Source Code Cross Referenced for Perl5Matcher.java in  » Development » Jakarta-ORO » org » apache » oro » text » regex » Java Source Code / Java DocumentationJava Source Code and Java Documentation

Java Source Code / Java Documentation
1. 6.0 JDK Core
2. 6.0 JDK Modules
3. 6.0 JDK Modules com.sun
4. 6.0 JDK Modules com.sun.java
5. 6.0 JDK Modules sun
6. 6.0 JDK Platform
7. Ajax
8. Apache Harmony Java SE
9. Aspect oriented
10. Authentication Authorization
11. Blogger System
12. Build
13. Byte Code
14. Cache
15. Chart
16. Chat
17. Code Analyzer
18. Collaboration
19. Content Management System
20. Database Client
21. Database DBMS
22. Database JDBC Connection Pool
23. Database ORM
24. Development
25. EJB Server geronimo
26. EJB Server GlassFish
27. EJB Server JBoss 4.2.1
28. EJB Server resin 3.1.5
29. ERP CRM Financial
30. ESB
31. Forum
32. GIS
33. Graphic Library
34. Groupware
35. HTML Parser
36. IDE
37. IDE Eclipse
38. IDE Netbeans
39. Installer
40. Internationalization Localization
41. Inversion of Control
42. Issue Tracking
43. J2EE
44. JBoss
45. JMS
46. JMX
47. Library
48. Mail Clients
49. Net
50. Parser
51. PDF
52. Portal
53. Profiler
54. Project Management
55. Report
56. RSS RDF
57. Rule Engine
58. Science
59. Scripting
60. Search Engine
61. Security
62. Sevlet Container
63. Source Control
64. Swing Library
65. Template Engine
66. Test Coverage
67. Testing
68. UML
69. Web Crawler
70. Web Framework
71. Web Mail
72. Web Server
73. Web Services
74. Web Services apache cxf 2.0.1
75. Web Services AXIS2
76. Wiki Engine
77. Workflow Engines
78. XML
79. XML UI
Java
Java Tutorial
Java Open Source
Jar File Download
Java Articles
Java Products
Java by API
Photoshop Tutorials
Maya Tutorials
Flash Tutorials
3ds-Max Tutorials
Illustrator Tutorials
GIMP Tutorials
C# / C Sharp
C# / CSharp Tutorial
C# / CSharp Open Source
ASP.Net
ASP.NET Tutorial
JavaScript DHTML
JavaScript Tutorial
JavaScript Reference
HTML / CSS
HTML CSS Reference
C / ANSI-C
C Tutorial
C++
C++ Tutorial
Ruby
PHP
Python
Python Tutorial
Python Open Source
SQL Server / T-SQL
SQL Server / T-SQL Tutorial
Oracle PL / SQL
Oracle PL/SQL Tutorial
PostgreSQL
SQL / MySQL
MySQL Tutorial
VB.Net
VB.Net Tutorial
Flash / Flex / ActionScript
VBA / Excel / Access / Word
XML
XML Tutorial
Microsoft Office PowerPoint 2007 Tutorial
Microsoft Office Excel 2007 Tutorial
Microsoft Office Word 2007 Tutorial
Java Source Code / Java Documentation » Development » Jakarta ORO » org.apache.oro.text.regex 
Source Cross Referenced  Class Diagram Java Document (Java Doc) 


0001:        /*
0002:         * $Id: Perl5Matcher.java,v 1.27 2003/11/07 20:16:25 dfs Exp $
0003:         *
0004:         * ====================================================================
0005:         * The Apache Software License, Version 1.1
0006:         *
0007:         * Copyright (c) 2000 The Apache Software Foundation.  All rights
0008:         * reserved.
0009:         *
0010:         * Redistribution and use in source and binary forms, with or without
0011:         * modification, are permitted provided that the following conditions
0012:         * are met:
0013:         *
0014:         * 1. Redistributions of source code must retain the above copyright
0015:         *    notice, this list of conditions and the following disclaimer.
0016:         *
0017:         * 2. Redistributions in binary form must reproduce the above copyright
0018:         *    notice, this list of conditions and the following disclaimer in
0019:         *    the documentation and/or other materials provided with the
0020:         *    distribution.
0021:         *
0022:         * 3. The end-user documentation included with the redistribution,
0023:         *    if any, must include the following acknowledgment:
0024:         *       "This product includes software developed by the
0025:         *        Apache Software Foundation (http://www.apache.org/)."
0026:         *    Alternately, this acknowledgment may appear in the software itself,
0027:         *    if and wherever such third-party acknowledgments normally appear.
