-Id: <199510170712.IAA17336@anna.mind.de> From: Andreas Koenig To: mouring@netnet.net Cc: perldb-interest@vix.com In-Reply-To: (message from Ben Lindstrom on Tue, 17 Oct 1995 00:01:04 -0500 (CDT)) Subject: Re: Question on SV/AV/etc Reply-To: andreas.koenig@mind.de >>>>> "ben" == Ben Lindstrom writes: ben> I'm to the point that I'm attempting to implement dbd_st_fetch() ben> Now, I need to have someone point me to some nice documention (besides ben> man perl* =) that describe how to populate and deal with AV* and SV* since ben> I am totally lost!! =) How to return an array from POSIX.xs: int uname() PPCODE: #ifdef HAS_UNAME struct utsname buf; if (uname(&buf) >= 0) { EXTEND(sp, 5); PUSHs(sv_2mortal(newSVpv(buf.sysname, 0))); PUSHs(sv_2mortal(newSVpv(buf.nodename, 0))); PUSHs(sv_2mortal(newSVpv(buf.release, 0))); PUSHs(sv_2mortal(newSVpv(buf.version, 0))); PUSHs(sv_2mortal(newSVpv(buf.machine, 0))); } #else How to populate an AV from 'man perlguts' void av_push(AV*, SV*); SV* av_pop(AV*); SV* av_shift(AV*); void av_unshift(AV*, I32 num); The whole perlguts manpage is really useful. ben> Any ideas? Hints? Or nice little padded cell? =) Hope, that helps, andreas ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <11134-2@ignite.demon.co.uk>; Tue, 17 Oct 1995 10:00:32 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 08:26:32 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa19566; 17 Oct 95 9:25 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa13765; 17 Oct 95 9:04 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA05814 for perldb-interest-real; Mon, 16 Oct 1995 23:52:35 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA05810 for ; Mon, 16 Oct 1995 23:52:33 -0700 Received: by gw.home.vix.com id AA04602; Mon, 16 Oct 95 23:54:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by anna.mind.de (940816.SGI.8.6.9/940406.SGI) id HAA17200; Tue, 17 Oct 1995 07:53:33 +0100 Date: Tue, 17 Oct 1995 07:53:33 +0100 Message-Id: <199510170653.HAA17200@anna.mind.de> From: Andreas Koenig To: mouring@netnet.net Cc: perldb-interest@vix.com In-Reply-To: (message from Ben Lindstrom on Mon, 16 Oct 1995 21:18:22 -0500 (CDT)) Subject: Re: Staticly linking DBI + QuickBase driver Reply-To: andreas.koenig@mind.de If I go e.g. into my TermReadKey-2.04 directory and say 'make perl', then the Makefile writes a brandnew perlmain.c for me which contains: static void xs_init() { char *file = __FILE__; { extern void boot_Term__ReadKey _((CV* cv)); newXS("Term::ReadKey::bootstrap", boot_Term__ReadKey, file); } AND I get a blib/auto/Term/ReadKey/ReadKey.a file which looks like so: k@anna:/usr/.../TermReadKey-2.04% nm -B blib/auto/Term/ReadKey/ReadKey.a | grep boot 00002c90 T boot_Term__ReadKey (the -B switch is for berkeley style output on my machine) You see, how they are considered to work together? Try it yourself with any of the well known solid extensions to find your deviation from the standard. I cannot guess it, sorry. andreas ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <13145-7@ignite.demon.co.uk>; Tue, 17 Oct 1995 12:43:03 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 11:12:11 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa24699; 17 Oct 95 12:11 +0100 Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id ae22962; 17 Oct 95 11:46 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <21398-0@frigate.doc.ic.ac.uk>; Tue, 17 Oct 1995 11:31:19 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA06218 for perldb-interest-real; Tue, 17 Oct 1995 01:59:17 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA06214 for ; Tue, 17 Oct 1995 01:59:11 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id JAA17660; Tue, 17 Oct 1995 09:57:12 +0100 From: Alligator Descartes Message-Id: <199510170857.JAA17660@fruitbat.mcqueen.com> Subject: Re: DBD for Oracle To: james.taylor@srs.gov Date: Tue, 17 Oct 1995 09:57:11 +0100 (BST) Cc: perldb-interest@fugue.com In-Reply-To: <199510162140.RAA12099@dark-knight.srs.gov> from "james.taylor@srs.gov" at Oct 16, 95 05:40:56 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 712 > I admit that I've haven't had time to test the code for quite some time... Is > anyone else doing anything? What is the current status of any improvements? You could check out: http://www.hermetica.com/technologia/DBI for ( reasonably ) up-to-date information on all aspects of perl and databases. The mSQL driver is relatively stable right now, and the drivers for Quickbase and Informix are nearly ready for alpha-testing. Some test programs are getting uploaded as well, including a Tk-based mSQL database tool thing by me, hopefully in a week or so. Hope that helps! > James Taylor > -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <18316-0@ignite.demon.co.uk>; Tue, 17 Oct 1995 16:53:37 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 15:48:20 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa14770; 17 Oct 95 16:47 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA07908 for perldb-interest-real; Tue, 17 Oct 1995 05:43:50 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA07904 for ; Tue, 17 Oct 1995 05:43:47 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id NAA21556; Tue, 17 Oct 1995 13:41:49 +0100 From: Alligator Descartes Message-Id: <199510171241.NAA21556@fruitbat.mcqueen.com> Subject: Re: Question on SV/AV/etc To: perldb-interest@fugue.com Date: Tue, 17 Oct 1995 13:41:48 +0100 (BST) Cc: mouring@netnet.net In-Reply-To: <9510171228.AA18982@dax.austin.ibm.com> from "mhm@austin.ibm.com" at Oct 17, 95 07:28:38 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1505 > |>Good news, as long as I don't have to return anything I can access the > |>database. =) Just fetchs and destruction problems seem to be the major > |>things I need to deal with (No, *blob* routines are not going to be > |>played with until I know at least what I'm doing with fetch. =) > |> > |>Any ideas? Hints? Or nice little padded cell? =) > > Having done the same for DB2....I'd vote for the padded cell! The only > things available are the perlguts pod and Tim's Oracle code. Yup. Definitely. I'm fixing (again!) the Informix code to do this. Funny how simple things like "It's returning a row!" give you such a kick! Usually only to discover it's returning the same row 30000 times.... 8-) Take a look at the DBD::mSQL code as well for another slightly different way to doing it. The way it's implemented in DBD::mSQL is that in dbdimp.c:dbd_describe, you fetch the row from the database and create an SV containing the data just before you return from dbd_describe. dbd_describe gets called in .xs:fetchrow and that just dumps what gets populated in dbd_describe. Short of seeing your code, I can't really give you much more help on it! I should add, that all this jiggery-pokery isn't a shortcoming of DBI, it's just a nightmare trying to fit wildly differing API's into the DBI-shaped hole........ > Mike > mhm@austin.ibm.com Regards. "Dubious-hacks-R-us" -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22090-5@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:06:47 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 19:35:23 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa12223; 17 Oct 95 20:34 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa13200; 17 Oct 95 17:47 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA08281 for perldb-interest-real; Tue, 17 Oct 1995 08:08:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA08277 for ; Tue, 17 Oct 1995 08:08:22 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id QAA26520; Tue, 17 Oct 1995 16:06:30 +0100 From: Alligator Descartes Message-Id: <199510171506.QAA26520@fruitbat.mcqueen.com> Subject: DBD::Informix available To: perldb-interest@fugue.com Date: Tue, 17 Oct 1995 16:06:29 +0100 (BST) Cc: Tim.Bunce@ig.co.uk, gbalfanz@netcom.com X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 611 A very developer release of DBD::Informix is now available at version 0.20pl0. Please don't use this driver against production databases right now! 8-) However, it does appear to work fairly reasonably so far. As I get feedback, more features and additions, fixes and more bugs will be added. So, go get it! Similarly, information is available from: http://www.hermetica.com/technologia/DBI Please let me know what happens if you try this driver! Feedback is essential at this stage. Thanks. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-5@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:06:56 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 17:47:22 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa27640; 17 Oct 95 18:46 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA08210 for perldb-interest-real; Tue, 17 Oct 1995 07:42:52 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA08206 for ; Tue, 17 Oct 1995 07:42:50 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id JAA96397; Tue, 17 Oct 1995 09:37:38 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for Tim.Bunce@ig.co.uk at austin.ibm.com; id AA27686; Tue, 17 Oct 1995 09:37:36 -0500 Message-Id: <9510171437.AA27686@dax.austin.ibm.com> To: perldb-interest@fugue.com Cc: chet@lusitania.watson.ibm.com, Tim.Bunce@ig.co.uk Subject: Re: Overriding malloc/free on AIX 3.2.5? Date: Tue, 17 Oct 95 09:37:36 -0500 Mike Moran writes: |> |> Chet Murthy writes: |> |> |> |>Hum ... OK -- a very ugly problem here. |> |> |> |>I was running a program using DBI::DB2, and had a bug which manifested |> |>itself as a segfault in free. I couldn't figure it out, but I noticed |> |>that free() in DB2.so didn't seem to be coming from perl -- that is, |> |>it wasn't being overridden. |> |> |> |>I tried various things, and finally concluded that, given that ld uses |> |>its import-list *last*, after all object-files and archive-files, it |> |>isn't possible to override free() so that all .so files will use |> |>Perl's implementation, unless free() is named something else. |> |> |> |>So I turned on the "HIDEMYMALLOC" flag, and put Myfree, Mymalloc, |> |>Myrealloc into perl.exp, recompiled everything, and all of a sudden, |> |>everything worked. |> |> |> |>I don't know what to do about this in the long run. It seems clear to |> |>me that Perl's standard method of overriding malloc with a private |> |>version (i.e., by just supplying malloc(), etc) isn't working for |> |>shared-library extensions (at least, not on AIX 3.2.5). |> |> |> |> Well, the problem is not AIX it is in either DBI or DB2. I haven't been |> far enough in the debugger yet to isolate which. It seems to be caused |> by a bad value in DBIc_NUM_FIELDS(imp_sth). |> |> More when I isolate it. Well, I was close. It turns out that the problem is in dbd_preparse which allocates a buffer in which it rewrites the sql statement with changes. Unfortunately, it used a statically determined string padding value, when Chet's code died it did so because he had long since overrun the malloced space. Below is a patch which should cure the problem without too much computational overhead and without too much wasted space. BTW, Tim, you need to apply this or a similar patch to Oracle/dbdimp.c Mike *** /tmp/T09xxIia Tue Oct 17 09:26:09 1995 --- dbdimp.c Tue Oct 17 09:20:24 1995 *************** *** 419,425 **** /* allocate room for copy of statement with spare capacity */ /* for editing ':1' into ':p1' so we can use obndrv. */ ! imp_sth->statement = (char*)safemalloc(strlen(statement) + 100); /* initialise phs ready to be cloned per placeholder */ memset(&phs_tpl, '\0',sizeof(phs_tpl)); --- 419,426 ---- /* allocate room for copy of statement with spare capacity */ /* for editing ':1' into ':p1' so we can use obndrv. */ ! imp_sth->statement = (char*)safemalloc(strlen(statement) + ! (DBIc_NUM_PARAMS(imp_sth)*4)); /* initialise phs ready to be cloned per placeholder */ memset(&phs_tpl, '\0',sizeof(phs_tpl)); ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22090-12@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:08:13 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 17 Oct 95 21:48:44 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa21446; 17 Oct 95 22:47 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa10613; 17 Oct 95 17:04 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA07867 for perldb-interest-real; Tue, 17 Oct 1995 05:26:25 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA07863 for ; Tue, 17 Oct 1995 05:26:23 -0700 From: mhm@austin.ibm.com Received: by gw.home.vix.com id AA26967; Tue, 17 Oct 95 05:28:41 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA80248; Tue, 17 Oct 1995 07:28:40 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@vix.com at austin.ibm.com; id AA18982; Tue, 17 Oct 1995 07:28:38 -0500 Message-Id: <9510171228.AA18982@dax.austin.ibm.com> To: Ben Lindstrom Cc: DBI perl Subject: Re: Question on SV/AV/etc In-Reply-To: Your message of Tue, 17 Oct 95 00:01:04 -0500. Date: Tue, 17 Oct 95 07:28:38 -0500 Ben Lindstrom writes: |> |>I'm to the point that I'm attempting to implement dbd_st_fetch() |> |>Now, I need to have someone point me to some nice documention (besides |>man perl* =) that describe how to populate and deal with AV* and SV* |>since I am totally lost!! =) |> |>Good news, as long as I don't have to return anything I can access the |>database. =) Just fetchs and destruction problems seem to be the major |>things I need to deal with (No, *blob* routines are not going to be |>played with until I know at least what I'm doing with fetch. =) |> |>Any ideas? Hints? Or nice little padded cell? =) |> Having done the same for DB2....I'd vote for the padded cell! The only things available are the perlguts pod and Tim's Oracle code. Mike mhm@austin.ibm.com ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22090-34@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:13:26 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 03:50:31 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa07647; 18 Oct 95 4:50 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa04365; 17 Oct 95 23:29 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA09325 for perldb-interest-real; Tue, 17 Oct 1995 14:12:09 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from tss0.bpt.sikorsky.com (tss0.bpt.sikorsky.com [140.76.28.5]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id OAA09321 for ; Tue, 17 Oct 1995 14:12:07 -0700 Received: by tss0.bpt.sikorsky.com (5.65/DEC-Ultrix/4.3) id AA15990; Tue, 17 Oct 1995 17:13:49 -0400 Received: by sun38.sikorsky (5.x/SMI-SVR4) id AA03216; Tue, 17 Oct 1995 17:14:10 -0400 Date: Tue, 17 Oct 1995 17:14:10 -0400 From: Dave Rattanni Message-Id: <9510172114.AA03216@sun38.sikorsky> To: perldb-interest@fugue.com Subject: Compiling DBD::Informix on Solaris 2.4 with ESQL 7.1 and GCC Cc: descarte@hermetica.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: V1DZEvBNoGbG7uI/05DkdQ== I was able to build DBD::Informix using Informix 7.1 ESQL libs. I suggest that the following be changed to account for the location and directory name differences 5.0 -> 7.1 ESQL. gcc -c -O -fpic -I/usr/local/lib/perl5/sun4-solaris/CORE -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion -Wtraditional -Wpointer-arith -Wcast-qual -I/opt/informix/include <- #### changed in ESQL 7.1 -I/usr/local/lib/perl5/sun4-solaris/DBI dbdimp.c dbdimp.c:1: sqlhdr.h: No such file or directory *** Error code 1 make: Fatal error: Command failed for target `dbdimp.o' In Makefile.PL: ********************************** 5. 0 ************************************** $opts{dynamic_lib} = { OTHERLDFLAGS => '-L$(INFORMIXDIR)/lib ' .$linkwith }; $opts{OBJECT} = '$(O_FILES) dbdimp.o'; $opts{macro} = { 'ESQL_LIB' => '$(INFORMIXDIR)/include' }; *********************************** 7.1 ************************************* $opts{dynamic_lib} = { OTHERLDFLAGS => '-L$(INFORMIXDIR)/lib ' .$linkwith }; $opts{OBJECT} = '$(O_FILES) dbdimp.o'; $opts{macro} = { 'ESQL_LIB' => '$(INFORMIXDIR)/incl/esql' }; Haven't run the tests yet. Dave ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22090-36@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:13:51 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 03:51:34 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa07912; 18 Oct 95 4:51 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa04409; 17 Oct 95 23:30 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA09259 for perldb-interest-real; Tue, 17 Oct 1995 13:56:10 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA09252 for ; Tue, 17 Oct 1995 13:56:05 -0700 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.ac02531; 16 Oct 95 11:01 +0100 Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id ad10200; 16 Oct 95 11:00 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <26193-0@ignite.demon.co.uk>; Mon, 16 Oct 1995 10:46:19 +0100 To: tim@www.cas.unt.edu, descarte@hermetica.com Subject: Re: DBD::mSQL driver expectancies. Cc: perldb-interest@fugue.com Date: Mon, 16 Oct 1995 10:46:19 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510161100.ad10200@relay-3.mail.demon.net> > From: Alligator Descartes > > > I'm trying to check out various interfaces to mSQL (for a WWW cgi > > implementation) and am having some problems with the current > > patch levels of DBI/DBD::mSQL. This is what I am working with: > > > > -rw-r--r-- 1 tim wheel 24775 Sep 20 21:11 /tmp/DBD-mSQL-0.60pl6.tar.gz > > -rw-r--r-- 1 tim wheel 45973 Sep 1 12:22 /tmp/DBI-0.63.tar.Z > > > > I receive this error when running ./test2.pl: > > > > DynaLoader.pm loaded (/lib /usr/lib /usr/local/lib) > > DynaLoader::bootstrap(DBI) > > DynaLoader::bootstrap(DBD::mSQL) > > install_driver(mSQL) failed: DBI version mismatch (DBI actual v3407879/s7, expec > > ted v4/s44) at /usr/local/lib/perl5/DynaLoader.pm line 463. > > > > Which leads me to believe that a newer version of DBI exists > > someplace (or I need to drop back a patch level or two on DBD::mSQL). > > Any clues? > > That combination of DBI/DBD::mSQL is the current working configuration ( bar a > couple of bugs in each piece! ), so you should be fine with it. > > What platform are you on? Also, what patchlevel of perl are you on. You need > to be at patchlevel 5.001m at least. > > If you could get that info to us, we can take a look from there. The message: DBI version mismatch (DBI actual v3407879/s7, expected v4/s44) means that when the driver was loaded and it located the DBI it checked the version and size fields in the dbistate structure and found them to be different from the values in the DBIXS.h file it was compiled with. See the definition of the DBISTATE_INIT macro in DBIXS.h. Looking at the 'DBI actual' values obviously something is corrupted. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-53@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:18:00 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 00:46:19 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id aj12232; 18 Oct 95 1:45 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <03488-0@frigate.doc.ic.ac.uk>; Tue, 17 Oct 1995 21:34:28 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA08905 for perldb-interest-real; Tue, 17 Oct 1995 12:03:17 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from srs.gov (bubba.srs.gov [192.33.240.5]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA08901 for ; Tue, 17 Oct 1995 12:02:33 -0700 From: james.taylor@srs.gov Received: by srs.gov id AA09878 (InterLock SMTP Gateway 1.1 for perldb-interest@fugue.com); Tue, 17 Oct 1995 15:03:14 -0400 Received: by srs.gov (Internal Mail Agent-1); Tue, 17 Oct 1995 15:03:14 -0400 Date: Tue, 17 Oct 1995 15:03:04 -0400 Message-Id: <199510171903.PAA13592@dark-knight.srs.gov> To: perldb-interest@fugue.com Subject: subscribe Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: S8woAZV8Lamk3swvYReagg== subscribe ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-65@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:19:41 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 00:46:26 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id an12232; 18 Oct 95 1:45 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <04994-0@frigate.doc.ic.ac.uk>; Tue, 17 Oct 1995 23:07:13 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA09171 for perldb-interest-real; Tue, 17 Oct 1995 13:30:41 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netnet1.netnet.net (netnet1.netnet.net [198.70.64.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA09167 for ; Tue, 17 Oct 1995 13:30:38 -0700 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id PAA13838; Tue, 17 Oct 1995 15:32:48 -0500 Date: Tue, 17 Oct 1995 15:32:48 -0500 (CDT) From: Ben Lindstrom To: Alligator Descartes cc: perldb-interest@fugue.com Subject: Re: Question on SV/AV/etc In-Reply-To: <199510171241.NAA21556@fruitbat.mcqueen.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 17 Oct 1995, Alligator Descartes wrote: > > Having done the same for DB2....I'd vote for the padded cell! The only > > things available are the perlguts pod and Tim's Oracle code. > > Yup. Definitely. I'm fixing (again!) the Informix code to do this. > Funny how simple things like "It's returning a row!" give you such a kick! > Usually only to discover it's returning the same row 30000 times.... > =) getting rows from that database in ObjC is a breeze, just since I'm not really too qualified to interface with perl. =) (Oh well..We all get stuck with new adventures. =) > > Take a look at the DBD::mSQL code as well for another slightly different > way to doing it. > Ok, I'm always willing to look at new code (And claim it's magic. =) > Short of seeing your code, I can't really give you much more help on it! > You'll get a chance. I'm releasing the first Alpha .02 for those to comment (and/or laugh at. =). > I should add, that all this jiggery-pokery isn't a shortcoming of DBI, it's > just a nightmare trying to fit wildly differing API's into the DBI-shaped > hole........ It's not really a matter of not understanding DBI. After I was able to get a copy up and running I was abe to see the interaction better. Just it's my lack of Perl compiler API that is dragging me over my head in quicksand. Someday it would be nice if a book was published that descibed the basics of writing perl modules with C, C++, and ObjC code added. Oh well...I can ONLY dream. =) --- Ben Lindstrom http://www.netnet.net/users/mouring "Trust me, I'm a better Network Admin then a programmer." ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-80@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:26:42 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 02:18:16 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa20929; 18 Oct 95 3:15 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA10268 for perldb-interest-real; Tue, 17 Oct 1995 18:11:02 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA10264 for ; Tue, 17 Oct 1995 18:11:00 -0700 Received: by gw.home.vix.com id AA22045; Tue, 17 Oct 95 18:13:14 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from disperse.demon.co.uk by relay-4.mail.demon.net id msg.gn05915; 18 Oct 95 2:05 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id at01607; 17 Oct 95 11:08 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa04698; 17 Oct 95 9:59 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <11111-0@ignite.demon.co.uk>; Tue, 17 Oct 1995 09:58:25 +0100 To: perldb-interest@vix.com, james.taylor@srs.gov Subject: Re: DBD for Oracle Date: Tue, 17 Oct 1995 09:58:25 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510171000.aa04698@relay-3.mail.demon.net> > From: james.taylor@srs.gov > > perldb-interest, > > I admit that I've haven't had time to test the code for quite some time... Is > anyone else doing anything? What is the current status of any improvements? How long is my piece of string? Grab a copy and read the Changes file. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-84@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:27:07 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 03:10:02 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa12406; 18 Oct 95 4:07 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA10437 for perldb-interest-real; Tue, 17 Oct 1995 19:12:38 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id TAA10433 for ; Tue, 17 Oct 1995 19:12:37 -0700 Received: by gw.home.vix.com id AA25789; Tue, 17 Oct 95 19:14:47 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by srs.gov id AA07037 (InterLock SMTP Gateway 1.1 for perldb-interest@vix.com); Tue, 17 Oct 1995 22:14:20 -0400 Received: by srs.gov (Internal Mail Agent-1); Tue, 17 Oct 1995 22:14:20 -0400 Date: Tue, 17 Oct 1995 22:10:57 -0400 From: James Taylor Subject: Re: DBD for Oracle To: Tim Bunce Cc: perldb-interest@vix.com Message-Id: <30846231.6697@dark-knight.srs.gov> Mime-Version: 1.0 X-Mailer: Mozilla 2.0b1J (X11; I; SunOS 5.4 sun4c) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit References: <9510171000.aa04698@relay-3.mail.demon.net> Tim Bunce wrote: > > > From: james.taylor@srs.gov > > > > perldb-interest, > > > > I admit that I've haven't had time to test the code for quite some time... Is > > anyone else doing anything? What is the current status of any improvements? > > How long is my piece of string? > > Grab a copy and read the Changes file. > > Tim. Silly, Silly, Silly me. I had supposed that since I hadn't seen any perldb-interest messages for a long time, that nothing was going on. However, I re-subscribed this afternoon and have already seen some messages. I must have somehow become unsubscribed! Now I just need to dig around and find the archive so that I can discover what good stuff I've missed in the last several months. Sorry for the bother, James ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-86@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:27:20 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 03:10:34 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa12653; 18 Oct 95 4:07 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA10443 for perldb-interest-real; Tue, 17 Oct 1995 19:12:43 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id TAA10439 for ; Tue, 17 Oct 1995 19:12:40 -0700 Received: by gw.home.vix.com id AA25792; Tue, 17 Oct 95 19:14:50 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by srs.gov id AA13435 (InterLock SMTP Gateway 1.1 for perldb-interest@vix.com); Tue, 17 Oct 1995 22:14:19 -0400 Received: by srs.gov (Internal Mail Agent-1); Tue, 17 Oct 1995 22:14:19 -0400 Date: Tue, 17 Oct 1995 22:11:01 -0400 From: James Taylor Subject: Re: DBD for Oracle To: Tim Bunce Cc: perldb-interest@vix.com Message-Id: <30846235.5B9B@dark-knight.srs.gov> Mime-Version: 1.0 X-Mailer: Mozilla 2.0b1J (X11; I; SunOS 5.4 sun4c) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit References: <9510171000.aa04698@relay-3.mail.demon.net> Tim Bunce wrote: > > > From: james.taylor@srs.gov > > > > perldb-interest, > > > > I admit that I've haven't had time to test the code for quite some time... Is > > anyone else doing anything? What is the current status of any improvements? > > How long is my piece of string? > > Grab a copy and read the Changes file. > > Tim. Silly, Silly, Silly me. I had supposed that since I hadn't seen any perldb-interest messages for a long time, that nothing was going on. However, I re-subscribed this afternoon and have already seen some messages. I must have somehow become unsubscribed! Now I just need to dig around and find the archive so that I can discover what good stuff I've missed in the last several months. Sorry for the bother, James ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-93@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:28:49 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 05:21:52 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ab04083; 18 Oct 95 6:19 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <15001-0@frigate.doc.ic.ac.uk>; Wed, 18 Oct 1995 05:27:49 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA10720 for perldb-interest-real; Tue, 17 Oct 1995 20:29:22 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA10716 for ; Tue, 17 Oct 1995 20:29:21 -0700 Received: by gw.home.vix.com id AA00642; Tue, 17 Oct 95 20:31:39 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id WAA06583; Tue, 17 Oct 1995 22:29:30 -0500 Date: Tue, 17 Oct 1995 22:29:30 -0500 (CDT) From: Ben Lindstrom To: DBI perl Subject: First release of QuickBase DBD for DBI. Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hopefully within the next day or so the driver will be accessable form the WWW page and the FTP site for DBI. But just wanted to let you all know I just put out my first Alpha of the driver if anyone cares to follow along when I beg for help. =) Soon..Soon..I'll have fetch.. =) ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <22089-95@ignite.demon.co.uk>; Wed, 18 Oct 1995 07:29:02 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 05:43:54 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa13301; 18 Oct 95 6:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA10903 for perldb-interest-real; Tue, 17 Oct 1995 21:37:17 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id VAA10899 for ; Tue, 17 Oct 1995 21:37:15 -0700 Received: by gw.home.vix.com id AA04513; Tue, 17 Oct 95 21:39:29 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from disperse.demon.co.uk by relay-4.mail.demon.net id msg.ax14369; 18 Oct 95 5:34 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id bz23789; 17 Oct 95 23:34 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ad20769; 17 Oct 95 17:38 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <19145-0@ignite.demon.co.uk>; Tue, 17 Oct 1995 17:38:36 +0100 To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: Question on SV/AV/etc Date: Tue, 17 Oct 1995 17:38:36 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510171738.ad20769@relay-3.mail.demon.net> > From: Ben Lindstrom > > I'm to the point that I'm attempting to implement dbd_st_fetch() > > Now, I need to have someone point me to some nice documention (besides > man perl* =) that describe how to populate and deal with AV* and SV* since > I am totally lost!! =) Post fragments of DBD::Oracle code and a description of what's not clear. > Any ideas? Hints? Or nice little padded cell? =) Sorry, my little padded cell is full right now (I just wish I could remember where I put the key, it's rather cramped in here :). Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <26709-10@ignite.demon.co.uk>; Wed, 18 Oct 1995 12:07:05 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 09:54:37 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa01011; 18 Oct 95 10:54 +0100 Received: from 204.254.239.2 by frigate.doc.ic.ac.uk with SMTP (PP) id <19580-0@frigate.doc.ic.ac.uk>; Wed, 18 Oct 1995 10:00:20 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA11211 for perldb-interest-real; Wed, 18 Oct 1995 00:00:03 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id XAA11200 for ; Tue, 17 Oct 1995 23:59:57 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.ac08439; 18 Oct 95 1:33 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id do25819; 18 Oct 95 0:57 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa15059; 17 Oct 95 17:07 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <18574-0@ignite.demon.co.uk>; Tue, 17 Oct 1995 17:07:05 +0100 To: perldb-interest@fugue.com, descarte@hermetica.com Subject: Re: Question on SV/AV/etc Cc: mouring@netnet.net Date: Tue, 17 Oct 1995 17:07:05 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510171707.aa15059@relay-3.mail.demon.net> > From: Alligator Descartes > > > |>Good news, as long as I don't have to return anything I can access the > > |>database. =) Just fetchs and destruction problems seem to be the major > > |>things I need to deal with (No, *blob* routines are not going to be > > |>played with until I know at least what I'm doing with fetch. =) > > |> > > |>Any ideas? Hints? Or nice little padded cell? =) > > > > Having done the same for DB2....I'd vote for the padded cell! The only > > things available are the perlguts pod and Tim's Oracle code. And I view the DBD::Oracle code as the 'reference implementation'. > Short of seeing your code, I can't really give you much more help on it! Ditto. > I should add, that all this jiggery-pokery isn't a shortcoming of DBI, it's > just a nightmare trying to fit wildly differing API's into the DBI-shaped > hole........ And the DBI-shaped hole will change shape again soon (ish). Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <27708-0@ignite.demon.co.uk>; Wed, 18 Oct 1995 12:55:40 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 11:53:47 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa25849; 18 Oct 95 12:52 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA12747 for perldb-interest-real; Wed, 18 Oct 1995 02:23:13 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id CAA12743 for ; Wed, 18 Oct 1995 02:23:08 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.an28756; 18 Oct 95 10:22 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id be22005; 17 Oct 95 22:54 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa18671; 17 Oct 95 17:26 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <19031-0@ignite.demon.co.uk>; Tue, 17 Oct 1995 17:25:53 +0100 To: perldb-interest@fugue.com, marcm@hpunix.gsfc.state.ga.us Subject: Re: Stored procedures and bind variables Date: Tue, 17 Oct 1995 17:25:53 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510171726.aa18671@relay-3.mail.demon.net> > From: "Marc M. from Tactics" > > Hi everybody! > > I am using DBI 0.63 and DBD:Oracle 0.25 on an Hp. > > Does anyone know how to run stored procedures and > pass values (from OUT, IN OUT parameters) back to perl? > > I cannot find any documentation on how to bind > variables. Is the only alternative to shell out > to sqlplus? There must be some way to do this. Change that to "There must be some way to do this in the future" and you'd be right. I've not implemented it yet. We're close though and I'll be working on DBD::Oracle again soon. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <00230-0@ignite.demon.co.uk>; Wed, 18 Oct 1995 17:01:24 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 12:49:03 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa08085; 18 Oct 95 13:48 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa21727; 18 Oct 95 13:33 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA12873 for perldb-interest-real; Wed, 18 Oct 1995 03:24:42 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA12869 for ; Wed, 18 Oct 1995 03:24:37 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id LAA24841; Wed, 18 Oct 1995 11:22:42 +0100 From: Alligator Descartes Message-Id: <199510181022.LAA24841@fruitbat.mcqueen.com> Subject: First release of QuickBase DBD for DBI. To: perldb-interest@fugue.com, mouring@netnet.net Date: Wed, 18 Oct 1995 11:22:41 +0100 (BST) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 621 > Hopefully within the next day or so the driver will be accessable form > the WWW page and the FTP site for DBI. > > But just wanted to let you all know I just put out my first Alpha of the > driver if anyone cares to follow along when I beg for help. =) > > Soon..Soon..I'll have fetch.. =) DBD::QBase ( as is appears to be called ) is available from: ftp://ftp.mcqueen.com/pub/dbperl/DBD/QBase/DBD-QBase-0.02.tar.gz Information on the driver is available at: http://www.hermetica.com/technologia/DBI -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <00247-1@ignite.demon.co.uk>; Wed, 18 Oct 1995 17:01:50 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 12:25:49 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa04525; 18 Oct 95 13:24 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <23036-0@frigate.doc.ic.ac.uk>; Wed, 18 Oct 1995 13:02:05 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA12793 for perldb-interest-real; Wed, 18 Oct 1995 02:45:26 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA12789 for ; Wed, 18 Oct 1995 02:45:20 -0700 Received: by gw.home.vix.com id AA23802; Wed, 18 Oct 95 02:47:36 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id EAA14667; Wed, 18 Oct 1995 04:45:58 -0500 Date: Wed, 18 Oct 1995 04:45:58 -0500 (CDT) From: Ben Lindstrom To: Tim Bunce Cc: perldb-interest@vix.com Subject: Re: Question on SV/AV/etc In-Reply-To: <9510171738.ad20769@relay-3.mail.demon.net> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 17 Oct 1995, Tim Bunce wrote: > > > From: Ben Lindstrom > > > > I'm to the point that I'm attempting to implement dbd_st_fetch() > > > > Now, I need to have someone point me to some nice documention (besides > > man perl* =) that describe how to populate and deal with AV* and SV* since > > I am totally lost!! =) > > Post fragments of DBD::Oracle code and a description of what's not clear. > It's not the matter of DBD::Oracle code I'm unsure of...Just I need to learn and understand how perl does it's dirty work for passing arrays around. Mainly the problem is learning the right seqents of commands to add items to an array. QuickBase request you to bind a column with a variable before the fetch. And then when the fetch occurs it knows where to put the data. This I understand and now how to implement. Just finding the command(s) to call to populate the AV array to pass back in my real problem. > > Any ideas? Hints? Or nice little padded cell? =) > > Sorry, my little padded cell is full right now (I just wish I could > remember where I put the key, it's rather cramped in here :). > That's ok...I think What Software will pay for my own 7ft by 4ft by 5ft padded cell 6 to 9 feet under.=) ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <01598-8@ignite.demon.co.uk>; Wed, 18 Oct 1995 17:56:10 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 16:48:26 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa21643; 18 Oct 95 17:47 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA13290 for perldb-interest-real; Wed, 18 Oct 1995 06:24:55 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA13286 for ; Wed, 18 Oct 1995 06:24:53 -0700 Received: by gw.home.vix.com id AA08091; Wed, 18 Oct 95 06:26:27 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.ao13185; 18 Oct 95 14:22 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id bw24097; 18 Oct 95 13:49 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ae09983; 18 Oct 95 12:06 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <26001-0@ignite.demon.co.uk>; Wed, 18 Oct 1995 10:45:04 +0100 To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: First release of QuickBase DBD for DBI. Date: Wed, 18 Oct 1995 10:45:04 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510181206.ae09983@relay-3.mail.demon.net> > From: Ben Lindstrom > > Hopefully within the next day or so the driver will be accessable form > the WWW page and the FTP site for DBI. > > But just wanted to let you all know I just put out my first Alpha of the > driver if anyone cares to follow along when I beg for help. =) Please register yourself and a URL for your module with the automated module server at http://franz.ww.tu-berlin.de/modulelist > Soon..Soon..I'll have fetch.. =) :-) ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <01880-0@ignite.demon.co.uk>; Wed, 18 Oct 1995 18:00:07 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 16:58:41 GMT Received: from [204.254.239.2] by punt-2.mail.demon.net id aa26671; 18 Oct 95 17:57 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA13446 for perldb-interest-real; Wed, 18 Oct 1995 07:35:51 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from rotor.sikorsky.com (rotor.sikorsky.com [140.76.24.221]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA13442 for ; Wed, 18 Oct 1995 07:35:47 -0700 Received: from dino.sikorsky.com by rotor.sikorsky.com (4.1/SMI-4.1) id AA12434; Wed, 18 Oct 95 10:33:45 EDT Received: from rotor.sikorsky.com by dino.sikorsky.com (4.1/SMI-4.1) id AA16747; Wed, 18 Oct 95 10:38:25 EDT Received: from dino.sikorsky.com by rotor.sikorsky.com (4.1/SMI-4.1) id AA12431; Wed, 18 Oct 95 10:33:37 EDT Received: from tss0.bpt.sikorsky.com by dino.sikorsky.com (4.1/SMI-4.1) id AA16731; Wed, 18 Oct 95 10:38:17 EDT Received: by tss0.bpt.sikorsky.com (5.65/DEC-Ultrix/4.3) id AA04294; Wed, 18 Oct 1995 07:50:26 -0400 Received: by sun38.sikorsky (5.x/SMI-SVR4) id AA03688; Wed, 18 Oct 1995 07:49:00 -0400 Date: Wed, 18 Oct 1995 07:49:00 -0400 From: Dave Rattanni Message-Id: <9510181149.AA03688@sun38.sikorsky> To: perldb-interest@fugue.com Subject: DBD::Informix on Solaris 2.4 with ESQL 7.1 Cc: descarte@hermetica.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: Ya6bhMT82Ajb+BiWJwn+0A== This is output from test.pl. >./test.pl install_driver(Informix) failed: Can't load '/usr/local/lib/perl5/sun4-solaris/auto/DBD/Informix/Informix.so' for module DBD::Informix: ld.so.1: /usr/bin/perl: fatal: relocation error: symbol not found: ASF_Call: referenced in /usr/local/lib/perl5/sun4-solaris/auto/DBD/Informix/Informix.so at line 450. at /usr/local/lib/perl5/DBD/Informix.pm line 18 at /usr/local/lib/perl5/DBI.pm line 167 DBI::install_driver called at ./test.pl line 12 ************************************************************************ The symbol lives in: >nm iasfs07a.so | grep ASF [299] | 40504| 1632|FUNC |GLOB |0 |7 |ASF_Call [450] | 71920| 416|FUNC |GLOB |0 |7 |nsASFTlAddr [495] | 46624| 44|FUNC |GLOB |0 |7 |getASFGLBCTL [553] | 131992| 192|FUNC |GLOB |0 |7 |ASF_SigSwitch [632] | 37048| 144|FUNC |GLOB |0 |7 |deleteASFCtl [727] | 35984| 380|FUNC |GLOB |0 |7 |defineASFCtl [732] | 43672| 228|FUNC |GLOB |0 |7 |ASF_Init Must this library also be loaded during the driver installation?? Dave ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <02600-1@ignite.demon.co.uk>; Wed, 18 Oct 1995 19:53:49 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 18:52:44 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa03379; 18 Oct 95 19:52 +0100 Received: from [204.254.239.2] by relay-1.mail.demon.net id aa12109; 18 Oct 95 18:37 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA13343 for perldb-interest-real; Wed, 18 Oct 1995 06:53:09 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA13339 for ; Wed, 18 Oct 1995 06:52:55 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.ac14633; 18 Oct 95 14:47 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id gi21243; 18 Oct 95 13:47 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ac09983; 18 Oct 95 12:06 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <25893-0@ignite.demon.co.uk>; Wed, 18 Oct 1995 10:31:15 +0100 To: descarte@hermetica.com Subject: Re: Happy with DBD-mSQL (fwd) Cc: perldb-interest@fugue.com Date: Wed, 18 Oct 1995 10:31:15 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510181206.ac09983@relay-3.mail.demon.net> > From: Alligator Descartes > > Tim, donno if you got one of these, but I reckon you should get a copy! > > It's stuff like this....... > > ------- > > Hi! > You haven't heard from me since I was struggling to get DBD-mSQL > running. > > Now I'm pleased to report that I've successfully completed an > application that makes heavy use of DBD-mSQL. It works on a big > database, uses 6 different cursors (three of them open simultaneously). > And it works like a charm! > > Use of the DBI::DBD features is still limited (I have some difficulty > to understand them -- like so many :-( ). I only use: $drh->connect, > $dbh->prepare(), $sth->execute, $sth->fetchrow, and $sth->finish. > But they seem to satify my needs. > > I'm impressed by the possibilities you and Tim Bunce have created for > me (and Larry Wall, of course). > > This is for moral support! > Greetings, > > ___ Marti Rijken ____ mrijken@inter.nl.net ____ Arnhem, Netherlands ___ > _______ Province of Gelderland, nature conservation department ______ > > Alligator Descartes | PGP key available on request! > descarte@hermetica.com | http://www.hermetica.com Thanks. And since it's also your good work I've CC'd it the list :-) Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <04404-9@ignite.demon.co.uk>; Thu, 19 Oct 1995 07:07:59 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 18 Oct 95 21:21:39 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa24467; 18 Oct 95 22:20 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA13906 for perldb-interest-real; Wed, 18 Oct 1995 10:17:47 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA13902 for ; Wed, 18 Oct 1995 10:17:43 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.bw19835; 18 Oct 95 17:00 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id bk26929; 18 Oct 95 14:54 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ac21751; 18 Oct 95 13:18 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <27893-0@ignite.demon.co.uk>; Wed, 18 Oct 1995 13:17:27 +0100 To: perldb-interest@fugue.com, mhm@austin.ibm.com Subject: Re: Overriding malloc/free on AIX 3.2.5? Cc: chet@lusitania.watson.ibm.com Date: Wed, 18 Oct 1995 13:17:27 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510181318.ac21751@relay-3.mail.demon.net> > From: mhm@austin.ibm.com > > Mike Moran writes: > |> > |> Chet Murthy writes: > |> |> > |> |>Hum ... OK -- a very ugly problem here. > |> |> > |> |>I was running a program using DBI::DB2, and had a bug which manifested > |> |>itself as a segfault in free. I couldn't figure it out, but I noticed > |> |>that free() in DB2.so didn't seem to be coming from perl -- that is, > |> |>it wasn't being overridden. > |> |> > |> |>I tried various things, and finally concluded that, given that ld uses > |> |>its import-list *last*, after all object-files and archive-files, it > |> |>isn't possible to override free() so that all .so files will use > |> |>Perl's implementation, unless free() is named something else. > |> |> > |> |>So I turned on the "HIDEMYMALLOC" flag, and put Myfree, Mymalloc, > |> |>Myrealloc into perl.exp, recompiled everything, and all of a sudden, > |> |>everything worked. > |> |> > |> |>I don't know what to do about this in the long run. It seems clear to > |> |>me that Perl's standard method of overriding malloc with a private > |> |>version (i.e., by just supplying malloc(), etc) isn't working for > |> |>shared-library extensions (at least, not on AIX 3.2.5). > |> |> > |> > |> Well, the problem is not AIX it is in either DBI or DB2. I haven't been > |> far enough in the debugger yet to isolate which. It seems to be caused > |> by a bad value in DBIc_NUM_FIELDS(imp_sth). > |> > |> More when I isolate it. > > Well, I was close. It turns out that the problem is in dbd_preparse > which allocates a buffer in which it rewrites the sql statement with > changes. Unfortunately, it used a statically determined string padding > value, when Chet's code died it did so because he had long since overrun > the malloced space. > > Below is a patch which should cure the problem without too much > computational overhead and without too much wasted space. > > BTW, Tim, you need to apply this or a similar patch to Oracle/dbdimp.c > > Mike > > /* allocate room for copy of statement with spare capacity */ > /* for editing ':1' into ':p1' so we can use obndrv. */ > ! imp_sth->statement = (char*)safemalloc(strlen(statement) + 100); > > --- 419,426 ---- > > /* allocate room for copy of statement with spare capacity */ > /* for editing ':1' into ':p1' so we can use obndrv. */ > ! imp_sth->statement = (char*)safemalloc(strlen(statement) + > ! (DBIc_NUM_PARAMS(imp_sth)*4)); Er, I don't think so. How is DBIc_NUM_PARAMS set before you've parsed the string ? Granted the +100 should actually be something bigger (or realloc inside the loop if we get close). Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <04405-27@ignite.demon.co.uk>; Thu, 19 Oct 1995 07:11:35 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 03:44:51 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa04367; 19 Oct 95 4:43 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa27825; 18 Oct 95 22:37 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA14014 for perldb-interest-real; Wed, 18 Oct 1995 11:03:31 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA14010 for ; Wed, 18 Oct 1995 11:03:29 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id NAA106293; Wed, 18 Oct 1995 13:05:47 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA11790; Wed, 18 Oct 1995 13:05:36 -0500 Message-Id: <9510181805.AA11790@dax.austin.ibm.com> To: Tim Bunce Cc: perldb-interest@fugue.com, mhm@austin.ibm.com, chet@lusitania.watson.ibm.com Subject: Re: Overriding malloc/free on AIX 3.2.5? In-Reply-To: Your message of Wed, 18 Oct 95 13:17:27 +0100. <9510181318.ac21751@relay-3.mail.demon.net> Date: Wed, 18 Oct 95 13:05:35 -0500 Tim Bunce writes: |> |> > From: mhm@austin.ibm.com |> > |> > Well, I was close. It turns out that the problem is in dbd_preparse |> > which allocates a buffer in which it rewrites the sql statement with |> > changes. Unfortunately, it used a statically determined string padding |> > value, when Chet's code died it did so because he had long since overrun |> > the malloced space. |> > |> > Below is a patch which should cure the problem without too much |> > computational overhead and without too much wasted space. |> > |> > BTW, Tim, you need to apply this or a similar patch to Oracle/dbdimp.c |> > |> > Mike |> > |> > /* allocate room for copy of statement with spare capacity */ |> > /* for editing ':1' into ':p1' so we can use obndrv. */ |> > ! imp_sth->statement = (char*)safemalloc(strlen(statement) + 100); |> > |> > --- 419,426 ---- |> > |> > /* allocate room for copy of statement with spare capacity */ |> > /* for editing ':1' into ':p1' so we can use obndrv. */ |> > ! imp_sth->statement = (char*)safemalloc(strlen(statement) + |> > ! (DBIc_NUM_PARAMS(imp_sth)*4)); |> |> Er, I don't think so. |> |> How is DBIc_NUM_PARAMS set before you've parsed the string ? |> |> Granted the +100 should actually be something bigger (or realloc |> inside the loop if we get close). |> At the point in time where DBD::Oracle calls oopen (dbdimp.c: db_st_prepare line 300 in version 0.25) DBD::DB2 has already parsed the statement and the number of parameters is available from the engine. Not knowing OCI I was assuming you had the same info available or at a minimun, needed to make one additional call to get it. From your comment above, I see I was mistaken. Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <04405-32@ignite.demon.co.uk>; Thu, 19 Oct 1995 07:12:18 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 04:05:10 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa09201; 19 Oct 95 5:04 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa03343; 18 Oct 95 23:49 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA14197 for perldb-interest-real; Wed, 18 Oct 1995 11:58:22 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA14193 for ; Wed, 18 Oct 1995 11:58:19 -0700 Received: by gw.home.vix.com id AA01252; Wed, 18 Oct 95 12:00:34 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from disperse.demon.co.uk by relay-4.mail.demon.net id g.ag19119; 18 Oct 95 16:30 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id ep26911; 18 Oct 95 15:27 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ab14131; 18 Oct 95 12:31 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <27513-0@ignite.demon.co.uk>; Wed, 18 Oct 1995 12:27:44 +0100 To: mouring@netnet.net Subject: Re: Question on SV/AV/etc Cc: perldb-interest@vix.com Date: Wed, 18 Oct 1995 12:27:44 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510181231.ab14131@relay-3.mail.demon.net> > From: Ben Lindstrom > > On Tue, 17 Oct 1995, Tim Bunce wrote: > > > > From: Ben Lindstrom > > > > > > I'm to the point that I'm attempting to implement dbd_st_fetch() > > > > > > Now, I need to have someone point me to some nice documention (besides > > > man perl* =) that describe how to populate and deal with AV* and SV* since > > > I am totally lost!! =) > > > > Post fragments of DBD::Oracle code and a description of what's not clear. > > It's not the matter of DBD::Oracle code I'm unsure of...Just I need to > learn and understand how perl does it's dirty work for passing arrays > around. > > Mainly the problem is learning the right seqents of commands to add > items to an array. > > QuickBase request you to bind a column with a variable before the fetch. > And then when the fetch occurs it knows where to put the data. This I > understand and now how to implement. Just finding the command(s) to > call to populate the AV array to pass back in my real problem. But that's what DBD::Oracle does. dbd_describe allocates an SV for each column and binds it to Oracle. dbd_st_fetch does the fetch and copies the columns to elements of the AV* returned by DBIS->get_fbav(imp_sth). All the code is there. Maybe you need to move/also bind in dbd_st_fetch but basically there should be little to change. NOTE: The further away your code is from DBD::Oracle's the harder it will be for you to track changes that I make to the implementation! Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <04406-49@ignite.demon.co.uk>; Thu, 19 Oct 1995 07:15:15 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 04:49:23 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id as14068; 19 Oct 95 5:48 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <06924-0@frigate.doc.ic.ac.uk>; Thu, 19 Oct 1995 01:55:42 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA15037 for perldb-interest-real; Wed, 18 Oct 1995 16:35:14 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA15033 for ; Wed, 18 Oct 1995 16:35:12 -0700 Received: by gw.home.vix.com id AA21600; Wed, 18 Oct 95 16:37:30 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from [205.149.164.200] (rmdudley.vip.best.com) by mail.crl.com with SMTP id AA12071 (5.65c/IDA-1.5 for ); Wed, 18 Oct 1995 16:36:34 -0700 Date: Wed, 18 Oct 1995 16:36:34 -0700 X-Sender: mrlizard@mail.crl.com Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: perldb-interest@vix.com From: Brian Valente Subject: website address? Hi, Sorry for the mundane questions - what's the URL of the companion web/ftp site to this list? Thanks, Brian =========================================================================== Brian Valente | brian_valente@rmdudley.com R.M. Dudley Corporation | mrlizard@crl.com 1799 Old Bayshore Suite 200 | http://www.rmdudley.com Burlingame, CA 94010 | 415-697-1650 tel. 415-697-1816 fax. =========================================================================== ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <04406-51@ignite.demon.co.uk>; Thu, 19 Oct 1995 07:15:40 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 04:50:27 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id at14068; 19 Oct 95 5:48 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <06957-0@frigate.doc.ic.ac.uk>; Thu, 19 Oct 1995 01:56:57 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA14979 for perldb-interest-real; Wed, 18 Oct 1995 16:18:37 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from spica.npac.syr.edu (root@spica-gw.npac.syr.edu [128.230.14.6]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA14975 for ; Wed, 18 Oct 1995 16:18:10 -0700 Received: from greatwall.npac.syr.edu by spica.npac.syr.edu (4.1/I-1.98K) id AA27370; Wed, 18 Oct 95 19:17:53 EDT Message-Id: <9510182317.AA27370@spica.npac.syr.edu> Date: Wed, 18 Oct 95 19:18:47 -0400 From: Gang Cheng X-Mailer: Mozilla 1.1N (X11; I; IRIX 5.3 IP22) Mime-Version: 1.0 To: hotjava-interest@java.sun.com, java-announce@java.eng.sun.com, java-interest@java.sun.com, java-porting@java.sun.com, perldb-interest@fugue.com, sp-discussion@mcs.anl.gov, WIN95-L@eva.dc.lsoft.com, www-vrml@wired.com, gcheng@npac.syr.edu Subject: A WWW archive system for USENET newsgroups and mailing-list(new and repost) X-Url: http://asknpac.npac.syr.edu/cgi-bin/news/showbody?bid=34000586&word=~1 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Please notice that we have dedicated a new server for this news archive and the previous URL () should be replaced by the new one: or . We have also added more newsgroups/mailing lists upon user's requests. --- Gang We at Northeast Parallel Architecture Center (NPAC), Syracue University has developed an archive system for searching/reading USENET newsgroups, mailing lists and personal mailbox from WWW. An Oracle database server is used to store/manage mails and two search/navigation interfaces accesible by any WWW browser to the archive are provided: one is an advanced search interface allowing queries with various options such as query by mail header, by date, by subject (keywords), by sender. The other is a Hypermail-like navigation interface for users familiar with Hypermail. We offer free access to this archive for the Internet community. Check the URL or . The whole archive is automatically updated in every 30 minutes. Currently the following newsgroups (total 67) are archived in the database, including this newsgroup/mailing-group: (..) shows the current total number of mails archived in the database alt.comp.compression (46) comp.ai (375) comp.ai.neural-nets (152) comp.ai.philosophy (464) comp.arch (549) comp.benchmarks (202) comp.compression (388) comp.compression.research (28) comp.databases (704) comp.databases.object (181) comp.databases.oracle (2461) comp.dcom.cell-relay (175) comp.dcom.frame-relay (149) comp.dcom.isdn (929) comp.dcom.lans.fddi (55) comp.dcom.lans.misc (71) comp.dcom.net-management (96) comp.dcom.servers (98) comp.dcom.sys.cisco (898) comp.dcom.telecom (380) comp.dcom.videoconf (149) comp.graphics.visualization (139) comp.infosystems (93) comp.infosystems.gis (406) comp.infosystems.www.advocacy (247) comp.infosystems.www.announce (869) comp.infosystems.www.authoring.cgi (1647) comp.infosystems.www.authoring.html (3995) comp.infosystems.www.authoring.images (679) comp.infosystems.www.authoring.misc (634) comp.infosystems.www.browsers.mac (681) comp.infosystems.www.browsers.misc (577) comp.infosystems.www.browsers.ms-windows (2095) comp.infosystems.www.browsers.x (467) comp.infosystems.www.misc (1157) comp.infosystems.www.servers.mac (256) comp.infosystems.www.servers.misc (209) comp.infosystems.www.servers.ms-windows (602) comp.infosystems.www.servers.unix (724) comp.lang.basic.visual.database (601) comp.lang.basic.visual.misc (2474) comp.lang.c++ (3764) comp.lang.java (1850) comp.lang.oberon (170) comp.lang.perl.announce (27) comp.lang.perl.misc (1702) comp.lang.rexx (265) comp.multimedia (1202) comp.os.ms-windows.programmer.misc (1546) comp.os.ms-windows.programmer.tools (493) comp.os.ms-windows.programmer.win32 (1163) comp.os.ms-windows.setup (608) comp.parallel (80) comp.parallel.mpi (41) comp.parallel.pvm (103) comp.protocols.tcp-ip (358) comp.security.misc (515) comp.software-eng (459) comp.sources.unix (8) comp.sys.sgi.graphics (188) comp.sys.sun.admin (948) comp.unix.admin (837) comp.unix.aix (1261) comp.unix.misc (337) comp.unix.programmer (658) comp.unix.solaris (1881) syr.general (218) The following mailing lists (total 8) are archived: hotjava-interest (mailing-list) (788) java-announce (mailing-list) (8) java-interest (mailing-list) (1479) java-porting (mailing-list) (279) perldb-interest (mailing-list) (173) sp-discussion (mailing-list) (51) win95-l@peach.ease.lsoft.com (mailing-list) (507) www-vrml (mailing-list) (3235) This system is still under development and more functions are planned to add to this system. Please send your comments to Gang Cheng (gcheng@npac.syr.edu). Thanks. --- Gang ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <05714-5@ignite.demon.co.uk>; Thu, 19 Oct 1995 11:59:10 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 08:55:46 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa20508; 19 Oct 95 9:53 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA15817 for perldb-interest-real; Wed, 18 Oct 1995 23:59:08 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA15813 for ; Wed, 18 Oct 1995 23:59:07 -0700 Received: by gw.home.vix.com id AA17557; Thu, 19 Oct 95 00:01:23 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from get.hooked.net (du07.inow.com [205.158.32.106]) by vader.inow.com (8.6.12/8.6.12) with SMTP id AAA11794 for ; Thu, 19 Oct 1995 00:01:16 -0700 Date: Thu, 19 Oct 1995 00:01:16 -0700 Message-Id: <199510190701.AAA11794@vader.inow.com> X-Sender: webworks@inow.com X-Mailer: Windows Eudora Light Version 1.5.2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: perldb-interest@vix.com From: Sarvesh Mathur Subject: subsc subscribe ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <07225-0@ignite.demon.co.uk>; Thu, 19 Oct 1995 15:12:40 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 11:33:23 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa17543; 19 Oct 95 12:32 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa12226; 19 Oct 95 12:19 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA16074 for perldb-interest-real; Thu, 19 Oct 1995 01:59:15 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA16070 for ; Thu, 19 Oct 1995 01:59:13 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id JAA06651; Thu, 19 Oct 1995 09:57:13 +0100 From: Alligator Descartes Message-Id: <199510190857.JAA06651@fruitbat.mcqueen.com> Subject: Re: website address? To: Brian Valente Date: Thu, 19 Oct 1995 09:57:11 +0100 (BST) Cc: perldb-interest@fugue.com In-Reply-To: from "Brian Valente" at Oct 18, 95 04:36:34 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 364 > Sorry for the mundane questions - what's the URL of the companion web/ftp > site to this list? I'm supposing it must be: http://www.hermetica.com/technologia/DBI and the general perl module one is: http://franz.ww.tu-berlin.de/modulelist > Brian -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <14676-0@ignite.demon.co.uk>; Thu, 19 Oct 1995 18:46:08 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 17:44:41 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa09398; 19 Oct 95 18:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA18053 for perldb-interest-real; Thu, 19 Oct 1995 08:41:11 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA18047 for ; Thu, 19 Oct 1995 08:41:09 -0700 Message-Id: <199510191541.IAA18047@toccata.fugue.com> To: perldb-interest@toccata.fugue.com Subject: Advertising the mailing list... Date: Thu, 19 Oct 1995 08:41:08 -0700 From: Ted Lemon The number of people who send mail to perldb-interest asking to be subscribed is too large to attribute to chance. I think the problem is bad advertising. If you mention the perldb mailing list to a friend, or if you advertise it in some regular posting, please don't describe it as ``perldb-interest@fugue.com.'' Instead, describe it as ``the Perldb mailing list (send mail to perldb-interest-request@fugue.com to subscribe).'' This way, you haven't mentioned the actual mailing list address in your message, so the recipient presumably won't think to send mail there. If you have a regular posting that you send out, please make the above change to it now so you don't forget. Thanks! _MelloN_ ---+++--- Received: from punt-4.mail.demon.net by ignite.demon.co.uk with SMTP (PP) id <16662-1@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:06:13 +0100 Received: from punt-4.mail.demon.net by mailstore for dbperl@ignite.demon.co.uk id <814153085:01674>; Fri, 20 Oct 95 01:38:05 GMT Received: from relay-1.mail.demon.net by punt-4.mail.demon.net id aa01569; 20 Oct 95 2:37 +0100 Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id ac17903; 19 Oct 95 22:51 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <28507-0@frigate.doc.ic.ac.uk>; Thu, 19 Oct 1995 22:01:59 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA18761 for perldb-interest-real; Thu, 19 Oct 1995 12:02:40 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA18757 for ; Thu, 19 Oct 1995 12:02:38 -0700 Received: by gw.home.vix.com id AA09540; Thu, 19 Oct 95 12:04:55 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from gandalf (gandalf [130.105.3.186]) by mailbox.osf.org (8.6.12/8.6.x) with SMTP id PAA01903; Thu, 19 Oct 1995 15:04:51 -0400 Date: Thu, 19 Oct 1995 15:04:51 -0400 Message-Id: <199510191904.PAA01903@mailbox.osf.org> X-Sender: dougm@postman.osf.org X-Mailer: Windows Eudora Version 2.0.3 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: perldb-interest@vix.com From: Doug MacEachern Subject: DBI/DBD::mSQL Cc: descarte@mcqueen.com OS: HP-UX A.09.05 Perl: 5.001m DBI: 0.63 DBD::mSQL: 0.60pl6 mSQL server: 1.0.8 We've got DBI installed, tested, working, using dynamic linking. The DBD::mSQL installation tries to force static linking for HP-UX, but we went ahead and commented out the following block: # Since HPUX can't mix -pic & non-pic (.a) in a dynamically loadable # file and oracle only supply .a libs we have to build static. if ($Config{osname} eq 'hpux' || $Config{osname} eq 'i386-sysv' ) { print "Warning: forced to build static not dynamic on $Config{osname}\n"; print " See README file for more information.\n"; $opts{LINKTYPE} = 'static'; } Should this be there? It looks like the Makefile.PL for mSQL used the Oracle Makfile.PL as a template and left this behind. ?? At HP-UX is not listed as a known working platform, has anyone else had success? It's installed, and the test2.pl scripts runs with a few warnings: -- /lib/dld.sl: Unresolved symbol: dbd_st_readblob (code) from /usr/local/lib/perl5/hpux/auto/DBD/mSQL/mSQL.sl -- But this is a known bug, correct? -- ** Disconnecting from second instance.... disconnect(DBI::db=HASH(0x400aa020)) invalidates 1 associated cursor(s) at test2.pl line 131. -- This doesn't seem to effect anything, and judging from the comments and code in mSQL.xs lines 120-126 or so, I should see this warning after a successful $dbh->disconnect, correct?? --- Summary of my perl5 (patchlevel 1) configuration: Platform: osname=hpux, osver=9, archname=hpux uname='hp-ux seagrams a.09.05 a 9000710 2007577379 two-user license ' hint=recommended Compiler: cc='cc', optimize='+O1' cppflags='-D_POSIX_SOURCE -D_HPUX_SOURCE -Aa' ccflags ='-D_POSIX_SOURCE -D_HPUX_SOURCE -Aa' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=sl libpth=/lib/pa1.1 /lib /usr/lib /usr/local/lib libs=-lm -ldld libc=/lib/libc.sl Dynamic Linking: dlsrc=dl_hpux.xs, dlext=sl, d_dlsymun= cccdlflags='+z', ccdlflags='-Wl,-E ', lddlflags='-b' --- So far through testing I'm very happy with the interface, and the overall DBI specification, thanks to all who have contributed to this development! -Doug ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <16665-17@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:11:14 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 00:50:16 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa09448; 20 Oct 95 1:49 +0100 Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id ab20740; 19 Oct 95 23:44 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <00593-0@frigate.doc.ic.ac.uk>; Thu, 19 Oct 1995 23:03:10 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA18906 for perldb-interest-real; Thu, 19 Oct 1995 12:55:49 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA18902 for ; Thu, 19 Oct 1995 12:55:44 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id UAA11893; Thu, 19 Oct 1995 20:53:42 +0100 From: Alligator Descartes Message-Id: <199510191953.UAA11893@fruitbat.mcqueen.com> Subject: Re: DBI/DBD::mSQL To: Doug MacEachern Date: Thu, 19 Oct 1995 20:53:41 +0100 (BST) Cc: perldb-interest@fugue.com, Tim.Bunce@ig.co.uk In-Reply-To: <199510191904.PAA01903@mailbox.osf.org> from "Doug MacEachern" at Oct 19, 95 03:04:51 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2261 > The DBD::mSQL installation tries to force static linking for HP-UX, but we > went ahead and commented out the following block: > > # Since HPUX can't mix -pic & non-pic (.a) in a dynamically loadable > # file and oracle only supply .a libs we have to build static. > if ($Config{osname} eq 'hpux' || $Config{osname} eq 'i386-sysv' ) { > print "Warning: forced to build static not dynamic on $Config{osname}\n"; > print " See README file for more information.\n"; > $opts{LINKTYPE} = 'static'; > } > > Should this be there? It looks like the Makefile.PL for mSQL used the > Oracle Makfile.PL as a template and left this behind. ?? Yup, that's exactly what's happened. If it makes no difference, I'll remove it from the Makefile.PL for the next release/patchlevel. > At > HP-UX is not listed as a known working platform, has anyone else had success? I have ( says the WWW site admin sheepishly... )......If others have, they haven't told me the platform info....... > It's installed, and the test2.pl scripts runs with a few warnings: > > -- > /lib/dld.sl: Unresolved symbol: dbd_st_readblob (code) from > /usr/local/lib/perl5/hpux/auto/DBD/mSQL/mSQL.sl > -- > But this is a known bug, correct? Yes, the dbd_st_readblob code got left in by mistake. It's out in the next release/patchlevel........ > -- > ** Disconnecting from second instance.... > disconnect(DBI::db=HASH(0x400aa020)) invalidates 1 associated cursor(s) at > test2.pl line 131. > -- > > This doesn't seem to effect anything, and judging from the comments and code > in mSQL.xs lines 120-126 or so, I should see this warning after a successful > $dbh->disconnect, correct?? Well, sort of. It's an unresolved issue in the DBI code, according to the Bunceman. To avoid the error, stick a: undef $whatever_your_cursor_is_called after the $whatever_you_cursor_is_called->finish; call. > So far through testing I'm very happy with the interface, and the overall > DBI specification, thanks to all who have contributed to this development! Wa-hoo! A satisfied customer! Thanks! > -Doug -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <16664-21@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:12:22 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 20:09:12 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ae00131; 19 Oct 95 21:08 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <24655-0@frigate.doc.ic.ac.uk>; Thu, 19 Oct 1995 18:46:35 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA18030 for perldb-interest-real; Thu, 19 Oct 1995 08:35:08 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA18026 for ; Thu, 19 Oct 1995 08:35:07 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id KAA29346 for ; Thu, 19 Oct 1995 10:37:24 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA23620; Thu, 19 Oct 1995 10:37:23 -0500 Message-Id: <9510191537.AA23620@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: How does Oracle handle LOBs? Date: Thu, 19 Oct 95 10:37:22 -0500 Looking at DBD::Oracle v0.25 it appears that OCI supports at a minimum piece meal return of a LOB. That brings up the question, does it support retrieving the entire LOB to a file? Since DB2 supports three types of LOBs (Binary, Character, Wide Character) there is more than one way to retrieve them. I was wondering If I should make suitable additions to the DB2.xs file for the additional variants or add a trailing argument to the existing readblob function. Any comments? Mike -------------------------------------------------------------------------- Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <16664-24@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:13:20 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 20:09:14 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id af00131; 19 Oct 95 21:08 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <24755-0@frigate.doc.ic.ac.uk>; Thu, 19 Oct 1995 18:48:50 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA18059 for perldb-interest-real; Thu, 19 Oct 1995 08:41:19 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA18048 for ; Thu, 19 Oct 1995 08:41:09 -0700 Received: by gw.home.vix.com id AA23592; Thu, 19 Oct 95 08:43:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us X400-Received: by mta bnr.ca in /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Thu, 19 Oct 1995 11:42:57 -0400 X400-Received: by /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Thu, 19 Oct 1995 11:42:48 -0400 X400-Received: by /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Thu, 19 Oct 1995 11:42:00 -0400 Date: Thu, 19 Oct 1995 11:42:00 -0400 X400-Originator: /dd.id=0146423/g=daniel/i=dg/s=cayouette/@bnr.ca X400-Mts-Identifier: [/PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/;bcars520.b.998:19.09.95.15.42.48] X400-Content-Type: P2-1984 (2) Content-Identifier: status of syb... From: "daniel (d.g.) cayouette" Message-Id: <"26001 Thu Oct 19 11:42:51 1995"@bnr.ca> To: perldb-interest@vix.com Cc: Tim.Bunce@ig.co.uk Subject: status of sybase perl? What is the status of perl for oracle, ingres and sybase? Daniel ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <16664-35@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:17:01 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 19 Oct 95 22:49:59 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa29880; 19 Oct 95 23:48 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA19194 for perldb-interest-real; Thu, 19 Oct 1995 14:17:32 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from spsgate.sps.mot.com (spsgate.sps.mot.com [192.70.231.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id OAA19190 for ; Thu, 19 Oct 1995 14:17:30 -0700 Received: from mogate (mogate.sps.mot.com) by spsgate.sps.mot.com (4.1/SMI-4.1/Email 2.1 10/25/93) id AA14596 for perldb-interest@fugue.com; Thu, 19 Oct 95 14:19:41 MST Received: from motsps by mogate (4.1/SMI-4.1/Email-2.0) id AA26401; Thu, 19 Oct 95 14:19:39 MST Received: from opus.psdc.sps.mot.com (opus-vlsi-az.sps.mot.com) by motsps (4.1/SMI-4.1/Email-2.1) id AA03855 for perldb-interest@fugue.com; Thu, 19 Oct 95 14:19:36 MST Received: from yosemite by opus.psdc.sps.mot.com with SMTP (1.37.109.6/16.2) id AA09729; Thu, 19 Oct 95 14:19:11 -0700 Received: by yosemite (1.38.193.4/16.2) id AA01397; Thu, 19 Oct 1995 14:19:33 -0700 From: Jim Searle Message-Id: <9510191419.ZM1395@yosemite.vlsi-az.sps.mot.com> Date: Thu, 19 Oct 1995 14:19:33 -0700 In-Reply-To: Alligator Descartes "Re: DBI/DBD::mSQL" (Oct 19, 8:53pm) References: <199510191953.UAA11893@fruitbat.mcqueen.com> X-Mailer: Z-Mail (3.2.1 31aug95) To: Alligator Descartes , Doug MacEachern Subject: Re: DBI/DBD::mSQL Cc: perldb-interest@fugue.com, Tim.Bunce@ig.co.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii On Oct 19, 8:53pm, Alligator Descartes wrote: > Subject: Re: DBI/DBD::mSQL > > > At > > HP-UX is not listed as a known working platform, has anyone else had > > success? > > I have ( says the WWW site admin sheepishly... )......If others have, they > haven't told me the platform info....... >From the help provided in this posting, and the help of Mark Pease, I too now have a working DBD::mSQL, and the test2.pl script passes with no warnings!! OS: HP-UX A.09.05 CC: HP92453-01 A.09.70 HP C Compiler Perl: 5.001m DBI: 0.63 DBD::mSQL: 0.60pl6 mSQL server: 1.0.8 > > So far through testing I'm very happy with the interface, and the overall > > DBI specification, thanks to all who have contributed to this development! > > Wa-hoo! A satisfied customer! > > Thanks! I agree, Thank You! jim -- | Jim Searle Email: Jim_Searle-ljs021@email.mot.com 602-752-2732 | | Motorola Phoenix Strategic Design Center (MS=BB106 LOC=AZ28) __ O | | 4625 South Ash Ave. Suite 12, Tempe Arizona 85282 _ \ <_ | |______________________________________________________________( )/ ( )__| ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <16664-44@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:19:27 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 00:05:50 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa05861; 20 Oct 95 1:04 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA19310 for perldb-interest-real; Thu, 19 Oct 1995 14:42:14 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA19306 for ; Thu, 19 Oct 1995 14:42:10 -0700 Received: by gw.home.vix.com id AA20949; Thu, 19 Oct 95 14:44:26 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from rijken.inter.NL.net by altrade.nijmegen.inter.nl.net via ah99-17.Arnhem.NL.net [193.79.228.242] with ESMTP id WAA17266 (8.6.10/1.8); Thu, 19 Oct 1995 22:44:19 +0100 Received: (from mrijken@localhost) by rijken.inter.NL.net (8.6.11/8.6.11) id WAA00490; Thu, 19 Oct 1995 22:43:05 +0100 Date: Thu, 19 Oct 1995 22:43:05 +0100 (MET) From: Marti Rijken To: Doug MacEachern Cc: perldb-interest@vix.com Subject: Re: DBI/DBD::mSQL In-Reply-To: <199510191904.PAA01903@mailbox.osf.org> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 19 Oct 1995, Doug MacEachern wrote: > [...] > Should this be there? It looks like the Makefile.PL for mSQL used the > Oracle Makfile.PL as a template and left this behind. ?? Yes, it seems to be left behind. DBD::mSQL was developed on a copy of DBD::Oracle, as is the recommended way to create a new driver. > /lib/dld.sl: Unresolved symbol: dbd_st_readblob (code) from > /usr/local/lib/perl5/hpux/auto/DBD/mSQL/mSQL.sl Here's a patch in which the reference is commented out. It works o.k. for me. The 'todo' list of Alligator Descartes mentions a clean-up of the code :-) -------------------------------------------------- *** mSQL.xs Thu Sep 21 20:02:10 1995 --- mSQL.xs.ORG Thu Sep 21 08:31:34 1995 *************** readblob(sth, field, offset, len, destsv *** 260,267 **** long len SV * destsv CODE: ! /* ST(0) = dbd_st_readblob(sth, field, offset, len, destsv); */ ! ST(0) = &sv_undef; /* discarded anyway, is it? */ void --- 260,266 ---- long len SV * destsv CODE: ! ST(0) = dbd_st_readblob(sth, field, offset, len, destsv); void -------------------------------------------------- > ** Disconnecting from second instance.... > disconnect(DBI::db=HASH(0x400aa020)) invalidates 1 associated cursor(s) at > test2.pl line 131. > > This doesn't seem to effect anything, and judging from the comments and code > in mSQL.xs lines 120-126 or so, I should see this warning after a successful > $dbh->disconnect, correct?? I've got the impression that all the application has to do is finish any open cursors. The DBI module will disconnect on exit. Can anyone confirm that this is as it should be? Please tell me if you get DBD::mSQL running on HPUX. I'm using DBD::mSQL to my full satisfaction at home on Linux 1.2.13, but it might even be an alternative for our Oracle database on HPUX at work... Regards, ___ Marti Rijken ____ mrijken@inter.nl.net ____ Arnhem, Netherlands ___ _______ Province of Gelderland, nature conservation department ______ ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <16664-53@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:20:51 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 00:52:55 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa25149; 20 Oct 95 1:50 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA19590 for perldb-interest-real; Thu, 19 Oct 1995 16:19:43 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA19586 for ; Thu, 19 Oct 1995 16:19:36 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id msg.ab08675; 20 Oct 95 0:16 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id fa00285; 19 Oct 95 23:29 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa05631; 19 Oct 95 17:23 +0100 Date: Thu, 19 Oct 1995 17:22:52 +100 (BST) From: Paul Mahoney X-Sender: ptm@xact4.xact.com To: perldb-interest@fugue.com Subject: Memory leak in perl 4.036 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I believe I've found a memory leak in perl 4.036. Please don't tell me to move over to perl 5, I can't at the moment :-( I haven't got the solution yet ... I'm hoping the team have a fix for this already :-) When you call a subroutine from within a "foreach" loop a little bit more memory is grabbed by perl and not released. My script executes a such a loop an EXTREMELY LARGE number of times... after 4 hours I was getting the following error from perl: Out of memory! Basically malloc() finally refused to given me any more memory. It's taken me a long time to pin the problem down and I'm VERY sure my summary of the bug is correct. Is this a known problem? Has someone got a patch to fix it? Thanks --- Paul Mahoney, X-Act Solutions Limited smail: Owlsmead, Meads Road, Little Common, Bexhill-on-Sea, East Sussex TN39 4SY email: ptm@xact.demon.co.uk ... pmahoney@cix.compulink.co.uk phone: +44 424 846368 ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <16664-72@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:23:46 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 04:56:05 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa25646; 20 Oct 95 5:55 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA19907 for perldb-interest-real; Thu, 19 Oct 1995 20:27:15 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA19903 for ; Thu, 19 Oct 1995 20:27:13 -0700 Received: by gw.home.vix.com id AA01900; Thu, 19 Oct 95 20:29:26 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from hpcc123.corp.hp.com (okamoto.corp.hp.com) by hp.com with ESMTP (1.37.109.16/15.5+ECS 3.3) id AA085943544; Thu, 19 Oct 1995 15:59:05 -0700 Received: by hpcc123.corp.hp.com (1.37.109.16/15.5+ECS 3.3) id AA129623543; Thu, 19 Oct 1995 15:59:03 -0700 From: Jeff Okamoto Message-Id: <199510192259.AA129623543@hpcc123.corp.hp.com> Subject: Re: DBI/DBD::mSQL To: dougm@osf.org Date: Thu, 19 Oct 95 15:59:03 PDT Cc: perldb-interest@vix.com, descarte@mcqueen.com In-Reply-To: <199510191904.PAA01903@mailbox.osf.org>; from "Doug MacEachern" at Oct 19, 95 3:04 pm Mailer: Elm [revision: 70.85] > # Since HPUX can't mix -pic & non-pic (.a) in a dynamically loadable > # file and oracle only supply .a libs we have to build static. > if ($Config{osname} eq 'hpux' || $Config{osname} eq 'i386-sysv' ) { > print "Warning: forced to build static not dynamic on $Config{osname}\n"; > print " See README file for more information.\n"; > $opts{LINKTYPE} = 'static'; > } This is somewhat incorrect. You *can* mix PIC and non-PIC object files in a .a library, but you cannot link a non-PIC object file into a shared library. I have received .a archives that contain PIC modules and can thus turn them into shared libraries. Perhaps the comment could be rewritten as: # HP-UX cannot link a non-PIC object file into a shared library. Since the # .a libs that Oracle supplies contain non-PIC object files, we have to # build static. Jeff ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <19623-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 07:30:46 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 06:27:01 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa22571; 20 Oct 95 7:26 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id WAA20234 for perldb-interest-real; Thu, 19 Oct 1995 22:11:13 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id WAA20230 for ; Thu, 19 Oct 1995 22:11:12 -0700 Received: by gw.home.vix.com id AA10156; Thu, 19 Oct 95 22:13:28 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id AAA12848; Fri, 20 Oct 1995 00:13:27 -0500 Date: Fri, 20 Oct 1995 00:13:26 -0500 (CDT) From: Ben Lindstrom To: DBI perl Subject: Another light comes on (QBase/Oracle) Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Since I have a working copy of QBase driver I printed both QBase and Oracle DBD and started going through the code line by line...And it's rather embarassing what I learned. =) You know all those NICE structures that were gracefully defined? Well.. I basicly commented them out and ignored them. Oh well..had to start somewhere. =) Now.. I mainly understand imp_sth_st structure...but since I don't know Oracle..I'm unsure what the following are used in imp_dbd_st. Lda_Def lda; ub1 hda[HDA_SIZE]; Also the whole imp_fbh_st. Niether one seems to have any direct affect to me. I see how imp_sth_st contains reference information to how to access the database (like currenty "file descripter" and other required global typedata). So far the rest I understand..The QuickBase ObjC object takes away a lot of the code and hides it...So most of my work is sorting through and figuring out what QuickBase handles and what I have to handle. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <25123-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 16:54:34 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 15:43:59 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa11403; 20 Oct 95 16:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA22108 for perldb-interest-real; Fri, 20 Oct 1995 06:32:23 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA22104 for ; Fri, 20 Oct 1995 06:32:18 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.az22363; 20 Oct 95 14:29 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id al09207; 20 Oct 95 13:39 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa13549; 20 Oct 95 12:53 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <22827-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 12:01:52 +0100 To: okamoto@hpcc123.corp.hp.com Subject: Re: DBI/DBD::mSQL Cc: perldb-interest@fugue.com Date: Fri, 20 Oct 1995 12:01:52 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510201253.aa13549@relay-3.mail.demon.net> > From: Jeff Okamoto > > Perhaps the comment could be rewritten as: > > # HP-UX cannot link a non-PIC object file into a shared library. Since the > # .a libs that Oracle supplies contain non-PIC object files, we have to > # build static. Done. Thanks. Tim. ---+++--- Received: from punt3.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03032-10@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:07:36 +0100 Received: from punt-3.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 19:17:10 GMT Received: from relay-1.mail.demon.net by punt-3.mail.demon.net id aa14373; 20 Oct 95 20:16 +0100 Received: from toccata.fugue.com by relay-1.mail.demon.net id aa06811; 20 Oct 95 19:22 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA22614 for perldb-interest-real; Fri, 20 Oct 1995 09:32:20 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA22610 for ; Fri, 20 Oct 1995 09:32:14 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id RAA18366; Fri, 20 Oct 1995 17:28:41 +0100 From: Alligator Descartes Message-Id: <199510201628.RAA18366@fruitbat.mcqueen.com> Subject: Re: First release of QuickBase DBD for DBI. To: Tim Bunce Date: Fri, 20 Oct 1995 17:28:40 +0100 (BST) Cc: perldb-interest@fugue.com In-Reply-To: <9510201507.ao05588@relay-3.mail.demon.net> from "Tim Bunce" at Oct 20, 95 02:46:59 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 798 > > DBD::QBase ( as is appears to be called ) is available from: > > > > ftp://ftp.mcqueen.com/pub/dbperl/DBD/QBase/DBD-QBase-0.02.tar.gz > > > > Information on the driver is available at: > > > > http://www.hermetica.com/technologia/DBI > > I think we should be using the CPAN to distribute database drivers. > > Can all driver developers please vist > > http://franz.ww.tu-berlin.de/modulelist > > and register themselves. If possible please provide a URL (maybe ftp://...) > from which your module can automatically be retrieved and mirrored > by the Module Server. As far as I know, the DBD::Informix and DBD::mSQL drivers are already there and get mirrored....... > Tim. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03031-12@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:08:15 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 17:37:34 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa10980; 20 Oct 95 18:36 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA22378 for perldb-interest-real; Fri, 20 Oct 1995 08:23:34 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA22373 for ; Fri, 20 Oct 1995 08:23:30 -0700 Received: by gw.home.vix.com id AA20811; Fri, 20 Oct 95 08:25:43 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.bu26681; 20 Oct 95 15:55 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id ad13862; 20 Oct 95 15:17 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ac05588; 20 Oct 95 15:07 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <23461-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 13:08:39 +0100 To: dougm@osf.org, mrijken@solair1.inter.nl.net Subject: Re: DBI/DBD::mSQL Cc: perldb-interest@vix.com Date: Fri, 20 Oct 1995 13:08:39 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510201507.ac05588@relay-3.mail.demon.net> > From: Marti Rijken > > On Thu, 19 Oct 1995, Doug MacEachern wrote: > > > disconnect(DBI::db=HASH(0x400aa020)) invalidates 1 associated cursor(s) at > > test2.pl line 131. > > > > This doesn't seem to effect anything, and judging from the comments and code > > in mSQL.xs lines 120-126 or so, I should see this warning after a successful > > $dbh->disconnect, correct?? > > I've got the impression that all the application has to do is finish any > open cursors. The DBI module will disconnect on exit. Can anyone > confirm that this is as it should be? That's okay. The warning should not count inactive cursors. It should get fixed. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03031-26@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:13:50 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 21 Oct 95 01:24:47 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa29706; 21 Oct 95 0:36 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA23893 for perldb-interest-real; Fri, 20 Oct 1995 15:24:00 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA23889 for ; Fri, 20 Oct 1995 15:23:59 -0700 Received: by gw.home.vix.com id AA27264; Fri, 20 Oct 95 15:26:13 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.br26681; 20 Oct 95 15:55 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id ad13878; 20 Oct 95 15:17 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ag05588; 20 Oct 95 15:07 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <23607-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 13:42:43 +0100 To: perldb-interest@vix.com, cayouett@bnr.ca Subject: Re: status of sybase perl? Date: Fri, 20 Oct 1995 13:42:43 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510201507.ag05588@relay-3.mail.demon.net> > From: "daniel (d.g.) cayouette" > > What is the status of perl for oracle, ingres and sybase? Check out http://www.hermetica.com/technologia/DBI/DBD/index.html Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03033-35@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:15:16 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 23:19:35 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa03358; 21 Oct 95 0:18 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <21475-0@frigate.doc.ic.ac.uk>; Fri, 20 Oct 1995 18:46:43 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA22291 for perldb-interest-real; Fri, 20 Oct 1995 08:06:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA22287 for ; Fri, 20 Oct 1995 08:06:36 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.al26824; 20 Oct 95 15:55 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id ad13917; 20 Oct 95 15:17 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id af05588; 20 Oct 95 15:07 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <23589-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 13:36:56 +0100 To: perldb-interest@fugue.com, mhm@austin.ibm.com Subject: Re: How does Oracle handle LOBs? Date: Fri, 20 Oct 1995 13:36:56 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510201507.af05588@relay-3.mail.demon.net> > From: mhm@austin.ibm.com > > Looking at DBD::Oracle v0.25 it appears that OCI supports at a minimum > piece meal return of a LOB. That brings up the question, does it > support retrieving the entire LOB to a file? No, but it could be added as a default method based on readblob which a driver could override if required. I'd probably do it as: $sth->copy_blob_to_file(1, $open_file_handle); I'll add it to the Todo file. > Since DB2 supports three types of LOBs (Binary, Character, Wide Character) > there is more than one way to retrieve them. I was wondering If I should > make suitable additions to the DB2.xs file for the additional variants > or add a trailing argument to the existing readblob function. The $sth should carry enough information for read_blob to know what type a given field is. Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03033-43@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:16:55 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 23:19:44 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ag03358; 21 Oct 95 0:18 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <22373-0@frigate.doc.ic.ac.uk>; Fri, 20 Oct 1995 19:23:31 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA22581 for perldb-interest-real; Fri, 20 Oct 1995 09:28:15 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA22577 for ; Fri, 20 Oct 1995 09:28:08 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id sg.ah26892; 20 Oct 95 15:56 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id af13853; 20 Oct 95 15:17 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ao05588; 20 Oct 95 15:07 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <24055-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 14:46:59 +0100 To: perldb-interest@fugue.com, mouring@netnet.net, descarte@hermetica.com Subject: Re: First release of QuickBase DBD for DBI. Cc: k@franz.ww.tu-berlin.de Date: Fri, 20 Oct 1995 14:46:59 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510201507.ao05588@relay-3.mail.demon.net> > From: Alligator Descartes > > > Hopefully within the next day or so the driver will be accessable form > > the WWW page and the FTP site for DBI. > > > > But just wanted to let you all know I just put out my first Alpha of the > > driver if anyone cares to follow along when I beg for help. =) > > > > Soon..Soon..I'll have fetch.. =) > > DBD::QBase ( as is appears to be called ) is available from: > > ftp://ftp.mcqueen.com/pub/dbperl/DBD/QBase/DBD-QBase-0.02.tar.gz > > Information on the driver is available at: > > http://www.hermetica.com/technologia/DBI I think we should be using the CPAN to distribute database drivers. Can all driver developers please vist http://franz.ww.tu-berlin.de/modulelist and register themselves. If possible please provide a URL (maybe ftp://...) from which your module can automatically be retrieved and mirrored by the Module Server. Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03033-47@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:17:45 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 23:19:47 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id am03358; 21 Oct 95 0:19 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <24896-0@frigate.doc.ic.ac.uk>; Fri, 20 Oct 1995 21:08:40 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA23094 for perldb-interest-real; Fri, 20 Oct 1995 11:19:03 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from eve.mscs.mu.edu (eve.mscs.mu.edu [134.48.4.166]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA23090 for ; Fri, 20 Oct 1995 11:19:01 -0700 Received: by eve.mscs.mu.edu (Smail3.1.28.1 #9) id m0t6M07-0002DlC; Fri, 20 Oct 95 13:17 CDT Message-Id: From: BCL HTML account Subject: DBI Installation To: perldb-interest@fugue.com Date: Fri, 20 Oct 1995 13:17:02 -0500 (CDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2519 Hi, We are getting segmentation fault while running make test of DBI module. perl Makefile.PL and make are working. Following is the output of make test. It is giving an error on test 5. Please HELP!! -Thanks Malathi eve:/usr/local/www# make test TEST_VERBOSE=1 Makefile:333: warning: overriding commands for target `/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib/DBI.pm' Makefile:298: warning: ignoring old commands for target `/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib/DBI.pm' /usr/local/bin/perl -I/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib -I/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib -I/usr/local/lib/perl5.0/lib/sun4-solaris -I/usr/local/lib/perl5.0/lib -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............1..6 ok 1 ok 2 ok 3 DBD::ExampleP in /usr/local/lib/perl5.0/lib is hidden by DBD::ExampleP in /rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib at t/min.t line 16 DBD::NullP in /usr/local/lib/perl5.0/lib is hidden by DBD::NullP in /rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib at t/min.t line 16 ok 4 FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.37 cusr 0.22 csys = 0.59 cpu) /usr/local/bin/perl -I/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib -I/rsch_oracle/workarea/www/tools/GATEWAY/DBI/blib -I/usr/local/lib/perl5.0/lib/sun4-solaris -I/usr/local/lib/perl5.0/lib test.pl test.pl DBI test application $Revision: 1.12 $ make: *** [test] Segmentation Fault (core dumped) script done on Fri Oct 20 13:06:49 1995 ******************************************************************************* ***** perl -e 'use Config; print Config::myconfig()' ******** ******************************************************************************* Summary of my perl5 (patchlevel 1) configuration: Platform: osname=solaris, osver=2.4, archname=sun4-solaris uname='sunos eve 5.4 generic_101945-27 sun4m sparc ' hint=recommended Compiler: cc='gcc', optimize='-O' cppflags='' ccflags ='' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/lib /usr/lib /usr/ccs/lib /usr/local/lib libs=-lsocket -lnsl -ldb -ldl -lm -lc -lcrypt libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-fpic', ccdlflags='', lddlflags='-G' ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03033-49@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:18:24 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 20 Oct 95 23:19:45 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ai03358; 21 Oct 95 0:18 +0100 Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <23500-0@frigate.doc.ic.ac.uk>; Fri, 20 Oct 1995 20:15:04 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA22778 for perldb-interest-real; Fri, 20 Oct 1995 10:16:58 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA22774 for ; Fri, 20 Oct 1995 10:16:56 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id MAA43059 for ; Fri, 20 Oct 1995 12:19:13 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA22024; Fri, 20 Oct 1995 12:19:12 -0500 Message-Id: <9510201719.AA22024@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: Anyone have successfully using blobs in DBD::Oracle? Date: Fri, 20 Oct 95 12:19:11 -0500 Just curious, I'm having problems getting the data from readblob back into my perl script variable under DB2. Thus, the obvious question has anyone tried it (readblob under Oracle)? Mike --------------------------------------------------------------------------- Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03031-48@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:18:37 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 21 Oct 95 08:05:43 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa01527; 21 Oct 95 9:05 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA24543 for perldb-interest-real; Sat, 21 Oct 1995 00:02:13 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id AAA24539 for ; Sat, 21 Oct 1995 00:02:10 -0700 Received: from disperse.demon.co.uk by relay-4.mail.demon.net id msg.cc26681; 20 Oct 95 15:56 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id ae13901; 20 Oct 95 15:17 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ap05588; 20 Oct 95 15:07 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <24075-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 14:50:04 +0100 To: mhm@austin.ibm.com Subject: Re: Overriding malloc/free on AIX 3.2.5? Cc: perldb-interest@fugue.com, chet@lusitania.watson.ibm.com Date: Fri, 20 Oct 1995 14:50:04 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510201507.ap05588@relay-3.mail.demon.net> > From: mhm@austin.ibm.com > > |> > /* allocate room for copy of statement with spare capacity */ > |> > /* for editing ':1' into ':p1' so we can use obndrv. */ > |> > ! imp_sth->statement = (char*)safemalloc(strlen(statement) + > |> > ! (DBIc_NUM_PARAMS(imp_sth)*4)); > |> > |> Er, I don't think so. > |> > |> How is DBIc_NUM_PARAMS set before you've parsed the string ? > > At the point in time where DBD::Oracle calls oopen (dbdimp.c: db_st_prepare > line 300 in version 0.25) DBD::DB2 has already parsed the statement and > the number of parameters is available from the engine. So what is dbd_preparse() doing? Why not get the info from the DB2 API? I presume DB2 use '?' as the placeholder. Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03033-267@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:47:17 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sat, 21 Oct 95 08:52:54 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa23402; 21 Oct 95 9:52 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA24650 for perldb-interest-real; Sat, 21 Oct 1995 00:51:09 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA24646 for ; Sat, 21 Oct 1995 00:51:07 -0700 Received: by gw.home.vix.com id AA08332; Sat, 21 Oct 95 00:53:22 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from disperse.demon.co.uk by relay-4.mail.demon.net id msg.bf21862; 20 Oct 95 14:21 +0100 Received: from post.demon.co.uk by relay-2.mail.demon.net id at09207; 20 Oct 95 13:39 +0100 Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ai13549; 20 Oct 95 12:54 +0100 Received: from ig.co.uk by ignite.demon.co.uk id <23137-0@ignite.demon.co.uk>; Fri, 20 Oct 1995 12:50:56 +0100 To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: Another light comes on (QBase/Oracle) Date: Fri, 20 Oct 1995 12:50:56 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510201254.ai13549@relay-3.mail.demon.net> > From: Ben Lindstrom > > Since I have a working copy of QBase driver I printed both QBase and > Oracle DBD and started going through the code line by line...And it's > rather embarassing what I learned. =) > > You know all those NICE structures that were gracefully defined? Well.. > I basicly commented them out and ignored them. Oh well..had to start > somewhere. =) > > Now.. I mainly understand imp_sth_st structure...but since I don't know > Oracle..I'm unsure what the following are used in imp_dbd_st. > Lda_Def lda; > ub1 hda[HDA_SIZE]; Basically a working area for an oracle database connection. It's oracle specific. > Also the whole imp_fbh_st. Niether one seems to have any direct affect > to me. Affect? > I see how imp_sth_st contains reference information to how to > access the database (like currenty "file descripter" and other required > global typedata). Are you asking a question? Anyway, imp_fbh_st stuff is very likely to change radically 'soon' :-) Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <03033-326@ignite.demon.co.uk>; Mon, 23 Oct 1995 07:55:04 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 22 Oct 95 16:41:35 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa06390; 22 Oct 95 16:40 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <04972-0@frigate.doc.ic.ac.uk>; Sun, 22 Oct 1995 17:38:41 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA29199 for perldb-interest-real; Sun, 22 Oct 1995 07:43:42 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from tss0.bpt.sikorsky.com (tss0.bpt.sikorsky.com [140.76.28.5]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA29195 for ; Sun, 22 Oct 1995 07:43:40 -0700 Received: by tss0.bpt.sikorsky.com (5.65/DEC-Ultrix/4.3) id AA03927; Sun, 22 Oct 1995 10:45:25 -0400 Received: by sun38.sikorsky (5.x/SMI-SVR4) id AA06918; Sun, 22 Oct 1995 10:45:51 -0400 Date: Sun, 22 Oct 1995 10:45:51 -0400 From: Dave Rattanni Message-Id: <9510221445.AA06918@sun38.sikorsky> To: perldb-interest@fugue.com Subject: DBD::Informix/Solaris 2.4 and ESQLC 7.1 As suggested, I ran test.pl on a local Informix 7.1 SE database and it ran all tests without a hitch. Congrats to Alligator, et al! ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <19842-0@ignite.demon.co.uk>; Mon, 23 Oct 1995 21:31:08 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 23 Oct 95 19:30:05 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa25190; 23 Oct 95 19:28 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA02231 for perldb-interest-real; Mon, 23 Oct 1995 06:21:38 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA02227 for ; Mon, 23 Oct 1995 06:21:37 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id IAA39718 for ; Mon, 23 Oct 1995 08:23:49 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA26902; Mon, 23 Oct 1995 08:23:48 -0500 Message-Id: <9510231323.AA26902@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: What am I doing wrong...got to be something stupid Date: Mon, 23 Oct 95 08:23:48 -0500 Below is my routine to read a blob from DB2, I seem to be getting the correct values back in bufsv along with the correct length getting set in the SV struct. int dbd_st_readblob(sth, field, offset, len, destrv, destoffset) SV *sth; int field; long offset; long len; SV *destrv; long destoffset; { D_imp_sth(sth); D_imp_dbh_from_sth; int retl=0; SV *bufsv; int rtval=0; char *msg; bufsv = SvRV(destrv); sv_setpvn(bufsv,"",0); /* ensure it's writable string */ SvGROW(bufsv, len+destoffset+1); /* SvGROW doesn't do +1 */ rtval = SQLGetData(imp_sth->phstmt,field,SQL_C_BINARY, ((void *)SvPVX(bufsv)),len+destoffset,&retl); msg = (ERRTYPE(rtval) ? "GetData failed to read blob":"Invalid Handle"); rtval = check_error(sth,rtval,henv,imp_dbh->hdbc,imp_sth->phstmt,msg); EOI(rtval); SvCUR_set(bufsv, len ); *SvEND(bufsv) = '\0'; /* consistent with perl sv_setpvn etc */ return 1; } However, where I get back to perl the variable contains nothing but SCALAR(0x200c8690)SCALAR(0x200c86b4)SCALAR(0x200c869c)SCALAR(0x200c86cc) SCALAR(0x20022254)SCALAR(0x20022260)SCALAR(0x2002226c)SCALAR(0x20022278) ... SCALAR(0x20022284)SCALAR(0x20022290)SC So obviously, I am either not providing the correct value for destrv or am not returning it correctly...it has to be something truly stupid on my part. Any one spot it? Mike mhm@austin.ibm.com MODULE = DBD::DB2 PACKAGE = DBD::DB2::st void readblob(sth, field, offset, len, destrv=Nullsv, destoffset=0) SV * sth int field long offset long len SV * destrv long destoffset CODE: if (!destrv) destrv = sv_2mortal(newRV(newSV(0))); if (dbd_st_readblob(sth, field, offset, len, destrv, destoffset)) ST(0) = destrv; else ST(0) = &sv_undef; ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <23709-0@ignite.demon.co.uk>; Tue, 24 Oct 1995 03:39:13 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 01:17:31 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa11527; 24 Oct 95 1:16 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA03191 for perldb-interest-real; Mon, 23 Oct 1995 15:30:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA03187 for ; Mon, 23 Oct 1995 15:30:04 -0700 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id SAA22721 for ; Mon, 23 Oct 1995 18:32:20 -0400 Message-Id: <199510232232.SAA22721@pilot.njin.net> To: perldb-interest@fugue.com Subject: CGI-Perl-DBI-Oracle problem Date: Mon, 23 Oct 1995 18:32:18 -0400 From: "Patricia H. Ju" Hi all, (Read subject :) I have a perl script that runs ok from the command line (I get a lot of: ERROR EVENT: DBI::db=HASH(0x158630)->DBI::default_handler: 12223 ORA-12223: TNS:internal limit restriction exceeded (DBD: login failed) at /usr/local/lib/perl5/DBD/Oracle.pm line 99 Can't call method "prepare" without a package or object reference at /usr/local/lib/perl5/Oraperl.pm line 90. errors, but once in a while it doesn't spit up and behaves nicely. However, when running this script via a browser (Netscape), I get: ORACLE_HOME set to /opt/oracle test at /home/patricia/public_html/fibbs/bin/test.cgi line 5. ERROR EVENT: DBI::db=HASH(0x157dd0)->DBI::default_handler: 1034 ORA-01034: ORACLE not available (DBD: login failed) at /usr/local/lib/perl5/DBD/Oracle.pm line 99 Can't call method "prepare" without a package or object reference at /usr/local/lib/perl5/Oraperl.pm line 90. and it NEVER works. Is there some environmental variable I must make the perl script set or something that's keeping the script from ever successfully logging into the Oracle database when using CGI ? (Any DBA's who can help me stop this annoying login failure problem would be much appreciated, too. I'm the only person who's using this database and I should *not* be constantly getting login errors!) Thanks. Patricia ............. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <23709-2@ignite.demon.co.uk>; Tue, 24 Oct 1995 03:39:35 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 01:18:41 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa11730; 24 Oct 95 1:17 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA03221 for perldb-interest-real; Mon, 23 Oct 1995 15:35:41 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA03217 for ; Mon, 23 Oct 1995 15:35:39 -0700 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id SAA22954; Mon, 23 Oct 1995 18:37:53 -0400 Message-Id: <199510232237.SAA22954@pilot.njin.net> To: perldb-interest@fugue.com cc: phj@pilot.njin.net Subject: Re: CGI-Perl-DBI-Oracle problem Date: Mon, 23 Oct 1995 18:37:52 -0400 From: "Patricia H. Ju" Oops. 5 seconds after sending my help plea, I realized the problem. Logged in as me, I have my ORACLE_SID variable set so I don't need to specify my database in the script. But when I run it as a CGI script, there's no default database. Points as to why I'm getting those TNS: internal limit reached errors would still be useful, though. (I imagine it's some kind of lack-of-logging off or something; does Perl or DBD automatically logoff or does one have to explicitly call ora_logoff ?) Thanks (: Patricia ........... "Men have forgotten this truth," said the fox. "But you must not forget it. You become responsible, forever, for what you have tamed . . ." ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <24893-32@ignite.demon.co.uk>; Tue, 24 Oct 1995 07:10:52 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 04:42:42 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id ad17255; 24 Oct 95 4:41 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <29636-0@frigate.doc.ic.ac.uk>; Mon, 23 Oct 1995 10:30:12 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA00461 for perldb-interest-real; Mon, 23 Oct 1995 01:01:17 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA00457 for ; Mon, 23 Oct 1995 01:01:15 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id HAA25339; Mon, 23 Oct 1995 07:58:53 +0100 From: Alligator Descartes Message-Id: <199510230658.HAA25339@fruitbat.mcqueen.com> Subject: Re: DBD::Informix To: Dave Rattanni Date: Mon, 23 Oct 1995 07:58:52 +0100 (BST) Cc: perldb-interest@fugue.com, Tim.Bunce@ig.co.uk In-Reply-To: <9510221500.AA06926@sun38.sikorsky> from "Dave Rattanni" at Oct 22, 95 11:00:03 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 558 > I posted mail to the mailing list, but I forgot > to CC you. Anyway, I ran test.pl as suggested > and it ran with no errors on my 7.1 SE database. > > Well done. Thanks Dave. Your help has been completely invaluable! I'll start adding more functionality into the driver this week and keep the list up to do date with progress. Similarly, progress will be indicated on the pages at: http://www.hermetica.com/technologia/DBI Regards. > Dave -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <24893-46@ignite.demon.co.uk>; Tue, 24 Oct 1995 07:12:38 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 04:44:08 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id am17255; 24 Oct 95 4:43 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <08709-0@frigate.doc.ic.ac.uk>; Mon, 23 Oct 1995 17:03:48 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA02116 for perldb-interest-real; Mon, 23 Oct 1995 05:04:29 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA02112 for ; Mon, 23 Oct 1995 05:04:27 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA21247 for ; Mon, 23 Oct 1995 07:06:44 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA24130; Mon, 23 Oct 1995 07:06:43 -0500 Message-Id: <9510231206.AA24130@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: Re: Overriding malloc/free on AIX 3.2.5? In-Reply-To: Your message of Fri, 20 Oct 95 14:50:04 +0100. <9510201507.ap05588@relay-3.mail.demon.net> Date: Mon, 23 Oct 95 07:06:43 -0500 Tim Bunce writes: |>> From: mhm@austin.ibm.com |>> |>> |> > /* allocate room for copy of statement with spare capacity */ |>> |> > /* for editing ':1' into ':p1' so we can use obndrv. */ |>> |> > ! imp_sth->statement = (char*)safemalloc(strlen(statement) + |>> |> > ! (DBIc_NUM_PARAMS(imp_sth)*4)); |>> |> |>> |> Er, I don't think so. |>> |> |>> |> How is DBIc_NUM_PARAMS set before you've parsed the string ? |>> |>> At the point in time where DBD::Oracle calls oopen (dbdimp.c: |>> db_st_prepare line 300 in version 0.25) DBD::DB2 has already |>> parsed the statement and the number of parameters is available |>> from the engine. |> |>So what is dbd_preparse() doing? Why not get the info from the DB2 API? Setting up the imp_sth->bind_names hash. Although, with this release of DB2 it is a little bit of overkill. My basic tenant has been to get as much of DBD::DB2 running with as little mutation to the reference implementation as possible. Conceivably, I could do away with the entire imp_sth->bind_names hash as everything is '?' in DB2, but then I would have significantly more changes to track during this alpha stage of DBI/DBD. I'll probable continue along the same path until the references (DBI/DBD::Oracle) go GA, then I'll look at things like above where I can make a significant performance enhancement by droping lots of redundant code. |>I presume DB2 use '?' as the placeholder. Currently, '?' is the only acceptable placeholder, but I expect that to change if what I am hearing about the X/Open CLI standard is true (haven't gotten my copy yet :-( ) Mike mhm@austin.ibm.com ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <24893-56@ignite.demon.co.uk>; Tue, 24 Oct 1995 07:13:57 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 05:50:53 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa03832; 24 Oct 95 5:49 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA04400 for perldb-interest-real; Mon, 23 Oct 1995 21:42:45 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id VAA04396 for ; Mon, 23 Oct 1995 21:42:32 -0700 Received: by relay-1.mail.demon.net id ao16465; 23 Oct 95 19:36 GMT Received: from relay-4.mail.demon.net by relay-1.mail.demon.net id ac12830; 23 Oct 95 18:29 GMT Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ad28410; 23 Oct 95 13:47 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id al28074; 23 Oct 95 13:46 GMT Received: from ig.co.uk by ignite.demon.co.uk id <14433-0@ignite.demon.co.uk>; Mon, 23 Oct 1995 14:23:09 +0100 To: cayouett@bnr.ca Subject: Re: status of sybase perl? Cc: perldb-interest@fugue.com Date: Mon, 23 Oct 1995 14:23:09 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510231346.al28074@relay-3.mail.demon.net> > From: "daniel (d.g.) cayouette" > > I guess this means that no-one is currently working on a sybase module. > > Daniel > > In message "Re: status of sybase perl?", you write: > > >> From: "daniel (d.g.) cayouette" > >> > >> What is the status of perl for oracle, ingres and sybase? > > > >Check out > > > > http://www.hermetica.com/technologia/DBI/DBD/index.html There has been talk but no action so far (I think). Sybase is a special case since it has a well established Perl5 Module already: Sybase:: ::DBlib bdcO Sybase DBlibrary interface MEWP ::Sybperl bdcf sybperl 1.0xx compatibility module MEWP ::CTlib cdcO Sybase CTlibrary intgerface MEWP MEWP Michael Peppler It seems sensible to write a DBD::Sybase that is actually an interface layer to Sybase::DBlib. Tim. ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <26572-0@ignite.demon.co.uk>; Tue, 24 Oct 1995 08:12:55 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 07:12:23 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa28262; 24 Oct 95 7:11 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA04643 for perldb-interest-real; Mon, 23 Oct 1995 23:11:53 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id XAA04639 for ; Mon, 23 Oct 1995 23:11:48 -0700 Received: from post.demon.co.uk by relay-1.mail.demon.net id ce00662; 23 Oct 95 23:47 GMT Received: from relay-1.mail.demon.net by relay-3.mail.demon.net id aa28733; 23 Oct 95 20:44 GMT Received: from ig.co.uk by ignite.demon.co.uk id <19962-0@ignite.demon.co.uk>; Mon, 23 Oct 1995 21:44:08 +0100 To: perldb-interest@fugue.com, mhm@austin.ibm.com Subject: Re: What am I doing wrong...got to be something stupid Date: Mon, 23 Oct 1995 21:44:08 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510232044.aa28733@relay-3.mail.demon.net> > From: mhm@austin.ibm.com > > Below is my routine to read a blob from DB2, I seem to be getting > the correct values back in bufsv along with the correct length getting > set in the SV struct. Could you post the portion of the application that does the select, fetch and print? Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <27113-5@ignite.demon.co.uk>; Tue, 24 Oct 1995 10:36:06 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 08:39:38 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id ab19497; 24 Oct 95 8:38 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <14412-0@frigate.doc.ic.ac.uk>; Mon, 23 Oct 1995 21:32:54 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA02694 for perldb-interest-real; Mon, 23 Oct 1995 09:43:38 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA02690 for ; Mon, 23 Oct 1995 09:43:34 -0700 Received: by relay-4.mail.demon.net id g.aa06841; 23 Oct 95 15:46 GMT Received: from post.demon.co.uk by relay-4.mail.demon.net id g.ag02324; 23 Oct 95 14:43 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa08959; 23 Oct 95 14:42 GMT Received: from ig.co.uk by ignite.demon.co.uk id <15211-0@ignite.demon.co.uk>; Mon, 23 Oct 1995 15:42:06 +0100 To: perldb-interest@fugue.com, mhm@austin.ibm.com Subject: Re: Anyone have successfully using blobs in DBD::Oracle? Date: Mon, 23 Oct 1995 15:42:06 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510231442.aa08959@relay-3.mail.demon.net> > From: mhm@austin.ibm.com > > Just curious, I'm having problems getting the data from readblob back > into my perl script variable under DB2. Thus, the obvious question has > anyone tried it (readblob under Oracle)? Yes. I think at least Alligator Descartes uses it. Tim. ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <27113-28@ignite.demon.co.uk>; Tue, 24 Oct 1995 10:37:36 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 08:42:19 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id an19497; 24 Oct 95 8:41 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <20465-0@frigate.doc.ic.ac.uk>; Tue, 24 Oct 1995 01:37:17 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA03416 for perldb-interest-real; Mon, 23 Oct 1995 16:25:35 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from kjahds.com (root@ringotty10.jvnc.net [192.67.236.210]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA03412 for ; Mon, 23 Oct 1995 16:25:29 -0700 Received: by kjahds.com (Linux Smail3.1.28.1 #1) id m0t7WDz-0000XMC; Mon, 23 Oct 95 19:24 EDT Date: Mon, 23 Oct 1995 19:24:11 -0400 (EDT) From: Kenneth Albanowski To: Tim Bunce cc: lwall@scalpel.netlabs.com, perldb-interest@fugue.com, perl5-porters@africa.nicoh.com Subject: Re: perl bug: array of undefs is falsely false In-Reply-To: <9510232242.aa23945@relay-3.mail.demon.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 23 Oct 1995, Tim Bunce wrote: > Was anything finalised on this issue? > > What behaviour could we reasonably expect in 5.002? (and/or later?) As a bizzare example, I just happened to write this: @a=(1,2,3,4,0,2,2,0,0,0); while(my($a,$b)=splice(@a,0,2)) {print "$a|$b\n"} Unfortunately, it doesn't work. This, however, does: @a=(1,2,3,4,0,2,2,0,0,0); my($a,$b); while(($a,$b)=splice(@a,0,2)) {print "$a|$b\n"}' I'd say reasonable behaviour is the latter working properly. Exemplarly behaviour would be the first working properly. > Tim. -- Kenneth Albanowski (kjahds@kjahds.com, CIS: 70705,126) ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <04905-2@ignite.demon.co.uk>; Tue, 24 Oct 1995 19:26:12 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 18:19:50 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id aa03734; 24 Oct 95 18:18 GMT Received: from [204.254.239.2] by relay-1.mail.demon.net id aa27195; 24 Oct 95 18:17 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA06773 for perldb-interest-real; Tue, 24 Oct 1995 07:51:41 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from hub.ucsb.edu (hub.ucsb.edu [128.111.24.40]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA06769 for ; Tue, 24 Oct 1995 07:51:37 -0700 Received: from dokoka (dokoka.ucsb.edu) by hub.ucsb.edu; id AA08498 sendmail 4.1/UCSB-2.1-sun Tue, 24 Oct 95 07:53:46 PDT for perldb-interest@fugue.com Received: from dokoka.ucsb.edu by dokoka via ESMTP (940816.SGI.8.6.9/940406.SGIv2) id HAA03758; Tue, 24 Oct 1995 07:53:38 -0700 Message-Id: <199510241453.HAA03758@dokoka> To: Tim Bunce Cc: cayouett@bnr.ca, perldb-interest@fugue.com Subject: Re: status of sybase perl? Reply-To: "Alan K. Stebbens" In-Reply-To: Your message of Mon, 23 Oct 1995 14:23:09 PDT. <9510231346.al28074@relay-3.mail.demon.net> Date: Tue, 24 Oct 1995 07:53:36 -0700 From: Alan Stebbens > Sybase is a special case since it has a well established Perl5 Module > already: > > Sybase:: > ::DBlib bdcO Sybase DBlibrary interface MEWP > ::Sybperl bdcf sybperl 1.0xx compatibility module MEWP > ::CTlib cdcO Sybase CTlibrary intgerface MEWP > > MEWP Michael Peppler > > It seems sensible to write a DBD::Sybase that is actually an interface > layer to Sybase::DBlib. Well, after looking over Sybase::DBlib and Sybase::CTlib, it is less obvious which should be considered the primary interface if you wish to use cursors. If cursors are not relevant to your application, then use DBlib. It is true that Sybase::DBlib does handle the majority of the database functions, but Sybase::CTlib seems to handle cursors, while Sybase::DBlib currently does not, although the C version of DBlib (from Sybase) does have cursor support. I would be glad to discover that I am confused and/or mistaken. Otherwise, perhaps the Sybase::DBlib needs an update to System 10.x? Alan ---+++--- Received: from punt2.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <08354-12@ignite.demon.co.uk>; Tue, 24 Oct 1995 22:58:44 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 24 Oct 95 21:27:17 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa09811; 24 Oct 95 21:24 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA07606 for perldb-interest-real; Tue, 24 Oct 1995 11:28:35 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-2.mail.demon.net (disperse.demon.co.uk [158.152.1.77]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA07602 for ; Tue, 24 Oct 1995 11:28:30 -0700 Received: from post.demon.co.uk by relay-2.mail.demon.net id ab04267; 24 Oct 95 18:30 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa12690; 24 Oct 95 18:29 GMT Received: from ig.co.uk by ignite.demon.co.uk id <05092-0@ignite.demon.co.uk>; Tue, 24 Oct 1995 19:28:30 +0100 To: mpeppler@itf.ch Subject: Re: status of sybase perl? Cc: perldb-interest@fugue.com, aks@hub.ucsb.edu Date: Tue, 24 Oct 1995 19:28:30 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510241829.aa12690@relay-3.mail.demon.net> I've forwarded this to Michael Peppler since I know little about Sybase. ----- Begin Included Message ----- From: Alan Stebbens > Sybase is a special case since it has a well established Perl5 Module > already: > > Sybase:: > ::DBlib bdcO Sybase DBlibrary interface MEWP > ::Sybperl bdcf sybperl 1.0xx compatibility module MEWP > ::CTlib cdcO Sybase CTlibrary intgerface MEWP > > MEWP Michael Peppler > > It seems sensible to write a DBD::Sybase that is actually an interface > layer to Sybase::DBlib. Well, after looking over Sybase::DBlib and Sybase::CTlib, it is less obvious which should be considered the primary interface if you wish to use cursors. If cursors are not relevant to your application, then use DBlib. It is true that Sybase::DBlib does handle the majority of the database functions, but Sybase::CTlib seems to handle cursors, while Sybase::DBlib currently does not, although the C version of DBlib (from Sybase) does have cursor support. I would be glad to discover that I am confused and/or mistaken. Otherwise, perhaps the Sybase::DBlib needs an update to System 10.x? Alan ----- End Included Message ----- ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <11740-0@ignite.demon.co.uk>; Wed, 25 Oct 1995 08:06:52 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 06:40:42 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id aa06738; 25 Oct 95 6:40 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <20996-0@frigate.doc.ic.ac.uk>; Tue, 24 Oct 1995 01:55:05 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA03258 for perldb-interest-real; Mon, 23 Oct 1995 15:46:04 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-2.mail.demon.net (disperse.demon.co.uk [158.152.1.77]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA03254 for ; Mon, 23 Oct 1995 15:45:59 -0700 Received: by relay-2.mail.demon.net id aa23344; 23 Oct 95 22:47 GMT Received: from post.demon.co.uk by relay-2.mail.demon.net id ae23002; 23 Oct 95 22:44 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa23945; 23 Oct 95 22:42 GMT Received: from ig.co.uk by ignite.demon.co.uk id <19999-0@ignite.demon.co.uk>; Mon, 23 Oct 1995 21:51:21 +0100 To: lwall@scalpel.netlabs.com Subject: Re: perl bug: array of undefs is falsely false Cc: perldb-interest@fugue.com, perl5-porters@nicoh.com Date: Mon, 23 Oct 1995 21:51:21 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510232242.aa23945@relay-3.mail.demon.net> Was anything finalised on this issue? What behaviour could we reasonably expect in 5.002? (and/or later?) Tim. > From: Larry Wall > > : > % Excerpts from the mail message of Mark Jefferys: > : > : > % ) 2. Make list-assignments return undef for zero objects assigned, > : > % ) and SV_NO for undef-only assignments. This allows: > : > % ) > : > % ) if (defined(@arr = func())) ... > : > : > % ) 3. Make list-assignments normal, but have list-slices cut off > : > % ) trailing undefs, so > : > % ) (getpwent)[0,2] > : > % ) returns an empty list. > : > % ) A. This will cause code which relies on list-slices > : > % ) returning a fixed number of items to break. Sub calls > : > % ) are a likely source of problems: > : > % ) func1((func2())[1,3], ++$cnt); > : > % ) # Expects $cnt in 3rd place, not last. > : > : > % ) 7. Make a list-slice of an empty list be empty. This is an example > : > % ) of minimal intervention: make sure empty lists stay empty, and > : > % ) thus false, but otherwise acts normally. This would be somewhat > : > % ) easier to implement for tied arrays (and hashes). > : > % > : > % This raises the problem of (A) again. > : > > : > Granted, although far less often than the others. I like this one > : > mostly because the exceptions I can think of are more contrived than > : > for the others, and pretty easy to work around: > : > > : > @arr ? @arr[@idx] : (undef) x @idx > : > > : > % ) I kinda like #2 or #7 at the moment; [...] > > I think that as we move into using prototypes, func1() above would > probably be declared as > > sub func1 ($$$); > > so you wouldn't be able to say > > func1((func2())[1,3], ++$cnt); > > in any event. (Unless I did something Really Strange with argument > processing to make it look inside a list slice.) > > : So, to summarise the situation (and check that _I_ understand it :) ... > : > : - Currently evaluation in scalar context of the assignment of a list > : of undefs is regarded as false. > : > : - It seems widely agreed that this is not ideal. > : > : - This was added to cater for: while (($user, $uid) = (getpwent)[0,2]) { } > : > : - It seems widely agreed that this is useful. > : > : I also like #7 "Make a list-slice of an empty list be empty". > : Feels kind'a like null-propagation in SQL expressions. > > A little, though I think Perl's notion is a little more fail-soft. > > : #2 seems too subtle for people to remember to use in their code. > : Especially where they don't expect to get nulls but might in error > : conditions. > : > : Does anyone have and hard objections or better alternatives? > : > : Larry, do you concur? > > Hail the concurring hero, eh? > > [sorry] > > I'm leaning toward #7 too. Especially since it makes array assignments > a little faster... > > On the other hand, doing something a bit #8ish wouldn't be too difficult. > It would be relatively trivial to inform the OP_LSLICE that it was on the > right side of an OP_AASSIGN. I don't think we'd even have to worry about > whether the assignment was in a scalar context, though even that wouldn't > be difficult to do. > > Larry > ---+++--- Received: from punt.demon.co.uk by ignite.demon.co.uk with SMTP (PP) id <13500-2@ignite.demon.co.uk>; Wed, 25 Oct 1995 12:44:29 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 11:10:29 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id aa09282; 25 Oct 95 11:09 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA08765 for perldb-interest-real; Wed, 25 Oct 1995 00:50:31 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from chsun.eunet.ch (chsun.eunet.ch [146.228.10.15]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id AAA08761 for ; Wed, 25 Oct 1995 00:50:24 -0700 Received: from itf1.UUCP by chsun.eunet.ch (8.6.10/1.34) id IAA13536; Wed, 25 Oct 1995 08:52:36 +0100 Received: from kiruna.itf.ch by itf.ch (4.1/SMI-4.1) id AA00443; Wed, 25 Oct 95 08:27:14 +0100 Date: Wed, 25 Oct 95 08:27:14 +0100 From: Michael Peppler Message-Id: <9510250727.AA00443@itf.ch> To: perldb-interest@fugue.com Subject: Re: status of sybase perl? Cc: aks@hub.ucsb.edu [ Note: I was unfortunately removed from the perldb-interest list when the address changed, so I haven't followed this discussion from the begining...] From: Alan Stebbens > > > Sybase is a special case since it has a well established Perl5 Module > > already: > > > > Sybase:: > > ::DBlib bdcO Sybase DBlibrary interface MEWP > > ::Sybperl bdcf sybperl 1.0xx compatibility module MEWP > > ::CTlib cdcO Sybase CTlibrary intgerface MEWP > > > > MEWP Michael Peppler A quick recap of the situation of the Sybase modules: ::DBlib and ::Sybperl will be taken out of beta (full 2.0 release) in a couple of days. The modules are stable, and I have refrained from adding new functionality to them. ::CTlib is quite advanced, and includes many features not found in the other modules, such as native handling of special datatypes such as DATETIME, MONEY and NUMERIC/DECIMAL, using operator overloading where possible. ::CTlib is where I will be concentrating my developpment efforts (for reasons that will probably become clear in a couple of weeks). The API is much cleaner than DBlib, and this is where Sybase will be enhancing their functionality as well. > > > > It seems sensible to write a DBD::Sybase that is actually an interface > > layer to Sybase::DBlib. > > Well, after looking over Sybase::DBlib and Sybase::CTlib, it is less > obvious which should be considered the primary interface if you wish to > use cursors. If cursors are not relevant to your application, then use > DBlib. I would write DBD::Sybase on top of Sybase::CTlib. It's bound to be quite a bit simpler. > It is true that Sybase::DBlib does handle the majority of the database > functions, but Sybase::CTlib seems to handle cursors, while > Sybase::DBlib currently does not, although the C version of DBlib (from > Sybase) does have cursor support. > > I would be glad to discover that I am confused and/or mistaken. > Otherwise, perhaps the Sybase::DBlib needs an update to System 10.x? Sybase::DBlib already includes 90+ API calls (out of a total of over 300). CTlib already offers better functionality (with the exception of the BCP API) using only 16 API calls... Michael ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <08754-1@oink>; Wed, 25 Oct 1995 17:55:11 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 16:44:53 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa14067; 25 Oct 95 16:44 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA10523 for perldb-interest-real; Wed, 25 Oct 1995 06:20:36 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA10519 for ; Wed, 25 Oct 1995 06:20:34 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id IAA69819 for ; Wed, 25 Oct 1995 08:22:53 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA29726; Wed, 25 Oct 1995 08:22:52 -0500 Message-Id: <9510251322.AA29726@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: Re: What am I doing wrong...got to be something stupid Date: Wed, 25 Oct 95 08:22:51 -0500 Sorry about the delay. I lost permissions to one directory in my mail hierarchy and as a result have not seen anything coming from perldb-interest for a few days. Tim Bunce writes: |> > From: mhm@austin.ibm.com |> > |> > Below is my routine to read a blob from DB2, I seem to be getting |> > the correct values back in bufsv along with the correct length getting |> > set in the SV struct. |> |> Could you post the portion of the application that does the select, |> fetch and print? |> Here is a small test case. Mike mhm@austin.ibm.com #!/usr/local/bin/perl use DBI; use DB2::Constants; use DBD::DB2 qw($attrib_char ); $DBI::dbi_debug=2; $dbh = DBI->connect('sample'); $stmt = "select empno, photo_format from emp_photo where photo_format ='gif';"; $sth = $dbh->prepare($stmt); $sth->execute(); while ((@row = $sth->fetchrow())) { open(OUTPUT,">$row[0].$row[1]") || die "Can't open $row[0].$row[1]"; $stmt = "select empno,picture from emp_photo where empno = ?;"; $pict_sth = $dbh->prepare($stmt); $pict_sth->bind_param(1,$row[0],$attrib_char); $pict_sth->execute(); $pict_sth->fetch(); while ($buff = $pict_sth->readblob(1,0,4096)) { print OUTPUT $buff; $buff = ""; } close(OUTPUT); } ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <10121-0@oink>; Wed, 25 Oct 1995 18:59:38 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 17:57:58 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa20481; 25 Oct 95 17:57 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA10598 for perldb-interest-real; Wed, 25 Oct 1995 07:16:52 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from hatch.sonalysts.com (hatch.sonalysts.com [198.6.208.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA10594 for ; Wed, 25 Oct 1995 07:16:50 -0700 Received: by hatch.sonalysts.com (1.37.109.16/16.2) id AA157200636; Wed, 25 Oct 1995 10:17:16 -0400 Date: Wed, 25 Oct 1995 10:17:16 -0400 (EDT) From: Matt Stahl To: jfox@slb.com Cc: ntperl@mail.hip.com, perldb-interest@fugue.com Subject: Re: [?] NTPerl5 Configure In-Reply-To: <9510242113.AA09707@jade> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 24 Oct 1995 fox@austin.asc.slb.com wrote: > > > Date: Tue, 24 Oct 1995 16:35:02 -0400 (EDT) > > From: Matt Stahl > > Precedence: bulk > > > > I recently downloaded the latest NTPerl5 for i386 from > > ntperl.hip.com (ntperl.5.091.i86.zip). Looking at the configure.pm module, > > I noticed all the harware specific info that was pertinent to the machine > > the src was compiled on, which of course, doesn't match my machine. > > I also noticed that there is no *configure* script (like there is > > in the 5.001m UNIX version). > > I'd like to be able to make some extensions (such as > > DBI/DBD::Oracle), but they use ExtUtils::MakeMaker (also not a part of the > > NTperl5 port, but will *hopefully* work as is), which uses Configure.pm. > > If I make perl from the src, will Configure.pm get updated? This > > is an option only if I get my hands on the Microsoft SDK (since the Perl > > Makefile needs some SDK libs...) > > Has anyone already tackled these problems? > > > > Thanks, > > Matt > > Hi, Matt= > > I had no luck with a quick hack attack at geting DBI/DBD::Oracle to > work on NT (because of the missing MakeMaker and, worse, lack of > support for dynamic loading. I didn't have much luck with my quick > attempt to make DBI/DBD static extensions. As this is not the primary > focus of my work, I didn't want to spend too much time on it. It would be nice if NTperl included MakeMaker, Configure, et al. It would certainly make all our jobs a little easier. (Hint, hint to hip.com) > > I can, however, build perl from the source on NT (Visual-C++ does it > pretty much by itself). If you have some experiments that fail for > lack of SDK libs, perhaps I can try them for you "in my spare time." > > =Jim Fox Thanks Jim! I'll continue my attempts (also in my *spare time*), and will keep your offer in mind. --Matt Stahl ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <11820-0@oink>; Wed, 25 Oct 1995 20:52:31 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 19:44:58 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa17377; 25 Oct 95 19:44 GMT Received: from [204.254.239.2] by relay-1.mail.demon.net id aa23044; 25 Oct 95 19:43 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA11109 for perldb-interest-real; Wed, 25 Oct 1995 10:45:30 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA11105 for ; Wed, 25 Oct 1995 10:45:28 -0700 Received: by gw.home.vix.com id AA09739; Wed, 25 Oct 95 10:47:46 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id MAA26022; Wed, 25 Oct 1995 12:47:44 -0500 Date: Wed, 25 Oct 1995 12:47:44 -0500 (CDT) From: Ben Lindstrom To: DBI perl Subject: I think I have fetch almost working.. Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I just need to verify my test perl code is correct. ...Assume datbase connected. $cursor = $dbh->prepare( "SELECT * FROM table1" ); $cursor->execute; while ( ($name,$age) = $cursor->fetchrow ) { print "Field: $name --> $age\n"; } $cursor->finish; I only get name returned currently. I need to know if I've done the Perl code right. If so I know it's in my DBD. Thanks. --- Closer we come to success..The more problems that occur. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <12007-0@oink>; Wed, 25 Oct 1995 21:13:31 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 20:10:57 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa01330; 25 Oct 95 20:10 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id aa24099; 25 Oct 95 20:09 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA11203 for perldb-interest-real; Wed, 25 Oct 1995 11:19:04 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-2.mail.demon.net (disperse.demon.co.uk [158.152.1.77]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA11199 for ; Wed, 25 Oct 1995 11:18:57 -0700 Received: from post.demon.co.uk by relay-2.mail.demon.net id ad27407; 25 Oct 95 18:13 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa26359; 25 Oct 95 18:13 GMT Received: from toad by oink with SMTP (PP) id <10247-0@oink>; Wed, 25 Oct 1995 19:11:50 +0100 Received: by toad (5.0/SMI-SVR4) id AA19547; Wed, 25 Oct 1995 19:11:27 +0000 Date: Wed, 25 Oct 1995 19:11:27 +0000 From: Tim Bunce Message-Id: <9510251811.AA19547@toad> To: jfox@slb.com, mgstahl@sonalysts.com Subject: Re: [?] NTPerl5 Configure Cc: ntperl@mail.hip.com, perldb-interest@fugue.com, makemaker@franz.ww.tu-berlin.de X-Sun-Charset: US-ASCII content-length: 974 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Matt Stahl > > > I had no luck with a quick hack attack at geting DBI/DBD::Oracle to > > work on NT (because of the missing MakeMaker and, worse, lack of > > support for dynamic loading. I didn't have much luck with my quick > > attempt to make DBI/DBD static extensions. As this is not the primary > > focus of my work, I didn't want to spend too much time on it. > > It would be nice if NTperl included MakeMaker, Configure, et al. It would > certainly make all our jobs a little easier. (Hint, hint to hip.com) MakeMaker is *vital* to any useful Perl port. Period. A new much-enhanced MakeMaker (5.00) is about to me released which includes improved support for OS/2. That should be a useful stepping stone for NT porters. Anyone interested in working on or with MakeMaker should subscribe to the mailing list via makemaker-request@franz.ww.tu-berlin.de. I urge the NT porters to join. You'll find us very willing to help. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15516-25@oink>; Thu, 26 Oct 1995 07:09:13 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 23:27:52 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id aa14711; 25 Oct 95 23:26 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA11993 for perldb-interest-real; Wed, 25 Oct 1995 15:10:08 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-2.mail.demon.net (disperse.demon.co.uk [158.152.1.77]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA11986 for ; Wed, 25 Oct 1995 15:09:59 -0700 Received: from post.demon.co.uk by relay-2.mail.demon.net id ac12058; 25 Oct 95 22:06 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa22533; 25 Oct 95 22:06 GMT Received: from toad by oink with SMTP (PP) id <14049-0@oink>; Wed, 25 Oct 1995 23:05:40 +0100 Received: by toad (5.0/SMI-SVR4) id AA19753; Wed, 25 Oct 1995 23:05:15 +0000 Date: Wed, 25 Oct 1995 23:05:15 +0000 From: Tim Bunce Message-Id: <9510252205.AA19753@toad> To: perldb-interest@fugue.com, mhm@austin.ibm.com Subject: Re: What am I doing wrong...got to be something stupid Cc: descarte@hermetica.com X-Sun-Charset: US-ASCII content-length: 1208 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: mhm@austin.ibm.com > > Tim Bunce writes: > |> > From: mhm@austin.ibm.com > |> > > |> > Below is my routine to read a blob from DB2, I seem to be getting > |> > the correct values back in bufsv along with the correct length getting > |> > set in the SV struct. > |> > |> Could you post the portion of the application that does the select, > |> fetch and print? > |> > > Here is a small test case. > > while ($buff = $pict_sth->readblob(1,0,4096)) { > print OUTPUT $buff; > $buff = ""; > } Ah, so your getting SCALAR(0x200c8690)SCALAR(0x200c86b4)SCALAR(0x200c869c)SCALAR(0x200c86cc) was actually one 'SCALAR(0x200c86b4)' per fetch. That makes more sense. It means that you're returning a reference to a scalar. That points me to this code in the xs file: if (!destrv) destrv = sv_2mortal(newRV(newSV(0))); if (dbd_st_readblob(sth, field, offset, len, destrv, destoffset)) ST(0) = destrv; <=== else ST(0) = &sv_undef; and I now see what's wrong. Make this change and let me know how it goes: - ST(0) = destrv; + ST(0) = SvRV(destrv); Alligator, why didn't this affect you? Did you use the readblob(..., \$dest) mechanism instead? Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <15515-6@oink>; Thu, 26 Oct 1995 07:06:28 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 23:21:41 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa09388; 25 Oct 95 23:20 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA11868 for perldb-interest-real; Wed, 25 Oct 1995 14:46:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA11864 for ; Wed, 25 Oct 1995 14:46:05 -0700 Received: by gw.home.vix.com id AA28747; Wed, 25 Oct 95 14:48:24 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from sunspot.filoli.com (root@sunspot.filoli.com [204.162.1.17]) by filoli.filoli.com (8.6.10/8.6.9) with ESMTP id OAA28225; Wed, 25 Oct 1995 14:47:52 -0700 Received: from orac.filoli.com (orac.filoli.com [204.162.1.137]) by sunspot.filoli.com (8.6.12/8.6.9) with SMTP id OAA12212; Wed, 25 Oct 1995 14:47:47 -0700 Message-Id: <199510252147.OAA12212@sunspot.filoli.com> Received: by orac.filoli.com (NX5.67e/NX3.0X) id AA04384; Wed, 25 Oct 95 14:47:45 -0700 Content-Type: text/plain Mime-Version: 1.0 (NeXT Mail 3.3 v118.2) Original-Received: by NeXT.Mailer (1.118.2) PP-warning: Illegal Received field on preceding line From: George Johnston Date: Wed, 25 Oct 95 14:47:43 -0700 To: Ben Lindstrom Subject: I think I have fetch almost working.. Cc: DBI perl , andy_lyman@filoli.com References: >Sender: perldb-interest-request@fugue.com > X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us > Date: Wed, 25 Oct 1995 12:47:44 -0500 (CDT) > From: Ben Lindstrom > To: DBI perl > Subject: I think I have fetch almost working.. > > > I just need to verify my test perl code is correct. > >...Assume datbase connected. > $cursor = $dbh->prepare( "SELECT * FROM table1" ); > > $cursor->execute; > > while ( ($name,$age) = $cursor->fetchrow ) { > print "Field: $name --> $age\n"; > } > > $cursor->finish; > > I only get name returned currently. > > I need to know if I've done the Perl code right. If so I know it's > in myDBD. > > Thanks. > >--- > Closer we come to success..The more problems that occur. > I had to go in and make fetch work on my own, we are operating on a tight deadline. The last bug that kept me up was in your test script qb.pl. while (@field = $cursor3->fetchrow) { print "Field: $field\n"; } I just changed it to while (@field = $cursor3->fetchrow) { print "Field: @field\n"; } and I got fetch to work all the way back to the calling perl program. I would suggest that you look to see that you are really returning an array from dbd_st_fetch in your dbdimp.c. George ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <15515-8@oink>; Thu, 26 Oct 1995 07:06:39 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 23:47:45 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa22074; 25 Oct 95 23:46 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA12092 for perldb-interest-real; Wed, 25 Oct 1995 15:33:18 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA12088 for ; Wed, 25 Oct 1995 15:33:15 -0700 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA24240; Wed, 25 Oct 1995 22:30:54 +0100 From: Alligator Descartes Message-Id: <199510252130.WAA24240@fruitbat.mcqueen.com> Subject: Re: What am I doing wrong...got to be something stupid To: Tim Bunce Date: Wed, 25 Oct 1995 22:30:53 +0100 (BST) Cc: perldb-interest@fugue.com, mhm@austin.ibm.com In-Reply-To: <9510252205.AA19753@toad> from "Tim Bunce" at Oct 25, 95 11:05:15 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 458 > Alligator, why didn't this affect you? Did you use the readblob(..., \$dest) > mechanism instead? The mSQL API doesn't support BLOBs currently. ( So I dropped the code! ) And I've been completely tied up with real work, so I haven't reached this issue with the Informix driver yet! So, I suppose that's why it didn't hit me....... 8-) > Tim. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15516-20@oink>; Thu, 26 Oct 1995 07:08:26 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 25 Oct 95 23:21:47 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id aa11582; 25 Oct 95 23:20 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA11972 for perldb-interest-real; Wed, 25 Oct 1995 15:07:13 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA11967 for ; Wed, 25 Oct 1995 15:07:11 -0700 Received: by gw.home.vix.com id AA00492; Wed, 25 Oct 95 15:09:29 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id RAA11388; Wed, 25 Oct 1995 17:09:14 -0500 Date: Wed, 25 Oct 1995 17:09:14 -0500 (CDT) From: Ben Lindstrom To: George Johnston Cc: DBI perl , andy_lyman@filoli.com Subject: Re: I think I have fetch almost working.. In-Reply-To: <199510252147.OAA12212@sunspot.filoli.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII > I had to go in and make fetch work on my own, we are operating on a > tight deadline. The last bug that kept me up was in your test > script qb.pl. > > while (@field = $cursor3->fetchrow) { > print "Field: $field\n"; > } > Hmm...I guess I don't rememver $field...Oh well.. > I just changed it to > while (@field = $cursor3->fetchrow) { > print "Field: @field\n"; > } > > and I got fetch to work all the way back to the calling perl program. > > I would suggest that you look to see that you are really returning > an array from dbd_st_fetch in your dbdimp.c. > Yes..I just changed my test suite to what you have and I do have a problem with passing things back. The first column is being passed back twice. I have the new QuickBase server for rollbacks and commits. You can send the transcation commands via prepre() or via makeCommand:. Have you started working on multiply cursors active? I don't think it's good for both of us to be working on the same functions. It might go quicker to work together. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15516-84@oink>; Thu, 26 Oct 1995 07:15:33 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 26 Oct 95 05:06:46 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id aa16866; 26 Oct 95 5:05 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA12962 for perldb-interest-real; Wed, 25 Oct 1995 20:44:04 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA12958 for ; Wed, 25 Oct 1995 20:43:59 -0700 Received: by gw.home.vix.com id AA23114; Wed, 25 Oct 95 20:46:12 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-2.mail.demon.net id ad17736; 25 Oct 95 19:56 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa24135; 25 Oct 95 19:55 GMT Received: from toad by oink with SMTP (PP) id <11929-0@oink>; Wed, 25 Oct 1995 20:55:33 +0100 Received: by toad (5.0/SMI-SVR4) id AA19658; Wed, 25 Oct 1995 20:55:11 +0000 Date: Wed, 25 Oct 1995 20:55:11 +0000 From: Tim Bunce Message-Id: <9510251955.AA19658@toad> To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: I think I have fetch almost working.. X-Sun-Charset: US-ASCII Content-Length: 226 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Ben Lindstrom > > I just need to verify my test perl code is correct. > > ...Assume datbase connected. > $cursor = $dbh->prepare( "SELECT * FROM table1" ); Add $cursor->debug(2); here. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <20508-62@oink>; Thu, 26 Oct 1995 14:29:54 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 26 Oct 95 08:39:31 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id bg15403; 26 Oct 95 8:39 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <24359-0@frigate.doc.ic.ac.uk>; Tue, 24 Oct 1995 05:26:33 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA04192 for perldb-interest-real; Mon, 23 Oct 1995 20:07:21 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-2.mail.demon.net (disperse.demon.co.uk [158.152.1.77]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA04188 for ; Mon, 23 Oct 1995 20:07:17 -0700 Received: from post.demon.co.uk by relay-2.mail.demon.net id aa25981; 24 Oct 95 2:58 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa24494; 24 Oct 95 2:57 GMT Received: from ig.co.uk by ignite.demon.co.uk id <24021-0@ignite.demon.co.uk>; Tue, 24 Oct 1995 03:54:10 +0100 To: perldb-interest@fugue.com, phj@pilot.njin.net Subject: Re: CGI-Perl-DBI-Oracle problem Date: Tue, 24 Oct 1995 03:54:10 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510240257.aa24494@relay-3.mail.demon.net> > From: "Patricia H. Ju" > > > Oops. 5 seconds after sending my help plea, I realized the problem. > Logged in as me, I have my ORACLE_SID variable set so I don't need > to specify my database in the script. But when I run it as a CGI > script, there's no default database. If I had a penny (or a cent) for every time someone ... > Points as to why I'm getting those TNS: internal limit reached > errors would still be useful, though. Talk to the person who configured TNS. > (I imagine it's some kind of > lack-of-logging off or something; does Perl or DBD automatically > logoff or does one have to explicitly call ora_logoff ?) DBD::Oracle will call ologof() when the $dbh object is destroyed. Perl will do that when it exits. > "Men have forgotten this truth," said the fox. "But you must not forget it. > You become responsible, forever, for what you have tamed . . ." Very true. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <20508-6@oink>; Thu, 26 Oct 1995 14:25:39 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 26 Oct 95 08:36:45 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id ab15403; 26 Oct 95 8:35 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <27225-0@frigate.doc.ic.ac.uk>; Tue, 24 Oct 1995 07:58:17 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id WAA04582 for perldb-interest-real; Mon, 23 Oct 1995 22:57:55 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-1.mail.demon.net (relay-1.mail.demon.net [158.152.1.140]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id WAA04578 for ; Mon, 23 Oct 1995 22:57:43 -0700 Received: from post.demon.co.uk by relay-1.mail.demon.net id ab00508; 23 Oct 95 23:28 GMT Received: from relay-1.mail.demon.net by relay-3.mail.demon.net id aa25586; 23 Oct 95 18:26 GMT Received: from ig.co.uk by ignite.demon.co.uk id <18592-0@ignite.demon.co.uk>; Mon, 23 Oct 1995 19:26:45 +0100 To: perldb-interest@fugue.com, webm@marque.mscs.mu.edu Subject: Re: DBI Installation Date: Mon, 23 Oct 1995 19:26:45 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-ID: <9510231826.aa25586@relay-3.mail.demon.net> > From: BCL HTML account > > We are getting segmentation fault while running make test of DBI module. > perl Makefile.PL and make are working. > Following is the output of make test. It is giving an error on test 5. > > FAILED on test 5 > DBI test application $Revision: 1.12 $ > make: *** [test] Segmentation Fault (core dumped) > > Summary of my perl5 (patchlevel 1) configuration: > Platform: > osname=solaris, osver=2.4, archname=sun4-solaris > uname='sunos eve 5.4 generic_101945-27 sun4m sparc ' > hint=recommended > Compiler: > cc='gcc', optimize='-O' > cppflags='' > ccflags ='' > ldflags ='' > stdchar='unsigned char', d_stdstdio=define, usevfork=false > voidflags=15, castflags=0, d_casti32=define, d_castneg=define > intsize=4, alignbytes=8, usemymalloc=y, randbits=15 > Libraries: > so=so > libpth=/lib /usr/lib /usr/ccs/lib /usr/local/lib > libs=-lsocket -lnsl -ldb -ldl -lm -lc -lcrypt > libc=/usr/lib/libc.so > Dynamic Linking: > dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= > cccdlflags='-fpic', ccdlflags='', lddlflags='-G' Thanks. Please also post... Perl version number (if not 5.001m then upgrade first) Release number of DBI (if not 0.63 then upgrade first) GCC version number (if not >=2.5.8 then upgrade first) Stack trace from core file (rebuild with -g & use gdb or adb) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <20508-133@oink>; Thu, 26 Oct 1995 14:36:10 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 26 Oct 95 08:43:27 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id cs15403; 26 Oct 95 8:42 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <09782-0@frigate.doc.ic.ac.uk>; Tue, 24 Oct 1995 19:35:40 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA07079 for perldb-interest-real; Tue, 24 Oct 1995 09:21:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA07075 for ; Tue, 24 Oct 1995 09:21:51 -0700 Received: by gw.home.vix.com id AA19443; Tue, 24 Oct 95 09:23:55 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-2.mail.demon.net id aa11657; 24 Oct 95 15:39 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id aa06912; 24 Oct 95 15:38 GMT Received: from ig.co.uk by ignite.demon.co.uk id <02386-0@ignite.demon.co.uk>; Tue, 24 Oct 1995 15:13:23 +0100 To: perldb-interest@fugue.com, mhm@austin.ibm.com Subject: Re: Overriding malloc/free on AIX 3.2.5? Date: Tue, 24 Oct 1995 15:13:23 +0100 From: Tim Bunce Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Message-Id: <9510241538.aa06912@relay-3.mail.demon.net> > From: mhm@austin.ibm.com > > Tim Bunce writes: > |>> From: mhm@austin.ibm.com > |>> > |>> |> > /* allocate room for copy of statement with spare capacity */ > |>> |> > /* for editing ':1' into ':p1' so we can use obndrv. */ > |>> |> > ! imp_sth->statement = (char*)safemalloc(strlen(statement) + > |>> |> > ! (DBIc_NUM_PARAMS(imp_sth)*4)); > |>> |> > |>> |> Er, I don't think so. > |>> |> > |>> |> How is DBIc_NUM_PARAMS set before you've parsed the string ? > |>> > |>> At the point in time where DBD::Oracle calls oopen (dbdimp.c: > |>> db_st_prepare line 300 in version 0.25) DBD::DB2 has already > |>> parsed the statement and the number of parameters is available > |>> from the engine. > |> > |>So what is dbd_preparse() doing? Why not get the info from the DB2 API? > > Setting up the imp_sth->bind_names hash. Although, with this release > of DB2 it is a little bit of overkill. My basic tenant has been to > get as much of DBD::DB2 running with as little mutation to the reference > implementation as possible. Conceivably, I could do away with the > entire imp_sth->bind_names hash as everything is '?' in DB2, but then > I would have significantly more changes to track during this alpha > stage of DBI/DBD. A wise policy :-) > |>I presume DB2 use '?' as the placeholder. > > Currently, '?' is the only acceptable placeholder, but I expect that > to change if what I am hearing about the X/Open CLI standard is true > (haven't gotten my copy yet :-( ) X/Open CLI uses '?'. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <25428-14@oink>; Thu, 26 Oct 1995 17:50:04 +0100 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 26 Oct 95 16:02:43 GMT Received: from relay-2.mail.demon.net by punt-2.mail.demon.net id aa09948; 26 Oct 95 16:01 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id aa04020; 26 Oct 95 16:01 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA14877 for perldb-interest-real; Thu, 26 Oct 1995 06:21:54 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA14873 for ; Thu, 26 Oct 1995 06:21:51 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id IAA56331 for ; Thu, 26 Oct 1995 08:24:10 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA23938; Thu, 26 Oct 1995 08:24:09 -0500 Message-Id: <9510261324.AA23938@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: Re: What am I doing wrong...got to be something stupid Date: Thu, 26 Oct 95 08:24:08 -0500 Tim Bunce writes: |> > From: mhm@austin.ibm.com |> > |> > Here is a small test case. |> > |> > while ($buff = $pict_sth->readblob(1,0,4096)) { |> > print OUTPUT $buff; |> > $buff = ""; |> > } |> |> It means that you're returning a reference to a scalar. That points me |> to this code in the xs file: |> |> if (!destrv) |> destrv = sv_2mortal(newRV(newSV(0))); |> if (dbd_st_readblob(sth, field, offset, len, destrv, destoffset)) |> ST(0) = destrv; <=== |> else ST(0) = &sv_undef; |> |> and I now see what's wrong. Make this change and let me know how it goes: |> |> - ST(0) = destrv; |> + ST(0) = SvRV(destrv); |> After making the above change, something is still not right with dbd_st_readblob or my understanding of perlguts is still severly lacking. From dbd_st_readblob, my assumption has been that the line bufsv = SvRV(destrv); causes destrv to be dereferenced, which means that bufsv and destrv would both be pointing at the same memory. However, when looking at both destrv and bufsv in the debugger after the SQLGetData call, I see the appropriate piece of the blob in bufsv->sv_any->xpvav.xav_array and an unchanged destrv->sv_any->xpvav.xav_array. So just what am I missing? Is it not getting passed in correctly? Mike mhm@austin.ibm.com void readblob(sth, field, offset, len, destrv=Nullsv, destoffset=0) SV * sth int field long offset long len SV * destrv long destoffset CODE: if (!destrv) destrv = sv_2mortal(newRV(newSV(0))); if (dbd_st_readblob(sth, field, offset, len, destrv, destoffset)) ST(0) = SvRV(destrv); else ST(0) = &sv_undef; int dbd_st_readblob(sth, field, offset, len, destrv, destoffset) SV *sth; int field; long offset; long len; SV *destrv; long destoffset; { D_imp_sth(sth); D_imp_dbh_from_sth; int retl=0; SV *bufsv; int rtval=0; char *msg; bufsv = SvRV(destrv); sv_setpvn(bufsv,"",0); /* ensure it's writable string */ SvGROW(bufsv, len+destoffset+1); /* SvGROW doesn't do +1 */ rtval = SQLGetData(imp_sth->phstmt,field,SQL_C_BINARY, ((void *)SvPVX(bufsv)),len+destoffset,&retl); msg = (ERRTYPE(rtval) ? "GetData failed to read blob":"Invalid Handle"); rtval = check_error(sth,rtval,henv,imp_dbh->hdbc,imp_sth->phstmt,msg); EOI(rtval); SvCUR_set(bufsv, len ); *SvEND(bufsv) = '\0'; /* consistent with perl sv_setpvn etc */ return 1; } ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <27567-12@oink>; Thu, 26 Oct 1995 19:59:08 +0100 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 26 Oct 95 18:41:18 GMT Received: from relay-2.mail.demon.net by punt-1.mail.demon.net id aa04910; 26 Oct 95 18:40 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id aa10793; 26 Oct 95 18:38 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <10364-0@frigate.doc.ic.ac.uk>; Tue, 24 Oct 1995 20:03:34 +0100 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA06903 for perldb-interest-real; Tue, 24 Oct 1995 08:28:18 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA06899 for ; Tue, 24 Oct 1995 08:28:12 -0700 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.112.95]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id KAA105938; Tue, 24 Oct 1995 10:30:18 -0500 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA22134; Tue, 24 Oct 1995 10:30:05 -0500 Message-Id: <9510241530.AA22134@dax.austin.ibm.com> To: mhm@austin.ibm.com Cc: perldb-interest@fugue.com Subject: Re: What am I doing wrong...got to be something stupid In-Reply-To: Your message of Mon, 23 Oct 95 08:23:48 -0500. <9510231323.AA26902@dax.austin.ibm.com> Date: Tue, 24 Oct 95 10:30:04 -0500 mhm@austin.ibm.com writes: |>int |>dbd_st_readblob(sth, field, offset, len, destrv, destoffset) |> SV *sth; |> int field; |> long offset; |> long len; |> SV *destrv; |> long destoffset; |>{ |> D_imp_sth(sth); |> D_imp_dbh_from_sth; |> int retl=0; |> SV *bufsv; |> int rtval=0; |> char *msg; |> |> bufsv = SvRV(destrv); |> sv_setpvn(bufsv,"",0); /* ensure it's writable string */ |> SvGROW(bufsv, len+destoffset+1); /* SvGROW doesn't do +1 */ |> |> rtval = SQLGetData(imp_sth->phstmt,field,SQL_C_BINARY, |> ((void *)SvPVX(bufsv)),len+destoffset,&retl); |> msg = (ERRTYPE(rtval) ? "GetData failed to readblob":"InvalidHandle"); |> rtval = check_error(sth,rtval,henv,imp_dbh->hdbc,imp_sth->phstmt,msg); |> EOI(rtval); |> |> SvCUR_set(bufsv, len ); |> *SvEND(bufsv) = '\0'; /* consistent with perl sv_setpvn etc */ |> |> return 1; |>} Ok, I need ideas...I'm using perl5.001m + patches, it passes all tests yet the above code doesn't work right. On the first call to imp_sth->readblob, destrv is created by the Xs routine and is passed in to dbd_st_readblob along with zero offset. bufsv is created and get properly filled by SQLGetData, but it is not reflected in destrv. Why? Is bufsv created wrong? Is destsv not created or passed properly? As you can tell I'm in over my head. So, I'm just trying to figure it out and learn something. Mike mhm@austin.ibm.com void readblob(sth, field, offset, len, destrv=Nullsv, destoffset=0) SV * sth int field long offset long len SV * destrv long destoffset CODE: if (!destrv) destrv = sv_2mortal(newRV(newSV(0))); if (dbd_st_readblob(sth, field, offset, len, destrv, destoffset)) ST(0) = destrv; else ST(0) = &sv_undef; ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <12024-0@oink>; Mon, 30 Oct 1995 19:18:07 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 30 Oct 95 18:48:06 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa00566; 30 Oct 95 18:46 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA28283 for perldb-interest-real; Mon, 30 Oct 1995 08:39:24 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA28279 for ; Mon, 30 Oct 1995 08:39:17 -0800 Received: by gw.home.vix.com id AA01052; Mon, 30 Oct 95 08:41:32 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from kstock@localhost) by nova.auspex.fr (8.6.9/8.6.9) id RAA00293; Robinson France, Mon, 30 Oct 1995 17:42:42 +0100 Date: Mon, 30 Oct 1995 17:42:42 +0100 Message-Id: <199510301642.RAA00293@nova.auspex.fr> To: perldb-interest@vix.com Subject: Re: OraPerl Cc: kstock@nova.auspex.fr, djb2413@is2.nyu.edu From: Kevin Stock Organisation: Auspex Systems European Technical Support Centre Phone: +33 1 41 07 95 05 Reply-To: KStock@auspex.com Hello Dan, > I'm wondering if anyone out there has been able to > compile OraPerl for a Solaris (v2.4) machine running > Perl (v5.0) and Oracle (v7.2.2)??? No, I can safely say that no-one has been able to do this. Oraperl will not work with Perl v5 at all. You could either build Oraperl with Perl 4 or use Tim Bunce' DBD:Oracle which works with v5. Kevin ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12023-7@oink>; Mon, 30 Oct 1995 19:18:59 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 30 Oct 95 18:55:36 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id af03063; 30 Oct 95 18:54 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <13536-0@frigate.doc.ic.ac.uk>; Mon, 30 Oct 1995 17:17:37 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA28048 for perldb-interest-real; Mon, 30 Oct 1995 06:16:28 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA28044 for ; Mon, 30 Oct 1995 06:16:26 -0800 Received: by gw.home.vix.com id AA20381; Mon, 30 Oct 95 06:18:45 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from PORT-048.DIAL.NET.NYU.EDU by is2.NYU.EDU; (5.65v3.0/1.1.8.2/23Sep94-1121PM) id AA32381; Mon, 30 Oct 1995 09:18:36 -0500 Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 30 Oct 1995 10:17:55 -0500 To: perldb-interest@vix.com From: Dan Boyle Subject: OraPerl I've just joined this list so excuse me if this thread has already woven its ugly cloth. I'm wondering if anyone out there has been able to compile OraPerl for a Solaris (v2.4) machine running Perl (v5.0) and Oracle (v7.2.2)??? ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12023-15@oink>; Mon, 30 Oct 1995 19:19:41 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 30 Oct 95 19:12:43 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa11886; 30 Oct 95 19:12 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <15598-0@frigate.doc.ic.ac.uk>; Mon, 30 Oct 1995 18:56:11 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA28358 for perldb-interest-real; Mon, 30 Oct 1995 09:03:24 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA28354 for ; Mon, 30 Oct 1995 09:03:22 -0800 Received: by gw.home.vix.com id AA02568; Mon, 30 Oct 95 09:05:32 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from romulus.InterNex.Net by starfleet (SMI-8.6.9/SMI-SVR4) id JAA01673; Mon, 30 Oct 1995 09:05:29 -0800 Received: by romulus.InterNex.Net (5.x/SMI-SVR4) id AA04545; Mon, 30 Oct 1995 09:05:23 -0800 Date: Mon, 30 Oct 1995 09:05:21 -0800 (PST) From: Matt Ho X-Sender: matt@romulus To: Dan Boyle Cc: perldb-interest@vix.com Subject: Re: OraPerl In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 30 Oct 1995, Dan Boyle wrote: > I've just joined this list so excuse me if this thread has > already woven its ugly cloth. > > I'm wondering if anyone out there has been able to > compile OraPerl for a Solaris (v2.4) machine running > Perl (v5.0) and Oracle (v7.2.2)??? Yeah, though the server is running on an NT box (bleh) ''' (o o) --------------------------oOO--(_)--OOo-------------------------- Matt Ho PGP available on request WWW Architect Email: matt@InterNex.Net InterNex Information Services http://www.InterNex.Net/ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12701-23@oink>; Mon, 30 Oct 1995 20:49:20 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 30 Oct 95 20:46:31 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa28156; 30 Oct 95 20:45 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id aa08252; 30 Oct 95 20:45 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA28672 for perldb-interest-real; Mon, 30 Oct 1995 10:33:21 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA28668 for ; Mon, 30 Oct 1995 10:33:20 -0800 Received: by gw.home.vix.com id AA10490; Mon, 30 Oct 95 10:35:38 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from romulus.InterNex.Net by starfleet (SMI-8.6.9/SMI-SVR4) id KAA02880; Mon, 30 Oct 1995 10:35:31 -0800 Received: by romulus.InterNex.Net (5.x/SMI-SVR4) id AA05093; Mon, 30 Oct 1995 10:35:25 -0800 Date: Mon, 30 Oct 1995 10:35:24 -0800 (PST) From: Matt Ho X-Sender: matt@romulus To: Kevin Stock Cc: perldb-interest@vix.com, kstock@nova.auspex.fr, djb2413@is2.nyu.edu Subject: Re: OraPerl In-Reply-To: <199510301642.RAA00293@nova.auspex.fr> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 30 Oct 1995, Kevin Stock wrote: > > I'm wondering if anyone out there has been able to > > compile OraPerl for a Solaris (v2.4) machine running > > Perl (v5.0) and Oracle (v7.2.2)??? > > No, I can safely say that no-one has been able to do this. Oraperl will > not work with Perl v5 at all. You could either build Oraperl with Perl > 4 or use Tim Bunce' DBD:Oracle which works with v5. Actually, he's right. I built the DBD:Oracle package which has OraPerl emulation built into it. ''' (o o) --------------------------oOO--(_)--OOo-------------------------- Matt Ho PGP available on request WWW Architect Email: matt@InterNex.Net InterNex Information Services http://www.InterNex.Net/ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13732-62@oink>; Mon, 30 Oct 1995 22:35:32 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 30 Oct 95 20:55:04 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa02230; 30 Oct 95 20:53 GMT Received: from relay-4.mail.demon.net by relay-2.mail.demon.net id ae02460; 30 Oct 95 20:52 GMT Received: from punt.demon.co.uk by relay-4.mail.demon.net id sg.aa26014; 28 Oct 95 15:53 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa29607; 28 Oct 95 15:52 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA22129 for perldb-interest-real; Sat, 28 Oct 1995 07:08:30 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-2.mail.demon.net (disperse.demon.co.uk [158.152.1.77]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA22125 for ; Sat, 28 Oct 1995 07:08:26 -0700 Received: from post.demon.co.uk by relay-2.mail.demon.net id aa08154; 26 Oct 95 13:22 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id ai06692; 26 Oct 95 13:21 GMT Received: from toad by oink with SMTP (PP) id <19974-0@oink>; Thu, 26 Oct 1995 13:25:34 +0100 Received: by toad (5.0/SMI-SVR4) id AA20240; Thu, 26 Oct 1995 13:25:12 +0000 Date: Thu, 26 Oct 1995 13:25:12 +0000 From: Tim Bunce Message-Id: <9510261225.AA20240@toad> To: brent@dino.cortex.net Subject: Re: (Comments|Request) Cc: perldb-interest@fugue.com X-Sun-Charset: US-ASCII content-length: 1177 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Brent Halliburton > > How do I unsubscribe? Mail to perldb-interest-REQUEST@fugue.com > Or more accurately, could this list be subdivided somewhat. I am > interested in finding out when new versions of the software are released, > or about things like the Hermetica site, but I am not interested in > problems that people on platforms I will never use are having compiling the > DBI stuff. > > It would be nice if there was an announcements list (tightly moderated), a > compiling list, and a usage list. > > Please take these remarks as constructive. I like reading the list, but I > get too much e-mail to read about Sybase problems on HP machines, etc.. I > can't help them, and I will probably never have similar problems. > > Your assistance in either of these arenas would be very helpful. I've been having similar thoughts. I was thinking of three lists like this: dbi-announce obvious dbi-users general + discussion of the DBI interface evolution dbi-dev DBI & DBD driver development and porting Installation problems would be mailed to dbi-dev but you need not subscribe. I've CC'd this to the list for comments. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13732-88@oink>; Mon, 30 Oct 1995 22:37:21 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 30 Oct 95 20:55:39 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id ab02821; 30 Oct 95 20:54 GMT Received: from relay-4.mail.demon.net by relay-2.mail.demon.net id ap02525; 30 Oct 95 20:53 GMT Received: from punt.demon.co.uk by relay-4.mail.demon.net id sg.aa04811; 28 Oct 95 18:44 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa15159; 28 Oct 95 18:43 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA22516 for perldb-interest-real; Sat, 28 Oct 1995 10:33:36 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from ubmail.ubalt.edu (ubmail.ubalt.edu [198.202.0.25]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA22512 for ; Sat, 28 Oct 1995 10:33:33 -0700 Received: from UBmail.ubalt.edu by UBmail.ubalt.edu (PMDF V5.0-3 #7370) id <01HWZ4KR9LVK00CI6U@UBmail.ubalt.edu> for perldb-interest@fugue.com; Sat, 28 Oct 1995 13:34:14 -0500 (EST) Date: Sat, 28 Oct 1995 13:34:14 -0500 (EST) From: SHAZAD MALIK Subject: link between two different platforms ???? To: perldb-interest@fugue.com Message-id: <01HWZ4KRAY3M00CI6U@UBmail.ubalt.edu> X-VMS-To: MX%"perldb-interest@fugue.com" X-VMS-Cc: SMALIK MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Content-transfer-encoding: 7BIT Hi there... I have a database called "PROCITE" installed on my PC. I wanna create a search browser which will be performed from the Unix server(SGI) on my PC database. My problem is how do I make the link between the two platform, so that they communicate with each other. Is there any software which can create the link. I just dont understand how to make the translation from my UNIX to PC, in order to get and store information on my PC database. I have figured out some options: 1) install a DOS device on the UNIX machine. 2) Install a DOS hard disk on the UNIX machine while duplicating the PC database on it. 3) OR is there something much easier from the above two options. But I just fail to see the picture about the link between the two platforms. Any suggestion ...plezzzzzzz !!!! Shazad. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13732-119@oink>; Mon, 30 Oct 1995 22:39:59 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Mon, 30 Oct 95 20:57:10 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa03649; 30 Oct 95 20:56 GMT Received: from relay-4.mail.demon.net by relay-2.mail.demon.net id ar02611; 30 Oct 95 20:55 GMT Received: from punt.demon.co.uk by relay-4.mail.demon.net id msg.aa23448; 28 Oct 95 15:43 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa25544; 28 Oct 95 15:43 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA22135 for perldb-interest-real; Sat, 28 Oct 1995 07:09:06 -0700 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA22131 for ; Sat, 28 Oct 1995 07:09:04 -0700 Received: by gw.home.vix.com id AA20009; Sat, 28 Oct 95 07:11:17 -0700 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-2.mail.demon.net id ae08069; 26 Oct 95 13:22 GMT Received: from relay-2.mail.demon.net by relay-3.mail.demon.net id af06692; 26 Oct 95 13:21 GMT Received: from toad by oink with SMTP (PP) id <19907-0@oink>; Thu, 26 Oct 1995 13:06:43 +0100 Received: by toad (5.0/SMI-SVR4) id AA20228; Thu, 26 Oct 1995 13:06:21 +0000 Date: Thu, 26 Oct 1995 13:06:21 +0000 From: Tim Bunce Message-Id: <9510261206.AA20228@toad> To: gj@filoli.com, mouring@netnet.net Subject: Re: I think I have fetch almost working.. Cc: perldb-interest@vix.com, andy_lyman@filoli.com X-Sun-Charset: US-ASCII Content-Length: 405 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Ben Lindstrom > > I have the new QuickBase server for rollbacks and commits. You can send the > transcation commands via prepre() or via makeCommand:. > > Have you started working on multiply cursors active? I don't think it's > good for both of us to be working on the same functions. It might go > quicker to work together. Yes, cooperate, don't duplicate. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <17755-1@oink>; Tue, 31 Oct 1995 01:24:12 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 31 Oct 95 01:18:36 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa01988; 31 Oct 95 1:17 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA29299 for perldb-interest-real; Mon, 30 Oct 1995 16:03:50 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA29295 for ; Mon, 30 Oct 1995 16:03:43 -0800 Received: by relay-4.mail.demon.net id sg.aa14360; 31 Oct 95 0:02 GMT Received: from post.demon.co.uk by relay-4.mail.demon.net id g.ae13398; 30 Oct 95 23:49 GMT Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id ah14873; 30 Oct 95 23:48 GMT Received: from toad by oink with SMTP (PP) id <16122-0@oink>; Mon, 30 Oct 1995 23:43:13 +0000 Received: by toad (5.0/SMI-SVR4) id AA24470; Mon, 30 Oct 1995 23:42:51 +0000 Date: Mon, 30 Oct 1995 23:42:51 +0000 From: Tim Bunce Message-Id: <9510302342.AA24470@toad> To: perldb-interest@fugue.com, SMALIK@ubmail.ubalt.edu Subject: Re: link between two different platforms ???? X-Sun-Charset: US-ASCII content-length: 511 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: SHAZAD MALIK > > Hi there... > > I have a database called "PROCITE" installed on my PC. I wanna > create a search browser which will be performed from the Unix server(SGI) on > my PC database. My problem is how do I make the link between the two > platform, so that they communicate with each other. > > Any suggestion ...plezzzzzzz !!!! Join the www-rdb@w3.org mailing list. Subscribe via www-rdb-request@w3.org. Also ask in comp.infosystems.www.authoring.cgi. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <19207-1@oink>; Tue, 31 Oct 1995 07:06:19 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 31 Oct 95 03:07:38 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa06017; 31 Oct 95 3:07 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA29535 for perldb-interest-real; Mon, 30 Oct 1995 17:51:29 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA29531 for ; Mon, 30 Oct 1995 17:51:27 -0800 Received: by gw.home.vix.com id AA18526; Mon, 30 Oct 95 17:53:46 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id TAA08053; Mon, 30 Oct 1995 19:53:42 -0600 Date: Mon, 30 Oct 1995 19:53:42 -0600 (CST) From: Ben Lindstrom To: DBI perl Subject: Openning database on another machine. Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I noticed in Oracle you have some information in your dbd_db_login() that describes how to use other databases on different machines (if I understand it correctly.). Is this going to be the standard for identifying another machine for pulling information from? Or should the argument line be explanded for login to include "host machine"? Also, is there a standard way of adding in non-standard database routines. Example: QuickBase lets you set what software is currently accessing the database. It would be nice to beable to let the user change it. I sorta have the test.pl...but I'm unsure by quickly looking at it what the tangled web of perl code is attempting to do. You can take a look at the output tommorrow when I release version .02 alpha. --- Quote of the day: Anyways..progress is made slowly...=) (Really slowly when your being over taken by finals. =) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <23041-2@oink>; Tue, 31 Oct 1995 16:55:08 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 31 Oct 95 14:34:28 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa12652; 31 Oct 95 14:33 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA01642 for perldb-interest-real; Tue, 31 Oct 1995 03:57:35 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA01638 for ; Tue, 31 Oct 1995 03:57:33 -0800 Received: by gw.home.vix.com id AA29441; Tue, 31 Oct 95 03:59:51 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aa16189; 31 Oct 95 11:58 GMT Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id aa24423; 31 Oct 95 11:57 GMT Received: from toad by oink with SMTP (PP) id <21646-0@oink>; Tue, 31 Oct 1995 11:56:44 +0000 Received: by toad (5.0/SMI-SVR4) id AA25558; Tue, 31 Oct 1995 11:56:22 +0000 Date: Tue, 31 Oct 1995 11:56:22 +0000 From: Tim Bunce Message-Id: <9510311156.AA25558@toad> To: perldb-interest@vix.com, mouring@netnet.net Subject: Re: Openning database on another machine. X-Sun-Charset: US-ASCII Content-Length: 1035 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Ben Lindstrom > > I noticed in Oracle you have some information in your dbd_db_login() that > describes how to use other databases on different machines (if I > understand it correctly.). > > Is this going to be the standard for identifying another machine for > pulling information from? No. > Or should the argument line be explanded for login to include "host machine"? Basically the DBI standard says what the other standards say... How you specify the database to connect to is an implementation issue. If you're designing something from scratch I'd suggest following a URL style or an ODBC style. > Also, is there a standard way of adding in non-standard database routines. Yes: $h->func(...., $func_name); > I sorta have the test.pl...but I'm unsure by quickly looking at it what > the tangled web of perl code is attempting to do. You can take a look > at the output tommorrow when I release version .02 alpha. Please mail a copy to me. I've no time to go ftp fishing. Thanks. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <24122-6@oink>; Tue, 31 Oct 1995 18:44:04 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 31 Oct 95 18:32:23 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa11647; 31 Oct 95 18:31 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <10691-0@frigate.doc.ic.ac.uk>; Tue, 31 Oct 1995 18:06:41 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA01916 for perldb-interest-real; Tue, 31 Oct 1995 07:20:19 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA01912 for ; Tue, 31 Oct 1995 07:20:17 -0800 Received: by gw.home.vix.com id AA12162; Tue, 31 Oct 95 07:22:36 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id JAA04221; Tue, 31 Oct 1995 09:12:09 -0600 Date: Tue, 31 Oct 1995 09:12:09 -0600 (CST) From: Ben Lindstrom To: Tim Bunce Cc: perldb-interest@vix.com Subject: Re: Openning database on another machine. In-Reply-To: <9510311156.AA25558@toad> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 31 Oct 1995, Tim Bunce wrote: > > > From: Ben Lindstrom > > > > Or should the argument line be explanded for login to include "host machine"? > > Basically the DBI standard says what the other standards say... > How you specify the database to connect to is an implementation issue. > > If you're designing something from scratch I'd suggest following a URL style > or an ODBC style. > I guess my thoughts are it should be decided what front interface all DBI drivers will use. If it be: $dbh->connect('hostmachine.com','database','user','pass','driver'); or $dbh->connect('hostmachine.com/database',''usr',pass',driver'); So if/when other people plan on building clients that access a server on another machine it will be setup. (Of course that means that the DBD will have to figure out how to do the physical connect, just the interface will be documented). > > I sorta have the test.pl...but I'm unsure by quickly looking at it what > > the tangled web of perl code is attempting to do. You can take a look > > at the output tommorrow when I release version .02 alpha. > > Please mail a copy to me. I've no time to go ftp fishing. Thanks. > Ok..I'll add you to the list of people I'm sending the release (that should be .03 alpha =). I'll also send a copy of the output+errors. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <25679-20@oink>; Tue, 31 Oct 1995 21:22:27 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Tue, 31 Oct 95 20:56:39 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa14476; 31 Oct 95 20:55 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id aa22553; 31 Oct 95 20:54 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <13751-0@frigate.doc.ic.ac.uk>; Tue, 31 Oct 1995 20:50:19 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA02373 for perldb-interest-real; Tue, 31 Oct 1995 11:08:16 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA02369 for ; Tue, 31 Oct 1995 11:08:12 -0800 Received: by gw.home.vix.com id AA27538; Tue, 31 Oct 95 10:46:41 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab17408; 31 Oct 95 18:45 GMT Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id ae11268; 31 Oct 95 18:43 GMT Received: from toad by oink with SMTP (PP) id <24007-0@oink>; Tue, 31 Oct 1995 18:00:30 +0000 Received: by toad (5.0/SMI-SVR4) id AA25830; Tue, 31 Oct 1995 18:00:09 +0000 Date: Tue, 31 Oct 1995 18:00:09 +0000 From: Tim Bunce Message-Id: <9510311800.AA25830@toad> To: mouring@netnet.net Subject: Re: Openning database on another machine. Cc: perldb-interest@vix.com X-Sun-Charset: US-ASCII Content-Length: 1397 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Ben Lindstrom > > On Tue, 31 Oct 1995, Tim Bunce wrote: > > > > From: Ben Lindstrom > > > > > > Or should the argument line be explanded for login to include "host machine"? > > > > Basically the DBI standard says what the other standards say... > > How you specify the database to connect to is an implementation issue. > > > > If you're designing something from scratch I'd suggest following a URL style > > or an ODBC style. > > I guess my thoughts are it should be decided what front interface all > DBI drivers will use. If it be: > > $dbh->connect('hostmachine.com','database','user','pass','driver'); > > or > > $dbh->connect('hostmachine.com/database',''usr',pass',driver'); > > So if/when other people plan on building clients that access a server on > another machine it will be setup. (Of course that means that the DBD will > have to figure out how to do the physical connect, just the interface will > be documented). Databases, from the applications point of view, are logical things. Exposing physical details such as host name as an explicit part of the interface is generally not a good idea. I have no plans to change $dbh->connect arguments or argument handling. It matches SQL CLI and other interfaces well. It _is_ possible that the DBI->connect may do something fancy one day with it's first argument. