001/* $Id: SimpleRegexMatcher.java 471661 2006-11-06 08:09:25Z skitching $
002 *
003 * Licensed to the Apache Software Foundation (ASF) under one or more
004 * contributor license agreements.  See the NOTICE file distributed with
005 * this work for additional information regarding copyright ownership.
006 * The ASF licenses this file to You under the Apache License, Version 2.0
007 * (the "License"); you may not use this file except in compliance with
008 * the License.  You may obtain a copy of the License at
009 * 
010 *      http://www.apache.org/licenses/LICENSE-2.0
011 * 
012 * Unless required by applicable law or agreed to in writing, software
013 * distributed under the License is distributed on an "AS IS" BASIS,
014 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
015 * See the License for the specific language governing permissions and
016 * limitations under the License.
017 */ 
018
019package org.apache.commons.digester;
020
021import org.apache.commons.logging.Log;
022import org.apache.commons.logging.LogFactory;
023
024/**
025 * <p>Simple regex pattern matching algorithm.</p>
026 * 
027 * <p>This uses just two wildcards:
028 * <ul>
029 *  <li><code>*</code> matches any sequence of none, one or more characters
030 *  <li><code>?</code> matches any one character 
031 * </ul>
032 * Escaping these wildcards is not supported .</p>
033 *
034 * @since 1.5
035 */
036
037public class SimpleRegexMatcher extends RegexMatcher {
038    
039    // --------------------------------------------------------- Fields
040    
041    /** Default log (class wide) */
042    private static final Log baseLog = LogFactory.getLog(SimpleRegexMatcher.class);
043    
044    /** Custom log (can be set per object) */
045    private Log log = baseLog;
046    
047    // --------------------------------------------------------- Properties
048    
049    /** 
050     * Gets the <code>Log</code> implementation.
051     */
052    public Log getLog() {
053        return log;
054    }
055    
056    /**
057     * Sets the current <code>Log</code> implementation used by this class.
058     */
059    public void setLog(Log log) {
060        this.log = log;
061    }
062    
063    // --------------------------------------------------------- Public Methods
064    
065    /** 
066     * Matches using simple regex algorithm.
067     * 
068     *
069     * @param basePattern the standard digester path representing the element
070     * @param regexPattern the regex pattern the path will be tested against
071     * @return true if the given pattern matches the given path
072     */
073    public boolean match(String basePattern, String regexPattern) {
074        // check for nulls
075        if (basePattern == null || regexPattern == null) {
076            return false;
077        }
078        return match(basePattern, regexPattern, 0, 0);
079    }
080    
081    // --------------------------------------------------------- Implementations Methods
082    
083    /**
084     * Implementation of regex matching algorithm.
085     * This calls itself recursively.
086     */
087    private boolean match(String basePattern, String regexPattern, int baseAt, int regexAt) {
088        if (log.isTraceEnabled()) {
089            log.trace("Base: " + basePattern);
090            log.trace("Regex: " + regexPattern);
091            log.trace("Base@" + baseAt);
092            log.trace("Regex@" + regexAt);
093        }
094        
095        // check bounds
096        if (regexAt >= regexPattern.length()) {
097            // maybe we've got a match
098            if (baseAt >= basePattern.length()) {
099                // ok!
100                return true;
101            }
102            // run out early
103            return false;
104            
105        } else {
106            if (baseAt >= basePattern.length()) {
107                // run out early
108                return false;
109            }
110        }
111        
112        // ok both within bounds
113        char regexCurrent = regexPattern.charAt(regexAt);
114        switch (regexCurrent) {
115            case '*':
116                // this is the tricky case
117                // check for terminal 
118                if (++regexAt >= regexPattern.length()) {
119                    // this matches anything let - so return true
120                    return true;
121                }
122                // go through every subsequent apperance of the next character
123                // and so if the rest of the regex matches
124                char nextRegex = regexPattern.charAt(regexAt);
125                if (log.isTraceEnabled()) {
126                    log.trace("Searching for next '" + nextRegex + "' char");
127                }
128                int nextMatch = basePattern.indexOf(nextRegex, baseAt);
129                while (nextMatch != -1) {
130                    if (log.isTraceEnabled()) {
131                        log.trace("Trying '*' match@" + nextMatch);
132                    }
133                    if (match(basePattern, regexPattern, nextMatch, regexAt)) {
134                        return true;
135                    }
136                    nextMatch = basePattern.indexOf(nextRegex, nextMatch + 1);
137                }
138                log.trace("No matches found.");
139                return false;
140                
141            case '?':
142                // this matches anything
143                return match(basePattern, regexPattern, ++baseAt, ++regexAt);
144            
145            default:
146                if (log.isTraceEnabled()) {
147                    log.trace("Camparing " + regexCurrent + " to " + basePattern.charAt(baseAt));
148                }
149                if (regexCurrent == basePattern.charAt(baseAt)) {
150                    // still got more to go
151                    return match(basePattern, regexPattern, ++baseAt, ++regexAt);
152                }
153                return false;
154        }
155    }
156}