0028:         *
0029:         * 4. The names "Apache" and "Apache Software Foundation", "Jakarta-Oro" 
0030:         *    must not be used to endorse or promote products derived from this
0031:         *    software without prior written permission. For written
0032:         *    permission, please contact apache@apache.org.
0033:         *
0034:         * 5. Products derived from this software may not be called "Apache" 
0035:         *    or "Jakarta-Oro", nor may "Apache" or "Jakarta-Oro" appear in their 
0036:         *    name, without prior written permission of the Apache Software Foundation.
0037:         *
0038:         * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
0039:         * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
0040:         * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
0041:         * DISCLAIMED.  IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
0042:         * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
0043:         * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
0044:         * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
0045:         * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
0046:         * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
0047:         * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
0048:         * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
0049:         * SUCH DAMAGE.
0050:         * ====================================================================
0051:         *
0052:         * This software consists of voluntary contributions made by many
0053:         * individuals on behalf of the Apache Software Foundation.  For more
0054:         * information on the Apache Software Foundation, please see
0055:         * <http://www.apache.org/>.
0056:         */
0057:
0058:        package org.apache.oro.text.regex;
0059:
0060:        import java.util.*;
0061:
0062:        /**
0063:         * The Perl5Matcher class is used to match regular expressions
0064:         * (conforming to the Perl5 regular expression syntax) generated by
0065:         * Perl5Compiler.
0066:         * <p>
0067:         * Perl5Compiler and Perl5Matcher are designed with the intent that
0068:         * you use a separate instance of each per thread to avoid the overhead
0069:         * of both synchronization and concurrent access (e.g., a match that takes
0070:         * a long time in one thread will block the progress of another thread with
0071:         * a shorter match).  If you want to use a single instance of each
0072:         * in a concurrent program, you must appropriately protect access to
0073:         * the instances with critical sections.  If you want to share Perl5Pattern
0074:         * instances between concurrently executing instances of Perl5Matcher, you
0075:         * must compile the patterns with {@link Perl5Compiler#READ_ONLY_MASK}.
0076:         *
0077:         * @version @version@
0078:         * @since 1.0
0079:         * @see PatternMatcher
0080:         * @see Perl5Compiler
0081:         */
0082:        public final class Perl5Matcher implements  PatternMatcher {
0083:            private static final char __EOS = Character.MAX_VALUE;
0084:            private static final int __INITIAL_NUM_OFFSETS = 20;
0085:
0086:            private boolean __multiline = false, __lastSuccess = false;
0087:            private boolean __caseInsensitive = false;
0088:            private char __previousChar, __input[], __originalInput[];
0089:            private Perl5Repetition __currentRep;
0090:            private int __numParentheses, __bol, __eol, __currentOffset,
0091:                    __endOffset;
0092:
0093:            private char[] __program;
0094:            private int __expSize, __inputOffset, __lastParen;
0095:            private int[] __beginMatchOffsets, __endMatchOffsets;
0096:            private Stack __stack = new Stack();
0097:            private Perl5MatchResult __lastMatchResult = null;
0098:
0099:            private static boolean __compare(char[] s1, int s1Offs, char[] s2,
0100:                    int s2Offs, int n) {
0101:                int cnt;
0102:
0103:                for (cnt = 0; cnt < n; cnt++, s1Offs++, s2Offs++) {
0104:                    if (s1Offs >= s1.length)
0105:                        return false;
0106:                    if (s2Offs >= s2.length)
0107:                        return false;
0108:                    if (s1[s1Offs] != s2[s2Offs])
0109:                        return false;
0110:                }
0111:
0112:                return true;
0113:            }
0114:
0115:            private static int __findFirst(char[] input, int current,
0116:                    int endOffset, char[] mustString) {
0117:                int count, saveCurrent;
0118:                char ch;
0119:
0120:                if (input.length == 0)
0121:                    return endOffset;
0122:
0123:                ch = mustString[0];
0124:                // Find the offset of the first character of the must string
0125:                while (current < endOffset) {
0126:                    if (ch == input[current]) {