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <00857-0@oink>; Wed, 1 Nov 1995 07:05:50 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 05:18:28 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa23614; 1 Nov 95 5:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA03508 for perldb-interest-real; Tue, 31 Oct 1995 20:16:31 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA03504 for ; Tue, 31 Oct 1995 20:16:29 -0800 Received: by gw.home.vix.com id AA08420; Tue, 31 Oct 95 20:18:47 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: by proxy.netlabs.com (4.1/SMI-4.1) id AA24674; Tue, 31 Oct 95 20:23:47 PST Received: from vaccine-bb.netlabs.com(192.190.199.1) by proxy.netlabs.com via smap (V1.3) id sma024671; Tue Oct 31 20:23:21 1995 Received: from tile (tile.netlabs.com) by netlabs.com (4.1/SMI-4.1) id AA29226; Tue, 31 Oct 95 20:20:28 PST From: Chuck Kung Message-Id: <9511010420.AA29226@netlabs.com> Subject: Re: OraPerl To: Matt Ho Date: Tue, 31 Oct 95 20:16:49 PST Cc: djb2413@is2.nyu.edu, perldb-interest@vix.com In-Reply-To: ; from "Matt Ho" at Oct 30, 95 9:05 am X-Mailer: ELM [version 2.3 PL11] hi, i'm interested in the current progress of DBperl. how far along is it? is it in the specification stage or implementation stage? i'm deciding now whether to use sybperl of isqlperl but could wait if DBperl is near. thanks, chuck ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <03046-2@oink>; Wed, 1 Nov 1995 12:01:42 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 11:41:02 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa01193; 1 Nov 95 11:40 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA05239 for perldb-interest-real; Wed, 1 Nov 1995 02:14:57 -0800 Precedence: junk Sender: perldb-interest-request@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA05235 for ; Wed, 1 Nov 1995 02:14:54 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id KAA23443; Wed, 1 Nov 1995 10:12:32 GMT From: Alligator Descartes Message-Id: <199511011012.KAA23443@fruitbat.mcqueen.com> Subject: Re: OraPerl To: Chuck Kung Date: Wed, 1 Nov 1995 10:12:31 +0000 (GMT) Cc: perldb-interest@fugue.com In-Reply-To: <9511010420.AA29226@netlabs.com> from "Chuck Kung" at Oct 31, 95 08:16:49 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 817 > i'm interested in the current progress of DBperl. > how far along is it? is it in the specification > stage or implementation stage? i'm deciding now > whether to use sybperl of isqlperl but could wait if > DBperl is near. You can find a load of information on DBI at: http://www.hermetica.com/technologia/DBI There are currently a few drivers implemented for it including: Oracle by Tim Bunce ( complete-ish ) mSQL by myself ( complete-ish ) Informix by myself ( early development, but looking good! ) Quickbase by Ben Lindstrom ( very early development ) Sybase has been allocated, but I don't know what stage it's at. Probably "about to be implemented" Hope this helps. > chuck -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <07848-19@oink>; Wed, 1 Nov 1995 21:21:28 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 21:14:35 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa01165; 1 Nov 95 21:14 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA05970 for perldb-interest-real; Wed, 1 Nov 1995 09:34:51 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA05965 for ; Wed, 1 Nov 1995 09:34:50 -0800 Message-Id: <199511011734.JAA05965@toccata.fugue.com> To: perldb-interest@toccata.fugue.com Subject: The perldb-interest mailing list is now obsolete. Date: Wed, 01 Nov 1995 09:34:49 -0800 From: Ted Lemon The perldb-interest mailing list has been split into three mailing lists; one for users, one for developers and one for announcements of interest to both. The perldb-interest mailing list will be gatewayed to the dbi-users mailing list for the time being. If you are involved in a developer-related discussion, please redirect it to dbi-dev@fugue.com. If you have an announcement to make, send it to dbi-announce@fugue.com. If you are engaging in discussions of interest to all DBI users, please redirect that to dbi-users@fugue.com. Once the activity on perldb-interest has died down, I'll be disabling the list and putting in an automatic responder telling people what's happened. _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <08649-0@oink>; Wed, 1 Nov 1995 22:00:27 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 21:57:50 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id ad06514; 1 Nov 95 21:56 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <05710-0@frigate.doc.ic.ac.uk>; Wed, 1 Nov 1995 21:19:57 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA06169 for dbi-users-real; Wed, 1 Nov 1995 10:13:55 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: (from root@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA06164 for dbi-users; Wed, 1 Nov 1995 10:13:54 -0800 Message-Id: <199511011813.KAA06164@toccata.fugue.com> To: dbi-users@toccata.fugue.com Subject: Welcome to the DBI Users' mailing list! Date: Wed, 01 Nov 1995 09:27:22 -0800 From: Ted Lemon The perldb-interest mailing list has been split into three mailing lists; one for users, one for developers and one for announcements of interest to both. The DBI Users' mailing list is intended for users who wish to share information and experience about the use of the perl DBI. Developers are obviously also welcome on the list, but the list is not intended for the discussion of issues strictly of interest to developers. If you do not wish to be on the DBI Users' mailing list, please send mail to dbi-users-request including your full name and email address and asking to be removed. _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <08649-11@oink>; Wed, 1 Nov 1995 22:01:25 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 21:57:53 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id ae06514; 1 Nov 95 21:56 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <05842-0@frigate.doc.ic.ac.uk>; Wed, 1 Nov 1995 21:26:45 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA06153 for dbi-announce-real; Wed, 1 Nov 1995 10:13:27 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: (from root@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA06148 for dbi-announce; Wed, 1 Nov 1995 10:13:26 -0800 Message-Id: <199511011813.KAA06148@toccata.fugue.com> To: dbi-announce@toccata.fugue.com Subject: Welcome to the DBI Announcement mailing list! Date: Wed, 01 Nov 1995 09:31:34 -0800 From: Ted Lemon The perldb-interest mailing list has been split into three mailing lists; one for users, one for developers and one for announcements of interest to both. The DBI Announcement mailing list is intended for announcements about the availability of new versions of DBI software, new discussions about DBI software, and new DBI-related resources. It is not intended for discussions of any kind. If you do not wish to be on the DBI Announcement mailing list, please send mail to dbi-announce-request including your full name and email address and asking to be removed. _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <08649-12@oink>; Wed, 1 Nov 1995 22:01:31 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 21:58:06 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id ag06514; 1 Nov 95 21:56 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <06211-0@frigate.doc.ic.ac.uk>; Wed, 1 Nov 1995 21:48:28 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA06016 for perldb-interest-real; Wed, 1 Nov 1995 09:47:28 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA06012 for ; Wed, 1 Nov 1995 09:47:24 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id LAA52678 for ; Wed, 1 Nov 1995 11:49:39 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for perldb-interest@fugue.com at austin.ibm.com; id AA19842; Wed, 1 Nov 1995 11:49:38 -0600 Message-Id: <9511011749.AA19842@dax.austin.ibm.com> To: perldb-interest@fugue.com Subject: Dealing with blobs Date: Wed, 01 Nov 95 11:49:37 -0600 My appologies if you have seen this before, but between Tim's mail problems and my department moving and changing IP address I have not seen a reply to the following message. Any one have any ideas about what is happening? Mike mhm@austin.ibm.com ------- Forwarded Message Subject: Re: What am I doing wrong...got to be something stupid In-Reply-To: Your message of Thu, 26 Oct 95 21:09:19 +0000. <9510262009.AA21540@toad> Date: Fri, 27 Oct 95 06:47:01 -0500 X-Filter: mailagent [version 3.0 PL44] for mhm@austin.ibm.com Tim Bunce writes: |> > From: mhm@austin.ibm.com |> > |> > Tim Bunce writes: |> > |> |> > |> - ST(0) = destrv; |> > |> + ST(0) = SvRV(destrv); |> > |> > After making the above change, something is still not right with |> > dbd_st_readblob or my understanding of perlguts is still severly lacking. |> > |> > From dbd_st_readblob, my assumption has been that the line |> > |> > bufsv = SvRV(destrv); |> > |> > causes destrv to be dereferenced, which means that bufsv and destrv |> > would both be pointing at the same memory. However, when looking at |> > both destrv and bufsv in the debugger after the SQLGetData call, I |> > see the appropriate piece of the blob in bufsv->sv_any->xpvav.xav_array |> > and an unchanged destrv->sv_any->xpvav.xav_array. |> |> You should be looking at destrv->sv_any->xpv.xpv_pv but no matter, |> they both have the same shape: |> |> struct xpvav { |> char* xav_array; /* pointer to malloced string */ |> |> struct xpv { |> char * xpv_pv; /* pointer to malloced string */ |> |> so, the data is there! |> |> > So just what am I missing? Is it not getting passed in correctly? |> |> You've not said what $sth->readblob is returning now! I don't know exactly what it is, but it looks like $sth->readblob is walking through the perl memory areas (I'll enclose a sample at the end). |> |> > rtval = SQLGetData(imp_sth->phstmt,field,SQL_C_BINARY, |> > ((void *)SvPVX(bufsv)),len+destoffset,&retl); |> > msg = (ERRTYPE(rtval) ? "GetData failed to read blob":"Invalid Handle"); |> > rtval = check_error(sth,rtval,henv,imp_dbh->hdbc,imp_sth->phstmt,msg); |> > EOI(rtval); |> > |> > SvCUR_set(bufsv, len ); |> > *SvEND(bufsv) = '\0'; /* consistent with perl sv_setpvn etc */ |> |> It's probably not relevant but |> |> SvCUR_set(bufsv, len ); |> should be |> SvCUR_set(bufsv, offset+retl ); Unfortunately, this is one of DB2 CLI's defeciencies. I don't know how many bytes were actually read. What I do know is how many bytes remaining. So far, I have not come up with a reasonable method of determining what retl is with out burning a lot of cycles. Mike mhm@austin.ibm.com As promised a sample of dump of what is being returned by $sth->readblob(): 0000000 0 0 0 1 3 0 \0 \0 \0 \0 \0 222 \0 \0 \0 T 0000020 \0 \0 \0 003 \0 \0 016 \0 \0 \0 002 002 \0 \0 016 T 0000040 \0 \0 \0 \0 \0 \0 \0 002 \0 \0 \0 \0 \0 \0 @ 004 0000060 \0 \0 \0 001 \0 \0 \0 \0 _ i o b \0 \0 \0 \0 0000100 \0 \0 \0 \0 \0 \0 @ 005 \0 \0 \0 001 \0 \0 \0 \0 0000120 s t r l e n \0 \0 \0 \0 \0 \0 \0 \0 @ \n 0000140 \0 \0 \0 001 \0 \0 \0 \0 m a l l o c \0 \0 0000160 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 001 \0 \0 \0 \0 0000200 f r e e \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 @ \n 0000220 \0 \0 \0 001 \0 \0 \0 \0 s p r i n t f \0 0000240 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 001 \0 \0 \0 \0 0000260 f p r i n t f \0 \0 \0 \0 \0 \0 \0 @ \n 0000300 \0 \0 \0 001 \0 \0 \0 \0 a t o i \0 \0 \0 \0 0000320 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 001 \0 \0 \0 \0 0000340 i s d i g i t \0 \0 \0 \0 \0 \0 \0 @ \n 0000360 \0 \0 \0 001 \0 \0 \0 \0 i s a l p h a \0 0000400 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 001 \0 \0 \0 \0 0000420 S v \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 @ 004 0000440 \0 \0 \0 \0 \0 \0 \0 \0 X p v \0 \0 \0 \0 \0 0000460 \0 \0 \0 \0 \0 \0 @ 004 \0 \0 \0 \0 \0 \0 \0 \0 0000500 \0 \0 \0 \0 \0 \0 \0 032 \0 \0 \0 \0 \0 \0 @ 004 0000520 \0 \0 \0 \0 \0 \0 \0 \0 n a \0 \0 \0 \0 \0 \0 0000540 \0 \0 \0 \0 \0 \0 @ 004 \0 \0 \0 \0 \0 \0 \0 \0 0000560 o p \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 @ 004 0000600 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0000620 \0 \0 \0 \0 \0 \0 @ 004 \0 \0 \0 \0 \0 \0 \0 \0 0000640 \0 \0 \0 \0 \0 \0 \0 7 \0 \0 \0 \0 \0 \0 @ 004 0000660 \0 \0 \0 \0 \0 \0 \0 \0 s t a c k _ s p 0000700 \0 \0 \0 \0 \0 \0 @ 004 \0 \0 \0 \0 \0 \0 \0 \0 0000720 s v _ n o \0 \0 \0 \0 \0 \0 \0 \0 \0 @ 004 0000740 \0 \0 \0 \0 \0 \0 \0 \0 s v _ u n d e f 0000760 \0 \0 \0 \0 \0 \0 @ 004 \0 \0 \0 \0 \0 \0 \0 \0 0001000 s v _ y e s \0 \0 \0 \0 \0 \0 \0 \0 @ 004 0001020 \0 \0 \0 \0 \0 \0 \0 \0 a v _ s t o r e 0001040 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001060 c r o a k \0 \0 \0 \0 \0 \0 \0 \0 \0 @ \n 0001100 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 C 0001120 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001140 h v _ f e t c h \0 \0 \0 \0 \0 \0 @ \n 0001160 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 Q 0001200 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001220 \0 \0 \0 \0 \0 \0 \0 _ \0 \0 \0 \0 \0 \0 @ \n 0001240 \0 \0 \0 \0 \0 \0 \0 \0 h v _ s t o r e 0001260 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001300 n e w A V \0 \0 \0 \0 \0 \0 \0 \0 \0 @ \n 0001320 \0 \0 \0 \0 \0 \0 \0 \0 n e w H V \0 \0 \0 0001340 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001360 n e w R V \0 \0 \0 \0 \0 \0 \0 \0 \0 @ \n 0001400 \0 \0 \0 \0 \0 \0 \0 \0 n e w S V \0 \0 \0 0001420 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001440 n e w S V i v \0 \0 \0 \0 \0 \0 \0 @ \n 0001460 \0 \0 \0 \0 \0 \0 \0 \0 n e w S V p v \0 0001500 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001520 n e w X S \0 \0 \0 \0 \0 \0 \0 \0 \0 @ \n 0001540 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 o 0001560 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001600 s v _ 2 b o o l \0 \0 \0 \0 \0 \0 @ \n 0001620 \0 \0 \0 \0 \0 \0 \0 \0 s v _ 2 i v \0 \0 0001640 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001660 \0 \0 \0 \0 \0 \0 \0 | \0 \0 \0 \0 \0 \0 @ \n 0001700 \0 \0 \0 \0 \0 \0 \0 \0 s v _ 2 p v \0 \0 0001720 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0001740 \0 \0 \0 \0 \0 \0 \0 211 \0 \0 \0 \0 \0 \0 @ \n 0001760 \0 \0 \0 \0 \0 \0 \0 \0 s v _ c a t p v 0002000 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0002020 s v _ f r e e \0 \0 \0 \0 \0 \0 \0 @ \n 0002040 \0 \0 \0 \0 \0 \0 \0 \0 s v _ g r o w \0 0002060 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0002100 \0 \0 \0 \0 \0 \0 \0 226 \0 \0 \0 \0 \0 \0 @ \n 0002120 \0 \0 \0 \0 \0 \0 \0 \0 s v _ s e t i v 0002140 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0002160 s v _ s e t p v \0 \0 \0 \0 \0 \0 @ \n 0002200 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 246 0002220 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0002240 s v _ s e t s v \0 \0 \0 \0 \0 \0 @ \n 0002260 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 262 0002300 \0 \0 \0 \0 \0 \0 @ \n \0 \0 \0 \0 \0 \0 \0 \0 0002320 w a r n \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 @ \n 0002340 \0 \0 \0 \0 \0 \0 \0 \0 d i r t y \0 \0 \0 ------- End of Forwarded Message ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <08649-15@oink>; Wed, 1 Nov 1995 22:01:47 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 21:58:01 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id ah06514; 1 Nov 95 21:56 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <06225-0@frigate.doc.ic.ac.uk>; Wed, 1 Nov 1995 21:48:30 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA06161 for dbi-dev-real; Wed, 1 Nov 1995 10:13:41 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: (from root@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA06156 for dbi-dev; Wed, 1 Nov 1995 10:13:41 -0800 Message-Id: <199511011813.KAA06156@toccata.fugue.com> To: dbi-dev@toccata.fugue.com Subject: Welcome to the DBI Developers' mailing list! Date: Wed, 01 Nov 1995 09:30:22 -0800 From: Ted Lemon The perldb-interest mailing list has been split into three mailing lists; one for users, one for developers and one for announcements of interest to both. The DBI Developers' mailing list is intended for DBI and DBD developers who wish to discuss implementation issues, standards issues and bug fixes. User-related issues such as how to build and/or install the DBI, where to get it, and so on, are not appropriate on this list - please direct such queries to the DBI Users' mailing list. If you do not wish to be on the DBI Developers' mailing list, please send mail to dbi-dev-request including your full name and email address and asking to be removed. _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <08649-16@oink>; Wed, 1 Nov 1995 22:02:01 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 01 Nov 95 21:58:02 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id ai06514; 1 Nov 95 21:57 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <06342-0@frigate.doc.ic.ac.uk>; Wed, 1 Nov 1995 21:52:15 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA06323 for dbi-announce-real; Wed, 1 Nov 1995 10:35:54 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA06318 for ; Wed, 1 Nov 1995 10:35:51 -0800 Message-Id: <199511011835.KAA06318@toccata.fugue.com> To: dbi-announce@toccata.fugue.com Subject: Do not send me subscription requests if you're already subscribed!!! Date: Wed, 01 Nov 1995 10:35:50 -0800 From: Ted Lemon I'm starting to get subscription requests from existing members for the new mailing list. Please don't. If you were on perldb-interest, you're already on the new lists. Only send me requests to unsubscribe to the lists you don't want to be on. Thanks! _MelloN_ ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <10903-4@oink>; Thu, 2 Nov 1995 07:06:12 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 02 Nov 95 00:13:36 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa23329; 2 Nov 95 0:13 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA07575 for dbi-dev-real; Wed, 1 Nov 1995 14:28:06 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA07571; Wed, 1 Nov 1995 14:28:03 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA29379; Wed, 1 Nov 1995 22:25:51 GMT From: Alligator Descartes Message-Id: <199511012225.WAA29379@fruitbat.mcqueen.com> Subject: DBD::Qbase 0.03 To: perldb-interest@fugue.com, dbi-dev@fugue.com, mouring@netnet.net, Tim.Bunce@ig.co.uk Date: Wed, 1 Nov 1995 22:25:49 +0000 (GMT) Cc: Alligator Descartes X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2064 I've just taken a quick peek at the code. Some comments: o Your build instructions are slightly odd in that you suggest building DBD::Qbase within the perl5 source tree. DBI and the other DBDs state you *shouldn't* build within the perl5 source tree. o You don't appear to be using any of the structures Tim has put in place for the purpose of using! Eg, the imp_dbh structures are to be used for connections. The imp_fbh structures for fetching data from the database. I don't see them. Maybe Quickbase is odd that way, I don't know, but you're not helping yourself by not using them! I'd suggest ( again ) crefully reading the source code for DBD::Oracle and DBD::mSQL ( and maybe DBD::Informix ). Those clearly show the function flow for doing database stuff. Excluding binds, a normal session could be: connect prepare a cursor execute fetch the rows finish cursor disconnect Not entirely unusual. Tell me if Quickbase is different. The connection is handled via the _login function in .xs. This calls dbd_db_login. That handles the connection and population of the lda ( which is not Oracle-specific, although, the Oracle implementtion of it is! Check DBD::mSQL for more hints on transporting it ). dbd_st_prepare is called next through the stubs in the XS code. This prepares the cursor for execution. In mSQL, the tests the statement for inserts et al, since they are handled differently than SELECTs. In DBD::Informix, it opens a cursor with the SQL provided. execute is called via the XS code again. You backend mileage will vary heavily here. dbd_describe I regard as the guts of the driver's fetch functionality. This reads the structures of the data returned by the cursor and creates SVs which are the data types that perl can cope with. This routine doesn't appear to exist in your code. disconnect merely frees the imp_dbh structure created above. Hope this helps a bit. Regards. A. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <10902-13@oink>; Thu, 2 Nov 1995 07:06:53 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 02 Nov 95 00:28:16 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id af11376; 2 Nov 95 0:27 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <09405-0@frigate.doc.ic.ac.uk>; Thu, 2 Nov 1995 00:08:07 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA07583 for perldb-interest-real; Wed, 1 Nov 1995 14:28:25 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA07571; Wed, 1 Nov 1995 14:28:03 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA29379; Wed, 1 Nov 1995 22:25:51 GMT From: Alligator Descartes Message-Id: <199511012225.WAA29379@fruitbat.mcqueen.com> Subject: DBD::Qbase 0.03 To: perldb-interest@fugue.com, dbi-dev@fugue.com, mouring@netnet.net, Tim.Bunce@ig.co.uk Date: Wed, 1 Nov 1995 22:25:49 +0000 (GMT) Cc: Alligator Descartes X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2064 I've just taken a quick peek at the code. Some comments: o Your build instructions are slightly odd in that you suggest building DBD::Qbase within the perl5 source tree. DBI and the other DBDs state you *shouldn't* build within the perl5 source tree. o You don't appear to be using any of the structures Tim has put in place for the purpose of using! Eg, the imp_dbh structures are to be used for connections. The imp_fbh structures for fetching data from the database. I don't see them. Maybe Quickbase is odd that way, I don't know, but you're not helping yourself by not using them! I'd suggest ( again ) crefully reading the source code for DBD::Oracle and DBD::mSQL ( and maybe DBD::Informix ). Those clearly show the function flow for doing database stuff. Excluding binds, a normal session could be: connect prepare a cursor execute fetch the rows finish cursor disconnect Not entirely unusual. Tell me if Quickbase is different. The connection is handled via the _login function in .xs. This calls dbd_db_login. That handles the connection and population of the lda ( which is not Oracle-specific, although, the Oracle implementtion of it is! Check DBD::mSQL for more hints on transporting it ). dbd_st_prepare is called next through the stubs in the XS code. This prepares the cursor for execution. In mSQL, the tests the statement for inserts et al, since they are handled differently than SELECTs. In DBD::Informix, it opens a cursor with the SQL provided. execute is called via the XS code again. You backend mileage will vary heavily here. dbd_describe I regard as the guts of the driver's fetch functionality. This reads the structures of the data returned by the cursor and creates SVs which are the data types that perl can cope with. This routine doesn't appear to exist in your code. disconnect merely frees the imp_dbh structure created above. Hope this helps a bit. Regards. A. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <10903-12@oink>; Thu, 2 Nov 1995 07:06:52 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 02 Nov 95 00:50:28 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa06041; 2 Nov 95 0:49 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA07877 for dbi-dev-real; Wed, 1 Nov 1995 15:17:51 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netnet1.netnet.net (netnet1.netnet.net [198.70.64.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA07861; Wed, 1 Nov 1995 15:17:15 -0800 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id RAA15312; Wed, 1 Nov 1995 17:18:01 -0600 Date: Wed, 1 Nov 1995 17:18:01 -0600 (CST) From: Ben Lindstrom To: Alligator Descartes cc: perldb-interest@fugue.com, dbi-dev@fugue.com, Tim.Bunce@ig.co.uk, Alligator Descartes Subject: Re: DBD::Qbase 0.03 In-Reply-To: <199511012225.WAA29379@fruitbat.mcqueen.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Wed, 1 Nov 1995, Alligator Descartes wrote: > > I've just taken a quick peek at the code. Some comments: > > o Your build instructions are slightly odd in that you suggest > building DBD::Qbase within the perl5 source tree. DBI and the other > DBDs state you *shouldn't* build within the perl5 source tree. > Since the DBD is built using Objective-Cism it does not currently lend itself being dynamicaly loaded. I've spent 20 to 30 hours of compiling and recompiling and modifying to try and make it work, and I've currently given up...I'm not sure if one can staticly compile the DBD and dynamicly compile the DBI. If you would have read the TODO it would have stated this problem. It's something I don't have time to play with due to dead lines, but from what I've heard from people in the NeXT comunity it's a problem with Perl's lack of knowledge of Objective-C. > o You don't appear to be using any of the structures Tim has put in > place for the purpose of using! > I know, the structures go hand and hand with multiply Cursors which is the next project. I felt it would be easier to debug a single staticly built connect while I learned how DBD's were built. Version .04 Alpha will fix this drawback and support multiply cursors. > Eg, the imp_dbh structures are to be used for connections. The > imp_fbh structures for fetching data from the database. I don't see > them. > > Maybe Quickbase is odd that way, I don't know, but you're not > helping yourself by not using them! > QuickBase is an Object. So most of the structures are not needed...However imp_dbh is needed for multiply database connects. (see above) > I'd suggest ( again ) crefully reading the source code for DBD::Oracle and > DBD::mSQL ( and maybe DBD::Informix ). Those clearly show the function > flow for doing database stuff. Excluding binds, a normal session could be: > binding is something I'm unsure of at the moment. But this driver works as a single cursor system. I have small test databases to prove it.=) > dbd_st_prepare is called next through the stubs in the XS code. This prepares > the cursor for execution. In mSQL, the tests the statement for inserts et al, > since they are handled differently than SELECTs. In DBD::Informix, it opens > a cursor with the SQL provided. > Correct me if I'm wrong but prepare takes as it's arguement an SQL statement. Correct? (Note: I'm leaving out bind shortcuts for the moment =) QuickBase takes as it's arguement SQL/Transcation commands. So the only thing that seems to be missing is multiply-cursors (coming) and user define binds/shortcuts. > dbd_describe I regard as the guts of the driver's fetch functionality. This > reads the structures of the data returned by the cursor and creates SVs which > are the data types that perl can cope with. This routine doesn't appear to > exist in your code. > That is becuase currently...(and might be very improper) split between the execute comand (see the for (....) [qb bindStr...] for binding string pointers to the internal QuickBase Object) and in fetch (does an newsSVpv/av_push for each column so it can be returned). > Hope this helps a bit. > =) Yes it does. It conforms most of the things that the DBD-QBase lacks that I know about. And also helps me to verify that changes need to be made before the next release. Sorry if the current version of the driver looks a bit strange compared to mSQL, Oracle, but most of the deep database work is hidden via Objective-C. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <10903-24@oink>; Thu, 2 Nov 1995 07:08:31 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 02 Nov 95 01:45:58 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id ae23556; 2 Nov 95 1:45 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <10640-0@frigate.doc.ic.ac.uk>; Thu, 2 Nov 1995 00:51:40 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA07866 for perldb-interest-real; Wed, 1 Nov 1995 15:17:20 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netnet1.netnet.net (netnet1.netnet.net [198.70.64.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA07861; Wed, 1 Nov 1995 15:17:15 -0800 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id RAA15312; Wed, 1 Nov 1995 17:18:01 -0600 Date: Wed, 1 Nov 1995 17:18:01 -0600 (CST) From: Ben Lindstrom To: Alligator Descartes cc: perldb-interest@fugue.com, dbi-dev@fugue.com, Tim.Bunce@ig.co.uk, Alligator Descartes Subject: Re: DBD::Qbase 0.03 In-Reply-To: <199511012225.WAA29379@fruitbat.mcqueen.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Wed, 1 Nov 1995, Alligator Descartes wrote: > > I've just taken a quick peek at the code. Some comments: > > o Your build instructions are slightly odd in that you suggest > building DBD::Qbase within the perl5 source tree. DBI and the other > DBDs state you *shouldn't* build within the perl5 source tree. > Since the DBD is built using Objective-Cism it does not currently lend itself being dynamicaly loaded. I've spent 20 to 30 hours of compiling and recompiling and modifying to try and make it work, and I've currently given up...I'm not sure if one can staticly compile the DBD and dynamicly compile the DBI. If you would have read the TODO it would have stated this problem. It's something I don't have time to play with due to dead lines, but from what I've heard from people in the NeXT comunity it's a problem with Perl's lack of knowledge of Objective-C. > o You don't appear to be using any of the structures Tim has put in > place for the purpose of using! > I know, the structures go hand and hand with multiply Cursors which is the next project. I felt it would be easier to debug a single staticly built connect while I learned how DBD's were built. Version .04 Alpha will fix this drawback and support multiply cursors. > Eg, the imp_dbh structures are to be used for connections. The > imp_fbh structures for fetching data from the database. I don't see > them. > > Maybe Quickbase is odd that way, I don't know, but you're not > helping yourself by not using them! > QuickBase is an Object. So most of the structures are not needed...However imp_dbh is needed for multiply database connects. (see above) > I'd suggest ( again ) crefully reading the source code for DBD::Oracle and > DBD::mSQL ( and maybe DBD::Informix ). Those clearly show the function > flow for doing database stuff. Excluding binds, a normal session could be: > binding is something I'm unsure of at the moment. But this driver works as a single cursor system. I have small test databases to prove it.=) > dbd_st_prepare is called next through the stubs in the XS code. This prepares > the cursor for execution. In mSQL, the tests the statement for inserts et al, > since they are handled differently than SELECTs. In DBD::Informix, it opens > a cursor with the SQL provided. > Correct me if I'm wrong but prepare takes as it's arguement an SQL statement. Correct? (Note: I'm leaving out bind shortcuts for the moment =) QuickBase takes as it's arguement SQL/Transcation commands. So the only thing that seems to be missing is multiply-cursors (coming) and user define binds/shortcuts. > dbd_describe I regard as the guts of the driver's fetch functionality. This > reads the structures of the data returned by the cursor and creates SVs which > are the data types that perl can cope with. This routine doesn't appear to > exist in your code. > That is becuase currently...(and might be very improper) split between the execute comand (see the for (....) [qb bindStr...] for binding string pointers to the internal QuickBase Object) and in fetch (does an newsSVpv/av_push for each column so it can be returned). > Hope this helps a bit. > =) Yes it does. It conforms most of the things that the DBD-QBase lacks that I know about. And also helps me to verify that changes need to be made before the next release. Sorry if the current version of the driver looks a bit strange compared to mSQL, Oracle, but most of the deep database work is hidden via Objective-C. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <10903-33@oink>; Thu, 2 Nov 1995 07:09:56 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 02 Nov 95 01:46:08 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aj23556; 2 Nov 95 1:45 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <09037-0@frigate.doc.ic.ac.uk>; Wed, 1 Nov 1995 23:51:43 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA07356 for perldb-interest-real; Wed, 1 Nov 1995 13:48:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA07347 for ; Wed, 1 Nov 1995 13:48:07 -0800 Received: by relay-4.mail.demon.net id sg.ab07018; 1 Nov 95 21:43 GMT Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ak05338; 1 Nov 95 21:23 GMT Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id aa17534; 1 Nov 95 21:19 GMT Received: from toad by oink with SMTP (PP) id <07787-0@oink>; Wed, 1 Nov 1995 20:50:57 +0000 Received: by toad (5.0/SMI-SVR4) id AA02021; Wed, 1 Nov 1995 20:50:36 +0000 Date: Wed, 1 Nov 1995 20:50:36 +0000 From: Tim Bunce Message-Id: <9511012050.AA02021@toad> To: perldb-interest@fugue.com Subject: Update X-Sun-Charset: US-ASCII content-length: 380 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 Just a quickie before I go away for a few days (till Tuesday). I did actually make a release last week (!) but our ftp server was, and still is, down. I won't tease you with the Changes (they were minor tidy-ups anyway). Meanwhile I've pressed ahead and I hope to release the Oraperl emulation (only) as a beta after just one more alpha release. Till next week (or so :) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <10902-40@oink>; Thu, 2 Nov 1995 07:11:21 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 02 Nov 95 04:01:49 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa18161; 2 Nov 95 4:00 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <14542-0@frigate.doc.ic.ac.uk>; Thu, 2 Nov 1995 03:50:31 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA08569 for dbi-dev-real; Wed, 1 Nov 1995 18:01:26 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from filoli.filoli.com (filoli.com [204.162.0.10]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA08565 for ; Wed, 1 Nov 1995 18:01:24 -0800 Received: from sunspot.filoli.com (root@sunspot.filoli.com [204.162.1.17]) by filoli.filoli.com (8.6.10/8.6.9) with ESMTP id SAA27015; Wed, 1 Nov 1995 18:03:03 -0800 Received: from orac.filoli.com (orac.filoli.com [204.162.1.137]) by sunspot.filoli.com (8.6.12/8.6.9) with SMTP id SAA20366; Wed, 1 Nov 1995 18:02:56 -0800 Message-Id: <199511020202.SAA20366@sunspot.filoli.com> Received: by orac.filoli.com (NX5.67e/NX3.0X) id AA12833; Wed, 1 Nov 95 18:02:46 -0800 Mime-Version: 1.0 (NeXT Mail 3.3 v118.2) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Original-Received: by NeXT.Mailer (1.118.2) PP-warning: Illegal Received field on preceding line From: George Johnston Date: Wed, 1 Nov 95 18:02:43 -0800 To: dbi-dev@fugue.com Subject: Re: DBD::Qbase 0.03 Cc: Alligator Descartes , Tim.Bunce@ig.co.uk, Ben Lindstrom References: Yes, Alligator you are right, we haven't used the internal stuff. = I had a quick deadline and am totally new to perl internals. I am = impressed by its used of memory handles and I need to learn the = nuances of when one uses sv_newmortal vs newSViv . Once I am familiar with the memory allocation and release routines = and knowing when to call which ones where, then I can use those = impish things to keep track of the connection information which = (because it is an object pointer as Ben pointed out) is trivial to = keep track of once I know where to put it without stepping on = memory that DBI is using for its purposes. Your email has gone a = long way to showing some of these nuances. Further enlightenment = (or a pointer to such) would be greatly appreciated. Because QuickBase is Objective-C then perlmain.c needs to be = compiled with the -ObjC option. I know for my users of = perl/QuickBase it was much more important for them to be able to = use QuickBase than to have it dynamically loaded thus I spent no = cycles on the dynamic loading problem especially since the Obj-C = runtime (I believe) needs to be linked in statically. Thanks for the code review and I know I will have more questions. George ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <17121-38@oink>; Fri, 3 Nov 1995 07:10:16 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Thu, 02 Nov 95 23:44:23 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa03575; 2 Nov 95 23:43 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA12225 for dbi-dev-real; Thu, 2 Nov 1995 14:02:55 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA12221 for ; Thu, 2 Nov 1995 14:02:51 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA09605; Thu, 2 Nov 1995 22:00:30 GMT From: Alligator Descartes Message-Id: <199511022200.WAA09605@fruitbat.mcqueen.com> Subject: Re: DBD::Qbase 0.03 To: George Johnston Date: Thu, 2 Nov 1995 22:00:29 +0000 (GMT) Cc: dbi-dev@fugue.com In-Reply-To: <199511020202.SAA20366@sunspot.filoli.com> from "George Johnston" at Nov 1, 95 06:02:43 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 956 > Because QuickBase is Objective-C then perlmain.c needs to be = > compiled with the -ObjC option. I know for my users of = > perl/QuickBase it was much more important for them to be able to = > use QuickBase than to have it dynamically loaded thus I spent no = > cycles on the dynamic loading problem especially since the Obj-C = > runtime (I believe) needs to be linked in statically. Yes, but if you specify: make LINKTYPE=static on building just the DBD::QBase module, surely that would allow you to load the other modules dynamically, no? Maybe not..... 8-) > Thanks for the code review and I know I will have more questions. Sure thing. I'm usually fairly helpful.....8-) I'm going to tidy up my "How-to-write-a-driver" document in the next day or so and make that available on Hermetica's WWW, so take a look at that.... > George -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <17121-49@oink>; Fri, 3 Nov 1995 07:11:36 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 03 Nov 95 00:11:41 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa16732; 3 Nov 95 0:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA12329 for dbi-dev-real; Thu, 2 Nov 1995 14:34:25 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netnet1.netnet.net (netnet1.netnet.net [198.70.64.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA12325 for ; Thu, 2 Nov 1995 14:34:21 -0800 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id QAA12851; Thu, 2 Nov 1995 16:36:31 -0600 Date: Thu, 2 Nov 1995 16:36:31 -0600 (CST) From: Ben Lindstrom To: Alligator Descartes cc: George Johnston , dbi-dev@fugue.com Subject: Re: DBD::Qbase 0.03 In-Reply-To: <199511022200.WAA09605@fruitbat.mcqueen.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 2 Nov 1995, Alligator Descartes wrote: > > > Because QuickBase is Objective-C then perlmain.c needs to be = > > compiled with the -ObjC option. I know for my users of = > > perl/QuickBase it was much more important for them to be able to = > > use QuickBase than to have it dynamically loaded thus I spent no = > > cycles on the dynamic loading problem especially since the Obj-C = > > runtime (I believe) needs to be linked in statically. > > Yes, but if you specify: > > make LINKTYPE=static > > on building just the DBD::QBase module, surely that would allow you to load > the other modules dynamically, no? Maybe not..... > I'll look into it this weekend when I have my next block of time for working on the code is. > > Thanks for the code review and I know I will have more questions. > > Sure thing. I'm usually fairly helpful.....8-) > > I'm going to tidy up my "How-to-write-a-driver" document in the next day or > so and make that available on Hermetica's WWW, so take a look at that.... > That was something I was thinking about as I was learning..."Gee...it would be nice to write down what I'm doing and later go back and put it in some decent order." Oh well..I'll leave it for the more experienced hackers. I'm just a poor college student learning to be a Network Admin. =) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21809-20@oink>; Fri, 3 Nov 1995 17:03:46 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 03 Nov 95 13:11:29 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa21120; 3 Nov 95 13:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA15980 for dbi-announce-real; Fri, 3 Nov 1995 03:05:48 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id DAA15975 for ; Fri, 3 Nov 1995 03:05:45 -0800 Message-Id: <199511031105.DAA15975@toccata.fugue.com> To: dbi-announce@fugue.com Subject: New DBI Mailing List administration interface... Date: Fri, 03 Nov 1995 03:05:44 -0800 From: Ted Lemon Power To The People I'm tired of hacking on my mailing list database, so I sat down over the course of the past two days and hacked together a WorldWide Web CGI Script that can be used to subscribe and unsubscribe from the DBI mailing lists. If you don't have WWW access, you can still send administrivia requests to dbi-request@fugue.com, but if you do have WWW access, I'd appreciate it if you'd use the Web page instead. The URL is http://www.fugue.com/dbi/. If you have any comments on how the thing works, please let me know. I'm feeling kind of smug about it right now, so somebody should pop my balloon ASAP. :') BTW, it uses forms, so it looks really ugly if you're running Lynx. When I feel ambitious, I'll conditionalize that, but for now, you're kind of stuck with it. Sorry about that. You can still use it from Lynx if you're clever and patient. _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <24602-3@oink>; Fri, 3 Nov 1995 21:04:20 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Fri, 03 Nov 95 19:55:49 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa26853; 3 Nov 95 19:53 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA16864 for dbi-users-real; Fri, 3 Nov 1995 09:55:17 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fns.com (ops1-fddi.fns.com [168.127.10.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA16860 for ; Fri, 3 Nov 1995 09:55:15 -0800 From: Scott=Alexander%se%FNS@fnsvines1.fns.com Received: from fnsvines1.fns.com by fns.com (5.0/SMI-SVR4) id AA28800; Fri, 3 Nov 1995 12:58:02 +0500 Message-Id: <9511031758.AA28800@fns.com> Received: by fnsvines1.fns.com with VINES ; Fri, 3 Nov 95 13:00:22 EST Date: Fri, 3 Nov 95 13:00:08 EST Subject: How to trap database error messages? To: dbi-users@fugue.com content-length: 268 Can someone provide me with example code on how to trap database error messages. I'm trying to trap a unique constraint message when inserting records into an Oracle db. It seems by default to output the message to the display. Thanks, Scott Alexander sga@fns.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01930-75@oink>; Mon, 6 Nov 1995 07:17:28 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 05 Nov 95 13:19:15 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa21652; 5 Nov 95 13:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA22401 for dbi-announce-real; Sun, 5 Nov 1995 04:20:09 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA22380; Sun, 5 Nov 1995 04:13:30 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id MAA20599; Sun, 5 Nov 1995 12:11:07 GMT From: Alligator Descartes Message-Id: <199511051211.MAA20599@fruitbat.mcqueen.com> Subject: DBD::mSQL-0.60pl7 released To: dbi-dev@fugue.com, dbi-announce@fugue.com, dbi-users@fugue.com, dbi-users@fugue.com, msql-list@bunyip.com, k@anna.mind.de, Sven Verdoolaege Date: Sun, 5 Nov 1995 12:11:05 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 526 The new release of DBD::mSQL ( 0.60pl7 ) has been released. This release is yet another developers release since I think Tim's planning to update the DBI code again. Thus, I'm holding off going to 0.61 until: a) The new driver works b) The new DBI stuff is out Anyway, some stuff has changed, and there's more features now, so go get! Further information at: http://www.hermetica.com/technologia/DBI Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01930-76@oink>; Mon, 6 Nov 1995 07:17:33 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 05 Nov 95 13:19:17 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa21704; 5 Nov 95 13:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA22384 for perldb-interest-real; Sun, 5 Nov 1995 04:13:34 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA22380; Sun, 5 Nov 1995 04:13:30 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id MAA20599; Sun, 5 Nov 1995 12:11:07 GMT From: Alligator Descartes Message-Id: <199511051211.MAA20599@fruitbat.mcqueen.com> Subject: DBD::mSQL-0.60pl7 released To: dbi-dev@fugue.com, dbi-announce@fugue.com, dbi-users@fugue.com, dbi-users@fugue.com, msql-list@bunyip.com, k@anna.mind.de, Sven Verdoolaege Date: Sun, 5 Nov 1995 12:11:05 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 526 The new release of DBD::mSQL ( 0.60pl7 ) has been released. This release is yet another developers release since I think Tim's planning to update the DBI code again. Thus, I'm holding off going to 0.61 until: a) The new driver works b) The new DBI stuff is out Anyway, some stuff has changed, and there's more features now, so go get! Further information at: http://www.hermetica.com/technologia/DBI Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01930-78@oink>; Mon, 6 Nov 1995 07:17:44 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 05 Nov 95 13:19:18 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id ab21766; 5 Nov 95 13:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA22406 for dbi-dev-real; Sun, 5 Nov 1995 04:21:33 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA22380; Sun, 5 Nov 1995 04:13:30 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id MAA20599; Sun, 5 Nov 1995 12:11:07 GMT From: Alligator Descartes Message-Id: <199511051211.MAA20599@fruitbat.mcqueen.com> Subject: DBD::mSQL-0.60pl7 released To: dbi-dev@fugue.com, dbi-announce@fugue.com, dbi-users@fugue.com, dbi-users@fugue.com, msql-list@bunyip.com, k@anna.mind.de, Sven Verdoolaege Date: Sun, 5 Nov 1995 12:11:05 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 526 The new release of DBD::mSQL ( 0.60pl7 ) has been released. This release is yet another developers release since I think Tim's planning to update the DBI code again. Thus, I'm holding off going to 0.61 until: a) The new driver works b) The new DBI stuff is out Anyway, some stuff has changed, and there's more features now, so go get! Further information at: http://www.hermetica.com/technologia/DBI Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01930-80@oink>; Mon, 6 Nov 1995 07:17:56 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Sun, 05 Nov 95 13:22:41 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa22975; 5 Nov 95 13:22 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA22391 for dbi-users-real; Sun, 5 Nov 1995 04:18:29 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA22380; Sun, 5 Nov 1995 04:13:30 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id MAA20599; Sun, 5 Nov 1995 12:11:07 GMT From: Alligator Descartes Message-Id: <199511051211.MAA20599@fruitbat.mcqueen.com> Subject: DBD::mSQL-0.60pl7 released To: dbi-dev@fugue.com, dbi-announce@fugue.com, dbi-users@fugue.com, dbi-users@fugue.com, msql-list@bunyip.com, k@anna.mind.de, Sven Verdoolaege Date: Sun, 5 Nov 1995 12:11:05 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 526 The new release of DBD::mSQL ( 0.60pl7 ) has been released. This release is yet another developers release since I think Tim's planning to update the DBI code again. Thus, I'm holding off going to 0.61 until: a) The new driver works b) The new DBI stuff is out Anyway, some stuff has changed, and there's more features now, so go get! Further information at: http://www.hermetica.com/technologia/DBI Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <18624-1@oink>; Wed, 8 Nov 1995 02:16:38 +0000 Received: from punt-2.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 08 Nov 95 00:26:34 GMT Received: from relay-4.mail.demon.net by punt-2.mail.demon.net id aa03237; 8 Nov 95 0:25 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA00920 for dbi-dev-real; Tue, 7 Nov 1995 14:27:45 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from wpgate.hqpacaf.af.mil (WPGATE.hqpacaf.af.mil [131.38.122.206]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id OAA00912 for ; Tue, 7 Nov 1995 14:27:40 -0800 Received: from HQPACAF-Message_Server by wpgate.hqpacaf.af.mil with Novell_GroupWise; Tue, 07 Nov 1995 12:12:19 -1000 Message-Id: X-Mailer: Novell GroupWise 4.1 Date: Tue, 07 Nov 1995 12:09:52 -1000 From: Damon LaCaille To: dbi-dev@fugue.com Subject: unsubscribe Damon L. LaCaille lacailld@wpgate.hqpacaf.af.mil unsubscribe Damon L. LaCaille lacailld@wpgate.hqpacaf.af.mil ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <18625-8@oink>; Wed, 8 Nov 1995 02:17:07 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 08 Nov 95 00:20:50 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa21241; 8 Nov 95 0:19 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa01336; 8 Nov 95 0:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA00138 for dbi-dev-real; Tue, 7 Nov 1995 09:26:34 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA00123 for ; Tue, 7 Nov 1995 09:26:17 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ae21082; 7 Nov 95 17:01 GMT Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id msg.ag00212; 7 Nov 95 17:01 GMT Received: from toad by oink with SMTP (PP) id <14967-0@oink>; Tue, 7 Nov 1995 16:23:59 +0000 Received: by toad (5.0/SMI-SVR4) id AA08604; Tue, 7 Nov 1995 16:23:36 +0000 Date: Tue, 7 Nov 1995 16:23:36 +0000 From: Tim Bunce Message-Id: <9511071623.AA08604@toad> To: dbi-dev@fugue.com Subject: Re: DBD::Qbase 0.03 X-Sun-Charset: US-ASCII content-length: 162 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Ben Lindstrom > > I'm not sure if one can staticly compile the DBD and dynamicly > compile the DBI. If you can't it's a bug. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <18625-10@oink>; Wed, 8 Nov 1995 02:17:16 +0000 Received: from punt-1.mail.demon.net via puntmail for dbperl@ignite.demon.co.uk; Wed, 08 Nov 95 00:42:42 GMT Received: from relay-4.mail.demon.net by punt-1.mail.demon.net id aa29727; 8 Nov 95 0:41 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA00820 for dbi-users-real; Tue, 7 Nov 1995 14:01:04 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA00816 for ; Tue, 7 Nov 1995 14:01:02 -0800 Received: by gw.home.vix.com id AA06325; Tue, 7 Nov 95 14:01:44 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ai14005; 7 Nov 95 21:55 GMT Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id sg.ai06695; 7 Nov 95 21:55 GMT Received: from toad by oink with SMTP (PP) id <17258-0@oink>; Tue, 7 Nov 1995 21:48:53 +0000 Received: by toad (5.0/SMI-SVR4) id AA09164; Tue, 7 Nov 1995 21:48:30 +0000 Date: Tue, 7 Nov 1995 21:48:30 +0000 From: Tim Bunce Message-Id: <9511072148.AA09164@toad> To: kerce@scri.fsu.edu Subject: Re: Well done! Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII Content-Length: 1515 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Kingsley Kerce > > If you're getting sick of bug reports, here's a success report. Yippe! > Platform: > AIX 3.2.5 on RS/6000 Model 375 > Oracle7 Server Release 7.1.3.0.0 - Production Release > With the distributed option > > I picked up DBD::Oracle 0.25 and DBI 0.63 via the following URL: > http://www.hermetica.com/technologia/DBI/DBD/index.html > > I had installed perl 5.001m a few weeks back, so I followed your > READMEs in DBI and Oracle, testing and installing with no problems at > all. > > Next, the big test, as far as I'm concerned: porting a Web-database > gateway based on oraperl (perl 4.036) to DBD::Oracle. I'd coded such > a gateway a month or so back and it's been running fine since. Well, > I simply changed the CGI scripts' #! paths to point to the perl5.001m > installation (instead of oraperl/perl 4.036) and inserted the > following lines in a commonly "require"d file: > > $ENV{ORACLE_HOME} = '/usr/oracle' unless $ENV{ORACLE_HOME}; > > eval 'use Oraperl; 1' || die $@ if $] >= 5; > > and IT WORKED ON FIRST TRY!!! > > Congratulations, > Kingsley Kerce Systems Development and Administration > Supercomputer Computations Research Institute kerce@scri.fsu.edu > Florida State University http://www.scri.fsu.edu/~kerce/ > Tallahassee, FL, USA 32306 Obtain my PGP public key via finger/WWW The next one will fix more of the bugs you've not yet tripped up on! :-) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01286-29@oink>; Thu, 9 Nov 1995 17:01:04 +0000 Received: from punt-1.mail.demon.net with mailstore id 815931629:08951:0; Thu, 09 Nov 95 15:40:29 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa08882; 9 Nov 95 15:40 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa05519; 9 Nov 95 15:39 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA05421 for dbi-users-real; Thu, 9 Nov 1995 02:09:46 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA05417 for ; Thu, 9 Nov 1995 02:09:40 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id KAA17548; Thu, 9 Nov 1995 10:07:04 GMT From: Alligator Descartes Message-Id: <199511091007.KAA17548@fruitbat.mcqueen.com> Subject: Re: DBD::mSQL To: Randy Terbush Date: Thu, 9 Nov 1995 10:07:03 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: <199511090357.VAA29787@sierra.zyzzyva.com> from "Randy Terbush" at Nov 8, 95 09:57:59 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 7959 [ Please note! Comments most welcome! I'll incorporate this stuff into a WWW page soon! -- A ] > I don't suppose you have a snippet of code to show how to do a > submit with mSQL and DBD? I'm going to assume you mean through a WWW form? Anyway, it's much of a muchness. Try this ( off the top of my head! ): -----------------8<----------------------------8<---------------------- #!/usr/bin/perl -w # use DBI; # Load the DBI Switch $drh = DBI->install_driver( 'mSQL' ); # Load the mSQL driver. Returns a driver # - handle in $drh $dbh = $drh->connect( '', 'rubbish' ); # Connects via the mSQL driver ( as designated # - by the driver handle $drh ) to a database # - on the local machine called 'rubbish'. die "Cannot connect: $DBI::errstr\n" unless $dbh; # Check we've connected. Die if not. &ReadParse; # A routine to parse a URL. See CGI.pm foreach $arg ( keys %in ) { # The args are returned in a hash %in print "$arg: $in{$arg}

\n"; } # Assume that we've submitted a form with the following fields: # # o Name # o Email address # # Assume we have a table in the 'rubbish' database called 'cgistuff' with the # - following fields: # # o name CHAR(64) # o email CHAR(128) $cursor = $dbh->do( " # Insert a row of data into the table INSERT INTO cgistuff VALUES # - via the database handle returned ( $in{'name'}, $in{'email'} )" ); # - above # A $dbh->do doesn't require an execute or finish phase. It just happens # - atomically. $dbh->disconnect; # Disconnect the database handle. This actually # - disconnects you from the database specified # - by $dbh exit 0; --------------------------8<--------------------------8<--------------------- > I am confused by the code. Where do I find a complete list of the > functions that are available to me. The 'test.pl' script in 0.60pl7 exercises all the functions in the driver. A basic problem to using drivers is understanding the architecture, which I'll take a moment to try and explain: $drh--------------------- Driver handles | | $dbh---$dbh---$dbh---$dbh Database handles | | | | $sth---$sth--+ $sth-+ $sth-+ $sth-+-$sth Statement handles ( Apologies for the bad drawing! ). What is says though is: o Before you do operations with a certain type of database you load the database driver, eg, if you want to use mSQL you call: $drh = DBI->install_driver( 'mSQL' ); If we thern wished to load the Oracle driver, another line: $oradrh = DBI->install_driver( 'Oracle' ); Notice they are assigned to different variables. The driver for each database *type* ( Oracle, mSQL ) is loaded only ONCE per script. Hence the one $drh above. All operations on that type of database are invoked indirectly through the $drh ( or DRIVER HANDLE ). o Now that we've got a driver loaded, we can start doing things with databases of that type, eg, connect! Obviously you have to connect to a specific database before you can start doing stuff, so these functions are called against the DRIVER HANDLE, eg: $dbh = $drh->connect( '', 'rubbish' ); This connects you to an mSQL database ( as defined by the driver handle you use to connect ), identified by the arguments passed to the 'connect' call. If successfull, it returns a variable $dbh ( DATABASE HANDLE ). The differences between a DRIVER HANDLE and a DATABASE HANDLE is that the driver handle is independent of actual databases. It just tells the DBI Switch which set of lower-level database routines to call, eg, the Switch thinks "I've just had a connect request with these arguments, and it's via the $drh driver handle. Now, the $drh driver handle is an mSQL one, so we'll call the mSQL connect primitive to do the actual work." Previously, we only had *one* driver handle per database type. Database handles are much more plentiful. You can have as many database handles as you need within one driver handle, eg: $dbh1 = $drh->connect( '', 'rubbish1' ); $dbh2 = $drh->connect( '', 'rubbish2' ); $dbh3 = $drh->connect( 'someotherhost', 'pants1' ); $dbh4 = $drh->connect( 'someotherhost2', 'pants2' ); Connects us simultaneously to 4 different databases ( all mSQL, using the *one* driver handle ). These database are located on 3 different physical machines. This can be done within one script. o Statement handles are another level down. Think of the whole structure as a tree. You can one DRIVER HANDLE, but many DATABASE HANDLES from that, and many STATEMENT HANDLES from *each* DATABASE HANDLE. Yes? Good! Now, a statement handle is a handle created on operations within the driver which act on statements! No, really! A statement is an operation on actual data within tables ( usually ), eg, $sth = $dbh1->do( "DELETE FROM table1" ); This executes the statement "DELETE FROM table1" in the database indicated previously by $dbh1 ( from above, the one on 'localhost' called 'rubbish1'. We could now do: $sth = $dbh4->do( "UPDATE cgistuff SET email = ''" ); In our CGI example, which would clear all the email fields. As in the relationship between DATABASE HANDLES and DRIVER HANDLES of the many to one, it exists between STATEMENT HANDLES and DATABASE HANDLES. Another aspect of STATEMENT HANDLES is the 'cursor', eg: $sth = $dbh1->prepare( "SELECT id, name, email FROM madeuptable" ); $sth->execute; while ( @row = $sth->fetchrow ) { print "Row: @row\n"; } $sth->finish; This chunk of code prepares a STATEMENT HANDLE $sth with regard to the statement "SELECT .....". If the database allows you to prepare it ( $sth is returned defined and not undef ), ( a successfull prepare implies the field name are correct, the database is connectable still and the table exists ), then you can continue working with that statement. The second part of using a cursor is to EXECUTE the statement. This sends it to the database engine to return the rows into a cache. At this point, you specifiy bind variables ( if any ), but these don't exist in DBD::mSQL, so don't worry! After a successful execution, the data is ready to fetch from the database. We can now execute a 'while' loop calling 'fetchrow' against the STATEMENT HANDLE. If the fetchrow returns 'undef', the loop will terminate. The 'fetchrow' will return undef when there's no more data to be fetched. A LIST is returned from fetchrow. In the example above, we're fetching all the fields into a LIST, but you could have done: while ( ( $id, $name, $email ) = $sth->fetchrow ) and fetched the fields directly into separate variables. The final call is: 'finish' which closes a cursor and is called usually after all the rows have been fetched. This frees up the memory used by the cursor. > Thanks for your help Hope this helps! Again, with this information in mind, re-read the 'test.pl' script with DBD::mSQL-0.60pl7. It'll show which functions are callable against which sort of HANDLE. Although here's another off the top of my head list. I may have missed one or two..... $drh is a DRIVER HANDLE, $dbh a DATABASE HANDLE and $sth a STATEMENT HANDLE. @ary is a LIST, $scalar is an ordinary scalar, %hash is a hash. $rv is a return value which is ignorable. (*) means non-standard and mSQL specific Invoking through Returns ---------------- ------- DBI->install_driver $drh $drh->connect $dbh $drh->func( '_ListDBs' ) @ary * $drh->disconnect_all ? $dbh->func( '_ListTables' ) @ary * $dbh->disconnect $rv $dbh->do $rv $dbh->prepare $sth $sth->execute $rv $sth->fetchrow @ary $sth->finish $rv Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05363-21@oink>; Fri, 10 Nov 1995 07:09:29 +0000 Received: from punt-1.mail.demon.net with mailstore id 815948860:08356:0; Thu, 09 Nov 95 20:27:40 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa08349; 9 Nov 95 20:27 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa25024; 9 Nov 95 20:26 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA06089 for dbi-users-real; Thu, 9 Nov 1995 09:58:24 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA06085 for ; Thu, 9 Nov 1995 09:58:19 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab08926; 9 Nov 95 17:59 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa29508; 9 Nov 95 17:59 GMT Received: from toad by oink with SMTP (PP) id <02884-0@oink>; Thu, 9 Nov 1995 17:54:44 +0000 Received: by toad (5.0/SMI-SVR4) id AA12673; Thu, 9 Nov 1995 17:54:22 +0000 Date: Thu, 9 Nov 1995 17:54:22 +0000 From: Tim Bunce Message-Id: <9511091754.AA12673@toad> To: randy@zyzzyva.com, descarte@hermetica.com Subject: Re: DBD::mSQL Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 3746 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Alligator Descartes > > [ Please note! Comments most welcome! I'll incorporate this stuff into a > WWW page soon! -- A ] > > > I don't suppose you have a snippet of code to show how to do a > > submit with mSQL and DBD? > > I'm going to assume you mean through a WWW form? Anyway, it's much of a > muchness. Try this ( off the top of my head! ): > > -----------------8<----------------------------8<---------------------- > #!/usr/bin/perl -w > # > > use DBI; # Load the DBI Switch > $drh = DBI->install_driver( 'mSQL' ); > # Load the mSQL driver. Returns a driver > # - handle in $drh > $dbh = $drh->connect( '', 'rubbish' ); > # Connects via the mSQL driver ( as designated > # - by the driver handle $drh ) to a database > # - on the local machine called 'rubbish'. > die "Cannot connect: $DBI::errstr\n" > unless $dbh; > # Check we've connected. Die if not. > > &ReadParse; # A routine to parse a URL. See CGI.pm General CGI issues: Ideally you should send the Content-type header _first_ and redirect STDERR to STDOUT (and/or use Tom's CGI::ErrorWrap module). That way any error with use/install/connect (which are _bound_ to happen) will still generate some output for the user. > $cursor = $dbh->do( " # Insert a row of data into the table > INSERT INTO cgistuff VALUES # - via the database handle returned > ( $in{'name'}, $in{'email'} )" ); # - above *Always* test the return status. It's almost certain that the default error handler in the DBI will *NOT* output anything in future releases. > o Before you do operations with a certain type of database you load > the database driver, eg, if you want to use mSQL you call: > > $drh = DBI->install_driver( 'mSQL' ); I strongly recommend that you *don't* explicitly load drivers. It's much better to use the Driver parameter to the DBI->connect() method. > > The differences between a DRIVER HANDLE and a DATABASE HANDLE is > that the driver handle is independent of actual databases. It just > tells the DBI Switch which set of lower-level database routines to > call, eg, the Switch thinks "I've just had a connect request with > these arguments, and it's via the $drh driver handle. Now, the $drh > driver handle is an mSQL one, so we'll call the mSQL connect primitive > to do the actual work." Think Methods and Objects. > o Statement handles are another level down. Think of the whole > structure as a tree. You can one DRIVER HANDLE, but many DATABASE > HANDLES from that, and many STATEMENT HANDLES from *each* DATABASE > HANDLE. Yes? > > Good! Now, a statement handle is a handle created on operations within > the driver which act on statements! No, really! Not quite. A statement handle is a handle representing a 'prepared statement'. > A statement is an > operation on actual data within tables ( usually ), eg, Nope, it could be an 'alter session set foo=bar'. Think Generality. I'd drop the sentance or put the usually upfront without the (). > $sth = $dbh1->do( "DELETE FROM table1" ); Ooops. The do method does NOT return a handle. It 'does' the statement and returns a status. > The second part of using a cursor is to EXECUTE the statement. This > sends it to the database engine to return the rows into a cache. The engine may already have it (eg Oracle). Say "asks the engine to execute the prepared statement". Adding || die "..." to everything would set a good example. > Hope this helps! A *big* thankyou for doing all this Alligator! It really needs to be done. > $drh->func( '_ListDBs' ) @ary * > $dbh->func( '_ListTables' ) @ary * These may have official methods very soon. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <05362-22@oink>; Fri, 10 Nov 1995 07:09:30 +0000 Received: from punt-2.mail.demon.net with mailstore id 815960016:24260:1; Thu, 09 Nov 95 23:33:36 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa22940; 9 Nov 95 23:33 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA06457 for dbi-users-real; Thu, 9 Nov 1995 13:13:16 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from srs.gov (bubba.srs.gov [192.33.240.5]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA06453 for ; Thu, 9 Nov 1995 13:12:23 -0800 From: james.taylor@srs.gov Received: by srs.gov id AA08214 (InterLock SMTP Gateway 1.1 for dbi-users@fugue.com); Thu, 9 Nov 1995 16:12:02 -0500 Received: by srs.gov (Internal Mail Agent-1); Thu, 9 Nov 1995 16:12:02 -0500 Date: Thu, 9 Nov 1995 16:11:56 -0500 Message-Id: <199511092111.QAA04943@dark-knight.srs.gov> To: Tim.Bunce@ig.co.uk Subject: DBD::Oracle Cc: dbi-users@fugue.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Content-Md5: pbzhVhIGjV9/TJgmd/61/A== Tim, I no longer need to make any changes to get DBD::Oracle to compile with = Solaris=20 2.4 SPARCompiler C. You might want to remove the note from the README = file. Thanks, James ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05363-32@oink>; Fri, 10 Nov 1995 07:11:15 +0000 Received: from punt-1.mail.demon.net with mailstore id 815951457:00223:0; Thu, 09 Nov 95 21:10:57 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ab00107; 9 Nov 95 21:10 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <10555-0@frigate.doc.ic.ac.uk>; Tue, 7 Nov 1995 23:49:25 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA00439 for dbi-dev-real; Tue, 7 Nov 1995 12:53:19 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA00435 for ; Tue, 7 Nov 1995 12:53:16 -0800 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id PAA28099; Tue, 7 Nov 1995 15:55:28 -0500 Message-Id: <199511072055.PAA28099@pilot.njin.net> To: dbi-dev@fugue.com cc: phj@pilot.njin.net Subject: sqlNet login Date: Tue, 07 Nov 1995 15:55:25 -0500 From: "Patricia H. Ju" Hi all, Sorry if this is inappropriate (I can't remember what is and what isn't) or can be found elsewhere (tell me where!) but I have no idea how to loginto another host's oracle server via Oracle DBI. The connect string should probably be: sqlplus mylogin/mypassword@T:the.server.com:database ? but I'm not even sure about that, as I can't get this working at the command line, even. Help ? Patricia phj@pilot.njin.net ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05363-89@oink>; Fri, 10 Nov 1995 07:20:10 +0000 Received: from punt-1.mail.demon.net with mailstore id 815977100:07796:1; Fri, 10 Nov 95 04:18:20 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa07780; 10 Nov 95 4:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA07091 for dbi-users-real; Thu, 9 Nov 1995 18:49:18 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sierra.zyzzyva.com (root@ppp0.zyzzyva.com [198.183.2.50]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA07087 for ; Thu, 9 Nov 1995 18:49:13 -0800 Received: from zyzzyva.com (randy@localhost.zyzzyva.com [127.0.0.1]) by sierra.zyzzyva.com (8.6.12/8.6.11) with ESMTP id UAA17450; Thu, 9 Nov 1995 20:48:17 -0600 Message-Id: <199511100248.UAA17450@sierra.zyzzyva.com> To: Tim Bunce Subject: Re: DBD::mSQL In-reply-to: Tim.Bunce's message of Thu, 09 Nov 1995 17:54:22 +0000. <9511091754.AA12673@toad> Cc: dbi-users@fugue.com X-uri: http://www.zyzzyva.com/ Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 09 Nov 1995 20:48:09 -0600 From: Randy Terbush > > > o Before you do operations with a certain type of database you load > > the database driver, eg, if you want to use mSQL you call: > > > > $drh = DBI->install_driver( 'mSQL' ); > > I strongly recommend that you *don't* explicitly load drivers. > > It's much better to use the Driver parameter to the DBI->connect() method. Could you give an example of the syntax for this? I cannot seem to get it to work from the comments in the code. > A *big* thankyou for doing all this Alligator! It really needs to be done. A strong second. This is very helpful. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05636-47@oink>; Fri, 10 Nov 1995 07:20:11 +0000 Received: from punt-1.mail.demon.net with mailstore id 815977211:08217:1; Fri, 10 Nov 95 04:20:11 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa08174; 10 Nov 95 4:20 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA07070 for perldb-interest-real; Thu, 9 Nov 1995 18:45:38 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA07066 for ; Thu, 9 Nov 1995 18:45:29 -0800 Received: by gw.home.vix.com id AA13833; Thu, 9 Nov 95 18:47:45 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from cthulhu.engr.sgi.com by sgi.sgi.com via ESMTP (950405.SGI.8.6.12/910110.SGI) id SAA27242; Thu, 9 Nov 1995 18:47:38 -0800 Received: from whitetiger.engr.sgi.com by cthulhu.engr.sgi.com via ESMTP (950511.SGI.8.6.12.PATCH526/911001.SGI) id RAA14383; Thu, 9 Nov 1995 17:29:22 -0800 Received: by whitetiger.engr.sgi.com (940816.SGI.8.6.9/940406.SGI.AUTO) id RAA05844; Thu, 9 Nov 1995 17:29:18 -0800 From: Todd Hoff Message-Id: <9511091729.ZM5842@whitetiger.engr.sgi.com> Date: Thu, 9 Nov 1995 17:29:16 -0800 X-Mailer: Z-Mail (3.2.0 26oct94 MediaMail) To: dbi-users@vix.com Subject: Oracle Login Problems Cc: Tim.Bunce@ig.co.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Congratulations on such an easy to install system, it is quite nice. The only problem is i can't get it to work :-) When i try to do a login i get this message: Oraperl Emulation Interface version 1.21 Oracle Driver 1.20 DBI Switch by Tim Bunce, version 1.45 ERROR EVENT: DBI::db=HASH(0x100a75dc)->DBI::default_handler: 1034 ORA-01034: ORACLE not available (DBD: login failed) at /usr/local/lib/perl5/DBD/Oracle.pm line 99 The code looks like: #!/bin/perl use Oraperl; ora_version(); ora_login("i18n", "sgi", "cool"); The same call in per4 oraperl works. I've tried lots of combinations and asked some people. Do you have any idea why it might not be working? Log: whitetiger 84% uname -a IRIX whitetiger 5.3 11091812 IP22 mips Oracle Version: SQL*Plus: Release 3.1.3.4.1 - Production on Thu Nov 9 16:51:17 1995 Copyright (c) Oracle Corporation 1979, 1994. All rights reserved. Connected to: Oracle7 Server Release 7.1.4.1.0 - Production Release With the distributed and parallel query options PL/SQL Release 2.1.4.0.0 - Production hitetiger 149% perl !$ perl Makefile.PL Configuring DBD::Oracle version 0.25... Remember to actually read the README file! Using Oracle in /hosts/furnace/db/oracle7.1.3 Oracle sysliblist: -lc -lm Using /hosts/furnace/db/oracle7.1.3/proc16/lib/proc16.mk version 1.6 System: perl5.001 irix hoshi 5.3 12200159 ip22 mips irix dl_dlopen.xs Compiler: cc -O -D_BSD_TYPES -D_BSD_SIGNALS -Olimit 3000 Oracle proc.mk would have used these values but we override them: CC: cc CFLAGS: -I. -O `if [ -d /usr/lib/cmplrs ] ; then echo "-systype svr4 -KPIC" ; fi ` LDFLAGS: -L$(LIBHOME) `if [ -x /usr/necbin/uversion ] ; then echo "-Zcommon_pic -znodefs" ; elif [ -d /usr/lib/cmplrs ] ; then echo "-systype svr4" ; fi ` Writing Makefile for DBD::Oracle whitetiger 150% make mkdir ./blib mkdir ./blib/DBD mkdir ./blib/auto mkdir ./blib/auto/DBD mkdir ./blib/auto/DBD/Oracle /bin/perl /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c cc -c -D_BSD_TYPES -D_BSD_SIGNALS -Olimit 3000 -O -I/usr/local/lib/perl5/irix/CORE -I/hosts/furnace/db/oracle7.1.3/rdbms/demo -I/usr/local/lib/perl5/irix/DBI Oracle.c cc -c -D_BSD_TYPES -D_BSD_SIGNALS -Olimit 3000 -O -I/usr/local/lib/perl5/irix/CORE -I/hosts/furnace/db/oracle7.1.3/rdbms/demo -I/usr/local/lib/perl5/irix/DBI dbdimp.c Running mkbootstrap for DBD::Oracle () ld -o ./blib/auto/DBD/Oracle/Oracle.so -shared Oracle.o dbdimp.o -L/hosts/furnace/db/oracle7.1.3/lib -locic /hosts/furnace/db/oracle7.1.3/lib/osntab.o -lsqlnet -lora -lsqlnet -lora -lnlsrtl -lcv6 -lcore -lnlsrtl -lcv6 -lcore `cat /hosts/furnace/db/oracle7.1.3/rdbms/lib/sysliblist` -lc -lm cp Oracle.bs ./blib/auto/DBD/Oracle/Oracle.bs cp Oraperl.pm ./blib/Oraperl.pm cp Oracle.pm ./blib/DBD/Oracle.pm whitetiger 151% make test /bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/irix -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/base..............ok t/main..............ok All tests successful. Files=2, Tests=6, 2 secs ( 0.60 cusr 0.15 csys = 0.75 cpu) /bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/irix -I/usr/local/lib/perl5 test.pl Oraperl test application $Revision: 1.17 $ Oraperl Emulation Interface version 1.21 Oracle Driver 1.20 DBI Switch by Tim Bunce, version 1.45 Connecting to '' (from command line, else uses ORACLE_SID/TWO_TASK as 'crgs/crgs' (via ORACLE_USERID environment var or default) ERROR EVENT: DBI::db=HASH(0x100b1068)->DBI::default_handler: 1017 ORA-01017: invalid username/password; logon denied (DBD: login failed) at blib/DBD/Oracle.pm line 99 ora_login: 1017: ORA-01017: invalid username/password; logon denied (DBD: login failed) *** Error code 255 (bu21) whitetiger 152% make install : perl5.000 and MM pre 3.8 autosplit into INST_ARCHLIB, we delete these old files here rm -f /usr/local/lib/perl5/irix/auto/DBD/Oracle/*.al rm -f /usr/local/lib/perl5/irix/auto/DBD/Oracle/*.ix make INST_LIB=/usr/local/lib/perl5 INST_ARCHLIB=/usr/local/lib/perl5/irix INST_EXE=/usr/local/bin ld -o /usr/local/lib/perl5/irix/auto/DBD/Oracle/Oracle.so -shared Oracle.o dbdimp.o -L/hosts/furnace/db/oracle7.1.3/lib -locic /hosts/furnace/db/oracle7.1.3/lib/osntab.o -lsqlnet -lora -lsqlnet -lora -lnlsrtl -lcv6 -lcore -lnlsrtl -lcv6 -lcore `cat /hosts/furnace/db/oracle7.1.3/rdbms/lib/sysliblist` -lc -lm Oraperl Emulation Interface version 1.21 Oracle Driver 1.20 DBI Switch by Tim Bunce, version 1.45 ERROR EVENT: DBI::db=HASH(0x100a75dc)->DBI::default_handler: 1034 ORA-01034: ORACLE not available (DBD: login failed) at /usr/local/lib/perl5/DBD/Oracle.pm line 99 whitetiger 128% perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=irix, osver=5, archname=irix uname='irix hoshi 5.3 12200159 ip22 mips ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-D_BSD_TYPES -D_BSD_SIGNALS' ccflags ='-D_BSD_TYPES -D_BSD_SIGNALS -Olimit 3000' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=1, d_casti32=define, d_castneg= intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/usr/lib /usr/local/lib /lib libs=-lcrypt -lm libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags=' ', ccdlflags=' ', lddlflags='-shared' ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <12121-1@oink>; Fri, 10 Nov 1995 15:36:25 +0000 Received: from punt-2.mail.demon.net with mailstore id 816010022:10119:0; Fri, 10 Nov 95 13:27:02 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id aa10112; 10 Nov 95 13:27 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <05805-1@frigate.doc.ic.ac.uk>; Thu, 9 Nov 1995 02:58:03 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA03227 for dbi-dev-real; Wed, 8 Nov 1995 16:00:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netnet1.netnet.net (netnet1.netnet.net [198.70.64.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA03223 for ; Wed, 8 Nov 1995 16:00:08 -0800 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id SAA22181; Wed, 8 Nov 1995 18:02:28 -0600 Date: Wed, 8 Nov 1995 18:02:28 -0600 (CST) From: Ben Lindstrom To: dbi-dev@fugue.com Subject: dbh to sth data migration In-Reply-To: <199511012225.WAA29379@fruitbat.mcqueen.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII As I'm looking around I'm trying to figure out how exactly perl/DBI knows what in the imp_dbh_st should be copied into imp_sth_st on a "D_imp_dbh_from_sth". I tracked it down to a #define that says it's really D_imp_from_child() But from there..I can't seem to find any references to that function. BTW, as I expected. It's always good to start simple and move to the harder stuff as you learn.=) --- Ben Lindstrom mouring@netnet.net Ben.a.Lindstrom@uwrf.edu (NeXT mail) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12128-5@oink>; Fri, 10 Nov 1995 15:37:48 +0000 Received: from punt-1.mail.demon.net with mailstore id 816007200:00570:1; Fri, 10 Nov 95 12:40:00 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa00354; 10 Nov 95 12:39 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA09405 for dbi-users-real; Fri, 10 Nov 1995 02:22:59 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id CAA09400 for ; Fri, 10 Nov 1995 02:22:45 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aa04301; 10 Nov 95 10:24 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa19797; 10 Nov 95 10:23 GMT Received: from toad by oink with SMTP (PP) id <10755-0@oink>; Fri, 10 Nov 1995 10:23:38 +0000 Received: by toad (5.0/SMI-SVR4) id AA14103; Fri, 10 Nov 1995 10:23:15 +0000 Date: Fri, 10 Nov 1995 10:23:15 +0000 From: Tim Bunce Message-Id: <9511101023.AA14103@toad> To: james.taylor@srs.gov Subject: Re: DBD::Oracle Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 241 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: james.taylor@srs.gov > > Tim, > > I no longer need to make any changes to get DBD::Oracle to compile with Solaris > 2.4 SPARCompiler C. You might want to remove the note from the README file. Done. Thanks for the update. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12128-11@oink>; Fri, 10 Nov 1995 15:38:13 +0000 Received: from punt-1.mail.demon.net with mailstore id 816009374:17654:0; Fri, 10 Nov 95 13:16:14 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa17625; 10 Nov 95 13:16 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP) id <11476-0@frigate.doc.ic.ac.uk>; Wed, 8 Nov 1995 00:27:54 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA00586 for dbi-users-real; Tue, 7 Nov 1995 13:13:52 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA00582 for ; Tue, 7 Nov 1995 13:13:42 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aa10198; 7 Nov 95 21:08 GMT Received: from relay-4.mail.demon.net by relay-3.mail.demon.net id sg.ab26996; 7 Nov 95 21:07 GMT Received: from toad by oink with SMTP (PP) id <16788-0@oink>; Tue, 7 Nov 1995 20:47:45 +0000 Received: by toad (5.0/SMI-SVR4) id AA09042; Tue, 7 Nov 1995 20:47:23 +0000 Date: Tue, 7 Nov 1995 20:47:23 +0000 From: Tim Bunce Message-Id: <9511072047.AA09042@toad> To: dbi-users@fugue.com, Scott=Alexander%se%FNS@fnsvines1.fns.com Subject: Re: How to trap database error messages? X-Sun-Charset: US-ASCII content-length: 517 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Scott=Alexander%se%FNS@fnsvines1.fns.com > > Can someone provide me with example code on how to trap database error > messages. I'm trying to trap a unique constraint message when inserting > records into an Oracle db. It seems by default to output the message to the > display. > > Thanks, > Scott Alexander > sga@fns.com Error handling needs more work. This will be addressed soon (this week maybe). Meanwhile I'd suggest just commenting out the Carp::carp line in default_handler in DBI.pm! Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12128-15@oink>; Fri, 10 Nov 1995 15:38:30 +0000 Received: from punt-1.mail.demon.net with mailstore id 816011646:07104:1; Fri, 10 Nov 95 13:54:06 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa06300; 10 Nov 95 13:53 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA09509 for dbi-dev-real; Fri, 10 Nov 1995 03:28:38 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id DAA09505 for ; Fri, 10 Nov 1995 03:28:35 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ac16980; 10 Nov 95 11:19 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa29594; 10 Nov 95 11:19 GMT Received: from toad by oink with SMTP (PP) id <10951-0@oink>; Fri, 10 Nov 1995 11:01:18 +0000 Received: by toad (5.0/SMI-SVR4) id AA14172; Fri, 10 Nov 1995 11:00:46 +0000 Date: Fri, 10 Nov 1995 11:00:46 +0000 From: Tim Bunce Message-Id: <9511101100.AA14172@toad> To: dbi-dev@fugue.com, phj@pilot.njin.net Subject: Re: sqlNet login X-Sun-Charset: US-ASCII content-length: 567 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: "Patricia H. Ju" > > Hi all, > > Sorry if this is inappropriate (I can't remember what is and what isn't) It's not really, it's an Oracle support issue. But anyway... > or can be found elsewhere (tell me where!) but I have no idea how to > loginto another host's oracle server via Oracle DBI. The > connect string should probably be: > sqlplus mylogin/mypassword@T:the.server.com:database That looks okay to me (assuming you have Sql*Net installed). Get it working from the command line first. Talk to Oracle if you can't. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12128-17@oink>; Fri, 10 Nov 1995 15:38:42 +0000 Received: from punt-1.mail.demon.net with mailstore id 816011821:08561:1; Fri, 10 Nov 95 13:57:01 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa08185; 10 Nov 95 13:56 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA09524 for dbi-dev-real; Fri, 10 Nov 1995 03:33:22 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA09520 for ; Fri, 10 Nov 1995 03:33:17 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id LAA25630 for dbi-dev@fugue.com; Fri, 10 Nov 1995 11:31:06 GMT From: Alligator Descartes Message-Id: <199511101131.LAA25630@fruitbat.mcqueen.com> Subject: Driver loading To: dbi-dev@fugue.com Date: Fri, 10 Nov 1995 11:31:05 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 814 All. A quick note about driver loading. Discussions welcome! A recently posted note of mine had the line: $drh = DBI->install_driver( 'mSQL' ); in it. Tim suggested: $dbh = connect( 'dbhost', 'dbname', 'mSQL' ); to load the drivers. Unfortunately, in DBD::mSQL there's a private method $drh->func( '_ListDBs' ) which operates against the driver and not any actual databases. How can this be done via the connect method instead of the explicit driver load? Also, how can we check the driver's actually been loaded? $dbh may be undef after a failed connect ( the *connect* may have failed ), but how can we tell the driver load didn't fail? ( Missing symbols or whatever? ) Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12128-21@oink>; Fri, 10 Nov 1995 15:39:01 +0000 Received: from punt-1.mail.demon.net with mailstore id 816012302:12309:1; Fri, 10 Nov 95 14:05:02 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa12243; 10 Nov 95 14:04 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA09492 for dbi-users-real; Fri, 10 Nov 1995 03:20:01 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id DAA09485 for ; Fri, 10 Nov 1995 03:19:58 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab16980; 10 Nov 95 11:19 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ac29594; 10 Nov 95 11:19 GMT Received: from toad by oink with SMTP (PP) id <10930-0@oink>; Fri, 10 Nov 1995 10:58:14 +0000 Received: by toad (5.0/SMI-SVR4) id AA14168; Fri, 10 Nov 1995 10:57:52 +0000 Date: Fri, 10 Nov 1995 10:57:52 +0000 From: Tim Bunce Message-Id: <9511101057.AA14168@toad> To: dbi-users@fugue.com, tmh@whitetiger.engr.sgi.com Subject: Re: Oracle Login Problems X-Sun-Charset: US-ASCII content-length: 859 Organisation: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Todd Hoff > To: perldb-interest@vix.com Please switch to using dbi-users@fugue.com > ERROR EVENT: DBI::db=HASH(0x100a75dc)->DBI::default_handler: > 1034 ORA-01034: ORACLE not available (DBD: login failed) > at /usr/local/lib/perl5/DBD/Oracle.pm line 99 > ora_login("i18n", "sgi", "cool"); > The same call in per4 oraperl works. I've tried lots of combinations > and asked some people. > > Do you have any idea why it might not be working? Maybe, but not why perl4 oraperl would work. I'll need to check the code. Is i18n an Sql*NET alias? Try: ora_login("", "sgi@i18n", "cool"); and also try ora_login("", "sgi", "cool"); with ORACLE_SID or TWO_TASK set as needed. Please let me know how both these work out since I need to get to the bottom of this. Has anyone else had similar problems? Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12128-23@oink>; Fri, 10 Nov 1995 15:39:08 +0000 Received: from punt-1.mail.demon.net with mailstore id 816012488:13812:1; Fri, 10 Nov 95 14:08:08 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa13641; 10 Nov 95 14:07 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA09558 for dbi-dev-real; Fri, 10 Nov 1995 03:41:41 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA09554 for ; Fri, 10 Nov 1995 03:41:39 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id LAA25755; Fri, 10 Nov 1995 11:39:16 GMT From: Alligator Descartes Message-Id: <199511101139.LAA25755@fruitbat.mcqueen.com> Subject: Re: Oracle Login Problems To: Tim.Bunce@ig.co.uk (Tim Bunce) Date: Fri, 10 Nov 1995 11:39:15 +0000 (GMT) Cc: dbi-dev@fugue.com In-Reply-To: <9511101057.AA14168@toad> from "Tim Bunce" at Nov 10, 95 10:57:52 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1007 > Maybe, but not why perl4 oraperl would work. I'll need to check the code. > > Is i18n an Sql*NET alias? > > Try: > ora_login("", "sgi@i18n", "cool"); > > and also try > > ora_login("", "sgi", "cool"); > > with ORACLE_SID or TWO_TASK set as needed. > > Please let me know how both these work out since I need to get to the > bottom of this. > > Has anyone else had similar problems? Not in the Oraperl emulation, just straight DBI. SQL*Net V2 lookups don't appear to work. If you define the connection descriptor to be a SQL*Net V2 connection descriptor in TNSNAMES.ORA, it doesn't appear to use it and attempts connection to a local database of the name of whatever you specified as the c.desc. I thought maybe fiddling the args ( as you did above ) or adding judicious '@'s in may have helped, but it doesn't. I haven't had a lot of time to play with it, right enough...... > Tim. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15841-10@oink>; Fri, 10 Nov 1995 23:24:58 +0000 Received: from punt-1.mail.demon.net with mailstore id 816035995:02022:0; Fri, 10 Nov 95 20:39:55 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa02017; 10 Nov 95 20:39 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id g.aa10820; 10 Nov 95 20:39 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA00145 for dbi-users-real; Fri, 10 Nov 1995 10:15:19 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sc.ZIB-Berlin.DE (sc.ZIB-Berlin.DE [130.73.108.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA00141 for ; Fri, 10 Nov 1995 10:15:13 -0800 Received: from franz.ww.TU-Berlin.DE by sc.ZIB-Berlin.DE (4.1/SMI-5.3-sc/22.08.93) id AA13452; Fri, 10 Nov 95 18:59:12 +0100 Received: by franz.ww.TU-Berlin.DE (NX5.67d/NeXT-1.0) id AA10583; Fri, 10 Nov 95 19:13:06 +0100 Date: Fri, 10 Nov 95 19:13:06 +0100 Message-Id: <9511101813.AA10583@franz.ww.TU-Berlin.DE> Subject: Welcome new user The DBI Mailing Lists To: dbi-users@fugue.com, koenig@franz.ww.tu-berlin.de, Tim.Bunce@ig.co.uk From: The Perl Module Server Welcome The DBI Mailing Lists, __PAUSE__, the perl authors upload server, has a userid for you: DBIML. The following credentials are in the database: Name: [The DBI Mailing Lists] email: [dbi-users@fugue.com] homepage: [http://www.fugue.com/dbi] You are welcome to change these credentials yourself and upload your work to ftp://franz.ww.tu-berlin.de/incoming/ whenever you feel the desire. See http://franz.ww.tu-berlin.de/modulelist/ for details. [entered by Tim Bunce] + DBIML The DBI Mailing Lists <dbi-users@fugue.com> ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19045-9@oink>; Sat, 11 Nov 1995 03:21:37 +0000 Received: from punt-1.mail.demon.net with mailstore id 816058186:21059:1; Sat, 11 Nov 95 02:49:46 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa20974; 11 Nov 95 2:49 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA01258 for dbi-dev-real; Fri, 10 Nov 1995 16:25:01 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA01254 for ; Fri, 10 Nov 1995 16:24:59 -0800 Received: by gw.home.vix.com id AA14018; Fri, 10 Nov 95 16:27:12 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aa01562; 11 Nov 95 0:12 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa24182; 11 Nov 95 0:12 GMT Received: from toad by oink with SMTP (PP) id <17913-0@oink>; Sat, 11 Nov 1995 00:12:19 +0000 Received: by toad (5.0/SMI-SVR4) id AA15607; Sat, 11 Nov 1995 00:11:56 +0000 Date: Sat, 11 Nov 1995 00:11:56 +0000 From: Tim Bunce Message-Id: <9511110011.AA15607@toad> To: perl5-porters@africa.nicoh.com, ak@ananke.s.bawue.de Subject: Re: ODBC extensions for Perl Cc: dbi-dev@fugue.com, mhm@austin.ibm.com X-Sun-Charset: US-ASCII Content-Length: 1253 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: ak@ananke.s.bawue.de (Andreas Kaiser) > > On Fri, 10 Nov 95 01:23:14 UT, you wrote: > > >One of my only reasons for using NT is "easy" access to the SQLServer > >database. If Win32 Perl had an ODBC or DBLIB extension similar to sybperl > >that would be fantastic! Is anyone working on this? > > I have implemented an extension for DB2/2 CLI which is pretty close to > ODBC. Sigh. A full DBI compatible DB2 driver has been developed already: DBI amcO Generic Database Interface (see DBD modules) DBIML DBD:: ::Oracle amcO Oracle Driver for DBI TIMB ::Ingres cmcO Ingres Driver for DBI TIMB ::Informix adcO Informix Driver for DBI ADESC ::mSQL amcO Msql Driver for DBI ADESC ::DB2 adcO DB2 Driver for DBI MHM ::Sybase idcO Sybase Driver for DBI MEWP ::QBase amcO QBase Driver for DBI BENLI MHM Mike Moran Take a look at http://www.hermetica.com/technologia/DBI Please subscribe to the dbi-dev mailing list via http://www.fugue.com/dbi I've CC'd this to the list and Mike. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <29508-3@oink>; Mon, 13 Nov 1995 07:08:17 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816131306:12756:1; Sat, 11 Nov 95 23:08:26 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa12638; 11 Nov 95 23:08 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA04463 for dbi-users-real; Sat, 11 Nov 1995 13:47:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sgi.sgi.com (SGI.COM [192.48.153.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA04459 for ; Sat, 11 Nov 1995 13:47:13 -0800 Received: from whitetiger.engr.sgi.com by sgi.sgi.com via ESMTP (950405.SGI.8.6.12/910110.SGI) id NAA12754; Sat, 11 Nov 1995 13:49:37 -0800 Received: by whitetiger.engr.sgi.com (940816.SGI.8.6.9/940406.SGI.AUTO) id NAA22618; Sat, 11 Nov 1995 13:49:29 -0800 From: Todd Hoff Message-Id: <9511111349.ZM22616@whitetiger.engr.sgi.com> Date: Sat, 11 Nov 1995 13:49:27 -0800 In-Reply-To: Tim Bunce "Re: Oracle Login Problems" (Nov 10, 10:57am) References: <9511101057.AA14168@toad> X-Mailer: Z-Mail (3.2.0 26oct94 MediaMail) To: Tim Bunce , dbi-users@fugue.com Subject: Re: Oracle Login Problems Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii On Nov 10, 10:57am, Tim Bunce wrote: > Is i18n an Sql*NET alias? No answer to this yet. > Try: > ora_login("", "sgi@i18n", "cool"); > > and also try > > ora_login("", "sgi", "cool"); > > with ORACLE_SID or TWO_TASK set as needed. > > Please let me know how both these work out since I need to get to the > bottom of this. > Both worked like a charm! Thanx! ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <29509-26@oink>; Mon, 13 Nov 1995 07:14:39 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816148970:07312:1; Sun, 12 Nov 95 04:02:50 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa07272; 12 Nov 95 4:02 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA04825 for dbi-users-real; Sat, 11 Nov 1995 18:48:23 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from medg.lcs.mit.edu (medg.lcs.mit.edu [18.30.0.179]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA04821 for ; Sat, 11 Nov 1995 18:48:21 -0800 Received: from martin.lcs.mit.edu (martin.lcs.mit.edu [18.30.0.175]) by medg.lcs.mit.edu (8.6.9/8.6.9) with ESMTP id VAA00569 for ; Sat, 11 Nov 1995 21:44:20 -0500 From: Eric Michael Jordan Received: (emjordan@localhost) by martin.lcs.mit.edu (8.6.9/8.6.9) id VAA12493; Sat, 11 Nov 1995 21:49:20 -0500 Date: Sat, 11 Nov 1995 21:49:20 -0500 Message-Id: <199511120249.VAA12493@martin.lcs.mit.edu> To: dbi-users@fugue.com Subject: error message when I use ora_logoff The following script gives me the following error message: Bad free() ignored at junk.pl line 10. #!/u/emjordan/bin/perl5 -w # -*- Perl -*- use DBI; use Oraperl; $lda = &ora_login("", "scott/tiger\@P:PROD", "") || die &badpw; &ora_logoff($lda); I realize that there are other ways to use ora_login, but I have to do it this way because: 1. This is going to be executed by a cgi script, so I will not be able to set up the environment ahead of time and ensure that $ENV{'oracle_sid'} has been set to PROD. 2. There is a bug in Oracle v. 7.1.4 on SunOS that sometimes causes ora_login to freeze if any application that uses OCI tries to change any environment variable before calling orlon. This manifests itself if the first argument to ora_login is non-empty. Eric ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <00657-4@oink>; Mon, 13 Nov 1995 15:48:43 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816271151:17633:1; Mon, 13 Nov 95 13:59:11 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa17500; 13 Nov 95 13:59 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA09331 for dbi-users-real; Mon, 13 Nov 1995 03:26:49 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id DAA09327 for ; Mon, 13 Nov 1995 03:26:44 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ah17778; 13 Nov 95 11:27 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.aa16252; 13 Nov 95 11:27 GMT Received: from toad by oink with SMTP (PP) id <02386-0@oink>; Mon, 13 Nov 1995 10:31:51 +0000 Received: by toad (5.0/SMI-SVR4) id AA20021; Mon, 13 Nov 1995 10:31:26 +0000 Date: Mon, 13 Nov 1995 10:31:26 +0000 From: Tim Bunce Message-Id: <9511131031.AA20021@toad> To: turbo@flern.onramp.net Subject: Re: DBD Oracle and DBI mailing lists Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 883 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > Date: Mon, 13 Nov 1995 01:06:38 -0600 (CST) > From: turbo@onramp.net > > I am sorry to bother you with this but I can't seem to find the new > mailing lists for DBD Oracle and DBI interest. Subscrive via http://www.fugue.com/dbi > I am working on a large > web interface to an oracle accounting system and have some questions and > comments. Currently about the apparent lack of support for sql select > for update queries (by apparent I mean I can't get them to work:). It's not been specified/implemented yet. Tim. p.s. Your message had broken headers. The blank line is the problem: From turbo@flern.onramp.net Mon Nov 13 07:40 GMT 1995 Date: Mon, 13 Nov 1995 01:06:38 -0600 (CST) To: Tim.Bunce@ig.co.uk X-Url: Content-Type: text X-Personal_Name: Chris Turbeville turbo@onramp.net From: turbo@onramp.net Subject: DBD Oracle and DBI mailing lists content-length: 365 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <02419-0@oink>; Mon, 13 Nov 1995 17:26:27 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816283554:07030:1; Mon, 13 Nov 95 17:25:54 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa06966; 13 Nov 95 17:25 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA09630 for dbi-users-real; Mon, 13 Nov 1995 07:16:05 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail.crl.com (mail.crl.com [165.113.1.22]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA09626 for ; Mon, 13 Nov 1995 07:16:03 -0800 Received: from [205.149.164.116] (bta.vip.best.com) by mail.crl.com with SMTP id AA18169 (5.65c/IDA-1.5 for ); Mon, 13 Nov 1995 07:16:26 -0800 Date: Mon, 13 Nov 1995 07:16:26 -0800 X-Sender: mrlizard@mail.crl.com Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: dbi-users@fugue.com From: mrlizard@crl.com (Brian Valente) Subject: DBD::mSQL test build problems So many mailing lists... I hope this is the right one. I have problems getting "make test" to compile correctly on my Sparc 20/Solaris 2.4. I have mSQL 1.0.9 installed successfully I have perl 5.001m installed successfully I have DBD 0.64 installed successfully when I run "make test" according to the instructions, I get: /usr/local/bin/perl -I./blib -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/bin/perl5 test.pl Testing: DBI->install_driver( 'mSQL' ): ok not ok: Testing: $drh->func( '', '_ListDBs' ): ***Error code 255 make: Fatal error: command failed for target 'test_dynamic' Originally, I got a MSQL_HOME not defined error, so I set it to /usr/local/Minerva - Hope that's right. Any help is deeply appreciated. Sincerely, Brian Valente ====================================================================== Brian Valente || brian_valente@rmdudley.com R.M. Dudley Corporation || mrlizard@crl.com 1799 Old Bayshore Suite 200 || applelink:RMDUDLEY Burlingame, CA 94010 || http://www.rmdudley.com/rmdudley/ (415) 697-1650 voice || (415) 697-1816 fax || ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03885-2@oink>; Mon, 13 Nov 1995 20:30:08 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816285998:28003:1; Mon, 13 Nov 95 18:06:38 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa27887; 13 Nov 95 18:06 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA09690 for dbi-users-real; Mon, 13 Nov 1995 07:48:44 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA09686 for ; Mon, 13 Nov 1995 07:48:41 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id PAA15991; Mon, 13 Nov 1995 15:46:21 GMT From: Alligator Descartes Message-Id: <199511131546.PAA15991@fruitbat.mcqueen.com> Subject: Re: DBD::mSQL test build problems To: mrlizard@crl.com (Brian Valente) Date: Mon, 13 Nov 1995 15:46:20 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: from "Brian Valente" at Nov 13, 95 07:16:26 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 998 > I have problems getting "make test" to compile correctly on my Sparc > 20/Solaris 2.4. > > I have mSQL 1.0.9 installed successfully > > I have perl 5.001m installed successfully > > I have DBD 0.64 installed successfully > > when I run "make test" according to the instructions, I get: > > /usr/local/bin/perl -I./blib -I/usr/local/lib/perl5/sun4-solaris > -I/usr/local/bin/perl5 test.pl > Testing: DBI->install_driver( 'mSQL' ): ok > not ok: > Testing: $drh->func( '', '_ListDBs' ): ***Error code 255 > make: Fatal error: command failed for target 'test_dynamic' > > Originally, I got a MSQL_HOME not defined error, so I set it to > /usr/local/Minerva - Hope that's right. Right, that's either, the driver failed to load, or the DB listing failed. I suspect the second. So, obvious question, do you have 'msqld' running? If you don't, that'll cause the problem. > Brian Valente -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03885-28@oink>; Mon, 13 Nov 1995 20:32:14 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816290933:14449:0; Mon, 13 Nov 95 19:28:53 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa14426; 13 Nov 95 19:28 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa26425; 13 Nov 95 19:28 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA10241 for dbi-dev-real; Mon, 13 Nov 1995 09:43:28 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netnet1.netnet.net (netnet1.netnet.net [198.70.64.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA10237 for ; Mon, 13 Nov 1995 09:43:27 -0800 Received: (from mouring@localhost) by netnet1.netnet.net (8.6.9/8.6.9) id LAA10673; Mon, 13 Nov 1995 11:45:50 -0600 Date: Mon, 13 Nov 1995 11:45:50 -0600 (CST) From: Ben Lindstrom To: dbi-dev@fugue.com Subject: Passing information from dbh to sth. Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I must be really losing it. =) I could have swore I send off a question about how to migrate data from dbh to sth (mainly the database connection). Anyways. I found: imp_dbh_from_sth (I think that's what it's call I don't have my code infront of me.) But when I track it down I get to *child() function that I can't find. Could so one point me in the direction of how the migration for the database information gets from dbh to sth? Thanks. --- Ben Lindstrom mouring@netnet.net Ben.a.Lindstrom@uwrf.edu (NeXT mail) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05266-3@oink>; Mon, 13 Nov 1995 22:55:16 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816296272:04799:0; Mon, 13 Nov 95 20:57:52 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa04788; 13 Nov 95 20:57 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa01546; 13 Nov 95 20:56 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA10532 for dbi-users-real; Mon, 13 Nov 1995 10:27:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from altrade.nijmegen.inter.nl.net (altrade.nijmegen.inter.nl.net [193.67.237.6]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA10528 for ; Mon, 13 Nov 1995 10:27:33 -0800 Received: from rijken.inter.NL.net by altrade.nijmegen.inter.nl.net via ah99-9.Arnhem.NL.net [193.79.228.234] with ESMTP id TAA23869 (8.6.10/1.8); Mon, 13 Nov 1995 19:29:26 +0100 Received: (from mrijken@localhost) by rijken.inter.NL.net (8.6.11/8.6.11) id TAA00184; Mon, 13 Nov 1995 19:27:15 +0100 Date: Mon, 13 Nov 1995 19:27:14 +0100 (MET) From: Marti Rijken To: Brian Valente cc: dbi-users@fugue.com Subject: Re: DBD::mSQL test build problems In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 13 Nov 1995, Brian Valente wrote: > So many mailing lists... I hope this is the right one. Yes, it is... > [...] > Testing: DBI->install_driver( 'mSQL' ): ok > not ok: > Testing: $drh->func( '', '_ListDBs' ): ***Error code 255 > make: Fatal error: command failed for target 'test_dynamic' I could simulate the error by creating an emty database directory $MSQL_HOME/msqldb, be it with: "*** [test_dynamic] Error 2". So perhaps you have no database defined yet? If so create a dummy database with msql, and see it it solves the problem. > Originally, I got a MSQL_HOME not defined error, so I set it to > /usr/local/Minerva - Hope that's right. That's the default. Setting MSQL_HOME to the default will get rid of the annoying but harmless warning. Hope this helps, ___ Marti Rijken ____ mrijken@inter.nl.net ____ Arnhem, Netherlands ___ _______ Province of Gelderland, nature conservation department ______ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <06849-0@oink>; Tue, 14 Nov 1995 00:54:33 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816310262:28727:1; Tue, 14 Nov 95 00:51:02 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa28518; 14 Nov 95 0:50 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA11386 for dbi-users-real; Mon, 13 Nov 1995 15:28:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA11382 for ; Mon, 13 Nov 1995 15:28:12 -0800 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id SAA00390 for ; Mon, 13 Nov 1995 18:30:38 -0500 Message-Id: <199511132330.SAA00390@pilot.njin.net> To: dbi-users@fugue.com Subject: Deletion weirdnesses Date: Mon, 13 Nov 1995 18:30:37 -0500 From: "Patricia H. Ju" Hi all, I'm having strangenesses deleting from an Oracle database. deleteRec: DELETE FROM Article WHERE MSGID = 1010 at Table.pm line 530. ERROR EVENT: DBI::st=HASH(0x207450)->DBI::default_handler: 60 ORA-00060: deadlock detected while waiting for resource (DBD: oexec error) at /usr/local/lib/perl5/Oraperl.pm line 93 History: 1. I was writing a CGI script to automate deletion of msgs and the areas to which they belonged; it was really slow (like, never finished) 2. After a few different ways of deleting things (deleting msgs one by one has worked for me before), I finally tried a straight delete in sqlplus. "DELETE FROM Article where MSGID = 1010" stopped working and "DELETE FROM Article where AREAID=10" just sat there looking at me. If there is a deadlock here, I have no idea where it would be. The first "DELETE FROM Article where AREAID=10" (which should retrieve a few rows and delete) generated an ora_fetch out of sequence message. Help ? Patricia ............. Who needs patience when you've got force ? ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <07656-0@oink>; Tue, 14 Nov 1995 02:44:30 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816316669:24696:1; Tue, 14 Nov 95 02:37:49 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa24609; 14 Nov 95 2:37 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA11790 for dbi-dev-real; Mon, 13 Nov 1995 17:32:47 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA11786 for ; Mon, 13 Nov 1995 17:32:44 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id TAA40789 for ; Mon, 13 Nov 1995 19:35:10 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dbi-dev@fugue.com at austin.ibm.com; id AA22766; Mon, 13 Nov 1995 19:35:09 -0600 Message-Id: <9511140135.AA22766@dax.austin.ibm.com> To: dbi-dev@fugue.com Subject: Blobs are now working on DB2perl Date: Mon, 13 Nov 95 19:35:09 -0600 A few tweaks in DBD::DB2 and big correction in my perl script and everyting works fine. A while back, I asked about other databases abilities to read a blob to a "file" and Tim suggested something like $sth->copy_blob_to_file(1, $open_file_handle); I thought about it for awhile and would like to suggest the following: $sth->readblob_to_file(field, file_or_device_name); First, the above is (atleast for DB2) more semantically correct, in that the DB2 engine will be doing al the I/O not perl. And secondly, it would allow the databases that support direct engine I/O to work without wasting cycles and file handles. Any serious downside that I'm missing? Mike ------------------------------------------------------------------------------ Michael H. Moran | Standard Disclaimer: The content of mhm@austin.ibm.com | this posting is independent of Commercial Performance | official IBM position. IBM Corporation, Austin, Texas | ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13542-9@oink>; Tue, 14 Nov 1995 18:02:27 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816371290:12210:0; Tue, 14 Nov 95 17:48:10 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa12160; 14 Nov 95 17:48 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 14 Nov 1995 17:26:03 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA14369 for dbi-users-real; Tue, 14 Nov 1995 05:43:30 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id FAA14365 for ; Tue, 14 Nov 1995 05:43:27 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aa05115; 14 Nov 95 13:45 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.am20059; 14 Nov 95 13:43 GMT Received: from toad by oink with SMTP (PP) id <10703-0@oink>; Tue, 14 Nov 1995 13:40:50 +0000 Received: by toad (5.0/SMI-SVR4) id AA04112; Tue, 14 Nov 1995 13:40:27 +0000 Date: Tue, 14 Nov 1995 13:40:27 +0000 From: Tim Bunce Message-Id: <9511141340.AA04112@toad> To: turbo@flern.onramp.net Subject: Re: DBD Oracle and DBI mailing lists Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 965 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Chris Turbeville > > > > Date: Mon, 13 Nov 1995 01:06:38 -0600 (CST) > > > From: turbo@onramp.net > > > > > > I am sorry to bother you with this but I can't seem to find the new > > > mailing lists for DBD Oracle and DBI interest. > > > > Subscrive via http://www.fugue.com/dbi > Have done thanks. > > > > I am working on a large > > > web interface to an oracle accounting system and have some questions and > > > comments. Currently about the apparent lack of support for sql select > > > for update queries (by apparent I mean I can't get them to work:). > > > > It's not been specified/implemented yet. > Hmm. Is there any way then in oraperl to do mutually exclusive > updates? (ie increment a counter and be gauranteed your the only one > with that count) > Thanks > -Chris Just $dbh->do() an update and that should create an exclusive lock (also take a look at "alter transaction ..." and "alter session ...") Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15375-0@oink>; Wed, 15 Nov 1995 07:05:49 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816372564:23168:0; Tue, 14 Nov 95 18:09:24 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa23164; 14 Nov 95 18:09 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id msg.aa28273; 14 Nov 95 18:08 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA14506 for dbi-users-real; Tue, 14 Nov 1995 06:55:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail.crl.com (mail.crl.com [165.113.1.22]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA14502 for ; Tue, 14 Nov 1995 06:55:39 -0800 Received: from [205.149.164.116] (bta.vip.best.com) by mail.crl.com with SMTP id AA22008 (5.65c/IDA-1.5 for ); Tue, 14 Nov 1995 06:56:18 -0800 Date: Tue, 14 Nov 1995 06:56:18 -0800 X-Sender: mrlizard@mail.crl.com Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: dbi-users@fugue.com From: Brian Valente Subject: DBD-mSQL Examples? Hi Folks, Well, seems that I have most of the big chunks up and running (thanks for all your help), so rather than grind on with detailed bits and pieces questions, does anyone have good example perl scripts? I've read the spec, but examples are always a better way for me to understand how it's _really_ done. Regards, Brian Valente ====================================================================== Brian Valente || brian_valente@rmdudley.com R.M. Dudley Corporation || mrlizard@crl.com 1799 Old Bayshore Suite 200 || applelink:RMDUDLEY Burlingame, CA 94010 || http://www.rmdudley.com/rmdudley/ (415) 697-1650 voice || (415) 697-1816 fax || ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15375-11@oink>; Wed, 15 Nov 1995 07:06:53 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816375530:18714:1; Tue, 14 Nov 95 18:58:50 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa18655; 14 Nov 95 18:58 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA14783 for dbi-users-real; Tue, 14 Nov 1995 08:53:25 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA14779 for ; Tue, 14 Nov 1995 08:53:19 -0800 Received: by relay-4.mail.demon.net id msg.aa20781; 14 Nov 95 16:53 GMT Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab24012; 14 Nov 95 16:18 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ad20901; 14 Nov 95 16:16 GMT Received: from toad by oink with SMTP (PP) id <11676-0@oink>; Tue, 14 Nov 1995 14:55:00 +0000 Received: by toad (5.0/SMI-SVR4) id AA05215; Tue, 14 Nov 1995 14:54:37 +0000 Date: Tue, 14 Nov 1995 14:54:37 +0000 From: Tim Bunce Message-Id: <9511141454.AA05215@toad> To: alan@lims.gel.com Subject: I know I'm cheating ... Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 807 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 Forwarded to the list: ----- Begin Included Message ----- Date: Mon, 13 Nov 1995 11:48:48 -0800 From: alan.johnson@gel.com (Alan Johnson) but I'm coming to the horse himself :-) You can swat me (like a horse does :-)) or ... Does ora_do (oraperl 4) return the return status of the PL/SQL procedure called? I.e., if I write: $sql_proc = "begin if blah then return(-100) else return (0) end if; end;"; $dummy = &ora_do($lda,$sql_proc); does $dummy have the returned value of the procedure called (i.e. -100 or 0) It doesn't appear to, but I'm just checking. TIA Alan -------------------------------- Alan Johnson MIS Director General Engineering Laboratories 2040 Savage Road Charleston, SC 29414 alan.johnson@gel.com (803) 556-8171 (voice) (803) 766-1178 (fax) ----- End Included Message ----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15375-30@oink>; Wed, 15 Nov 1995 07:10:05 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816382414:17687:1; Tue, 14 Nov 95 20:53:34 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa17603; 14 Nov 95 20:53 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA15497 for dbi-users-real; Tue, 14 Nov 1995 11:17:39 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from ns.ge.com (ns.ge.com [192.35.39.24]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA15493 for ; Tue, 14 Nov 1995 11:17:36 -0800 Received: from mobster.cit.ge.com ([3.47.4.100]) by ns.ge.com (8.6.12/8.6.11) with ESMTP id OAA25215 for ; Tue, 14 Nov 1995 14:19:51 -0500 Received: from mobster (localhost.cit.ge.com [127.0.0.1]) by mobster.cit.ge.com (8.6.12/8.6.12) with ESMTP id OAA20349 for ; Tue, 14 Nov 1995 14:21:37 -0500 Message-Id: <199511141921.OAA20349@mobster.cit.ge.com> X-Mailer: exmh version 1.6.2 7/18/95 To: dbi-users@fugue.com Subject: Oracle DBD installation Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 14 Nov 1995 14:21:35 -0500 From: Bob Snyder This may sound stupid, but is it possible to configure/compile the Oracle DBD without having the Oracle software installed? I'm trying to set up CGI scripts to access a remote Oracle server on a different platform, so I don't have Oracle installed on the machine I'm using for web development (solaris 2.4). If Oracle is required, I'd appreciate someone telling me exactly what I need, or if they have any better ideas. I have a good bit of web and CGI knowledge, a fair bit of perl knowledge, and very little Oracle knowledge. Bob ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15375-42@oink>; Wed, 15 Nov 1995 07:12:01 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816385796:18118:1; Tue, 14 Nov 95 21:49:56 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa17873; 14 Nov 95 21:49 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA15686 for dbi-users-real; Tue, 14 Nov 1995 12:22:38 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA15680 for ; Tue, 14 Nov 1995 12:22:32 -0800 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id PAA23166 for ; Tue, 14 Nov 1995 15:24:56 -0500 Message-Id: <199511142024.PAA23166@pilot.njin.net> To: dbi-users@fugue.com Subject: ofetch out of sequence (was deletion weirdnesses) Date: Tue, 14 Nov 1995 15:24:53 -0500 From: "Patricia H. Ju" Rokey. This is what's happening: query: DELETE ARTICLE WHERE AREAID = 10 at Table.pm line 360. ERROR EVENT: DBI::st=HASH(0x1a8af0)->DBI::default_handler: 1002 ORA-01002: fetch out of sequence (DBD: ofetch error) at Table.pm line 362 Mind you, the articles I'm deleting have no dependenceies other than referrring to the areaID of table Area. So I figured if I delete all the articles that are of area 10, *then* delete area 10, I should be ok, right ? Apparently not. Anyone know what's going on ? Patricia ............. Who needs patience when you've got force ? ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <00523-26@oink>; Wed, 15 Nov 1995 11:51:13 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816434532:03311:1; Wed, 15 Nov 95 11:22:12 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa03087; 15 Nov 95 11:21 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA17081 for dbi-users-real; Wed, 15 Nov 1995 01:10:54 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from nova.auspex.fr ([194.2.216.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA17077 for ; Wed, 15 Nov 1995 01:10:48 -0800 Received: (from kstock@localhost) by nova.auspex.fr (8.6.9/8.6.9) id KAA00476; Robinson France, Wed, 15 Nov 1995 10:13:35 +0100 Date: Wed, 15 Nov 1995 10:13:35 +0100 Message-Id: <199511150913.KAA00476@nova.auspex.fr> To: alan@lims.gel.com Subject: Re: I know I'm cheating ... Cc: dbi-users@fugue.com From: KStock@auspex.com (Kevin Stock) Organisation: Auspex Systems European Technical Support Centre Phone: +33 1 41 07 95 05 Reply-To: KStock@auspex.com > Date: Mon, 13 Nov 1995 11:48:48 -0800 > From: alan.johnson@gel.com (Alan Johnson) > > but I'm coming to the horse himself :-) You can swat me (like a horse does > :-)) or ... Tim's been called a lot of things, most of them by me :-) but that's a new one... > Does ora_do (oraperl 4) return the return status of the PL/SQL procedure > called? I.e., if I write: > > $sql_proc = "begin if blah then return(-100) else return (0) end if; end;"; > > $dummy = &ora_do($lda,$sql_proc); > > does $dummy have the returned value of the procedure called (i.e. -100 or 0) > > It doesn't appear to, but I'm just checking. Oraperl 4 doesn't know anything about PL/SQL. All right, _I_ didn't know anything about PL/SQL when I wrote Oraperl :-). &ora_do returns the number of rows affected, unless it's 0, in which case it returns OK (which tests as true but hase a numerical value of 0). I think you'd need to store the return value in the database and then select it. Or, perhaps there's some way of making the procedure appear to fail with some particular error status which would be made available in $ora_errno... Kevin ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <02226-0@oink>; Wed, 15 Nov 1995 18:19:58 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816457627:14797:0; Wed, 15 Nov 95 17:47:07 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa14787; 15 Nov 95 17:47 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa11033; 15 Nov 95 17:46 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA18917 for dbi-users-real; Wed, 15 Nov 1995 07:06:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail.crl.com (mail.crl.com [165.113.1.22]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA18913 for ; Wed, 15 Nov 1995 07:06:55 -0800 Received: from [205.149.164.116] (bta.vip.best.com) by mail.crl.com with SMTP id AA01244 (5.65c/IDA-1.5 for ); Wed, 15 Nov 1995 07:07:34 -0800 Date: Wed, 15 Nov 1995 07:07:34 -0800 X-Sender: mrlizard@mail.crl.com Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: dbi-users@fugue.com From: Brian Valente Subject: Error codes and such For all the help I've received on this list, I'd like to provide some scripts and html pages to introduce people to using DBD/mSQL for back-ending a web site. As soon as I can get these working fairly bug-free, I'll do that. If you're interested, please e-mail me. In the meantime, I keep running into problems with getting a null value returned from a $dbh->prepare($querystring) statement. Is there a way to find out what went wrong? Right now, All I know is that the prepare statement didn't return a valid statement handle. Speaking of which, is there any documentation for error codes and handling _in general_? It'd sure help... Thanks again, Brian ====================================================================== Brian Valente || brian_valente@rmdudley.com R.M. Dudley Corporation || mrlizard@crl.com 1799 Old Bayshore Suite 200 || applelink:RMDUDLEY Burlingame, CA 94010 || http://www.rmdudley.com/rmdudley/ (415) 697-1650 voice || (415) 697-1816 fax || ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <02808-3@oink>; Wed, 15 Nov 1995 20:07:42 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816460947:13918:0; Wed, 15 Nov 95 18:42:27 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa13906; 15 Nov 95 18:42 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 15 Nov 1995 18:30:02 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA19020 for dbi-users-real; Wed, 15 Nov 1995 08:18:00 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA19014 for ; Wed, 15 Nov 1995 08:17:57 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id QAA11204; Wed, 15 Nov 1995 16:15:42 GMT From: Alligator Descartes Message-Id: <199511151615.QAA11204@fruitbat.mcqueen.com> Subject: Re: Error codes and such To: mrlizard@crl.com (Brian Valente) Date: Wed, 15 Nov 1995 16:15:41 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: from "Brian Valente" at Nov 15, 95 07:07:34 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1303 > For all the help I've received on this list, I'd like to provide some > scripts and html pages to introduce people to using DBD/mSQL for > back-ending a web site. As soon as I can get these working fairly > bug-free, I'll do that. If you're interested, please e-mail me. What a guy! If you want to send me the pages and scripts, I'll put them on the Hermetica pages for all to see. > In the meantime, I keep running into problems with getting a null value > returned from a $dbh->prepare($querystring) statement. Is there a way to > find out what went wrong? Right now, All I know is that the prepare > statement didn't return a valid statement handle. Speaking of which, is > there any documentation for error codes and handling _in general_? It'd > sure help... Use $DBI::errstr. The error from mSQL populates that variable. Simiarly, the error number goes into $DBI::errnum ( or $DBI::errno, I can never remeber ) but in DBD::mSQL that's always -1, so it's not much use. Do something like: $cursor = $dbh->prepare( "SELECT blah FROM blah" ); die "$DBI::errstr\n" unless $cursor; to check the prepare, then display the error if it blew out. Hope that helps. > Brian -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <04501-0@oink>; Wed, 15 Nov 1995 23:12:37 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816476201:01088:1; Wed, 15 Nov 95 22:56:41 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa00774; 15 Nov 95 22:56 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA19852 for dbi-users-real; Wed, 15 Nov 1995 13:23:26 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA19848 for ; Wed, 15 Nov 1995 13:23:22 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ac12019; 15 Nov 95 21:21 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa09240; 15 Nov 95 21:20 GMT Received: from toad by oink with SMTP (PP) id <03903-0@oink>; Wed, 15 Nov 1995 21:20:14 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA08777; Wed, 15 Nov 1995 21:19:51 +0000 Date: Wed, 15 Nov 1995 21:19:51 +0000 From: Tim Bunce Message-Id: <9511152119.AA08777@toad.ig.co.uk> To: dbi-users@fugue.com, rsnyder@mobster.cit.ge.com Subject: Re: Oracle DBD installation X-Sun-Charset: US-ASCII content-length: 570 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Bob Snyder > > This may sound stupid, but is it possible to configure/compile the Oracle DBD > without having the Oracle software installed? Nope. At least not without more effort than you'd probably want to contemplate. > If Oracle is required, I'd appreciate someone telling me exactly what I need, > or if they have any better ideas. I have a good bit of web and CGI knowledge, > a fair bit of perl knowledge, and very little Oracle knowledge. Follow what the Makefile.PL does. Basically lots of headers and libs. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04617-2@oink>; Wed, 15 Nov 1995 23:18:00 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816473502:09399:0; Wed, 15 Nov 95 22:11:42 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa09394; 15 Nov 95 22:11 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa02746; 15 Nov 95 22:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA19621 for dbi-dev-real; Wed, 15 Nov 1995 12:49:09 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA19617 for ; Wed, 15 Nov 1995 12:49:02 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aj08539; 15 Nov 95 20:43 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ab01049; 15 Nov 95 20:43 GMT Received: from toad by oink with SMTP (PP) id <03373-0@oink>; Wed, 15 Nov 1995 20:37:24 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA08722; Wed, 15 Nov 1995 20:37:01 +0000 Date: Wed, 15 Nov 1995 20:37:01 +0000 From: Tim Bunce Message-Id: <9511152037.AA08722@toad.ig.co.uk> To: dbi-dev@fugue.com, mouring@netnet.net Subject: Re: Passing information from dbh to sth. X-Sun-Charset: US-ASCII content-length: 1070 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Ben Lindstrom > > I must be really losing it. =) I could have swore I send off a question > about how to migrate data from dbh to sth (mainly the database connection). > > I found: imp_dbh_from_sth (I think that's what it's call I don't have my > code infront of me.) But when I track it down I get to *child() function > that I can't find. > > Could so one point me in the direction of how the migration for the database > information gets from dbh to sth? Er, you don't. The sth contains a pointer to the dbh. Anyway, that's all hidden by macros. The D_imp_dbh_from_sth macro is the one you want (I think, your message isn't really very clear). If you really want to know how it _currently_ works then study DBIXS.h: #define D_imp_dbh_from_sth D_imp_from_child(imp_dbh, imp_dbh_t, imp_sth) #define D_imp_from_child(name,type,child) \ type *name = (type*)(DBIc_PARENT_COM(child)) #define DBIc_PARENT_COM(imp) _imp2com(imp, std.parent_com) #define _imp2com(p,f) ((p)->com.f) Macros'R'us :-) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04617-4@oink>; Wed, 15 Nov 1995 23:18:09 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816473535:09699:0; Wed, 15 Nov 95 22:12:15 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa09695; 15 Nov 95 22:12 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa02839; 15 Nov 95 22:12 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA19534 for dbi-users-real; Wed, 15 Nov 1995 12:22:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA19530 for ; Wed, 15 Nov 1995 12:22:05 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.ab06056; 15 Nov 95 20:07 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ac21671; 15 Nov 95 20:06 GMT Received: from toad by oink with SMTP (PP) id <02779-0@oink>; Wed, 15 Nov 1995 20:05:56 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA08597; Wed, 15 Nov 1995 20:05:32 +0000 Date: Wed, 15 Nov 1995 20:05:32 +0000 From: Tim Bunce Message-Id: <9511152005.AA08597@toad.ig.co.uk> To: dbi-users@fugue.com, mrlizard@crl.com Subject: Re: Error codes and such X-Sun-Charset: US-ASCII content-length: 589 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Brian Valente > > In the meantime, I keep running into problems with getting a null value > returned from a $dbh->prepare($querystring) statement. Is there a way to > find out what went wrong? Your driver should put useful info in $DBI::err and $DBI::errstr > Speaking of which, is there any documentation for error codes and > handling _in general_? It'd sure help... Currently values in $DBI::err are driver specific. It's probable that a $DBI::state variable based on the SQL 'SQLSTATE' variable with defined values will be added sometime soon. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04617-11@oink>; Wed, 15 Nov 1995 23:18:36 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816474846:21691:0; Wed, 15 Nov 95 22:34:06 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa21687; 15 Nov 95 22:34 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa04228; 15 Nov 95 22:33 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA19806 for dbi-users-real; Wed, 15 Nov 1995 13:17:39 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA19802 for ; Wed, 15 Nov 1995 13:17:30 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.ae09167; 15 Nov 95 21:18 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ad08099; 15 Nov 95 21:16 GMT Received: from toad by oink with SMTP (PP) id <03668-0@oink>; Wed, 15 Nov 1995 21:12:11 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA08760; Wed, 15 Nov 1995 21:11:48 +0000 Date: Wed, 15 Nov 1995 21:11:48 +0000 From: Tim Bunce Message-Id: <9511152111.AA08760@toad.ig.co.uk> To: alan@lims.gel.com, KStock@auspex.com Subject: Re: I know I'm cheating ... Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 1338 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: KStock@auspex.com (Kevin Stock) > > > From: alan.johnson@gel.com (Alan Johnson) > > > > but I'm coming to the horse himself :-) You can swat me (like a horse does > > :-)) or ... > > Tim's been called a lot of things, most of them by me :-) but that's a > new one... :-) > > Does ora_do (oraperl 4) return the return status of the PL/SQL procedure > > called? I.e., if I write: > > > > $sql_proc = "begin if blah then return(-100) else return (0) end if; end;"; > > > > $dummy = &ora_do($lda,$sql_proc); > > > > does $dummy have the returned value of the procedure called (i.e. -100 or 0) > > > > It doesn't appear to, but I'm just checking. > > Oraperl 4 doesn't know anything about PL/SQL. All right, _I_ didn't > know anything about PL/SQL when I wrote Oraperl :-). &ora_do returns > the number of rows affected, unless it's 0, in which case it returns > OK (which tests as true but hase a numerical value of 0). > > I think you'd need to store the return value in the database and then > select it. Or, perhaps there's some way of making the procedure appear > to fail with some particular error status which would be made available > in $ora_errno... Maybe raise_error (or raise_exception?) would work for that. I'd be happy to tweak DBD::Oracle if needed but I don't have time to look into it right now. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04617-14@oink>; Wed, 15 Nov 1995 23:18:45 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816476285:04196:1; Wed, 15 Nov 95 22:58:05 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa02911; 15 Nov 95 22:58 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA19888 for dbi-users-real; Wed, 15 Nov 1995 13:30:16 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA19884 for ; Wed, 15 Nov 1995 13:30:11 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab08649; 15 Nov 95 21:17 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa08099; 15 Nov 95 21:16 GMT Received: from toad by oink with SMTP (PP) id <03620-0@oink>; Wed, 15 Nov 1995 20:59:35 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA08751; Wed, 15 Nov 1995 20:59:11 +0000 Date: Wed, 15 Nov 1995 20:59:11 +0000 From: Tim Bunce Message-Id: <9511152059.AA08751@toad.ig.co.uk> To: dbi-users@fugue.com, phj@pilot.njin.net Subject: Re: ofetch out of sequence (was deletion weirdnesses) X-Sun-Charset: US-ASCII content-length: 792 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: "Patricia H. Ju" > > Rokey. > > This is what's happening: > > query: DELETE ARTICLE WHERE AREAID = 10 at Table.pm line 360. > ERROR EVENT: DBI::st=HASH(0x1a8af0)->DBI::default_handler: > 1002 ORA-01002: fetch out of sequence (DBD: ofetch error) > at Table.pm line 362 > > Mind you, the articles I'm deleting have no dependenceies other than > referrring to the areaID of table Area. So I figured if I delete all > the articles that are of area 10, *then* delete area 10, I should > be ok, right ? Apparently not. The fetch out of sequence is more of an internal issue than anything else. Did you use $dbh->do("...") to execute this? If not then you should. If you did then please post a real code fragment and $dbh->debug(2); trace log. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <05455-0@oink>; Thu, 16 Nov 1995 01:47:13 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816481905:08712:1; Thu, 16 Nov 95 00:31:45 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa08490; 16 Nov 95 0:31 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA20438 for perldb-interest-real; Wed, 15 Nov 1995 15:12:26 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA20434 for ; Wed, 15 Nov 1995 15:12:24 -0800 Received: by gw.home.vix.com id AA29297; Wed, 15 Nov 95 15:14:48 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from dn800e0.fingerhut.com (daemon@localhost) by dn800e0.fingerhut.com (8.6.12/8.6.12) with ESMTP id RAA14415; Wed, 15 Nov 1995 17:00:14 -0600 Received: from seag.fingerhut.com (GF007E0.SEAG.FINGERHUT.COM [151.210.140.7]) by dn800e0.fingerhut.com (8.6.12/8.6.12) with SMTP id RAA14411; Wed, 15 Nov 1995 17:00:14 -0600 Received: from g0084.fingerhut.com. by seag.fingerhut.com (5.0/SMI-SVR4) id AA01065; Wed, 15 Nov 1995 17:14:05 -0600 Received: by g0084.fingerhut.com. (5.x/SMI-SVR4) id AA07203; Wed, 15 Nov 1995 17:14:01 -0600 Date: Wed, 15 Nov 1995 17:14:01 -0600 From: Bruce.Albrecht@seag.fingerhut.com (Bruce Albrecht) Message-Id: <9511152314.AA07203@g0084.fingerhut.com.> To: dbi-users@vix.com Cc: Tim.Bunce@ig.co.uk Subject: Scripts dump core whenever I get an oracle error Content-Length: 2278 I'm running Perl-5.001m, DBD-Oracle0.25, DBI-0.62 (or DBI-0.64), and my script uses Oraperl.pm. It seems like I always get a core dump whenever there's an error in my ora_open statement. I got different results from DBI-0.62 and DBI-0.64: DBI-0.62 displayed the error text before dumping core, and DBI-0.64 just dumped core. The following stack frames were from the same script, and the perl binary was identical, just the location of the Perl libraries and the DBI interface was changed. Is this a known problem (perhaps even fixed), a new problem, or did I install this incorrectly? DBI-0.62: (dbx) where =>[1] XS_DBD__Oracle__st__prepare(0x2700eaa8, 0x46, 0x1efc88, 0x142f98, 0x4, 0x0), at 0xef42e8f8 [2] pp_entersub(0xab410, 0x1, 0x13eaa8, 0xa50b4, 0x0, 0x0), at 0x4e83c [3] run(0xf5548, 0xfffff7ff, 0x14aa88, 0x0, 0xffffffff, 0xf5548), at 0x49204 [4] perl_call_sv(0xf5548, 0x0, 0x40, 0x40, 0x42, 0x0), at 0x1da38 [5] XS_DBI_dispatch(0x0, 0x143a98, 0xef5e6d9c, 0xef5e6d84, 0xb7408, 0x14b1ac), at 0xef5e46c8 [6] pp_entersub(0xab410, 0x1, 0x144eb4, 0xa50b4, 0x0, 0x0), at 0x4e83c [7] run(0x0, 0x1d188, 0x153fa8, 0x153fa8, 0xa8800, 0x0), at 0x49204 [8] perl_run(0x0, 0x0, 0x0, 0xa2000, 0x0, 0xa2000), at 0x1d28c [9] main(0x4, 0xeffff24c, 0xeffff260, 0xa7c00, 0x0, 0x0), at 0x1bd5c DBI-0.64: (dbx) where =>[1] sv_setiv(0x1bf0, 0x388, 0x6d650a00, 0x7efefeff, 0x81010100, 0xff00), at 0x502c0 [2] ora_error(0x142fa4, 0x145a48, 0x388, 0xefffec44, 0x145908, 0xd5050), at 0xef42fb94 [3] dbd_st_prepare(0x142fa4, 0x1, 0x0, 0x0, 0x145a08, 0x40000), at 0xef4305fc [4] XS_DBD__Oracle__st__prepare(0x2700eab4, 0x46, 0x1ef588, 0x142fa4, 0x4, 0x0), at 0xef42e8e8 [5] pp_entersub(0xab410, 0x1, 0x13eab4, 0xa50b4, 0x0, 0x0), at 0x4e83c [6] run(0xf5748, 0xfffff7ff, 0x14be88, 0x0, 0xffffffff, 0xf5748), at 0x49204 [7] perl_call_sv(0xf5748, 0x0, 0x40, 0x40, 0x42, 0x0), at 0x1da38 [8] XS_DBI_dispatch(0x0, 0x143a78, 0xef5e704c, 0xef5e7034, 0xb7408, 0x14b1b8), at 0xef5e4860 [9] pp_entersub(0xab410, 0x1, 0x144ec0, 0xa50b4, 0x0, 0x0), at 0x4e83c [10] run(0x0, 0x1d188, 0x154728, 0x154728, 0xa8800, 0x0), at 0x49204 [11] perl_run(0x0, 0x0, 0x0, 0xa2000, 0x0, 0xa2000), at 0x1d28c [12] main(0x4, 0xeffff204, 0xeffff218, 0xa7c00, 0x0, 0x0), at 0x1bd5c ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05456-9@oink>; Thu, 16 Nov 1995 01:47:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816483365:04080:1; Thu, 16 Nov 95 00:56:05 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa03855; 16 Nov 95 0:56 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA20568 for dbi-users-real; Wed, 15 Nov 1995 15:53:59 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA20564 for ; Wed, 15 Nov 1995 15:53:55 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ac13441; 15 Nov 95 23:50 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.aa09828; 15 Nov 95 23:49 GMT Received: from toad by oink with SMTP (PP) id <05092-0@oink>; Wed, 15 Nov 1995 23:49:01 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA09763; Wed, 15 Nov 1995 23:48:28 +0000 Date: Wed, 15 Nov 1995 23:48:28 +0000 From: Tim Bunce Message-Id: <9511152348.AA09763@toad.ig.co.uk> To: dbi-users@fugue.com Subject: Vacation! Cc: perl5-porters@nicoh.com X-Sun-Charset: US-ASCII content-length: 700 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 I'm going to be away for the next two weeks (18th Nov thru 3rd Dec). I'm going to sit on a beach in southern Portugal and read and relax (and have nothing to do with computers or perl*) I hope to make a new DBI & DBD release before I go (since ftp.demon is still down I'll probably send it directly into the CPAN archives). The Oraperl.pm included will be the last version before it goes beta. Tim. * Actually I will spend some time on ODBC / SQL CLI issues. We need to establish a strategy soon. A DBD::ODBC looks very probable. The DBI/DBD interface will almost certainly change. It's a thorny area. I need time to think. CC'd to perl5-porters to save me writing the same thing again there. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05456-11@oink>; Thu, 16 Nov 1995 01:48:01 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816483461:04569:1; Thu, 16 Nov 95 00:57:41 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa04248; 16 Nov 95 0:56 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA20530 for dbi-users-real; Wed, 15 Nov 1995 15:47:43 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA20526 for ; Wed, 15 Nov 1995 15:47:33 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aa12571; 15 Nov 95 23:49 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.aa09614; 15 Nov 95 23:48 GMT Received: from toad by oink with SMTP (PP) id <05041-0@oink>; Wed, 15 Nov 1995 23:38:53 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA09760; Wed, 15 Nov 1995 23:38:30 +0000 Date: Wed, 15 Nov 1995 23:38:30 +0000 From: Tim Bunce Message-Id: <9511152338.AA09760@toad.ig.co.uk> To: dbi-users@fugue.com, Bruce.Albrecht@seag.fingerhut.com Subject: Re: Scripts dump core whenever I get an oracle error X-Sun-Charset: US-ASCII content-length: 2791 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Bruce.Albrecht@seag.fingerhut.com (Bruce Albrecht) > To: perldb-interest@vix.com Please start using the new dbi-users@fugue.com address. > I'm running Perl-5.001m, DBD-Oracle0.25, DBI-0.62 (or DBI-0.64), and > my script uses Oraperl.pm. It seems like I always get a core dump > whenever there's an error in my ora_open statement. I got different > results from DBI-0.62 and DBI-0.64: DBI-0.62 displayed the error text > before dumping core, and DBI-0.64 just dumped core. Be aware that it's generally a very bad idea to mix-and-match pre-compiled versions of DBI's and DBD's. To be safe a DBD should only be used with the DBI that was installed when it was compiled. > The following > stack frames were from the same script, and the perl binary was > identical, just the location of the Perl libraries and the DBI > interface was changed. Is this a known problem (perhaps even fixed), > a new problem, or did I install this incorrectly? > DBI-0.64: > > (dbx) where > =>[1] sv_setiv(0x1bf0, 0x388, 0x6d650a00, 0x7efefeff, 0x81010100, 0xff00), at 0x502c0 > [2] ora_error(0x142fa4, 0x145a48, 0x388, 0xefffec44, 0x145908, 0xd5050), at 0xef42fb94 > [3] dbd_st_prepare(0x142fa4, 0x1, 0x0, 0x0, 0x145a08, 0x40000), at 0xef4305fc This is much more helpful than most people who report a core dump without a stack trace. Thanks. (Recompiling the DBD::Oracle module with -g and without -O would show the line numbers but in this case it doesn't matter since there's only one sv_setiv in ora_error) Please rebuild DBI-0.64 (with -g) and install it. Then rebuild DBD::Oracle (with -g) using the newly installed DBI. The problem maybe that the string is too long. I've recently changed the code so it can deal with strings of any length. Try this: void ora_error(h, lda, rc, what) SV *h; Lda_Def *lda; int rc; char *what; { D_imp_xxh(h); sb2 rc_sb2 = rc; /* stop GCC complaining about arg2 of oerhms() */ SV *errstr = DBIc_ERRSTR(imp_xxh); if (lda) { /* is oracle error (allow for non-oracle errors) */ int len; char msg[1024]; /* Oracle oerhms can do duplicate free if connect fails */ oerhms(lda, rc_sb2, (text*)msg, sizeof(msg)); len = strlen(msg); if (len && msg[len-1] == '\n') msg[len-1] = '\0'; /* trim off \n from end of message */ sv_setpv(errstr, (char*)msg); } else sv_setpv(errstr, what); sv_setiv(DBIc_ERR(imp_xxh), (IV)rc); if (what && lda) { sv_catpv(errstr, " (DBD: "); sv_catpv(errstr, what); sv_catpv(errstr, ")"); } DBIh_EVENT2(h, ERROR_event, DBIc_ERR(imp_xxh), errstr); if (dbis->debug >= 2) fprintf(DBILOGFP, "%s error %d recorded: %s\n", what, rc, SvPV(errstr,na)); } Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <06320-4@oink>; Thu, 16 Nov 1995 03:14:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816490094:16585:1; Thu, 16 Nov 95 02:48:14 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa16514; 16 Nov 95 2:48 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA21152 for dbi-dev-real; Wed, 15 Nov 1995 17:46:00 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA21148 for ; Wed, 15 Nov 1995 17:45:54 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab08721; 16 Nov 95 1:47 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa25406; 16 Nov 95 1:47 GMT Received: from toad by oink with SMTP (PP) id <05315-0@oink>; Thu, 16 Nov 1995 00:41:16 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA09893; Thu, 16 Nov 1995 00:40:53 +0000 Date: Thu, 16 Nov 1995 00:40:53 +0000 From: Tim Bunce Message-Id: <9511160040.AA09893@toad.ig.co.uk> To: dbi-dev@fugue.com, mhm@austin.ibm.com Subject: Re: Blobs are now working on DB2perl X-Sun-Charset: US-ASCII content-length: 1687 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: mhm@austin.ibm.com > > A few tweaks in DBD::DB2 and big correction in my perl script and > everyting works fine. > > A while back, I asked about other databases abilities to read a blob to > a "file" and Tim suggested something like > > $sth->copy_blob_to_file(1, $open_file_handle); > > I thought about it for awhile and would like to suggest the following: > > $sth->readblob_to_file(field, file_or_device_name); > > First, the above is (atleast for DB2) more semantically correct, in that > the DB2 engine will be doing al the I/O not perl. And secondly, it > would allow the databases that support direct engine I/O to work without > wasting cycles and file handles. > > Any serious downside that I'm missing? Nope. I've renamed readblob to blob_read in anticipation of other blob_* methods and implemented a default blob_copy_to_file using blob_read. (A readblob stub is provided for old code as a migration aid. It will complain once but still call blob_read for you.) blob_copy_to_file will copy to either a filename or a filehandle passed as a globref: sub blob_copy_to_file { # returns length or undef on error my($self, $field, $filename_or_handleref, $blocksize) = @_; my $fh = $filename_or_handleref; my($len, $buf) = (0, ""); $blocksize ||= 512; # not too ambitious local(*FH); unless(ref $fh) { open(FH, ">$fh") || return undef; $fh = \*FH; } while(defined($self->blob_read($field, $len, $blocksize, \$buf)) { print $fh $buf; $len += length $buf; } close(FH); $len; } Okay? (speak soon :-) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <07018-1@oink>; Thu, 16 Nov 1995 07:06:00 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816492876:29697:1; Thu, 16 Nov 95 03:34:36 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa29652; 16 Nov 95 3:34 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA21360 for dbi-dev-real; Wed, 15 Nov 1995 18:39:32 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA21356 for ; Wed, 15 Nov 1995 18:39:28 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id UAA43026 for ; Wed, 15 Nov 1995 20:41:57 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dbi-dev@fugue.com at austin.ibm.com; id AA04158; Wed, 15 Nov 1995 20:41:54 -0600 Message-Id: <9511160241.AA04158@dax.austin.ibm.com> To: dbi-dev@fugue.com Subject: Re: Blobs are now working on DB2perl In-Reply-To: Your message of Thu, 16 Nov 95 00:40:53 +0000. <9511160040.AA09893@toad.ig.co.uk> Date: Wed, 15 Nov 95 20:41:54 -0600 Tim Bunce writes: |> > From: mhm@austin.ibm.com |> > it would allow the databases that support direct engine I/O to work |> > without wasting cycles and file handles. |> > |> |> Nope. |> |> I've renamed readblob to blob_read in anticipation of other blob_* methods |> and implemented a default blob_copy_to_file using blob_read. |> |> (A readblob stub is provided for old code as a migration aid. It will |> complain once but still call blob_read for you.) |> |> blob_copy_to_file will copy to either a filename or a filehandle |> passed as a globref: |> |> sub blob_copy_to_file { # returns length or undef on error |> my($self, $field, $filename_or_handleref, $blocksize) = @_; |> my $fh = $filename_or_handleref; |> my($len, $buf) = (0, ""); |> $blocksize ||= 512; # not too ambitious |> local(*FH); |> unless(ref $fh) { |> open(FH, ">$fh") || return undef; |> $fh = \*FH; |> } |> while(defined($self->blob_read($field, $len, $blocksize, \$buf)) { |> print $fh $buf; |> $len += length $buf; |> } |> close(FH); |> $len; |> } |> |> Okay? (speak soon :-) Well.....I think we are having paradigm difficulties. From my point of view, blob_copy_to_file is closer to blob_fetch_by_user_defined_block and what I want is blob_straight_to_file_bypassing_perl. Or more compactly, DB2 provides a method of reading in user defined chunks of a blob and a method of slurping the whole blob in one read. So for me with DB2, I hate to be provided arguments for which I have no need or use (block size). Let me sleep on it, as I am somewhat scatter brained between the hour, the cold and the medication. Mike ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <10226-0@oink>; Thu, 16 Nov 1995 16:42:29 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816533176:09168:1; Thu, 16 Nov 95 14:46:16 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa08921; 16 Nov 95 14:46 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id EAA23426 for dbi-dev-real; Thu, 16 Nov 1995 04:43:38 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id EAA23422 for ; Thu, 16 Nov 1995 04:43:36 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id GAA45496 for ; Thu, 16 Nov 1995 06:46:05 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dbi-dev@fugue.com at austin.ibm.com; id AA21326; Thu, 16 Nov 1995 06:46:04 -0600 Message-Id: <9511161246.AA21326@dax.austin.ibm.com> To: dbi-dev@fugue.com Subject: Re: Error codes and such In-Reply-To: Your message of Wed, 15 Nov 95 20:05:32 +0000. <9511152005.AA08597@toad.ig.co.uk> Date: Thu, 16 Nov 95 06:46:04 -0600 Tim Bunce writes: |> |>It's probable that a $DBI::state variable based on the SQL 'SQLSTATE' |>variable with defined values will be added sometime soon. |> Thanks, that will make life with DB2 much easier. Mike ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <10226-3@oink>; Thu, 16 Nov 1995 16:42:47 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816534186:21290:0; Thu, 16 Nov 95 15:03:06 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id aa21225; 16 Nov 95 15:03 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Thu, 16 Nov 1995 14:58:22 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA23457 for dbi-dev-real; Thu, 16 Nov 1995 05:07:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA23453 for ; Thu, 16 Nov 1995 05:07:13 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA29338 for ; Thu, 16 Nov 1995 07:09:42 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dbi-dev@fugue.com at austin.ibm.com; id AA22646; Thu, 16 Nov 1995 07:09:42 -0600 Message-Id: <9511161309.AA22646@dax.austin.ibm.com> To: dbi-dev@fugue.com Subject: 2nd attempt at commenting on blob_copy_to_file In-Reply-To: Your message of Thu, 16 Nov 95 00:40:53 +0000. <9511160040.AA09893@toad.ig.co.uk> Date: Thu, 16 Nov 95 07:09:41 -0600 After reading what I posted last night, I apologize for being way too scatter brained.......So lets try again. Tim Bunce writes: |> |>I've renamed readblob to blob_read in anticipation of other blob_* |>methods and implemented a default blob_copy_to_file using blob_read. |> |>(A readblob stub is provided for old code as a migration aid. It will |>complain once but still call blob_read for you.) |> |>blob_copy_to_file will copy to either a filename or a filehandle |>passed as a globref: |> |> sub blob_copy_to_file { # returns length or undef on error |> my($self, $field, $filename_or_handleref, $blocksize) = @_; |> my $fh = $filename_or_handleref; |> my($len, $buf) = (0, ""); |> $blocksize ||= 512; # not too ambitious |> local(*FH); |> unless(ref $fh) { |> open(FH, ">$fh") || return undef; |> $fh = \*FH; |> } |> while(defined($self->blob_read($field, $len, $blocksize, \$buf)){ |> print $fh $buf; |> $len += length $buf; |> } |> close(FH); |> $len; |> } Ok, the problem I have with this implementation is the possibility of getting a reference to an open file handle. With DB2, I have the ability to bind the incoming blob column directly to a "file", thus by passing any I/O by perl and requiring only one fetch from the database. I guess the next questions are: When perl opens a file is it opened in O_EXCL mode? If not, can I retrieve the original file name if I have the reference? Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12918-4@oink>; Thu, 16 Nov 1995 20:51:05 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816554884:04735:0; Thu, 16 Nov 95 20:48:04 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa04640; 16 Nov 95 20:48 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA24157 for dbi-dev-real; Thu, 16 Nov 1995 11:04:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA24153 for ; Thu, 16 Nov 1995 11:04:55 -0800 Received: by gw.home.vix.com id AA21566; Thu, 16 Nov 95 11:07:19 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ae29802; 16 Nov 95 19:01 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.aa05855; 16 Nov 95 19:00 GMT Received: from toad by oink with SMTP (PP) id <11200-0@oink>; Thu, 16 Nov 1995 17:30:18 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA12274; Thu, 16 Nov 1995 17:29:54 +0000 Date: Thu, 16 Nov 1995 17:29:54 +0000 From: Tim Bunce Message-Id: <9511161729.AA12274@toad.ig.co.uk> To: dbi-dev@fugue.com, mhm@austin.ibm.com Subject: Re: 2nd attempt at commenting on blob_copy_to_file X-Sun-Charset: US-ASCII Content-Length: 1102 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: mhm@austin.ibm.com > > |> sub blob_copy_to_file { # returns length or undef on error > |> my($self, $field, $filename_or_handleref, $blocksize) = @_; > > Ok, the problem I have with this implementation is the possibility of > getting a reference to an open file handle. With DB2, I have the > ability to bind the incoming blob column directly to a "file", thus > by passing any I/O by perl and requiring only one fetch from the > database. Ah, so what you want is something like an extended bind_column() that establishes the fetch-into-file mechanism. Umm. Is that part of ODBC? Note that the blob_copy_to_file is a default 'fallback' method. I'd expect you'd probably implement a better one in DBD::DB2 (but keeping the same interface). > I guess the next questions are: > > When perl opens a file is it opened in O_EXCL mode? Nope (currently you need to do fancy stuff with POSIX for that) > If not, can I retrieve the original file name if I have the reference? Yes, if you saved the name as part of the glob being referenced. > Mike > mhm@austin.ibm.com Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13726-2@oink>; Thu, 16 Nov 1995 22:27:25 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816560089:22683:0; Thu, 16 Nov 95 22:14:49 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa22678; 16 Nov 95 22:14 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id sg.aa14816; 16 Nov 95 22:14 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Thu, 16 Nov 1995 22:09:32 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA24311 for dbi-dev-real; Thu, 16 Nov 1995 12:08:11 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA24307 for ; Thu, 16 Nov 1995 12:08:07 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id OAA12747; Thu, 16 Nov 1995 14:10:27 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dbi-dev@fugue.com at austin.ibm.com; id AA18202; Thu, 16 Nov 1995 14:10:24 -0600 Message-Id: <9511162010.AA18202@dax.austin.ibm.com> To: Tim Bunce Cc: dbi-dev@fugue.com, mhm@austin.ibm.com Subject: Re: 2nd attempt at commenting on blob_copy_to_file In-Reply-To: Your message of Thu, 16 Nov 95 17:29:54 +0000. <9511161729.AA12274@toad.ig.co.uk> Date: Thu, 16 Nov 95 14:10:23 -0600 Tim Bunce writes: |>> From: mhm@austin.ibm.com |>> |>> |> sub blob_copy_to_file { # returns length or undef on error |>> |> my($self, $field, $filename_or_handleref, $blocksize) = @_; |>> |>> Ok, the problem I have with this implementation is the possibility of |>> getting a reference to an open file handle. With DB2, I have the |>> ability to bind the incoming blob column directly to a "file", thus |>> by passing any I/O by perl and requiring only one fetch from the |>> database. |> |>Ah, so what you want is something like an extended bind_column() that |>establishes the fetch-into-file mechanism. Umm. What I want is to be able to use the DB2 CLI mechanisms to do it for me. In essence, I would like to be able to do the following sub blob_copy_to_file { my($self, $field, $filename, $blocksize) = @_; my ($len); $len = $self->read_blob_to_device($field, $filename, $blocksize); } dbd_blob2dev(IV field, SV *name, IV blksz) { int rt_len; SQLBindFileToCol(imp_sth->phstmt,field,SvPVx(name->sv), arg4,SQL_FILE_OVERWRITE,MAX_NAME_LENGTH,&rt_len,&blob_locator); SQLFetch(imp_sth->phstmt); return (rt_len); } Granted the above doesn't contain everything (no error checking and I sure I don't have the all the correct magic) but merely the gist of what I would like to be able to do. |>Is that part of ODBC? Not yet..... |>Note that the blob_copy_to_file is a default 'fallback' method. |>I'd expect you'd probably implement a better one in DBD::DB2 (but |>keeping the same interface). |> |>> I guess the next questions are: |>> |>> When perl opens a file is it opened in O_EXCL mode? Ok, that implies that if I was handed a ref to an open FH I could have the database engine reopen it and dump the blob in question. |>Nope (currently you need to do fancy stuff with POSIX for that) |> |>> If not, can I retrieve the original file name if I have the reference? |> |>Yes, if you saved the name as part of the glob being referenced. Good, now all I have to do is figure out how to save the name as part of the glob....unfortunately, following all the discussions on globs has been very low on the ToDo list. So is it safe to assume the new version will be out sometime on the 17th? Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <14010-0@oink>; Thu, 16 Nov 1995 22:57:37 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816561544:05762:0; Thu, 16 Nov 95 22:39:04 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa05746; 16 Nov 95 22:39 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Thu, 16 Nov 1995 22:35:46 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA24493 for dbi-dev-real; Thu, 16 Nov 1995 12:44:06 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA24489 for ; Thu, 16 Nov 1995 12:44:00 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.ai09414; 16 Nov 95 20:38 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa01945; 16 Nov 95 20:38 GMT Received: from toad by oink with SMTP (PP) id <12663-0@oink>; Thu, 16 Nov 1995 20:23:31 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA13110; Thu, 16 Nov 1995 20:23:03 +0000 Date: Thu, 16 Nov 1995 20:23:03 +0000 From: Tim Bunce Message-Id: <9511162023.AA13110@toad.ig.co.uk> To: mhm@austin.ibm.com Subject: Re: 2nd attempt at commenting on blob_copy_to_file Cc: dbi-dev@fugue.com X-Sun-Charset: US-ASCII content-length: 788 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: mhm@austin.ibm.com > > SQLBindFileToCol(imp_sth->phstmt,field,SvPVx(name->sv), > arg4,SQL_FILE_OVERWRITE,MAX_NAME_LENGTH,&rt_len,&blob_locator); Thanks. > |>> If not, can I retrieve the original file name if I have the reference? > |> > |>Yes, if you saved the name as part of the glob being referenced. > > Good, now all I have to do is figure out how to save the name as part > of the glob....unfortunately, following all the discussions on globs has > been very low on the ToDo list. $FH = "/etc/motd"; @FH = (...); %FH = (...); open(FH, $FH); $fhref = \*FH; ... print $$fhref; > So is it safe to assume the new version will be out sometime on the > 17th? If it's not then it won't appear for a couple of weeks! (Thanks for the reminder! :-) Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <15351-7@oink>; Fri, 17 Nov 1995 07:06:34 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816573133:28424:0; Fri, 17 Nov 95 01:52:13 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa28298; 17 Nov 95 1:52 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA25059 for dbi-announce-real; Thu, 16 Nov 1995 16:05:32 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA25050; Thu, 16 Nov 1995 16:05:19 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.af01057; 17 Nov 95 0:04 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa17722; 17 Nov 95 0:03 GMT Received: from toad by oink with SMTP (PP) id <14505-0@oink>; Fri, 17 Nov 1995 00:03:07 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA14346; Fri, 17 Nov 1995 00:02:38 +0000 Date: Fri, 17 Nov 1995 00:02:38 +0000 From: Tim Bunce Message-Id: <9511170002.AA14346@toad.ig.co.uk> To: dbi-announce@fugue.com Subject: DBI-0.65.tar.Z and DBD-Oracle-0.27.tar.Z Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 4509 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 New DBI and DBD::Oracle releases: --- Changes in DBI 0.65, 16 Nov 1995 Added $DBI::state to hold SQL CLI / ODBC SQLSTATE value. SQLSTATE "00000" (success) is returned as "" (false), all else is true. If a driver does not explicitly initialise it (via $h->{State} or DBIc_STATE(imp_xxh) then $DBI::state will automatically return "" if $DBI::err is false otherwise "S1000" (general error). As always, this is a new feature and liable to change. The is *no longer* a default error handler! You can add your own using push(@{$h->{Handlers}}, sub { ... }) but be aware that this interface may change (or go away). The DBI now automatically clears $DBI::err, errstr and state before calling most DBI methods. Previously error conditions would persist. Added DBIh_CLEAR_ERROR(imp_xxh) macro. DBI now EXPORT_OK's some utility functions, neat($value), neat_list(@values) and dump_results($sth). Slightly enhanced t/min.t minimal test script in an effort to help narrow down the few stray core dumps that some porters still report. Renamed readblob to blob_read (old name still works but warns). Added default blob_copy_to_file method. Added $sth = $dbh->tables method. This returns an $sth for a query which has these columns: TABLE_CATALOGUE, TABLE_OWNER, TABLE_NAME, TABLE_TYPE, REMARKS in that order. The TABLE_CATALOGUE column should be ignored for now. Changes in DBI 0.64, 23 Oct 1995 Fixed 'disconnect invalidates 1 associated cursor(s)' problem. Drivers using DBIc_ACTIVE_on/off() macros should not need any changes other than to test for DBIc_ACTIVE_KIDS() instead of DBIc_KIDS(). Fixed possible core dump in dbih_clearcom during global destruction. --- Changes in DBD::Oracle 0.27, 16 Nov 1995: Improved oraperl compatibility in assorted ways. Added the original oraperl manual to Oraperl.pm as pod. (perldoc Oraperl will display the manual). The manual highlights remaining differences in the emulation. Added original oraperl examples into oraperl.ex directory. Improved error reporting in test.pl. Removed some internal limits. Renamed readblob to blob_read and fixed return value. Implemented $sth = $dbh->tables method (see DBI changes file). Uses $Config{archlibexp} not $Config{archlib} in makefile.PL Requires DBI-0.65. I've still not got around to implementing a row cache (array fetch) but it's next on my list. This should be the last alpha release of the oraperl emulation. (The DBD::Oracle module itself will remain alpha for awhile yet.) Changes in DBD::Oracle 0.26, 23 Oct 1995: Removed error message (sql) size limits in parse and ora_error. Fixed FETCH NUM_OF_PARAMS so it doesn't trigger a describe. The ora_lengths and ora_types attributes work in non-oraperl mode. --- These files are *NOT* available from ftp.demon.co.uk because the system is still down (long story, don't ask). The files have been uploaded to the Perl Module Server site from where they will be replicated into the CPAN archives: CPAN master site ftp://ftp.funet.fi/pub/languages/perl/CPAN/ Africa ftp://ftp.is.co.za/programming/perl/CPAN/ Asia ftp://dongpo.math.ncu.edu.tw/perl/CPAN ftp://ftp.lab.kdd.co.jp/lang/perl/CPAN/ Australasia ftp://coombs.anu.edu.au/pub/perl/ ftp://ftp.mame.mu.oz.au/pub/perl/CPAN/ Europe ftp://ftp.ci.uminho.pt/pub/lang/perl/ ftp://ftp.cs.ruu.nl/pub/PERL/CPAN/ ftp://ftp.demon.co.uk/pub/mirrors/perl/CPAN/ ftp://ftp.pasteur.fr/pub/computing/unix/perl/CPAN/ ftp://ftp.rz.ruhr-uni-bochum.de/pub/programming/languages/perl/CPAN/ ftp://ftp.sunet.se/pub/lang/perl/CPAN/ ftp://ftp.switch.ch/mirror/CPAN/ ftp://orpheu.ci.uminho.pt/pub/lang/perl/ North America ftp://ftp.cis.ufl.edu/pub/perl/CPAN/ ftp://ftp.delphi.com/pub/mirrors/packages/perl/CPAN/ ftp://ftp.sedl.org/pub/mirrors/CPAN/ ftp://ftp.sterling.com/programming/languages/perl/ ftp://ftp.uoknor.edu/mirrors/CPAN/ ftp://uiarchive.cso.uiuc.edu/pub/lang/perl/CPAN/ but it may take a day or so. Since I'm going to be away for two weeks, if this release has problems (or you have problems with it) then you'll just have to help each other. Enjoy. Tim. p.s. The DBI Changes file in the release has an incorrect 'Changes in...' line. Ignore it. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15352-21@oink>; Fri, 17 Nov 1995 07:08:56 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816573025:06572:0; Fri, 17 Nov 95 01:50:25 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa06565; 17 Nov 95 1:50 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Fri, 17 Nov 1995 01:50:15 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA25065 for dbi-users-real; Thu, 16 Nov 1995 16:08:44 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA25050; Thu, 16 Nov 1995 16:05:19 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.af01057; 17 Nov 95 0:04 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa17722; 17 Nov 95 0:03 GMT Received: from toad by oink with SMTP (PP) id <14505-0@oink>; Fri, 17 Nov 1995 00:03:07 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA14346; Fri, 17 Nov 1995 00:02:38 +0000 Date: Fri, 17 Nov 1995 00:02:38 +0000 From: Tim Bunce Message-Id: <9511170002.AA14346@toad.ig.co.uk> To: dbi-announce@fugue.com Subject: DBI-0.65.tar.Z and DBD-Oracle-0.27.tar.Z Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 4509 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 New DBI and DBD::Oracle releases: --- Changes in DBI 0.65, 16 Nov 1995 Added $DBI::state to hold SQL CLI / ODBC SQLSTATE value. SQLSTATE "00000" (success) is returned as "" (false), all else is true. If a driver does not explicitly initialise it (via $h->{State} or DBIc_STATE(imp_xxh) then $DBI::state will automatically return "" if $DBI::err is false otherwise "S1000" (general error). As always, this is a new feature and liable to change. The is *no longer* a default error handler! You can add your own using push(@{$h->{Handlers}}, sub { ... }) but be aware that this interface may change (or go away). The DBI now automatically clears $DBI::err, errstr and state before calling most DBI methods. Previously error conditions would persist. Added DBIh_CLEAR_ERROR(imp_xxh) macro. DBI now EXPORT_OK's some utility functions, neat($value), neat_list(@values) and dump_results($sth). Slightly enhanced t/min.t minimal test script in an effort to help narrow down the few stray core dumps that some porters still report. Renamed readblob to blob_read (old name still works but warns). Added default blob_copy_to_file method. Added $sth = $dbh->tables method. This returns an $sth for a query which has these columns: TABLE_CATALOGUE, TABLE_OWNER, TABLE_NAME, TABLE_TYPE, REMARKS in that order. The TABLE_CATALOGUE column should be ignored for now. Changes in DBI 0.64, 23 Oct 1995 Fixed 'disconnect invalidates 1 associated cursor(s)' problem. Drivers using DBIc_ACTIVE_on/off() macros should not need any changes other than to test for DBIc_ACTIVE_KIDS() instead of DBIc_KIDS(). Fixed possible core dump in dbih_clearcom during global destruction. --- Changes in DBD::Oracle 0.27, 16 Nov 1995: Improved oraperl compatibility in assorted ways. Added the original oraperl manual to Oraperl.pm as pod. (perldoc Oraperl will display the manual). The manual highlights remaining differences in the emulation. Added original oraperl examples into oraperl.ex directory. Improved error reporting in test.pl. Removed some internal limits. Renamed readblob to blob_read and fixed return value. Implemented $sth = $dbh->tables method (see DBI changes file). Uses $Config{archlibexp} not $Config{archlib} in makefile.PL Requires DBI-0.65. I've still not got around to implementing a row cache (array fetch) but it's next on my list. This should be the last alpha release of the oraperl emulation. (The DBD::Oracle module itself will remain alpha for awhile yet.) Changes in DBD::Oracle 0.26, 23 Oct 1995: Removed error message (sql) size limits in parse and ora_error. Fixed FETCH NUM_OF_PARAMS so it doesn't trigger a describe. The ora_lengths and ora_types attributes work in non-oraperl mode. --- These files are *NOT* available from ftp.demon.co.uk because the system is still down (long story, don't ask). The files have been uploaded to the Perl Module Server site from where they will be replicated into the CPAN archives: CPAN master site ftp://ftp.funet.fi/pub/languages/perl/CPAN/ Africa ftp://ftp.is.co.za/programming/perl/CPAN/ Asia ftp://dongpo.math.ncu.edu.tw/perl/CPAN ftp://ftp.lab.kdd.co.jp/lang/perl/CPAN/ Australasia ftp://coombs.anu.edu.au/pub/perl/ ftp://ftp.mame.mu.oz.au/pub/perl/CPAN/ Europe ftp://ftp.ci.uminho.pt/pub/lang/perl/ ftp://ftp.cs.ruu.nl/pub/PERL/CPAN/ ftp://ftp.demon.co.uk/pub/mirrors/perl/CPAN/ ftp://ftp.pasteur.fr/pub/computing/unix/perl/CPAN/ ftp://ftp.rz.ruhr-uni-bochum.de/pub/programming/languages/perl/CPAN/ ftp://ftp.sunet.se/pub/lang/perl/CPAN/ ftp://ftp.switch.ch/mirror/CPAN/ ftp://orpheu.ci.uminho.pt/pub/lang/perl/ North America ftp://ftp.cis.ufl.edu/pub/perl/CPAN/ ftp://ftp.delphi.com/pub/mirrors/packages/perl/CPAN/ ftp://ftp.sedl.org/pub/mirrors/CPAN/ ftp://ftp.sterling.com/programming/languages/perl/ ftp://ftp.uoknor.edu/mirrors/CPAN/ ftp://uiarchive.cso.uiuc.edu/pub/lang/perl/CPAN/ but it may take a day or so. Since I'm going to be away for two weeks, if this release has problems (or you have problems with it) then you'll just have to help each other. Enjoy. Tim. p.s. The DBI Changes file in the release has an incorrect 'Changes in...' line. Ignore it. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <28352-4@oink>; Mon, 20 Nov 1995 07:17:17 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816718061:06683:0; Sat, 18 Nov 95 18:07:41 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa06534; 18 Nov 95 18:07 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA03400 for dbi-dev-real; Sat, 18 Nov 1995 08:43:27 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from aladdin.rotterdam.luna.net (schuller@aladdin.rotterdam.luna.net [194.151.24.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA03396 for ; Sat, 18 Nov 1995 08:43:24 -0800 Received: (from schuller@localhost) by aladdin.rotterdam.luna.net (8.6.12/8.6.12) id RAA16365 for dbi-dev@toccata.fugue.com; Sat, 18 Nov 1995 17:46:05 -0100 Message-Id: <199511181846.RAA16365@aladdin.rotterdam.luna.net> Subject: Postgres95? To: dbi-dev@toccata.fugue.com Date: Sat, 18 Nov 1995 17:46:03 -0100 (GMT) From: Schuller@Luna.nl (Bart Schuller) X-Mailer: ELM [version 2.4 PL24 ME8b] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 127 Hello, Has anyone thought about writing DBD:Pg95 ? If not, would it be possible to use part of another adapter (mSQL)? Bart. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <07789-80@oink>; Tue, 21 Nov 1995 12:04:56 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816943111:25691:0; Tue, 21 Nov 95 08:38:31 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa25659; 21 Nov 95 8:38 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA01185 for dbi-dev-real; Mon, 20 Nov 1995 23:16:23 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA01181; Mon, 20 Nov 1995 23:16:11 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id NAA06855; Sun, 19 Nov 1995 13:00:23 GMT From: Alligator Descartes Message-Id: <199511191300.NAA06855@fruitbat.mcqueen.com> Subject: New DBI release To: dbi-users@fugue.com, dbi-dev@fugue.com Date: Sun, 19 Nov 1995 13:00:22 +0000 (GMT) Cc: Tim.Bunce@ig.co.uk X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 365 I've tested the new release of the DBI Switch ( 0.65 ) against the following drivers: DBD-Oracle-0.27 compiles fine on Solaris2.3 DBD-mSQL-0.60pl7 compiles fine on Solaris2.3 and Linux 1.2.13 DBD-Informix-0.20pl0 compiles fine on Solaris2.3 Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <07789-82@oink>; Tue, 21 Nov 1995 12:05:06 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816943493:28648:0; Tue, 21 Nov 95 08:44:53 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa28611; 21 Nov 95 8:44 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA01222 for dbi-users-real; Mon, 20 Nov 1995 23:21:21 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA01181; Mon, 20 Nov 1995 23:16:11 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id NAA06855; Sun, 19 Nov 1995 13:00:23 GMT From: Alligator Descartes Message-Id: <199511191300.NAA06855@fruitbat.mcqueen.com> Subject: New DBI release To: dbi-users@fugue.com, dbi-dev@fugue.com Date: Sun, 19 Nov 1995 13:00:22 +0000 (GMT) Cc: Tim.Bunce@ig.co.uk X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 365 I've tested the new release of the DBI Switch ( 0.65 ) against the following drivers: DBD-Oracle-0.27 compiles fine on Solaris2.3 DBD-mSQL-0.60pl7 compiles fine on Solaris2.3 and Linux 1.2.13 DBD-Informix-0.20pl0 compiles fine on Solaris2.3 Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <01579-0@oink>; Tue, 21 Nov 1995 17:21:36 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816968648:15450:0; Tue, 21 Nov 95 15:44:08 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa15348; 21 Nov 95 15:44 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA03426 for dbi-users-real; Tue, 21 Nov 1995 05:32:29 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA03417; Tue, 21 Nov 1995 05:31:11 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA36396; Tue, 21 Nov 1995 07:33:30 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dbi-users@fugue.com at austin.ibm.com; id AA21432; Tue, 21 Nov 1995 07:33:21 -0600 Message-Id: <9511211333.AA21432@dax.austin.ibm.com> To: Alligator Descartes Cc: dbi-users@fugue.com, dbi-dev@fugue.com, Tim.Bunce@ig.co.uk Subject: Re: New DBI release In-Reply-To: Your message of Sun, 19 Nov 95 13:00:22 +0000. <199511191300.NAA06855@fruitbat.mcqueen.com> Date: Tue, 21 Nov 95 07:33:20 -0600 Alligator Descartes writes: |> |>I've tested the new release of the DBI Switch ( 0.65 ) against the |>following drivers: |> |> DBD-Oracle-0.27 compiles fine on Solaris2.3 |> DBD-mSQL-0.60pl7 compiles fine on Solaris2.3 |> and Linux 1.2.13 |> DBD-Informix-0.20pl0 compiles fine on Solaris2.3 |> You can AIX 4.1.4 and DB::DB2 Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01581-2@oink>; Tue, 21 Nov 1995 17:21:59 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816967880:27248:0; Tue, 21 Nov 95 15:31:20 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa27125; 21 Nov 95 15:31 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA03421 for dbi-dev-real; Tue, 21 Nov 1995 05:31:17 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail1.austin.ibm.com (netmail1.austin.ibm.com [129.35.208.96]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA03417; Tue, 21 Nov 1995 05:31:11 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail1.austin.ibm.com (8.6.12/8.6.11) with SMTP id HAA36396; Tue, 21 Nov 1995 07:33:30 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dbi-users@fugue.com at austin.ibm.com; id AA21432; Tue, 21 Nov 1995 07:33:21 -0600 Message-Id: <9511211333.AA21432@dax.austin.ibm.com> To: Alligator Descartes Cc: dbi-users@fugue.com, dbi-dev@fugue.com, Tim.Bunce@ig.co.uk Subject: Re: New DBI release In-Reply-To: Your message of Sun, 19 Nov 95 13:00:22 +0000. <199511191300.NAA06855@fruitbat.mcqueen.com> Date: Tue, 21 Nov 95 07:33:20 -0600 Alligator Descartes writes: |> |>I've tested the new release of the DBI Switch ( 0.65 ) against the |>following drivers: |> |> DBD-Oracle-0.27 compiles fine on Solaris2.3 |> DBD-mSQL-0.60pl7 compiles fine on Solaris2.3 |> and Linux 1.2.13 |> DBD-Informix-0.20pl0 compiles fine on Solaris2.3 |> You can AIX 4.1.4 and DB::DB2 Mike mhm@austin.ibm.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <03925-3@oink>; Wed, 22 Nov 1995 07:11:52 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 816998563:14413:0; Wed, 22 Nov 95 00:02:43 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa14100; 22 Nov 95 0:02 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA04368 for dbi-users-real; Tue, 21 Nov 1995 14:03:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA04359; Tue, 21 Nov 1995 14:00:53 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id VAA28990; Tue, 21 Nov 1995 21:58:29 GMT From: Alligator Descartes Message-Id: <199511212158.VAA28990@fruitbat.mcqueen.com> Subject: Can't see the woods.....$DBI::errstr woes.... To: dbi-users@fugue.com, dbi-dev@fugue.com Date: Tue, 21 Nov 1995 21:58:28 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 724 DBI-0.65 DBD::Oracle-0.27 Simple script. ------- use DBI; $drh = DBI->install_driver( 'Oracle' ); $dbh = $drh->connect( 'database', 'user', 'pass' ); die "Cannot connect: $DBI::errstr\n" unless $dbh; print "Got here!\n"; ------- where database, user and pass are complete rubbish. I would expect the script to barf with the 'Cannot connect' message plus the appropriate Oracle error code string? ( Expecting an ORA-1034 usually.... ) However! I get: Got here! instead.................. Am I missing something big here? Or have I been developing drivers so long now that I can't program in them?! Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03924-9@oink>; Wed, 22 Nov 1995 07:13:05 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 816997651:07456:0; Tue, 21 Nov 95 23:47:31 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa07452; 21 Nov 95 23:47 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa27094; 21 Nov 95 23:47 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA04363 for dbi-dev-real; Tue, 21 Nov 1995 14:00:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA04359; Tue, 21 Nov 1995 14:00:53 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id VAA28990; Tue, 21 Nov 1995 21:58:29 GMT From: Alligator Descartes Message-Id: <199511212158.VAA28990@fruitbat.mcqueen.com> Subject: Can't see the woods.....$DBI::errstr woes.... To: dbi-users@fugue.com, dbi-dev@fugue.com Date: Tue, 21 Nov 1995 21:58:28 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 724 DBI-0.65 DBD::Oracle-0.27 Simple script. ------- use DBI; $drh = DBI->install_driver( 'Oracle' ); $dbh = $drh->connect( 'database', 'user', 'pass' ); die "Cannot connect: $DBI::errstr\n" unless $dbh; print "Got here!\n"; ------- where database, user and pass are complete rubbish. I would expect the script to barf with the 'Cannot connect' message plus the appropriate Oracle error code string? ( Expecting an ORA-1034 usually.... ) However! I get: Got here! instead.................. Am I missing something big here? Or have I been developing drivers so long now that I can't program in them?! Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <09771-43@oink>; Wed, 22 Nov 1995 17:06:32 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817054857:27362:0; Wed, 22 Nov 95 15:40:57 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa27352; 22 Nov 95 15:40 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa21787; 22 Nov 95 15:40 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA07051 for dbi-dev-real; Wed, 22 Nov 1995 03:45:36 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA07047; Wed, 22 Nov 1995 03:45:34 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id DAA22599; Wed, 22 Nov 1995 03:47:07 -0800 Message-Id: <199511221147.DAA22599@voro.lbl.gov> X-Mailer: exmh version 1.5.3 12/28/94 To: Alligator Descartes cc: dbi-users@fugue.com, dbi-dev@fugue.com Subject: Re: Can't see the woods.....$DBI::errstr woes.... In-reply-to: Your message of "Tue, 21 Nov 1995 21:58:28 GMT." <199511212158.VAA28990@fruitbat.mcqueen.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 22 Nov 1995 03:47:07 -0800 From: Mark Dedlow X-SMTP-Posting-Host: toccata.fugue.com [Wed, 22 Nov 95 15:40:11 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 22 Nov 95 15:40:56 GMT] Same DBI/DBD::Oracle versions (Solaris 2.4), I get: Database handle DBI::db=HASH(0x196e00) DESTROY ignored - never set up at ./testDBI.pl line 7. Abort (core dumped) if the user/pass is garbage. It works if user/pass OK... Mark > > DBI-0.65 > DBD::Oracle-0.27 > > Simple script. > > ------- > use DBI; > > $drh = DBI->install_driver( 'Oracle' ); > $dbh = $drh->connect( 'database', 'user', 'pass' ); > die "Cannot connect: $DBI::errstr\n" unless $dbh; > > print "Got here!\n"; > ------- > > where database, user and pass are complete rubbish. I would expect the > script to barf with the 'Cannot connect' message plus the appropriate Oracle > error code string? ( Expecting an ORA-1034 usually.... ) > > However! I get: > > Got here! > > instead.................. > > Am I missing something big here? Or have I been developing drivers so long > now that I can't program in them?! > > Regards. > > -- > Alligator Descartes | PGP key available on request! > descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <09771-54@oink>; Wed, 22 Nov 1995 17:08:25 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817058403:27872:0; Wed, 22 Nov 95 16:40:03 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa27862; 22 Nov 95 16:40 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa26709; 22 Nov 95 16:39 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA07059 for dbi-users-real; Wed, 22 Nov 1995 03:52:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA07047; Wed, 22 Nov 1995 03:45:34 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id DAA22599; Wed, 22 Nov 1995 03:47:07 -0800 Message-Id: <199511221147.DAA22599@voro.lbl.gov> X-Mailer: exmh version 1.5.3 12/28/94 To: Alligator Descartes cc: dbi-users@fugue.com, dbi-dev@fugue.com Subject: Re: Can't see the woods.....$DBI::errstr woes.... In-reply-to: Your message of "Tue, 21 Nov 1995 21:58:28 GMT." <199511212158.VAA28990@fruitbat.mcqueen.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 22 Nov 1995 03:47:07 -0800 From: Mark Dedlow X-SMTP-Posting-Host: toccata.fugue.com [Wed, 22 Nov 95 16:39:43 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 22 Nov 95 16:40:03 GMT] Same DBI/DBD::Oracle versions (Solaris 2.4), I get: Database handle DBI::db=HASH(0x196e00) DESTROY ignored - never set up at ./testDBI.pl line 7. Abort (core dumped) if the user/pass is garbage. It works if user/pass OK... Mark > > DBI-0.65 > DBD::Oracle-0.27 > > Simple script. > > ------- > use DBI; > > $drh = DBI->install_driver( 'Oracle' ); > $dbh = $drh->connect( 'database', 'user', 'pass' ); > die "Cannot connect: $DBI::errstr\n" unless $dbh; > > print "Got here!\n"; > ------- > > where database, user and pass are complete rubbish. I would expect the > script to barf with the 'Cannot connect' message plus the appropriate Oracle > error code string? ( Expecting an ORA-1034 usually.... ) > > However! I get: > > Got here! > > instead.................. > > Am I missing something big here? Or have I been developing drivers so long > now that I can't program in them?! > > Regards. > > -- > Alligator Descartes | PGP key available on request! > descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <13388-3@oink>; Thu, 23 Nov 1995 07:11:59 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817084963:21905:0; Thu, 23 Nov 95 00:02:43 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa21247; 23 Nov 95 0:01 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA11024 for dbi-users-real; Wed, 22 Nov 1995 14:02:21 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from jumbo.Read.TASC.COM (jumbo.read.tasc.com [147.81.240.3]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id OAA11020 for ; Wed, 22 Nov 1995 14:02:14 -0800 Received: from viper.is.rest.tasc.com by jumbo.Read.TASC.COM (4.1/TASC-CLIENT-1.5) id AA19241; Wed, 22 Nov 95 17:05:13 EST Received: from irnet.rest.tasc.com (stowe.irnet.rest.tasc.com) by viper.is.rest.tasc.com (4.1/TASCnet-4.1R0.0) id AA24099; Wed, 22 Nov 95 17:03:06 EST Received: from cannon.rest.tasc.com (cannon.irnet.rest.tasc.com) by irnet.rest.tasc.com (4.1/SMI-4.1) id AA18298; Wed, 22 Nov 95 17:04:09 EST From: mthurn@stowe.irnet.rest.tasc.com (Martin Thurn) Message-Id: <9511222204.AA18298@irnet.rest.tasc.com> Subject: cannot build DBI 0.65 on SunOS To: dbi-users@fugue.com Date: Wed, 22 Nov 1995 17:04:07 -0500 (EST) Cc: mthurn@stowe.irnet.rest.tasc.com (Martin Thurn), Tim.Bunce@ig.co.uk, thurn@cis.ohio-state.edu (Martin Thurn) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 8056 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 0:01:54 GMT] Thanks to those who pointed me to the latest version 0.65 of DBI. However, I get exactly the same error when I try to make it: ========== Script started on Wed Nov 22 16:57:38 1995 cannon:/fs81/TIR/lib/perl5/DBI-0.65[1]: make realclean rm -rf *~ t/*~ *.o *.a mon.out core so_locations DBI.bs DBI.bso DBI.exp DBI.c ./blib mv Makefile Makefile.old 2>/dev/null rm -rf /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI rm -f /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI/DBI.so /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI/DBI.bs rm -f /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI/DBI.a /fs81/TIR/lib/perl5/DBI-0.65/blib/DBI/DBIXS.h /fs81/TIR/lib/perl5/DBI-0.65/blib/DBD/ExampleP.pm /fs81/TIR/lib/perl5/DBI-0.65/blib/DBD/NullP.pm /fs81/TIR/lib/perl5/DBI-0.65/blib/DBD/Sponge.pm /fs81/TIR/lib/perl5/DBI-0.65/blib/DBI.pm rm -rf Makefile Makefile.old rm -f ./perlmain.o ./perlmain.c Makefile cannon:/fs81/TIR/lib/perl5/DBI-0.65[2]: perl Makefile.PL Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker DEFINE => '-g -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion' NAME => 'DBI' VERSION => '0.65' Using header files found in /usr/local/lib/perl5/sun4-sunos/CORE Looking for perl 5 by these names: miniperl perl, in these dirs: . /home2/accts/mthurn/bin /usr/local/bin /usr/local/etc /usr/bin/X11 /usr/openwin/bin /usr/ucb /usr/bin /bin /usr/etc /usr/local/ioffice/bin /fs41/inquery/bin /fs81/TIR/bin /usr/local/bin Executing /usr/local/bin/perl Writing Makefile for DBI cannon:/fs81/TIR/lib/perl5/DBI-0.65[3]: make mkdir /fs81/TIR/lib/perl5/DBI-0.65/blib mkdir /fs81/TIR/lib/perl5/DBI-0.65/blib/auto mkdir /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI /usr/local/bin/perl /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c gcc -c -O -fpic -I/usr/local/lib/perl5/sun4-sunos/CORE -g -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion DBI.c DBI.c: In function `dbih_getcom': DBI.c:232: warning: cast increases required alignment of target type DBI.c:240: warning: cast increases required alignment of target type DBI.c: In function `dbih_setup_handle': DBI.c:290: warning: implicit declaration of function `fprintf' DBI.c:318: warning: cast increases required alignment of target type DBI.c:330: warning: value computed is not used DBI.c:343: warning: value computed is not used DBI.c:349: warning: value computed is not used DBI.c:350: warning: value computed is not used DBI.c:351: warning: value computed is not used DBI.c:352: warning: value computed is not used DBI.c: In function `dbih_sth_bind_col': DBI.c:538: warning: value computed is not used DBI.c: In function `dbih_set_attr': DBI.c:589: warning: value computed is not used DBI.c: In function `dbih_get_attr': DBI.c:624: warning: value computed is not used DBI.c: In function `XS_DBI_dispatch': DBI.c:844: warning: value computed is not used DBI.c:844: warning: value computed is not used DBI.c: In function `XS_DBI__debug_dispatch': DBI.c:1152: warning: implicit declaration of function `fclose' DBI.c: In function `XS_DBD_____st_fetchrow': DBI.c:1423: warning: cast increases required alignment of target type DBI.c:1423: warning: cast increases required alignment of target type DBI.c:1427: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD_____st_fetch': DBI.c:1460: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD___mem__common_DESTROY': DBI.c:1552: warning: cast increases required alignment of target type Running mkbootstrap for DBI () ld -o /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI/DBI.so -assert nodefinitions DBI.o cp DBI.bs /fs81/TIR/lib/perl5/DBI-0.65/blib/auto/DBI/DBI.bs mkdir /fs81/TIR/lib/perl5/DBI-0.65/blib/DBI cp DBIXS.h /fs81/TIR/lib/perl5/DBI-0.65/blib/DBI/DBIXS.h mkdir /fs81/TIR/lib/perl5/DBI-0.65/blib/DBD cp lib/DBD/ExampleP.pm /fs81/TIR/lib/perl5/DBI-0.65/blib/DBD/ExampleP.pm cp lib/DBD/NullP.pm /fs81/TIR/lib/perl5/DBI-0.65/blib/DBD/NullP.pm cp lib/DBD/Sponge.pm /fs81/TIR/lib/perl5/DBI-0.65/blib/DBD/Sponge.pm cp DBI.pm /fs81/TIR/lib/perl5/DBI-0.65/blib/DBI.pm cannon:/fs81/TIR/lib/perl5/DBI-0.65[4]: make test /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=11, 1 secs ( 0.42 cusr 0.15 csys = 0.57 cpu) /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ make: *** [test] Segmentation fault (core dumped) cannon:/fs81/TIR/lib/perl5/DBI-0.65[5]: make test TEST_VERBOSE=1 /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............1..11 ok 1 ok 2 ok 3 ok 4 FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=11, 1 secs ( 0.35 cusr 0.18 csys = 0.53 cpu) /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ make: *** [test] Segmentation fault (core dumped) cannon:/fs81/TIR/lib/perl5/DBI-0.65[6]: exit cannon:/fs81/TIR/lib/perl5/DBI-0.65[7]: script done on Wed Nov 22 16:58:43 1995 ========== Script started on Wed Nov 22 16:59:00 1995 cannon:/fs81/TIR/lib/perl5/DBI-0.65[1]: perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=sunos, osver=4.1.3, archname=sun4-sunos uname='sunos stowe 4.1.3 1 sun4m ' hint=recommended Compiler: cc='gcc', optimize='-O' cppflags='' ccflags ='' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=31 Libraries: so=so libpth=/lib /usr/lib /usr/ucblib /usr/local/lib libs=-lnsl -ldbm -ldl -lm -lc -lposix libc=/usr/lib/libc.so.1.8 Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-fpic', ccdlflags='', lddlflags='-assert nodefinitions' cannon:/fs81/TIR/lib/perl5/DBI-0.65[2]: exit cannon:/fs81/TIR/lib/perl5/DBI-0.65[3]: script done on Wed Nov 22 16:59:06 1995 ========== Script started on Wed Nov 22 16:59:25 1995 cannon:/fs81/TIR/lib/perl5/DBI-0.65[1]: make test PERL_DL_DEBUG=2 /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............DynaLoader.pm loaded (/lib /usr/lib /usr/ucblib /usr/local/lib /usr/lib/X11/lib /usr/openwin/lib) DynaLoader::bootstrap(DBI) FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=11, 2 secs ( 0.48 cusr 0.22 csys = 0.70 cpu) /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ DynaLoader.pm loaded (/lib /usr/lib /usr/ucblib /usr/local/lib /usr/lib/X11/lib /usr/openwin/lib) DynaLoader::bootstrap(DBI) make: *** [test] Segmentation fault (core dumped) cannon:/fs81/TIR/lib/perl5/DBI-0.65[2]: exit cannon:/fs81/TIR/lib/perl5/DBI-0.65[3]: script done on Wed Nov 22 16:59:42 1995 ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <13388-4@oink>; Thu, 23 Nov 1995 07:12:21 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817085483:26493:0; Thu, 23 Nov 95 00:11:23 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa26376; 23 Nov 95 0:11 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA11143 for dbi-users-real; Wed, 22 Nov 1995 14:25:08 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA11137 for ; Wed, 22 Nov 1995 14:24:51 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA21711; Wed, 22 Nov 1995 22:21:53 GMT From: Alligator Descartes Message-Id: <199511222221.WAA21711@fruitbat.mcqueen.com> Subject: Re: cannot build DBI 0.65 on SunOS To: mthurn@stowe.irnet.rest.tasc.com (Martin Thurn) Date: Wed, 22 Nov 1995 22:21:52 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: <9511222204.AA18298@irnet.rest.tasc.com> from "Martin Thurn" at Nov 22, 95 05:04:07 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 243 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 0:11:16 GMT] > Thanks to those who pointed me to the latest version 0.65 of DBI. > However, I get exactly the same error when I try to make it: Can you send me a stack trace from the core dump? Thanks. A. -- Alligator Descartes descarte@hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13275-53@oink>; Thu, 23 Nov 1995 07:15:11 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817081314:02960:0; Wed, 22 Nov 95 23:01:54 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa02789; 22 Nov 95 23:01 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 22 Nov 1995 22:58:34 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA10721 for dbi-users-real; Wed, 22 Nov 1995 12:57:21 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from emerald.empress.com (emerald.empress.com [192.43.219.6]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA10717 for ; Wed, 22 Nov 1995 12:57:14 -0800 Received: (from fjones@localhost) by emerald.empress.com (8.6.11/8.6.9) id QAA01098 for dbi-users@fugue.com; Wed, 22 Nov 1995 16:02:03 -0500 From: Francis Jones Message-Id: <199511222102.QAA01098@emerald.empress.com> Subject: perl5 DBI manual pages To: dbi-users@fugue.com Date: Wed, 22 Nov 1995 16:02:02 -0500 (EST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 774 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 22 Nov 95 23:01:53 GMT] hi, I've just acquired and installed the Perl5 DBI. I've had a minor problem: after you create the Makefile via % perl Makefile.PL you get a Makefile which wants to install the manual pages in $(ROOTDIR)/lib/perl5/man, rather than in $(ROOTDIR)/man which seems to me a more standard spot. If every package (e.g., bash, perl, perl DBI,...) installs in its own 'man' subdirectory, then fine... but most install in a common spot (such as /usr/local/man), so it seems that barring any extraordinary circumstances, that the perl DBI should install there too. IMHO, it makes more sense to install in a common place since you don't have to keep altering your MANPATH to find the stuff. In any case, is there a reason for this choice that I should be aware of? thanks, fj ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13486-0@oink>; Thu, 23 Nov 1995 07:16:12 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817082018:09645:0; Wed, 22 Nov 95 23:13:38 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa09635; 22 Nov 95 23:13 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 22 Nov 1995 23:12:38 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA10864 for dbi-users-real; Wed, 22 Nov 1995 13:28:51 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA10860 for ; Wed, 22 Nov 1995 13:28:45 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id VAA21006; Wed, 22 Nov 1995 21:26:24 GMT From: Alligator Descartes Message-Id: <199511222126.VAA21006@fruitbat.mcqueen.com> Subject: Re: DBI make failed To: mthurn@irnet.rest.tasc.com (Martin Thurn) Date: Wed, 22 Nov 1995 21:26:24 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: <9511222054.AA18163@irnet.rest.tasc.com> from "Martin Thurn" at Nov 22, 95 03:54:34 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 453 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 22 Nov 95 23:13:38 GMT] > Making DBI failed on my machine. Can you help me fix it? Thank you. > Attached below is the info requested in the README file (that which I > could get, at least). I recommend upgrading to DBI-0.65 and try again with that. The version you have is fairly out of date. > - - Martin Thurn > mthurn@tasc.com or thurn@cis.ohio-state.edu -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <13274-27@oink>; Thu, 23 Nov 1995 07:10:35 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817073386:06022:0; Wed, 22 Nov 95 20:49:46 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ab05940; 22 Nov 95 20:49 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 22 Nov 1995 20:48:57 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA07767 for dbi-dev-real; Wed, 22 Nov 1995 10:24:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA07763 for ; Wed, 22 Nov 1995 10:24:05 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id SAA17901; Wed, 22 Nov 1995 18:21:18 GMT From: Alligator Descartes Message-Id: <199511221821.SAA17901@fruitbat.mcqueen.com> Subject: Re: perl DBI To: fjones@empress.com (Francis Jones) Date: Wed, 22 Nov 1995 18:21:16 +0000 (GMT) Cc: dbi-dev@fugue.com, Tim.Bunce@ig.co.uk In-Reply-To: <199511221802.NAA00883@emerald.empress.com> from "Francis Jones" at Nov 22, 95 01:02:58 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 5808 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 22 Nov 95 20:49:45 GMT] > Naturally, everything is now fine... I'll be honest. I just heard of a routing failure this afternoon in MCI, so that would probably have cut transatlantic service for a while...... > > Can I put you down as the forthcoming author of the Empress driver then? > > Hmm. Well, I'm starting work on it, but you can never tell when > things get snagged as a result of more pressing (ie. commercial) > issues, since we're a money-making company. So, I think it's > fair to say that: > ... _someone_ will be working on it, > ... for the moment it's me, > ... I don't see anything on the horizon at this point > to pull me away from it. That'll do! 8-) Just remember to keep checking your entry in the DBD index to make sure nothing's wrong, or if something changes, to let me know. > I hope you don't mind if I ask a couple more questions... > > 1. Is there a mailing list/newsgroup for the DBI? I tried > perdb-interest-REQUEST@vix.com, but got a reply back saying > that it was defunct. I'm looking for a site/list/etc. which > I can monitor to stay up to date. Yup, there are three now. There's: dbi-dev-request@fugue.com # DBI/DBD development issues dbi-users-request@fugue.com # DBI/DBD user issues dbi-announce-request@fugue.com # DBI/DBD announcements &c There's a form at: http://www.fugue.com/dbi which allows you to subscribe. That, or mail the above addresses. The list is hand-maintained, so allow a few days to get subscribed, although traffic is low just now. > 2. I have a copy of the "Perl 5 Database Interface (DBI) API specification" > (Version 0.6). From the diagram in that document, the usage of databases > within perl seems to be something like this: > > Perl Script --> API --> Switch --> Driver --> DB Engine > > and this document covers the 'API' and 'Switch' portions of this. > So, it seemed to me that I needed to try to implement Empress versions > of the 'lower-level' functions which I expected to see in the Switch > portion (Ch. 5) of the document. But, it seems that this was wrong, and that > instead I should be attempting to implement Empress versions of the > functions defined in the API portion (Ch. 3) of the document. > I have downloaded the oracle, mSQL & informix packages from your > site, so I guess I will see what they have done, but could you clear > up my confusion? I think I don't understand what the purpose of the > switch module is. Do you know if there's a document on 'how to write a driver > for the DB Perl API' that is more detailed for neophytes to Perl 5 > (like me) than the document on hermetica? (e.g., where does the C code go?) Right, the Empress functions comes into play at the 'Driver' level. o The 'Perl Script' is obvious. o The 'API' is the DBI API, eg, 'connect', 'fetchrow' and so on. o The 'Switch' basically holds data on what drivers are loaded ( since you can load multiple drivers simultaneously ), what handles are assigned to which drivers, what statements are assigned to which handles and the like. o The 'Driver' is the bit we're interested in..........See below. o The 'DB Engine' is whichever database engine you're connecting to. The DBD::Empress module would follow the form of the DBD::Oracle, DBD::mSQL and DBD::Informix modules, in that, the Empress.xs file ( see the .xs files from the above distributions ) would define the entry points from the DBI API to the low-level database vendor-specific code, eg, take a look at Oracle.xs. There's a function in there called 'connect'. This in turn calls 'dbd_db_login'. Taking a look at dbdimp.c, this is where all the seriously nasty vendor code goes. So, following on from our example above, looking at dbd_db_login() we see that it calls the database-vendor API function to connect to a database. In Oracle this is ologon, in mSQL msqlConnect, in Informix $database. So, in that function call the Empress API function to connect to a database. The translation from DBI API call to that function happens through the .xs file. Thus, the flow goes: $dbh = $drh->connect( .... ); # $dbh = database handle # $drh = driver handle Empress.xs:connect() ---> dbdimp.c: dbd_db_login() ---> EmpressAPI: empressConnect() # or whatever it's called So, the actual data returned goes from the Empress API call back up through the .xs file into the perl script ( to simplify horrendously ). Because the .xs file is identical ( nearly! ) between all the drivers, it provides a standard hook-point for your vendor-specific code. > 3. What are the names of the API functions? I'm a bit confused by > the quote (e.g., "db'....") naming convention. In ch.10.1 there is > a summary of the functions, but there seems to be an inconsistency: > > e.g., db'setvalue() vs. db_setvalue() Take a look at 'test.pl' from the DBD::mSQL distribution. It provides a fairly complete exercise of the DBI API. Plus some special mSQL specific implemented functions.... > As you can guess I'm a newcomer to the perl DB (and to perl 5, as well) > so I can imagine that some of the answers to these questions are obvious. > But, I'm not sure where to look, and any help in the way of either > direct response, or reference to a document, would be a huge help. Well, it's a nebulous area just now since Tim's occasionally reconstructing bits of the DBI interface itself, plus I can't get documentation online fast enough at Hermetica, but I hope the above helps a bit. I'll forward a document on explaining something about the various handles and such to you too, so that can give you a better grip from the perl side of using it rather than the driver coding side! > Francis Jones Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13385-28@oink>; Thu, 23 Nov 1995 07:16:38 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817083104:19705:0; Wed, 22 Nov 95 23:31:44 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa19701; 22 Nov 95 23:31 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa06716; 22 Nov 95 23:28 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA10790 for dbi-users-real; Wed, 22 Nov 1995 13:16:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from jumbo.Read.TASC.COM (jumbo.read.tasc.com [147.81.240.3]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA10786 for ; Wed, 22 Nov 1995 13:16:53 -0800 Received: from viper.is.rest.tasc.com by jumbo.Read.TASC.COM (4.1/TASC-CLIENT-1.5) id AA18348; Wed, 22 Nov 95 16:19:52 EST Received: from irnet.rest.tasc.com (stowe.irnet.rest.tasc.com) by viper.is.rest.tasc.com (4.1/TASCnet-4.1R0.0) id AA23369; Wed, 22 Nov 95 16:17:48 EST Received: from cannon.rest.tasc.com (cannon.irnet.rest.tasc.com) by irnet.rest.tasc.com (4.1/SMI-4.1) id AA18219; Wed, 22 Nov 95 16:18:53 EST Received: from fandago.Read.TASC.COM by irnet.rest.tasc.com (4.1/SMI-4.1) id AA18172; Wed, 22 Nov 95 15:57:40 EST Received: from jumbo.Read.TASC.COM by fandago.Read.TASC.COM (4.1/TASC-1.9) id AA26599; Wed, 22 Nov 95 15:53:55 EST Received: from viper.is.rest.tasc.com by jumbo.Read.TASC.COM (4.1/TASC-CLIENT-1.5) id AA17848; Wed, 22 Nov 95 15:55:37 EST Received: from irnet.rest.tasc.com (stowe.irnet.rest.tasc.com) by viper.is.rest.tasc.com (4.1/TASCnet-4.1R0.0) id AA23083; Wed, 22 Nov 95 15:53:32 EST Received: from cannon.rest.tasc.com (cannon.irnet.rest.tasc.com) by irnet.rest.tasc.com (4.1/SMI-4.1) id AA18163; Wed, 22 Nov 95 15:54:36 EST From: Martin Thurn Message-Id: <9511222054.AA18163@irnet.rest.tasc.com> Subject: DBI make failed To: dbi-users@vix.com, Tim.Bunce@ig.co.uk, Martin Thurn Date: Wed, 22 Nov 1995 15:54:34 -0500 (EST) Cc: Martin Thurn X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 7788 X-SMTP-Posting-Host: toccata.fugue.com [Wed, 22 Nov 95 23:29:00 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 22 Nov 95 23:31:44 GMT] Making DBI failed on my machine. Can you help me fix it? Thank you. Attached below is the info requested in the README file (that which I could get, at least). - - Martin Thurn mthurn@tasc.com or thurn@cis.ohio-state.edu ===== Script started on Wed Nov 22 15:36:09 1995 cannon:/fs81/TIR/lib/perl5/DBI[1]: make realclean rm -rf *~ t/*~ *.o *.a mon.out core so_locations DBI.bs DBI.bso DBI.exp DBI.c ./blib mv Makefile Makefile.old 2>/dev/null rm -rf /fs81/TIR/lib/perl5/DBI/blib/auto/DBI /fs81/TIR/lib/perl5/DBI/blib/auto/DBI rm -f /fs81/TIR/lib/perl5/DBI/blib/auto/DBI/DBI.so /fs81/TIR/lib/perl5/DBI/blib/auto/DBI/DBI.bs rm -f /fs81/TIR/lib/perl5/DBI/blib/auto/DBI/DBI.a /fs81/TIR/lib/perl5/DBI/blib/DBI/DBIXS.h /fs81/TIR/lib/perl5/DBI/blib/DBD/ExampleP.pm /fs81/TIR/lib/perl5/DBI/blib/DBD/NullP.pm /fs81/TIR/lib/perl5/DBI/blib/DBI.pm rm -rf Makefile Makefile.old rm -f ./perlmain.o ./perlmain.c Makefile cannon:/fs81/TIR/lib/perl5/DBI[2]: perl Makefile.PL Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker DEFINE => '-Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion' NAME => 'DBI' VERSION => '0.63' Using header files found in /usr/local/lib/perl5/sun4-sunos/CORE Looking for perl 5 by these names: miniperl perl, in these dirs: . /home2/accts/mthurn/bin /usr/local/bin /usr/local/etc /usr/bin/X11 /usr/openwin/bin /usr/ucb /usr/bin /bin /usr/etc /usr/local/ioffice/bin /fs41/inquery/bin /fs81/TIR/bin /usr/local/bin Executing /usr/local/bin/perl Writing Makefile for DBI cannon:/fs81/TIR/lib/perl5/DBI[3]: make mkdir /fs81/TIR/lib/perl5/DBI/blib mkdir /fs81/TIR/lib/perl5/DBI/blib/auto mkdir /fs81/TIR/lib/perl5/DBI/blib/auto/DBI /usr/local/bin/perl /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c gcc -c -O -fpic -I/usr/local/lib/perl5/sun4-sunos/CORE -Wall -pedantic -Wno-comment -Wpointer-arith -Wcast-align -Wconversion DBI.c DBI.c: In function `dbih_getcom': DBI.c:220: warning: cast increases required alignment of target type DBI.c:228: warning: cast increases required alignment of target type DBI.c: In function `dbih_setup_handle': DBI.c:278: warning: implicit declaration of function `fprintf' DBI.c:306: warning: cast increases required alignment of target type DBI.c:318: warning: value computed is not used DBI.c:331: warning: value computed is not used DBI.c:337: warning: value computed is not used DBI.c:338: warning: value computed is not used DBI.c:339: warning: value computed is not used DBI.c: In function `dbih_sth_bind_col': DBI.c:518: warning: value computed is not used DBI.c: In function `dbih_set_attr': DBI.c:569: warning: value computed is not used DBI.c: In function `dbih_get_attr': DBI.c:604: warning: value computed is not used DBI.c: In function `XS_DBI__debug_dispatch': DBI.c:1084: warning: implicit declaration of function `fclose' DBI.c: In function `XS_DBD_____st_fetchrow': DBI.c:1330: warning: cast increases required alignment of target type DBI.c:1330: warning: cast increases required alignment of target type DBI.c:1334: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD_____st_fetch': DBI.c:1367: warning: cast increases required alignment of target type DBI.c: In function `XS_DBD___mem__common_DESTROY': DBI.c:1457: warning: cast increases required alignment of target type Running mkbootstrap for DBI () ld -o /fs81/TIR/lib/perl5/DBI/blib/auto/DBI/DBI.so -assert nodefinitions DBI.o cp DBI.bs /fs81/TIR/lib/perl5/DBI/blib/auto/DBI/DBI.bs mkdir /fs81/TIR/lib/perl5/DBI/blib/DBI cp DBIXS.h /fs81/TIR/lib/perl5/DBI/blib/DBI/DBIXS.h mkdir /fs81/TIR/lib/perl5/DBI/blib/DBD cp lib/DBD/ExampleP.pm /fs81/TIR/lib/perl5/DBI/blib/DBD/ExampleP.pm cp lib/DBD/NullP.pm /fs81/TIR/lib/perl5/DBI/blib/DBD/NullP.pm cp DBI.pm /fs81/TIR/lib/perl5/DBI/blib/DBI.pm cannon:/fs81/TIR/lib/perl5/DBI[4]: make test /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI/blib -I/fs81/TIR/lib/perl5/DBI/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.28 cusr 0.23 csys = 0.52 cpu) /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI/blib -I/fs81/TIR/lib/perl5/DBI/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ make: *** [test] Segmentation fault (core dumped) cannon:/fs81/TIR/lib/perl5/DBI[5]: exit cannon:/fs81/TIR/lib/perl5/DBI[6]: script done on Wed Nov 22 15:36:55 1995 ========== Script started on Wed Nov 22 15:37:46 1995 cannon:/fs81/TIR/lib/perl5/DBI[1]: perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=sunos, osver=4.1.3, archname=sun4-sunos uname='sunos stowe 4.1.3 1 sun4m ' hint=recommended Compiler: cc='gcc', optimize='-O' cppflags='' ccflags ='' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=31 Libraries: so=so libpth=/lib /usr/lib /usr/ucblib /usr/local/lib libs=-lnsl -ldbm -ldl -lm -lc -lposix libc=/usr/lib/libc.so.1.8 Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-fpic', ccdlflags='', lddlflags='-assert nodefinitions' cannon:/fs81/TIR/lib/perl5/DBI[2]: exit cannon:/fs81/TIR/lib/perl5/DBI[3]: script done on Wed Nov 22 15:37:54 1995 ========== Script started on Wed Nov 22 15:42:19 1995 cannon:/fs81/TIR/lib/perl5/DBI[1]: make test PERL_DL_DEBUG=2 /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI/blib -I/fs81/TIR/lib/perl5/DBI/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/min...............DynaLoader.pm loaded (/lib /usr/lib /usr/ucblib /usr/local/lib /usr/lib/X11/lib /usr/openwin/lib) DynaLoader::bootstrap(DBI) FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.33 cusr 0.17 csys = 0.50 cpu) /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI/blib -I/fs81/TIR/lib/perl5/DBI/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ DynaLoader.pm loaded (/lib /usr/lib /usr/ucblib /usr/local/lib /usr/lib/X11/lib /usr/openwin/lib) DynaLoader::bootstrap(DBI) make: *** [test] Segmentation fault (core dumped) cannon:/fs81/TIR/lib/perl5/DBI[2]: exit cannon:/fs81/TIR/lib/perl5/DBI[3]: script done on Wed Nov 22 15:42:30 1995 ========= Script started on Wed Nov 22 15:46:22 1995 cannon:/fs81/TIR/lib/perl5/DBI[1]: make test TEST_VERBOSE=1 /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI/blib -I/fs81/TIR/lib/perl5/DBI/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t t/min...............1..6 ok 1 ok 2 ok 3 ok 4 FAILED on test 5 Failed 1 test, 0.00% okay. Files=1, Tests=6, 1 secs ( 0.33 cusr 0.15 csys = 0.48 cpu) /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI/blib -I/fs81/TIR/lib/perl5/DBI/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ make: *** [test] Segmentation fault (core dumped) cannon:/fs81/TIR/lib/perl5/DBI[2]: exit cannon:/fs81/TIR/lib/perl5/DBI[3]: script done on Wed Nov 22 15:46:33 1995 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13486-6@oink>; Thu, 23 Nov 1995 07:17:35 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817085632:13464:0; Thu, 23 Nov 95 00:13:52 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa13459; 23 Nov 95 0:13 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa10667; 23 Nov 95 0:13 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA11320 for dbi-users-real; Wed, 22 Nov 1995 14:49:54 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA11314; Wed, 22 Nov 1995 14:49:44 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA22006; Wed, 22 Nov 1995 22:47:27 GMT From: Alligator Descartes Message-Id: <199511222247.WAA22006@fruitbat.mcqueen.com> Subject: DBI Mailing List Archives To: dbi-announce@fugue.com, dbi-users@fugue.com Date: Wed, 22 Nov 1995 22:47:27 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 377 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 0:13:22 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Thu, 23 Nov 95 0:13:51 GMT] All three DBI mailing lists: DBI Announcements DBI Development DBI Users are being archived as hypermail archives at: http://www.hermetica.com as of now! The archives go about 3-4 weeks back on all three lists. There's also more documentation online now too. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13504-0@oink>; Thu, 23 Nov 1995 07:18:22 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817086561:21918:0; Thu, 23 Nov 95 00:29:21 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa21910; 23 Nov 95 0:29 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa11869; 23 Nov 95 0:28 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA11330 for dbi-announce-real; Wed, 22 Nov 1995 14:50:03 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA11314; Wed, 22 Nov 1995 14:49:44 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA22006; Wed, 22 Nov 1995 22:47:27 GMT From: Alligator Descartes Message-Id: <199511222247.WAA22006@fruitbat.mcqueen.com> Subject: DBI Mailing List Archives To: dbi-announce@fugue.com, dbi-users@fugue.com Date: Wed, 22 Nov 1995 22:47:27 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 377 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 0:28:25 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Thu, 23 Nov 95 0:29:20 GMT] All three DBI mailing lists: DBI Announcements DBI Development DBI Users are being archived as hypermail archives at: http://www.hermetica.com as of now! The archives go about 3-4 weeks back on all three lists. There's also more documentation online now too. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13522-3@oink>; Thu, 23 Nov 1995 07:19:48 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817103253:07321:0; Thu, 23 Nov 95 05:07:33 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa07289; 23 Nov 95 5:07 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA13470 for dbi-users-real; Wed, 22 Nov 1995 19:39:32 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail.crl.com (mail.crl.com [165.113.1.22]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id TAA13466 for ; Wed, 22 Nov 1995 19:39:30 -0800 Received: from [205.149.164.116] (bta.vip.best.com) by mail.crl.com with SMTP id AA20060 (5.65c/IDA-1.5 for ); Wed, 22 Nov 1995 19:39:55 -0800 Date: Wed, 22 Nov 1995 19:39:55 -0800 X-Sender: mrlizard@mail.crl.com Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: msql-list@bunyip.com From: mrlizard@crl.com (Brian Valente) Subject: mSQL API available through DBI? Cc: dbi-users@fugue.com X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 5:07:31 GMT] I've read through Alligators ramblings (which are rather informative, BTW), but I still haven't been able to figure out if you can access the mSQL API (as defined in the documentation) through DBI. For example, I want to find out how many rows were returned by my last query: Perhaps it is: $result = $dbh->func( '_NumRows' )?? I saw the $dbh->func() but I'm not sure what that does for me. Thanks, Brian ====================================================================== Brian Valente || brian_valente@rmdudley.com R.M. Dudley Corporation || mrlizard@crl.com 1799 Old Bayshore Suite 200 || applelink:RMDUDLEY Burlingame, CA 94010 || http://www.rmdudley.com/rmdudley/ (415) 697-1650 voice || (415) 697-1816 fax || ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <19022-6@oink>; Thu, 23 Nov 1995 16:11:26 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817136392:20714:0; Thu, 23 Nov 95 14:19:52 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa20652; 23 Nov 95 14:19 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA21021 for dbi-users-real; Thu, 23 Nov 1995 03:47:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA21017 for ; Thu, 23 Nov 1995 03:47:06 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id LAA25567; Thu, 23 Nov 1995 11:44:33 GMT From: Alligator Descartes Message-Id: <199511231144.LAA25567@fruitbat.mcqueen.com> Subject: Re: [mSQL] mSQL API available through DBI? To: mrlizard@crl.com Date: Thu, 23 Nov 1995 11:44:32 +0000 (GMT) Cc: dbi-users@fugue.com, msql-list@bunyip.com In-Reply-To: from "Brian Valente" at Nov 22, 95 07:39:55 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1237 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 14:19:45 GMT] > I've read through Alligators ramblings (which are rather informative, BTW), > but I still haven't been able to figure out if you can access the mSQL API > (as defined in the documentation) through DBI. Well, you can't. That's the point of DBI. It abstracts the database junk away from the native API! Of course, if we haven't fully implemented the DBI side of things, then it's a complete pain. However, see below. > For example, I want to find out how many rows were returned by my last query: > > Perhaps it is: > > $result = $dbh->func( '_NumRows' )?? > > I saw the $dbh->func() but I'm not sure what that does for me. That bit of the DBD::mSQL driver currently "under construction" for release 0.61 ( in a week or two... ) The missing functionality from the mSQL API currently is: ( off the top of my head ): Create a database <-- Fixed in my copy. Accessible through 'func' Drop a database <-- Ditto # of row returned <-- Under implementation, probably accessible through a variable.... List table meta-data <-- Sort of working. These'll all be in the next major(-ish) release. > Brian -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19021-29@oink>; Thu, 23 Nov 1995 16:13:01 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817130225:24367:0; Thu, 23 Nov 95 12:37:05 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa24022; 23 Nov 95 12:37 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA20831 for dbi-dev-real; Thu, 23 Nov 1995 02:25:03 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay.inferentia.it (dns.inferentia.it [151.99.220.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA20827 for ; Thu, 23 Nov 1995 02:24:46 -0800 Received: from yquem.inferentia.it (yquem.inferentia.it [194.184.62.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id LAA26057 for ; Thu, 23 Nov 1995 11:26:56 +0100 Posted-Date: Thu, 23 Nov 1995 11:26:56 +0100 Received: by yquem.inferentia.it with Microsoft Mail id <01BAB998.0E3C2460@yquem.inferentia.it>; Thu, 23 Nov 1995 11:37:23 +-100 Message-ID: <01BAB998.0E3C2460@yquem.inferentia.it> From: Davide Migliavacca To: "'dbi-dev@fugue.com'" Subject: Latest DBI and DBD::Oracle OK on AIX 3.2.5, OSF1 3.2 Date: Thu, 23 Nov 1995 11:37:15 +-100 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 12:37:04 GMT] Sorry if this is already known - been away from the lists for a while. I just grabbed DBI 0.65 and DBD::Oracle 0.27, compiled, tested, = installed and field-tested on existing code on both AIX 3.2.5 and DEC = OSF/1 3.2. So this may be considered "known working platforms" as well. Cheers Davide -------------------------------------------------------- Davide Migliavacca (aka shari) Inferentia S.r.l. - Milano ITALY +39 2 59928.1 (FAX .221) * Standard disclaimer here * -------------------------------------------------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19021-37@oink>; Thu, 23 Nov 1995 16:13:42 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817133033:18817:0; Thu, 23 Nov 95 13:23:53 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa18643; 23 Nov 95 13:23 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA20859 for dbi-users-real; Thu, 23 Nov 1995 02:32:09 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay.inferentia.it (dns.inferentia.it [151.99.220.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA20855 for ; Thu, 23 Nov 1995 02:32:06 -0800 Received: from yquem.inferentia.it (yquem.inferentia.it [194.184.62.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id LAA28229 for ; Thu, 23 Nov 1995 11:34:37 +0100 Posted-Date: Thu, 23 Nov 1995 11:34:37 +0100 Received: by yquem.inferentia.it with Microsoft Mail id <01BAB999.21665780@yquem.inferentia.it>; Thu, 23 Nov 1995 11:45:04 +-100 Message-ID: <01BAB999.21665780@yquem.inferentia.it> From: Davide Migliavacca To: "'dbi-users@fugue.com'" Subject: WOWPERL call for user feedback Date: Thu, 23 Nov 1995 11:45:02 +-100 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Thu, 23 Nov 95 13:23:46 GMT] Is anybody *actually* using wowperl (ftp://ftp.inferentia.it/pub/perl/db/wowperl.tar)? I have received almost no feedback about it. Just wanted to know if it's worth maintaining. I was thinking about adding samples but I will do that only if someone benefits from it. IMHO replies should be placed directly to me rather than to list. Davide -------------------------------------------------------- Davide Migliavacca (aka shari) Inferentia S.r.l. - Milano ITALY +39 2 59928.1 (FAX .221) * Standard disclaimer here * -------------------------------------------------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04603-0@oink>; Mon, 27 Nov 1995 07:07:14 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817242620:04468:0; Fri, 24 Nov 95 19:50:20 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa03743; 24 Nov 95 19:50 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA02345 for dbi-users-real; Fri, 24 Nov 1995 09:33:59 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA02299; Fri, 24 Nov 1995 09:26:36 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id RAA11217; Fri, 24 Nov 1995 17:23:45 GMT From: Alligator Descartes Message-Id: <199511241723.RAA11217@fruitbat.mcqueen.com> Subject: DBD::mSQL-0.60pl8 released To: dbi-users@fugue.com, dbi-dev@fugue.com, dbi-announce@fugue.com Date: Fri, 24 Nov 1995 17:23:43 +0000 (GMT) Cc: msql-list@bunyip.com, k@anna.mind.de, Tim.Bunce@ig.co.uk, SVEN.VERDOOLAEGE@student.kuleuven.ac.be (Sven Verdoolaege), mrlizard@crl.com X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1014 X-SMTP-Posting-Host: toccata.fugue.com [Fri, 24 Nov 95 19:50:14 GMT] DBD::mSQL-0.60pl8 has been released. Here's the extract from the Changes file: o Added error string and number to 'func' methods o Completed implementation of 'ListFields' invoked through the 'func' method o Added implementation of 'CreateDB' and 'DropDB' through the 'func' method o Added a method 'NumRows' against $sth to show the number of rows returned by a statement. The mSQL API is now fully featured within the DBD::mSQL driver, so it's well worth upgrading. Information on DBI in general is available from: http://www.hermetica.com/technologia/DBI The code is available from the usual places: ftp://ftp.mcqueen.com/pub/dbperl/DBD/mSQL/DBD-mSQL-0.60pl8.tar.gz being one of them. CPAN should have it by tomorrow ( Saturday ). Persons using Msqlperl are now advised to think about upgrading onto DBI/DBD::mSQL instead. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <04604-1@oink>; Mon, 27 Nov 1995 07:07:20 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817243227:19122:0; Fri, 24 Nov 95 20:00:27 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ab19019; 24 Nov 95 20:00 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Fri, 24 Nov 1995 19:49:16 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA02303 for dbi-announce-real; Fri, 24 Nov 1995 09:26:47 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA02299; Fri, 24 Nov 1995 09:26:36 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id RAA11217; Fri, 24 Nov 1995 17:23:45 GMT From: Alligator Descartes Message-Id: <199511241723.RAA11217@fruitbat.mcqueen.com> Subject: DBD::mSQL-0.60pl8 released To: dbi-users@fugue.com, dbi-dev@fugue.com, dbi-announce@fugue.com Date: Fri, 24 Nov 1995 17:23:43 +0000 (GMT) Cc: msql-list@bunyip.com, k@anna.mind.de, Tim.Bunce@ig.co.uk, SVEN.VERDOOLAEGE@student.kuleuven.ac.be (Sven Verdoolaege), mrlizard@crl.com X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1014 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Fri, 24 Nov 95 20:00:26 GMT] DBD::mSQL-0.60pl8 has been released. Here's the extract from the Changes file: o Added error string and number to 'func' methods o Completed implementation of 'ListFields' invoked through the 'func' method o Added implementation of 'CreateDB' and 'DropDB' through the 'func' method o Added a method 'NumRows' against $sth to show the number of rows returned by a statement. The mSQL API is now fully featured within the DBD::mSQL driver, so it's well worth upgrading. Information on DBI in general is available from: http://www.hermetica.com/technologia/DBI The code is available from the usual places: ftp://ftp.mcqueen.com/pub/dbperl/DBD/mSQL/DBD-mSQL-0.60pl8.tar.gz being one of them. CPAN should have it by tomorrow ( Saturday ). Persons using Msqlperl are now advised to think about upgrading onto DBI/DBD::mSQL instead. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <04604-3@oink>; Mon, 27 Nov 1995 07:07:33 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817243229:19131:0; Fri, 24 Nov 95 20:00:29 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ac19019; 24 Nov 95 20:00 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Fri, 24 Nov 1995 19:46:45 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA02340 for dbi-dev-real; Fri, 24 Nov 1995 09:33:24 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA02299; Fri, 24 Nov 1995 09:26:36 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id RAA11217; Fri, 24 Nov 1995 17:23:45 GMT From: Alligator Descartes Message-Id: <199511241723.RAA11217@fruitbat.mcqueen.com> Subject: DBD::mSQL-0.60pl8 released To: dbi-users@fugue.com, dbi-dev@fugue.com, dbi-announce@fugue.com Date: Fri, 24 Nov 1995 17:23:43 +0000 (GMT) Cc: msql-list@bunyip.com, k@anna.mind.de, Tim.Bunce@ig.co.uk, SVEN.VERDOOLAEGE@student.kuleuven.ac.be (Sven Verdoolaege), mrlizard@crl.com X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1014 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Fri, 24 Nov 95 20:00:28 GMT] DBD::mSQL-0.60pl8 has been released. Here's the extract from the Changes file: o Added error string and number to 'func' methods o Completed implementation of 'ListFields' invoked through the 'func' method o Added implementation of 'CreateDB' and 'DropDB' through the 'func' method o Added a method 'NumRows' against $sth to show the number of rows returned by a statement. The mSQL API is now fully featured within the DBD::mSQL driver, so it's well worth upgrading. Information on DBI in general is available from: http://www.hermetica.com/technologia/DBI The code is available from the usual places: ftp://ftp.mcqueen.com/pub/dbperl/DBD/mSQL/DBD-mSQL-0.60pl8.tar.gz being one of them. CPAN should have it by tomorrow ( Saturday ). Persons using Msqlperl are now advised to think about upgrading onto DBI/DBD::mSQL instead. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04698-7@oink>; Mon, 27 Nov 1995 07:18:54 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817408321:01319:0; Sun, 26 Nov 95 17:52:01 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa01112; 26 Nov 95 17:51 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA02310 for dbi-users-real; Sun, 26 Nov 1995 08:06:17 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA02306 for ; Sun, 26 Nov 1995 08:06:12 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id QAA16884; Sun, 26 Nov 1995 16:03:48 GMT From: Alligator Descartes Message-Id: <199511261603.QAA16884@fruitbat.mcqueen.com> Subject: Ramble on DBD::mSQL: $sth->func( '_NumRows' ) To: mrlizard@crl.com (Brian Valente) Date: Sun, 26 Nov 1995 16:03:47 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: from "Brian Valente" at Nov 26, 95 06:59:22 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2846 X-SMTP-Posting-Host: toccata.fugue.com [Sun, 26 Nov 95 17:52:00 GMT] > I have not been able to get the NumRows function to work. I assume it goes > something like this: > > $dbh = $drh->connect(blah blah); > $sh = $dbh->prepare($querystring); > $sh->execute; > $found = $sh->NumRows; > > However, not only do I not get anything back, but it closes the cursor(?) > so I don't receive any subsequent rows. Let me give an example to > illustrate the last: > > I used a search from/html page to display the results. Before it simply > listed the records. I wanted to insert "# records found that matched you > criteria" at the start, but when I inserted the NumRows function, it didn't > print anything out at all! > > I hope it's only because I'm not using it right. Surely is. It was a slightly misleading advert in the 'Changes' file. Although, you could've checked the test.pl to see the method in there. You should be using: $dbh = $drh->connect( blah blah ); $sh = $dbh->prepare( $querystring ); $found = $sh->func( '_NumRows' ); print "Found: $found\n"; $sh->execute; $sh->finish; Hope that clears things up. I've also attached a list of functions/methods you can invoke within DBD::mSQL. This is 'under construction', but will be eventually released with the driver at the next patchlevels/majors. Let me know how things go. > Brian ------------------8<--------------------------8<------------------------- =title DBI Driver Methods =head DBI Specific Handles $drh is a DRIVER HANDLE $dbh a DATABASE HANDLE $sth a STATEMENT HANDLE $rv is a RETURN VALUE =head Error Variables $DBI::errnum $DBI::errstr =head perl Variable Types ( see perl??? ) @ary is a LIST $scalar is an ordinary scalar %hash is a hash =head Methods (*) means non-standard and DBD::mSQL specific (?) means I'm not sure about Invoking through Returns ---------------- ------- Driver Handle Methods: ---------------------- DBI->install_driver $drh $drh->func( '_ListDBs' ) @ary * $drh->func( '_CreateDB' ) $rv * $drh->func( '_DropDB' ) $rv $drh->connect $dbh $drh->disconnect_all $rv (?) Database Handle Methods: ------------------------ $dbh->func( '_ListTables' ) @ary * $dbh->func( '_ListFields' ) %hash * $dbh->disconnect $rv $dbh->do $rv $dbh->prepare $sth Statement Handle Methods: ------------------------- $sth->func( '_NumRows' ) $rv * $sth->execute $rv $sth->fetchrow @ary $sth->finish $rv ----------------8<-----------------------8<------------------------ -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <18457-10@oink>; Tue, 28 Nov 1995 13:11:07 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817562785:22165:0; Tue, 28 Nov 95 12:46:25 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa21915; 28 Nov 95 12:46 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id CAA00692 for dbi-users-real; Tue, 28 Nov 1995 02:01:53 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id CAA00687 for ; Tue, 28 Nov 1995 02:01:47 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id JAA29888; Tue, 28 Nov 1995 09:59:13 GMT From: Alligator Descartes Message-Id: <199511280959.JAA29888@fruitbat.mcqueen.com> Subject: Re: Informix Interface to PERL To: vblack@Candle.com (Vendetta Black) Date: Tue, 28 Nov 1995 09:59:11 +0000 (GMT) Cc: dbi-users@fugue.com, Tim.Bunce@ig.co.uk In-Reply-To: <9510278175.AA817531955@SMTPGWY.CANDLE.COM> from "Vendetta Black" at Nov 27, 95 08:02:22 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1705 X-SMTP-Posting-Host: toccata.fugue.com [Tue, 28 Nov 95 12:46:12 GMT] > I am having a problem with the Informix interface. I can retrieve > data but am unable to retrieve BLOBS particuliarly any text BLOBS. BLOBs aren't supported yet in DBD::Informix. The driver is at a very experimental stage just now, and I haven't had the time to devote to bringing it up to the state of being particularly useful yet. The usefulness of it will increase along with the version number fairly soon, I hope. > I am also receiving a couple messages during the make test. I do > know the what the two hidden messages are. I'll leave this to Tim. [ test results deleted ] > Also, this may be a version 7.1 to 6.0 difference but > there were no INFORMIX libraries in the Makefiles. Well, I only have access to Informix 5.04 ( I think ), so the libraries in the Makefile.PL script reflect that. With help from Dave Rattanni, support for Informix 7 has been added, but if you could send me the necessary library names for Informix 6, we can cover that now, as well. > When I attempt to a SELECT that contains a text BLOB, I do > retrieve the other columns but not the column containing the BLOB. Again, they're not supported yet............... RSN is the best I can say, I'm afraid. Keep watching the mailing lists, and the Hermetica web pages. Also, you should resubscribe to the DBI Mailing Lists. THe perldb-interest list is defunct ( or soon will be ). The new list addresses are either: http://www.fugue.com/dbi via the WWW ( preferred ), or: dbi-users-request@fugue.com dbi-dev-request@fugue.com dbi-announce@fugue.com Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <22464-60@oink>; Wed, 29 Nov 1995 07:17:36 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817600641:17002:0; Tue, 28 Nov 95 23:17:21 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ab10179; 28 Nov 95 23:17 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 28 Nov 1995 22:42:46 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA02859 for dbi-users-real; Tue, 28 Nov 1995 12:16:17 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from clothos.candle.com (clothos.Candle.Com [204.217.194.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA02855 for ; Tue, 28 Nov 1995 12:16:13 -0800 Received: from phobos.Candle.Com by clothos.candle.com with SMTP id AA14725 (5.67b/IDA-1.5 for ); Tue, 28 Nov 1995 12:12:29 -0800 Received: from smtpgwy.Candle.Com by phobos.Candle.Com with SMTP id AA26528 (5.67b/IDA-1.5); Tue, 28 Nov 1995 12:12:22 -0800 Received: from ccMail by SMTPGWY.CANDLE.COM (SMTPLINK V2.10.05) id AA817589585; Tue, 28 Nov 95 12:05:25 PST Date: Tue, 28 Nov 95 12:05:25 PST From: Vendetta Black Message-Id: <9510288175.AA817589585@SMTPGWY.CANDLE.COM> To: Alligator Descartes Cc: dbi-users@fugue.com, Tim.Bunce@ig.co.uk Subject: Re[2]: Informix Interface to PERL X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 28 Nov 95 23:17:20 GMT] From INFORMIX /informix/lib/libasf.a /informix/lib/netlib.a Also from good ole AIX: lbsd ltli The libtli.a is the network piece (ISTAR) which is required under 6.0. I don't think there is a lot of differences between 6.X and 7.X but I will know in about 2 months when I install here. Thanks for telling me about the new mailing lists. Vendetta ______________________________ Reply Separator _________________________________ Subject: Re: Informix Interface to PERL Author: Alligator Descartes at INTERNET Date: 11/28/95 2:05 AM > I am having a problem with the Informix interface. I can retrieve > data but am unable to retrieve BLOBS particuliarly any text BLOBS. BLOBs aren't supported yet in DBD::Informix. The driver is at a very experimental stage just now, and I haven't had the time to devote to bringing it up to the state of being particularly useful yet. The usefulness of it will increase along with the version number fairly soon, I hope. > I am also receiving a couple messages during the make test. I do > know the what the two hidden messages are. I'll leave this to Tim. [ test results deleted ] > Also, this may be a version 7.1 to 6.0 difference but > there were no INFORMIX libraries in the Makefiles. Well, I only have access to Informix 5.04 ( I think ), so the libraries in the Makefile.PL script reflect that. With help from Dave Rattanni, support for Informix 7 has been added, but if you could send me the necessary library names for Informix 6, we can cover that now, as well. > When I attempt to a SELECT that contains a text BLOB, I do > retrieve the other columns but not the column containing the BLOB. Again, they're not supported yet............... RSN is the best I can say, I'm afraid. Keep watching the mailing lists, and the Hermetica web pages. Also, you should resubscribe to the DBI Mailing Lists. THe perldb-interest list is defunct ( or soon will be ). The new list addresses are either: http://www.fugue.com/dbi via the WWW ( preferred ), or: dbi-users-request@fugue.com dbi-dev-request@fugue.com dbi-announce@fugue.com Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <22464-86@oink>; Wed, 29 Nov 1995 07:20:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817607911:21283:0; Wed, 29 Nov 95 01:18:31 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa21272; 29 Nov 95 1:18 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa11283; 29 Nov 95 1:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA03200 for dbi-users-real; Tue, 28 Nov 1995 14:21:30 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA03196 for ; Tue, 28 Nov 1995 14:21:28 -0800 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id RAA04473; Tue, 28 Nov 1995 17:23:52 -0500 Message-Id: <199511282223.RAA04473@pilot.njin.net> To: dbi-users@fugue.com cc: phj@pilot.njin.net Subject: DBI/Oracle speed problems Date: Tue, 28 Nov 1995 17:23:50 -0500 From: "Patricia H. Ju" X-SMTP-Posting-Host: toccata.fugue.com [Wed, 29 Nov 95 1:18:15 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 29 Nov 95 1:18:30 GMT] Hi dbi-users, I have slowness. I am trying to ascertain whether it is a tuning problem (tests seem to imply it is not) or slowness in DBI. The main slowness seems to come from the login connection. On both machines, after invoking 'sqlplus' there's a slight pause (longer on the machine with less memory), then the username prompt, and the password prompt following quickly. Another noticeable lag (3 seconds) before the 'SQL>' prompt shows. From here, queries are quite speedy. This leads me to believe something's stuck in making the connection ? But I don't know where to look to fix something like that. Anyone have speed problems connecting to Oracle DBI ? Do you think it's worth trying oraperl instead ? If anyone recognizes the symptoms, please, please help. Thanks much. Patricia ............. Who needs patience when you've got force ? ------- End of Forwarded Message ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <22464-88@oink>; Wed, 29 Nov 1995 07:21:11 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817607996:21893:0; Wed, 29 Nov 95 01:19:56 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa21889; 29 Nov 95 1:19 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 29 Nov 1995 01:19:44 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA03390 for dbi-users-real; Tue, 28 Nov 1995 15:42:21 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA03382 for ; Tue, 28 Nov 1995 15:42:15 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id XAA14837; Tue, 28 Nov 1995 23:39:53 GMT From: Alligator Descartes Message-Id: <199511282339.XAA14837@fruitbat.mcqueen.com> Subject: Re: DBI/Oracle speed problems To: phj@pilot.njin.net (Patricia H. Ju) Date: Tue, 28 Nov 1995 23:39:51 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: <199511282223.RAA04473@pilot.njin.net> from "Patricia H. Ju" at Nov 28, 95 05:23:50 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2003 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 29 Nov 95 1:19:55 GMT] > I have slowness. I am trying to ascertain whether it is a tuning > problem (tests seem to imply it is not) or slowness in DBI. Could be................ > The main slowness seems to come from the login connection. On both > machines, after invoking 'sqlplus' there's a slight pause (longer on the > machine with less memory), then the username prompt, and the password > prompt following quickly. Another noticeable lag (3 seconds) before > the 'SQL>' prompt shows. From here, queries are quite speedy. Yes, the login stuff is a noticed problem from this person's point of view also. I can, however, tell you that the speed problem doesn't appear to lie within DBI, per se, but in the quantity of memory required to kick off an Oracle-based process. The DBD::Oracle library is 2.6M ( on a Sparc Solaris2.3 box ) which causes my machine to swap it immediately. You essentially then have to wait until it comes back in before it'll connect. Once you're in, though, it's fairly quick. After some testing, I found that all Oracle processes took the same length of time to connect, including DBI ones. > This leads me to believe something's stuck in making the connection ? > But I don't know where to look to fix something like that. Add dispatchers to your instance. That may help you somewhat. It helped here a little bit. > Anyone have speed problems connecting to Oracle DBI ? Do you think > it's worth trying oraperl instead ? Well, remember that oraperl for perl5 is actually another emulation layer on top of DBI on top of the DBD::Oracle driver, so that wouldn't help you at all! > If anyone recognizes the symptoms, please, please help. I bought more memory. That helped considerably! Also, the problems are not as noticable with the Informix or mSQL drivers, but then, their shared library, at least on my machine are approximately 1/20th the size........ > Patricia -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <22464-99@oink>; Wed, 29 Nov 1995 07:22:32 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817624357:25206:0; Wed, 29 Nov 95 05:52:37 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa25202; 29 Nov 95 5:52 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 29 Nov 1995 05:51:59 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA03955 for dbi-users-real; Tue, 28 Nov 1995 20:08:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA03951 for ; Tue, 28 Nov 1995 20:08:37 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id WAA63323; Tue, 28 Nov 1995 22:10:55 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for phj@pilot.njin.net at austin.ibm.com; id AA27178; Tue, 28 Nov 1995 22:10:45 -0600 Message-Id: <9511290410.AA27178@dax.austin.ibm.com> To: Alligator Descartes Cc: phj@pilot.njin.net (Patricia H. Ju), dbi-users@fugue.com Subject: Re: DBI/Oracle speed problems In-Reply-To: Your message of Tue, 28 Nov 95 23:39:51 +0000. <199511282339.XAA14837@fruitbat.mcqueen.com> Date: Tue, 28 Nov 95 22:10:44 -0600 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 29 Nov 95 5:52:37 GMT] Alligator Descartes writes: |> |> > I have slowness. I am trying to ascertain whether it is a tuning |> > problem (tests seem to imply it is not) or slowness in DBI. |> |> Could be................ |> |> > The main slowness seems to come from the login connection. On both |> > machines, after invoking 'sqlplus' there's a slight pause (longer on the |> > machine with less memory), then the username prompt, and the password |> > prompt following quickly. Another noticeable lag (3 seconds) before |> > the 'SQL>' prompt shows. From here, queries are quite speedy. |> |> Yes, the login stuff is a noticed problem from this person's point of view also. |> I can, however, tell you that the speed problem doesn't appear to lie within |> DBI, per se, but in the quantity of memory required to kick off an Oracle-based |> process. The DBD::Oracle library is 2.6M ( on a Sparc Solaris2.3 box ) which |> causes my machine to swap it immediately. You essentially then have to wait |> until it comes back in before it'll connect. Once you're in, though, it's |> fairly quick. |> |> After some testing, I found that all Oracle processes took the same length of |> time to connect, including DBI ones. This has been a problem in Oracle for a really long time and how bad it is depends on who did the port of Oracle. The problem is that Oracle sets up context areas for each connection, which can frequently make the machine page. If you are running a version built by Oracle it will be reasonably fast. However, if you are running a vendored port it might take forever ....... |> > This leads me to believe something's stuck in making the connection ? |> > But I don't know where to look to fix something like that. |> |> Add dispatchers to your instance. That may help you somewhat. It helped |> here a little bit. |> |> > Anyone have speed problems connecting to Oracle DBI ? Do you think |> > it's worth trying oraperl instead ? |> |>Well, remember that oraperl for perl5 is actually another emulation layer on |>top of DBI on top of the DBD::Oracle driver, so that wouldn't help you at all! |> |> > If anyone recognizes the symptoms, please, please help. |> |> I bought more memory. That helped considerably! That is probably the best solution ( is that a :-) or a :-( ). Failing that you could write a daemon which opens an holds a connection to the database, and then have all your other work talk to the daemon. The other thing to check is the isolation level the database is running at. I don't know what the default is for Oracle, but if I had to guess I would say it is "Cursor Stability" by default. Depending on what you are doing this can slow you down considerably. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <26923-14@oink>; Wed, 29 Nov 1995 14:46:41 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817647344:10791:0; Wed, 29 Nov 95 12:15:44 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa10594; 29 Nov 95 12:15 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA05375 for dbi-users-real; Wed, 29 Nov 1995 01:20:19 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay.inferentia.it (dns.inferentia.it [151.99.220.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA05371 for ; Wed, 29 Nov 1995 01:20:16 -0800 Received: from yquem.inferentia.it (yquem.inferentia.it [194.184.62.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id KAA19400; Wed, 29 Nov 1995 10:22:55 +0100 Posted-Date: Wed, 29 Nov 1995 10:22:55 +0100 Received: by yquem.inferentia.it with Microsoft Mail id <01BABE46.1E277340@yquem.inferentia.it>; Wed, 29 Nov 1995 10:33:26 +-100 Message-ID: <01BABE46.1E277340@yquem.inferentia.it> From: Davide Migliavacca To: "dbi-users@fugue.com" Cc: "'Patricia H. Ju'" Subject: RE: DBI/Oracle speed problems Date: Wed, 29 Nov 1995 10:33:25 +-100 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable X-SMTP-Posting-Host: toccata.fugue.com [Wed, 29 Nov 95 12:15:43 GMT] >The main slowness seems to come from the login connection. On both > machines, after invoking 'sqlplus' there's a slight pause (longer on = the Are you using the "user@SID" form of login? Don't; use the ORACLE_SID environment variable. AFAIK, this changes the = underlying SQL*Net transport. In the "user@SID" form, you'll get a = network connection from your host to your host, while dropping the @SID = obtains a local, and speedier, connection. Another possible workaround, if you have enough memory, might be running = a multithreaded server with a couple of dispatchers instead of a = dedicated server. BTW, while you need more memory to set up, it may end = in a lesser memory requirement since it may help optimizing session = creation. This usually helps with lots of connections/disconnections = which is typical of CGI applications. Notice the number of conditionals = in this paragraph 8-) Definitely, however, it's unlikely to be a DBD problem. When you solve = the problem for SQL*Plus, you'll have improved connection times for ALL = possibile client applications. -------------------------------------------------------- Davide Migliavacca (aka shari) Inferentia S.r.l. - Milano ITALY +39 2 59928.1 (FAX .221) * Standard disclaimer here * -------------------------------------------------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <29884-11@oink>; Thu, 30 Nov 1995 07:06:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817672287:22055:0; Wed, 29 Nov 95 19:11:27 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa22050; 29 Nov 95 19:11 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa17068; 29 Nov 95 19:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA07604 for dbi-users-real; Wed, 29 Nov 1995 08:15:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from office.demon.net (office.demon.net [193.195.224.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA07600 for ; Wed, 29 Nov 1995 08:15:30 -0800 Received: from xact.demon.co.uk by office.demon.net id aa23400; 29 Nov 95 16:13 GMT Date: Wed, 29 Nov 1995 16:09:54 +000 (GMT) From: Paul Mahoney X-Sender: ptm@xact4.xact.com To: dbi-users@toccata.fugue.com Subject: DBI and SCO 3.2.4.2 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: toccata.fugue.com [Wed, 29 Nov 95 19:10:55 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 29 Nov 95 19:11:27 GMT] I've looked at some faq's but couldn't find an answer ... yet I'm sure I'm not the first to want to try out perl5-DBI-DBD(Informix) on SCO UNIX 3.2.4.2 Ok, perl5.001m compiled, tested and installed effortlessly :-) DBI compiled and tested out ok (I think, I was difficult to tell) DBD (Informix 0.02pl0) ran ./build got ***** Removing old dbdimp.c ***** ***** Generating new Makefile ***** Can't load module DBI, dynamic loading not available in this perl. (You may need to build a new perl executable which either supports dynamic loading or has the DBI module statically linked into it.) at /usr/local/lib/perl5/DBI.pm line 30 BEGIN failed--compilation aborted at Makefile.PL line 11. ***** Attaching preprocessor information to Makefile ***** ***** Generating MAXCURSOR information ***** Now, when configuring perl5 I took the default choice of NO for dynamic loading ... I don't thick SCO 3.2.4.2 supports it. I haven't found any instructions on how to statically link DBI into perl5. Probably obvious but I thought I ask if someone passed this way before. PS: Is there a Perl 5 book yet? --- Paul Mahoney, X-Act Solutions Limited smail: Owlsmead, Meads Road, Little Common, Bexhill-on-Sea, East Sussex TN39 4SY email: ptm@xact.demon.co.uk phone: +44 424 846368 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <29884-30@oink>; Thu, 30 Nov 1995 07:09:45 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 817688637:26268:0; Wed, 29 Nov 95 23:43:57 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa26262; 29 Nov 95 23:43 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id msg.ad11334; 29 Nov 95 23:43 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 29 Nov 1995 21:44:04 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA08256 for dbi-users-real; Wed, 29 Nov 1995 11:27:18 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA08252 for ; Wed, 29 Nov 1995 11:27:16 -0800 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id OAA06500; Wed, 29 Nov 1995 14:29:05 -0500 Message-Id: <199511291929.OAA06500@pilot.njin.net> To: dbi-users@fugue.com cc: phj@pilot.njin.net Subject: re:Oracle slowness Date: Wed, 29 Nov 1995 14:28:59 -0500 From: "Patricia H. Ju" X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 29 Nov 95 23:43:32 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 29 Nov 95 23:43:57 GMT] Hi all, Thanks for responding. I increased my mts_max_dispatchers to 10 but this hasn't seemed to help.. After altering my CGI perl scripts to make its connections once only, at the beginning of the script, instead of on an as-needed basis farther down, my programs began to respond quite well. 4-5 second delays only. Unfortunately, these changes didn't seem to help on the production machine. The two machines have the same amount of memory and are otherwise similar except the production machine has a much, much higher load. Out of curiosity, I ran the same cgi script on each machine, at the command line. Turns out perl takes about 2 seconds to load the Oracle module on the non-production machine.. and >= 6 seconds (!) to load on the production machine. Is there any way (recompiling with diff options ?) to make this loading quicker, besides buying more memory ? Moving to Msql is my least favorite solution.. but one I might have to take tomorrow. Patricia ........... "Men have forgotten this truth," said the fox. "But you must not forget it. You become responsible, forever, for what you have tamed . . ." ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <00541-6@oink>; Thu, 30 Nov 1995 07:17:08 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817674053:23927:0; Wed, 29 Nov 95 19:40:53 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id aa23778; 29 Nov 95 19:40 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 29 Nov 1995 19:32:58 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA07650 for dbi-users-real; Wed, 29 Nov 1995 08:38:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id IAA07646 for ; Wed, 29 Nov 1995 08:38:12 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id QAA22783; Wed, 29 Nov 1995 16:35:58 GMT From: Alligator Descartes Message-Id: <199511291635.QAA22783@fruitbat.mcqueen.com> Subject: Re: DBI and SCO 3.2.4.2 To: ptm@xact.demon.co.uk (Paul Mahoney) Date: Wed, 29 Nov 1995 16:35:56 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: from "Paul Mahoney" at Nov 29, 95 04:09:54 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1541 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 29 Nov 95 19:40:51 GMT] > I've looked at some faq's but couldn't find an answer ... yet I'm sure I'm > not the first to want to try out perl5-DBI-DBD(Informix) on SCO UNIX 3.2.4.2 > > perl5.001m compiled, tested and installed effortlessly :-) > > DBI compiled and tested out ok (I think, I was difficult to tell) > > DBD (Informix 0.02pl0) > > ran > ./build > got > ***** Removing old dbdimp.c ***** > ***** Generating new Makefile ***** > Can't load module DBI, dynamic loading not available in this perl. > (You may need to build a new perl executable which either supports > dynamic loading or has the DBI module statically linked into it.) > at /usr/local/lib/perl5/DBI.pm line 30 > BEGIN failed--compilation aborted at Makefile.PL line 11. > ***** Attaching preprocessor information to Makefile ***** > ***** Generating MAXCURSOR information ***** > > Now, when configuring perl5 I took the default choice of NO for dynamic > loading ... I don't thick SCO 3.2.4.2 supports it. > > I haven't found any instructions on how to statically link DBI into > perl5. Probably obvious but I thought I ask if someone passed this way > before. The DBI README file tells you, I think: make LINKTYPE=static ought to do it. Then: make perl to build a new perl executable with the module built into it. > PS: Is there a Perl 5 book yet? Haven't seen one, but I could be wrong. Hope that helps. > Paul Mahoney, X-Act Solutions Limited -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <18995-22@oink>; Mon, 4 Dec 1995 07:09:15 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 817855789:02953:0; Fri, 01 Dec 95 22:09:49 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa01169; 1 Dec 95 22:08 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA02883 for dbi-users-real; Fri, 1 Dec 1995 09:16:47 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from eol1 (eol1.eo.lu [193.93.48.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA02879 for ; Fri, 1 Dec 1995 09:16:01 -0800 Received: from [193.93.48.253] by eol1 with smtp (Smail3.1.28.1 #4) id m0tLYNZ-000bKNC; Fri, 1 Dec 95 17:32 MET Message-Id: To: "dbi-users@fugue.com" Subject: Coredump on empty handler Date: Fri, 01 Dec 95 17:30:03 --0100 From: Massimo Balestra X-Mailer: E-Mail Connection v2.5.03 X-SMTP-Posting-Host: toccata.fugue.com [Fri, 1 Dec 95 22:09:01 GMT] -- [ From: Massimo Balestra * EMC.Ver #2.5.02 ] -- Hi everybody, Some weeks ago I wrote a mail like this: > I use DBI 0.63 with the DBD::Oracle 0.25 with the Oraperl emulation. > The platform is an IBM AIX (R6000) RISC machine. > ................... > My problem is: when I am developing the functions to > access the oracle Database and, in the firs testing there is a > syntax error in my SQL statement, I get a Coredump > ................... I tested it with DBI 0.65 and DBD::Oracle 0.27 and seems that this problem is solved :-))))))) I am very happy. But..... I still get a coredump in this case: Look at this example #!/usr/local/bin/perl use Oraperl; $ORA_LDA = &ORA_LOGON; if ($ORA_LDA == -4) { print "Error in the database during the connection\n"; exit 0; } $SELECT = "SELECT field FROM table "; $SH = &ora_open($ORA_LDA, $SELECT); if ($ora_errno > 0) { print "Error\n"; exit -4; } @NR = &ora_fetch($SN); # I wrong the variable name. I passed an empty handler. # In this situation I get the coredump. &ora_close($SH); print "$NR[0]\n"; &ORA_LOGOFF; You know, it is not a real DBD/DBD bug but..... get back an error is better than a coredump. Let look at all informations: Result of perl perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=aix, osver=3.2.5, archname=aix uname='aix bgspectrum2 2 3 002016317100 ' hint=previous Compiler: cc='cc', optimize='-O -g' cppflags='-D_ALL_SOURCE' ccflags ='-D_ALL_SOURCE' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=1, d_casti32=define, d_castneg= intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/lib /usr/lib /usr/ccs/lib /usr/local/lib libs=-ldbm -lld -lm -lc -lbsd -lPW libc=/usr/lib/libc.a Dynamic Linking: dlsrc=dl_aix.xs, dlext=so, d_dlsymun= cccdlflags=' ', ccdlflags='-bE:perl.exp', lddlflags='-H512 -T512 -bhalt: 4 -bM:SRE -bI:$(PERL_INC)/perl.exp -bE:$(BASEEXT).exp -e _nostart -lc' ----------------------------------------------------------------- ----------------------------------------------------------------- ----------------------------------------------------------------- Installation of DBI-0.65 Result of perl Makefile.PL Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker (v4.16) Checking if your kit is complete... Looks good NAME => 'DBI' VERSION => '0.65' Using PERL=/usr/bin/perl Writing Makefile for DBI ----------------------------------------------------------------- Result of make mkdir ./blib mkdir ./blib/auto mkdir ./blib/auto/DBI /usr/bin/perl "-I/usr/local/lib/perl5/aix" "-I/usr/local/lib/perl5" -e 'use ExtUtils::MakeMaker qw(&mksymlists); \ &mksymlists(DL_FUNCS => "", DL_VARS => "", NAME => "DBI")' Useless use of reference constructor in void context at -e line 2. /usr/bin/perl -I/usr/local/lib/perl5/aix -I/usr/local/lib/perl5 /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c cc -c -D_ALL_SOURCE -O -g -I/usr/local/lib/perl5/aix/CORE DBI.c Running Mkbootstrap for DBI () chmod 644 DBI.bs LD_RUN_PATH="" ld -o ./blib/auto/DBI/DBI.so -H512 -T512 -bhalt:4 -bM:SRE - bI:/usr/local/lib/perl5/aix/CORE/perl.exp -bE:DBI.exp -e _nostart -lc DBI.o chmod 755 ./blib/auto/DBI/DBI.so cp DBI.bs ./blib/auto/DBI/DBI.bs chmod 644 ./blib/auto/DBI/DBI.bs mkdir ./blib/DBI cp DBIXS.h ./blib/DBI/DBIXS.h chmod 644 ./blib/DBI/DBIXS.h mkdir ./blib/DBD cp lib/DBD/ExampleP.pm ./blib/DBD/ExampleP.pm chmod 644 ./blib/DBD/ExampleP.pm cp lib/DBD/NullP.pm ./blib/DBD/NullP.pm chmod 644 ./blib/DBD/NullP.pm cp lib/DBD/Sponge.pm ./blib/DBD/Sponge.pm chmod 644 ./blib/DBD/Sponge.pm cp DBI.pm ./blib/DBI.pm chmod 644 ./blib/DBI.pm ----------------------------------------------------------------- Result of make test TEST_VERBOSE=1 /usr/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/aix - I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose =1; runtests @ARGV;' t/*.t t/min...............1..11 DBD::ExampleP in /usr/local/lib/perl5 is hidden by DBD::ExampleP in ./blib at t/min.t line 16 DBD::NullP in /usr/local/lib/perl5 is hidden by DBD::NullP in ./blib at t/min.t line 16 DBD::Sponge in /usr/local/lib/perl5 is hok 1 ok 2 ok 3 idden by DBD::Sponge in ./blib at t/min.t line 16 ok 4 ok 4 ok 5 ok 6 ok 7 ok 8 ok 9 ok 10 ok 11 ok All tests successful. Files=1, Tests=11, 0 secs ( 0.26 cusr 0.05 csys = 0.31 cpu) /usr/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/aix - I/usr/local/lib/perl5 test.pl test.pl DBI test application $Revision: 1.12 $ Switch: DBI Switch by Tim Bunce, 1.48 DebugDispatch: 0 DBD::ExampleP in /usr/local/lib/perl5 is hidden by DBD::ExampleP in ./blib at test.pl line 49 DBD::NullP in /usr/local/lib/perl5 is hidden by DBD::NullP in ./blib at test.pl line 49 DBD::Sponge in /usr/local/lib/perl5 is hidden by DBD::Sponge in ./blib at test.pl line 49 Available Drivers: ExampleP, NullP, Sponge, Oracle Read DBI special variables (expect 0, 99, 99): err: err: 99 errstr: 99 Attempt to modify DBI special variables. Expect a 'Can't modify' error message: Can't modify $DBI::&rows special variable at test.pl line 59 Driver installed as DBI::dr=HASH(0x2010797c) Connected as DBI::db=HASH(0x2012b510) Test error handling: prepare invalid query. Expect an ERROR EVENT message: prepare returned UNDEF. $DBI::err= Cursor prepare'd as DBI::st=HASH(0x2012b678) Prepared as DBI::st=HASH(0x2012b750) Number of fields: 3 Data type of first field: 1 Driver name: ExampleP Fetching data from both cursors. Expect several rows of data: 82431 2 ., 8 512 . 82431 2 .., 8 512 .. 82413 2048 lpp, 8 512 cron 82431 65868 bin, 8 512 lpd 82413 51808 ccs, 8 512 mail 82413 68112 include, 8 512 mqueue 82413 70203 lbin, 8 512 qdaemon 82431 111154 lib, 8 512 secretmail 82368 127990 lost+found, 8 512 writesrv 82413 127991 mbin, 8 512 uucp 82413 127999 sbin, 8 512 uucppublic 82413 129062 share, 8 512 rwho Automatic method parameter usage check. Expect a 'DBI ... invalid' error and a 'Usage: ...' message: DBI commit: invalid number of parameters: handle + 1 Usage: $h->commit() at test.pl line 130. Preparing new $cursor_a to replace current $cursor_a. We enable debugging on current to watch it's destruction. Expect several lines of DBI trace information: DBI::st=HASH(0x2012b774) debug level set to 2 -> DESTROY for DBD::ExampleP::st (DBI::st=HASH(0x2012b774)) <- DESTROY= undef Executing via func redirect: $h->func(..., 'execute') Binding columns of $cursor_a to variables. Fetching one row from new $cursor_a with a bound column. Expect a large number follwed by a dot: 817052704 . bind_col worked Cursor tests done (scoped objects will be destroyed now) test.pl done (global destruction will follow) ----------------------------------------------------------------- ----------------------------------------------------------------- ----------------------------------------------------------------- Installation of DBD-Oracle-0.27 Result of perl Makefile.PL /u01/home/oracle/product/7.1.4/rdbms/lib/oracle. mk Configuring DBD::Oracle version 0.27... Remember to actually read the README file! Using Oracle in /u01/home/oracle/product/7.1.4 Oracle sysliblist: -lm -lld -lm Using /u01/home/oracle/product/7.1.4/rdbms/lib/oracle.mk version 7.87 System: perl5.001 aix bgspectrum2 2 3 002016317100 aix dl_aix.xs Compiler: cc -O -g -D_ALL_SOURCE Oracle proc.mk would have used these values but we override them: CC: oraxlc CFLAGS: -I. -O LDFLAGS: -L$(LIBHOME) Checking if your kit is complete... Looks good Writing Makefile for DBD::Oracle ----------------------------------------------------------------- Result of make mkdir ./blib mkdir ./blib/DBD mkdir ./blib/auto mkdir ./blib/auto/DBD mkdir ./blib/auto/DBD/Oracle /usr/bin/perl "-I/usr/local/lib/perl5/aix" "-I/usr/local/lib/perl5" -e 'use ExtUtils::MakeMaker qw(&mksymlists); \ &mksymlists(DL_FUNCS => "", DL_VARS => "", NAME => "DBD::Oracle")' Useless use of reference constructor in void context at -e line 2. /usr/bin/perl -I/usr/local/lib/perl5/aix -I/usr/local/lib/perl5 /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c cc -c -D_ALL_SOURCE -O -g -I/usr/local/lib/perl5/aix/CORE - I/u01/home/oracle/product/7.1.4/rdbms/demo -I/u01/home/oracle/product/7.1. 4/rdbms/demo -I/usr/local/lib/perl5/aix/DBI Oracle.c cc -c -D_ALL_SOURCE -O -g -I/usr/local/lib/perl5/aix/CORE - I/u01/home/oracle/product/7.1.4/rdbms/demo -I/u01/home/oracle/product/7.1. 4/rdbms/demo -I/usr/local/lib/perl5/aix/DBI dbdimp.c Running Mkbootstrap for DBD::Oracle () chmod 644 Oracle.bs LD_RUN_PATH="" ld -H512 -T512 -bhalt:4 -o ./blib/auto/DBD/Oracle/Oracle.so -H512 -T512 -bhalt:4 -bM:SRE -bI:/usr/local/lib/perl5/aix/CORE/perl.exp -bE: Oracle.exp -e _nostart -lc Oracle.o dbdimp.o -L/u01/home/oracle/product/7.1 .4/lib /u01/home/oracle/product/7.1.4/lib/libocic.a /u01/home/oracle/product/7.1.4/lib/osntab.o -lsqlnet -lora -lsqlnet -lora -lnlsrtl -lcv6 -lcore -lnlsrtl -lcv6 -lcore `cat /u01/home/oracle/product/7.1.4/rdbms/lib/sysliblist` /lib/crt0.o -lc -bI: /u01/home/oracle/product/7.1. chmod 755 ./blib/auto/DBD/Oracle/Oracle.so cp Oracle.bs ./blib/auto/DBD/Oracle/Oracle.bs chmod 644 ./blib/auto/DBD/Oracle/Oracle.bs cp Oraperl.pm ./blib/Oraperl.pm chmod 644 ./blib/Oraperl.pm cp oraperl.ph ./blib/oraperl.ph chmod 644 ./blib/oraperl.ph cp Oracle.pm ./blib/DBD/Oracle.pm chmod 644 ./blib/DBD/Oracle.pm ----------------------------------------------------------------- Result of make test TEST_VERBOSE=1 /usr/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/aix - I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose =1; runtests @ARGV;' t/*.t t/base..............1..5 ok 1 ok 2 ok 3 ok 4 ok 5 ok t/main..............t/main.t DBI test application $Revision: 1.1 $ 1..1 ok 1 ok All tests successful. Files=2, Tests=6, 1 secs ( 0.61 cusr 0.13 csys = 0.74 cpu) /usr/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/aix - I/usr/local/lib/perl5 test.pl Oraperl test application $Revision: 1.19 $ Oraperl Emulation Interface version 1.23 Oracle Driver 1.21 DBI Switch by Tim Bunce, version 1.48 Connecting to 'ATMNET' (from command line, else uses ORACLE_SID/TWO_TASK) as '/' (via ORACLE_USERID environment var or default) Fields: 4 Names: 'NUM_T', 'DATE_T', 'CHAR_T', 'NULL_T' Lengths: '40', '75', '30', '0' Types: '2', '12', '1', '1' Data rows: fetch: 79.2 22-NOV-95 OPS$MBA NULL csr reassigned (forces destruction)... Fetch list of tables: Fetched: DUAL Fetched: SYSTEM_PRIVILEGE_MAP Fetched: TABLE_PRIVILEGE_MAP Fetched: STMT_AUDIT_OPTION_MAP Test ora_do with harmless non-select statement (set transaction read only) Expect error message: ora_do: 1453: ORA-01453: SET TRANSACTION must be first statement of transaction (DBD: oexec error) csr out of scope... ora_logoff... lda out of scope... Testing repetitive connect/open/close/disconnect: Expect sequence of digits, no other messages: 1 2 3 4 5 Test interaction of explicit close/logoff and implicit DESTROYs Expect just 'done.', no other messages: done. Test complete (1 seconds). ----------------------------------------------------------------- ----------------------------------------------------------------- ----------------------------------------------------------------- Running the test program (the example you saw before) I get the following informations: dbx Version 3.1 Type 'help' for help. reading symbolic information ... (dbx) r segmentation violation in dbih_getcom at line 238 in file "DBI.c" 238 if (SvRMAGICAL(sv) && (mg=SvMAGIC(sv))->mg_type == DBI_MAGIC) { (dbx) where dbih_getcom(hrv = 0x20072c0c), line 238 in "DBI.c" XS_DBD__Oracle__st_fetchrow(cv = 0x2006eb08), line 382 in "Oracle.c" pp_entersub(), line 1658 in "pp_hot.c" run.run(), line 29 in "run.c" perl_run(sv_interp = (nil)), line 516 in "perl.c" main(argc = 0, argv = (nil), env = (nil)), line 34 in "perlmain.c" (dbx) q ----------------------------------------------------------------- I hope I gave you enough informations. Thank you Max -- __________________________________________________________________ o _______________ /\_ _| | Massimo Balestra _\__`[_______________| Email mbalestra@flower.it ] [ \, ][ ][ __________________________________________________________________ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <28240-19@oink>; Tue, 5 Dec 1995 07:09:22 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818109293:20643:0; Mon, 04 Dec 95 20:34:53 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ad19692; 4 Dec 95 20:34 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Mon, 4 Dec 1995 20:04:20 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA05242 for dbi-users-real; Mon, 4 Dec 1995 09:07:03 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA05238 for ; Mon, 4 Dec 1995 09:07:00 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id RAA04490; Mon, 4 Dec 1995 17:02:22 GMT From: Alligator Descartes Message-Id: <199512041702.RAA04490@fruitbat.mcqueen.com> Subject: Re: DBI/Oracle speed problems To: Tim.Bunce@ig.co.uk (Tim Bunce) Date: Mon, 4 Dec 1995 17:02:21 +0000 (GMT) Cc: dbi-users@fugue.com, phj@pilot.njin.net In-Reply-To: <9512041255.AA27326@toad.ig.co.uk> from "Tim Bunce" at Dec 4, 95 12:55:05 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 494 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Mon, 4 Dec 95 20:34:53 GMT] > Several people have replied with good thoughts. No one mentioned the > README: The README? Now, Tim, that's the *last* place any of us would expect to see helpful advice. Do the normal thing. Don't include it, or stick it in as a comment at line 1026 of some random bit of source code......... 8-) > Tim. > > p.s. Standby for post-vacation announcements :-) *dives for cover* -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <28240-23@oink>; Tue, 5 Dec 1995 07:10:04 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818109315:20867:0; Mon, 04 Dec 95 20:35:15 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ad19692; 4 Dec 95 20:35 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Mon, 4 Dec 1995 20:01:36 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA05206 for dbi-users-real; Mon, 4 Dec 1995 08:40:20 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA05202 for ; Mon, 4 Dec 1995 08:40:15 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ai17854; 4 Dec 95 16:40 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ao04233; 4 Dec 95 16:37 GMT Received: from toad by oink with SMTP (PP) id <24899-0@oink>; Mon, 4 Dec 1995 12:55:32 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA27326; Mon, 4 Dec 1995 12:55:05 +0000 Date: Mon, 4 Dec 1995 12:55:05 +0000 From: Tim Bunce Message-Id: <9512041255.AA27326@toad.ig.co.uk> To: dbi-users@fugue.com, phj@pilot.njin.net Subject: Re: DBI/Oracle speed problems X-Sun-Charset: US-ASCII content-length: 1595 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Mon, 4 Dec 95 16:37:34 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Mon, 4 Dec 95 16:40:53 GMT] X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Mon, 4 Dec 95 20:35:14 GMT] > From: "Patricia H. Ju" > > Hi dbi-users, > > I have slowness. I am trying to ascertain whether it is a tuning > problem (tests seem to imply it is not) or slowness in DBI. > > The main slowness seems to come from the login connection. On both > machines, after invoking 'sqlplus' there's a slight pause (longer on the > machine with less memory), then the username prompt, and the password > prompt following quickly. Another noticeable lag (3 seconds) before > the 'SQL>' prompt shows. From here, queries are quite speedy. > > This leads me to believe something's stuck in making the connection ? > But I don't know where to look to fix something like that. Several people have replied with good thoughts. No one mentioned the README: Oracle 7.1.3/4: Connection takes a long time and may coredump Oracle bug number: 227321 related to changing the environment before connecting to oracle. To work around this bug, do not set any environment variables in your oraperl script before you call ora_login, and when you do call ora_login, the first argument must be the empty string. This means that you have to be sure that your environment variables ORACLE_SID and ORACLE_HOME are set properly before you execute any oraperl script. It is probably also possible to pass the SID to ora_login as part of the username (for example, ora_login("", "SCOTT/TIGER@PROD", "")), although I have not tested this. This workaround is based on information from Kevin Stock. That's something else worth thinking about. Tim. p.s. Standby for post-vacation announcements :-) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <28240-34@oink>; Tue, 5 Dec 1995 07:12:52 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818116746:05183:0; Mon, 04 Dec 95 22:39:06 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa05171; 4 Dec 95 22:39 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id sg.aa02278; 4 Dec 95 22:38 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Mon, 4 Dec 1995 22:34:13 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA05592 for dbi-users-real; Mon, 4 Dec 1995 11:37:39 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from colossus.cse.psu.edu (root@colossus.cse.psu.edu [130.203.1.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA05588 for ; Mon, 4 Dec 1995 11:37:36 -0800 Received: from localhost by colossus.cse.psu.edu with SMTP id <78536>; Mon, 4 Dec 1995 14:40:06 -0500 X-Mailer: exmh version 1.6.1 5/23/95 To: Tim Bunce cc: dbi-users@fugue.com Subject: Re: DBI/Oracle speed problems In-reply-to: Your message of "Mon, 04 Dec 1995 07:55:05 EST." <9512041255.AA27326@toad.ig.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 4 Dec 1995 14:39:09 -0500 From: John D Groenveld Message-Id: <95Dec4.144006est.78536@colossus.cse.psu.edu> X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Mon, 4 Dec 95 22:38:26 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Mon, 4 Dec 95 22:39:05 GMT] ora_login("", "SCOTT/TIGER\@SID", "") dies... ora_login("", "SCOTT/TIGER\@SQLNet2ServiceName", "") and ora_login("", "SCOTT\@SQLNet2ServiceName", "") work fine. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <28240-41@oink>; Tue, 5 Dec 1995 07:14:02 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818119606:03894:0; Mon, 04 Dec 95 23:26:46 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa03888; 4 Dec 95 23:26 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id sg.aa07661; 4 Dec 95 23:25 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Mon, 4 Dec 1995 23:25:22 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA05856 for dbi-users-real; Mon, 4 Dec 1995 13:27:23 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA05852 for ; Mon, 4 Dec 1995 13:27:15 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id VAA06390; Mon, 4 Dec 1995 21:24:23 GMT From: Alligator Descartes Message-Id: <199512042124.VAA06390@fruitbat.mcqueen.com> Subject: Re: FWD>Returned Mail To: Norman Ho Date: Mon, 4 Dec 1995 21:24:21 +0000 (GMT) Cc: dbi-users@fugue.com, Tim.Bunce@ig.co.uk In-Reply-To: from "Norman Ho" at Dec 4, 95 02:06:55 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 921 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Mon, 4 Dec 95 23:25:56 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Mon, 4 Dec 95 23:26:45 GMT] > As I understood after looking at the DBI specification that DBI can > not call PL/SQL functions/procedures which is integrated into a specific > Oracle Database... Am I right? Can Oraperl do this? Would you please > give me a hint?? Thank you very much for your time, DBI, per se, cannot call PL/SQL, however, the DBD::Oracle can ( or at least will be able to soon ). I think this is a 'about-to-be-implemented' feature of Oracle. It would appear as a private function to the Oracle driver. Under perl5, oraperl wouldn't be able to do it, because oraperl is simply a translation module that sits on top of the raw DBD::Oracle driver. However, I'd double check with Tim Bunce about this one, since I'm not 100% sure on it. His email address is: Tim.Bunce@ig.co.uk > Norman Ho from FermiLab Regards. -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01999-7@oink>; Tue, 5 Dec 1995 15:12:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818157980:15307:0; Tue, 05 Dec 95 10:06:20 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa15275; 5 Dec 95 10:06 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA07247 for dbi-users-real; Mon, 4 Dec 1995 23:03:33 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA07243 for ; Mon, 4 Dec 1995 23:03:30 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id XAA06742; Mon, 4 Dec 1995 23:05:18 -0800 Message-Id: <199512050705.XAA06742@voro.lbl.gov> X-Mailer: exmh version 1.5.3 12/28/94 To: Tim Bunce cc: dbi-users@fugue.com, phj@pilot.njin.net Subject: Re: DBI/Oracle speed problems In-reply-to: Your message of "Mon, 04 Dec 1995 12:55:05 GMT." <9512041255.AA27326@toad.ig.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 04 Dec 1995 23:05:18 -0800 From: Mark Dedlow X-SMTP-Posting-Host: toccata.fugue.com [Tue, 5 Dec 95 10:06:18 GMT] For what it's worth, I tried a few tests to compare startup times. I compared DBI to SQLPLUS on, and between, 2 systems. The results are all good in my opinion, but some interesting. Client: Sparc10 80MB Solaris 2.4 Oracle 7.2.2 Server: Sparc1000 704MB Solaris 2.4 Oracle 7.1.6 These are avg times (seconds) of 10 tests each. Exec on connect to DBI SQLPLUS ------- --------------------- ------ --------- Client client db 1.1 0.7 Client client db via sqlnet 1.1 0.7 Client server db 1.4 0.8 Server server db 2.2 0.7 Server server db via sqlnet 2.0 0.8 The things that are interesting to me are: 1) My Sparc10 is twice as fast as my Sparc1000, for DBI, but no difference with SQLPLUS. 2) No slower to a remote database than to a local one (thru 2 routers) 3) No slower going via sqlnet than going via IPC on a local database. Here is the test fragment: #!/usr/local/bin/perl use DBI; $drh = DBI->install_driver("Oracle"); $dbh = $drh->connect("", "username\@sqlnetv2alias", "password"); ## $dbh = $drh->connect("", "username", "password"); (for local db) $query = "select sysdate from dual"; $cursor = $dbh->prepare($query); $cursor->execute; print (@field,"\n") while (@field = $cursor->fetchrow); $cursor->finish; $dbh->disconnect; sqlplus was tested like this: sqlplus username/password@sqlnetv2alias @doit.sql (remote) sqlplus username/password @doit.sql (local) doit.sql: select sysdate from dual; exit; Mark > > From: "Patricia H. Ju" > > > > Hi dbi-users, > > > > I have slowness. I am trying to ascertain whether it is a tuning > > problem (tests seem to imply it is not) or slowness in DBI. > > > > The main slowness seems to come from the login connection. On both > > machines, after invoking 'sqlplus' there's a slight pause (longer on the > > machine with less memory), then the username prompt, and the password > > prompt following quickly. Another noticeable lag (3 seconds) before > > the 'SQL>' prompt shows. From here, queries are quite speedy. > > > > This leads me to believe something's stuck in making the connection ? > > But I don't know where to look to fix something like that. > > Several people have replied with good thoughts. No one mentioned the > README: > > Oracle 7.1.3/4: Connection takes a long time and may coredump > > Oracle bug number: 227321 related to changing the environment before > connecting to oracle. > > To work around this bug, do not set any environment variables in your > oraperl script before you call ora_login, and when you do call > ora_login, the first argument must be the empty string. This means > that you have to be sure that your environment variables ORACLE_SID > and ORACLE_HOME are set properly before you execute any oraperl > script. It is probably also possible to pass the SID to ora_login as > part of the username (for example, ora_login("", "SCOTT/TIGER@PROD", > "")), although I have not tested this. > This workaround is based on information from Kevin Stock. > > That's something else worth thinking about. > > Tim. > > p.s. Standby for post-vacation announcements :-) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05112-15@oink>; Wed, 6 Dec 1995 07:07:12 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818192828:04677:0; Tue, 05 Dec 95 19:47:08 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa04672; 5 Dec 95 19:47 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa15125; 5 Dec 95 19:46 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA09534 for dbi-users-real; Tue, 5 Dec 1995 07:13:41 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA09530 for ; Tue, 5 Dec 1995 07:13:38 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.ab07606; 5 Dec 95 15:13 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ab10648; 5 Dec 95 15:12 GMT Received: from toad by oink with SMTP (PP) id <01170-0@oink>; Tue, 5 Dec 1995 10:30:14 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA00191; Tue, 5 Dec 1995 10:29:43 +0000 Date: Tue, 5 Dec 1995 10:29:43 +0000 From: Tim Bunce Message-Id: <9512051029.AA00191@toad.ig.co.uk> To: descarte@hermetica.com Subject: Re: DBI/Oracle speed problems Cc: dbi-users@fugue.com, phj@pilot.njin.net X-Sun-Charset: US-ASCII content-length: 449 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 5 Dec 95 15:12:03 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 5 Dec 95 15:13:50 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 5 Dec 95 19:46:55 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 5 Dec 95 19:47:07 GMT] > From: Alligator Descartes > > > Several people have replied with good thoughts. No one mentioned the > > README: > > The README? Now, Tim, that's the *last* place any of us would expect to > see helpful advice. > > Do the normal thing. Don't include it, or stick it in as a comment at line > 1026 of some random bit of source code......... > > 8-) Jack Shirazi recently suggested renaming it to FREE.SEX :-) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05112-25@oink>; Wed, 6 Dec 1995 07:08:12 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818196137:07860:0; Tue, 05 Dec 95 20:42:17 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa07855; 5 Dec 95 20:42 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa20538; 5 Dec 95 20:41 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA09808 for dbi-users-real; Tue, 5 Dec 1995 09:20:55 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA09804 for ; Tue, 5 Dec 1995 09:20:53 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ae22216; 5 Dec 95 17:18 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa07111; 5 Dec 95 17:17 GMT Received: from toad by oink with SMTP (PP) id <02713-0@oink>; Tue, 5 Dec 1995 17:16:09 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA01494; Tue, 5 Dec 1995 17:15:41 +0000 Date: Tue, 5 Dec 1995 17:15:41 +0000 From: Tim Bunce Message-Id: <9512051715.AA01494@toad.ig.co.uk> To: dbi-users@fugue.com, mbalestra@flower.it Subject: Re: Coredump on empty handler X-Sun-Charset: US-ASCII content-length: 627 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 5 Dec 95 17:17:04 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 5 Dec 95 17:18:36 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 5 Dec 95 20:41:55 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 5 Dec 95 20:42:17 GMT] > From: Massimo Balestra > > $SH = &ora_open($ORA_LDA, $SELECT); > .... > &ora_fetch($SN) In the name of efficiency I have sinned. (Many have sinned in the name of efficiency, but that's no excuse :-) Basically, by default, ora_close is an alias for DBD::Oracle::st::finish(). It is thus not called as a method and so bypasses the method lookup that would have otherwise failed. Adding the following line to your application _above_ use Oraperl should prove this: BEGIN { $Oraperl::safe = 1 } See Oraperl.pm for more explanation. Fast or safe? You pays ya money, you takes ya choice. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <05113-4@oink>; Wed, 6 Dec 1995 07:06:16 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818202127:16485:0; Tue, 05 Dec 95 22:22:07 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa16377; 5 Dec 95 22:22 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA10201 for dbi-users-real; Tue, 5 Dec 1995 11:30:01 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA10194 for ; Tue, 5 Dec 1995 11:29:57 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ag04534; 5 Dec 95 18:45 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ab01236; 5 Dec 95 18:43 GMT Received: from toad by oink with SMTP (PP) id <03621-0@oink>; Tue, 5 Dec 1995 18:22:47 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA01902; Tue, 5 Dec 1995 18:22:19 +0000 Date: Tue, 5 Dec 1995 18:22:19 +0000 From: Tim Bunce Message-Id: <9512051822.AA01902@toad.ig.co.uk> To: Chris_Wolcott@ccmail.va.grci.com Subject: ORAPERL, Sorry to bother you directly Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 2456 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 5 Dec 95 18:43:37 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 5 Dec 95 18:45:40 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 5 Dec 95 22:22:05 GMT] Forwarded to the dbi-users list. ----- Begin Included Message ----- From Chris_Wolcott@ccmail.va.grci.com Fri Nov 17 16:19:06 1995 Date: Fri, 17 Nov 95 11:04:58 EST From: Chris Wolcott To: tim.bunce@ig.co.uk Return-Receipt-To: Chris_Wolcott@ccmail.va.grci.com Subject: ORAPERL, Sorry to bother you directly Question about ORAPERL and LONG RAW types. PERL Version 4.0.1.8 Patch 36 ORAPERL Version 2.004 ORACLE Server Version 7.1.3 SYSTEM SCO Unix and DEC Alpha I posted an article on COMP.DATABASES.ORACLE and COMP.LANG.PERL about this very question and received a lot of responses that stated when I found the answer could I forward it to them. I have kept all the hungry animals emails in preparation for an answer but none every came, so I am now asking you directly in hopes that my requirement and everyone elses appetite can be satisfied. ** QUESTION ** -------------- Do you have any simple examples on inserting and selecting information from ORACLE with column types of LONG RAW? Specifically I wish to put a word document into the column. We have done this through PRO C but have not had success with ORAPERL. One of my room mates is getting something like "datatypes inconsistant" when trying to fetch and I seem to hang when trying to insert. ** SAMPLE INSERT CODE BEING USED ** ----------------------------------- #!/usr/bin/oraperl $lda = &ora_login ($DB, $USER, $PASS) || die "$ora_errstr\n"; #------------------------------------------------------------ $ora_long = 32000; undef $/; open (longfile, "./facsend.doc"); binmode longfile; # Not need since on UNIX $data = ; $/ = "\n"; close (longfile); $SQL "insert into TEMP_LONG_RAW values (1005, '$data')"; #$SQL "insert into TEMP_LONG_RAW values (1005, $data)"; &ora_do($lda, $SQL) || die "$ora_errstr\n"; #------------------------------------------------------------- &ora_logoff($lda); exit; Again sorry to bother you directly, but any help would be appreciated and passed on to others. Thanks. Chris Wolcott (cwolcott@grci.com) (703) 506-5422 (www.grci.com) ----- End Included Message ----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05112-73@oink>; Wed, 6 Dec 1995 07:11:01 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818219349:05978:0; Wed, 06 Dec 95 03:09:09 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa05937; 6 Dec 95 3:09 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA11194 for dbi-users-real; Tue, 5 Dec 1995 17:36:11 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from motgate.mot.com (motgate.mot.com [129.188.136.100]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA11190 for ; Tue, 5 Dec 1995 17:36:05 -0800 Received: from pobox.mot.com (pobox.mot.com [129.188.137.100]) by motgate.mot.com (8.7.1/8.6.10/MOT-3.8) with ESMTP id TAA24546 for ; Tue, 5 Dec 1995 19:38:10 -0600 (CST) Received: from po_box.cig.mot.com (po_box.cig.mot.com [136.182.15.5]) by pobox.mot.com (8.7.1/8.6.10/MOT-3.8) with SMTP id TAA17263 for ; Tue, 5 Dec 1995 19:38:09 -0600 (CST) Message-Id: <199512060138.UAA23687@po_box.cig.mot.com> Received: (chinay@localhost) by booboo.cig.mot.com (8.6.9/SCERG-1.12B) id TAA02533 for dbi-users@fugue.com; Tue, 5 Dec 1995 19:38:08 -0600 From: Albert Chin-A-Young Subject: problems with DBI-Oracle-0.27 To: dbi-users@fugue.com Date: Tue, 5 Dec 1995 19:38:08 -0600 (CST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Wed, 6 Dec 95 3:09:08 GMT] When running the test supplied with DBI-Oracle-0.27 with perl-5.001n on a Motorola SVR4 box, I get the following error: PERL_DL_NONLAZY=1 /usr/misc/bin/perl5 -I./blib/m88k-svr4 -I./blib -I/usr/misc/perl5/lib/m88k-svr4 -I/usr/misc/perl5/lib -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/base..............install_driver(Oracle) failed: Can't load './blib/m88k-svr4/auto/DBD/Oracle/Oracle.so' for module DBD::Oracle: dynamic linker: /usr/misc/bin/perl5: relocation error: symbol not found: oerhms at /usr/misc/perl5/lib/DynaLoader.pm line 136. at blib/DBD/Oracle.pm line 18 at /usr/misc/perl5/lib/DBI.pm line 180 DBI::install_driver called at t/base.t line 16 FAILED on test 4 t/main..............ok Failed 1 test, 50.00% okay. -- albert chin (its service team 2) ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <12147-9@oink>; Wed, 6 Dec 1995 20:40:56 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818276580:26818:0; Wed, 06 Dec 95 19:03:00 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa26744; 6 Dec 95 19:02 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA14314 for dbi-users-real; Wed, 6 Dec 1995 07:49:11 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from torga.ci.uminho.pt (torga.ci.uminho.pt [193.136.16.251]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA14310 for ; Wed, 6 Dec 1995 07:48:47 -0800 Received: from pessoa by torga.ci.uminho.pt (5.4R3.10/140.2) id AA06134; Wed, 6 Dec 1995 16:47:05 -0500 Received: from athena by pessoa.ci.uminho.pt (5.4R3.10/140.2) id AA24019; Wed, 6 Dec 1995 16:45:16 +0100 Received: from athena (localhost) by athena.ci.uminho.pt (5.0/SMI-SVR4) id AA20270; Wed, 6 Dec 1995 16:48:21 +0100 Message-Id: <9512061548.AA20270@athena.ci.uminho.pt> Date: Wed, 06 Dec 95 16:48:21 0100 From: Luis Gaspar Organization: Universidade do Minho X-Mailer: Mozilla 1.12 (X11; I; SunOS 5.3 sun4d) Mime-Version: 1.0 To: dbi-users@fugue.com Subject: Sockets X-Url: http://www.hermetica.com/technologia/DBI/ Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Content-Length: 209 X-SMTP-Posting-Host: toccata.fugue.com [Wed, 6 Dec 95 19:02:54 GMT] I'am having troubles with sockets, to implement a client-server aplication. Can anyone show me some examples or related code that might help me. Thanks, Luis Gaspar University of Minho - Portugal ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12168-9@oink>; Wed, 6 Dec 1995 20:41:24 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818276399:13125:0; Wed, 06 Dec 95 18:59:59 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa13115; 6 Dec 95 18:59 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa03928; 6 Dec 95 18:59 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA14308 for dbi-users-real; Wed, 6 Dec 1995 07:48:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from borcore.usbr.gov (borcore.usbr.gov [140.215.19.5]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA14304 for ; Wed, 6 Dec 1995 07:48:11 -0800 Received: from MTREVINO (matrevino.do.usbr.gov) by borcore.usbr.gov (4.1/SMI-4.1) id AA12616; Wed, 6 Dec 95 08:48:05 MST Message-Id: <9512061548.AA12616@borcore.usbr.gov> Date: Wed, 06 Dec 95 08:48:06 -0700 From: "Mark A. Trevino" Organization: U.S. Bureau of Reclamation X-Mailer: Mozilla 1.2N (Windows; I; 16bit) Mime-Version: 1.0 To: dbi-users@fugue.com Subject: (no subject) Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii X-SMTP-Posting-Host: toccata.fugue.com [Wed, 6 Dec 95 18:59:26 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 6 Dec 95 18:59:58 GMT] subscribe ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <13009-2@oink>; Wed, 6 Dec 1995 21:22:58 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818284904:18385:0; Wed, 06 Dec 95 21:21:44 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa18276; 6 Dec 95 21:21 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA14766 for dbi-users-real; Wed, 6 Dec 1995 10:45:54 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA14762 for ; Wed, 6 Dec 1995 10:45:36 -0800 Received: by relay-4.mail.demon.net id msg.aa09826; 6 Dec 95 18:42 GMT Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ag22961; 6 Dec 95 17:14 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ac19784; 6 Dec 95 17:11 GMT Received: from toad by oink with SMTP (PP) id <11186-0@oink>; Wed, 6 Dec 1995 17:05:24 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA05389; Wed, 6 Dec 1995 17:04:55 +0000 Date: Wed, 6 Dec 1995 17:04:55 +0000 From: Tim Bunce Message-Id: <9512061704.AA05389@toad.ig.co.uk> To: dedlow@voro.lbl.gov Subject: Re: DBI/Oracle speed problems Cc: dbi-users@fugue.com, phj@pilot.njin.net X-Sun-Charset: US-ASCII content-length: 1442 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Wed, 6 Dec 95 17:11:03 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Wed, 6 Dec 95 17:14:08 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Wed, 6 Dec 95 21:21:42 GMT] > From: Mark Dedlow > > For what it's worth, I tried a few tests to compare startup times. > I compared DBI to SQLPLUS on, and between, 2 systems. The results are > all good in my opinion, but some interesting. > > Client: Sparc10 80MB Solaris 2.4 Oracle 7.2.2 > Server: Sparc1000 704MB Solaris 2.4 Oracle 7.1.6 > > These are avg times (seconds) of 10 tests each. > > Exec on connect to DBI SQLPLUS > ------- --------------------- ------ --------- > Client client db 1.1 0.7 > Client client db via sqlnet 1.1 0.7 > Client server db 1.4 0.8 > Server server db 2.2 0.7 > Server server db via sqlnet 2.0 0.8 > > The things that are interesting to me are: > > 1) My Sparc10 is twice as fast as my Sparc1000, for DBI, but > no difference with SQLPLUS. > 2) No slower to a remote database than to a local one (thru 2 routers) > 3) No slower going via sqlnet than going via IPC on a local database. Strange. To get a more complete picture you'd need to run an empty test. E.g., just connect/disconnect in perl and an empty file for sqlplus. The difference then gives you the prepare/execute/fetch speed. For full marks you could also try: $dbh->prepare($query, { ora_parse_defer => 1 }); but I doubt it'll make a measurable difference on one query. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <13546-7@oink>; Wed, 6 Dec 1995 23:19:26 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818291598:20398:0; Wed, 06 Dec 95 23:13:18 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id aa20113; 6 Dec 95 23:13 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 6 Dec 1995 23:11:24 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA15073 for dbi-users-real; Wed, 6 Dec 1995 12:56:21 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA15069; Wed, 6 Dec 1995 12:56:18 -0800 Received: by gw.home.vix.com id AA00500; Wed, 6 Dec 95 12:56:07 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id g.af15601; 6 Dec 95 20:48 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id g.af17767; 6 Dec 95 20:47 GMT Received: from toad by oink with SMTP (PP) id <12111-0@oink>; Wed, 6 Dec 1995 20:38:18 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA06012; Wed, 6 Dec 1995 20:37:51 +0000 Date: Wed, 6 Dec 1995 20:37:51 +0000 From: Tim Bunce Message-Id: <9512062037.AA06012@toad.ig.co.uk> To: descarte@hermetica.com, dedlow@voro.lbl.gov Subject: Re: Can't see the woods.....$DBI::errstr woes.... Cc: dbi-users@fugue.com, dbi-dev@fugue.com X-Sun-Charset: US-ASCII Content-Length: 1748 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-Smtp-Posting-Host: ignite.demon.co.uk [Wed, 6 Dec 95 20:47:12 GMT] X-Smtp-Posting-Host: post.demon.co.uk [Wed, 6 Dec 95 20:48:47 GMT] X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 6 Dec 95 23:13:16 GMT] > From: Mark Dedlow > > Same DBI/DBD::Oracle versions (Solaris 2.4), I get: > > Database handle DBI::db=HASH(0x196e00) DESTROY ignored - never set up at > ./testDBI.pl line 7. > Abort (core dumped) > > if the user/pass is garbage. It works if user/pass OK... Here are a couple of patches (sorry they're straight diffs but I'm using rcsdiff and I've no time to fiddle). The fist fixes the warning and should be applied to Oracle.xs (and all the other DBD *.xs files - take note developers). diff -r1.36 Oracle.xs 131c131 < if (DBIc_WARN(imp_dbh) && !dirty) --- > if (DBIc_WARN(imp_dbh) && !dirty && dbis->debug >= 2) 299c299 < if (DBIc_WARN(imp_sth) && !dirty) --- > if (DBIc_WARN(imp_sth) && !dirty && dbis->debug >= 2) The second is a patch to DBI-0.65 which _may_ fix the core dump: diff -r1.49 DBI.xs 1146d1145 < SV *lhp = DBI_LAST_HANDLE; 1147a1147 > imp_xxh_t *imp_xxh = (ok) ? DBIh_COM(DBI_LAST_HANDLE) : NULL; 1149,1155c1149,1157 < D_imp_xxh(DBI_LAST_HANDLE); < < if (dbis->debug >= 2 || DBIc_DEBUGIV(imp_xxh) >= 2) { < SvROK_on(DBI_LAST_HANDLE); < fprintf(DBILOGFP," <> $DBI::%s (%c) FETCH from lasth=%s\n", < meth, type, (ok)?SvPV(lhp,na):"none"); < SvROK_off(DBI_LAST_HANDLE); --- > if (dbis->debug >= 2 || (ok && DBIc_DEBUGIV(imp_xxh) >= 2)) { > fprintf(DBILOGFP," <> $DBI::%s (%c) FETCH from lasth=", meth, type); > if (ok) { > SvROK_on(DBI_LAST_HANDLE); > fprintf(DBILOGFP,"%s\n", SvPV(DBI_LAST_HANDLE,na)); > SvROK_off(DBI_LAST_HANDLE); > } else { > fprintf(DBILOGFP,"none\n"); > } 1213c1216 < ST(0) = lhp; --- > ST(0) = DBI_LAST_HANDLE; Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13544-11@oink>; Wed, 6 Dec 1995 23:19:43 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818291360:11736:0; Wed, 06 Dec 95 23:09:20 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa11722; 6 Dec 95 23:09 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id msg.aa28305; 6 Dec 95 23:08 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA15080 for dbi-dev-real; Wed, 6 Dec 1995 12:58:48 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA15069; Wed, 6 Dec 1995 12:56:18 -0800 Received: by gw.home.vix.com id AA00500; Wed, 6 Dec 95 12:56:07 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id g.af15601; 6 Dec 95 20:48 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id g.af17767; 6 Dec 95 20:47 GMT Received: from toad by oink with SMTP (PP) id <12111-0@oink>; Wed, 6 Dec 1995 20:38:18 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA06012; Wed, 6 Dec 1995 20:37:51 +0000 Date: Wed, 6 Dec 1995 20:37:51 +0000 From: Tim Bunce Message-Id: <9512062037.AA06012@toad.ig.co.uk> To: descarte@hermetica.com, dedlow@voro.lbl.gov Subject: Re: Can't see the woods.....$DBI::errstr woes.... Cc: dbi-users@fugue.com, dbi-dev@fugue.com X-Sun-Charset: US-ASCII Content-Length: 1748 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-Smtp-Posting-Host: ignite.demon.co.uk [Wed, 6 Dec 95 20:47:12 GMT] X-Smtp-Posting-Host: post.demon.co.uk [Wed, 6 Dec 95 20:48:47 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Wed, 6 Dec 95 23:08:39 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 6 Dec 95 23:09:19 GMT] > From: Mark Dedlow > > Same DBI/DBD::Oracle versions (Solaris 2.4), I get: > > Database handle DBI::db=HASH(0x196e00) DESTROY ignored - never set up at > ./testDBI.pl line 7. > Abort (core dumped) > > if the user/pass is garbage. It works if user/pass OK... Here are a couple of patches (sorry they're straight diffs but I'm using rcsdiff and I've no time to fiddle). The fist fixes the warning and should be applied to Oracle.xs (and all the other DBD *.xs files - take note developers). diff -r1.36 Oracle.xs 131c131 < if (DBIc_WARN(imp_dbh) && !dirty) --- > if (DBIc_WARN(imp_dbh) && !dirty && dbis->debug >= 2) 299c299 < if (DBIc_WARN(imp_sth) && !dirty) --- > if (DBIc_WARN(imp_sth) && !dirty && dbis->debug >= 2) The second is a patch to DBI-0.65 which _may_ fix the core dump: diff -r1.49 DBI.xs 1146d1145 < SV *lhp = DBI_LAST_HANDLE; 1147a1147 > imp_xxh_t *imp_xxh = (ok) ? DBIh_COM(DBI_LAST_HANDLE) : NULL; 1149,1155c1149,1157 < D_imp_xxh(DBI_LAST_HANDLE); < < if (dbis->debug >= 2 || DBIc_DEBUGIV(imp_xxh) >= 2) { < SvROK_on(DBI_LAST_HANDLE); < fprintf(DBILOGFP," <> $DBI::%s (%c) FETCH from lasth=%s\n", < meth, type, (ok)?SvPV(lhp,na):"none"); < SvROK_off(DBI_LAST_HANDLE); --- > if (dbis->debug >= 2 || (ok && DBIc_DEBUGIV(imp_xxh) >= 2)) { > fprintf(DBILOGFP," <> $DBI::%s (%c) FETCH from lasth=", meth, type); > if (ok) { > SvROK_on(DBI_LAST_HANDLE); > fprintf(DBILOGFP,"%s\n", SvPV(DBI_LAST_HANDLE,na)); > SvROK_off(DBI_LAST_HANDLE); > } else { > fprintf(DBILOGFP,"none\n"); > } 1213c1216 < ST(0) = lhp; --- > ST(0) = DBI_LAST_HANDLE; Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15171-7@oink>; Thu, 7 Dec 1995 03:00:12 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818295851:21889:0; Thu, 07 Dec 95 00:24:11 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ab21848; 7 Dec 95 0:24 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 6 Dec 1995 23:17:29 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA15060 for dbi-users-real; Wed, 6 Dec 1995 12:54:46 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA15056 for ; Wed, 6 Dec 1995 12:54:39 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id g.ac15711; 6 Dec 95 20:48 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.ac17767; 6 Dec 95 20:46 GMT Received: from toad by oink with SMTP (PP) id <11674-0@oink>; Wed, 6 Dec 1995 18:36:31 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA05862; Wed, 6 Dec 1995 18:36:05 +0000 Date: Wed, 6 Dec 1995 18:36:05 +0000 From: Tim Bunce Message-Id: <9512061836.AA05862@toad.ig.co.uk> To: dbi-users@fugue.com, mthurn@stowe.irnet.rest.tasc.com Subject: Re: cannot build DBI 0.65 on SunOS Cc: thurn@cis.ohio-state.edu X-Sun-Charset: US-ASCII content-length: 1235 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Wed, 6 Dec 95 20:46:58 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Wed, 6 Dec 95 20:48:38 GMT] X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Thu, 7 Dec 95 0:24:11 GMT] > From: mthurn@stowe.irnet.rest.tasc.com (Martin Thurn) > > Thanks to those who pointed me to the latest version 0.65 of DBI. > However, I get exactly the same error when I try to make it: > ========== > > /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=1; runtests @ARGV;' t/*.t > t/min...............1..11 > ok 1 > ok 2 > ok 3 > ok 4 > FAILED on test 5 > Failed 1 test, 0.00% okay. > Files=1, Tests=11, 1 secs ( 0.35 cusr 0.18 csys = 0.53 cpu) > /usr/local/bin/perl -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/fs81/TIR/lib/perl5/DBI-0.65/blib -I/usr/local/lib/perl5/sun4-sunos -I/usr/local/lib/perl5 test.pl > test.pl > DBI test application $Revision: 1.12 $ > make: *** [test] Segmentation fault (core dumped) This problem only seems to affect very few people. The vast majority build the DBI with no problem. The fact you're using gcc on a sun just compounds the puzzle. I'd like to get to the bottom of this. If you're GCC is earlier than 2.5.8 then I'd recommend that you upgrade. I'll need a stack trace from a DBI built with -g before I can help you further. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <15169-12@oink>; Thu, 7 Dec 1995 03:00:42 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818299766:00710:0; Thu, 07 Dec 95 01:29:26 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa00611; 7 Dec 95 1:29 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA15672 for dbi-users-real; Wed, 6 Dec 1995 15:41:13 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA15667 for ; Wed, 6 Dec 1995 15:41:10 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aj14827; 6 Dec 95 23:20 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.ab24247; 6 Dec 95 23:18 GMT Received: from toad by oink with SMTP (PP) id <13362-0@oink>; Wed, 6 Dec 1995 22:00:29 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA06272; Wed, 6 Dec 1995 22:00:04 +0000 Date: Wed, 6 Dec 1995 22:00:04 +0000 From: Tim Bunce Message-Id: <9512062200.AA06272@toad.ig.co.uk> To: dbi-users@fugue.com, chinay@cig.mot.com Subject: Re: problems with DBI-Oracle-0.27 X-Sun-Charset: US-ASCII content-length: 790 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Wed, 6 Dec 95 23:18:19 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Wed, 6 Dec 95 23:20:22 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Thu, 7 Dec 95 1:29:18 GMT] > From: Albert Chin-A-Young > > When running the test supplied with DBI-Oracle-0.27 with perl-5.001n on > a Motorola SVR4 box, I get the following error: > > PERL_DL_NONLAZY=1 /usr/misc/bin/perl5 -I./blib/m88k-svr4 -I./blib > -I/usr/misc/perl5/lib/m88k-svr4 -I/usr/misc/perl5/lib -e > 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests > @ARGV;' t/*.t > t/base..............install_driver(Oracle) failed: Can't load > './blib/m88k-svr4/auto/DBD/Oracle/Oracle.so' for module DBD::Oracle: > dynamic linker: /usr/misc/bin/perl5: relocation error: symbol not > found: oerhms at /usr/misc/perl5/lib/DynaLoader.pm line 136. No idea. Did it work on a previous release of perl and/or DBD::Oracle ? Does it work for anyone else on SVR4 ? Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15171-43@oink>; Thu, 7 Dec 1995 03:02:51 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818298977:20038:0; Thu, 07 Dec 95 01:16:17 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa20014; 7 Dec 95 1:16 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA15625 for dbi-users-real; Wed, 6 Dec 1995 15:22:55 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA15621 for ; Wed, 6 Dec 1995 15:22:30 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id g.ac15499; 6 Dec 95 23:20 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ad24247; 6 Dec 95 23:18 GMT Received: from toad by oink with SMTP (PP) id <13402-0@oink>; Wed, 6 Dec 1995 22:19:02 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA06297; Wed, 6 Dec 1995 22:18:37 +0000 Date: Wed, 6 Dec 1995 22:18:37 +0000 From: Tim Bunce Message-Id: <9512062218.AA06297@toad.ig.co.uk> To: aburlison@cix.compulink.co.uk Subject: Re: Problems building Oracle DBD v0.27 on Unisys U6000 svr4 Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 1014 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Wed, 6 Dec 95 23:18:25 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Wed, 6 Dec 95 23:20:19 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Thu, 7 Dec 95 1:16:17 GMT] > From: aburlison@cix.compulink.co.uk (Alan Burlison) > > Hi all, > > Somehow I seem to have 'fallen off' the PerlDB mailing list - hopefully > I've now resubscribed (thought it had gone quiet...) > > I have just got around to updating to 5.002beta1, DBI 0.65 and Oracle DBD > 0.27, and had a few problems building. The build itself worked fine, but > 'make test' failed, with the following output: > > Can't load './blib/PENTIUM-svr4.0/auto/DBD/Oracle/Oracle.so' for module > DBD::Oracle: dynamic linker: /usr/bin/perl: relocation error: symbol not > found: vfork at /opt/lib/perl5/DynaLoader.pm line 136. > > This is (I think) caused by incorrect processing by MakeMaker. MakeMaker > has moved to using $self instead of %att for specifying config options > and hints. Unfortunately, the old mechanism seems to have been broken in > the process. The folowing patches put things right in DBD: Thanks. I won't apply that utill 5.002 is out and then I'll add a require 5.002; to Makefile.PL. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15171-49@oink>; Thu, 7 Dec 1995 03:03:18 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818300666:03977:0; Thu, 07 Dec 95 01:44:26 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa03903; 7 Dec 95 1:44 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA15776 for dbi-users-real; Wed, 6 Dec 1995 16:07:16 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA15772 for ; Wed, 6 Dec 1995 16:07:14 -0800 Received: by gw.home.vix.com id AA17141; Wed, 6 Dec 95 16:07:07 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ai21516; 6 Dec 95 23:27 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa26076; 6 Dec 95 23:26 GMT Received: from toad by oink with SMTP (PP) id <13939-0@oink>; Wed, 6 Dec 1995 23:25:44 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA06479; Wed, 6 Dec 1995 23:25:19 +0000 Date: Wed, 6 Dec 1995 23:25:19 +0000 From: Tim Bunce Message-Id: <9512062325.AA06479@toad.ig.co.uk> To: groenvel@cse.psu.edu Subject: Re: DBD-Oracle-0.27 blob_read Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII Content-Length: 245 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-Smtp-Posting-Host: ignite.demon.co.uk [Wed, 6 Dec 95 23:26:02 GMT] X-Smtp-Posting-Host: post.demon.co.uk [Wed, 6 Dec 95 23:27:49 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Thu, 7 Dec 95 1:44:21 GMT] > From: John D Groenveld > > &ora_readblob references readblob and not blob_read. The pod should include an > example of long and long raw DML ops. > Thanks, > John > groenvel@cse.psu.edu Ooops. Patches welcome. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <16975-3@oink>; Thu, 7 Dec 1995 06:08:17 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818309817:06639:0; Thu, 07 Dec 95 04:16:57 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa06619; 7 Dec 95 4:16 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA16236 for dbi-users-real; Wed, 6 Dec 1995 18:38:34 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from chan.crim.ca (chan.crim.ca [132.218.1.4]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id SAA16232 for ; Wed, 6 Dec 1995 18:38:31 -0800 Received: from zorro.crim.ca by chan.crim.ca (4.1/SMI-4.1) id AA06224; Wed, 6 Dec 95 21:38:31 EST Received: by zorro.crim.ca (5.0/SMI-SVR4) id AA02471; Wed, 6 Dec 1995 21:38:37 +0500 Date: Wed, 6 Dec 1995 21:38:37 +0500 From: maparent@zorro.crim.ca (Marc-Antoine Parent) Message-Id: <9512070238.AA02471@zorro.crim.ca> To: dbi-users@fugue.com Subject: return values from Oracle stored functions? X-Sun-Charset: US-ASCII Content-Length: 596 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 7 Dec 95 4:16:56 GMT] I hope this is not the umpteenth time this question comes on this group, but I did not find a faq. I saw a thread ('I know I'm cheating...') regarding oraperl4, but I hope things are different with DBD/DBI. Is it possible to get (somehow!) the return value from a stored function? Or at least get the IN/OUT parmeters to return me something? (I tried that and got a hang for 1 min, followed by a 'connection cut' message.) Otherwise, my installation tested fine. (DBI .64 and DBD-Oracle .26 on a Solaris 2.3. Will upgrading help?) I will do without if need be, but I would appreciate a hint! ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <18351-0@oink>; Thu, 7 Dec 1995 08:31:47 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818324397:02752:0; Thu, 07 Dec 95 08:19:57 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa02732; 7 Dec 95 8:19 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id WAA16634 for dbi-users-real; Wed, 6 Dec 1995 22:18:12 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id WAA16630 for ; Wed, 6 Dec 1995 22:18:07 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ac21787; 7 Dec 95 6:17 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa05135; 7 Dec 95 6:15 GMT Received: from toad by oink with SMTP (PP) id <17167-0@oink>; Thu, 7 Dec 1995 06:15:41 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA07023; Thu, 7 Dec 1995 06:15:15 +0000 Date: Thu, 7 Dec 1995 06:15:15 +0000 From: Tim Bunce Message-Id: <9512070615.AA07023@toad.ig.co.uk> To: dbi-users@fugue.com, maparent@zorro.crim.ca Subject: Re: return values from Oracle stored functions? X-Sun-Charset: US-ASCII content-length: 971 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Thu, 7 Dec 95 6:15:59 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Thu, 7 Dec 95 6:17:57 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Thu, 7 Dec 95 8:19:56 GMT] > From: maparent@zorro.crim.ca (Marc-Antoine Parent) > > I hope this is not the umpteenth time this question comes on this group, but > I did not find a faq. > I saw a thread ('I know I'm cheating...') regarding oraperl4, but I hope things > are different with DBD/DBI. > > Is it possible to get (somehow!) the return value from a stored function? Not currently. > Or at least get the IN/OUT parmeters to return me something? Not currently. > (I tried that and got a hang for 1 min, followed by a 'connection cut' message.) > > Otherwise, my installation tested fine. > (DBI .64 and DBD-Oracle .26 on a Solaris 2.3. Will upgrading help?) Nope. > I will do without if need be, but I would appreciate a hint! It is patchable but is raises some nasty issues about efficiency (you'd have to copy-back the value to perl all the time even when not required since you can't tell what's an output parameter) It will get supported - somehow. It's too important. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05536-9@oink>; Fri, 8 Dec 1995 00:09:37 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818366802:04528:0; Thu, 07 Dec 95 20:06:42 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa04523; 7 Dec 95 20:06 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id msg.aa18500; 7 Dec 95 20:05 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA18869 for dbi-users-real; Thu, 7 Dec 1995 09:20:45 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from chan.crim.ca (chan.crim.ca [132.218.1.4]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA18865 for ; Thu, 7 Dec 1995 09:20:43 -0800 Received: from zorro.crim.ca by chan.crim.ca (4.1/SMI-4.1) id AA14762; Thu, 7 Dec 95 12:20:42 EST Received: by zorro.crim.ca (5.0/SMI-SVR4) id AA05526; Thu, 7 Dec 1995 12:20:51 +0500 Date: Thu, 7 Dec 1995 12:20:51 +0500 From: Marc-Antoine Parent Message-Id: <9512071720.AA05526@zorro.crim.ca> To: Tim.Bunce@ig.co.uk Subject: Re: return values from Oracle stored functions? Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII Content-Length: 1555 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 7 Dec 95 20:05:50 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Thu, 7 Dec 95 20:06:41 GMT] First, thank you for a VERY prompt answer! > > Is it possible to get (somehow!) the return value from a stored function? > > Not currently. > > > Or at least get the IN/OUT parmeters to return me something? > > Not currently. > It is patchable but is raises some nasty issues about efficiency > (you'd have to copy-back the value to perl all the time even when not > required since you can't tell what's an output parameter) > > It will get supported - somehow. It's too important. > > Tim. > I note that last comment applies to IN/OUT parameters, but not to functions. Is that not much less of a patch? (with little or no performance overhead) I'm thinking of a construct of the type: my($csr) = $lda->prepare( << ) || die "drh->do: $DBI::err: $DBI::errstr\n"; BEGIN :1 := somefunction('sdhasfdjh'); END; # somefunctions RETURNs a VARCHAR2 $returnValue = "aaaaa"; @result = $csr->execute($returnValue); Or even more simply: @result = $lda->evalStringFunction("somefunction('sdhasfdjh')") || die "drh->do: $DBI::err: $DBI::errstr\n"; (Which, of course, alters DBI, but probably in a reasonable way, as getting a string value from a function SHOULD be "universal" enough...) Of course, another possibility would be to somewhat have two arrays passed to "execute", one for IN parameters and the other for OUT (and another for IN/OUT?) but that rings too database-specific to my ears. Do you see other ways to solve that problem? (is this group the right place to discuss this, or maybe the implementors'?) ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <20857-3@oink>; Mon, 11 Dec 1995 07:06:43 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818448328:26880:4; Fri, 08 Dec 95 18:45:28 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa26248; 8 Dec 95 18:44 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA22258 for dbi-users-real; Fri, 8 Dec 1995 07:35:38 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA22254 for ; Fri, 8 Dec 1995 07:35:35 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id g.ag03836; 8 Dec 95 15:26 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.aa14138; 8 Dec 95 15:25 GMT Received: from toad by oink with SMTP (PP) id <10888-0@oink>; Fri, 8 Dec 1995 15:25:00 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA07754; Fri, 8 Dec 1995 15:24:34 +0000 Date: Fri, 8 Dec 1995 15:24:34 +0000 From: Tim Bunce Message-Id: <9512081524.AA07754@toad.ig.co.uk> To: mthurn@stowe.irnet.rest.tasc.com Subject: Re: cannot build DBI 0.65 on SunOS Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 320 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Fri, 8 Dec 95 15:25:56 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Fri, 8 Dec 95 15:26:43 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Fri, 8 Dec 95 18:44:56 GMT] > From: mthurn@irnet.rest.tasc.com (Martin Thurn) > > > > FAILED on test 5 > > > DBI test application $Revision: 1.12 $ > > > make: *** [test] Segmentation fault (core dumped) > > We upgraded to the newest perl 5.002(?) and now it works fine! Wonderful news. Thanks. I hope that's the end of the core dumps. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <20858-5@oink>; Mon, 11 Dec 1995 07:07:07 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818447059:02419:0; Fri, 08 Dec 95 18:24:19 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa02414; 8 Dec 95 18:24 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa06027; 8 Dec 95 18:24 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA22217 for dbi-users-real; Fri, 8 Dec 1995 07:22:48 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail.crl.com (mail.crl.com [165.113.1.22]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA22213 for ; Fri, 8 Dec 1995 07:22:47 -0800 Received: from [205.149.164.116] (bta.vip.best.com) by mail.crl.com with SMTP id AA19294 (5.65c/IDA-1.5 for ); Fri, 8 Dec 1995 07:20:27 -0800 Date: Fri, 8 Dec 1995 07:20:27 -0800 X-Sender: mrlizard@mail.crl.com Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: dbi-users@fugue.com From: Brian Valente Subject: Installing DBI/Driver locally X-SMTP-Posting-Host: toccata.fugue.com [Fri, 8 Dec 95 18:24:06 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Fri, 8 Dec 95 18:24:18 GMT] Hello all, How best to install DBD in a local directory? I can't install it in the usual perl library directories, and I'm wondering if it's going to break? Thanks, Regards, Brian Valente ====================================================================== Brian Valente || brian_valente@rmdudley.com R.M. Dudley Corporation || mrlizard@crl.com 1799 Old Bayshore Suite 200 || applelink:RMDUDLEY Burlingame, CA 94010 || http://www.rmdudley.com/rmdudley/ (415) 697-1650 voice || (415) 697-1816 fax || ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <20858-16@oink>; Mon, 11 Dec 1995 07:10:47 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818449487:22316:0; Fri, 08 Dec 95 19:04:47 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa22309; 8 Dec 95 19:04 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id msg.aa08585; 8 Dec 95 18:50 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA22208 for dbi-users-real; Fri, 8 Dec 1995 07:20:00 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA22204 for ; Fri, 8 Dec 1995 07:19:55 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ac03040; 8 Dec 95 15:17 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ah11866; 8 Dec 95 15:16 GMT Received: from toad by oink with SMTP (PP) id <09730-0@oink>; Fri, 8 Dec 1995 12:17:03 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA06947; Fri, 8 Dec 1995 12:16:36 +0000 Date: Fri, 8 Dec 1995 12:16:36 +0000 From: Tim Bunce Message-Id: <9512081216.AA06947@toad.ig.co.uk> To: maparent@zorro.crim.ca Subject: Re: return values from Oracle stored functions? Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 851 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Fri, 8 Dec 95 15:16:16 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Fri, 8 Dec 95 15:17:25 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Fri, 8 Dec 95 18:50:03 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Fri, 8 Dec 95 19:04:46 GMT] > From: maparent@zorro.crim.ca (Marc-Antoine Parent) > > First, thank you for a VERY prompt answer! > > > > Is it possible to get (somehow!) the return value from a stored function? > > > > Not currently. > > > > > Or at least get the IN/OUT parmeters to return me something? > > > > Not currently. > > It is patchable but is raises some nasty issues about efficiency > > (you'd have to copy-back the value to perl all the time even when not > > required since you can't tell what's an output parameter) > > > > It will get supported - somehow. It's too important. > > I note that last comment applies to IN/OUT parameters, but not to functions. I meant both. > (is this group the right place to discuss this, or maybe the implementors'?) This is the right group. The dbi-dev group is for developers to discuss implementation issues. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <20857-20@oink>; Mon, 11 Dec 1995 07:10:57 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818457774:07800:0; Fri, 08 Dec 95 21:22:54 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa07718; 8 Dec 95 21:22 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA23069 for dbi-dev-real; Fri, 8 Dec 1995 11:47:04 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from paul.spu.edu (Paul.spu.edu [192.190.33.3]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA23064 for ; Fri, 8 Dec 1995 11:47:01 -0800 Received: by paul.spu.edu; (5.65/1.1.8.2/17Sep95-0400PM) id AA08607; Fri, 8 Dec 1995 11:45:31 -0800 Date: Fri, 8 Dec 1995 11:45:29 -0800 (PST) From: Phil Rand To: DBI/DBD developers , VMS perl list Subject: dbi/dbd-oracle port to DEC OpenVMS ? In-Reply-To: <9512081524.AA07754@toad.ig.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: toccata.fugue.com [Fri, 8 Dec 95 21:22:48 GMT] Hi, Is anybody else already working on porting dbi (perl database interface) to VMS? How about DBD-Oracle? I've just barely gotten started, using perl5_001_950925B. Actually, I've only attempted an installation of dbi, so far. If nobody else on the dbi-dev list is interested in VMS, I'll do my best to target future cries for help to one list or the other, whichever seems most appropriate, as I continue working on this. -- -- Phil Rand aka -- Computer & Information Systems (206) 281-2428 -- Seattle Pacific University, 3307 3rd Ave W, Seattle, WA 98119 -- http://paul.spu.edu/~prand/ -- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <20858-32@oink>; Mon, 11 Dec 1995 07:12:42 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818456417:29442:0; Fri, 08 Dec 95 21:00:17 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa28458; 8 Dec 95 21:00 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA22862 for dbi-users-real; Fri, 8 Dec 1995 10:26:23 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sultan.hsii.ccare.com ([192.101.181.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA22858 for ; Fri, 8 Dec 1995 10:26:18 -0800 Received: from msgw.hsii.ccare.com by SULTAN.HSII.CCARE.COM (PMDF V5.0-5 #4597) id <01HYKC6MY7UO0000XE@SULTAN.HSII.CCARE.COM> for dbi-users@fugue.com; Fri, 08 Dec 1995 12:25:40 -0600 (CST) Received: by msgw.hsii.ccare.com with Microsoft Mail id <30C8832A@msgw.hsii.ccare.com>; Fri, 08 Dec 1995 12:25:46 -0600 (CST) Date: Fri, 08 Dec 1995 12:22:00 -0600 (CST) From: "Li, Alexander" Subject: HELP building DBI-0.65 To: DBI User List Message-id: <30C8832A@msgw.hsii.ccare.com> X-Mailer: Microsoft Mail V3.0 Content-transfer-encoding: 7BIT Encoding: 284 TEXT X-SMTP-Posting-Host: toccata.fugue.com [Fri, 8 Dec 95 21:00:14 GMT] Dear Perl/DBI experts, Attached at the end is a problem log when our programmer(I'm the sys. admin) try to build DBI-0.65. This happened only *after* I recompiled perl 5 with dynamic loading. i.e. The first time was fine. Our goal is to build a Perl 5 to Oracle 7 interface. Perl 5 tested fine. The platform is a Sun Voyager running Solaris 2.4. Compiler is Sun's SPARCompiler C v3.0. I'm relative new to this stuff and probably made a stupid mistake somewhere. Thanks for you kind help. ---------------------------------- Alex Li Health Systems Integration, Inc. ali@hsii.ccare.com work: 612-858-7570 ---------------------------------- Script started on Fri Dec 08 09:46:13 1995 gh753> perl Makefile.PL Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker (v4.16) Checking if your kit is complete... Looks good NAME => 'DBI' VERSION => '0.65' Using PERL=/usr/local/bin/perl Writing Makefile for DBI gh753> gh753> make mkdir ./blib mkdir ./blib/auto mkdir ./blib/auto/DBI /usr/local/bin/perl -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/lib/perl5 /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c cc -c -DDEBUGGING -O -I/usr/local/lib/perl5/sun4-solaris/CORE DBI.c Running Mkbootstrap for DBI () chmod 644 DBI.bs LD_RUN_PATH="" ld -o blib/auto/DBI/DBI.none DBI.o Undefined first referenced symbol in file sv_2iv DBI.o sv_magic DBI.o stack_sp DBI.o markstack_max DBI.o sv_free DBI.o gv_fetchpv DBI.o sv_undef DBI.o av_fetch DBI.o sv_2pv DBI.o newSVpv DBI.o sv_setiv DBI.o mg_find DBI.o __ctype DBI.o gv_fetchmethod DBI.o abort DBI.o hv_fetch DBI.o Xpv DBI.o sv_catpv DBI.o strerror DBI.o newRV DBI.o newSViv DBI.o newSV DBI.o dirty DBI.o malloc DBI.o sprintf DBI.o perl_destruct_level DBI.o av_store DBI.o newXS DBI.o newSVsv DBI.o runlevel DBI.o sv_no DBI.o sv_bless DBI.o sv_2bool DBI.o gv_stashpv DBI.o Sv DBI.o hv_store DBI.o fclose DBI.o perl_call_sv DBI.o sv_mortalcopy DBI.o savepv DBI.o sv_setsv DBI.o sv_catpvn DBI.o strncmp DBI.o sv_backoff DBI.o __iob DBI.o dowantarray DBI.o strcat DBI.o warn DBI.o sv_newmortal DBI.o errno DBI.o stack_base DBI.o stack_max DBI.o av_len DBI.o sv_setpv DBI.o sv_2mortal DBI.o strcmp DBI.o fopen DBI.o memset DBI.o croak DBI.o strcpy DBI.o markstack_grow DBI.o stack_grow DBI.o na DBI.o sv_dump DBI.o markstack_ptr DBI.o op DBI.o perl_get_sv DBI.o strlen DBI.o perl_call_method DBI.o fprintf DBI.o newAV DBI.o sv_yes DBI.o ld: fatal: Symbol referencing errors. No output written to blib/auto/DBI/DBI.none *** Error code 1 make: Fatal error: Command failed for target `blib/auto/DBI/DBI.none' gh753> gh753> gh753> perl -e ` 'use Config; print Conig  fig::myconif  fig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=solaris, osver=2.4, archname=sun4-solaris uname='sunos gh753 5.4 generic_101945-10 sun4m sparc ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-DDEBUGGING' ccflags ='-DDEBUGGING' ldflags ='' stdchar='unsigned char', d_stdstdio=, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/lib /usr/lib /usr/ccs/lib libs=-lsocket -lnsl -ldl -lm -lc -lcrypt libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_none.xs, dlext=none, d_dlsymun= cccdlflags='', ccdlflags='', lddlflags='' gh753> script done on Fri Dec 08 09:51:39 1995 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <26211-4@oink>; Mon, 11 Dec 1995 12:12:20 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818673771:24347:1; Mon, 11 Dec 95 09:22:51 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa24310; 11 Dec 95 9:22 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id AAA29950 for dbi-users-real; Mon, 11 Dec 1995 00:22:47 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from monge.brunel.ac.uk (pp@monge.brunel.ac.uk [134.83.72.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id AAA29946 for ; Mon, 11 Dec 1995 00:22:43 -0800 From: cs92bbm@brunel.ac.uk Received: from ccsp-14.brunel.ac.uk by monge.brunel.ac.uk with SMTP (PP) id <03166-0@monge.brunel.ac.uk>; Mon, 11 Dec 1995 08:22:37 +0000 Message-Id: <690.9512110822@ccsp-14.brunel.ac.uk> Subject: Oracle6 and Perl5 To: dbi-users@fugue.com Date: Mon, 11 Dec 1995 08:22:34 +0000 (GMT) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Mon, 11 Dec 95 9:22:49 GMT] Just wondering whether anyone has had any luck with this combination? Tim suggested that a DBI hacker could change the calls appropriately... does such a hacker exist then? Any help would be appreciated - this is an important one! Thanks, Ben. [ Ben Middleton cs92bbm@brunel.ac.uk ] ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <26211-11@oink>; Mon, 11 Dec 1995 12:12:47 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818678191:03114:0; Mon, 11 Dec 95 10:36:31 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa03108; 11 Dec 95 10:36 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id msg.aa05967; 11 Dec 95 10:35 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id BAA00151 for dbi-users-real; Mon, 11 Dec 1995 01:27:37 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay.inferentia.it (dns.inferentia.it [151.99.220.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id BAA00147 for ; Mon, 11 Dec 1995 01:27:33 -0800 Received: from yquem.inferentia.it (yquem.inferentia.it [194.184.62.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id KAA11720; Mon, 11 Dec 1995 10:26:35 +0100 Posted-Date: Mon, 11 Dec 1995 10:26:35 +0100 Received: by yquem.inferentia.it with Microsoft Mail id <01BAC7B4.CBFE6EC0@yquem.inferentia.it>; Mon, 11 Dec 1995 10:38:23 +-100 Message-ID: <01BAC7B4.CBFE6EC0@yquem.inferentia.it> From: Davide Migliavacca To: 'Marc-Antoine Parent' Cc: "'dbi-users@fugue.com'" Subject: RE: return values from Oracle stored functions? Date: Mon, 11 Dec 1995 10:38:22 +-100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-SMTP-Posting-Host: toccata.fugue.com [Mon, 11 Dec 95 10:35:28 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Mon, 11 Dec 95 10:36:30 GMT] You can use functions like this: select function(par,...) from dual (a normal select statemente, palatable to DBD) IFF you declare your = function like function line (i integer) return varchar2; -- for instance! PRAGMA RESTRICT_REFERENCES(line, WNDS, WNPS, RNDS); -- may be less = strict Check your Oracle server documentation addendum. wowperl (ftp://ftp.inferentia.it/pub/perl/dbperl/wowperl.tar) does = exactly that. It should be noted that, IF you are desperate enough to write wrappers = for your store procedures to use dbms_output, you can use wowperl to = retrieve output from procedures. Not very elegant, though. Just until SP = find their way in DBD. Davide ---------- From: Marc-Antoine Parent[SMTP:maparent@zorro.crim.ca] Sent: gioved=EC 7 dicembre 1995 8.20 To: Tim.Bunce@ig.co.uk Cc: dbi-users@fugue.com Subject: Re: return values from Oracle stored functions? First, thank you for a VERY prompt answer! > > Is it possible to get (somehow!) the return value from a stored = function? >=20 > Not currently. >=20 > > Or at least get the IN/OUT parmeters to return me something? >=20 > Not currently. > It is patchable but is raises some nasty issues about efficiency > (you'd have to copy-back the value to perl all the time even when not > required since you can't tell what's an output parameter) >=20 > It will get supported - somehow. It's too important. >=20 > Tim. >=20 I note that last comment applies to IN/OUT parameters, but not to = functions. Is that not much less of a patch? (with little or no performance = overhead) I'm thinking of a construct of the type: my($csr) =3D $lda->prepare( << ) || die "drh->do: $DBI::err: = $DBI::errstr\n"; BEGIN :1 :=3D somefunction('sdhasfdjh'); END; # somefunctions RETURNs a VARCHAR2 $returnValue =3D "aaaaa"; @result =3D $csr->execute($returnValue); Or even more simply: @result =3D $lda->evalStringFunction("somefunction('sdhasfdjh')") || die "drh->do: $DBI::err: $DBI::errstr\n"; (Which, of course, alters DBI, but probably in a reasonable way, as = getting a string value from a function SHOULD be "universal" enough...) Of course, another possibility would be to somewhat have two arrays = passed to "execute", one for IN parameters and the other for OUT (and another for = IN/OUT?) but that rings too database-specific to my ears. Do you see other ways to solve that problem?=20 (is this group the right place to discuss this, or maybe the = implementors'?) -------------------------------------------------------- Davide Migliavacca (aka shari) Inferentia S.r.l. - Milano ITALY +39 2 59928.1 (FAX .221) * Standard disclaimer here * -------------------------------------------------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03440-5@oink>; Tue, 12 Dec 1995 03:39:42 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818735179:26710:1; Tue, 12 Dec 95 02:26:19 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa26608; 12 Dec 95 2:26 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA03365 for dbi-users-real; Mon, 11 Dec 1995 17:49:38 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA03348; Mon, 11 Dec 1995 17:44:44 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aj22882; 12 Dec 95 1:38 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ad10782; 12 Dec 95 1:38 GMT Received: from toad by oink with SMTP (PP) id <01955-0@oink>; Tue, 12 Dec 1995 01:30:01 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA27073; Tue, 12 Dec 1995 01:29:27 +0000 Date: Tue, 12 Dec 1995 01:29:27 +0000 From: Tim Bunce Message-Id: <9512120129.AA27073@toad.ig.co.uk> To: dbi-announce@fugue.com, dbi-users@fugue.com Subject: New directions for DBI - Important Cc: ak@ananke.s.bawue.de, kejin@lithium.empress.com X-Sun-Charset: US-ASCII content-length: 4705 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 12 Dec 95 1:38:16 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 12 Dec 95 1:38:56 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 2:26:15 GMT] I've been putting off writing this message because I've little time to deal with the fallout it's bound to create... but I don't want to leave it any longer. It's too important. As you may remember I went away for a couple of weeks vacation recently saying that I'd be thinking about ODBC and how it fitted in with the DBI. To cut a long story short I've decided that we should throw away a significant chunk of the work that we've done and reengineer an implementation based very strongly around the SQL CLI / ODBC API. My reasoning goes something like this: - The SQL CLI has recently been adopted as an international standard. - ODBC is based on the SQL CLI standard and Microsoft has commited to make ODBC v3 a proper superset of the standard. - Database vendors (include Infomix, IBM, Watcom, etc.) are adopting ODBC as the native CLI for their database engine(s). - ODBC has *huge* market share on PC's and a very rapidly growing presence on UNIX (Solaris 2.5 comes with an ODBC driver manager, other vendors are likely to do the same). A large number of commercial drivers already exist. Non-commercial drivers and being developed. - The DBI currently lacks much of the breadth required for a full featured API. Defining the API would duplicate much of the work standardised by the SQL CLI / ODBC API. - The DBI needs to support ODBC well for it to be widely accepted as the standard perl database interface. Specifically it would need to expose the majority of the API and associated concepts. A DBD::ODBC offering only the current DBI API is not sufficient. - A free driver manager exists (Ke Jin's iODBC) and free drivers can be implemented with approximately the same skill level as DBD drivers. And finally some carrots, in case you still need them: - I believe that we will be able to implement ODBC drivers for loading into perl which are implemented as Perl scripts. - I also believe that it should be possible to implement ODBC drivers in Perl which can be loaded into non-perl applications! - The DBI will no longer be incomplete and undocumented! Okay, so what's the plan? What will the new 'thing' look like? Will my current DBI scripts still run? Questions, questions. I'm sure you'll have many more than this. The initial step is to carefully define a fairly formal 'perl language binding' for the SQL CLI / ODBC API. Basically this involves some easy decisions like "string input parameters do not need a separate length parameter to we'll delete it for all functions" and some hard and/or subtle ones, like how best to model handles or deal with descriptors. (Handles will almost certainly be objects, unlike existing ODBC modules.) There will be some tough decisions to make but I'm very keen to avoid limiting the access to the underlying API functionality. Basically anything that you can do in C (or COBOL!) you should be able to do in Perl. Once that's agreed I'll implement it. It should be very straight forward. (Much less work than I'm throwing away :-) I'll implement it over at least one commercial driver manager and also over the free iODBC manager. I may borrow some code from Andreas Kaiser's DB2CLI module if he's happy for me to do that and I decide I want to do some things the same way. Over this complete but low level API we'll layer some high level methods that will give us our familiar DBI interface back (as far as possible). At this point we'll have complete API useable by anyone who can acquire an ODBC diver. Note that some (many?) database vendors are giving away drivers. The next stage is to develop some drivers for databases and/or platforms where free drivers are not available. I envisage a scheme whereby I define and implement as much support structure as possible to make it easier for others to follow in my footsteps. Similar to DBD::Oracle. The final component is the infrastructure for implementing drivers in perl. It would be of *great* help to have the *active* contribution of people who have used ODBC in real world applications. I hope Andreas Kaiser and Ke Jin will join the mailing list if the're not already on it. Meanwhile I suggest that everyone else start learning the SQL CLI / ODBC API! Many books are available from many bookstores. I'll stop here and leave you to absorb the meaning of all this. Feel free to comment on the above but please avoid getting into technical issues just yet. The time for that will come. And please note that I'm going to be very short of time till mid-Jan. I have real-work-deliverables to deliver :-( Tim. (Sympathy for throwing away much of the DBI and DBD::Oracle would be appreciated. Merry Christmas :-) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03440-7@oink>; Tue, 12 Dec 1995 03:39:59 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818735523:29278:1; Tue, 12 Dec 95 02:32:03 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa29215; 12 Dec 95 2:32 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA03352 for dbi-announce-real; Mon, 11 Dec 1995 17:44:58 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA03348; Mon, 11 Dec 1995 17:44:44 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aj22882; 12 Dec 95 1:38 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ad10782; 12 Dec 95 1:38 GMT Received: from toad by oink with SMTP (PP) id <01955-0@oink>; Tue, 12 Dec 1995 01:30:01 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA27073; Tue, 12 Dec 1995 01:29:27 +0000 Date: Tue, 12 Dec 1995 01:29:27 +0000 From: Tim Bunce Message-Id: <9512120129.AA27073@toad.ig.co.uk> To: dbi-announce@fugue.com, dbi-users@fugue.com Subject: New directions for DBI - Important Cc: ak@ananke.s.bawue.de, kejin@lithium.empress.com X-Sun-Charset: US-ASCII content-length: 4705 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 12 Dec 95 1:38:16 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 12 Dec 95 1:38:56 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 2:32:02 GMT] I've been putting off writing this message because I've little time to deal with the fallout it's bound to create... but I don't want to leave it any longer. It's too important. As you may remember I went away for a couple of weeks vacation recently saying that I'd be thinking about ODBC and how it fitted in with the DBI. To cut a long story short I've decided that we should throw away a significant chunk of the work that we've done and reengineer an implementation based very strongly around the SQL CLI / ODBC API. My reasoning goes something like this: - The SQL CLI has recently been adopted as an international standard. - ODBC is based on the SQL CLI standard and Microsoft has commited to make ODBC v3 a proper superset of the standard. - Database vendors (include Infomix, IBM, Watcom, etc.) are adopting ODBC as the native CLI for their database engine(s). - ODBC has *huge* market share on PC's and a very rapidly growing presence on UNIX (Solaris 2.5 comes with an ODBC driver manager, other vendors are likely to do the same). A large number of commercial drivers already exist. Non-commercial drivers and being developed. - The DBI currently lacks much of the breadth required for a full featured API. Defining the API would duplicate much of the work standardised by the SQL CLI / ODBC API. - The DBI needs to support ODBC well for it to be widely accepted as the standard perl database interface. Specifically it would need to expose the majority of the API and associated concepts. A DBD::ODBC offering only the current DBI API is not sufficient. - A free driver manager exists (Ke Jin's iODBC) and free drivers can be implemented with approximately the same skill level as DBD drivers. And finally some carrots, in case you still need them: - I believe that we will be able to implement ODBC drivers for loading into perl which are implemented as Perl scripts. - I also believe that it should be possible to implement ODBC drivers in Perl which can be loaded into non-perl applications! - The DBI will no longer be incomplete and undocumented! Okay, so what's the plan? What will the new 'thing' look like? Will my current DBI scripts still run? Questions, questions. I'm sure you'll have many more than this. The initial step is to carefully define a fairly formal 'perl language binding' for the SQL CLI / ODBC API. Basically this involves some easy decisions like "string input parameters do not need a separate length parameter to we'll delete it for all functions" and some hard and/or subtle ones, like how best to model handles or deal with descriptors. (Handles will almost certainly be objects, unlike existing ODBC modules.) There will be some tough decisions to make but I'm very keen to avoid limiting the access to the underlying API functionality. Basically anything that you can do in C (or COBOL!) you should be able to do in Perl. Once that's agreed I'll implement it. It should be very straight forward. (Much less work than I'm throwing away :-) I'll implement it over at least one commercial driver manager and also over the free iODBC manager. I may borrow some code from Andreas Kaiser's DB2CLI module if he's happy for me to do that and I decide I want to do some things the same way. Over this complete but low level API we'll layer some high level methods that will give us our familiar DBI interface back (as far as possible). At this point we'll have complete API useable by anyone who can acquire an ODBC diver. Note that some (many?) database vendors are giving away drivers. The next stage is to develop some drivers for databases and/or platforms where free drivers are not available. I envisage a scheme whereby I define and implement as much support structure as possible to make it easier for others to follow in my footsteps. Similar to DBD::Oracle. The final component is the infrastructure for implementing drivers in perl. It would be of *great* help to have the *active* contribution of people who have used ODBC in real world applications. I hope Andreas Kaiser and Ke Jin will join the mailing list if the're not already on it. Meanwhile I suggest that everyone else start learning the SQL CLI / ODBC API! Many books are available from many bookstores. I'll stop here and leave you to absorb the meaning of all this. Feel free to comment on the above but please avoid getting into technical issues just yet. The time for that will come. And please note that I'm going to be very short of time till mid-Jan. I have real-work-deliverables to deliver :-( Tim. (Sympathy for throwing away much of the DBI and DBD::Oracle would be appreciated. Merry Christmas :-) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03440-10@oink>; Tue, 12 Dec 1995 03:40:18 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818736512:06675:1; Tue, 12 Dec 95 02:48:32 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa06566; 12 Dec 95 2:48 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA03446 for dbi-users-real; Mon, 11 Dec 1995 18:15:39 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id SAA03442 for ; Mon, 11 Dec 1995 18:15:37 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.af25792; 12 Dec 95 2:14 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.aa14843; 12 Dec 95 2:13 GMT Received: from toad by oink with SMTP (PP) id <03073-0@oink>; Tue, 12 Dec 1995 02:13:25 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA27306; Tue, 12 Dec 1995 02:12:59 +0000 Date: Tue, 12 Dec 1995 02:12:59 +0000 From: Tim Bunce Message-Id: <9512120212.AA27306@toad.ig.co.uk> To: dbi-users@fugue.com Subject: Perl5 & informix X-Sun-Charset: US-ASCII content-length: 389 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 12 Dec 95 2:13:41 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 12 Dec 95 2:14:26 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 2:48:23 GMT] ----- Begin Included Message ----- Date: Mon, 11 Dec 1995 08:51:25 -0500 From: Bahram Samani To: Tim.Bunce@ig.co.uk Subject: Perl5 & informix X-Url: http://www.hermetica.com/technologia/DBI/ We are not able to do a complete make for perl5, informix, and AIX, or HPUX at Disney. Is there a new version of the beta? Thanks! Bahram ----- End Included Message ----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04179-5@oink>; Tue, 12 Dec 1995 04:28:01 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818741408:08998:1; Tue, 12 Dec 95 04:10:08 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa08976; 12 Dec 95 4:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA03828 for dbi-announce-real; Mon, 11 Dec 1995 19:39:55 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from spica.npac.syr.edu (root@spica-gw.npac.syr.edu [128.230.14.6]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id TAA03824; Mon, 11 Dec 1995 19:39:52 -0800 Received: from nova.npac.syr.edu by spica.npac.syr.edu (4.1/I-1.98K) id AA06923; Mon, 11 Dec 95 22:39:44 EST Received: (from gcheng@localhost) by nova.npac.syr.edu (8.7.1/8.7.1) id WAA29249; Mon, 11 Dec 1995 22:39:30 -0500 (EST) Date: Mon, 11 Dec 1995 22:39:30 -0500 (EST) From: Gang Cheng Message-Id: <199512120339.WAA29249@nova.npac.syr.edu> To: dbi-announce@fugue.com, dbi-users@fugue.com Subject: DBI mailing list archive X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 4:10:08 GMT] check URL: http://asknpac.npac.syr.edu. This archive is online (updated in every 30 minutes). Total about 120 USENET newsgroups and mailing lists are archived there (total ~300,000 articles up to now). --- Gang ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04179-7@oink>; Tue, 12 Dec 1995 04:28:07 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818741763:11023:1; Tue, 12 Dec 95 04:16:03 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa10962; 12 Dec 95 4:15 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA03842 for dbi-users-real; Mon, 11 Dec 1995 19:41:06 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from spica.npac.syr.edu (root@spica-gw.npac.syr.edu [128.230.14.6]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id TAA03824; Mon, 11 Dec 1995 19:39:52 -0800 Received: from nova.npac.syr.edu by spica.npac.syr.edu (4.1/I-1.98K) id AA06923; Mon, 11 Dec 95 22:39:44 EST Received: (from gcheng@localhost) by nova.npac.syr.edu (8.7.1/8.7.1) id WAA29249; Mon, 11 Dec 1995 22:39:30 -0500 (EST) Date: Mon, 11 Dec 1995 22:39:30 -0500 (EST) From: Gang Cheng Message-Id: <199512120339.WAA29249@nova.npac.syr.edu> To: dbi-announce@fugue.com, dbi-users@fugue.com Subject: DBI mailing list archive X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 4:15:54 GMT] check URL: http://asknpac.npac.syr.edu. This archive is online (updated in every 30 minutes). Total about 120 USENET newsgroups and mailing lists are archived there (total ~300,000 articles up to now). --- Gang ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04179-9@oink>; Tue, 12 Dec 1995 04:28:13 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818741802:11294:1; Tue, 12 Dec 95 04:16:42 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa11255; 12 Dec 95 4:16 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA03837 for dbi-users-real; Mon, 11 Dec 1995 19:40:32 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id TAA03833 for ; Mon, 11 Dec 1995 19:40:23 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab02843; 12 Dec 95 3:39 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ad21596; 12 Dec 95 3:39 GMT Received: from toad by oink with SMTP (PP) id <03190-0@oink>; Tue, 12 Dec 1995 02:37:49 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA27342; Tue, 12 Dec 1995 02:37:23 +0000 Date: Tue, 12 Dec 1995 02:37:23 +0000 From: Tim Bunce Message-Id: <9512120237.AA27342@toad.ig.co.uk> To: dbi-users@fugue.com, AlexLi@msmail.hsii.ccare.com Subject: Re: HELP building DBI-0.65 X-Sun-Charset: US-ASCII content-length: 1830 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 12 Dec 95 3:39:22 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 12 Dec 95 3:39:55 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 4:16:42 GMT] > From: "Li, Alexander" > > Dear Perl/DBI experts, > > Attached at the end is a problem log when our programmer(I'm the sys. admin) > try to build DBI-0.65. This happened only *after* I recompiled perl 5 > with dynamic loading. i.e. The first time was fine. > > Our goal is to build a Perl 5 to Oracle 7 interface. Perl 5 tested fine. > > The platform is a Sun Voyager running Solaris 2.4. Compiler is Sun's > SPARCompiler C v3.0. > > I'm relative new to this stuff and probably made a stupid mistake somewhere. > LD_RUN_PATH="" ld -o blib/auto/DBI/DBI.none DBI.o The .none is a sign of a broken perl config. > Undefined first referenced > symbol in file > sv_2iv DBI.o > [...] > sv_yes DBI.o > > ld: fatal: Symbol referencing errors. No output written to > blib/auto/DBI/DBI.none > Summary of my perl5 (patchlevel 1) configuration: > > Platform: > osname=solaris, osver=2.4, archname=sun4-solaris > Compiler: > cc='cc', optimize='-O' > cppflags='-DDEBUGGING' > ccflags ='-DDEBUGGING' > ldflags ='' > stdchar='unsigned char', d_stdstdio=, usevfork=false > voidflags=15, castflags=0, d_casti32=define, d_castneg=define > intsize=4, alignbytes=8, usemymalloc=y, randbits=15 > Libraries: > so=so > libpth=/lib /usr/lib /usr/ccs/lib > libs=-lsocket -lnsl -ldl -lm -lc -lcrypt > libc=/usr/lib/libc.so > Dynamic Linking: > dlsrc=dl_none.xs, dlext=none, d_dlsymun= > cccdlflags='', ccdlflags='', lddlflags='' You don't have dynamic loading enabled according to this config. I recommend rebuilding perl using make distclean Configure -des make make test installperl (never just copy a perl binary) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05066-5@oink>; Tue, 12 Dec 1995 07:06:18 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818744559:27969:1; Tue, 12 Dec 95 05:02:39 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa27940; 12 Dec 95 5:02 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA04145 for dbi-users-real; Mon, 11 Dec 1995 20:40:02 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mailgw.claircom.com (mailgw.claircom.com [199.5.241.51]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA04138; Mon, 11 Dec 1995 20:40:00 -0800 Received: from nimo.claircom.com by mailgw.claircom.com with smtp (Smail3.1.26.7 #2) id m0tPMVU-0004thC; Mon, 11 Dec 95 20:40 PST Received: from airone.claircom.com by nimo.claircom.com with smtp (Smail3.1.26.7 #2) id m0tPMVT-0005VkC; Mon, 11 Dec 95 20:39 PST Received: from soc5.airone.claircom.com by airone.claircom.com (5.x/SMI-SVR4) id AA28045; Mon, 11 Dec 1995 20:36:22 -0800 Received: by soc5.airone.claircom.com (4.1/SMI-4.1) id AA03096; Mon, 11 Dec 95 20:20:19 PST Date: Mon, 11 Dec 95 20:20:19 PST From: billw@airone.claircom.com (William Warner) Message-Id: <9512120420.AA03096@soc5.airone.claircom.com> To: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk Subject: Re: New directions for DBI - Important Cc: ak@ananke.s.bawue.de, kejin@lithium.empress.com X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 5:02:38 GMT] I'm lacking a little background information. What is SQL CLI/ODBC ? William Warner billw@airone.claircom.com portable: 206-915-1157 home: 206-720-6414 page: 206-986-0399 work: 206-389-8210 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05066-9@oink>; Tue, 12 Dec 1995 07:06:40 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818744968:00394:1; Tue, 12 Dec 95 05:09:28 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa00377; 12 Dec 95 5:09 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA04151 for dbi-announce-real; Mon, 11 Dec 1995 20:40:16 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mailgw.claircom.com (mailgw.claircom.com [199.5.241.51]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA04138; Mon, 11 Dec 1995 20:40:00 -0800 Received: from nimo.claircom.com by mailgw.claircom.com with smtp (Smail3.1.26.7 #2) id m0tPMVU-0004thC; Mon, 11 Dec 95 20:40 PST Received: from airone.claircom.com by nimo.claircom.com with smtp (Smail3.1.26.7 #2) id m0tPMVT-0005VkC; Mon, 11 Dec 95 20:39 PST Received: from soc5.airone.claircom.com by airone.claircom.com (5.x/SMI-SVR4) id AA28045; Mon, 11 Dec 1995 20:36:22 -0800 Received: by soc5.airone.claircom.com (4.1/SMI-4.1) id AA03096; Mon, 11 Dec 95 20:20:19 PST Date: Mon, 11 Dec 95 20:20:19 PST From: billw@airone.claircom.com (William Warner) Message-Id: <9512120420.AA03096@soc5.airone.claircom.com> To: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk Subject: Re: New directions for DBI - Important Cc: ak@ananke.s.bawue.de, kejin@lithium.empress.com X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 5:09:27 GMT] I'm lacking a little background information. What is SQL CLI/ODBC ? William Warner billw@airone.claircom.com portable: 206-915-1157 home: 206-720-6414 page: 206-986-0399 work: 206-389-8210 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05066-15@oink>; Tue, 12 Dec 1995 07:07:18 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818745342:02558:1; Tue, 12 Dec 95 05:15:42 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa02540; 12 Dec 95 5:15 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA04229 for dbi-users-real; Mon, 11 Dec 1995 20:49:49 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA04225 for ; Mon, 11 Dec 1995 20:49:47 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ac08291; 12 Dec 95 4:48 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa26909; 12 Dec 95 4:48 GMT Received: from toad by oink with SMTP (PP) id <04625-0@oink>; Tue, 12 Dec 1995 04:47:03 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA27767; Tue, 12 Dec 1995 04:46:33 +0000 Date: Tue, 12 Dec 1995 04:46:33 +0000 From: Tim Bunce Message-Id: <9512120446.AA27767@toad.ig.co.uk> To: dbi-users@fugue.com, billw@airone.claircom.com Subject: Re: New directions for DBI - Important X-Sun-Charset: US-ASCII content-length: 291 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 12 Dec 95 4:48:33 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 12 Dec 95 4:48:51 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 5:15:41 GMT] > From: billw@airone.claircom.com (William Warner) > > I'm lacking a little background information. > > What is SQL CLI/ODBC ? An International Standard 'Call Level Interface' for access to databases from programs. Basically it's a standard API. Exactly what the DBI wants to be. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <08976-13@oink>; Tue, 12 Dec 1995 16:56:59 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818784057:19272:0; Tue, 12 Dec 95 16:00:57 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa19266; 12 Dec 95 16:00 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa26613; 12 Dec 95 15:56 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA06664 for dbi-users-real; Tue, 12 Dec 1995 05:47:18 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mailhost.onramp.net (mailhost.onramp.net [199.1.11.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA06660 for ; Tue, 12 Dec 1995 05:47:15 -0800 Received: from elcaro (elcaro.com [199.1.166.62]) by mailhost.onramp.net (8.6.12/8.6.5) with SMTP id HAA29897; Tue, 12 Dec 1995 07:47:02 -0600 Message-ID: <30CD8790.EB@hometown.com> Date: Tue, 12 Dec 1995 05:45:52 -0800 From: Steve Livingston X-Mailer: Mozilla 2.0b3 (X11; I; SunOS 5.4 sun4m) MIME-Version: 1.0 To: "dbi-users@fugue.com" CC: Tim.Bunce@ig.co.uk Subject: Building DBD-0.27 on Solaris/Sparc Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 15:56:45 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 16:00:56 GMT] Environment ----------- Perl5.001n...tests and runs OK DBI-0.65...builds and tests OK Solaris 2.4 on Sparc 20 GCC 2.6.3 Oracle 7.3 Beta First Problem ------------- make Makefile.pl -> can't find proc.mk Note: O7.3 beta is working OK (make realclean...) % perl Makefile.PL Configuring DBD::Oracle version 0.27... Remember to actually read the README file! Using Oracle in /private/slivings/oracle/orahome Oracle sysliblist: -lsocket -lnsl -lm -ldl -laio Unable to locate proc.mk (use -m /path/to/proc.mk to specify) Next Try -------- Found proc.mk on distribution CD and copied to (OH)/proc/lib (make realclean...) % perl Makefile.PL Configuring DBD::Oracle version 0.27... Remember to actually read the README file! Using Oracle in /private/slivings/oracle/orahome Oracle sysliblist: -lsocket -lnsl -lm -ldl -laio Using /private/slivings/oracle/orahome/proc/lib/proc.mk version System: perl5.001 sunos elcaro 5.4 generic_101945-29 sun4m sparc sun4-solaris dl_dlopen.xs Compiler: gcc -O -I/usr/local/include Oracle proc.mk would have used these values but we override them: CC: cc -O -xcg92 CFLAGS: $(INCLUDE) Checking if your kit is complete... Looks good Warning (non-fatal): Importing of %att is depreciated [PM] use $self instead Warning (non-fatal): Importing of %att is depreciated [PM] use $self instead Warning (non-fatal): Importing of %att is depreciated [EXTRALIBS] use $self instead Warning (non-fatal): Importing of %att is depreciated [EXTRALIBS][-laio -L$(LIBHOME) ] use $self instead CCCMD = gcc -c -I/usr/local/include -O Writing Makefile for DBD::Oracle And Now the MAKE ---------------- % perl Makefile.PL % /usr/ccs/bin/make umask 0 && cp Oraperl.pm ../../lib/Oraperl.pm umask 0 && cp oraperl.ph ../../lib/oraperl.ph umask 0 && cp Oracle.pm ../../lib/DBD/Oracle.pm ../../perl -I../../lib -I../../lib ../../lib/ExtUtils/xsubpp -typemap ../../lib/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c ld -dy /private/slivings/oracle/orahome/lib/crti.o /private/slivings/oracle/orahome/lib/crt1.o /private/slivings/oracle/orahome/lib/__fstd.o -c -c Oracle.c -o Oracle.o /usr/ccs/bin/ld: illegal option -- c /usr/ccs/bin/ld: illegal option -- c usage: ld [-abd:e:h:il:mo:rstu:z:B:D:F:GI:L:M:Q:R:S:VY:] file(s) ... [-a] create an absolute file [-b] do not do special PIC relocations in a.out [-d y|n] operate in dynamic|static mode [-e sym] use `sym' as entry point address [-h name] use `name' as internal shared object identifier [-i] ignore LD_LIBRARY_PATH setting [-l x] search for libx.so or libx.a [-m] print memory map [-o outfile] name the output file `outfile' [-r] create a relocatable object [-s] strip any symbol and debugging information [-t] do not warn of multiply defined symbols of different sizes [-u sym] create an undefined symbol `sym' [-z defs|nodefs] disallow|allow undefined symbols [-z muldefs] allow multiply defined symbols [-z text] disallow output relocations against text [-B dynamic|static] search for shared libraries|archives [-B symbolic] bind external references to definitions when creating shared objects [-D options] print diagnostic messages [-F name] specify library for which this file is a filter [-G] create a shared object [-I interp] use `interp' as path name of interpreter [-L path] search for libraries in directory `path' [-M mapfile] use processing directives contained in `mapfile' [-Q y|n] do|do not place version information in output file [-R path] specify a library search path to be used at run time [-V] print version information [-Y P,dirlist] use `dirlist' as a default path when searching for libraries *** Error code 1 make: Fatal error: Command failed for target `Oracle.o' More Info --------- % perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=solaris, osver=2.4, archname=sun4-solaris uname='sunos elcaro 5.4 generic_101945-29 sun4m sparc ' hint=recommended Compiler: cc='gcc', optimize='-O' cppflags='-I/usr/local/include' ccflags ='-I/usr/local/include' ldflags =' -L/usr/local/lib' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/usr/local/lib /lib /usr/lib /usr/ccs/lib libs=-lsocket -lnsl -ldl -lm -lc -lcrypt libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-fpic', ccdlflags=' ', lddlflags='-G -L/usr/local/lib' Thanks for your contributions to the Web. Cheers, Steve mouche@hometown.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <09291-0@oink>; Tue, 12 Dec 1995 17:02:23 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818787444:21718:0; Tue, 12 Dec 95 16:57:24 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa21712; 12 Dec 95 16:57 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa02449; 12 Dec 95 16:52 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA06712 for dbi-users-real; Tue, 12 Dec 1995 06:18:00 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sultan.hsii.ccare.com ([192.101.181.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id GAA06702 for ; Tue, 12 Dec 1995 06:17:55 -0800 Received: from msgw.hsii.ccare.com by SULTAN.HSII.CCARE.COM (PMDF V5.0-5 #4597) id <01HYPOO5BXF40018RB@SULTAN.HSII.CCARE.COM> for dbi-users@fugue.com; Tue, 12 Dec 1995 08:17:20 -0600 (CST) Received: by msgw.hsii.ccare.com with Microsoft Mail id <30CD8EF4@msgw.hsii.ccare.com>; Tue, 12 Dec 1995 08:17:24 -0600 (CST) Date: Mon, 11 Dec 1995 20:03:00 -0600 (CST) From: "Li, Alexander" Subject: FW: HELP building DBI-0.65 To: DBI User List Message-id: <30CD8EF4@msgw.hsii.ccare.com> X-Mailer: Microsoft Mail V3.0 Content-transfer-encoding: 7BIT Encoding: 297 TEXT X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 16:52:17 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 16:57:23 GMT] Sorry for duplicating a post, but I'm in dire need to solve the following problem. Thanks. Alex Li ---------- From: owner-dbi-users To: DBI User List Subject: HELP building DBI-0.65 Date: Friday, December 08, 1995 12:22PM Dear Perl/DBI experts, Attached at the end is a problem log when our programmer(I'm the sys. admin) try to build DBI-0.65. This happened only *after* I recompiled perl 5 with dynamic loading. i.e. The first time was fine. Our goal is to build a Perl 5 to Oracle 7 interface. Perl 5 tested fine. The platform is a Sun Voyager running Solaris 2.4. Compiler is Sun's SPARCompiler C v3.0. I'm relative new to this stuff and probably made a stupid mistake somewhere. Thanks for you kind help. ---------------------------------- Alex Li Health Systems Integration, Inc. ali@hsii.ccare.com work: 612-858-7570 ---------------------------------- Script started on Fri Dec 08 09:46:13 1995 gh753> perl Makefile.PL Build, TEST and INSTALL Perl 5 first. Use version 5.001m or later. DO NOT build the DBI under the Perl source tree. Use 'make test' to execute self tests. MakeMaker (v4.16) Checking if your kit is complete... Looks good NAME => 'DBI' VERSION => '0.65' Using PERL=/usr/local/bin/perl Writing Makefile for DBI gh753> gh753> make mkdir ./blib mkdir ./blib/auto mkdir ./blib/auto/DBI /usr/local/bin/perl -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/lib/perl5 /usr/local/lib/perl5/ExtUtils/xsubpp -typemap /usr/local/lib/perl5/ExtUtils/typemap DBI.xs >DBI.tc && mv DBI.tc DBI.c cc -c -DDEBUGGING -O -I/usr/local/lib/perl5/sun4-solaris/CORE DBI.c Running Mkbootstrap for DBI () chmod 644 DBI.bs LD_RUN_PATH="" ld -o blib/auto/DBI/DBI.none DBI.o Undefined first referenced symbol in file sv_2iv DBI.o sv_magic DBI.o stack_sp DBI.o markstack_max DBI.o sv_free DBI.o gv_fetchpv DBI.o sv_undef DBI.o av_fetch DBI.o sv_2pv DBI.o newSVpv DBI.o sv_setiv DBI.o mg_find DBI.o __ctype DBI.o gv_fetchmethod DBI.o abort DBI.o hv_fetch DBI.o Xpv DBI.o sv_catpv DBI.o strerror DBI.o newRV DBI.o newSViv DBI.o newSV DBI.o dirty DBI.o malloc DBI.o sprintf DBI.o perl_destruct_level DBI.o av_store DBI.o newXS DBI.o newSVsv DBI.o runlevel DBI.o sv_no DBI.o sv_bless DBI.o sv_2bool DBI.o gv_stashpv DBI.o Sv DBI.o hv_store DBI.o fclose DBI.o perl_call_sv DBI.o sv_mortalcopy DBI.o savepv DBI.o sv_setsv DBI.o sv_catpvn DBI.o strncmp DBI.o sv_backoff DBI.o __iob DBI.o dowantarray DBI.o strcat DBI.o warn DBI.o sv_newmortal DBI.o errno DBI.o stack_base DBI.o stack_max DBI.o av_len DBI.o sv_setpv DBI.o sv_2mortal DBI.o strcmp DBI.o fopen DBI.o memset DBI.o croak DBI.o strcpy DBI.o markstack_grow DBI.o stack_grow DBI.o na DBI.o sv_dump DBI.o markstack_ptr DBI.o op DBI.o perl_get_sv DBI.o strlen DBI.o perl_call_method DBI.o fprintf DBI.o newAV DBI.o sv_yes DBI.o ld: fatal: Symbol referencing errors. No output written to blib/auto/DBI/DBI.none *** Error code 1 make: Fatal error: Command failed for target `blib/auto/DBI/DBI.none' gh753> gh753> gh753> perl -e ` 'use Config; print Conig  fig::myconif  fig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=solaris, osver=2.4, archname=sun4-solaris uname='sunos gh753 5.4 generic_101945-10 sun4m sparc ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-DDEBUGGING' ccflags ='-DDEBUGGING' ldflags ='' stdchar='unsigned char', d_stdstdio=, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/lib /usr/lib /usr/ccs/lib libs=-lsocket -lnsl -ldl -lm -lc -lcrypt libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_none.xs, dlext=none, d_dlsymun= cccdlflags='', ccdlflags='', lddlflags='' gh753> script done on Fri Dec 08 09:51:39 1995 ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <01794-8@oink>; Wed, 13 Dec 1995 07:06:38 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818816457:05541:0; Wed, 13 Dec 95 01:00:57 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa05451; 13 Dec 95 1:00 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA00124 for dbi-users-real; Tue, 12 Dec 1995 14:06:55 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sparc20.cs.uiuc.edu (dswallow@sparc20.cs.uiuc.edu [128.174.244.30]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA08345 for ; Tue, 12 Dec 1995 13:13:10 -0800 From: dswallow@sparc20.cs.uiuc.edu Received: by sparc20.cs.uiuc.edu id AA08697 (5.67b/IDA-1.5 for dbi-users@fugue.com); Tue, 12 Dec 1995 15:13:03 -0600 Date: Tue, 12 Dec 1995 15:13:03 -0600 Message-Id: <199512122113.AA08697@sparc20.cs.uiuc.edu> To: dbi-users@fugue.com Content-Length: 5046 X-SMTP-Posting-Host: toccata.fugue.com [Wed, 13 Dec 95 1:00:52 GMT] ***** Removing old dbdimp.c ***** ***** Generating new Makefile ***** Configuring DBD::Informix version 0.20pl0... Remember to actually read the README file! Using Informix in /usr/local/informix Informix sysliblist: -L/usr/local/informix/lib/esql -lsql -lgen -los -lm Warning: forced to build static not dynamic on hpux See README file for more information. System: perl5.001 hp-ux hp47 a.09.05 a 9000725 2011486631 two-user license 9000/725-hpux dl_hpux.xs Compiler: cc -O -D_HPUX_SOURCE -Aa Writing Makefile for DBD::Informix ***** Attaching preprocessor information to Makefile ***** ***** Generating MAXCURSOR information ***** umask 0 && cp Informix.pm ./blib/DBD/Informix.pm umask 0 && cp mSQLtest.pl ./blib/DBD/mSQLtest.pl ./perl -I/tmp_mnt/home/student9/marcn/dev/cs292/lib/9000/725-hpux -I/tmp_mnt/home/student9/marcn/dev/cs292/lib /tmp_mnt/home/student9/marcn/dev/cs292/lib/ExtUtils/xsubpp -typemap /tmp_mnt/home/student9/marcn/dev/cs292/lib/ExtUtils/typemap Informix.xs >Informix.tc && mv Informix.tc Informix.c cc -c -D_HPUX_SOURCE -Aa -O -DVERSION=\"0.20pl0\" +z -I/tmp_mnt/home/student9/marcn/dev/cs292/lib/9000/725-hpux/CORE -I/usr/local/informix/incl -I/tmp_mnt/home/student9/marcn/dev/cs292/lib/9000/725-hpux/DBI Informix.c cc: "Informix.c", line 133: warning 608: Illegal integer-pointer combination in assignment. cc: "Informix.c", line 324: warning 608: Illegal integer-pointer combination in assignment. cc: "Informix.c", line 356: warning 608: Illegal integer-pointer combination in assignment. esql -c -I/usr/local/informix/incl -I/tmp_mnt/home/student9/marcn/dev/cs292/lib -I/tmp_mnt/home/student9/marcn/dev/cs292/lib/9000/725-hpux -I/tmp_mnt/home/student9/marcn/dev/cs292/lib/9000/725-hpux/DBI -I/tmp_mnt/home/student9/marcn/dev/cs292/lib/9000/725-hpux/CORE dbdimp.ec dbdimp.ec: In function `dbd_describe': dbdimp.ec:561: warning: overflow in implicit constant conversion gcc: /usr/local/informix/lib/esql/libsql.a: linker input file unused since linking not done gcc: /usr/local/informix/lib/esql/libgen.a: linker input file unused since linking not done gcc: /usr/local/informix/lib/esql/libos.a: linker input file unused since linking not done ar cr ./blib/9000/725-hpux/auto/DBD/Informix/Informix.a Informix.o dbdimp.o && : ./blib/9000/725-hpux/auto/DBD/Informix/Informix.a chmod 755 ./blib/9000/725-hpux/auto/DBD/Informix/Informix.a install_driver(Informix) failed: Can't load './Informix.o' for module DBD::Informix: Exec format error at /tmp_mnt/home/student9/marcn/dev/cs292/lib/DynaLoader.pm line 136. at /tmp_mnt/home/student9/marcn/dev/cs292/lib/DBD/Informix.pm line 18 at /tmp_mnt/home/student9/marcn/dev/cs292/lib/DBI.pm line 180 DBI::install_driver called at ./test.pl line 12 DBI dispatch debug level set to 2 DBI->install_driver DBI Informix at ./test.pl line 12 install_driver(Informix) failed: Can't load './Informix.o' for module DBD::Informix: Exec format error at /tmp_mnt/home/student9/marcn/dev/cs292/lib/DynaLoader.pm line 136. at /tmp_mnt/home/student9/marcn/dev/cs292/lib/DBD/Informix.pm line 18 at /tmp_mnt/home/student9/marcn/dev/cs292/lib/DBI.pm line 180 DBI::install_driver called at ./test.pl line 12 DBI::END DBI::disconnect_all DBI DBI::END complete >> DESTROY DISPATCH (DBI::dr=HASH(0x40063498) @1 g0 a4009cd68 r0) (outer handle DESTROY ignored) >> DESTROY DISPATCH (DBI::dr=HASH(0x4004fc38) @1 g0 a4009cd68 r0) dbih_clearcom 0x40074208 done Bad free() ignored during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Bad free() ignored during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. Attempt to free unreferenced scalar during global destruction. I am using perl 5.001m and esql 5.00. Note that it was necessary to use gcc instead of cc in the esql compilation. Summary of my perl5 (patchlevel 1) configuration: Platform: osname=hpux, osver=9, archname=9000/725-hpux uname='hp-ux hp47 a.09.05 a 9000725 2011486631 two-user license ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-D_HPUX_SOURCE -Aa' ccflags ='-D_HPUX_SOURCE -Aa' ldflags =' -L/usr/local/lib' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=sl libpth=/usr/local/lib /lib/pa1.1 /lib /usr/lib libs=-lnet -ldld -lm -lc -lndir -lcrypt libc=/lib/libc.sl Dynamic Linking: dlsrc=dl_hpux.xs, dlext=sl, d_dlsymun= cccdlflags='+z', ccdlflags='-Wl,-E ', lddlflags='-b -L/usr/local/lib' ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <01794-11@oink>; Wed, 13 Dec 1995 07:07:12 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818817553:13846:0; Wed, 13 Dec 95 01:19:13 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa13821; 13 Dec 95 1:19 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA00516 for dbi-users-real; Tue, 12 Dec 1995 15:21:33 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mailhost.onramp.net (mailhost.onramp.net [199.1.11.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA00512 for ; Tue, 12 Dec 1995 15:21:31 -0800 Received: from elcaro (elcaro.com [199.1.166.62]) by mailhost.onramp.net (8.6.12/8.6.5) with SMTP id RAA16863; Tue, 12 Dec 1995 17:21:33 -0600 Message-ID: <30CE0E56.24F9@hometown.com> Date: Tue, 12 Dec 1995 15:20:54 -0800 From: Steve Livingston X-Mailer: Mozilla 2.0b3 (X11; I; SunOS 5.4 sun4m) MIME-Version: 1.0 To: dbi-users@fugue.com CC: Tim.Bunce@ig.co.uk Subject: Building DBD-0.27 on Solaris/Sparc...again Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Wed, 13 Dec 95 1:19:10 GMT] Thanks for your suggestions... I've installed Pro*C, and the O7beta install log says it was successful, although I can't run the Pro*C demo program because it wants the SUNWspro compiler with the -xcg92 flag. I still receive the same fatal error for "perl Makefile.PL" I looked in vain for a MakeMaker on CPAN. Cheers, Steve ---------- % make realclean rm -rf Oracle.c ./blib Makeaperlfile ../../lib/auto/DBD/Oracle/extralibs.all perlmain.c mon.out core so_locations *~ */*~ */*/*~ *.o *.a perl.exe Oracle.bs Oracle.bso Oracle.def Oracle.exp mv Makefile Makefile.old 2>/dev/null rm -rf ../../lib/auto/DBD/Oracle ../../lib/auto/DBD/Oracle rm -f ../../lib/auto/DBD/Oracle/Oracle.so ../../lib/auto/DBD/Oracle/Oracle.bs rm -f ../../lib/auto/DBD/Oracle/Oracle.a ../../lib/Oraperl.pm ../../lib/oraperl.ph ../../lib/DBD/Oracle.pm rm -rf Makefile Makefile.old ---------- elcaro% perl Makefile.PL Configuring DBD::Oracle version 0.27... Remember to actually read the README file! Using Oracle in /private/slivings/oracle/orahome Oracle sysliblist: -lsocket -lnsl -lm -ldl -laio Using /private/slivings/oracle/orahome/proc/lib/proc.mk version System: perl5.001 sunos elcaro 5.4 generic_101945-29 sun4m sparc sun4-solaris dl_dlopen.xs Compiler: gcc -O -I/usr/local/include Oracle proc.mk would have used these values but we override them: CC: cc -O -xcg92 CFLAGS: $(INCLUDE) Checking if your kit is complete... Looks good Warning (non-fatal): Importing of %att is depreciated [PM] use $self instead Warning (non-fatal): Importing of %att is depreciated [PM] use $self instead Warning (non-fatal): Importing of %att is depreciated [EXTRALIBS] use $self instead Warning (non-fatal): Importing of %att is depreciated [EXTRALIBS][-laio -L$(LIBHOME) ] use $self instead CCCMD = gcc -c -I/usr/local/include -O Writing/make umask 0 && cp Oraperl.pm ../../lib/Oraperl.pm umask 0 && cp oraperl.ph ../../lib/oraperl.ph umask 0 && cp Oracle.pm ../../lib/DBD/Oracle.pm ../../perl -I../../lib -I../../lib ../../lib/ExtUtils/xsubpp -typemap ../../lib/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c ld -dy /private/slivings/oracle/orahome/lib/crti.o /private/slivings/oracle/orahome/lib/crt1.o /private/slivings/oracle/orahome/lib/__fstd.o -c -c Oracle.c -o Oracle.o /usr/ccs/bin/ld: illegal option -- c /usr/ccs/bin/ld: illegal option -- c usage: ld [-abd:e:h:il:mo:rstu:z:B:D:F:GI:L:M:Q:R:S:VY:] file(s) ... [-a] create an absolute file [-b] do not do special PIC relocations in a.out [-d y|n] operate in dynamic|static mode ... *** Error code 1 make: Fatal error: Command failed for target `Oracle.o' Makefile for DBD::Oracle --------- ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <01794-14@oink>; Wed, 13 Dec 1995 07:08:15 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818818316:18919:0; Wed, 13 Dec 95 01:31:56 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa18902; 13 Dec 95 1:31 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA00593 for dbi-users-real; Tue, 12 Dec 1995 15:44:13 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from paul.spu.edu (Paul.spu.edu [192.190.33.3]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA00582; Tue, 12 Dec 1995 15:42:55 -0800 Received: by paul.spu.edu; (5.65/1.1.8.2/17Sep95-0400PM) id AA18304; Tue, 12 Dec 1995 15:39:26 -0800 Date: Tue, 12 Dec 1995 15:39:26 -0800 (PST) From: Phil Rand To: Tim Bunce Cc: dbi-announce@fugue.com, dbi-users@fugue.com, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important In-Reply-To: <9512120129.AA27073@toad.ig.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: toccata.fugue.com [Wed, 13 Dec 95 1:31:55 GMT] On Tue, 12 Dec 1995, Tim Bunce wrote: > To cut a long story short I've decided that we should throw away a > significant chunk of the work that we've done and reengineer an > implementation based very strongly around the SQL CLI / ODBC API. Hooray!!!! When will you have it done? (Just kidding.) I doubt whether I have the technical skills to contribute much, but I'll be following along as best I can. As I recall, ODBC defines two or three different implementation levels, from simple to complex. Would you expect to follow those? Or are the easier ones too simple to be worth doing by themselves? You might want to consider abstracting out, somehow, the manor in which you interface with platform-dependent facilities. On Win32, for example, I believe one could go directly to platform-supplied odbc routines from c, or one could go through an OCX (OLE control). I assume most Unix platforms would need a much lower-level interface to ODBC. If the new api could gracefully adapt to all these circumstances, that would be a big win. -- -- Phil Rand aka -- Computer & Information Systems (206) 281-2428 -- Seattle Pacific University, 3307 3rd Ave W, Seattle, WA 98119 -- http://paul.spu.edu/~prand/ -- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01795-18@oink>; Wed, 13 Dec 1995 07:08:43 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818791956:05257:0; Tue, 12 Dec 95 18:12:36 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa05250; 12 Dec 95 18:12 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa01963; 12 Dec 95 16:45 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA06809 for dbi-users-real; Tue, 12 Dec 1995 07:02:13 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA06805 for ; Tue, 12 Dec 1995 07:02:08 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ad28990; 12 Dec 95 15:01 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ac15832; 12 Dec 95 15:01 GMT Received: from toad by oink with SMTP (PP) id <08342-0@oink>; Tue, 12 Dec 1995 14:52:19 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA29909; Tue, 12 Dec 1995 14:51:51 +0000 Date: Tue, 12 Dec 1995 14:51:51 +0000 From: Tim Bunce Message-Id: <9512121451.AA29909@toad.ig.co.uk> To: dbi-users@fugue.com, mouche@hometown.com Subject: Re: Building DBD-0.27 on Solaris/Sparc X-Sun-Charset: US-ASCII content-length: 420 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Tue, 12 Dec 95 15:01:34 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Tue, 12 Dec 95 15:01:53 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 16:45:33 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 18:12:35 GMT] > From: Steve Livingston > > Next Try > -------- > Found proc.mk on distribution CD and copied to (OH)/proc/lib Don't do that! You should properly install the whole Pro*C package. > Warning (non-fatal): Importing of %att is depreciated [PM] > use $self instead Anyone who gets this message and has build problems should upgrade to the latest MakeMaker available from a CPAN site. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01795-37@oink>; Wed, 13 Dec 1995 07:11:45 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818800173:27284:0; Tue, 12 Dec 95 20:29:33 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa27278; 12 Dec 95 20:29 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa14416; 12 Dec 95 18:30 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA07208 for dbi-users-real; Tue, 12 Dec 1995 09:12:54 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from ns2.eds.com (ns2.eds.com [199.228.142.78]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA07204 for ; Tue, 12 Dec 1995 09:12:53 -0800 Received: by ns2.eds.com (hello) id MAA21964; Tue, 12 Dec 1995 12:12:57 -0500 Received: by nnsp.eds.com (hello) id MAA01786; Tue, 12 Dec 1995 12:12:23 -0500 Received: from falcon (falcon.mmts.eds.com) by geronimo (5.x/SMI.SVR4) id AA09877; Tue, 12 Dec 1995 11:11:06 -0600 Received: by falcon (940816.SGI.8.6.9/940406.SGI) id LAA18736; Tue, 12 Dec 1995 11:12:58 -0900 Date: Tue, 12 Dec 1995 11:12:58 -0900 Message-Id: <199512122012.LAA18736@falcon> To: dbi-users@fugue.com Subject: can DBI return rows as hashes keyed by column names? From: Andy Cohan X-Attribution: AFC X-Face: "P!:r4>2dA)dWK/L(NxSG)-Ffvhe/]5y%uxD}zzNJn{omcc&K8oU>ZXTvJQ,.&Z4C}m)m-8 mzKkq--xzt@CUf'FIp2^Jz(=Fj,t^rU@TfMex\1V=;+j-=o~bv$h3 Reply-To: afc@mmts.eds.com X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 18:30:31 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 20:29:33 GMT] I'm trying to get our own perl5 DB class to use DBI for an Oracle back end and I would like to support the Perl5 Sybase facility of optionally returning rows as associative arrays keyed by the result column names (we have used it extensively). Does DBI and DBD::Oracle support this? If not, is it planned? AFC P.S. Please forgive any manifest ignorance; I just started looking at this stuff yesterday (but I did try all the doco at http://www.hermetica.com/technologia/DBI) Andrew F. Cohan Senior Systems Engineer afc@mmts.eds.com Advanced Technology Group A-(on its way AND LATE) EDS Internet & New Media Services Bones heal, Chicks dig scars, Glory is forever, so hook it in! ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01795-45@oink>; Wed, 13 Dec 1995 07:13:10 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818804207:07893:0; Tue, 12 Dec 95 21:36:47 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa07887; 12 Dec 95 21:36 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id msg.aa21923; 12 Dec 95 19:32 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA07341 for dbi-users-real; Tue, 12 Dec 1995 09:43:37 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from austin.asc.slb.com (PIGPEN-CLEAR.AUSTIN.ASC.SLB.COM [163.185.74.6]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA07337 for ; Tue, 12 Dec 1995 09:43:34 -0800 From: guthery@austin.sar.slb.com Received: from asterix (ASTERIX.AUSTIN.ASC.SLB.COM) by austin.asc.slb.com (4.1/relay.950801a) id AA16297; Tue, 12 Dec 95 11:43:00 CST Received: by asterix (4.1/client.nfs.930922a) id AA24868; Tue, 12 Dec 95 11:42:55 CST Date: Tue, 12 Dec 95 11:42:55 CST Message-Id: <9512121742.AA24868@asterix> To: Vern.Martin@bellhow.bellhow.com Cc: dbi-users@fugue.com, beecher@bellhow.bellhow.com, rich@bellhow.bellhow.com In-Reply-To: <199512121350.IAA28356@bh2.bellhow.com> (message from Vern Martin on Tue, 12 Dec 1995 08:50:11 -0500 (EST)) Subject: Re: Making oraperl for Solaris X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 19:32:54 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 21:36:47 GMT] The following makefile worked builds oraperl for me on Solaris. Ignore the ph stuff. It just provides access to our ph server from inside oraperl. # Makefile for Oraperl and Coraperl # Change these to your ORACLE installation directory and Perl source directory # ORACLE_HOME = /usr/local/oracle-7.0.16 SRC = /home/ugr0/perl/perl-4.035 # Oracle Definitions, copied from $(ORACLE_HOME)/c/demo/proc.mk # ALL_ORA_LIBS is the only entry that the Makefile actually uses; # change it to whatever you need to link Pro*C programs # OTHERLIBS = `cat $(ORACLE_HOME)/rdbms/lib/sysliblist` CLIBS = $(OTHERLIBS) OCILIB = -locic NETLIBS = $(ORACLE_HOME)/lib/osntab.o -lsqlnet -lnetwork ORALIBS = -lora -lcv6 -lcore -lsqlnet -lnlsrtl -lnetwork -lcore ALL_ORA_LIBS = -L$(ORACLE_HOME)/lib $(CLIBS) $(OCILIB) $(NETLIBS) $(ORALIBS) # Perl Definitions, taken from $SRC/usub/Makefile # Don't include the curses libraries here - they go in CURSELIB # GLOBINCS = LOCINCS = LIBS = `. $(SRC)/config.sh; echo $$libs` # Oraperl Definition # Set DEBUG to -DDEBUGGING, -DPERL_DEBUGGING or leave blank (see orafns.h) # If it is not blank, uncomment the definition of DBUG_O # #DEBUG = -DPERL_DEBUGGING #DBUG_O = dbug/dbug.o # Curses libraries, only required if you want to build Coraperl # You may also need -ltermlib or -ltermcap # CURSELIB = -lcurses # Row cache size for SELECT statements. # If you want to change the default, uncomment this and set the value you want # #CACHE = -DCACHE_SIZE=5 # Bind variable padding. # If you want binding an empty string to provoke an error, uncomment this line. # #BIND = -DNO_BIND_PADDING # Some system-specific things # # If your system library does not include strtoul, uncomment the next line STRTOUL = strtoul.o # # If your malloc() returns anything other than a char *, set the appropriate # type here (don't include the *) # MALLOC_PTR_TYPE=void # # If you are using Perl v3 instead of v4, uncomment the next line # STR_2MORTAL = -Dstr_2mortal=str_2static # Database, username and password to use for testing # TESTDATA = # # Leave these blank lines so that patches to what is below # won't be upset by your changes to the setups above. # # From here on, you shouldn't need to change anything. If you do, let me know. SRCS = oracle.mus orafns.c getcursor.c colons.c usersub.c \ debug.c strtoul.c ph-slb.c ph.c fcrypt.c cryptit.c OBJS = oracle.o orafns.o ph-slb.o ph.o fcrypt.o cryptit.o \ getcursor.o colons.o debug.o $(STRTOUL) #OOBJS = $(OBJS) usersub.o $(DBUG_O) $(ORACLE_HOME)/rdbms/lib/kpundf.o OOBJS = $(OBJS) usersub.o $(DBUG_O) COBJS = $(OBJS) cusersub.o $(DBUG_O) HDRS = patchlevel.h orafns.h DEFS = $(STRTOL) $(PUTENV) $(STR_2MORTAL) $(DEBUG) $(CACHE) $(BIND) CC = cc CFLAGS = -I. -Idbug -I$(SRC) $(GLOBINCS) $(LOCINCS) $(DEFS) -DBUILTIN_CLEAR -DFCNTL_LOCK -DSYSV -DWAIT_INT -DPH -DI_DIRENT -DHASSTRTOK -I. -DEMAIL_AUTH=\"1\" -DMAILDOMAIN=\"slb.com\" -DFALLBACKPORT=\"105\" -DHOST=\"ph.slb.com\" -DNSSERVICE=\"csnet-ns\" -DFALLBACKADDR=\"163.185.74.8\" oraperl: $(SRC)/uperl.o $(OOBJS) $(CC) -o oraperl $(SRC)/uperl.o $(OOBJS) \ -lm $(ALL_ORA_LIBS) $(LIBS) coraperl: $(SRC)/uperl.o $(COBJS) $(SRC)/usub/curses.o $(CC) -o coraperl $(SRC)/uperl.o $(COBJS) $(SRC)/usub/curses.o \ -lm $(ALL_ORA_LIBS) $(LIBS) $(CURSELIB) all: oraperl coraperl test: oraperl @oraperl -e '&ora_version' @(cd testdir ; \ rm -f My-Results ; \ echo "Testing oraperl, please wait ..." ; \ for i in *.pl ; do ../oraperl $$i $(TESTDATA) ; done > My-Results ; \ if cmp -s Standard-Results My-Results ; \ then echo "Test successful" ; \ else echo "Test failed - compare My-Results with Standard-Results" ;\ fi; echo) # We use oraperl although perl would suffice, because we know where it is! install: oraperl install.pl @./oraperl ./install.pl ${SRC} cusersub.c: usersub.c @rm -f cusersub.c ln usersub.c cusersub.c cusersub.o: cusersub.c $(CC) -c $(CFLAGS) -DCURSES cusersub.c oracle.c: $(SRC)/usub/mus oracle.mus $(SRC)/usub/mus oracle.mus >oracle.c $(OOBJS) $(COBJS): $(HDRS) dbug/dbug.o: (cd dbug ; $(MAKE) dbug.o) @echo " (back to main directory)" clean: (cd dbug ; $(MAKE) clean) @echo " (back to main directory)" rm -f nohup.out *.o oracle.c cusersub.c rm -f testdir/My-Results listing tags core realclean clobber: clean (cd dbug ; $(MAKE) clobber) (cd doc ; $(MAKE) clobber) @echo " (back to main directory)" rm -f oraperl coraperl listing: pr -fn Makefile $(HDRS) $(SRCS) >listing docs: (cd doc ; $(MAKE) docs) @echo " (back to main directory)" shar: clean shar -n oraperl-v2 -a -s kstock@encore.com -F -o :Part -l 64 \ Readme [C-Q]* Row* [S-z]* ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01928-26@oink>; Wed, 13 Dec 1995 07:16:35 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818811461:21110:0; Tue, 12 Dec 95 23:37:41 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa21105; 12 Dec 95 23:37 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa12957; 12 Dec 95 18:17 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA06674 for dbi-users-real; Tue, 12 Dec 1995 05:50:48 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from bellhow.bellhow.com (bellhow.bellhow.com [198.30.176.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA06670 for ; Tue, 12 Dec 1995 05:50:39 -0800 Received: from bh2.bellhow.com (bh2.bellhow.com [198.30.176.8]) by bellhow.bellhow.com (8.7/8.7) with ESMTP id IAA09311; Tue, 12 Dec 1995 08:50:10 -0500 (EST) From: Vern Martin Received: (from vern@localhost) by bh2.bellhow.com (8.7/8.7) id IAA28356; Tue, 12 Dec 1995 08:50:12 -0500 (EST) Message-Id: <199512121350.IAA28356@bh2.bellhow.com> Subject: making oraper for Solaris To: dbi-users@fugue.com Date: Tue, 12 Dec 1995 08:50:11 -0500 (EST) Cc: Bob Beecher , Rich Woodland X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 18:17:49 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 23:37:41 GMT] We have not succeeded in making oraperl on Solaris. Have any of you? If so, could you send a make file? ................................. Thanks, Vern ----------------------------------------------------------------------------- Vern.Martin@bellhow.com ph. 216-642-9060 x. 6216 ----------------------------------------------------------------------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01928-27@oink>; Wed, 13 Dec 1995 07:16:48 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818811484:21345:0; Tue, 12 Dec 95 23:38:04 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa21338; 12 Dec 95 23:38 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa11104; 12 Dec 95 18:04 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA07036 for dbi-users-real; Tue, 12 Dec 1995 08:19:47 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from clothos.candle.com (clothos.Candle.Com [204.217.194.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA07032 for ; Tue, 12 Dec 1995 08:19:44 -0800 Received: from phobos.Candle.Com by clothos.candle.com with SMTP id AA01078 (5.67b/IDA-1.5 for ); Tue, 12 Dec 1995 08:19:29 -0800 Received: from smtpgwy.Candle.Com by phobos.Candle.Com with SMTP id AA12943 (5.67b/IDA-1.5); Tue, 12 Dec 1995 08:19:27 -0800 Received: from ccMail by SMTPGWY.CANDLE.COM (SMTPLINK V2.10.05) id AA818785176; Tue, 12 Dec 95 08:12:39 PST Date: Tue, 12 Dec 95 08:12:39 PST From: Vendetta Black Message-Id: <9511128187.AA818785176@SMTPGWY.CANDLE.COM> To: dbi-users@fugue.com, Tim Bunce Subject: Re: Perl5 & informix X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 18:04:16 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 23:38:04 GMT] You did not mention either the version of AIX or INFORMIX. perl5.001m compiles cleanly on AIX 3.2.5 If you are using version 6.X (and likely 7.X) of INFORMIX Add the following libraries: -lbsd -lmsaa -lc -ltli (AIX) /informix/lib/libasf.a /informix/lib/netlib.a /informix/lib/nettlistub.a Also change /informix/incl to /informix/incl/esql Sorry, I am not very experienced when it comes to PERL so I had to cheat and modify Makefile instead of Makefile.PL. :) ______________________________ Reply Separator _________________________________ Subject: Perl5 & informix Author: Tim Bunce at INTERNET Date: 12/11/95 7:34 PM ----- Begin Included Message ----- Date: Mon, 11 Dec 1995 08:51:25 -0500 From: Bahram Samani To: Tim.Bunce@ig.co.uk Subject: Perl5 & informix X-Url: http://www.hermetica.com/technologia/DBI/ We are not able to do a complete make for perl5, informix, and AIX, or HPUX at Disney. Is there a new version of the beta? Thanks! Bahram ----- End Included Message ----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <02010-4@oink>; Wed, 13 Dec 1995 07:17:54 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818812313:29484:0; Tue, 12 Dec 95 23:51:53 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa29477; 12 Dec 95 23:51 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa26493; 12 Dec 95 20:08 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA07685 for dbi-users-real; Tue, 12 Dec 1995 10:52:20 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from colossus.cse.psu.edu (root@colossus.cse.psu.edu [130.203.1.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA07681 for ; Tue, 12 Dec 1995 10:52:15 -0800 Received: from localhost by colossus.cse.psu.edu with SMTP id <78374>; Tue, 12 Dec 1995 13:52:05 -0500 X-Mailer: exmh version 1.6.1 5/23/95 To: "Li, Alexander" cc: DBI User List Subject: Re: FW: HELP building DBI-0.65 In-reply-to: Your message of "Mon, 11 Dec 1995 21:03:00 EST." <30CD8EF4@msgw.hsii.ccare.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 12 Dec 1995 13:51:20 -0500 From: John D Groenveld Message-Id: <95Dec12.135205est.78374@colossus.cse.psu.edu> X-SMTP-Posting-Host: toccata.fugue.com [Tue, 12 Dec 95 20:08:10 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 12 Dec 95 23:51:53 GMT] Perhaps, you should be using perl5.001n and gcc2.7.2? John groenvel@cse.psu.edu ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01795-81@oink>; Wed, 13 Dec 1995 07:23:33 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818819572:08148:0; Wed, 13 Dec 95 01:52:52 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa07945; 13 Dec 95 1:52 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 13 Dec 1995 01:51:35 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA00586 for dbi-announce-real; Tue, 12 Dec 1995 15:42:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from paul.spu.edu (Paul.spu.edu [192.190.33.3]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA00582; Tue, 12 Dec 1995 15:42:55 -0800 Received: by paul.spu.edu; (5.65/1.1.8.2/17Sep95-0400PM) id AA18304; Tue, 12 Dec 1995 15:39:26 -0800 Date: Tue, 12 Dec 1995 15:39:26 -0800 (PST) From: Phil Rand To: Tim Bunce Cc: dbi-announce@fugue.com, dbi-users@fugue.com, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important In-Reply-To: <9512120129.AA27073@toad.ig.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 13 Dec 95 1:52:24 GMT] On Tue, 12 Dec 1995, Tim Bunce wrote: > To cut a long story short I've decided that we should throw away a > significant chunk of the work that we've done and reengineer an > implementation based very strongly around the SQL CLI / ODBC API. Hooray!!!! When will you have it done? (Just kidding.) I doubt whether I have the technical skills to contribute much, but I'll be following along as best I can. As I recall, ODBC defines two or three different implementation levels, from simple to complex. Would you expect to follow those? Or are the easier ones too simple to be worth doing by themselves? You might want to consider abstracting out, somehow, the manor in which you interface with platform-dependent facilities. On Win32, for example, I believe one could go directly to platform-supplied odbc routines from c, or one could go through an OCX (OLE control). I assume most Unix platforms would need a much lower-level interface to ODBC. If the new api could gracefully adapt to all these circumstances, that would be a big win. -- -- Phil Rand aka -- Computer & Information Systems (206) 281-2428 -- Seattle Pacific University, 3307 3rd Ave W, Seattle, WA 98119 -- http://paul.spu.edu/~prand/ -- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <02080-5@oink>; Wed, 13 Dec 1995 07:25:08 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818826611:29300:0; Wed, 13 Dec 95 03:50:11 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa29297; 13 Dec 95 3:50 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id sg.ab01777; 13 Dec 95 3:20 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 13 Dec 1995 01:51:35 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA00586 for dbi-announce-real; Tue, 12 Dec 1995 15:42:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from paul.spu.edu (Paul.spu.edu [192.190.33.3]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA00582; Tue, 12 Dec 1995 15:42:55 -0800 Received: by paul.spu.edu; (5.65/1.1.8.2/17Sep95-0400PM) id AA18304; Tue, 12 Dec 1995 15:39:26 -0800 Date: Tue, 12 Dec 1995 15:39:26 -0800 (PST) From: Phil Rand To: Tim Bunce Cc: dbi-announce@fugue.com, dbi-users@fugue.com, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important In-Reply-To: <9512120129.AA27073@toad.ig.co.