0127:                        saveCurrent = current;
0128:                        count = 0;
0129:
0130:                        while (current < endOffset && count < mustString.length) {
0131:                            if (mustString[count] != input[current])
0132:                                break;
0133:                            ++count;
0134:                            ++current;
0135:                        }
0136:
0137:                        current = saveCurrent;
0138:
0139:                        if (count >= mustString.length)
0140:                            break;
0141:                    }
0142:                    ++current;
0143:                }
0144:
0145:                return current;
0146:            }
0147:
0148:            private void __pushState(int parenFloor) {
0149:                int[] state;
0150:                int stateEntries, paren;
0151:
0152:                stateEntries = 3 * (__expSize - parenFloor);
0153:                if (stateEntries <= 0)
0154:                    state = new int[3];
0155:                else
0156:                    state = new int[stateEntries + 3];
0157:
0158:                state[0] = __expSize;
0159:                state[1] = __lastParen;
0160:                state[2] = __inputOffset;
0161:
0162:                for (paren = __expSize; paren > parenFloor; --paren, stateEntries -= 3) {
0163:                    state[stateEntries] = __endMatchOffsets[paren];
0164:                    state[stateEntries + 1] = __beginMatchOffsets[paren];
0165:                    state[stateEntries + 2] = paren;
0166:                }
0167:
0168:                __stack.push(state);
0169:            }
0170:
0171:            private void __popState() {
0172:                int[] state;
0173:                int entry, paren;
0174:
0175:                state = (int[]) __stack.pop();
0176:
0177:                __expSize = state[0];
0178:                __lastParen = state[1];
0179:                __inputOffset = state[2];
0180:
0181:                for (entry = 3; entry < state.length; entry += 3) {
0182:                    paren = state[entry + 2];
0183:                    __beginMatchOffsets[paren] = state[entry + 1];
0184:
0185:                    if (paren <= __lastParen)
0186:                        __endMatchOffsets[paren] = state[entry];
0187:                }
0188:
0189:                for (paren = __lastParen + 1; paren <= __numParentheses; paren++) {
0190:                    if (paren > __expSize)
0191:                        __beginMatchOffsets[paren] = OpCode._NULL_OFFSET;
0192:                    __endMatchOffsets[paren] = OpCode._NULL_OFFSET;
0193:                }
0194:            }
0195:
0196:            // Initialize globals needed before calling __tryExpression for first time
0197:            private void __initInterpreterGlobals(Perl5Pattern expression,
0198:                    char[] input, int beginOffset, int endOffset,
0199:                    int currentOffset) {
0200:                // Remove this hack after more efficient case-folding and unicode
0201:                // character classes are implemented
0202:                __caseInsensitive = expression._isCaseInsensitive;
0203:                __input = input;
0204:                __endOffset = endOffset;
0205:                __currentRep = new Perl5Repetition();
0206:                __currentRep._numInstances = 0;
0207:                __currentRep._lastRepetition = null;
0208:                __program = expression._program;
0209:                __stack.setSize(0);
0210:
0211:                // currentOffset should always be >= beginOffset and should
0212:                // always be equal to zero when beginOffset equals 0, but we
0213:                // make a weak attempt to protect against a violation of this
0214:                // precondition
0215:                if (currentOffset == beginOffset || currentOffset <= 0)
0216:                    __previousChar = '\n';
0217:                else {
0218:                    __previousChar = input[currentOffset - 1];
0219:                    if (!__multiline && __previousChar == '\n')
0220:                        __previousChar = '\0';
0221:                }
0222:
0223:                __numParentheses = expression._numParentheses;
0224:                __currentOffset = currentOffset;
0225:
0226:                __bol = beginOffset;
0227:                __eol = endOffset;
0228:
0229:                // Ok, here we're using endOffset as a temporary variable.
0230:                endOffset = __numParentheses + 1;
0231:                if (__beginMatchOffsets == null
0232:                        || endOffset > __beginMatchOffsets.length) {
0233:                    if (endOffset < __INITIAL_NUM_OFFSETS)
0234:                        endOffset = __INITIAL_NUM_OFFSETS;
0235:                    __beginMatchOffsets = new int[endOffset];
0236:                    __endMatchOffsets = new int[endOffset];
0237:                }
0238:            }
0239:
0240:            // Set the match result information.  Only call this if we successfully
0241:            // matched.
0242:            private void __setLastMatchResult() {
0243:                int offs, maxEndOffs = 0;
0244:
0245:                //endOffset+=dontTry;
0246:
0247:                __lastMatchResult = new Perl5MatchResult(__numParentheses + 1);
0248:
0249:                // This can happen when using Perl5StreamInput
0250:                if (__endMatchOffsets[0] > __originalInput.length)
0251:                    throw new ArrayIndexOutOfBoundsException();
0252:
0253:                __lastMatchResult._matchBeginOffset = __beginMatchOffsets[0];
0254:
0255:                while (__numParentheses >= 0) {
0256:                    offs = __beginMatchOffsets[__numParentheses];
0257:
0258:                    if (offs >= 0)
0259:                        __lastMatchResult._beginGroupOffset[__numParentheses] = offs
0260:                                - __lastMatchResult._matchBeginOffset;
0261:                    else
0262:                        __lastMatchResult._beginGroupOffset[__numParentheses] = OpCode._NULL_OFFSET;
0263:
0264:                    offs = __endMatchOffsets[__numParentheses];
0265:
0266:                    if (offs >= 0) {
0267:                        __lastMatchResult._endGroupOffset[__numParentheses] = offs
0268:                                - __lastMatchResult._matchBeginOffset;
0269:                        if (offs > maxEndOffs && offs <= __originalInput.length)
0270:                            maxEndOffs = offs;
0271:                    } else
0272:                        __lastMatchResult._endGroupOffset[__numParentheses] = OpCode._NULL_OFFSET;
0273:
0274:                    --__numParentheses;
0275:                }
0276:
0277:                __lastMatchResult._match = new String(__originalInput,
0278:                        __beginMatchOffsets[0], maxEndOffs
0279:                                - __beginMatchOffsets[0]);
0280:
0281:                // Free up for garbage collection
0282:                __originalInput = null;
0283:            }
0284:
0285:            // Expects to receive a valid regular expression program.  No checking
0286:            // is done to ensure validity.
0287:            // __originalInput must be set before calling this method for
0288:            // __lastMatchResult to be set correctly.
0289:            // beginOffset marks the beginning of the string
0290:            // currentOffset marks where to start the pattern search
0291:            private boolean __interpret(Perl5Pattern expression, char[] input,
0292:                    int beginOffset, int endOffset, int currentOffset) {
0293:                boolean success;
0294:                int minLength = 0, dontTry = 0, offset;
0295:                char ch, mustString[];
0296:
0297:                __initInterpreterGlobals(expression, input, beginOffset,
0298:                        endOffset, currentOffset);
0299:
0300:                success = false;
0301:                mustString = expression._mustString;
0302:
0303:                _mainLoop: while (true) {
0304:
0305:                    if (mustString != null
0306:                            && ((expression._anchor & Perl5Pattern._OPT_ANCH) == 0 || ((__multiline || (expression._anchor & Perl5Pattern._OPT_ANCH_MBOL) != 0) && expression._back >= 0))) {
0307:
0308:                        __currentOffset = __findFirst(__input, __currentOffset,
0309:                                endOffset, mustString);
0310:
0311:                        if (__currentOffset >= endOffset) {
0312:                            if ((expression._options & Perl5Compiler.READ_ONLY_MASK) == 0)
0313:                                expression._mustUtility++;
0314:                            success = false;
0315:                            break _mainLoop;
0316:                        } else if (expression._back >= 0) {
0317:                            __currentOffset -= expression._back;
0318:                            if (__currentOffset < currentOffset)
0319:                                __currentOffset = currentOffset;
0320:                            minLength = expression._back + mustString.length;
0321:                        } else if (!expression._isExpensive
0322:                                && (expression._options & Perl5Compiler.READ_ONLY_MASK) == 0
0323:                                && (--expression._mustUtility < 0)) {
0324:                            // Be careful!  The preceding logical expression is constructed
0325:                            // so that mustUtility is only decremented if the expression is
0326:                            // compiled without READ_ONLY_MASK.
0327:                            mustString = expression._mustString = null;
0328:                            __currentOffset = currentOffset;
0329:                        } else {
0330:                            __currentOffset = currentOffset;
0331:                            minLength = mustString.length;
0332:                        }
0333:                    }
0334:
0335:                    if ((expression._anchor & Perl5Pattern._OPT_ANCH) != 0) {
0336:                        if (__currentOffset == beginOffset
0337:                                && __tryExpression(beginOffset)) {
0338:                            success = true;
0339:                            break _mainLoop;
0340:                        } else if (__multiline
0341:                                || (expression._anchor & Perl5Pattern._OPT_ANCH_MBOL) != 0
0342:                                || (expression._anchor & Perl5Pattern._OPT_IMPLICIT) != 0) {
0343:                            if (minLength > 0)
0344:                                dontTry = minLength - 1;
0345:                            endOffset -= dontTry;
0346:
0347:                            if (__currentOffset > currentOffset)
0348:                                --__currentOffset;
0349:
0350:                            while (__currentOffset < endOffset) {
0351:                                if (__input[__currentOffset++] == '\n') {
0352:                                    if (__currentOffset < endOffset
0353:                                            && __tryExpression(__currentOffset)) {