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 13 Dec 95 3:20:54 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Wed, 13 Dec 95 3:50:10 GMT] On Tue, 12 Dec 1995, Tim Bunce wrote: > To cut a long story short I've decided that we should throw away a > significant chunk of the work that we've done and reengineer an > implementation based very strongly around the SQL CLI / ODBC API. Hooray!!!! When will you have it done? (Just kidding.) I doubt whether I have the technical skills to contribute much, but I'll be following along as best I can. As I recall, ODBC defines two or three different implementation levels, from simple to complex. Would you expect to follow those? Or are the easier ones too simple to be worth doing by themselves? You might want to consider abstracting out, somehow, the manor in which you interface with platform-dependent facilities. On Win32, for example, I believe one could go directly to platform-supplied odbc routines from c, or one could go through an OCX (OLE control). I assume most Unix platforms would need a much lower-level interface to ODBC. If the new api could gracefully adapt to all these circumstances, that would be a big win. -- -- Phil Rand aka -- Computer & Information Systems (206) 281-2428 -- Seattle Pacific University, 3307 3rd Ave W, Seattle, WA 98119 -- http://paul.spu.edu/~prand/ -- ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <06541-1@oink>; Wed, 13 Dec 1995 13:46:08 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818850387:08639:0; Wed, 13 Dec 95 10:26:27 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id aa08598; 13 Dec 95 10:26 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Wed, 13 Dec 1995 05:51:22 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA02173 for dbi-users-real; Tue, 12 Dec 1995 20:04:41 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from colossus.cse.psu.edu (root@colossus.cse.psu.edu [130.203.1.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA02169 for ; Tue, 12 Dec 1995 20:04:39 -0800 Received: by colossus.cse.psu.edu id <78398>; Tue, 12 Dec 1995 23:04:32 -0500 From: John D Groenveld To: mouche@hometown.com Subject: Re: Building DBD-0.27 on Solaris/Sparc...again Cc: dbi-users@fugue.com Message-Id: <95Dec12.230432est.78398@colossus.cse.psu.edu> Date: Tue, 12 Dec 1995 23:04:27 -0500 X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Wed, 13 Dec 95 10:26:27 GMT] I believe there's a note in the package to use gcc instead of ld. Try make LD=gcc ? I may be reaching into the fragmented slice of my brain however. John groenvel@cse.psu.edu ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <06569-0@oink>; Wed, 13 Dec 1995 13:48:31 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818861662:23485:2; Wed, 13 Dec 95 13:34:22 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa23165; 13 Dec 95 13:33 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id DAA04264 for dbi-users-real; Wed, 13 Dec 1995 03:28:51 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mailhost.onramp.net (mailhost.onramp.net [199.1.11.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id DAA04260 for ; Wed, 13 Dec 1995 03:28:49 -0800 Received: from elcaro (elcaro.com [199.1.166.62]) by mailhost.onramp.net (8.6.12/8.6.5) with SMTP id FAA24525; Wed, 13 Dec 1995 05:28:52 -0600 Message-ID: <30CEB8D3.414@hometown.com> Date: Wed, 13 Dec 1995 05:28:19 -0600 From: Steve Livingston X-Mailer: Mozilla 2.0b3 (X11; I; SunOS 5.4 sun4m) MIME-Version: 1.0 To: dbi-users@fugue.com CC: Tim.Bunce@ig.co.uk Subject: Building DBD-0.27 on Solaris 2.4...more Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Wed, 13 Dec 95 13:33:37 GMT] My make (with a new version of ld) for DBD-027 shows: umask 0 && cp Oraperl.pm ../../lib/Oraperl.pm umask 0 && cp oraperl.ph ../../lib/oraperl.ph umask 0 && cp Oracle.pm ../../lib/DBD/Oracle.pm ../../perl -I../../lib -I../../lib ../../lib/ExtUtils/xsubpp -typemap ../../lib/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c ld -dy /private/slivings/oracle/orahome/lib/crti.o /private/slivings/oracle/orahome/lib/crt1.o /private/slivings/oracle/orahome/lib/__fstd.o -c -c Oracle.c -o Oracle.o ld: cannot open linker script file -c: No such file or directory Hmm.. looks like "-c -c Oracle.c" is offensive. Removing that string gives undefined references in crt1.o Any ideas? Cheers, Steve ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13872-1@oink>; Thu, 14 Dec 1995 12:21:15 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818895841:24634:1; Wed, 13 Dec 95 23:04:01 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa23942; 13 Dec 95 23:03 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA04728 for dbi-users-real; Wed, 13 Dec 1995 11:22:43 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA04724 for ; Wed, 13 Dec 1995 11:22:17 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ab06580; 13 Dec 95 18:57 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa23714; 13 Dec 95 18:57 GMT Received: from toad by oink with SMTP (PP) id <10229-0@oink>; Wed, 13 Dec 1995 18:56:47 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA04916; Wed, 13 Dec 1995 18:56:21 +0000 Date: Wed, 13 Dec 1995 18:56:21 +0000 From: Tim Bunce Message-Id: <9512131856.AA04916@toad.ig.co.uk> To: prand@paul.spu.edu Subject: Re: New directions for DBI - Important Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 355 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Wed, 13 Dec 95 18:57:04 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Wed, 13 Dec 95 18:57:43 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Wed, 13 Dec 95 23:03:14 GMT] > From: Phil Rand > > As I recall, ODBC defines two or three different implementation levels, > from simple to complex. Would you expect to follow those? Or are the > easier ones too simple to be worth doing by themselves? Initially it'll be the API defined by the standard. Other M$ specifics can be added easily later. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <13875-3@oink>; Thu, 14 Dec 1995 12:22:09 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818896125:26704:1; Wed, 13 Dec 95 23:08:45 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa26395; 13 Dec 95 23:08 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA04718 for dbi-users-real; Wed, 13 Dec 1995 11:18:30 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA04708 for ; Wed, 13 Dec 1995 11:18:07 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ai07454; 13 Dec 95 19:06 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa26186; 13 Dec 95 19:05 GMT Received: from toad by oink with SMTP (PP) id <10322-0@oink>; Wed, 13 Dec 1995 19:05:32 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA04939; Wed, 13 Dec 1995 19:05:05 +0000 Date: Wed, 13 Dec 1995 19:05:05 +0000 From: Tim Bunce Message-Id: <9512131905.AA04939@toad.ig.co.uk> To: dbi-users@fugue.com, afc@mmts.eds.com Subject: Re: can DBI return rows as hashes keyed by column names? X-Sun-Charset: US-ASCII content-length: 448 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Wed, 13 Dec 95 19:05:50 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Wed, 13 Dec 95 19:06:16 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Wed, 13 Dec 95 23:08:12 GMT] > From: Andy Cohan > > I'm trying to get our own perl5 DB class to use DBI for an Oracle back end > and I would like to support the Perl5 Sybase facility of optionally > returning rows as associative arrays keyed by the result column names (we > have used it extensively). > > Does DBI and DBD::Oracle support this? If not, is it planned? @fieldnames = (...); while(@hash{@fieldnames} = $sth->fetchrow) { ... } Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13872-33@oink>; Thu, 14 Dec 1995 12:24:27 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818908932:25699:1; Thu, 14 Dec 95 02:42:12 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa25583; 14 Dec 95 2:41 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA00788 for dbi-users-real; Wed, 13 Dec 1995 16:47:28 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from CHILD.PRODIGY.CS.CMU.EDU (CHILD.PRODIGY.CS.CMU.EDU [128.2.242.220]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA00784 for ; Wed, 13 Dec 1995 16:47:26 -0800 Received: from [127.0.0.1] by CHILD.PRODIGY.CS.CMU.EDU id aa27486; 13 Dec 95 19:47 EST Reply-To: desilva+@cmu.edu To: dbi-users@fugue.com Subject: Oraperl: Updating LONG RAWs Date: Wed, 13 Dec 1995 19:47:13 -0500 Message-ID: <27484.818902033@CHILD.PRODIGY.CS.CMU.EDU> From: Rujith de Silva X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 2:41:53 GMT] -----BEGIN PGP SIGNED MESSAGE----- How does one insert or update LONG RAWs using Oraperl? I've found a way to do it by converting the data into a hex string, but that only works if the hex string doesn't come out longer than 2000 characters. There's got to be a better way. What I've found so far: $ora_long = 500000; # Dunno if this has any effect on UPDATEs. if ($types[0] == 24) # LONG RAW data { my $newdata = ""; foreach (unpack ('C*', $data)) # Convert it to hex format. { $newdata .= sprintf ("%2.2x", $_); # Ain't Perl great! } $data = $newdata; } my $stmt = "UPDATE $table_name SET $field_name = :1"; $stmt .= " WHERE $where_clause" if $where_clause; $csr = ora_open ($lda, $stmt); ora_bind ($csr, $data); ora_close ($csr); All information appreciated. Later, Rujith. Rujith de Silva. 1-(412) 268-3620. desilva+@cmu.edu http://www.cs.cmu.edu/~desilva/ PGP public key available by finger. -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCUAwUBMM90CokharavxuYRAQFytwP1GPjI51wW0FctqMzUfhj6KUjMDYoelQR3 Bt61+JWB4GDh94eFkhF6xLAJ7UQ3nMWx1+WOSKRh1k94tZo32xIpOI8GjQN+1+kz Qdu1///+qAbDOg3e5PO3HZ5U3F4+uc8EhJJmFvNIDRwIDOa6ZULZUxv/Ucu6Z/wu Uuov+sS3GQ== =cb5e -----END PGP SIGNATURE----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13872-39@oink>; Thu, 14 Dec 1995 12:24:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818911028:07751:0; Thu, 14 Dec 95 03:17:08 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa07273; 14 Dec 95 3:16 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA00932 for dbi-users-real; Wed, 13 Dec 1995 17:59:33 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from metronet.com (root@mail.metronet.com [192.245.137.6]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA00928 for ; Wed, 13 Dec 1995 17:59:30 -0800 Received: from hometown.com by metronet.com with SMTP id AA19589 (5.67a/IDA1.5hp for ); Wed, 13 Dec 1995 19:59:16 -0600 Message-Id: <199512140159.AA19589@metronet.com> X-Sender: mouche@mail.metronet.com X-Mailer: Windows Eudora Version 2.1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 13 Dec 1995 19:55:17 -0600 To: dbi-users@fugue.com From: Steve Livingston Subject: Re: Building DBD-0.27 on Solaris/Sparc Cc: Tim.Bunce@ig.co.uk X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 3:16:09 GMT] Well...I've started from scratch: Fresh install of Oracle 7 beta (including Pro*C) Installed SUNWspro compiler (no longer using gcc) Fresh install of perl5.001n with cc Re-built DBI-0.65 When installing DBD-0.27: make realclean perl Makefile.PL make (?) The make, shown below, gives up on the same ld statement as before. I've had folks with more Unix skills than I look at: Makefile proc.mk env_precomp.mk ...and everything else They say that this ld statement is odd, and perhaps it should be a cc...but to ask you folks. Ideas? Has anyone installed DBD-0.27 with O7beta? Cheers, Steve -------------- elcaro% make umask 0 && cp Oraperl.pm ../../lib/Oraperl.pm umask 0 && cp oraperl.ph ../../lib/oraperl.ph umask 0 && cp Oracle.pm ../../lib/DBD/Oracle.pm ../../perl -I../../lib -I../../lib ../../lib/ExtUtils/xsubpp -typemap ../../lib/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c ld -dy /private/slivings/oracle/orahome/lib/crti.o /private/slivings/oracle/orahome/lib/crt1.o /private/slivings/oracle/orahome/lib/__fstd.o -c -c Oracle.c -o Oracle.o ld: illegal option -- c ld: illegal option -- c usage: ld [-abd:e:h:il:mo:rstu:z:B:D:F:GI:L:M:Q:R:S:VY:] file(s) ... [-a] create an absolute file [-b] do not do special PIC relocations in a.out [-d y|n] operate in dynamic|static mode ... make: *** [Oracle.o] Error 1 --------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <14113-3@oink>; Thu, 14 Dec 1995 12:28:09 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818922307:09326:0; Thu, 14 Dec 95 06:25:07 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa09193; 14 Dec 95 6:24 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA01379 for dbi-users-real; Wed, 13 Dec 1995 20:42:13 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA01375 for ; Wed, 13 Dec 1995 20:42:11 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id WAA54463; Wed, 13 Dec 1995 22:42:13 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for prand@paul.spu.edu at austin.ibm.com; id AA23918; Wed, 13 Dec 1995 22:42:10 -0600 Message-Id: <9512140442.AA23918@dax.austin.ibm.com> To: Tim Bunce Cc: prand@paul.spu.edu, dbi-users@fugue.com Subject: Re: New directions for DBI - Important In-Reply-To: Your message of Wed, 13 Dec 95 18:56:21 +0000. <9512131856.AA04916@toad.ig.co.uk> Date: Wed, 13 Dec 95 22:42:10 -0600 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 6:24:38 GMT] Tim Bunce writes: |>> From: Phil Rand |>> |>> As I recall, ODBC defines two or three different implementation levels, |>> from simple to complex. Would you expect to follow those? Or are the |>> easier ones too simple to be worth doing by themselves? |> |>Initially it'll be the API defined by the standard. Other M$ specifics |>can be added easily later. |> I would suggest mimicking the SQL CLI standard which has a Core level, Level 1, and Level 2 compliance. This would give a nice level of flexibility to the end user, as he would only have to "use" the level sufficient to get his desired functionality. Mike mhm@austin.ibm.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <13872-63@oink>; Thu, 14 Dec 1995 12:29:06 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818934618:29896:4; Thu, 14 Dec 95 09:50:18 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa29722; 14 Dec 95 9:49 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id XAA01724 for dbi-users-real; Wed, 13 Dec 1995 23:36:36 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from auspex-gw.auspex.com (auspex-gw.auspex.com [144.48.1.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id XAA01720 for ; Wed, 13 Dec 1995 23:36:34 -0800 Received: from auspex.Auspex.com (auspex-e6.auspex.com [144.48.8.10]) by auspex-gw.auspex.com (8.6.12/8.6.12) with SMTP id XAA06043; Wed, 13 Dec 1995 23:36:40 -0800 Received: from alpha1 by auspex.Auspex.com (4.1/SMI-4.1) id AA15221; Wed, 13 Dec 95 23:36:39 PST Date: Wed, 13 Dec 95 23:36:39 PST From: kstock@auspex.com (Kevin Stock) Message-Id: <9512140736.AA15221@auspex.Auspex.com> Reply-To: kstock@auspex.com To: dbi-users@fugue.com Subject: Re: Oraperl: Updating LONG RAWs Cc: Rujith_S_DeSilva@CHILD.PRODIGY.CS.CMU.EDU X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 9:49:45 GMT] > How does one insert or update LONG RAWs using Oraperl? I've found a way to > do it by converting the data into a hex string, but that only works if the > hex string doesn't come out longer than 2000 characters. There's got to be > a better way. I hate to admit it, but the 'better way' is to fix Oraperl. As I never used RAW data, I didn't test Oraperl with it, and just assumed (that terrible word) that the conversion to string would work. Someone did send me some patches to make Oraperl work with RAW data, but as I don't have access to Oracle any more, I'm not in a position to make a new release. If someone wants to develop and release patches for this, I'd be more than happy to offer any help I can. Kevin P.S.: > $ora_long = 500000; # Dunno if this has any effect on UPDATEs. It hasn't. $ora_long is only used to allocate a buffer for fetching data. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <18521-18@oink>; Thu, 14 Dec 1995 19:31:08 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818965774:22792:0; Thu, 14 Dec 95 18:29:34 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa22303; 14 Dec 95 18:28 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id msg.aa07905; 14 Dec 95 18:00 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA03559 for dbi-users-real; Thu, 14 Dec 1995 06:42:18 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA03555 for ; Thu, 14 Dec 1995 06:42:12 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.aa13542; 14 Dec 95 14:38 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ad05647; 14 Dec 95 14:37 GMT Received: from toad by oink with SMTP (PP) id <16030-0@oink>; Thu, 14 Dec 1995 14:23:56 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA08814; Thu, 14 Dec 1995 14:23:29 +0000 Date: Thu, 14 Dec 1995 14:23:29 +0000 From: Tim Bunce Message-Id: <9512141423.AA08814@toad.ig.co.uk> To: dbi-users@fugue.com, desilva+@cmu.edu Subject: Re: Oraperl: Updating LONG RAWs X-Sun-Charset: US-ASCII content-length: 649 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Thu, 14 Dec 95 14:37:47 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Thu, 14 Dec 95 14:38:26 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 18:00:47 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Thu, 14 Dec 95 18:28:32 GMT] > From: Rujith de Silva > > -----BEGIN PGP SIGNED MESSAGE----- > > How does one insert or update LONG RAWs using Oraperl? I've found a way to > do it by converting the data into a hex string, but that only works if the > hex string doesn't come out longer than 2000 characters. There's got to be > a better way. Recent versions of DBD::Oracle may help. > foreach (unpack ('C*', $data)) # Convert it to hex format. > { > $newdata .= sprintf ("%2.2x", $_); # Ain't Perl great! > } Using push and a later join or a s/(.)/sprintf("%2.2x",$1)/e would probably be faster here. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19660-0@oink>; Thu, 14 Dec 1995 20:12:22 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818969771:25780:2; Thu, 14 Dec 95 19:36:11 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa25683; 14 Dec 95 19:35 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id sg.aa12362; 14 Dec 95 19:02 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA03592 for dbi-users-real; Thu, 14 Dec 1995 07:03:01 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA03588 for ; Thu, 14 Dec 1995 07:02:41 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id g.af13523; 14 Dec 95 14:38 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ac05647; 14 Dec 95 14:37 GMT Received: from toad by oink with SMTP (PP) id <16021-0@oink>; Thu, 14 Dec 1995 14:21:23 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA08808; Thu, 14 Dec 1995 14:20:56 +0000 Date: Thu, 14 Dec 1995 14:20:56 +0000 From: Tim Bunce Message-Id: <9512141420.AA08808@toad.ig.co.uk> To: dbi-users@fugue.com, mouche@hometown.com Subject: Re: Building DBD-0.27 on Solaris/Sparc X-Sun-Charset: US-ASCII content-length: 415 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Thu, 14 Dec 95 14:37:45 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Thu, 14 Dec 95 14:38:28 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 19:02:20 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Thu, 14 Dec 95 19:35:49 GMT] > From: Steve Livingston > > The make, shown below, gives up on the same ld statement as before. > > They say that this ld statement is odd, and perhaps it should be a cc...but > to ask you folks. > > Ideas? Has anyone installed DBD-0.27 with O7beta? I'd guess that the very early proc.mk is causing the problem. Perhaps you'll need to upgrade - the beta had many other problems. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <20271-3@oink>; Thu, 14 Dec 1995 20:40:05 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818973076:22584:0; Thu, 14 Dec 95 20:31:16 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa22199; 14 Dec 95 20:30 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA04035 for dbi-dev-real; Thu, 14 Dec 1995 10:16:12 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from emerald.empress.com (emerald.empress.com [192.43.219.6]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA04031; Thu, 14 Dec 1995 10:16:07 -0800 Received: (from fjones@localhost) by emerald.empress.com (8.6.11/8.6.9) id NAA08386; Thu, 14 Dec 1995 13:17:40 -0500 From: Francis Jones Message-Id: <199512141817.NAA08386@emerald.empress.com> Subject: Re: New directions for DBI - Important To: dbi-dev@fugue.com Date: Thu, 14 Dec 1995 13:17:38 -0500 (EST) Cc: dvi-users@fugue.com In-Reply-To: <9512140442.AA23918@dax.austin.ibm.com> from "mhm@austin.ibm.com" at Dec 13, 95 10:42:10 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 3650 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 20:30:36 GMT] Hello all, here are my thoughts on the proposed ODBC project: ODBC offers "core", "level 1" and "level 2" compliance levels, and these offer increasingly more complex functionality. As far as a perl interface is concerned, my feeling is that if you are going to call the new interface 'ODBC', then you should offer complete packages at successive levels. i.e., you can offer the core, core + level 1, etc. packages. These can be developed over time, with 'core' coming first. I don't think it is reasonable to detract from or augment the standard set of "ODBC" calls. After all, they are a well-defined standard which many people are becoming proficient with, and which (unlike the current perl DB interface) offers a well-defined response even to 'unsupported' features of any particular DB engine. So, for this part of the work, it looks pretty straight-forward. Just offer _all_ the ODBC calls in whichever levels it is desired to support. There is no further work needed to support any particular DB vendor, since they will have already created their own ODBC drivers which handle all the various ODBC calls in whatever fashion is appropriate for them. But, this doesn't mean that we shouldn't _also_ offer some sort of composite functionality. This is more difficult, since it involves trying to establish consensus amongst the perl user community. But, we could start with the perl DB interface document which seems to have been floating around for some time now, and then hash it out from there. The underlying code would contain ODBC calls to do the low level DB operations. The technical difficulty with this is that, even though a given DB vendor may be listed as 'Level 1' compliant, and even if the level of compliance can be determined by one of these composite routines, the problem is that lots of ODBC calls have lots of options. These options refine the requests for information, actions, etc., and may not be possible with one DB vs. another DB. However, in all cases the DB driver must respond with a message indicating things like "Operations unsupported", "Operation failed", "Operation succeeded", "Operation succeeded but with parameters other than those supplied; look in the standard place for further information", ... and so on. All of these responses are fixed in the ODBC standard. So, while the ODBC calls themselves are straight-forward to implement, and should be a universal set of functions, the 'composite' functionality akin to the existing perl DB interface might be difficult to develop if a too-ambitious set of composite functions were attempted. It would be highly desirable to have, for each 'composite' DB function, a routine which called ODBC underneath, and which was aware of the different characteristics of ODBC implementations by different vendors, and which handled all the exceptional responses from the ODBC servers in a sane manner. I do _not_ advocate having a different routine (e.g., to do a high-level 'fetch') tuned for each DB vendor, since this will probably not lead to the universality that ODBC is aimed at. In my opinion, the current DB interface suffers somewhat from the fact that the interfaces for the various DB vendors currently on offer have different sets of interface routines which are maintained in different source trees. These seem to have deviated from the published standard somewhat (either by omission, addition, or quirks in DB vendor functionality), so that perl programs written for one DB engine can't be guaranteed to work with another. This needs to be considered carefully. So, there's _my_ $0.02 worth. Cheers, Francis Jones ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21186-10@oink>; Thu, 14 Dec 1995 23:40:49 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 818976242:18208:0; Thu, 14 Dec 95 21:24:02 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa17779; 14 Dec 95 21:23 GMT Received: from toccata.fugue.com by relay-1.mail.demon.net id g.aa22886; 14 Dec 95 21:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA04100 for dbi-users-real; Thu, 14 Dec 1995 10:48:11 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from emerald.empress.com (emerald.empress.com [192.43.219.6]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA04096 for ; Thu, 14 Dec 1995 10:48:06 -0800 Received: (from fjones@localhost) by emerald.empress.com (8.6.11/8.6.9) id NAA08459 for dbi-users@fugue.com; Thu, 14 Dec 1995 13:49:44 -0500 From: Francis Jones Message-Id: <199512141849.NAA08459@emerald.empress.com> Subject: Re: New directions for DBI - Important To: dbi-users@fugue.com Date: Thu, 14 Dec 1995 13:49:43 -0500 (EST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 3696 X-SMTP-Posting-Host: toccata.fugue.com [Thu, 14 Dec 95 21:18:36 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Thu, 14 Dec 95 21:23:06 GMT] Hello all, here are my thoughts on the proposed ODBC project: ODBC offers "core", "level 1" and "level 2" compliance levels, and these offer increasingly more complex functionality. As far as a perl interface is concerned, my feeling is that if you are going to call the new interface 'ODBC', then you should offer complete packages at successive levels. i.e., you can offer the core, core + level 1, etc. packages. These can be developed over time, with 'core' coming first. I don't think it is reasonable to detract from or augment the standard set of "ODBC" calls. After all, they are a well-defined standard which many people are becoming proficient with, and which (unlike the current perl DB interface) offers a well-defined response even to 'unsupported' features of any particular DB engine. So, for this part of the work, it looks pretty straight-forward. Just offer _all_ the ODBC calls in whichever levels it is desired to support. There is no further work needed to support any particular DB vendor, since they will have already created their own ODBC drivers which handle all the various ODBC calls in whatever fashion is appropriate for them. But, this doesn't mean that we shouldn't _also_ offer some sort of composite functionality. This is more difficult, since it involves trying to establish consensus amongst the perl user community. But, we could start with the perl DB interface document which seems to have been floating around for some time now, and then hash it out from there. The underlying code would contain ODBC calls to do the low level DB operations. The technical difficulty with this is that, even though a given DB vendor may be listed as 'Level 1' compliant, and even if the level of compliance can be determined by one of these composite routines, the problem is that lots of ODBC calls have lots of options. These options refine the requests for information, actions, etc., and may not be possible with one DB vs. another DB. However, in all cases the DB driver must respond with a message indicating things like "Operations unsupported", "Operation failed", "Operation succeeded", "Operation succeeded but with parameters other than those supplied; look in the standard place for further information", ... and so on. All of these responses are fixed in the ODBC standard. So, while the ODBC calls themselves are straight-forward to implement, and should be a universal set of functions, the 'composite' functionality akin to the existing perl DB interface might be difficult to develop if a too-ambitious set of composite functions were attempted. It would be highly desirable to have, for each 'composite' DB function, a routine which called ODBC underneath, and which was aware of the different characteristics of ODBC implementations by different vendors, and which handled all the exceptional responses from the ODBC servers in a sane manner. I do _not_ advocate having a different routine (e.g., to do a high-level 'fetch') tuned for each DB vendor, since this will probably not lead to the universality that ODBC is aimed at. In my opinion, the current DB interface suffers somewhat from the fact that the interfaces for the various DB vendors currently on offer have different sets of interface routines which are maintained in different source trees. These seem to have deviated from the published standard somewhat (either by omission, addition, or quirks in DB vendor functionality), so that perl programs written for one DB engine can't be guaranteed to work with another. This needs to be considered carefully. So, there's _my_ $0.02 worth. Cheers, Francis Jones --NAA08388.818965121/emerald.empress.com-- ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <21185-17@oink>; Thu, 14 Dec 1995 23:41:37 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 818983537:10165:0; Thu, 14 Dec 95 23:25:37 GMT Received: from frigate.doc.ic.ac.uk by punt-2.mail.demon.net id ai09447; 14 Dec 95 23:24 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Thu, 14 Dec 1995 22:29:08 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA04394 for dbi-users-real; Thu, 14 Dec 1995 12:13:00 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail.crl.com (mail.crl.com [165.113.1.22]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id MAA04389 for ; Thu, 14 Dec 1995 12:12:58 -0800 Received: from [205.149.164.200] (rmdudley.vip.best.com) by mail.crl.com with SMTP id AA00589 (5.65c/IDA-1.5 for ); Thu, 14 Dec 1995 12:10:42 -0800 Date: Thu, 14 Dec 1995 12:10:42 -0800 X-Sender: mrlizard@mail.crl.com Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: dbi-users@fugue.com From: mrlizard@crl.com (Brian Valente) Subject: DBD core dumping Cc: descarte@hermetica.com X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Thu, 14 Dec 95 23:24:45 GMT] If there's still anyone out there who can help- I've run into unfortunate problems with DBI 0.65, DBD mSQL 0.18, and perl5.001n. Here's the setup ---Machine 1--- ---Machine 2--- SGI/IRIX 5.3 FreeBSD 2.2 msqld running DBD/DBI->msql installed fine on machine web server running here Here's the situation: Machine #1 - msqld running fine - can use msql monitor to access server no problem. DBD/DBI->msql libraries installed as test, and test script runs fine (both mine and the supplied one). Machine #2 - can use msql monitor to connect to machine #1 fine. Everything works okay. Ran test script for DBD/DBI->msql installation - everything's fine. However, when I run my own test script (same one that works fine on machine #1), I get a perl5.001 core dump: [ output from error checking code, so I know driver is installed okay ] $drh after install driver is DBI::dr=HASH(0xa6bdc) [my own error output ] Sorry, an Error Occured For some reason, your request failed. The error code reported was: Couldn't connect to db ctiregistry DBD internal error codes are: IOT trap (core dumped) Here is a code fragment from my script: $drh = DBI->install_driver($driver); &HandleError("Couldn't install driver $driver") unless $drh; print $OUTPUT "\$drh after install driver is $drh\n" if $DEBUG; # DBI->internal->{DebugDispatch} = 0; (commented out the debug stuff) # DBI->internal->debug(2); DEATH $dbh = $drh->connect( $hostname, $db ); &HandleError("Couldn't connect to db $db") unless $dbh; print $OUTPUT "\$dbh after connect is $dbh\n" if $DEBUG; I am running version 1.0.10 of msqld Version 0.65 of DBD Version p.018 of mSQL driver. If there is a god, please help me! I have the core dump if someone needs it. Thanks, Brian Valente =========================================================================== Brian Valente | brian_valente@rmdudley.com R.M. Dudley Corporation | mrlizard@crl.com 1799 Old Bayshore Suite 200 | http://www.rmdudley.com Burlingame, CA 94010 | 415-697-1650 tel. 415-697-1816 fax. =========================================================================== ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <23887-28@oink>; Fri, 15 Dec 1995 07:09:40 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819002690:16438:0; Fri, 15 Dec 95 04:44:50 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa16102; 15 Dec 95 4:43 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id TAA06341 for dbi-users-real; Thu, 14 Dec 1995 19:00:50 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from grctechs.va.grci.com (grctechs.va.grci.com [192.246.37.10]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id TAA06337 for ; Thu, 14 Dec 1995 19:00:42 -0800 Received: by grctechs.va.grci.com (5.65/25-eef) id AA26807; Thu, 14 Dec 95 21:37:29 -0500 My-Date: Thu, 14 Dec 1995 21:37:28 -0500 (EST) My-From: Greg Gerber X-Full-Name: Greg Gerber Date: Thu, 14 Dec 1995 21:37:28 -0500 (EST) From: Greg Gerber To: Kevin Stock Cc: dbi-users@fugue.com, Rujith_S_DeSilva@CHILD.PRODIGY.CS.CMU.EDU Subject: Re: Oraperl: Updating LONG RAWs In-Reply-To: <9512140736.AA15221@auspex.Auspex.com> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: toccata.fugue.com [Fri, 15 Dec 95 4:44:26 GMT] On Wed, 13 Dec 1995, Kevin Stock wrote: > > How does one insert or update LONG RAWs using Oraperl? I've found a way to > > do it by converting the data into a hex string, but that only works if the > > hex string doesn't come out longer than 2000 characters. There's got to be > > a better way. > > I hate to admit it, but the 'better way' is to fix Oraperl. As I never > used RAW data, I didn't test Oraperl with it, and just assumed (that > terrible word) that the conversion to string would work. Someone did > send me some patches to make Oraperl work with RAW data, but as I don't > have access to Oracle any more, I'm not in a position to make a new > release. > > If someone wants to develop and release patches for this, I'd be more > than happy to offer any help I can. > Don't know if this is what Kevin is refering to, but I did send him patches for long raws several weeks ago, that version had a small bug and I have a new version. I'll release it (here and to comp.lang.perl I guess) as soon as I have a chance to wrap it up. All I need is just a little more time ... :( ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <29897-6@oink>; Fri, 15 Dec 1995 19:33:52 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819053345:26546:3; Fri, 15 Dec 95 18:49:05 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa26353; 15 Dec 95 18:48 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA09320 for dbi-users-real; Fri, 15 Dec 1995 09:40:39 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA09316 for ; Fri, 15 Dec 1995 09:40:08 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.aj13293; 15 Dec 95 17:03 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id msg.ab00102; 15 Dec 95 17:01 GMT Received: from toad by oink with SMTP (PP) id <27423-0@oink>; Fri, 15 Dec 1995 14:45:12 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA17786; Fri, 15 Dec 1995 14:44:41 +0000 Date: Fri, 15 Dec 1995 14:44:41 +0000 From: Tim Bunce Message-Id: <9512151444.AA17786@toad.ig.co.uk> To: kstock@auspex.com, ggerber@grctechs.va.grci.com Subject: Re: Oraperl: Updating LONG RAWs Cc: dbi-users@fugue.com, Rujith_S_DeSilva@child.prodigy.cs.cmu.edu X-Sun-Charset: US-ASCII content-length: 1329 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Fri, 15 Dec 95 17:01:18 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Fri, 15 Dec 95 17:03:18 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Fri, 15 Dec 95 18:48:36 GMT] > From: Greg Gerber > > On Wed, 13 Dec 1995, Kevin Stock wrote: > > > > How does one insert or update LONG RAWs using Oraperl? I've found a way to > > > do it by converting the data into a hex string, but that only works if the > > > hex string doesn't come out longer than 2000 characters. There's got to be > > > a better way. > > > > I hate to admit it, but the 'better way' is to fix Oraperl. As I never > > used RAW data, I didn't test Oraperl with it, and just assumed (that > > terrible word) that the conversion to string would work. Someone did > > send me some patches to make Oraperl work with RAW data, but as I don't > > have access to Oracle any more, I'm not in a position to make a new > > release. > > > > If someone wants to develop and release patches for this, I'd be more > > than happy to offer any help I can. > > > > Don't know if this is what Kevin is refering to, but I did send him patches > for long raws several weeks ago, that version had a small bug and I have > a new version. I'll release it (here and to comp.lang.perl I guess) as > soon as I have a chance to wrap it up. All I need is just a little more > time ... :( You could send it to me and I'll add it to the oraperl archive on ftp.demon.co.uk (which is back now and soon to be updated). Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <02860-2@oink>; Sat, 16 Dec 1995 01:21:42 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819072650:10255:0; Sat, 16 Dec 95 00:10:50 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa09906; 16 Dec 95 0:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA10164 for dbi-users-real; Fri, 15 Dec 1995 14:25:56 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from monge.brunel.ac.uk (pp@monge.brunel.ac.uk [134.83.72.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id OAA10160 for ; Fri, 15 Dec 1995 14:25:51 -0800 From: cs92bbm@brunel.ac.uk Received: from ccsp-28.brunel.ac.uk by monge.brunel.ac.uk with SMTP (PP) id <17769-0@monge.brunel.ac.uk>; Fri, 15 Dec 1995 22:25:29 +0000 Message-Id: <8496.9512152225@ccsp-28.brunel.ac.uk> Subject: mSQL-1.0.10, DBI-0.65, DBD-0.60pl8 To: dbi-users@fugue.com Date: Fri, 15 Dec 1995 22:25:27 +0000 (GMT) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: toccata.fugue.com [Sat, 16 Dec 95 0:10:13 GMT] Hello... I am getting a number of spurrious errors, but they are not stopping fuctionality... presumaeably I can simply ignore them? disconnect(DBI::db=HASH(0x314a00)) invalidates 1 associated cursor(s) at ... Statement handle DBI::st=HASH(0x314b8c) destroyed without finish() at ... Any ideas? Ben [ Ben Middleton cs92bbm@brunel.ac.uk ] [ Interact with global communication at ] [ http://www.brunel.ac.uk:8080/~cs92bbm/gc.html ] ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03299-0@oink>; Sat, 16 Dec 1995 01:41:15 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819077391:15198:0; Sat, 16 Dec 95 01:29:51 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa14886; 16 Dec 95 1:29 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA10308 for dbi-users-real; Fri, 15 Dec 1995 15:46:49 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from lancelot.st.nepean.uws.edu.au (lancelot.st.nepean.uws.EDU.AU [137.154.148.30]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA10302 for ; Fri, 15 Dec 1995 15:46:29 -0800 Received: from arthur.st.nepean.uws.edu.au (arthur) by lancelot.st.nepean.uws.edu.au with SMTP id AA25569 (5.65c/IDA-1.5 for ); Sat, 16 Dec 1995 10:45:52 +1100 Received: by arthur.st.nepean.uws.edu.au id AA20240 (5.65c/IDA-1.4.4); Sat, 16 Dec 1995 10:45:29 +1100 Date: Sat, 16 Dec 1995 10:45:29 +1100 (EST) From: David Begley X-Sender: dbegley@arthur To: cs92bbm@brunel.ac.uk Cc: dbi-users@fugue.com Subject: Re: mSQL-1.0.10, DBI-0.65, DBD-0.60pl8 In-Reply-To: <8496.9512152225@ccsp-28.brunel.ac.uk> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: toccata.fugue.com [Sat, 16 Dec 95 1:29:02 GMT] On Fri, 15 Dec 1995 cs92bbm@brunel.ac.uk wrote: > I am getting a number of spurrious errors, but they are not stopping > fuctionality... presumaeably I can simply ignore them? Similarly, I'm getting "Bad free ignored..." whenever &ora_logoff is called in the Oraperl emulation module; ideas? dave ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <03299-1@oink>; Sat, 16 Dec 1995 01:41:18 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819077394:15202:1; Sat, 16 Dec 95 01:29:54 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa15012; 16 Dec 95 1:29 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA10419 for dbi-users-real; Fri, 15 Dec 1995 16:14:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA10415 for ; Fri, 15 Dec 1995 16:14:38 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id AAA06420; Sat, 16 Dec 1995 00:08:40 GMT From: Alligator Descartes Message-Id: <199512160008.AAA06420@fruitbat.mcqueen.com> Subject: Re: mSQL-1.0.10, DBI-0.65, DBD-0.60pl8 To: dbegley@st.nepean.uws.edu.au (David Begley) Date: Sat, 16 Dec 1995 00:08:39 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: from "David Begley" at Dec 16, 95 10:45:29 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 831 X-SMTP-Posting-Host: toccata.fugue.com [Sat, 16 Dec 95 1:29:13 GMT] > > I am getting a number of spurrious errors, but they are not stopping > > fuctionality... presumaeably I can simply ignore them? > > Similarly, I'm getting "Bad free ignored..." whenever &ora_logoff is > called in the Oraperl emulation module; ideas? The first problem of 'disconnect' gibberish is ignorable. If you do: undef ; after 'finish'ing each cursor, it'll stop the problem. It won't stop the problem on 'do' statements, though....... The 'bad free' thing is a different issue entirely and doesn't relate to the DBD::mSQL module at all. It relates to DBI. I think there was some reasoning about it wrt some issue with perl itself, but I don't recall the details. > dave -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04305-0@oink>; Sat, 16 Dec 1995 02:59:16 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819082604:19541:0; Sat, 16 Dec 95 02:56:44 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa19304; 16 Dec 95 2:55 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA10833 for dbi-users-real; Fri, 15 Dec 1995 17:45:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA10829 for ; Fri, 15 Dec 1995 17:45:12 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id sg.ac19228; 16 Dec 95 1:44 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.aa26395; 16 Dec 95 1:44 GMT Received: from toad by oink with SMTP (PP) id <03387-0@oink>; Sat, 16 Dec 1995 01:43:41 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA20823; Sat, 16 Dec 1995 01:43:08 +0000 Date: Sat, 16 Dec 1995 01:43:08 +0000 From: Tim Bunce Message-Id: <9512160143.AA20823@toad.ig.co.uk> To: cs92bbm@brunel.ac.uk, dbegley@st.nepean.uws.edu.au Subject: Re: mSQL-1.0.10, DBI-0.65, DBD-0.60pl8 Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 444 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Sat, 16 Dec 95 1:44:07 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Sat, 16 Dec 95 1:44:52 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Sat, 16 Dec 95 2:55:57 GMT] > From: David Begley > > On Fri, 15 Dec 1995 cs92bbm@brunel.ac.uk wrote: > > > I am getting a number of spurrious errors, but they are not stopping > > fuctionality... presumaeably I can simply ignore them? > > Similarly, I'm getting "Bad free ignored..." whenever &ora_logoff is > called in the Oraperl emulation module; ideas? Complain to Oracle. I've not yet found one that wasn't Oracle's fault. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04305-2@oink>; Sat, 16 Dec 1995 02:59:24 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819082609:19511:2; Sat, 16 Dec 95 02:56:49 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa19427; 16 Dec 95 2:56 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA10811 for dbi-users-real; Fri, 15 Dec 1995 17:43:25 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA10807 for ; Fri, 15 Dec 1995 17:43:19 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.aa19023; 16 Dec 95 1:43 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.ag25977; 16 Dec 95 1:41 GMT Received: from toad by oink with SMTP (PP) id <03285-0@oink>; Sat, 16 Dec 1995 01:40:45 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA20802; Sat, 16 Dec 1995 01:40:19 +0000 Date: Sat, 16 Dec 1995 01:40:19 +0000 From: Tim Bunce Message-Id: <9512160140.AA20802@toad.ig.co.uk> To: dbi-users@fugue.com, cs92bbm@brunel.ac.uk Subject: Re: mSQL-1.0.10, DBI-0.65, DBD-0.60pl8 X-Sun-Charset: US-ASCII content-length: 686 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Sat, 16 Dec 95 1:41:44 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Sat, 16 Dec 95 1:43:06 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Sat, 16 Dec 95 2:56:30 GMT] > From: cs92bbm@brunel.ac.uk > > Hello... > > I am getting a number of spurrious errors, but they are not stopping > fuctionality... presumaeably I can simply ignore them? > > disconnect(DBI::db=HASH(0x314a00)) invalidates 1 associated cursor(s) at ... > Statement handle DBI::st=HASH(0x314b8c) destroyed without finish() at ... Also, they're not really spurious. It's saying "You are disconnecting from a database on which you have an active cursor. That's probably an error (or lazyness) so I'm going to warn you about it". Here's a workaround I forgot to give in the other message: You could explicitly destroy the $sth before disconnecting. undef $sth; would work fine. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04510-1@oink>; Sat, 16 Dec 1995 03:04:52 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819082748:20289:1; Sat, 16 Dec 95 02:59:08 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa20091; 16 Dec 95 2:58 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA10817 for dbi-users-real; Fri, 15 Dec 1995 17:43:58 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id RAA10813 for ; Fri, 15 Dec 1995 17:43:54 -0800 Received: from post.demon.co.uk by relay-4.mail.demon.net id msg.aa18971; 16 Dec 95 1:43 GMT Received: from ignite.demon.co.uk by relay-3.mail.demon.net id sg.af25977; 16 Dec 95 1:41 GMT Received: from toad by oink with SMTP (PP) id <03266-0@oink>; Sat, 16 Dec 1995 01:35:54 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA20795; Sat, 16 Dec 1995 01:35:27 +0000 Date: Sat, 16 Dec 1995 01:35:27 +0000 From: Tim Bunce Message-Id: <9512160135.AA20795@toad.ig.co.uk> To: cs92bbm@brunel.ac.uk Subject: Re: mSQL-1.0.10, DBI-0.65, DBD-0.60pl8 Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 494 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 X-SMTP-Posting-Host: ignite.demon.co.uk [Sat, 16 Dec 95 1:41:42 GMT] X-SMTP-Posting-Host: post.demon.co.uk [Sat, 16 Dec 95 1:43:06 GMT] X-SMTP-Posting-Host: toccata.fugue.com [Sat, 16 Dec 95 2:58:30 GMT] > From: cs92bbm@brunel.ac.uk > > Hello... > > I am getting a number of spurrious errors, but they are not stopping > fuctionality... presumaeably I can simply ignore them? > > disconnect(DBI::db=HASH(0x314a00)) invalidates 1 associated cursor(s) at ... > Statement handle DBI::st=HASH(0x314b8c) destroyed without finish() at ... > > Any ideas? The mSQL driver hasn't been updated yet to use the new DBIc_ACTIVE_on and DBIc_ACTIVE_off macros. Complain (gently) to your 'vendor' :-) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12772-70@oink>; Mon, 18 Dec 1995 21:02:23 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819232479:08969:1; Sun, 17 Dec 95 20:34:39 GMT Received: from toccata.fugue.com by punt-1.mail.demon.net id aa08725; 17 Dec 95 20:33 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA16547 for dbi-users-real; Sun, 17 Dec 1995 11:04:37 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA16543 for ; Sun, 17 Dec 1995 11:04:34 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id SAA11254; Sun, 17 Dec 1995 18:59:31 GMT From: Alligator Descartes Message-Id: <199512171859.SAA11254@fruitbat.mcqueen.com> Subject: Re: Perl 5 & informix( DBI, DBD ) To: bsamani@iag.net (Bahram Samani) Date: Sun, 17 Dec 1995 18:59:29 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: <30D3BDB6.6222@iag.net> from "Bahram Samani" at Dec 17, 95 01:50:30 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 513 X-SMTP-Posting-Host: toccata.fugue.com [Sun, 17 Dec 95 20:33:59 GMT] > We are trying to get DBI, and the DBD working for informix, and > perl5 on HPUX, and IBM AIX 3.2.5. We have problems with the DBD, > and DBI compiling. Do you have a new version? Can you get the DBI module to compile, install and test correctly? Also, please subscribe to the DBI mailing lists if you haven't already done so. See: http://www.fugue.com/dbi for more details. Regards. > Bahram -- Alligator Descartes | PGP key available on request! descarte@hermetica.com | http://www.hermetica.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <19211-2@oink>; Tue, 19 Dec 1995 07:06:11 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819341339:02725:0; Tue, 19 Dec 95 02:48:59 GMT Received: from toccata.fugue.com by punt-2.mail.demon.net id aa02511; 19 Dec 95 2:48 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA22390 for dbi-users-real; Mon, 18 Dec 1995 13:46:47 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from metronet.com (root@mail.metronet.com [192.245.137.6]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA22386 for ; Mon, 18 Dec 1995 13:46:44 -0800 Received: from hometown.com by metronet.com with SMTP id AA29857 (5.67a/IDA1.5hp); Mon, 18 Dec 1995 15:35:39 -0600 Message-Id: <199512182135.AA29857@metronet.com> X-Sender: mouche@mail.metronet.com X-Mailer: Windows Eudora Version 2.1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 18 Dec 1995 15:31:26 -0600 To: dbi-users@fugue.com From: Steve Livingston Subject: Installing DBD-0.27 on Sparc/Solaris Cc: Tim.Bunce@ig.co.uk X-SMTP-Posting-Host: toccata.fugue.com [Tue, 19 Dec 95 2:48:13 GMT] (I have abandoned Oracle 7.3 beta and installed Oracle 7.2.2.) Installed DBI-0.65...tests OK Using: perl5.001n solaris 2.4 Shown below: (1) perl -e 'use Config; print Config::myconfig()' OK- (2) make realclean OK- (3) perl Makefile.PL OK- (4) make OK- (5) make test (failed) (6) make installed The "make install" fails with: sh: /usr/local/lib/perl5/sun4-solaris/auto/DBD/Oracle/.packlist: cannot create *** Error code 1 Prizes will be awarded for prompt response. Cheers, Steve ----------- (1) perl -e 'use Config; print Config::myconfig()' elcaro% perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=solaris, osver=2.4, archname=sun4-solaris uname='sunos elcaro 5.4 generic_101945-29 sun4m sparc ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-I/usr/local/include' ccflags ='-I/usr/local/include' ldflags =' -L/usr/local/lib' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/usr/local/lib /lib /usr/lib /usr/ccs/lib libs=-lsocket -lnsl -ldl -lm -lc -lcrypt libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-Kpic', ccdlflags=' ', lddlflags='-G -L/usr/local/lib' ----------- (2) make realclean elcaro% make realclean rm -rf Oracle.c ./blib Makeaperlfile ../../lib/auto/DBD/Oracle/extralibs.all perlmain.c mon.out core so_locations *~ */*~ */*/*~ *.o *.a perl.exe Oracle.bs Oracle.bso Oracle.def Oracle.exp mv Makefile Makefile.old 2>/dev/null rm -rf ../../lib/auto/DBD/Oracle ../../lib/auto/DBD/Oracle rm -f ../../lib/auto/DBD/Oracle/Oracle.so ../../lib/auto/DBD/Oracle/Oracle.bs rm -f ../../lib/auto/DBD/Oracle/Oracle.a ../../lib/Oraperl.pm ../../lib/oraperl.ph ../../lib/DBD/Oracle.pm rm -rf Makefile Makefile.old elcaro% ----------- (3) perl Makefile.PL elcaro% perl Makefile.PL Configuring DBD::Oracle version 0.27... Remember to actually read the README file! Using Oracle in /private/slivings/oracle/orahome Oracle sysliblist: -lsocket -lnsl -lm -ldl -laio -lposix4 Using /private/slivings/oracle/orahome/proc/lib/proc.mk version 1.6 System: perl5.001 sunos elcaro 5.4 generic_101945-29 sun4m sparc sun4-solaris dl_dlopen.xs Compiler: cc -O -I/usr/local/include Oracle proc.mk would have used these values but we override them: CC: ld -dy $(COMPOBJS) $(LDSTRING) LDFLAGS: -L$(LIBHOME) LDSTRING: -R /opt/SUNWcluster/lib -Y P,$(LD_LIBRARY_PATH):/opt/SUNWcluster/lib:/usr/ccs/lib:/usr/lib -Qy -lc $(COMPOBJ)/crtn.o See README notes about SPARCompiler on Solaris Checking if your kit is complete... Looks good CONFIG key 'ar' does not exist in Config.pm CONFIG key 'lib_ext' does not exist in Config.pm CONFIG key 'obj_ext' does not exist in Config.pm Warning (non-fatal): Importing of %att is deprecated [PM] use $self instead Warning (non-fatal): Importing of %att is deprecated [PM] use $self instead Warning (non-fatal): Importing of %att is deprecated [EXTRALIBS] use $self instead Warning (non-fatal): Importing of %att is deprecated [EXTRALIBS][-lsocket -lnsl -lm -ldl -laio -lposix4 -L$(LIBHOME) $(LLIBOCIC) $(TTLIBS)] use $self instead Writing Makefile for DBD::Oracle elcaro% ----------- (4) make elcaro% make umask 0 && cp Oraperl.pm ../../lib/Oraperl.pm umask 0 && cp oraperl.ph ../../lib/oraperl.ph umask 0 && cp Oracle.pm ../../lib/DBD/Oracle.pm ../../perl -I../../lib -I../../lib ../../lib/ExtUtils/xsubpp -typemap ../../lib/ExtUtils/typemap Oracle.xs >Oracle.tc && mv Oracle.tc Oracle.c cc -c -I/private/slivings/oracle/orahome/rdbms/demo -I/usr/local/lib/perl5/sun4-solaris/DBI -I/usr/local/include -O -DVERSION=\"0.27\" -Kpic -I../.. Oracle.c cc -c -I/private/slivings/oracle/orahome/rdbms/demo -I/usr/local/lib/perl5/sun4-solaris/DBI -I/usr/local/include -O -DVERSION=\"0.27\" -Kpic -I../.. dbdimp.c Running Mkbootstrap for DBD::Oracle () chmod 644 Oracle.bs LD_RUN_PATH="/lib" cc -o ../../lib/auto/DBD/Oracle/Oracle.so -G -L/usr/local/lib Oracle.o dbdimp.o -L/private/slivings/oracle/orahome/lib /private/slivings/oracle/orahome/lib/__fstd.o -locic /private/slivings/oracle/orahome/lib/osntab.o -lsqlnet -lora -lsqlnet -lora -lnlsrtl3 -lc3v6 -lcore3 -lnlsrtl3 -lcore3 `cat /private/slivings/oracle/orahome/rdbms/lib/sysliblist` -lsocket -lnsl -lm -ldl -laio -lposix4 chmod 755 ../../lib/auto/DBD/Oracle/Oracle.so cp Oracle.bs ../../lib/auto/DBD/Oracle/Oracle.bs chmod 644 ../../lib/auto/DBD/Oracle/Oracle.bs Installing ./blib/man/man3/DBD::Oraperl.3 elcaro% ----------- (5) make test elcaro% make test PERL_DL_NONLAZY=1 ../../perl -I../../lib -I../../lib -I../../lib -I../../lib -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/base..............ok t/main..............Identifier "DBI::PERL_ENDING" used only once: possible typo at ../../lib/DBI.pm line 124. Identifier "DBD::Switch::dr::imp_data_size" used only once: possible typo at ../../lib/DBI.pm line 382. ok All tests successful. Files=2, Tests=6, 1 secs ( 0.56 cusr 0.18 csys = 0.74 cpu) PERL_DL_NONLAZY=1 ../../perl -I../../lib -I../../lib -I../../lib -I../../lib test.pl Oraperl test application $Revision: 1.19 $ Oraperl Emulation Interface version 1.23 Oracle Driver 1.21 DBI Switch by Tim Bunce, version 1.48 Connecting to '' (from command line, else uses ORACLE_SID/TWO_TASK) as 'ctx1user/demo1' (via ORACLE_USERID environment var or default) Fields: 4 Names: 'NUM_T', 'DATE_T', 'CHAR_T', 'NULL_T' Lengths: '40', '75', '30', '0' Types: '2', '12', '1', '1' Data rows: fetch: 79.2 18-DEC-95 CTX1USER NULL csr reassigned (forces destruction)... Fetch list of tables: Fetched: DUAL Fetched: SYSTEM_PRIVILEGE_MAP Fetched: TABLE_PRIVILEGE_MAP Fetched: STMT_AUDIT_OPTION_MAP Test ora_do with harmless non-select statement (set transaction read only) Expect error message: ora_do: 1453: ORA-01453: SET TRANSACTION must be first statement of transaction (DBD: oexec error) csr out of scope... ora_logoff... lda out of scope... Testing repetitive connect/open/close/disconnect: Expect sequence of digits, no other messages: 1 2 3 4 5 Test interaction of explicit close/logoff and implicit DESTROYs Expect just 'done.', no other messages: done. Test complete (2 seconds). elcaro% ----------- (6) make install elcaro% make install make INST_LIB=/usr/local/lib/perl5 INST_ARCHLIB=/usr/local/lib/perl5/sun4-solaris INST_EXE=/usr/local/bin INST_MAN1DIR=/usr/local/man/man1 INST_MAN3DIR=/usr/local/lib/perl5/man/man3 all umask 0 && cp Oraperl.pm /usr/local/lib/perl5/Oraperl.pm sh: /usr/local/lib/perl5/sun4-solaris/auto/DBD/Oracle/.packlist: cannot create *** Error code 1 make: Fatal error: Command failed for target `/usr/local/lib/perl5/Oraperl.pm' Current working directory /private/slivings/perl5.001n/ext/DBD-Oracle-0.27 *** Error code 1 make: Fatal error: Command failed for target `pure_install' elcaro% ----------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19560-8@oink>; Tue, 19 Dec 1995 07:13:20 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819347157:19348:1; Tue, 19 Dec 95 04:25:57 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa19224; 19 Dec 95 4:25 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id sg.aa25694; 19 Dec 95 4:25 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Mon, 18 Dec 1995 20:17:33 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA20315 for dbi-users-real; Mon, 18 Dec 1995 10:01:47 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fox.earthweb.com (fox.earthweb.com [206.152.10.2]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA20311 for ; Mon, 18 Dec 1995 10:01:40 -0800 Received: (from patricia@localhost) by fox.earthweb.com (8.6.12/8.6.10) id MAA16897; Mon, 18 Dec 1995 12:58:28 -0500 Date: Mon, 18 Dec 1995 12:58:27 -0500 (EST) From: Patricia Ju To: dbi-users@fugue.com Subject: DBD::Oracle 'make test' fails Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 4:25:28 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 19 Dec 95 4:25:40 GMT] Hi all. I recently compiled and installed perl 5.001n and DBI 0.63. I was successful in compiling and installing DBD::mSQL. Compiling DBD::Oracle seemed to be okay as well. However, when running 'make test' I get the following errors. Please help. I've got live scripts that are hurting due to the lack of Oracle.. Thanks. Patricia ---------- Forwarded message ---------- Date: Mon, 18 Dec 1995 12:54:03 -0500 From: patricia@fox.earthweb.com To: patricia@fox.earthweb.com PERL_DL_NONLAZY=1 /usr/local/bin/perl -I./blib/sun4-solaris -I./blib -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/lib/perl5 -e 'use Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t t/base..............install_driver(Oracle) failed: Can't load './blib/sun4-solaris/auto/DBD/Oracle/Oracle.so' for module DBD::Oracle: ld.so.1: /usr/local/bin/perl: fatal: relocation error: symbol not found: DBIc_ACTIVE_KIDS: referenced in ./blib/sun4-solaris/auto/DBD/Oracle/Oracle.so at /usr/local/lib/perl5/DynaLoader.pm line 136. at blib/DBD/Oracle.pm line 18 at /usr/local/lib/perl5/DBI.pm line 167 DBI::install_driver called at t/base.t line 16 FAILED on test 4 t/main..............ok Failed 1 test, 50.00% okay. Files=2, Tests=6, 2 secs ( 0.75 cusr 0.30 csys = 1.05 cpu) PERL_DL_NONLAZY=1 /usr/local/bin/perl -I./blib/sun4-solaris -I./blib -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/lib/perl5 test.pl Oraperl test application $Revision: 1.19 $ install_driver(Oracle) failed: Can't load './blib/sun4-solaris/auto/DBD/Oracle/Oracle.so' for module DBD::Oracle: ld.so.1: /usr/local/bin/perl: fatal: relocation error: symbol not found: DBIc_ACTIVE_KIDS: referenced in ./blib/sun4-solaris/auto/DBD/Oracle/Oracle.so at /usr/local/lib/perl5/DynaLoader.pm line 136. at blib/DBD/Oracle.pm line 18 at /usr/local/lib/perl5/DBI.pm line 167 DBI::install_driver called at blib/Oraperl.pm line 50 (eval) called at (eval 2) line 1 main::BEGIN called at /usr/local/lib/perl5/DBI.pm line 0 (eval) called at /usr/local/lib/perl5/DBI.pm line 0 (eval) called at test.pl line 28 BEGIN failed--compilation aborted at (eval 2) line 1. make: *** [test_dynamic] Error 2 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19560-22@oink>; Tue, 19 Dec 1995 07:15:52 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819347693:22339:0; Tue, 19 Dec 95 04:34:53 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa21931; 19 Dec 95 4:33 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id msg.ab26759; 19 Dec 95 4:33 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 03:07:48 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA22917 for dbi-users-real; Mon, 18 Dec 1995 15:48:10 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from pilot.njin.net (pilot.njin.net [128.6.7.38]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA22913 for ; Mon, 18 Dec 1995 15:48:08 -0800 Received: from localhost (phj@localhost) by pilot.njin.net (8.6.12+bestmx+oldruq+newsunq+grosshack/8.6.12) with ESMTP id SAA26011; Mon, 18 Dec 1995 18:48:08 -0500 Message-Id: <199512182348.SAA26011@pilot.njin.net> To: Steve Livingston cc: dbi-users@fugue.com Subject: Re: Installing DBD-0.27 on Sparc/Solaris In-reply-to: Your message of "Mon, 18 Dec 1995 15:31:26 CST." <199512182135.AA29857@metronet.com> Date: Mon, 18 Dec 1995 18:48:07 -0500 From: "Patricia H. Ju" X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 4:33:12 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 19 Dec 95 4:33:40 GMT] On Mon, 18 Dec 1995 15:31:26 CST Steve Livingston wrote: | (failed) (6) make installed | |The "make install" fails with: | sh: /usr/local/lib/perl5/sun4-solaris/auto/DBD/Oracle/.packlist: |cannot create | *** Error code 1 Do you have permissions to install ? (Probably need root.) Patricia ............. Who needs patience when you've got force ? ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19741-1@oink>; Tue, 19 Dec 1995 07:17:07 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819347774:22729:2; Tue, 19 Dec 95 04:36:14 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa22574; 19 Dec 95 4:35 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id sg.ac26759; 19 Dec 95 4:35 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 03:59:16 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA23316 for dbi-users-real; Mon, 18 Dec 1995 17:45:09 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sultan.hsii.ccare.com ([192.101.181.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA23312 for ; Mon, 18 Dec 1995 17:45:07 -0800 Received: from msgw.hsii.ccare.com by SULTAN.HSII.CCARE.COM (PMDF V5.0-5 #4597) id <01HYYQAIMJHS0005RP@SULTAN.HSII.CCARE.COM>; Mon, 18 Dec 1995 19:40:45 -0600 (CST) Received: by msgw.hsii.ccare.com with Microsoft Mail id <30D61826@msgw.hsii.ccare.com>; Mon, 18 Dec 1995 19:40:54 -0600 (CST) Date: Mon, 18 Dec 1995 19:36:00 -0600 (CST) From: "Li, Alexander" Subject: Trying to install DBD::Oracle To: DBI User List Cc: Tim Bunce Message-id: <30D61826@msgw.hsii.ccare.com> X-Mailer: Microsoft Mail V3.0 Content-transfer-encoding: 7BIT Encoding: 57 TEXT X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 4:35:05 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 19 Dec 95 4:35:27 GMT] Hello all, I have Perl 5 and DBI-0.65 installed and attemping to install DBD::Oracle. ----- gh753> perl Makefile.PL Configuring DBD::Oracle version 0.21... Remember to actually read the README file! Using Oracle in /usr/local/oracle Warning: OCI (Pro*C) does not appear to be installed. Oracle sysliblist: -lsocket -lnsl -lm -ldl -laio -lposix4 Unable to locate proc.mk (use -m /path/to/proc.mk to specify) ---- Ok, I don't have OCI (Pro*C), which is correct. Now, do I need *all* the Oracle components installed or just the OCI (Pro*C) module will do? Here's my config. info: ---- gh753> perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=solaris, osver=2.4, archname=sun4-solaris uname='sunos gh753 5.4 generic_101945-10 sun4m sparc ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-DDEBUGGING' ccflags ='-DDEBUGGING' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/lib /usr/lib /usr/ccs/lib /usr/local/lib libs=-lsocket -lnsl -ldl -lm -lc -lcrypt libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-Kpic', ccdlflags=' ', lddlflags='-G' ------ Thanks for any help. ps: Tim, the DBI-0.65 build just fine after I rebuild Perl 5. Thanks. -------------------------------- Alex Li Sys. Admin Health Systems Integration, Inc. -------------------------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19560-28@oink>; Tue, 19 Dec 1995 07:17:08 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819347775:22797:0; Tue, 19 Dec 95 04:36:15 GMT Received: from relay-1.mail.demon.net by punt-1.mail.demon.net id aa22605; 19 Dec 95 4:35 GMT Received: from frigate.doc.ic.ac.uk by relay-1.mail.demon.net id sg.ae26759; 19 Dec 95 4:35 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 04:02:12 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA22835 for dbi-users-real; Mon, 18 Dec 1995 15:17:49 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from metronet.com (root@mail.metronet.com [192.245.137.6]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id PAA22831 for ; Mon, 18 Dec 1995 15:17:47 -0800 Received: from hometown.com by metronet.com with SMTP id AA00696 (5.67a/IDA1.5hp for ); Mon, 18 Dec 1995 17:17:37 -0600 Message-Id: <199512182317.AA00696@metronet.com> X-Sender: mouche@mail.metronet.com X-Mailer: Windows Eudora Version 2.1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 18 Dec 1995 17:13:25 -0600 To: dbi-users@fugue.com From: Steve Livingston Subject: Installing DBD-0.27 on Sparc/Solaris Cc: Tim.Bunce@ig.co.uk X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 4:35:26 GMT] X-SMTP-Posting-Host: relay-1.mail.demon.net [Tue, 19 Dec 95 4:35:36 GMT] Thanks to those who replied about checking permissions. The solution was along those lines: I had write permission to the directory, but... .packlist had been created in a previous installation attempt, and... .packlist was read-only to the user/owner. Deleting the old .packlist gave me a clean "make install" Thanks to all who replied. Cheers, Steve > >The "make install" fails with: > sh: /usr/local/lib/perl5/sun4-solaris/auto/DBD/Oracle/.packlist: cannot create > *** Error code 1 ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19741-16@oink>; Tue, 19 Dec 1995 07:21:08 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819356399:10611:4; Tue, 19 Dec 95 06:59:59 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id au09945; 19 Dec 95 6:59 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 05:43:42 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA23891 for dbi-announce-real; Mon, 18 Dec 1995 20:02:58 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from srvr7.engin.umich.edu (root@srvr7.engin.umich.edu [141.212.2.69]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA23887; Mon, 18 Dec 1995 20:02:55 -0800 Received: from srvr5.engin.umich.edu (srvr5.engin.umich.edu [141.212.2.89]) by srvr7.engin.umich.edu (8.6.12/8.6.4) with ESMTP id WAA09469; Mon, 18 Dec 1995 22:51:05 -0500 Received: from zaphod (pm035-27.dialip.mich.net [141.211.7.38]) by srvr5.engin.umich.edu (8.6.12/8.6.4) with SMTP id WAA19432; Mon, 18 Dec 1995 22:48:00 -0500 Message-ID: <30D636F2.5B21@umich.edu> Date: Mon, 18 Dec 1995 22:52:18 -0500 From: Dan DeMaggio Organization: Completely Disorganized X-Mailer: Mozilla 2.0b3 (WinNT; I) MIME-Version: 1.0 To: William Warner CC: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important References: <9512120420.AA03096@soc5.airone.claircom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 6:59:34 GMT] William Warner wrote: > What is SQL CLI/ODBC ? ODBC is a 'standard' API for talking to databases. It is similar to Winsock or POSIX in that it replaced a mish-mash of propritary APIs with just ONE. On the windows platform, it is quite good now (inital drivers were slow and buggy). In fact, Micro$oft has stated that their next SQL Server will talk ODBC 'natively' and they will use ODBC for benchmarking. ODBC allows multiple connections (even at the same time) to multiple data sources. It is deeply rooted in MS Windows, but has been ported Unix by 3rd parties. Mac ODBC is also in the works. To the user, it's pretty slick. First, you buy a database package that comes with ODBC Drivers (FoxPro, Access, Paradox, Oracle, Etc.) which you install on your client. Then take any application (Excel, Word, Crystal Reports, etc. etc.) that can use ODBC. In the app, you can browse your datasources, tables, fields, etc. and Import or Export data. Most PC databases come with an ODBC driver. The non-PC-centric ones may have 3rd party ($$) drivers avaliable. (i.e. big mainframe type databases) There is even a company selling ODBC middleware so you don't have to install all the drivers for all your datasources on all your clients. To the developer, ODBC Works like this: Application -> Driver Manager (DLL) -> DBMS Driver (DLL) -> Data Your application makes calls to the ODBC API thru the ODBC Driver Manager DLL (Dynamic Link Library). You ask to driver manager to enumerate datasources or to connect to a particular data source. The Driver Manager loads the appropriate DBMS driver. You can make calls to the driver manager to see what functions the driver supports (there are several levels of conformance). You can then establish a connection to a database, and create statement handles, do queries, get results, get information ABOUT results (i.e. data types of the columns) and such. I count 56 ODBC API functions, (not including the ODBC Install functions, ODBC translation functions, etc.). Defining a standard based on ODBC would make a simple 'select' take several lines of Perl. And if you care about column lengths, etc. then you will add quite a few more lines. And then there's the asynchronus stuff. On the other hand, if you only care about getting (mostly) text in and out of the database, you don't want to be bogged down with a lot of useless hand-waving calls. [i.e. in ODBC, the first thing you do is allocate an 'enviroment' from the driver manager before you even tell it what driver you want to talk to. Total waste of CPU cycles, if you ask me, but you didn't.] I am not an expert in ODBC by anymeans (I just have MSVC). But I'm sure I don't want to write Perl programs that make all those Messy calls. Maybe the perl database packages should all implement ODBC, and we can write a GENERIC database class that will make the nasty calls for us. That way, we know we have the full ODBC functionality (i.e. full database functionality) but only if we need it... What do you think? -- Dan DeMaggio (dmag@umich.edu) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19741-17@oink>; Tue, 19 Dec 1995 07:21:23 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819356401:10658:2; Tue, 19 Dec 95 07:00:01 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id av09945; 19 Dec 95 6:59 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 05:45:46 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA23913 for dbi-users-real; Mon, 18 Dec 1995 20:06:52 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from srvr7.engin.umich.edu (root@srvr7.engin.umich.edu [141.212.2.69]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA23887; Mon, 18 Dec 1995 20:02:55 -0800 Received: from srvr5.engin.umich.edu (srvr5.engin.umich.edu [141.212.2.89]) by srvr7.engin.umich.edu (8.6.12/8.6.4) with ESMTP id WAA09469; Mon, 18 Dec 1995 22:51:05 -0500 Received: from zaphod (pm035-27.dialip.mich.net [141.211.7.38]) by srvr5.engin.umich.edu (8.6.12/8.6.4) with SMTP id WAA19432; Mon, 18 Dec 1995 22:48:00 -0500 Message-ID: <30D636F2.5B21@umich.edu> Date: Mon, 18 Dec 1995 22:52:18 -0500 From: Dan DeMaggio Organization: Completely Disorganized X-Mailer: Mozilla 2.0b3 (WinNT; I) MIME-Version: 1.0 To: William Warner CC: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important References: <9512120420.AA03096@soc5.airone.claircom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 6:59:36 GMT] William Warner wrote: > What is SQL CLI/ODBC ? ODBC is a 'standard' API for talking to databases. It is similar to Winsock or POSIX in that it replaced a mish-mash of propritary APIs with just ONE. On the windows platform, it is quite good now (inital drivers were slow and buggy). In fact, Micro$oft has stated that their next SQL Server will talk ODBC 'natively' and they will use ODBC for benchmarking. ODBC allows multiple connections (even at the same time) to multiple data sources. It is deeply rooted in MS Windows, but has been ported Unix by 3rd parties. Mac ODBC is also in the works. To the user, it's pretty slick. First, you buy a database package that comes with ODBC Drivers (FoxPro, Access, Paradox, Oracle, Etc.) which you install on your client. Then take any application (Excel, Word, Crystal Reports, etc. etc.) that can use ODBC. In the app, you can browse your datasources, tables, fields, etc. and Import or Export data. Most PC databases come with an ODBC driver. The non-PC-centric ones may have 3rd party ($$) drivers avaliable. (i.e. big mainframe type databases) There is even a company selling ODBC middleware so you don't have to install all the drivers for all your datasources on all your clients. To the developer, ODBC Works like this: Application -> Driver Manager (DLL) -> DBMS Driver (DLL) -> Data Your application makes calls to the ODBC API thru the ODBC Driver Manager DLL (Dynamic Link Library). You ask to driver manager to enumerate datasources or to connect to a particular data source. The Driver Manager loads the appropriate DBMS driver. You can make calls to the driver manager to see what functions the driver supports (there are several levels of conformance). You can then establish a connection to a database, and create statement handles, do queries, get results, get information ABOUT results (i.e. data types of the columns) and such. I count 56 ODBC API functions, (not including the ODBC Install functions, ODBC translation functions, etc.). Defining a standard based on ODBC would make a simple 'select' take several lines of Perl. And if you care about column lengths, etc. then you will add quite a few more lines. And then there's the asynchronus stuff. On the other hand, if you only care about getting (mostly) text in and out of the database, you don't want to be bogged down with a lot of useless hand-waving calls. [i.e. in ODBC, the first thing you do is allocate an 'enviroment' from the driver manager before you even tell it what driver you want to talk to. Total waste of CPU cycles, if you ask me, but you didn't.] I am not an expert in ODBC by anymeans (I just have MSVC). But I'm sure I don't want to write Perl programs that make all those Messy calls. Maybe the perl database packages should all implement ODBC, and we can write a GENERIC database class that will make the nasty calls for us. That way, we know we have the full ODBC functionality (i.e. full database functionality) but only if we need it... What do you think? -- Dan DeMaggio (dmag@umich.edu) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19210-85@oink>; Tue, 19 Dec 1995 07:21:31 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819356475:11174:0; Tue, 19 Dec 95 07:01:15 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id ax09945; 19 Dec 95 6:59 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 06:07:04 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA24063 for dbi-users-real; Mon, 18 Dec 1995 20:40:43 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mailgw.claircom.com (mailgw.claircom.com [199.5.241.51]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA24042; Mon, 18 Dec 1995 20:39:04 -0800 Received: from nimo.claircom.com by mailgw.claircom.com with smtp (Smail3.1.26.7 #2) id m0tRtpQ-0004tpC; Mon, 18 Dec 95 20:39 PST Received: from airone.claircom.com by nimo.claircom.com with smtp (Smail3.1.26.7 #2) id m0tRtpP-0005VkC; Mon, 18 Dec 95 20:39 PST Received: from soc4.airone.claircom.com by airone.claircom.com (5.x/SMI-SVR4) id AA14395; Mon, 18 Dec 1995 20:35:21 -0800 Received: by soc4.airone.claircom.com (4.1/SMI-4.1) id AA10131; Mon, 18 Dec 95 20:11:31 PST Date: Mon, 18 Dec 95 20:11:31 PST From: billw@airone.claircom.com (William Warner) Message-Id: <9512190411.AA10131@soc4.airone.claircom.com> To: billw@airone.claircom.com, dmag@umich.edu Subject: Re: New directions for DBI - Important Cc: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 6:59:41 GMT] Very informative! Thank you! My two cents: DBI's first priority should be to conform to useful, accepted standards, because Perl is such good "glue" for patching disparate data/applications together. Perl needs to conform to those standards where market forces and myopic planning cannot. DBI's second priority should be to reduce the first priority to a one-liner. ww ----- Begin Included Message ----- From dmag@umich.edu Mon Dec 18 19:59 PST 1995 Date: Mon, 18 Dec 1995 22:52:18 -0500 From: Dan DeMaggio Organization: Completely Disorganized X-Mailer: Mozilla 2.0b3 (WinNT; I) Mime-Version: 1.0 To: William Warner Cc: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Content-Length: 3059 William Warner wrote: > What is SQL CLI/ODBC ? ODBC is a 'standard' API for talking to databases. It is similar to Winsock or POSIX in that it replaced a mish-mash of propritary APIs with just ONE. On the windows platform, it is quite good now (inital drivers were slow and buggy). In fact, Micro$oft has stated that their next SQL Server will talk ODBC 'natively' and they will use ODBC for benchmarking. ODBC allows multiple connections (even at the same time) to multiple data sources. It is deeply rooted in MS Windows, but has been ported Unix by 3rd parties. Mac ODBC is also in the works. To the user, it's pretty slick. First, you buy a database package that comes with ODBC Drivers (FoxPro, Access, Paradox, Oracle, Etc.) which you install on your client. Then take any application (Excel, Word, Crystal Reports, etc. etc.) that can use ODBC. In the app, you can browse your datasources, tables, fields, etc. and Import or Export data. Most PC databases come with an ODBC driver. The non-PC-centric ones may have 3rd party ($$) drivers avaliable. (i.e. big mainframe type databases) There is even a company selling ODBC middleware so you don't have to install all the drivers for all your datasources on all your clients. To the developer, ODBC Works like this: Application -> Driver Manager (DLL) -> DBMS Driver (DLL) -> Data Your application makes calls to the ODBC API thru the ODBC Driver Manager DLL (Dynamic Link Library). You ask to driver manager to enumerate datasources or to connect to a particular data source. The Driver Manager loads the appropriate DBMS driver. You can make calls to the driver manager to see what functions the driver supports (there are several levels of conformance). You can then establish a connection to a database, and create statement handles, do queries, get results, get information ABOUT results (i.e. data types of the columns) and such. I count 56 ODBC API functions, (not including the ODBC Install functions, ODBC translation functions, etc.). Defining a standard based on ODBC would make a simple 'select' take several lines of Perl. And if you care about column lengths, etc. then you will add quite a few more lines. And then there's the asynchronus stuff. On the other hand, if you only care about getting (mostly) text in and out of the database, you don't want to be bogged down with a lot of useless hand-waving calls. [i.e. in ODBC, the first thing you do is allocate an 'enviroment' from the driver manager before you even tell it what driver you want to talk to. Total waste of CPU cycles, if you ask me, but you didn't.] I am not an expert in ODBC by anymeans (I just have MSVC). But I'm sure I don't want to write Perl programs that make all those Messy calls. Maybe the perl database packages should all implement ODBC, and we can write a GENERIC database class that will make the nasty calls for us. That way, we know we have the full ODBC functionality (i.e. full database functionality) but only if we need it... What do you think? -- Dan DeMaggio (dmag@umich.edu) ----- End Included Message ----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19741-18@oink>; Tue, 19 Dec 1995 07:21:37 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819356472:11226:0; Tue, 19 Dec 95 07:01:12 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id bb09945; 19 Dec 95 7:00 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 06:44:06 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA24260 for dbi-users-real; Mon, 18 Dec 1995 21:10:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from srvr7.engin.umich.edu (root@srvr7.engin.umich.edu [141.212.2.69]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id VAA24240; Mon, 18 Dec 1995 21:08:10 -0800 Received: from srvr5.engin.umich.edu (srvr5.engin.umich.edu [141.212.2.89]) by srvr7.engin.umich.edu (8.6.12/8.6.4) with ESMTP id AAA10206; Tue, 19 Dec 1995 00:00:56 -0500 Received: from zaphod (pm035-27.dialip.mich.net [141.211.7.38]) by srvr5.engin.umich.edu (8.6.12/8.6.4) with SMTP id XAA22502; Mon, 18 Dec 1995 23:58:00 -0500 Message-ID: <30D64759.7D1A@umich.edu> Date: Tue, 19 Dec 1995 00:02:17 -0500 From: Dan DeMaggio Organization: Completely Disorganized X-Mailer: Mozilla 2.0b3 (WinNT; I) MIME-Version: 1.0 To: William Warner CC: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important References: <9512190411.AA10131@soc4.airone.claircom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 7:00:27 GMT] William Warner wrote: > DBI's second priority should be to reduce the first priority to a one-liner. Couldn't have said it better myself. I've been learning Java the past few months. It's better than C (no pointers!), but it shares a commom property with C: A lot of your code turns out to be detailed housekeeping. Even when you understand something and have written it a million times, it still takes a lot of code. There's no compressing it to a smaller program. In Perl, the language allows experienced programmers to take 'shortcuts' (and hence, compress your program.) To me, the Database interface must be usable in 4 or 5 calls (open database, send SQL, fetch results, close database). On the otherhand, this ignores fancy things like "will my 5 MB string fit into that fixed-length database field?". This requires a full-featured DB interface like ODBC. > ----- Begin Included Message ----- > >From dmag@umich.edu Mon Dec 18 19:59 PST 1995 [gratitutious article quoting deleted to keep the Internet from collapsing] > ----- End Included Message ----- -- Dan DeMaggio (dmag@umich.edu) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <19560-48@oink>; Tue, 19 Dec 1995 07:21:41 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819356472:11216:0; Tue, 19 Dec 95 07:01:12 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id az09945; 19 Dec 95 7:00 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 06:14:14 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA24046 for dbi-announce-real; Mon, 18 Dec 1995 20:39:09 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mailgw.claircom.com (mailgw.claircom.com [199.5.241.51]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id UAA24042; Mon, 18 Dec 1995 20:39:04 -0800 Received: from nimo.claircom.com by mailgw.claircom.com with smtp (Smail3.1.26.7 #2) id m0tRtpQ-0004tpC; Mon, 18 Dec 95 20:39 PST Received: from airone.claircom.com by nimo.claircom.com with smtp (Smail3.1.26.7 #2) id m0tRtpP-0005VkC; Mon, 18 Dec 95 20:39 PST Received: from soc4.airone.claircom.com by airone.claircom.com (5.x/SMI-SVR4) id AA14395; Mon, 18 Dec 1995 20:35:21 -0800 Received: by soc4.airone.claircom.com (4.1/SMI-4.1) id AA10131; Mon, 18 Dec 95 20:11:31 PST Date: Mon, 18 Dec 95 20:11:31 PST From: billw@airone.claircom.com (William Warner) Message-Id: <9512190411.AA10131@soc4.airone.claircom.com> To: billw@airone.claircom.com, dmag@umich.edu Subject: Re: New directions for DBI - Important Cc: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 7:00:24 GMT] Very informative! Thank you! My two cents: DBI's first priority should be to conform to useful, accepted standards, because Perl is such good "glue" for patching disparate data/applications together. Perl needs to conform to those standards where market forces and myopic planning cannot. DBI's second priority should be to reduce the first priority to a one-liner. ww ----- Begin Included Message ----- From dmag@umich.edu Mon Dec 18 19:59 PST 1995 Date: Mon, 18 Dec 1995 22:52:18 -0500 From: Dan DeMaggio Organization: Completely Disorganized X-Mailer: Mozilla 2.0b3 (WinNT; I) Mime-Version: 1.0 To: William Warner Cc: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Content-Length: 3059 William Warner wrote: > What is SQL CLI/ODBC ? ODBC is a 'standard' API for talking to databases. It is similar to Winsock or POSIX in that it replaced a mish-mash of propritary APIs with just ONE. On the windows platform, it is quite good now (inital drivers were slow and buggy). In fact, Micro$oft has stated that their next SQL Server will talk ODBC 'natively' and they will use ODBC for benchmarking. ODBC allows multiple connections (even at the same time) to multiple data sources. It is deeply rooted in MS Windows, but has been ported Unix by 3rd parties. Mac ODBC is also in the works. To the user, it's pretty slick. First, you buy a database package that comes with ODBC Drivers (FoxPro, Access, Paradox, Oracle, Etc.) which you install on your client. Then take any application (Excel, Word, Crystal Reports, etc. etc.) that can use ODBC. In the app, you can browse your datasources, tables, fields, etc. and Import or Export data. Most PC databases come with an ODBC driver. The non-PC-centric ones may have 3rd party ($$) drivers avaliable. (i.e. big mainframe type databases) There is even a company selling ODBC middleware so you don't have to install all the drivers for all your datasources on all your clients. To the developer, ODBC Works like this: Application -> Driver Manager (DLL) -> DBMS Driver (DLL) -> Data Your application makes calls to the ODBC API thru the ODBC Driver Manager DLL (Dynamic Link Library). You ask to driver manager to enumerate datasources or to connect to a particular data source. The Driver Manager loads the appropriate DBMS driver. You can make calls to the driver manager to see what functions the driver supports (there are several levels of conformance). You can then establish a connection to a database, and create statement handles, do queries, get results, get information ABOUT results (i.e. data types of the columns) and such. I count 56 ODBC API functions, (not including the ODBC Install functions, ODBC translation functions, etc.). Defining a standard based on ODBC would make a simple 'select' take several lines of Perl. And if you care about column lengths, etc. then you will add quite a few more lines. And then there's the asynchronus stuff. On the other hand, if you only care about getting (mostly) text in and out of the database, you don't want to be bogged down with a lot of useless hand-waving calls. [i.e. in ODBC, the first thing you do is allocate an 'enviroment' from the driver manager before you even tell it what driver you want to talk to. Total waste of CPU cycles, if you ask me, but you didn't.] I am not an expert in ODBC by anymeans (I just have MSVC). But I'm sure I don't want to write Perl programs that make all those Messy calls. Maybe the perl database packages should all implement ODBC, and we can write a GENERIC database class that will make the nasty calls for us. That way, we know we have the full ODBC functionality (i.e. full database functionality) but only if we need it... What do you think? -- Dan DeMaggio (dmag@umich.edu) ----- End Included Message ----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <23154-0@oink>; Tue, 19 Dec 1995 12:54:38 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819377323:22421:2; Tue, 19 Dec 95 12:48:43 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa22197; 19 Dec 95 12:48 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 07:15:49 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA24260 for dbi-users-real; Mon, 18 Dec 1995 21:10:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from srvr7.engin.umich.edu (root@srvr7.engin.umich.edu [141.212.2.69]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id VAA24240; Mon, 18 Dec 1995 21:08:10 -0800 Received: from srvr5.engin.umich.edu (srvr5.engin.umich.edu [141.212.2.89]) by srvr7.engin.umich.edu (8.6.12/8.6.4) with ESMTP id AAA10206; Tue, 19 Dec 1995 00:00:56 -0500 Received: from zaphod (pm035-27.dialip.mich.net [141.211.7.38]) by srvr5.engin.umich.edu (8.6.12/8.6.4) with SMTP id XAA22502; Mon, 18 Dec 1995 23:58:00 -0500 Message-ID: <30D64759.7D1A@umich.edu> Date: Tue, 19 Dec 1995 00:02:17 -0500 From: Dan DeMaggio Organization: Completely Disorganized X-Mailer: Mozilla 2.0b3 (WinNT; I) MIME-Version: 1.0 To: William Warner CC: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important References: <9512190411.AA10131@soc4.airone.claircom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 12:48:08 GMT] William Warner wrote: > DBI's second priority should be to reduce the first priority to a one-liner. Couldn't have said it better myself. I've been learning Java the past few months. It's better than C (no pointers!), but it shares a commom property with C: A lot of your code turns out to be detailed housekeeping. Even when you understand something and have written it a million times, it still takes a lot of code. There's no compressing it to a smaller program. In Perl, the language allows experienced programmers to take 'shortcuts' (and hence, compress your program.) To me, the Database interface must be usable in 4 or 5 calls (open database, send SQL, fetch results, close database). On the otherhand, this ignores fancy things like "will my 5 MB string fit into that fixed-length database field?". This requires a full-featured DB interface like ODBC. > ----- Begin Included Message ----- > >From dmag@umich.edu Mon Dec 18 19:59 PST 1995 [gratitutious article quoting deleted to keep the Internet from collapsing] > ----- End Included Message ----- -- Dan DeMaggio (dmag@umich.edu) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <23154-11@oink>; Tue, 19 Dec 1995 12:55:32 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819377328:22410:4; Tue, 19 Dec 95 12:48:48 GMT Received: from frigate.doc.ic.ac.uk by punt-1.mail.demon.net id aa22280; 19 Dec 95 12:48 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Tue, 19 Dec 1995 07:27:52 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id VAA24244 for dbi-announce-real; Mon, 18 Dec 1995 21:08:12 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from srvr7.engin.umich.edu (root@srvr7.engin.umich.edu [141.212.2.69]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id VAA24240; Mon, 18 Dec 1995 21:08:10 -0800 Received: from srvr5.engin.umich.edu (srvr5.engin.umich.edu [141.212.2.89]) by srvr7.engin.umich.edu (8.6.12/8.6.4) with ESMTP id AAA10206; Tue, 19 Dec 1995 00:00:56 -0500 Received: from zaphod (pm035-27.dialip.mich.net [141.211.7.38]) by srvr5.engin.umich.edu (8.6.12/8.6.4) with SMTP id XAA22502; Mon, 18 Dec 1995 23:58:00 -0500 Message-ID: <30D64759.7D1A@umich.edu> Date: Tue, 19 Dec 1995 00:02:17 -0500 From: Dan DeMaggio Organization: Completely Disorganized X-Mailer: Mozilla 2.0b3 (WinNT; I) MIME-Version: 1.0 To: William Warner CC: dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important References: <9512190411.AA10131@soc4.airone.claircom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SMTP-Posting-Host: frigate.doc.ic.ac.uk [Tue, 19 Dec 95 12:48:15 GMT] William Warner wrote: > DBI's second priority should be to reduce the first priority to a one-liner. Couldn't have said it better myself. I've been learning Java the past few months. It's better than C (no pointers!), but it shares a commom property with C: A lot of your code turns out to be detailed housekeeping. Even when you understand something and have written it a million times, it still takes a lot of code. There's no compressing it to a smaller program. In Perl, the language allows experienced programmers to take 'shortcuts' (and hence, compress your program.) To me, the Database interface must be usable in 4 or 5 calls (open database, send SQL, fetch results, close database). On the otherhand, this ignores fancy things like "will my 5 MB string fit into that fixed-length database field?". This requires a full-featured DB interface like ODBC. > ----- Begin Included Message ----- > >From dmag@umich.edu Mon Dec 18 19:59 PST 1995 [gratitutious article quoting deleted to keep the Internet from collapsing] > ----- End Included Message ----- -- Dan DeMaggio (dmag@umich.edu) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <27201-3@oink>; Tue, 19 Dec 1995 17:15:04 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819393073:00799:4; Tue, 19 Dec 95 17:11:13 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa00497; 19 Dec 95 17:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA02726 for dbi-users-real; Tue, 19 Dec 1995 08:17:31 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA02722 for ; Tue, 19 Dec 1995 08:17:27 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id af08935; 19 Dec 95 16:13 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id aa02182; 19 Dec 95 16:10 GMT Received: from toad by oink with SMTP (PP) id <24815-0@oink>; Tue, 19 Dec 1995 14:46:34 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA28679; Tue, 19 Dec 1995 14:46:07 +0000 Date: Tue, 19 Dec 1995 14:46:07 +0000 From: Tim Bunce Message-Id: <9512191446.AA28679@toad.ig.co.uk> To: dmag@umich.edu Subject: Re: New directions for DBI - Important Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 765 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Dan DeMaggio > > I am not an expert in ODBC by anymeans (I just have MSVC). But I'm sure I > don't want to write Perl programs that make all those Messy calls. Maybe > the perl database packages should all implement ODBC, and we can write a > GENERIC database class that will make the nasty calls for us. That way, > we know we have the full ODBC functionality (i.e. full database functionality) > but only if we need it... > > What do you think? Let me assure you that while the new DBI will include access to all ODBC functions you will almost always be able to use high-level utility methods, including the familiar: $sth => $dbh->prepare("..."); $sth->execute; @row = $sth->fetchrow; Native ODBC is ugly and verbose. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <27720-0@oink>; Tue, 19 Dec 1995 17:48:03 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819395049:02545:1; Tue, 19 Dec 95 17:44:09 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa02155; 19 Dec 95 17:43 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA02877 for dbi-announce-real; Tue, 19 Dec 1995 08:52:15 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from proxy.netlabs.com (ns.netlabs.com [192.94.48.47]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA02866; Tue, 19 Dec 1995 08:52:02 -0800 Received: by proxy.netlabs.com (4.1/SMI-4.1) id AA06854; Tue, 19 Dec 95 08:57:23 PST Received: from vaccine-bb.sems.com(192.190.199.1) by proxy.netlabs.com via smap (V1.3) id sma006852; Tue Dec 19 08:57:11 1995 Received: from scalpel (scalpel.netlabs.com) by netlabs.com (4.1/SMI-4.1) id AA08317; Tue, 19 Dec 95 08:54:27 PST Received: from localhost.netlabs.com by scalpel (4.1/SMI-4.1) id AA05876; Tue, 19 Dec 95 08:54:25 PST Message-Id: <9512191654.AA05876@scalpel> To: dmag@umich.edu (Dan DeMaggio) Cc: William Warner , dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important In-Reply-To: Your message of "Tue, 19 Dec 95 00:02:17 EST." <30D64759.7D1A@umich.edu> Date: Tue, 19 Dec 95 08:54:25 -0800 From: Larry Wall : William Warner wrote: : > DBI's second priority should be to reduce the first priority to a one-liner. : : Couldn't have said it better myself. I've been learning Java the past few : months. It's better than C (no pointers!), but it shares a commom property : with C: A lot of your code turns out to be detailed housekeeping. Even when : you understand something and have written it a million times, it still takes : a lot of code. There's no compressing it to a smaller program. : : In Perl, the language allows experienced programmers to take 'shortcuts' : (and hence, compress your program.) To me, the Database interface must be : usable in 4 or 5 calls (open database, send SQL, fetch results, close database). : On the otherhand, this ignores fancy things like "will my 5 MB string fit into : that fixed-length database field?". This requires a full-featured DB interface : like ODBC. Easy things should be easy. Hard things should be possible. Most computer languages fail on one or the other of these counts. Larry ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <27772-2@oink>; Tue, 19 Dec 1995 17:50:04 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819393789:06540:3; Tue, 19 Dec 95 17:23:09 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa04878; 19 Dec 95 17:20 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA02697 for dbi-users-real; Tue, 19 Dec 1995 08:05:35 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from eol1 (eol1.eo.lu [193.93.48.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA02693 for ; Tue, 19 Dec 1995 08:05:21 -0800 Received: from [193.93.48.253] by eol1 with smtp (Smail3.1.28.1 #4) id m0tS4XG-000bKRC; Tue, 19 Dec 95 17:05 MET Message-Id: To: "dbi-users@fugue.com" Subject: Re: Trying to install DBD::Oracle Date: Tue, 19 Dec 95 17:02:29 --0100 From: Massimo Balestra X-Mailer: E-Mail Connection v2.5.03 CC: "Li, Alexander" -- [ From: Massimo Balestra * EMC.Ver #2.5.02 ] -- -------- REPLY, Original message follows -------- > Date: Monday, 18-Dec-95 07:36 PM > > From: Li, Alexander \ Internet: (alexli@msmail.hsii.ccare. com) > To: dbi-users@fugue.com \ Internet: (dbi-users@fugue.com) > cc: Tim Bunce \ Internet: (tim.bunce@ig.co.uk) > > Subject: Trying to install DBD::Oracle > > Hello all, > > I have Perl 5 and DBI-0.65 installed and attemping to install DBD:: Oracle. > > ----- > gh753> perl Makefile.PL > > Configuring DBD::Oracle version 0.21... > Remember to actually read the README file! > > Using Oracle in /usr/local/oracle > Warning: OCI (Pro*C) does not appear to be installed. > Oracle sysliblist: -lsocket -lnsl -lm -ldl -laio -lposix4 > Unable to locate proc.mk (use -m /path/to/proc.mk to specify) > ---- > > Ok, I don't have OCI (Pro*C), which is correct. Now, do I need *all* the > Oracle > components installed or just the OCI (Pro*C) module will do? > > Here's my config. info: ............. > -------------------------------- > Alex Li > Sys. Admin > Health Systems Integration, Inc. > -------------------------------- > -------- REPLY, End of original message -------- Hi, I don't have ProC installed too but taking a look at the Makefile.PL file I found that we have to tell the installation to use oracle.mk instead of proc .mk. To do this you have only to pass the following parameter: -m /path/to/oracle .mk to the makefile maker. The complete command I type is: perl Makefile.PL -m path/to/oracle.mk Then I Install all the rest in the standard way. On my machine it works. Try and let me know. Bye Massimo -- __________________________________________________________________ o _______________ /\_ _| | Massimo Balestra _\__`[_______________| Email mbalestra@flower.it ] [ \, ][ ][ __________________________________________________________________ ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <28481-0@oink>; Tue, 19 Dec 1995 19:15:36 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819396156:11529:2; Tue, 19 Dec 95 18:02:36 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa11113; 19 Dec 95 18:02 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id IAA02870 for dbi-users-real; Tue, 19 Dec 1995 08:52:04 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from proxy.netlabs.com (ns.netlabs.com [192.94.48.47]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id IAA02866; Tue, 19 Dec 1995 08:52:02 -0800 Received: by proxy.netlabs.com (4.1/SMI-4.1) id AA06854; Tue, 19 Dec 95 08:57:23 PST Received: from vaccine-bb.sems.com(192.190.199.1) by proxy.netlabs.com via smap (V1.3) id sma006852; Tue Dec 19 08:57:11 1995 Received: from scalpel (scalpel.netlabs.com) by netlabs.com (4.1/SMI-4.1) id AA08317; Tue, 19 Dec 95 08:54:27 PST Received: from localhost.netlabs.com by scalpel (4.1/SMI-4.1) id AA05876; Tue, 19 Dec 95 08:54:25 PST Message-Id: <9512191654.AA05876@scalpel> To: dmag@umich.edu (Dan DeMaggio) Cc: William Warner , dbi-announce@fugue.com, dbi-users@fugue.com, Tim.Bunce@ig.co.uk, ak@ananke.s.bawue.de, kejin@lithium.empress.com Subject: Re: New directions for DBI - Important In-Reply-To: Your message of "Tue, 19 Dec 95 00:02:17 EST." <30D64759.7D1A@umich.edu> Date: Tue, 19 Dec 95 08:54:25 -0800 From: Larry Wall : William Warner wrote: : > DBI's second priority should be to reduce the first priority to a one-liner. : : Couldn't have said it better myself. I've been learning Java the past few : months. It's better than C (no pointers!), but it shares a commom property : with C: A lot of your code turns out to be detailed housekeeping. Even when : you understand something and have written it a million times, it still takes : a lot of code. There's no compressing it to a smaller program. : : In Perl, the language allows experienced programmers to take 'shortcuts' : (and hence, compress your program.) To me, the Database interface must be : usable in 4 or 5 calls (open database, send SQL, fetch results, close database). : On the otherhand, this ignores fancy things like "will my 5 MB string fit into : that fixed-length database field?". This requires a full-featured DB interface : like ODBC. Easy things should be easy. Hard things should be possible. Most computer languages fail on one or the other of these counts. Larry ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <28481-15@oink>; Tue, 19 Dec 1995 19:17:00 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819398901:07280:1; Tue, 19 Dec 95 18:48:21 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa06178; 19 Dec 95 18:47 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA03125 for dbi-dev-real; Tue, 19 Dec 1995 10:05:17 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay.inferentia.it (dns.inferentia.it [151.99.220.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA03114; Tue, 19 Dec 1995 10:04:26 -0800 Received: from yquem.inferentia.it (yquem.inferentia.it [194.184.62.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id TAA22686; Tue, 19 Dec 1995 19:03:33 +0100 Posted-Date: Tue, 19 Dec 1995 19:03:33 +0100 Received: by yquem.inferentia.it with Microsoft Mail id <01BACE46.5CF1E820@yquem.inferentia.it>; Tue, 19 Dec 1995 19:15:30 +-100 Message-ID: <01BACE46.5CF1E820@yquem.inferentia.it> From: Davide Migliavacca To: "'dbi-users@fugue.com'" Cc: "'dbi-dev@fugue.com'" Subject: RE: Trying to install DBD::Oracle Date: Tue, 19 Dec 1995 19:15:28 +-100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable It seems Massimo is right. I have looked and the .map files for the various products. The libocic.a = library is part of ocommon, so it should be necessary to install Pro*C = in order to compile DBD. Tested with Oracle 7.1.6 on both AIX 3.2.5 and DEC OSF/1 3.2.=20 Fair warning: I have NOT disinstalled Pro*C to test it completely. I think it is worth a look since it seems people starts using DBI from = scratch instead of slowly substituing Pro*C... On marted=EC 19 dicembre 1995 18.02, Massimo = Balestra[SMTP:mbalestra@flower.it] wrote: > -- [ From: Massimo Balestra * EMC.Ver #2.5.02 ] -- [...] > To do this you have only to pass the following parameter: -m = /path/to/oracle.mk to the makefile maker. [...] > Then I Install all the rest in the standard way. >=20 > On my machine it works.=20 >=20 > Try and let me know. >=20 > Bye > Massimo ---------------------------------------- Davide Migliavacca - Inferentia (Milano, IT) Phone +39 (2) 59928.1 FAX .221 *** opinions above are (you guess!) mine *** ---------------------------------------- ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <28481-22@oink>; Tue, 19 Dec 1995 19:17:34 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819399670:14178:2; Tue, 19 Dec 95 19:01:10 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa13785; 19 Dec 95 19:00 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA03118 for dbi-users-real; Tue, 19 Dec 1995 10:04:33 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay.inferentia.it (dns.inferentia.it [151.99.220.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA03114; Tue, 19 Dec 1995 10:04:26 -0800 Received: from yquem.inferentia.it (yquem.inferentia.it [194.184.62.17]) by relay.inferentia.it (8.6.12/8.6.12) with SMTP id TAA22686; Tue, 19 Dec 1995 19:03:33 +0100 Posted-Date: Tue, 19 Dec 1995 19:03:33 +0100 Received: by yquem.inferentia.it with Microsoft Mail id <01BACE46.5CF1E820@yquem.inferentia.it>; Tue, 19 Dec 1995 19:15:30 +-100 Message-ID: <01BACE46.5CF1E820@yquem.inferentia.it> From: Davide Migliavacca To: "'dbi-users@fugue.com'" Cc: "'dbi-dev@fugue.com'" Subject: RE: Trying to install DBD::Oracle Date: Tue, 19 Dec 1995 19:15:28 +-100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable It seems Massimo is right. I have looked and the .map files for the various products. The libocic.a = library is part of ocommon, so it should be necessary to install Pro*C = in order to compile DBD. Tested with Oracle 7.1.6 on both AIX 3.2.5 and DEC OSF/1 3.2.=20 Fair warning: I have NOT disinstalled Pro*C to test it completely. I think it is worth a look since it seems people starts using DBI from = scratch instead of slowly substituing Pro*C... On marted=EC 19 dicembre 1995 18.02, Massimo = Balestra[SMTP:mbalestra@flower.it] wrote: > -- [ From: Massimo Balestra * EMC.Ver #2.5.02 ] -- [...] > To do this you have only to pass the following parameter: -m = /path/to/oracle.mk to the makefile maker. [...] > Then I Install all the rest in the standard way. >=20 > On my machine it works.=20 >=20 > Try and let me know. >=20 > Bye > Massimo ---------------------------------------- Davide Migliavacca - Inferentia (Milano, IT) Phone +39 (2) 59928.1 FAX .221 *** opinions above are (you guess!) mine *** ---------------------------------------- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <29955-34@oink>; Tue, 19 Dec 1995 20:31:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819404259:07815:2; Tue, 19 Dec 95 20:17:39 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa07382; 19 Dec 95 20:16 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA03971 for dbi-users-real; Tue, 19 Dec 1995 11:46:25 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA03955; Tue, 19 Dec 1995 11:45:52 -0800 Message-Id: <199512191945.LAA03955@toccata.fugue.com> To: dbi-announce@fugue.com, dbi-users@fugue.com Subject: Re: New directions for DBI - Important In-reply-to: Your message of "Tue, 19 Dec 1995 08:54:25 PST." <9512191654.AA05876@scalpel> Date: Tue, 19 Dec 1995 11:45:51 -0800 From: Ted Lemon I've noticed that this discussion has been going on for a long time, Cc'd to both dbi-users and dbi-announce. Unfortunately, dbi-announce is not moderated, which means that this entire discussion has been going on on dbi-announce as well as dbi-users. It's clearly not an announcement, and is therefore not appropriate for dbi-announce. I wish I had some mechanical means of resolving this, but for now could people please remember not to direct replies to dbi-announce? Thanks! _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <29955-35@oink>; Tue, 19 Dec 1995 20:31:59 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819404260:07850:1; Tue, 19 Dec 95 20:17:40 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa07332; 19 Dec 95 20:16 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA03960 for dbi-announce-real; Tue, 19 Dec 1995 11:45:53 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id LAA03955; Tue, 19 Dec 1995 11:45:52 -0800 Message-Id: <199512191945.LAA03955@toccata.fugue.com> To: dbi-announce@fugue.com, dbi-users@fugue.com Subject: Re: New directions for DBI - Important In-reply-to: Your message of "Tue, 19 Dec 1995 08:54:25 PST." <9512191654.AA05876@scalpel> Date: Tue, 19 Dec 1995 11:45:51 -0800 From: Ted Lemon I've noticed that this discussion has been going on for a long time, Cc'd to both dbi-users and dbi-announce. Unfortunately, dbi-announce is not moderated, which means that this entire discussion has been going on on dbi-announce as well as dbi-users. It's clearly not an announcement, and is therefore not appropriate for dbi-announce. I wish I had some mechanical means of resolving this, but for now could people please remember not to direct replies to dbi-announce? Thanks! _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01303-0@oink>; Tue, 19 Dec 1995 21:18:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819406938:00651:3; Tue, 19 Dec 95 21:02:18 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa00353; 19 Dec 95 21:01 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id MAA04350 for dbi-users-real; Tue, 19 Dec 1995 12:20:30 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from ns2.eds.com (ns2.eds.com [199.228.142.78]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id MAA04346 for ; Tue, 19 Dec 1995 12:20:28 -0800 Received: by ns2.eds.com (hello) id PAA32222; Tue, 19 Dec 1995 15:20:39 -0500 Received: by nnsp.eds.com (hello) id PAA09192; Tue, 19 Dec 1995 15:20:08 -0500 Received: from falcon (falcon.mmts.eds.com) by geronimo (5.x/SMI.SVR4) id AA12666; Tue, 19 Dec 1995 14:18:49 -0600 Received: by falcon (940816.SGI.8.6.9/940406.SGI) id OAA00651; Tue, 19 Dec 1995 14:20:32 -0900 Date: Tue, 19 Dec 1995 14:20:32 -0900 Message-Id: <199512192320.OAA00651@falcon> To: Tim Bunce Cc: dbi-users@fugue.com, mpeppler@itf.CH Subject: Re: can DBI return rows as hashes keyed by column names? In-Reply-To: <9512131905.AA04939@toad.ig.co.uk> References: <9512131905.AA04939@toad.ig.co.uk> Reply-To: afc@mmts.eds.com From: afc@mmts.eds.com (Andy Cohan) X-Attribution: AFC X-Face: "P!:r4>2dA)dWK/L(NxSG)-Ffvhe/]5y%uxD}zzNJn{omcc&K8oU>ZXTvJQ,.&Z4C}m)m-8 mzKkq--xzt@CUf'FIp2^Jz(=Fj,t^rU@TfMex\1V=;+j-=o~bv$h3 >> From: Andy Cohan >> >> I'm trying to get our own perl5 DB class to use DBI for an Oracle back end >> and I would like to support the Perl5 Sybase facility of optionally >> returning rows as associative arrays keyed by the result column names (we >> have used it extensively). >> >> Does DBI and DBD::Oracle support this? If not, is it planned? >>>>> "Tim" == Tim Bunce writes: Tim> @fieldnames = (...); Tim> while(@hash{@fieldnames} = $sth->fetchrow) { Tim> ... Tim> } Thanks! This is *definitely* helpful. Now, if the query is something like "SELECT * FROM one_table, two_table ...", how can I know what order the columns are coming back in so that I can set up the fieldnames array? AFC ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <01680-13@oink>; Tue, 19 Dec 1995 22:34:32 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819412201:16353:0; Tue, 19 Dec 95 22:30:01 GMT Received: from relay-1.mail.demon.net ([158.152.1.140]) by punt-1.mail.demon.net id aa15716; 19 Dec 95 22:28 GMT Received: from toccata.fugue.com ([204.254.239.2]) by relay-1.mail.demon.net id aa23836; 19 Dec 95 22:28 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA04783 for dbi-users-real; Tue, 19 Dec 1995 13:37:50 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [129.35.208.98]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA04779 for ; Tue, 19 Dec 1995 13:37:48 -0800 From: mhm@austin.ibm.com Received: from dax.austin.ibm.com (dax.austin.ibm.com [129.35.86.169]) by netmail.austin.ibm.com (8.6.12/8.6.11) with SMTP id PAA58201; Tue, 19 Dec 1995 15:37:47 -0600 Received: by dax.austin.ibm.com (AIX 4.1/UCB 5.64/4.03-client-2.6) for dmag@umich.edu at austin.ibm.com; id AA20074; Tue, 19 Dec 1995 15:37:45 -0600 Message-Id: <9512192137.AA20074@dax.austin.ibm.com> To: Tim Bunce Cc: dmag@umich.edu, dbi-users@fugue.com Subject: Re: New directions for DBI - Important In-Reply-To: Your message of Tue, 19 Dec 95 14:46:07 +0000. <9512191446.AA28679@toad.ig.co.uk> Date: Tue, 19 Dec 95 15:37:44 -0600 Tim Bunce writes: |> Let me assure you that while the new DBI will include access to all ODBC |> functions you will almost always be able to use high-level utility |> methods, including the familiar: |> |> $sth => $dbh->prepare("..."); |> $sth->execute; |> @row = $sth->fetchrow; |> |> Native ODBC is ugly and verbose. |> Reminds me of COBOL :-( Mike mhm@austin.ibm.com ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <02194-2@oink>; Tue, 19 Dec 1995 23:02:48 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819413952:19812:1; Tue, 19 Dec 95 22:59:12 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa18647; 19 Dec 95 22:57 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id OAA05043 for dbi-dev-real; Tue, 19 Dec 1995 14:24:20 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from coffee.stanford.edu (coffee.Stanford.EDU [36.60.0.11]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id OAA05039 for ; Tue, 19 Dec 1995 14:24:19 -0800 Received: from isl.Stanford.EDU (isl [36.60.0.10]) by coffee.stanford.edu (8.7.1/8.7.1) with ESMTP id OAA03410; Tue, 19 Dec 1995 14:24:29 -0800 (PST) From: "Sanford M. Dickert" Received: (from ncc1701a@localhost) by isl.Stanford.EDU (8.7.1/8.7.1) id OAA02872; Tue, 19 Dec 1995 14:24:29 -0800 (PST) Date: Tue, 19 Dec 1995 14:24:29 -0800 (PST) Message-Id: <199512192224.OAA02872@isl.Stanford.EDU> To: Davide.Migliavacca@inferentia.it Subject: RE: Trying to install DBD::Oracle Cc: dbi-dev@fugue.com Question - I utilized Massimo's suggestion (thanks again) but I found another error in the make when I compiled. The error is as follows: ------------------------------------------------------------------------ LD_RUN_PATH="" gcc -o blib/auto/DBD/Oracle/Oracle.so -G Oracle.o dbdimp.o -L/opt/oracle/v7.1.6/lib /opt/oracle/v7.1.6/lib/__fstd.o /opt/oracle/v7.1.6/lib/libocic.a /opt/oracle/v7.1.6/lib/osntab.o -lsqlnet -lora -lsqlnet -lora -lnlsrtl -lcv6 -lcore -lnlsrtl -lcv6 -lcore `cat /opt/oracle/v7.1.6/rdbms/lib/sysliblist` -R /opt/SUNWcluster/lib -Y P,/opt/SUNWspro/lib:/usr/openwin/lib:/opt/SUNWcluster/lib:/usr/ccs/lib:/usr/lib -Qy -lc /opt/oracle/v7.1.6/lib/crtn.o -lcore -lsocket -lnsl -lm -ldl -laio gcc: P,/opt/SUNWspro/lib:/usr/openwin/lib:/opt/SUNWcluster/lib:/usr/ccs/lib:/usr/lib: No such file or directory gcc: unrecognized option `-Y' make: *** [blib/auto/DBD/Oracle/Oracle.so] Error 1 ------------------------------------------------------------------------ On line 250 of the Makefile, it reads: LDSTRING= -R /opt/SUNWcluster/lib -Y P,$(LD_LIBRARY_PATH):/opt/SUNWcluster/lib:/usr/ccs/lib:/usr/lib -Qy -lc $(COMPOBJ)/crtn.o Is this correct? Sanford Dickert ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <02743-7@oink>; Wed, 20 Dec 1995 00:06:54 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819417921:05228:3; Wed, 20 Dec 95 00:05:21 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa04909; 20 Dec 95 0:04 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id PAA05491 for dbi-users-real; Tue, 19 Dec 1995 15:33:17 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sultan.hsii.ccare.com ([192.101.181.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id PAA05486 for ; Tue, 19 Dec 1995 15:33:13 -0800 Received: from msgw.hsii.ccare.com by SULTAN.HSII.CCARE.COM (PMDF V5.0-5 #4597) id <01HZ004G0U6O00069W@SULTAN.HSII.CCARE.COM> for dbi-users@fugue.com; Tue, 19 Dec 1995 17:32:58 -0600 (CST) Received: by msgw.hsii.ccare.com with Microsoft Mail id <30D74BB3@msgw.hsii.ccare.com>; Tue, 19 Dec 1995 17:33:07 -0600 (CST) Date: Tue, 19 Dec 1995 17:33:00 -0600 (CST) From: "Li, Alexander" Subject: Where is "imp_fbh_t" defined? To: DBI User List Message-id: <30D74BB3@msgw.hsii.ccare.com> X-Mailer: Microsoft Mail V3.0 Content-transfer-encoding: 7BIT Encoding: 23 TEXT Hello, Anyone know where is "imp_fbh_t" defined? I got this when I try to make DBD::Oracle: --- gh753> make cc -c -DDEBUGGING -O -Kpic -I/usr/local/lib/perl5/sun4-solaris/CORE -I/sparcy/o racle/rdbms/demo -I/sparcy/oracle/rdbms/demo -I/usr/local/lib/perl5/sun4-solari s/DBI Oracle.c "./dbdimp.h", line 47: syntax error before or at: imp_fbh_t "./dbdimp.h", line 47: cannot recover from previous errors cc: acomp failed for Oracle.c *** Error code 2 make: Fatal error: Command failed for target `Oracle.o' gh753> ---- Thanks again. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <04104-0@oink>; Wed, 20 Dec 1995 07:08:02 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819421059:16064:1; Wed, 20 Dec 95 00:57:39 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa15832; 20 Dec 95 0:57 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA05740 for dbi-users-real; Tue, 19 Dec 1995 16:25:25 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA05736 for ; Tue, 19 Dec 1995 16:25:20 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id aa16078; 20 Dec 95 0:09 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id ac12171; 20 Dec 95 0:05 GMT Received: from toad by oink with SMTP (PP) id <02685-0@oink>; Tue, 19 Dec 1995 23:53:04 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA03089; Tue, 19 Dec 1995 23:52:37 +0000 Date: Tue, 19 Dec 1995 23:52:37 +0000 From: Tim Bunce Message-Id: <9512192352.AA03089@toad.ig.co.uk> To: afc@mmts.eds.com Subject: Re: can DBI return rows as hashes keyed by column names? Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 977 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: afc@mmts.eds.com (Andy Cohan) > > >> From: Andy Cohan > >> > >> I'm trying to get our own perl5 DB class to use DBI for an Oracle back end > >> and I would like to support the Perl5 Sybase facility of optionally > >> returning rows as associative arrays keyed by the result column names (we > >> have used it extensively). > >> > >> Does DBI and DBD::Oracle support this? If not, is it planned? > > >>>>> "Tim" == Tim Bunce writes: > Tim> @fieldnames = (...); > > Tim> while(@hash{@fieldnames} = $sth->fetchrow) { > Tim> ... > Tim> } > Thanks! This is *definitely* helpful. Now, if the query is something like > "SELECT * FROM one_table, two_table ...", how can I know what order the > columns are coming back in so that I can set up the fieldnames array? If you don't know the names in advance why bother sticking them in a hash? Never mind. Use: @fieldnames = @{$sth->{'NAME'}}; Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04108-6@oink>; Wed, 20 Dec 1995 07:09:24 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819420759:26609:1; Wed, 20 Dec 95 00:52:39 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa26254; 20 Dec 95 0:51 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA05771 for dbi-dev-real; Tue, 19 Dec 1995 16:30:56 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id QAA05767 for ; Tue, 19 Dec 1995 16:30:51 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id aa17584; 20 Dec 95 0:20 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id aa14434; 20 Dec 95 0:16 GMT Received: from toad by oink with SMTP (PP) id <03074-0@oink>; Wed, 20 Dec 1995 00:16:32 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA03173; Wed, 20 Dec 1995 00:16:05 +0000 Date: Wed, 20 Dec 1995 00:16:05 +0000 From: Tim Bunce Message-Id: <9512200016.AA03173@toad.ig.co.uk> To: ncc1701a@isl.stanford.edu Subject: Re: New directions for DBI - Important Cc: dbi-dev@fugue.com X-Sun-Charset: US-ASCII content-length: 309 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: "Sanford M. Dickert" > > Tim - check out my last message - I think you have a typo on that particular line. I fixed it myself (changing -Y P, to -YP,). > > Sanford The typo is not mine but Oracle's (as usual). Makefile.PL reads that line from Oracle's proc.mk. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <04107-19@oink>; Wed, 20 Dec 1995 07:10:31 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819427969:14939:2; Wed, 20 Dec 95 02:52:49 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa14883; 20 Dec 95 2:52 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA06364 for dbi-users-real; Tue, 19 Dec 1995 18:23:35 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sultan.hsii.ccare.com ([192.101.181.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA06360 for ; Tue, 19 Dec 1995 18:23:33 -0800 Received: from msgw.hsii.ccare.com by SULTAN.HSII.CCARE.COM (PMDF V5.0-5 #4597) id <01HZ062EANJK0007DB@SULTAN.HSII.CCARE.COM>; Tue, 19 Dec 1995 20:23:07 -0600 (CST) Received: by msgw.hsii.ccare.com with Microsoft Mail id <30D77396@msgw.hsii.ccare.com>; Tue, 19 Dec 1995 20:23:18 -0600 (CST) Date: Tue, 19 Dec 1995 20:20:00 -0600 (CST) From: "Li, Alexander" Subject: DBD::Oracle failed on make test -- ora_login fail To: DBI User List Cc: Tim Bunce Message-id: <30D77396@msgw.hsii.ccare.com> X-Mailer: Microsoft Mail V3.0 Content-transfer-encoding: 7BIT Encoding: 78 TEXT Here's a little twist about my install process: I don't have enough disk space on the same machine for both Oracle and the works. So I NFS mounted a Oracle directory during the install. But I need Pro*C stuff ( I tried making using oracle.mk, didn't work, I still owe M. Balestra a reply). When Pro*C is installed and before relinking all the Oracle executables, I ran out of space on that machine too. Would this *twist* have any effect on the following problem? Anyway, I pass the make stage (w/ proc.mk) 'make test' give me this: ---- gh753> make test /usr/local/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/sun4-solaris -I/ust t/base..............ok t/main..............ok All tests successful. Files=2, Tests=6, 3 secs ( 1.63 cusr 0.45 csys = 2.08 cpu) /usr/local/bin/perl -I./blib -I./blib -I/usr/local/lib/perl5/sun4-solaris -I/usl Oraperl test application $Revision: 1.19 $ Oraperl Emulation Interface version 1.23 Oracle Driver 1.21 DBI Switch by Tim Bunce, version 1.48 Connecting to 'v718' (from command line, else uses ORACLE_SID/TWO_TASK) as 'next/nextstep' (via ORACLE_USERID environment var or default) ora_login: 1034: ORA-01034: ORACLE not available (DBD: login failed) ORACLE_SID or TWO_TASK possibly not right, or server not running. Could not connect to database. Test aborted. *** Error code 2 make: Fatal error: Command failed for target `test_dynamic' --------- I'm able to connect to 'v718' using the same user/passwd via SQLPlus. My config looks like: -------- gh753> perl -e 'use Config; print Config::myconfig()' Summary of my perl5 (patchlevel 1) configuration: Platform: osname=solaris, osver=2.4, archname=sun4-solaris uname='sunos gh753 5.4 generic_101945-10 sun4m sparc ' hint=recommended Compiler: cc='cc', optimize='-O' cppflags='-DDEBUGGING' ccflags ='-DDEBUGGING' ldflags ='' stdchar='unsigned char', d_stdstdio=define, usevfork=false voidflags=15, castflags=0, d_casti32=define, d_castneg=define intsize=4, alignbytes=8, usemymalloc=y, randbits=15 Libraries: so=so libpth=/lib /usr/lib /usr/ccs/lib /usr/local/lib libs=-lsocket -lnsl -ldl -lm -lc -lcrypt libc=/usr/lib/libc.so Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun= cccdlflags='-Kpic', ccdlflags=' ', lddlflags='-G' ---------- Thanks again. Alex Li ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <09131-14@oink>; Wed, 20 Dec 1995 16:52:10 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819477276:22200:2; Wed, 20 Dec 95 16:34:36 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa21946; 20 Dec 95 16:34 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA09208 for dbi-users-real; Wed, 20 Dec 1995 07:09:09 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id HAA09204 for ; Wed, 20 Dec 1995 07:08:49 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id ac20935; 20 Dec 95 15:04 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id aa23616; 20 Dec 95 15:01 GMT Received: from toad by oink with SMTP (PP) id <08245-0@oink>; Wed, 20 Dec 1995 14:40:14 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA05001; Wed, 20 Dec 1995 14:39:47 +0000 Date: Wed, 20 Dec 1995 14:39:47 +0000 From: Tim Bunce Message-Id: <9512201439.AA05001@toad.ig.co.uk> To: dbi-users@fugue.com, AlexLi@msmail.hsii.ccare.com Subject: Re: DBD::Oracle failed on make test -- ora_login fail X-Sun-Charset: US-ASCII content-length: 670 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: "Li, Alexander" > > > Connecting > to 'v718' (from command line, else uses ORACLE_SID/TWO_TASK) > as 'next/nextstep' (via ORACLE_USERID environment var or default) > ora_login: 1034: ORA-01034: ORACLE not available (DBD: login failed) > > ORACLE_SID or TWO_TASK possibly not right, or server not running. > > I'm able to connect to 'v718' using the same user/passwd via SQLPlus. Try not specifying v718 on the command line, i.e., let DBD::Oracle use the env vars in the same way that SQLplus does. I'd guess that v718 is an Sql*NET alias - which don't work currently since DBD::Oracle can't tell that it's an alias. Tim. ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <12069-15@oink>; Thu, 21 Dec 1995 07:08:21 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819493693:28882:0; Wed, 20 Dec 95 21:08:13 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa28103; 20 Dec 95 21:07 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA09860 for dbi-users-real; Wed, 20 Dec 1995 10:53:04 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from sultan.hsii.ccare.com ([192.101.181.3]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id KAA09856 for ; Wed, 20 Dec 1995 10:53:01 -0800 Received: from msgw.hsii.ccare.com by SULTAN.HSII.CCARE.COM (PMDF V5.0-5 #4597) id <01HZ14M6NIJK00078Y@SULTAN.HSII.CCARE.COM>; Wed, 20 Dec 1995 12:52:36 -0600 (CST) Received: by msgw.hsii.ccare.com with Microsoft Mail id <30D85B78@msgw.hsii.ccare.com>; Wed, 20 Dec 1995 12:52:40 -0600 (CST) Date: Wed, 20 Dec 1995 12:51:00 -0600 (CST) From: "Li, Alexander" Subject: Re: DBD::Oracle failed on make test -- ora_login fail To: "Sanford M. Dickert" Cc: DBI User List , Tim Bunce Message-id: <30D85B78@msgw.hsii.ccare.com> X-Mailer: Microsoft Mail V3.0 Content-transfer-encoding: 7BIT Encoding: 31 TEXT I just realize one thing that I'm doing is *very* different from everyone else. That is I'm running on a true Oracle client machine. i.e. I have no Oracle server installed on the client machine. Thanks to Tim, I found out sqlplus works only like sqlplus next/next@v718 but never just "sqlplus" even my ORACLE_SID is set. After talking to someone who is more an expert in Oracle, you cannot run sqlplus on a client machine w/o a "fully qualified" name and connect string. I'm going to move all the tools to the Oracle server box for now. But is there anyone out there tried running the DBD tool on a client machine? Thanks for all who have help and responed to my messages. ---------- From: Sanford M. Dickert To: AlexLi Subject: Re: DBD::Oracle failed on make test -- ora_login fail Date: Wednesday, December 20, 1995 9:25AM Question - I had the same problem but it was solved once I used the correct database SID. Also, do you have the server running? (I know - silly question, but I didn't the first time I ran Tim's script). Sanford ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12068-16@oink>; Thu, 21 Dec 1995 07:08:25 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819483178:09918:2; Wed, 20 Dec 95 18:12:58 GMT Received: from relay-1.mail.demon.net ([158.152.1.140]) by punt-1.mail.demon.net id aa09856; 20 Dec 95 18:12 GMT Received: from toccata.fugue.com ([204.254.239.2]) by relay-1.mail.demon.net id aa16746; 20 Dec 95 18:12 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id GAA09151 for dbi-users-real; Wed, 20 Dec 1995 06:37:56 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from CHILD.PRODIGY.CS.CMU.EDU (CHILD.PRODIGY.CS.CMU.EDU [128.2.242.220]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id GAA09147 for ; Wed, 20 Dec 1995 06:37:54 -0800 Received: from [127.0.0.1] by CHILD.PRODIGY.CS.CMU.EDU id aa05330; 20 Dec 95 9:37 EST Reply-To: desilva+@cmu.edu To: dbi-users@fugue.com Subject: Keeping an Oracle connection open Date: Wed, 20 Dec 1995 09:37:34 -0500 Message-ID: <5328.819470254@CHILD.PRODIGY.CS.CMU.EDU> From: Rujith de Silva -----BEGIN PGP SIGNED MESSAGE----- Is there any way to have an exiting process NOT close an open lda? I have a WWW server that gets an Oracle connection (lda), then accepts an HTTP request, forks off a child, and lets the child use the lda. The trouble is, after processing the request, the child automatically closes the lda when exiting (its DESTROY method is probably getting called). This makes the lda unusable for the parent as well. Is there any way to get around this? Versions: SunOS 4.1.3_U1, perl 5.002 beta, DBI 0.65, DBD::Oracle 0.27, Oraperl 1.23 Later, Rujith. Rujith de Silva. 1-(412) 268-3620. desilva+@cmu.edu http://www.cs.cmu.edu/~desilva/ PGP public key available by finger. -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMNgfqYkharavxuYRAQFUaAQAsDHNLoCr+bMLbunymfJ+I47gsuRHNbFD Ucsz1atvsZ/UDjHy6syoNMj4IoDMm8o3ruGdJUFLktZ5HG3+4fGEHO78H6J+MilK oz/Kd7tCYNXrXnSjEg4qYe4GEFX/rYkPK9b5eO54yBfuALtqf6FZrABZjbHnBp+c MPrHAQTg84s= =qZxC -----END PGP SIGNATURE----- ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12260-21@oink>; Thu, 21 Dec 1995 07:20:40 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819501050:10033:1; Wed, 20 Dec 95 23:10:50 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa09786; 20 Dec 95 23:10 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA10433 for dbi-users-real; Wed, 20 Dec 1995 13:11:39 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from colossus.cse.psu.edu (root@colossus.cse.psu.edu [130.203.1.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id NAA10429 for ; Wed, 20 Dec 1995 13:11:37 -0800 Received: from localhost by colossus.cse.psu.edu with SMTP id <78355>; Wed, 20 Dec 1995 16:11:31 -0500 X-Mailer: exmh version 1.6.1 5/23/95 To: "Li, Alexander" cc: dbi-users@fugue.com Subject: Re: DBD::Oracle failed on make test -- ora_login fail In-reply-to: Your message of "Wed, 20 Dec 1995 13:51:00 EST." <30D85B78@msgw.hsii.ccare.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 20 Dec 1995 16:10:21 -0500 From: John D Groenveld Message-Id: <95Dec20.161131est.78355@colossus.cse.psu.edu> > but never just "sqlplus" even my ORACLE_SID is set. After talking to > someone who > is more an expert in Oracle, you cannot run sqlplus on a client machine w/o > a "fully qualified" name and connect string. Wait, now you have me confused. With SQLNet V1, there was a neat variable TWO_TASK that would let you do this. TWO_TASK has gone a way with SQLNet V2 because you can now have aliases which specify the driver, host, and database name. Perhaps Tim should add "which version of SQLNet" to the vital information needed when posting a problem. Then again, its been a long time since I used Oracle RDBMS v6 and SQLNetv1. :) > I'm going to move all the tools to the Oracle server box for now. But is > there anyone > out there tried running the DBD tool on a client machine? Yes, what does the below code give you on your client? I'm assuming you are using SQLNetV2, so the only Oracle environment variable you need is TNS_ADMIN. What I plan to do is set up my Solaris X86 machine as my webserver. I'll install Oracle on the PC and then remove it after DBPerl is working. Then, I'll have my CGI scripts on the X86 machine use SQLNet2 to access my Oracle SPARCServers. Good luck, John groenvel@cse.psu.edu #!/usr/bin/perl use Oraperl; $lda = &ora_login("", "scott\@SQLNET2Alias","tiger") || die "ora_login: $ora_errno: $ora_errstr\n"; $csr = &ora_open($lda,"select sysdate from dual"); while (@row = &ora_fetch($csr)) { print @row,"\n"; } &ora_close($csr); &ora_logoff($lda); ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12260-22@oink>; Thu, 21 Dec 1995 07:20:49 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819501254:11815:1; Wed, 20 Dec 95 23:14:14 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa11643; 20 Dec 95 23:13 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA10564 for dbi-users-real; Wed, 20 Dec 1995 13:37:16 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA10560 for ; Wed, 20 Dec 1995 13:37:14 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id NAA01182; Wed, 20 Dec 1995 13:33:02 -0800 Message-Id: <199512202133.NAA01182@voro.lbl.gov> X-Mailer: exmh version 1.6.2 7/18/95 To: "Li, Alexander" cc: "Sanford M. Dickert" , DBI User List , Tim Bunce Subject: Re: DBD::Oracle failed on make test -- ora_login fail In-reply-to: Your message of "Wed, 20 Dec 1995 12:51:00 CST." <30D85B78@msgw.hsii.ccare.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 20 Dec 1995 13:33:02 -0800 From: Mark Dedlow > I just realize one thing that I'm doing is *very* different from everyone > else. That is > I'm running on a true Oracle client machine. i.e. I have no Oracle server > installed > on the client machine. Thanks to Tim, I found out sqlplus works only like Actually, I imagine many people do this... I know I do. > sqlplus next/next@v718 > > but never just "sqlplus" even my ORACLE_SID is set. Think about it for a moment. Your ORACLE_SID is just a simple string, like "payroll" perhaps. How could sqlplus possibly know where this database resides? On what network host is it? sqlplus needs more info... but first let's step back a bit. Oracle client tools (e.g. sqlplus) use the combination of ORACLE_HOME and ORACLE_SID environment variables to locate a shared memory segment set up by the Oracle server. Running in this mode, the client-server communication is entirely via IPC (Inter Process Communication, e.g. shared memory, semaphores, etc.) If the server is on another machine, clearly the processes can't communicate via IPC. Accordingly, the *environment variable* ORACLE_SID has no real meaning on client that's on a different machine than the server. (Note that ORACLE_SID is often used to refer to the actual environment variable, and often just as a casual "name" of a database.) If the server is on a different machine, there needs to be a different syntax, which with sqlnet v1 is: username/password@T:hostname:ORACLE_SID Where ORACLE_SID is the ORACLE_SID *on* the remote machine. Note that what you are really doing in not connecting to the database server, but connecting to the sqlnet daemon, which is of course running on the same machine as the server. The sqlnet daemon uses the ORACLE_SID *string* you pass it to "find" the database on the remote server, much the same way sqlplus would find the server running on the same machine (i.e. IPC). If you are using sqlnet v2, the syntax is simpler looking, because the hostname, network protocol, and remote database are hidden away in a configuration file. E.g.: username/password@payroll There is a file (/etc/tnsnames.ora, or /var/opt/oracle/tnsnames.ora) that has something like this: payroll = (DESCRIPTION = (ADDRESS_LIST = (ADDRESS = (PROTOCOL = TCP) (Host = hostname.net) (Port = 1521) ) (CONNECT_DATA = (SID = payroll)) ) Finally, on a client, there is an env variable to sort of serve the function of ORACLE_SID for remote connections. It is TWO_TASK. Thus, this connects me to a remote database: % setenv TWO_TASK T:hostname:ORACLE_SID % sqlplus username/password Note that if you have *both* local and remote databases, and you have ORACLE_SID *and* TWO_TASK set, and you don't specify a fully qualified connect string on the command line, TWO_TASK takes precedence over ORACLE_SID (i.e. you get connected to remote system). Hope this all makes some sense and helps. Mark ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12324-0@oink>; Thu, 21 Dec 1995 07:21:09 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819530325:26965:1; Thu, 21 Dec 95 07:18:45 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa26836; 21 Dec 95 7:18 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id UAA11769 for dbi-users-real; Wed, 20 Dec 1995 20:16:26 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id UAA11765 for ; Wed, 20 Dec 1995 20:16:24 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id UAA02303; Wed, 20 Dec 1995 20:12:12 -0800 Message-Id: <199512210412.UAA02303@voro.lbl.gov> X-Mailer: exmh version 1.5.3 12/28/94 To: John D Groenveld cc: "Li, Alexander" , dbi-users@fugue.com Subject: Re: DBD::Oracle failed on make test -- ora_login fail In-reply-to: Your message of "Wed, 20 Dec 1995 16:10:21 EST." <95Dec20.161131est.78355@colossus.cse.psu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 20 Dec 1995 20:12:12 -0800 From: Mark Dedlow > Wait, now you have me confused. With SQLNet V1, there was a neat variable > TWO_TASK that would let you do this. TWO_TASK has gone a way with SQLNet V2 It's still there on SunOS and Solaris, and probably other flavors of unix. Just set TWO_TASK to your sqlnet v2 alias. On MS Win and Macs, there are the LOCAL/REMOTE aliases in the CONFIG.ORA that have similar effect. > Then again, its been a long time since I used Oracle RDBMS v6 and SQLNetv1 And Oracle will stop supporting it pretty soon. It won't ship with 7.3 and forward. v2 has many advantages too. > using SQLNetV2, so the only Oracle environment variable you need is TNS_ADMIN Actually, you don't *require* that either. It's only function is to tell a tool where to find config files (tnsnames.ora, etc.) if they are not in any of the standard places. Mark ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <12324-12@oink>; Thu, 21 Dec 1995 07:25:29 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819508208:05301:1; Thu, 21 Dec 95 01:10:08 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa04973; 21 Dec 95 1:09 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id QAA11183 for dbi-users-real; Wed, 20 Dec 1995 16:23:57 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail1.is.net (root@mail1.is.net [198.69.24.4]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id QAA11179 for ; Wed, 20 Dec 1995 16:23:55 -0800 Received: from stimpy.encompass.is.net ([204.180.18.228]) by mail1.is.net (8.6.11/8.6.12) with SMTP id TAA05925; Wed, 20 Dec 1995 19:15:19 -0500 Date: Wed, 20 Dec 1995 19:15:19 -0500 Message-Id: <199512210015.TAA05925@mail1.is.net> X-Sender: terry@yakko X-Mailer: Windows Eudora Light Version 1.5.2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: dbi-users@fugue.com From: Terry Greenlaw Subject: Re: DBD::Oracle failed on make test -- ora_login fail Cc: Mark Dedlow At 01:33 PM 12/20/95 -0800, you wrote: > [sql*net info deleted] >Finally, on a client, there is an env variable to sort of serve the >function of ORACLE_SID for remote connections. It is TWO_TASK. >Thus, this connects me to a remote database: > >% setenv TWO_TASK T:hostname:ORACLE_SID >% sqlplus username/password > >Note that if you have *both* local and remote databases, and you >have ORACLE_SID *and* TWO_TASK set, and you don't specify a fully >qualified connect string on the command line, TWO_TASK takes precedence >over ORACLE_SID (i.e. you get connected to remote system). > >Hope this all makes some sense and helps. > >Mark ORACLE_SID is really unnecessary to set since TWO_TASK provides the same functionality in addition to allowing remote connections. TWO_TASK = P:sid will use the pipe driver for local connections SQL*Net 1.x TWO_TASK = T:machine:sid will use TCP/IP (or D for DECNET, etc.) for remote SQL*Net 1.x conn. TWO_TASK = dbname will use the info stored in the SQL*Net 2.x configuration file for local or remote connections. ORACLE_HOME can also be left unset if you aren't using any of Oracle's executables, but error messages may not display, and the current DBD::Oracle generates a warning message (can this be disabled Tim ?). I've found discouraging the use of ORACLE_SID makes it easier on the users to see what is going on. I just wish that TWO_TASK could be renamed, since it makes no sense to the end user, and doesn't have the ORACLE prefix on it. Hope this doesn't fog the issue at hand too much ;-} tog Terry Greenlaw terry@encompass.is.net Encompass Technologies (770) 591-8903 ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <18913-0@oink>; Thu, 21 Dec 1995 17:05:48 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819565506:19959:0; Thu, 21 Dec 95 17:05:06 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa19091; 21 Dec 95 17:04 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA14603 for dbi-users-real; Thu, 21 Dec 1995 05:23:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from colossus.cse.psu.edu (root@colossus.cse.psu.edu [130.203.1.2]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id FAA14599 for ; Thu, 21 Dec 1995 05:23:12 -0800 Received: from localhost by colossus.cse.psu.edu with SMTP id <78451>; Thu, 21 Dec 1995 08:23:12 -0500 X-Mailer: exmh version 1.6.1 5/23/95 To: Mark Dedlow cc: dbi-users@fugue.com Subject: Re: DBD::Oracle failed on make test -- ora_login fail In-reply-to: Your message of "Wed, 20 Dec 1995 23:12:12 EST." <199512210412.UAA02303@voro.lbl.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 21 Dec 1995 08:21:51 -0500 From: John D Groenveld Message-Id: <95Dec21.082312est.78451@colossus.cse.psu.edu> > > Wait, now you have me confused. With SQLNet V1, there was a neat variable > > TWO_TASK that would let you do this. TWO_TASK has gone a way with SQLNet V2 > > It's still there on SunOS and Solaris, and probably other flavors of unix. > Just set TWO_TASK to your sqlnet v2 alias. On MS Win and Macs, there > are the LOCAL/REMOTE aliases in the CONFIG.ORA that have similar effect. > > > Then again, its been a long time since I used Oracle RDBMS v6 and SQLNetv1 > > And Oracle will stop supporting it pretty soon. It won't ship with > 7.3 and forward. v2 has many advantages too. All the documentation I've read is that TWO_TASK is a V1 variable. Since V1 is going away and TWO_TASK is non-descriptive anyway, I recommend that people phase out its use. Nice to see Oracle try to stay backwards compatable, though. > > using SQLNetV2, so the only Oracle environment variable you need is TNS_ADMIN > Actually, you don't *require* that either. It's only function is to tell > a tool where to find config files (tnsnames.ora, etc.) if they are not in > any of the standard places. Be careful with this one. Under Solaris 2.4 and Oracle 7.1.6 (but not 7.1.4), Oracle tools are shipped to check /var/opt/oracle for SQLNetV2 config files, yet still fail when TNS_ADMIN is not set. In fact, when you truss these tools, you see it stat /var/opt/oracle and still die. Oracle Support's response was that TNS_ADMIN is a required variable. :) ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <20670-2@oink>; Fri, 22 Dec 1995 07:06:23 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819567309:03937:0; Thu, 21 Dec 95 17:35:09 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa03601; 21 Dec 95 17:34 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id FAA14611 for dbi-users-real; Thu, 21 Dec 1995 05:26:03 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from actcom.co.il (root@actcom.co.il [192.114.47.1]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id FAA14607 for ; Thu, 21 Dec 1995 05:25:55 -0800 Received: from astea.actcom.co.il by actcom.co.il with SMTP (8.6.12/actcom-0.1) id PAA26621 for ; Thu, 21 Dec 1995 15:27:08 +0200 (rfc931-sender: p1.haifa1.actcom.co.il [192.114.47.31]) Message-ID: <30D96124.49C3@actcom.co.il> Date: Thu, 21 Dec 1995 15:29:08 +0200 From: Leonid Borisov Organization: Astea Israel X-Mailer: Mozilla 2.0b3 (Win95; I) MIME-Version: 1.0 To: dbi-users@fugue.com Subject: (no subject) X-URL: http://www.hermetica.com/technologia/DBI/doc/faq.html Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit unsubscibe Leonid Borisov ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <20670-7@oink>; Fri, 22 Dec 1995 07:06:56 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819572165:12119:1; Thu, 21 Dec 95 18:56:05 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa11591; 21 Dec 95 18:54 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA15428 for dbi-users-real; Thu, 21 Dec 1995 09:58:39 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA15424 for ; Thu, 21 Dec 1995 09:58:37 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id JAA04819; Thu, 21 Dec 1995 09:57:46 -0800 Message-Id: <199512211757.JAA04819@voro.lbl.gov> X-Mailer: exmh version 1.6.2 7/18/95 To: Alligator Descartes cc: groenvel@cse.psu.edu (John D Groenveld), dbi-users@fugue.com Subject: Re: DBD::Oracle failed on make test -- ora_login fail In-reply-to: Your message of "Thu, 21 Dec 1995 15:01:54 GMT." <199512211501.PAA20918@fruitbat.mcqueen.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 21 Dec 1995 09:57:46 -0800 From: Mark Dedlow > Yeah? I've not had any hassle with 7.1.4/6 on Solaris2.4 for finding the > V2 config files. I seem to recall it searches $ORACLE_HOME/network/admin > and /etc by default. /var/opt/oracle was where the V1 stuff went, like > the oratab file............... > > I just leave my stuff in /etc usually and symlink to $ORACLE_HOME/network/adm in > so's we can upgrade without any real hassles........ Yup, either sym link from /etc (4.1.x or /var/opt/oracle on SVR4) to $ORACLE_HOME/network/admin, or just make sure there are no bogus (e.g. samples) copies there, since the search order is: $HOME/.tnsnames.ora $ORACLE_HOME/network/admin/tnsnames.ora /var/opt/oracle/tnsnames.ora if TNS_ADMIN is not set. Mark ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <20670-10@oink>; Fri, 22 Dec 1995 07:07:56 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819575512:08202:2; Thu, 21 Dec 95 19:51:52 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-2.mail.demon.net id aa08059; 21 Dec 95 19:51 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA15711 for dbi-users-real; Thu, 21 Dec 1995 10:51:32 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from localhost (mellon@localhost) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA15705; Thu, 21 Dec 1995 10:51:30 -0800 Message-Id: <199512211851.KAA15705@toccata.fugue.com> To: Leonid Borisov cc: dbi-users@fugue.com Subject: Administrivia requests... In-reply-to: Your message of "Thu, 21 Dec 1995 15:29:08 +0200." <30D96124.49C3@actcom.co.il> Date: Thu, 21 Dec 1995 10:51:29 -0800 From: Ted Lemon To unsubscribe to the mailing list, please visit the DBI mailing list subscription URL at http://www.fugue.com/dbi. _MelloN_ ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21063-7@oink>; Fri, 22 Dec 1995 07:15:21 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819569972:09184:2; Thu, 21 Dec 95 18:19:32 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa08638; 21 Dec 95 18:19 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA15282 for dbi-users-real; Thu, 21 Dec 1995 09:10:53 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id JAA15278 for ; Thu, 21 Dec 1995 09:10:51 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id JAA04515; Thu, 21 Dec 1995 09:10:05 -0800 Message-Id: <199512211710.JAA04515@voro.lbl.gov> X-Mailer: exmh version 1.6.2 7/18/95 To: John D Groenveld cc: dbi-users@fugue.com Subject: Re: DBD::Oracle failed on make test -- ora_login fail In-reply-to: Your message of "Thu, 21 Dec 1995 08:21:51 EST." <95Dec21.082312est.78451@colossus.cse.psu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 21 Dec 1995 09:10:05 -0800 From: Mark Dedlow > Be careful with this one. Under Solaris 2.4 and Oracle 7.1.6 (but not 7.1.4), > Oracle tools are shipped to check /var/opt/oracle for SQLNetV2 config files, > yet still fail when TNS_ADMIN is not set. In fact, when you truss these tools > you see it stat /var/opt/oracle and still die. Oracle Support's response was > that TNS_ADMIN is a required variable. :) My experience is different. I've been supporting Oracle on Solaris continuously thru Solaris 2.1-2.4 and Oracle 7.0-7.2 (and all in between versions) on many machines with dozens of applications, and I rarely set TNS_ADMIN (only where I don't want to use /var/opt/oracle). Here is an excerpt from a truss I just did: % unsetenv TNS_ADMIN % truss sqlplus username/password@remotesystem access("/var/opt/oracle/tnsnames.ora", 0) = 0 stat("/var/opt/oracle/tnsnames.ora", 0xDFFFC7E0) = 0 open("/var/opt/oracle/tnsnames.ora", O_RDONLY) = 6 fcntl(6, F_SETFD, 0x00000001) = 0 fstat(6, 0xDFFFC558) = 0 read(6, " F N D D E V = \n ".., 8192) = 5240 Clearly, it's working just fine. In any case, we're getting off into territory probably not of interest to dbi-users, since TNS_ADMIN isn't used by it (is it?). Mark ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21063-13@oink>; Fri, 22 Dec 1995 07:16:18 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819571585:22045:2; Thu, 21 Dec 95 18:46:25 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa21754; 21 Dec 95 18:46 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id JAA15301 for dbi-users-real; Thu, 21 Dec 1995 09:18:36 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from mail-relay1 (mail-relay1.cis.yale.edu [130.132.21.199]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id JAA15297 for ; Thu, 21 Dec 1995 09:18:34 -0800 Received: from capitoline.cis.yale.edu by mail-relay1 with SMTP id AA10589 (5.67a/IDA-1.5 for ); Thu, 21 Dec 1995 12:17:27 -0500 Received: from morpheus.cis.yale.edu (morpheus [130.132.143.248]) by capitoline.cis.yale.edu (8.6.12/8.6.12) with ESMTP id MAA08601 for ; Thu, 21 Dec 1995 12:18:42 -0500 Received: (from rivory@localhost) by morpheus.cis.yale.edu (8.6.12/8.6.12) id MAA08754; Thu, 21 Dec 1995 12:18:40 -0500 Date: Thu, 21 Dec 1995 12:18:40 -0500 (EST) From: Royal Hansen X-Sender: rivory@morpheus To: dbi-users@fugue.com Subject: unsubscribe In-Reply-To: <30D96124.49C3@actcom.co.il> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII unsubscibe ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21404-0@oink>; Fri, 22 Dec 1995 07:16:21 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819571303:19789:1; Thu, 21 Dec 95 18:41:43 GMT Received: from relay-1.mail.demon.net ([158.152.1.140]) by punt-1.mail.demon.net id aa19553; 21 Dec 95 18:41 GMT Received: from toccata.fugue.com ([204.254.239.2]) by relay-1.mail.demon.net id aa10446; 21 Dec 95 18:40 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA14857 for dbi-users-real; Thu, 21 Dec 1995 07:07:43 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA14853 for ; Thu, 21 Dec 1995 07:07:40 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id PAA20918; Thu, 21 Dec 1995 15:01:56 GMT From: Alligator Descartes Message-Id: <199512211501.PAA20918@fruitbat.mcqueen.com> Subject: Re: DBD::Oracle failed on make test -- ora_login fail To: John D Groenveld Date: Thu, 21 Dec 1995 15:01:54 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: <95Dec21.082312est.78451@colossus.cse.psu.edu> from "John D Groenveld" at Dec 21, 95 08:21:51 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1555 > > And Oracle will stop supporting it pretty soon. It won't ship with > > 7.3 and forward. v2 has many advantages too. > > All the documentation I've read is that TWO_TASK is a V1 variable. Since V1 is > going away and TWO_TASK is non-descriptive anyway, I recommend that people > phase out its use. Nice to see Oracle try to stay backwards compatable, though. Nup, TWO_TASK does still seem to work under V2.............Obvious variable name to pick, eh?! > > Actually, you don't *require* that either. It's only function is to tell > > a tool where to find config files (tnsnames.ora, etc.) if they are not in > > any of the standard places. > > Be careful with this one. Under Solaris 2.4 and Oracle 7.1.6 (but not 7.1.4), > Oracle tools are shipped to check /var/opt/oracle for SQLNetV2 config files, > yet still fail when TNS_ADMIN is not set. In fact, when you truss these tools, > you see it stat /var/opt/oracle and still die. Oracle Support's response was > that TNS_ADMIN is a required variable. :) Yeah? I've not had any hassle with 7.1.4/6 on Solaris2.4 for finding the V2 config files. I seem to recall it searches $ORACLE_HOME/network/admin and /etc by default. /var/opt/oracle was where the V1 stuff went, like the oratab file............... I just leave my stuff in /etc usually and symlink to $ORACLE_HOME/network/admin so's we can upgrade without any real hassles........ -- Alligator Descartes | "...Nil posse creari De nilo" descarte@hermetica.com | -- Lucretius, De Rerum Natura I, l.155 http://www.hermetica.com | ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21063-23@oink>; Fri, 22 Dec 1995 07:18:19 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819575145:20598:2; Thu, 21 Dec 95 19:45:45 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id ab20004; 21 Dec 95 19:45 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id KAA15715 for dbi-users-real; Thu, 21 Dec 1995 10:51:35 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from netserve.lasc.lockheed.com (netserve.lasc.lockheed.com [134.5.45.1]) by toccata.fugue.com (8.6.9/8.6.11) with SMTP id KAA15706 for ; Thu, 21 Dec 1995 10:51:30 -0800 Message-Id: <199512211851.KAA15706@toccata.fugue.com> Received: from mip.lasc.lockheed.com by netserve.lasc.lockheed.com with SMTP (1.37.109.4/16.2) id AA20424; Thu, 21 Dec 95 13:49:56 -0500 Received: by mip.lasc.lockheed.com (1.38.193.4/16.2) id AA06477; Thu, 21 Dec 1995 13:49:25 -0500 Date: Thu, 21 Dec 1995 13:49:25 -0500 From: Terry Greenlaw To: dbi-users@fugue.com Subject: Re: DBD::Oracle failed on make test -- ora_login fail > From owner-dbi-users@fugue.com Thu Dec 21 13:24 EST 1995 > Received: from marble.litc.lockheed.com by mip.lasc.lockheed.com with SMTP > (1.38.193.4/16.2) id AA04812; Thu, 21 Dec 1995 13:24:07 -0500 > Return-Path: > Received: from toccata.fugue.com (toccata.fugue.com [204.254.239.2]) by marble.litc.lockheed.com (8.6.12/8.6.9) with ESMTP id KAA16842; Thu, 21 Dec 1995 10:32:53 -0800 > Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id HAA14857 for dbi-users-real; Thu, 21 Dec 1995 07:07:43 -0800 > Precedence: junk > Sender: owner-perldb-interest@fugue.com > Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id HAA14853 for ; Thu, 21 Dec 1995 07:07:40 -0800 > Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id PAA20918; Thu, 21 Dec 1995 15:01:56 GMT > From: Alligator Descartes > Message-Id: <199512211501.PAA20918@fruitbat.mcqueen.com> > Subject: Re: DBD::Oracle failed on make test -- ora_login fail > To: groenvel@cse.psu.edu (John D Groenveld) > Date: Thu, 21 Dec 1995 15:01:54 +0000 (GMT) > Cc: dbi-users@fugue.com > In-Reply-To: <95Dec21.082312est.78451@colossus.cse.psu.edu> from "John D Groenveld" at Dec 21, 95 08:21:51 am > X-Pants: ** The return address of this email should be: descarte@hermetica.com > X-Mailer: ELM [version 2.4 PL24] > Content-Type: text > Content-Length: 1555 > Status: R > > > And Oracle will stop supporting it pretty soon. It won't ship with > > > 7.3 and forward. v2 has many advantages too. > > > > All the documentation I've read is that TWO_TASK is a V1 variable. Since V1 is > > going away and TWO_TASK is non-descriptive anyway, I recommend that people > > phase out its use. Nice to see Oracle try to stay backwards compatable, though. > Nup, TWO_TASK does still seem to work under V2.............Obvious variable > name to pick, eh?! TWO_TASK is still the one to use until Oracle comes up with something a little more understandable like ORACLE_DB. Boring details of Oracle history follow.... On many platforms, Oracle can be configured in 2 seperate ways. The front and back end code can be linked together into one executable, but this has alot of problems with security since user applications are running DB kernel calls as the user. To solve this, the front and back end code were seperated. The front end is run in the user's process space, and the back end runs as a privledged user. The two tasks communicate through an O.S. specific IPC mechanism. On VMS, it's the mailbox driver; on Unix, it's the pipe driver (or fast pipe mechanism). The driver to use is stored in the environment variable TWO_TASK. This split also allows the front and back ends to reside on seperate machines enabling client-server computing. Many people inside Oracle believe the early adoption of this split was the primary reason that Oracle became #1. They weren't forced to redesign base components when client/server computing got hot. tog > -- > Alligator Descartes | "...Nil posse creari De nilo" > descarte@hermetica.com | -- Lucretius, De Rerum Natura I, l.155 > http://www.hermetica.com | Terry Greenlaw (on-site @ Lockheed Martin) Encompass Technologies z50816@mip.lasc.lockheed.com terry@encompass.is.net ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21404-13@oink>; Fri, 22 Dec 1995 07:18:51 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819577331:07980:1; Thu, 21 Dec 95 20:22:11 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa07747; 21 Dec 95 20:21 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id LAA16008 for dbi-users-real; Thu, 21 Dec 1995 11:32:28 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from cs.utah.edu (cs.utah.edu [128.110.4.21]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id LAA16004 for ; Thu, 21 Dec 1995 11:32:26 -0800 Received: from lal.cs.utah.edu by cs.utah.edu (8.6.12/utah-2.21-cs) id MAA09885; Thu, 21 Dec 1995 12:32:33 -0700 Received: by lal.cs.utah.edu (8.6.12/utah-2.15sun-leaf) id MAA23464; Thu, 21 Dec 1995 12:32:31 -0700 Date: Thu, 21 Dec 1995 12:32:30 -0700 (MST) From: Jonathan Olson To: dbi-users@fugue.com Subject: unsubscribe Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII unsubscribe me ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21063-35@oink>; Fri, 22 Dec 1995 07:21:23 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 819591840:25516:1; Fri, 22 Dec 95 00:24:00 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa24056; 22 Dec 95 0:23 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id NAA16623 for dbi-users-real; Thu, 21 Dec 1995 13:17:26 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id NAA16619 for ; Thu, 21 Dec 1995 13:17:20 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id VAA24741; Thu, 21 Dec 1995 21:11:54 GMT From: Alligator Descartes Message-Id: <199512212111.VAA24741@fruitbat.mcqueen.com> Subject: Re: TNS_ADMIN (was Re: DBD::Oracle failed on make test -- ora_login To: dedlow@voro.lbl.gov (Mark Dedlow) Date: Thu, 21 Dec 1995 21:11:53 +0000 (GMT) Cc: groenvel@cse.psu.edu, dbi-users@fugue.com In-Reply-To: <199512211952.LAA05507@voro.lbl.gov> from "Mark Dedlow" at Dec 21, 95 11:52:23 am X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2097 [cc'd back into the list] > > With SQL*DBA: Release 7.1.6.2.0 I can reproduce my error. My log book > > Oh, SQL*DBA!! I thought we were talking about *users*, > meaning to me, developers and end-users, not DBA's. > > SQL*DBA often behaves differently than all the other "client" > tools (sqlplus, reports, forms, etc, which generally behave > consistently as a family). Indeed so. In fact, against my logbook, SQL*DBA has almost *twice* the number of reported bugs and patches than any other product ( with the exception of Oracle Applications! ). > And speaking of "going away"... SQL*DBA is no longer officially > supported, at least according to some documentation. Other documentation > could be seen as conflicting (i.e. it says "use SQL*DBA", and they > ship it. But then I just installed Forms 2.3 from the *latest* > version of Oracle Financials!!!!, so "support" is a pretty fuzzy term; > they are still shipping stuff you haven't been able to even *buy* for > years) Again, yes. Forms2.3 stopped shipping years back apart from on Financials... I'll tell you this, once we lost one tape from our Financials suite, coule we get Oracle to find a replacement for the Forms module? Uh-uh. Half the Oracle support guys didn't believe Financials ran with 2.3. The other half thought I was winding them up! 8-) > In any case, I advise against using SQL*DBA as the "reference" behavior... Agreed. It's about the most un-clienty app bundled with the RDBMS. > Mark > > PS. Have you tried Server Manager yet? ("next-generation" DBA tool) Complete and utter rubbish, IMHO. I ended up writing one of my own. If anyone's interested in seeing what stage it was at around 7 months ago when I last had a minute to work on it, they're welcome to the source code! I've also got one that runs under DBI and DBD::Oracle which may be more worthwhile? Anyone interested in pooling resource to complete the design and write? Tk-based? -- Alligator Descartes | "...Nil posse creari De nilo" descarte@hermetica.com | -- Lucretius, De Rerum Natura I, l.155 http://www.hermetica.com | ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <02553-20@oink>; Mon, 25 Dec 1995 07:10:34 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819656243:09176:1; Fri, 22 Dec 95 18:17:23 GMT Received: from frigate.doc.ic.ac.uk ([155.198.191.3]) by punt-2.mail.demon.net id as07847; 22 Dec 95 18:16 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Fri, 22 Dec 1995 03:15:33 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id SAA17415 for dbi-users-real; Thu, 21 Dec 1995 18:05:25 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from voro.lbl.gov (voro.lbl.gov [131.243.64.29]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id SAA17411 for ; Thu, 21 Dec 1995 18:05:22 -0800 Received: from voro (localhost [127.0.0.1]) by voro.lbl.gov (8.6.12/G) with ESMTP id SAA06873; Thu, 21 Dec 1995 18:04:32 -0800 Message-Id: <199512220204.SAA06873@voro.lbl.gov> X-Mailer: exmh version 1.6.2 7/18/95 To: Alligator Descartes cc: groenvel@cse.psu.edu, dbi-users@fugue.com Subject: Re: TNS_ADMIN (was Re: DBD::Oracle failed on make test -- ora_login In-reply-to: Your message of "Thu, 21 Dec 1995 21:11:53 GMT." <199512212111.VAA24741@fruitbat.mcqueen.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 21 Dec 1995 18:04:32 -0800 From: Mark Dedlow > From: Alligator Descartes > > PS. Have you tried Server Manager yet? ("next-generation" DBA tool) > > Complete and utter rubbish, IMHO. I ended up writing one of my own. If Hence my quotations around "next-generation" (those are Oracle's words). I think your description more accurate. > I've also got one that runs under DBI and DBD::Oracle which may be more > worthwhile? Anyone interested in pooling resource to complete the design > and write? Tk-based? I'd be curious what you've done in DBI, and might be able/interested to contribute. Mark ---+++--- Received: from punt2.demon.co.uk by oink with SMTP (PP) id <02553-27@oink>; Mon, 25 Dec 1995 07:11:56 +0000 Received: from punt-2.mail.demon.net by mailstore for dbperl@ig.co.uk id 819656245:09182:1; Fri, 22 Dec 95 18:17:25 GMT Received: from frigate.doc.ic.ac.uk ([155.198.191.3]) by punt-2.mail.demon.net id at07847; 22 Dec 95 18:16 GMT Received: from toccata.fugue.com by frigate.doc.ic.ac.uk with SMTP (PP); Fri, 22 Dec 1995 03:25:38 +0000 Received: (from daemon@localhost) by toccata.fugue.com (8.6.9/8.6.11) id RAA17314 for dbi-users-real; Thu, 21 Dec 1995 17:38:06 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.9/8.6.11) with ESMTP id RAA17310 for ; Thu, 21 Dec 1995 17:38:00 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id BAA25783; Fri, 22 Dec 1995 01:32:23 GMT From: Alligator Descartes Message-Id: <199512220132.BAA25783@fruitbat.mcqueen.com> Subject: Re: TNS_ADMIN (was Re: DBD::Oracle failed on make test -- To: terry@encompass.is.net (Terry Greenlaw) Date: Fri, 22 Dec 1995 01:32:22 +0000 (GMT) Cc: dbi-users@fugue.com In-Reply-To: <199512212344.SAA20787@mail1.is.net> from "Terry Greenlaw" at Dec 21, 95 06:44:58 pm X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1491 [Oracle DBA products are.....] > > descarte@hermetica.com wrote: > > > >Complete and utter rubbish, IMHO. I ended up writing one of my own. If > >anyone's interested in seeing what stage it was at around 7 months ago > >when I last had a minute to work on it, they're welcome to the source code! > > > >I've also got one that runs under DBI and DBD::Oracle which may be more > >worthwhile? Anyone interested in pooling resource to complete the design > >and write? Tk-based? > > Sounds great. DBI/TkPerl is the first tool that would actually make this a > fun project. I also think Severe Mangler, oops I mean Server Manager is > completely useless. Let me know what you've got so far. I have some stuff > from my days at Oracle that may come in handy. Right. Here's the plan. Over the period between Xmas and New Year, I'll put a WWW page up on Hermetica with the sources and stuff we'd like to see in it. If anyone's interested in doing a bit of it, we can arrange that between us and divvy up the work accordingly. Howzat sound? I'll also put the Pro*C code up there too since it was more developed, so we could get extra ideas from that, I suppose........ Course, the name "Severe Mangler" sounds just about right........I vote for it! > -- > Terry Greenlaw terry@encompass.is.net -- Alligator Descartes | "...Nil posse creari De nilo" descarte@hermetica.com | -- Lucretius, De Rerum Natura I, l.155 http://www.hermetica.com | ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15736-1@oink>; Thu, 28 Dec 1995 07:05:55 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820088469:11658:1; Wed, 27 Dec 95 18:21:09 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa11497; 27 Dec 95 18:20 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id XAA06407 for dbi-users-real; Tue, 26 Dec 1995 23:00:10 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from spsgate.sps.mot.com (spsgate.sps.mot.com [192.70.231.1]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id XAA06403 for ; Tue, 26 Dec 1995 23:00:08 -0800 Received: from mogate (mogate.sps.mot.com) by spsgate.sps.mot.com (4.1/SMI-4.1/Email 2.1 10/25/93) id AA23844 for dbi-users@fugue.com; Wed, 27 Dec 95 09:12:51 MST Received: from emailchd by mogate (4.1/SMI-4.1/Email-2.0) id AA17166; Wed, 27 Dec 95 09:12:46 MST Received: by sterling.sps.mot.com (SMI-8.6/SMI-SVR4) id JAA03271; Wed, 27 Dec 1995 09:10:36 -0700 Date: Wed, 27 Dec 1995 09:10:36 -0700 From: sosnicki@sterling.sps.mot.com (Jerry Sosnicki x5426) Message-Id: <199512271610.JAA03271@sterling.sps.mot.com> To: dbi-users@fugue.com X-Mailer: Siren Mail (Motif 2.02 95/05/15) Mime-Version: 1.0 Content-Id: <252_291_820080635_5@sterling> Content-Type: text/plain; charset="us-ascii" unsubscribe sosnicki@sterling.sps.mot.com ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15736-24@oink>; Thu, 28 Dec 1995 07:08:21 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820099552:05992:1; Wed, 27 Dec 95 21:25:52 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa05867; 27 Dec 95 21:25 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id CAA02176 for dbi-users-real; Wed, 27 Dec 1995 02:50:22 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from eve.mscs.mu.edu (eve.mscs.mu.edu [134.48.4.166]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id CAA02172 for ; Wed, 27 Dec 1995 02:50:20 -0800 Received: by eve.mscs.mu.edu (Smail3.1.28.1 #9) id m0tV1zB-0002DUC; Wed, 27 Dec 95 13:58 CST Message-Id: From: zobair@marque.mscs.mu.edu (Saima Zobair) Subject: unsubscribe To: dbi-users@fugue.com Date: Wed, 27 Dec 1995 13:58:05 -0600 (CST) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 38 unsubscribe me from the mailing list ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15736-26@oink>; Thu, 28 Dec 1995 07:08:36 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820099946:08833:0; Wed, 27 Dec 95 21:32:26 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa08365; 27 Dec 95 21:31 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id CAA02122 for dbi-users-real; Wed, 27 Dec 1995 02:22:52 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from coffee.stanford.edu (coffee.Stanford.EDU [36.60.0.11]) by toccata.fugue.com (8.6.12/8.6.11) with ESMTP id CAA02118 for ; Wed, 27 Dec 1995 02:22:51 -0800 Received: from isl.Stanford.EDU (isl [36.60.0.10]) by coffee.stanford.edu (8.7.1/8.7.1) with ESMTP id LAA25780 for ; Wed, 27 Dec 1995 11:35:38 -0800 (PST) From: "Sanford M. Dickert" Received: (from ncc1701a@localhost) by isl.Stanford.EDU (8.7.1/8.7.1) id LAA28342 for dbi-users@fugue.com; Wed, 27 Dec 1995 11:35:37 -0800 (PST) Date: Wed, 27 Dec 1995 11:35:37 -0800 (PST) Message-Id: <199512271935.LAA28342@isl.Stanford.EDU> To: dbi-users@fugue.com Subject: How about working with Oracle 7.2.2.3? I am in a quandry about whether or not I should upgrade to 7.2.2.3 if I want to use DBPerl. I currently have 7.1.6.2 and I am about to install one or the other. Is there any potential problems with using 7.2 with DBPerl? Thanks in advance. Sanford Dickert ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <15736-28@oink>; Thu, 28 Dec 1995 07:08:50 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820100081:09770:2; Wed, 27 Dec 95 21:34:41 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa09620; 27 Dec 95 21:34 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id CAA02182 for dbi-announce-real; Wed, 27 Dec 1995 02:51:23 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from eve.mscs.mu.edu (eve.mscs.mu.edu [134.48.4.166]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id CAA02178 for ; Wed, 27 Dec 1995 02:51:22 -0800 Received: by eve.mscs.mu.edu (Smail3.1.28.1 #9) id m0tV209-0002DUC; Wed, 27 Dec 95 13:59 CST Message-Id: From: zobair@marque.mscs.mu.edu (Saima Zobair) Subject: unsubscribe To: dbi-announce@fugue.com Date: Wed, 27 Dec 1995 13:59:05 -0600 (CST) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 46 unsubscribe me from the mailing list thanks ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <21175-9@oink>; Thu, 28 Dec 1995 16:10:25 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820159748:13373:0; Thu, 28 Dec 95 14:09:08 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa13035; 28 Dec 95 14:08 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id TAA03342 for dbi-users-real; Wed, 27 Dec 1995 19:31:06 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id TAA03338 for ; Wed, 27 Dec 1995 19:31:04 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id aa10044; 28 Dec 95 12:43 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id aa14579; 28 Dec 95 12:43 GMT Received: from toad by oink with SMTP (PP) id <20154-0@oink>; Thu, 28 Dec 1995 12:42:48 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA23504; Thu, 28 Dec 1995 12:42:13 +0000 Date: Thu, 28 Dec 1995 12:42:13 +0000 From: Tim Bunce Message-Id: <9512281242.AA23504@toad.ig.co.uk> To: dbi-users@fugue.com, ncc1701a@isl.stanford.edu Subject: Re: How about working with Oracle 7.2.2.3? X-Sun-Charset: US-ASCII content-length: 375 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: "Sanford M. Dickert" > > I am in a quandry about whether or not I should upgrade to 7.2.2.3 if I want > to use DBPerl. I currently have 7.1.6.2 and I am about to install one or > the other. Is there any potential problems with using 7.2 with DBPerl? > > Thanks in advance. > > Sanford Dickert Not that I'm aware of, off-hand. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <24167-5@oink>; Fri, 29 Dec 1995 07:06:27 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820175134:01028:0; Thu, 28 Dec 95 18:25:34 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa00875; 28 Dec 95 18:25 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id XAA03598 for dbi-users-real; Wed, 27 Dec 1995 23:03:13 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id XAA03588 for ; Wed, 27 Dec 1995 23:03:06 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id ac26809; 28 Dec 95 16:07 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id aa19659; 28 Dec 95 16:07 GMT Received: from toad by oink with SMTP (PP) id <20301-0@oink>; Thu, 28 Dec 1995 13:05:40 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA23561; Thu, 28 Dec 1995 13:05:12 +0000 Date: Thu, 28 Dec 1995 13:05:12 +0000 From: Tim Bunce Message-Id: <9512281305.AA23561@toad.ig.co.uk> To: rick@ccomp.com Subject: Re: DBI Question Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 1089 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: rick@ccomp.com (Rick Clark) > > Tim, > > Please excuse the direct mail. You probably are inundated with questions like > this. True. Ideally people should ask questions through the mailing lists that way others can help answer or learn from the answers of others. > Question: How stable is the Oracle DBI interface for perl 5? Last I heard, > and according to the README in ftp.demon.co.uk:/pub/perl/db, it's in the > Alpha stage. Is this information current? I need a perl/Oracle interface for > a production system. Should I use perl4 and oraperl at this time, or perl 5 > and DBI? Perl4 oraperl is stable (read 'not actively maintained') but is showing its age (it uses old OCI calls). The Perl5 DBI interface is unstable (read 'being actively changed') but the DBD::Oracle driver comes with an Oraperl emulation that is stable and as near complete as anyone is likely to care. Tim. p.s. For the record, I intend to start heavy development on the new ODBC based DBI mid Jan (approx). Expect lots of noise as we try to resolve design tradeoffs around that time. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05913-4@oink>; Mon, 1 Jan 1996 07:06:14 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820271409:09799:1; Fri, 29 Dec 95 21:10:09 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa09479; 29 Dec 95 21:09 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id MAA02655 for dbi-users-real; Fri, 29 Dec 1995 12:31:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id MAA02634 for ; Fri, 29 Dec 1995 12:30:11 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id ab23884; 29 Dec 95 20:29 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id af04188; 29 Dec 95 20:28 GMT Received: from toad by oink with SMTP (PP) id <28819-0@oink>; Fri, 29 Dec 1995 19:22:01 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA26334; Fri, 29 Dec 1995 19:21:31 +0000 Date: Fri, 29 Dec 1995 19:21:31 +0000 From: Tim Bunce Message-Id: <9512291921.AA26334@toad.ig.co.uk> To: dbi-users@fugue.com, desilva+@cmu.edu Subject: Re: Keeping an Oracle connection open X-Sun-Charset: US-ASCII content-length: 819 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Rujith de Silva > > -----BEGIN PGP SIGNED MESSAGE----- > > Is there any way to have an exiting process NOT close an open lda? > > I have a WWW server that gets an Oracle connection (lda), then accepts an > HTTP request, forks off a child, and lets the child use the lda. The > trouble is, after processing the request, the child automatically closes > the lda when exiting (its DESTROY method is probably getting called). This > makes the lda unusable for the parent as well. Is there any way to get > around this? Not currently. I'll make a note of it though. Meanwhile you could hack in something along these lines: int dbd_db_mark_inactive_hack(dbh) SV *dbh; { D_imp_dbh(dbh); DBIc_ACTIVE_off(imp_dbh); } and call it just before exit. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05913-8@oink>; Mon, 1 Jan 1996 07:07:03 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820271545:10783:0; Fri, 29 Dec 95 21:12:25 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa10229; 29 Dec 95 21:11 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id MAA02658 for dbi-users-real; Fri, 29 Dec 1995 12:31:14 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id MAA02645 for ; Fri, 29 Dec 1995 12:30:58 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id aa23928; 29 Dec 95 20:29 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id ak04188; 29 Dec 95 20:28 GMT Received: from toad by oink with SMTP (PP) id <28907-0@oink>; Fri, 29 Dec 1995 19:52:07 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA26386; Fri, 29 Dec 1995 19:51:39 +0000 Date: Fri, 29 Dec 1995 19:51:39 +0000 From: Tim Bunce Message-Id: <9512291951.AA26386@toad.ig.co.uk> To: IPlatacis@aol.com Subject: Re: Help with perl/oraperl for VAX/VMS Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 2255 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: IPlatacis@aol.com > > Tim, > > I came across your name as I was looking for software needed to run perl on > our machine. Perhaps you could provide me with some help to get started. > > I need to create a simple web page that displays information retrieved from > an Oracle 7 database which is on a VAX/VMS system. We already have perl5 > loaded. I have not worked with perl before and the system manager is > unfamiliar with it also; therefore, I have some questions about what we > really need to get set up. > > 1. Can perl be used alone to connect to the database and retrieve the > records, or is it necessary to get either oraperl or DBperl? Without something like oraperl or DBperl you have to use pipes. It's messay and may not work on VMS. > 2. If I do need to get oraperl or DBperl, are there versions compatible for > VAX/VMS? Not that I know of. > Where could I find these? If there are no compatible versions, > what do you suggest I do to accomplish the task? Time and/or skill (in inversely proportional amounts). > 3. It seems as if oraperl is an extension of perl4 and DBperl is an > extension of perl5. Is this true, or can they be interchanged? Here's the standard reply: --- Perl4 database interfaces such as oraperl and ingperl will not work with Perl5. A new database independent standard interface is being developed for Perl5. An oracle module which uses that interface is being developed. The oracle module includes an oraperl emulation interface. Drivers are also currently being developed for Ingres, mSQL, DB2, Informix, and QuickBase. Other drivers will be developed over time. Alpha releases of the DBI and DBD::Oracle modules are available from ftp.demon.co.uk in /pub/perl/db/ but are not recommended for production use at this time. The archive is also mirrored by CPAN sites. For more information and to keep informed about progress you can join the a mailing list by sending a message to dbi-request@fugue.com or dbi-request@fugue.com (it may take a few days to be processed) or preferably subscribe via http://www.fugue.com/dbi All discussion related to this work should take place on the mailing list. --- > Thank you for your time, > > Ina Platacis > >>> IPlatacis@aol.com <<< ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05913-10@oink>; Mon, 1 Jan 1996 07:07:19 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820271545:10761:1; Fri, 29 Dec 95 21:12:25 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa10321; 29 Dec 95 21:11 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id MAA02688 for dbi-users-real; Fri, 29 Dec 1995 12:36:16 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from gw.home.vix.com (gw.home.vix.com [192.5.5.1]) by toccata.fugue.com (8.6.12/8.6.11) with ESMTP id MAA02684 for ; Fri, 29 Dec 1995 12:36:15 -0800 Received: by gw.home.vix.com id AA05263; Fri, 29 Dec 95 12:36:15 -0800 X-Btw: vix.com is also gw.home.vix.com and vixie.sf.ca.us Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id aa23889; 29 Dec 95 20:29 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id ae04188; 29 Dec 95 20:28 GMT Received: from toad by oink with SMTP (PP) id <28754-0@oink>; Fri, 29 Dec 1995 18:47:33 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA26297; Fri, 29 Dec 1995 18:47:06 +0000 Date: Fri, 29 Dec 1995 18:47:06 +0000 From: Tim Bunce Message-Id: <9512291847.AA26297@toad.ig.co.uk> To: dbi-users@fugue.com, patricia@earthweb.com Subject: Re: DBD::Oracle 'make test' fails X-Sun-Charset: US-ASCII Content-Length: 1305 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Patricia Ju > > I recently compiled and installed perl 5.001n and DBI 0.63. I was > successful in compiling and installing DBD::mSQL. Compiling DBD::Oracle > seemed to be okay as well. However, when running 'make test' I get > the following errors. > > Please help. I've got live scripts that are hurting due to the lack > of Oracle.. Sorry for the delay (I can't recall if I've replied to this or not). > PERL_DL_NONLAZY=1 /usr/local/bin/perl -I./blib/sun4-solaris -I./blib > -I/usr/local/lib/perl5/sun4-solaris -I/usr/local/lib/perl5 -e 'use > Test::Harness qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t > t/base..............install_driver(Oracle) failed: Can't load > './blib/sun4-solaris/auto/DBD/Oracle/Oracle.so' for module DBD::Oracle: > ld.so.1: /usr/local/bin/perl: fatal: relocation error: symbol not found: > DBIc_ACTIVE_KIDS: referenced in > ./blib/sun4-solaris/auto/DBD/Oracle/Oracle.so at > /usr/local/lib/perl5/DynaLoader.pm line 136. Upgrade to the latest DBI and DBD::Oracle. The tend to get released together and DBD::Oracle tends to use the latest features of the DBI. (It's good to see PERL_DL_NONLAZY=1 is proving useful) Also Perl5.001n is not recommended for general use. It has problems. Stick with 5.001m for now. Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05913-12@oink>; Mon, 1 Jan 1996 07:07:38 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820271740:12321:0; Fri, 29 Dec 95 21:15:40 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa11613; 29 Dec 95 21:14 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id MAA02729 for dbi-users-real; Fri, 29 Dec 1995 12:42:24 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from relay-4.mail.demon.net (relay-4.mail.demon.net [158.152.1.64]) by toccata.fugue.com (8.6.12/8.6.11) with SMTP id MAA02725 for ; Fri, 29 Dec 1995 12:42:16 -0800 Received: from post.demon.co.uk ([158.152.1.72]) by relay-4.mail.demon.net id ab24319; 29 Dec 95 20:33 GMT Received: from ignite.demon.co.uk ([158.152.8.99]) by relay-3.mail.demon.net id aa05237; 29 Dec 95 20:33 GMT Received: from toad by oink with SMTP (PP) id <29126-0@oink>; Fri, 29 Dec 1995 20:32:06 +0000 Received: by toad.ig.co.uk (5.0/SMI-SVR4) id AA26438; Fri, 29 Dec 1995 20:31:04 +0000 Date: Fri, 29 Dec 1995 20:31:04 +0000 From: Tim Bunce Message-Id: <9512292031.AA26438@toad.ig.co.uk> To: shahram@smartt.com Subject: Re: DBPerl Cc: dbi-users@fugue.com X-Sun-Charset: US-ASCII content-length: 666 Organization: Paul Ingram Group, Software Systems, +44 1 483 424424 > From: Shahram > > -- [ From: Shahram * EMC.Ver #2.5.02 ] -- > > Hello Tim > > I have read the document on DBperl. In on part of document you have > mentioned the > ftp site: ftp.demon.co.uk /pub/perl/db/dbperl Just /pub/perl/db > Can I use DBperl? You could. > what would you suggest for a shareware database engine? I don't use any shareware database engines. Perhaps someone else could send you some suggestions. > Sorry if my questions are too basic. For more info see http://www.hermetica.com/technologia/DBI and http://www.fugue.com/dbi. > Thanks > > Shahram Mojtabavi > Vancouver, Canada > (shahram@smartt.com) Tim. ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <06234-22@oink>; Mon, 1 Jan 1996 07:14:46 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820432764:07732:2; Sun, 31 Dec 95 17:59:24 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa07579; 31 Dec 95 17:59 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id JAA02522 for dbi-users-real; Sun, 31 Dec 1995 09:26:17 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.12/8.6.11) with ESMTP id JAA02518 for ; Sun, 31 Dec 1995 09:26:13 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id RAA17369 for dbi-users@fugue.com; Sun, 31 Dec 1995 17:21:12 GMT From: Alligator Descartes Message-Id: <199512311721.RAA17369@fruitbat.mcqueen.com> Subject: WIP To: dbi-users@fugue.com Date: Sun, 31 Dec 1995 17:21:11 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 533 Right guys, gals and Tim. I've drawn up a list of stuff I'm currently doing wrt DBI which is at: http://www.hermetica.com/descarte/wip.html If anyone's interested in helping ( I hope so! ), then let me know and we'll open discussions somewhere sometime soon. Some of the stuff could be quite spiffy. If someone's interested that hasn't got WWW access, let me know. Regards. -- Alligator Descartes | "...Nil posse creari De nilo" descarte@hermetica.com | -- Lucretius, De Rerum Natura I, l.155 http://www.hermetica.com | ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <06234-24@oink>; Mon, 1 Jan 1996 07:15:05 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820438738:21752:1; Sun, 31 Dec 95 19:38:58 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa21299; 31 Dec 95 19:38 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id LAA02625 for dbi-users-real; Sun, 31 Dec 1995 11:15:40 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from srvr7.engin.umich.edu (root@srvr7.engin.umich.edu [141.212.2.69]) by toccata.fugue.com (8.6.12/8.6.11) with ESMTP id LAA02621 for ; Sun, 31 Dec 1995 11:15:38 -0800 Received: from srvr5.engin.umich.edu (srvr5.engin.umich.edu [141.212.2.89]) by srvr7.engin.umich.edu (8.6.12/8.6.4) with ESMTP id OAA10224 for ; Sun, 31 Dec 1995 14:16:00 -0500 Received: from zaphod (pm002-07.dialip.mich.net [141.211.7.143]) by srvr5.engin.umich.edu (8.6.12/8.6.4) with SMTP id OAA10961 for ; Sun, 31 Dec 1995 14:15:36 -0500 Message-ID: <30E6EFF2.7237@umich.edu> Date: Sun, 31 Dec 1995 14:17:54 -0600 From: Dan DeMaggio Organization: None X-Mailer: Mozilla 2.0b3 (WinNT; I) MIME-Version: 1.0 To: dbi-users@fugue.com Subject: (no subject) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I have been doing some thinking about the DBI spec and ODBC. I think it needs to 'refocus' for lack of a better term. Here's why: 1) ODBC will/has become the cross-platform database interface standard (there are MANY companies shipping Mac and Unix ODBC drivers). Therefore, DBI should _not_ try to compete with it. 2) Programatically, ODBC stinks. It's not an elegant standard. It has 56 Functions, some of wich have 20 or 30 sub-functions. It makes you think in terms of statement handles. What do statement handles (as a concept) have to do with databases? Nothing! Therefore, what we need is an Elegant and Logical way to get to databases, *NOT* another 'standard'. Once we design this nice interface, THEN we write the 'back-end' code to implement this interface for all kinds of datbases (including ODBC, Oracle, mSQL, etc.) --------------------------------- Architecture: I propose an object-oriented approach. Imagine the following objects: - Database objects (allows login, etc) - Table objects (created by queries or pre-existing) - Field objects (contain data, but also have 'metadata' properties) To quote Larry: "Easy things should be easy. Hard things should be possible." Therefore, the first program will be simple: $d = new DBI::Oracle::Database "DSN=database;UID=dmag;PWD=none;"; $t = $d->query("select count(*) from table"); print $t->field(0)->value; Real perl hackers would write: print DBI::Oracle::Database::new("DSN=blah;")-> query("select 'hello world'")->field(0); I don't know about you, but it gives me a warm fuzzy feeling inside when I can write it all on one line like that. ;) Next, we want to do complicated things like parameter queries, non-sequential reads, access metadata... $d = new DBI::Oracle::Database "DSN=database;UID=dmag;PWD=none;"; $t = $d->query("select ? from table",PARAMETER_QUERY); $t->parameter(0) = "foobar"; $t->execute; while( $t->fetchrow(+2) ) { print $t->field(0)->name . $t->field("fred")->datatype; } It's extensable because you know *WHERE* to extend it. You don't just add another function, you add properties and methods to an object.. [The above is just an example. I'm not trying to define the standard, just show what it could look like. My next release of the ODBC driver for NT Perl will look much like this.] ---------------------------------- Dan's DBI Commandments: The DBI base package should: - Be a package 100% entirely written in Perl only. Did I mention that this would not involve using C code in any way whatsoever? - Probably be the base class for all Drivers. (like TIEHASH) - Define constants and provide stubs for missing functionality. - Contain default implementations of some methods that can be decomposed into more basic functions. Of course, this behavior can be over-ridden in the individual driver modules. - Should let the autoloader load the drivers (via 'use'). - May assist the user in dynamically discovering the drivers at run-time. Question: Can I find out what implementations of TIEHASH are installed on my computer?? - Provide simple and understandable functionality. Preformance is secondary. The DBI base package should _not_: - Include any C whatsoever. One could write a driver that connects to a database using TCP/IP and sends textual SQL queries. Thus, everything on the client would be 100% written in perl. - Specify the format of ANY 'C' functions or structures. - Be the central switch to dispatch calls to the drivers. - Hold data or know ANYTHING about which handles are assigned to which drivers, or what statements are assigned to which handles. See below. Each DBI::Driver should: - Be completely object oriented. In fact, it should probably inherit from DBI. (i.e. "@ISA = ('DBI')" - Override (provide implementations for) some or all DBI methods. - Keep track of all open connections and handles. - Be the only code that knows or cares what functions are in the native (i.e. machiene specific) C library. - Be the only one to know about C structures. [i.e. C structs should not be part of the DBI API] - Pay me millions of dollars in royalties. Ok, maybe not. - store connection handle information anywhere it wants (i.e. $self->{'handle'} might be a good place. But then again, maybe it's in $self->{'connection'}. Only the driver writer knows or cares. - Only return object refrences and user data (i.e. no actual statement handles floating around the main program). ---------------------------------- Conclusion: Maybe the above is just a polite way of saying "let's throw away the current DBI spec." Or maybe I'm talking about a front-end to DBI. Or maybe what I'm talking about isn't what you are talking about. Discussion? P.S.: I'm sending this to dbi-users because, well, I think it makes DBI easier to use. I hope all the developers are on this list. (if not, they can e-mail me ;) -- Dan "http:www-personal.umich.edu/~dmag" DeMaggio (dmag@umich.edu) ---+++--- Received: from punt.demon.co.uk by oink with SMTP (PP) id <05913-62@oink>; Mon, 1 Jan 1996 07:15:36 +0000 Received: from punt-1.mail.demon.net by mailstore for dbperl@ig.co.uk id 820450152:04301:0; Sun, 31 Dec 95 22:49:12 GMT Received: from toccata.fugue.com ([204.254.239.2]) by punt-1.mail.demon.net id aa04094; 31 Dec 95 22:48 GMT Received: (from daemon@localhost) by toccata.fugue.com (8.6.12/8.6.11) id OAA00695 for dbi-users-real; Sun, 31 Dec 1995 14:17:56 -0800 Precedence: junk Sender: owner-perldb-interest@fugue.com Received: from fruitbat.mcqueen.com (pillshovel.mcqueen.com [194.72.106.246]) by toccata.fugue.com (8.6.12/8.6.11) with ESMTP id OAA00691 for ; Sun, 31 Dec 1995 14:17:54 -0800 Received: (from descarte@localhost) by fruitbat.mcqueen.com (8.6.12/8.6.12) id WAA19070 for dbi-users@fugue.com; Sun, 31 Dec 1995 22:12:58 GMT From: Alligator Descartes Message-Id: <199512312212.WAA19070@fruitbat.mcqueen.com> Subject: Oracle Severe Mangler To: dbi-users@fugue.com Date: Sun, 31 Dec 1995 22:12:56 +0000 (GMT) X-Pants: ** The return address of this email should be: descarte@hermetica.com X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 680 Hi. I've setup some mailing lists pertaining solely to the Oracle Severe Mangler product that is being currently prototyped. The product is akin to Oracle's Server Manager, but is written in perl5, Tk and DBI/DBD::Oracle. The addresses of the mailing lists, for anyone that is interested are: severe-dev-request@hermetica.com severe-users-request@hermetica.com severe-announce-request@hermetica.com If you wish to subscribe, please send a message body containing: subscribe your_email_address Regards. -- Alligator Descartes | "...Nil posse creari De nilo" descarte@hermetica.com | -- Lucretius, De Rerum Natura I, l.155 http://www.hermetica.com/descarte | ---+++---