0354:                                        success = true;
0355:                                        break _mainLoop;
0356:                                    }
0357:                                }
0358:                            }
0359:                        }
0360:
0361:                        break _mainLoop;
0362:                    }
0363:
0364:                    if (expression._startString != null) {
0365:                        mustString = expression._startString;
0366:                        if ((expression._anchor & Perl5Pattern._OPT_SKIP) != 0) {
0367:                            ch = mustString[0];
0368:
0369:                            while (__currentOffset < endOffset) {
0370:                                if (ch == __input[__currentOffset]) {
0371:                                    if (__tryExpression(__currentOffset)) {
0372:                                        success = true;
0373:                                        break _mainLoop;
0374:                                    }
0375:                                    ++__currentOffset;
0376:                                    while (__currentOffset < endOffset
0377:                                            && __input[__currentOffset] == ch)
0378:                                        ++__currentOffset;
0379:                                }
0380:                                ++__currentOffset;
0381:                            }
0382:                        } else {
0383:
0384:                            while ((__currentOffset = __findFirst(__input,
0385:                                    __currentOffset, endOffset, mustString)) < endOffset) {
0386:                                if (__tryExpression(__currentOffset)) {
0387:                                    success = true;
0388:                                    break _mainLoop;
0389:                                }
0390:                                ++__currentOffset;
0391:                            }
0392:                        }
0393:
0394:                        break _mainLoop;
0395:                    }
0396:
0397:                    if ((offset = expression._startClassOffset) != OpCode._NULL_OFFSET) {
0398:                        boolean doEvery, tmp;
0399:                        char op;
0400:
0401:                        doEvery = ((expression._anchor & Perl5Pattern._OPT_SKIP) == 0);
0402:
0403:                        if (minLength > 0)
0404:                            dontTry = minLength - 1;
0405:                        endOffset -= dontTry;
0406:                        tmp = true;
0407:
0408:                        switch (op = __program[offset]) {
0409:                        case OpCode._ANYOF:
0410:                            offset = OpCode._getOperand(offset);
0411:                            while (__currentOffset < endOffset) {
0412:                                ch = __input[__currentOffset];
0413:
0414:                                if (ch < 256
0415:                                        && (__program[offset + (ch >> 4)] & (1 << (ch & 0xf))) == 0) {
0416:                                    if (tmp && __tryExpression(__currentOffset)) {
0417:                                        success = true;
0418:                                        break _mainLoop;
0419:                                    } else
0420:                                        tmp = doEvery;
0421:                                } else
0422:                                    tmp = true;
0423:                                ++__currentOffset;
0424:                            }
0425:
0426:                            break;
0427:
0428:                        case OpCode._ANYOFUN:
0429:                        case OpCode._NANYOFUN:
0430:                            offset = OpCode._getOperand(offset);
0431:                            while (__currentOffset < endOffset) {
0432:                                ch = __input[__currentOffset];
0433:
0434:                                if (__matchUnicodeClass(ch, __program, offset,
0435:                                        op)) {
0436:                                    if (tmp && __tryExpression(__currentOffset)) {
0437:                                        success = true;
0438:                                        break _mainLoop;
0439:                                    } else
0440:                                        tmp = doEvery;
0441:                                } else
0442:                                    tmp = true;
0443:                                ++__currentOffset;
0444:                            }
0445:
0446:                            break;
0447:
0448:                        case OpCode._BOUND:
0449:                            if (minLength > 0) {
0450:                                ++dontTry;
0451:                                --endOffset;
0452:                            }
0453:
0454:                            if (__currentOffset != beginOffset) {
0455:                                ch = __input[__currentOffset - 1];
0456:                                tmp = OpCode._isWordCharacter(ch);
0457:                            } else
0458:                                tmp = OpCode._isWordCharacter(__previousChar);
0459:
0460:                            while (__currentOffset < endOffset) {
0461:                                ch = __input[__currentOffset];
0462:                                if (tmp != OpCode._isWordCharacter(ch)) {
0463:                                    tmp = !tmp;
0464:                                    if (__tryExpression(__currentOffset)) {
0465:                                        success = true;
0466:                                        break _mainLoop;
0467:                                    }
0468:                                }
0469:                                ++__currentOffset;
0470:                            }
0471:
0472:                            if ((minLength > 0 || tmp)
0473:                                    && __tryExpression(__currentOffset)) {
0474:                                success = true;
0475:                                break _mainLoop;
0476:                            }
0477:                            break;
0478:
0479:                        case OpCode._NBOUND:
0480:                            if (minLength > 0) {
0481:                                ++dontTry;
0482:                                --endOffset;
0483:                            }
0484:
0485:                            if (__currentOffset != beginOffset) {
0486:                                ch = __input[__currentOffset - 1];
0487:                                tmp = OpCode._isWordCharacter(ch);
0488:                            } else
0489:                                tmp = OpCode._isWordCharacter(__previousChar);
0490:
0491:                            while (__currentOffset < endOffset) {
0492:                                ch = __input[__currentOffset];
0493:                                if (tmp != OpCode._isWordCharacter(ch))
0494:                                    tmp = !tmp;
0495:                                else if (__tryExpression(__currentOffset)) {
0496:                                    success = true;
0497:                                    break _mainLoop;
0498:                                }
0499:
0500:                                ++__currentOffset;
0501:                            }
0502:
0503:                            if ((minLength > 0 || !tmp)
0504:                                    && __tryExpression(__currentOffset)) {
0505:                                success = true;
0506:                                break _mainLoop;
0507:                            }
0508:                            break;
0509:
0510:                        case OpCode._ALNUM:
0511:                            while (__currentOffset < endOffset) {
0512:                                ch = __input[__currentOffset];
0513:                                if (OpCode._isWordCharacter(ch)) {
0514:                                    if (tmp && __tryExpression(__currentOffset)) {
0515:                                        success = true;
0516:                                        break _mainLoop;
0517:                                    } else
0518:                                        tmp = doEvery;
0519:                                } else
0520:                                    tmp = true;
0521:                                ++__currentOffset;
0522:                            }
0523:                            break;
0524:
0525:                        case OpCode._NALNUM:
0526:                            while (__currentOffset < endOffset) {
0527:                                ch = __input[__currentOffset];
0528:                                if (!OpCode._isWordCharacter(ch)) {
0529:                                    if (tmp && __tryExpression(__currentOffset)) {
0530:                                        success = true;
0531:                                        break _mainLoop;
0532:                                    } else
0533:                                        tmp = doEvery;
0534:                                } else
0535:                                    tmp = true;
0536:                                ++__currentOffset;
0537:                            }
0538:                            break;
0539:
0540:                        case OpCode._SPACE:
0541:                            while (__currentOffset < endOffset) {
0542:                                if (Character
0543:                                        .isWhitespace(__input[__currentOffset])) {
0544:                                    if (tmp && __tryExpression(__currentOffset)) {
0545:                                        success = true;
0546:                                        break _mainLoop;
0547:                                    } else
0548:                                        tmp = doEvery;
0549:                                } else
0550:                                    tmp = true;
0551:                                ++__currentOffset;
0552:                            }
0553:                            break;
0554:
0555:                        case OpCode._NSPACE:
0556:                            while (__currentOffset < endOffset) {
0557:                                if (!Character
0558:                                        .isWhitespace(__input[__currentOffset])) {
0559:                                    if (tmp && __tryExpression(__currentOffset)) {
0560:                                        success = true;
0561:                                        break _mainLoop;
0562:                                    } else
0563:                                        tmp = doEvery;
0564:                                } else
0565:                                    tmp = true;
0566:                                ++__currentOffset;
0567:                            }
0568:                            break;
0569:
0570:                        case OpCode._DIGIT:
0571:                            while (__currentOffset < endOffset) {
0572:                                if (Character.isDigit(__input[__currentOffset])) {
0573:                                    if (tmp && __tryExpression(__currentOffset)) {
0574:                                        success = true;
0575:                                        break _mainLoop;
0576:                                    } else
0577:                                        tmp = doEvery;
0578:                                } else
0579:                                    tmp = true;
0580:                                ++__currentOffset;
0581:                            }
0582:                            break;
0583:
0584:                        case OpCode._NDIGIT:
0585:                            while (__currentOffset < endOffset) {
0586:                                if (!Character
0587:                                        .isDigit(__input[__currentOffset])) {
0588:                                    if (tmp && __tryExpression(__currentOffset)) {
0589:                                        success = true;
0590:                                        break _mainLoop;
0591:                                    } else
0592:                                        tmp = doEvery;
0593:                                } else
0594:                                    tmp = true;
0595:                                ++__currentOffset;
0596:                            }
0597:                            break;
0598:                        } // end switch
0599:
0600:                    } else {
0601:                        if (minLength > 0)
0602:                            dontTry = minLength - 1;
0603:                        endOffset -= dontTry;
0604:
0605:                        do {
0606:                            if (__tryExpression(__currentOffset)) {
0607:                                success = true;
0608:                                break _mainLoop;
0609:                            }
0610:                        } while (__currentOffset++ < endOffset);
0611:
0612:                    }
0613:
0614:                    break _mainLoop;
0615:                } // end while
0616:
0617:                __lastSuccess = success;
0618:                __lastMatchResult = null;
0619:
0620:                return success;
0621:            }
0622:
0623:            private boolean __matchUnicodeClass(char code, char __program[],
0624:                    int offset, char opcode) {
0625:                boolean isANYOF = (opcode == OpCode._ANYOFUN);
0626:
0627:                while (__program[offset] != OpCode._END) {
0628:                    if (__program[offset] == OpCode._RANGE) {
0629:                        offset++;
0630:                        if ((code >= __program[offset])
0631:                                && (code <= __program[offset + 1])) {
0632:                            return isANYOF;
0633:                        } else {
0634:                            offset += 2;
0635:                        }
0636:
0637:                    } else if (__program[offset] == OpCode._ONECHAR) {
0638:                        offset++;
0639:                        if (__program[offset++] == code)
0640:                            return isANYOF;
0641:
0642:                    } else {
0643:                        isANYOF = (__program[offset] == OpCode._OPCODE) ? isANYOF
0644:                                : !isANYOF;
0645:
0646:                        offset++;
0647:                        switch (__program[offset++]) {
0648:                        case OpCode._ALNUM:
0649:                            if (OpCode._isWordCharacter(code))
0650:                                return isANYOF;
0651:                            break;
0652:                        case OpCode._NALNUM:
0653:                            if (!OpCode._isWordCharacter(code))
0654:                                return isANYOF;
0655:                            break;
0656:                        case OpCode._SPACE:
0657:                            if (Character.isWhitespace(code))
0658:                                return isANYOF;
0659:                            break;
0660:                        case OpCode._NSPACE:
0661:                            if (!Character.isWhitespace(code))
0662:                                return isANYOF;
0663:                            break;
0664:                        case OpCode._DIGIT:
0665:                            if (Character.isDigit(code))
0666:                                return isANYOF;
0667:                            break;
0668:                        case OpCode._NDIGIT:
0669:                            if (!Character.isDigit(code))
0670:                                return isANYOF;
0671:                            break;
0672:                        case OpCode._ALNUMC:
0673:                            if (Character.isLetterOrDigit(code))
0674:                                return isANYOF;
0675:                            break;
0676:                        case OpCode._ALPHA:
0677:                            if (Character.isLetter(code))
0678:                                return isANYOF;
0679:                            break;
0680:                        case OpCode._BLANK:
0681:                            if (Character.isSpaceChar(code))
0682:                                return isANYOF;
0683:                            break;
0684:                        case OpCode._CNTRL:
0685:                            if (Character.isISOControl(code))
0686:                                return isANYOF;
0687:                            break;
0688:                        case OpCode._LOWER:
0689:                            if (Character.isLowerCase(code))
0690:                                return isANYOF;
0691:                            // Remove this hack after more efficient case-folding and unicode
0692:                            // character classes are implemented
0693:                            if (__caseInsensitive
0694:                                    && Character.isUpperCase(code))
0695:                                return isANYOF;
0696:                            break;
0697:                        case OpCode._UPPER:
0698:                            if (Character.isUpperCase(code))
0699:                                return isANYOF;
0700:                            // Remove this hack after more efficient case-folding and unicode
0701:                            // character classes are implemented
0702:                            if (__caseInsensitive
0703:                                    && Character.isLowerCase(code))
0704:                                return isANYOF;
0705:                            break;
0706:                        case OpCode._PRINT:
0707:                            if (Character.isSpaceChar(code))
0708:                                return isANYOF;
0709:                            // Fall through to check if the character is alphanumeric,
0710:                            // or a punctuation mark.  Printable characters are either
0711:                            // alphanumeric, punctuation marks, or spaces.
0712:                        case OpCode._GRAPH:
0713:                            if (Character.isLetterOrDigit(code))
0714:                                return isANYOF;
0715:                            // Fall through to check if the character is a punctuation mark.
0716:                            // Graph characters are either alphanumeric or punctuation.
0717:                        case OpCode._PUNCT:
0718:                            switch (Character.getType(code)) {
0719:                            case Character.DASH_PUNCTUATION:
0720:                            case Character.START_PUNCTUATION:
0721:                            case Character.END_PUNCTUATION:
0722:                            case Character.CONNECTOR_PUNCTUATION:
0723:                            case Character.OTHER_PUNCTUATION:
0724:                            case Character.MATH_SYMBOL:
0725:                            case Character.CURRENCY_SYMBOL:
0726:                            case Character.MODIFIER_SYMBOL:
0727:                                return isANYOF;
0728:                            default:
0729:                                break;
0730:                            }
0731:                            break;
0732:                        case OpCode._XDIGIT:
0733:                            if ((code >= '0' && code <= '9')
0734:                                    || (code >= 'a' && code <= 'f')
0735:                                    || (code >= 'A' && code <= 'F'))
0736:                                return isANYOF;
0737:                            break;
0738:                        case OpCode._ASCII:
0739:                            if (code < 0x80)
0740:                                return isANYOF;
0741:                        }
0742:                    }
0743:                }
0744:                return !isANYOF;
0745:            }
0746:
0747:            private boolean __tryExpression(int offset) {
0748:                int count;
0749:
0750:                __inputOffset = offset;
0751:                __lastParen = 0;
0752:                __expSize = 0;
0753:
0754:                if (__numParentheses > 0) {
0755:                    for (count = 0; count <= __numParentheses; count++) {
0756:                        __beginMatchOffsets[count] = OpCode._NULL_OFFSET;
0757:                        __endMatchOffsets[count] = OpCode._NULL_OFFSET;
0758:                    }
0759:                }
0760:
0761:                if (__match(1)) {
0762:                    __beginMatchOffsets[0] = offset;
0763:                    __endMatchOffsets[0] = __inputOffset;
0764:                    return true;
0765:                }
0766:
0767:                return false;
0768:            }
0769:
0770:            private int __repeat(int offset, int max) {
0771:                int scan, eol, operand, ret;
0772:                char ch;
0773:                char op;
0774:
0775:                scan = __inputOffset;
0776:                eol = __eol;
0777:
0778:                if (max != Character.MAX_VALUE && max < eol - scan)
0779:                    eol = scan + max;
0780:
0781:                operand = OpCode._getOperand(offset);
0782:
0783:                switch (op = __program[offset]) {
0784:
0785:                case OpCode._ANY:
0786:                    while (scan < eol && __input[scan] != '\n')
0787:                        ++scan;
0788:                    break;
0789:
0790:                case OpCode._SANY:
0791:                    scan = eol;
0792:                    break;
0793:
0794:                case OpCode._EXACTLY:
0795:                    ++operand;
0796:                    while (scan < eol && __program[operand] == __input[scan])
0797:                        ++scan;
0798:                    break;
0799:
0800:                case OpCode._ANYOF:
0801:                    if (scan < eol && (ch = __input[scan]) < 256) {
0802:                        while ((ch < 256)
0803:                                && (__program[operand + (ch >> 4)] & (1 << (ch & 0xf))) == 0) {
0804:                            if (++scan < eol)
0805:                                ch = __input[scan];
0806:                            else
0807:                                break;
0808:                        }
0809:                    }
0810:                    break;
0811:
0812:                case OpCode._ANYOFUN:
0813:                case OpCode._NANYOFUN:
0814:                    if (scan < eol) {
0815:                        ch = __input[scan];
0816:                        while (__matchUnicodeClass(ch, __program, operand, op)) {
0817:                            if (++scan < eol)
0818:                                ch = __input[scan];
0819:                            else
0820:                                break;
0821:                        }
0822:                    }
0823:                    break;
0824:
0825:                case OpCode._ALNUM:
0826:                    while (scan < eol && OpCode._isWordCharacter(__input[scan]))
0827:                        ++scan;
0828:                    break;
0829:
0830:                case OpCode._NALNUM:
0831:                    while (scan < eol
0832:                            && !OpCode._isWordCharacter(__input[scan]))
0833:                        ++scan;
0834:                    break;
0835:
0836:                case OpCode._SPACE:
0837:                    while (scan < eol && Character.isWhitespace(__input[scan]))
0838:                        ++scan;
0839:                    break;
0840:
0841:                case OpCode._NSPACE:
0842:                    while (scan < eol && !Character.isWhitespace(__input[scan]))
0843:                        ++scan;
0844:                    break;
0845:
0846:                case OpCode._DIGIT:
0847:                    while (scan < eol && Character.isDigit(__input[scan]))
0848:                        ++scan;
0849:                    break;
0850:
0851:                case OpCode._NDIGIT:
0852:                    while (scan < eol && !Character.isDigit(__input[scan]))
0853:                        ++scan;
0854:                    break;
0855:
0856:                default:
0857:                    break;
0858:
0859:                }
0860:
0861:                ret = scan - __inputOffset;
0862:                __inputOffset = scan;
0863:
0864:                return ret;
0865:            }
0866:
0867:            private boolean __match(int offset) {
0868:                char nextChar, op;
0869:                int scan, next, input, maxScan, current, line, arg;
0870:                boolean inputRemains = true, minMod = false;
0871:                Perl5Repetition rep;
0872:
0873:                input = __inputOffset;
0874:                inputRemains = (input < __endOffset);
0875:                nextChar = (inputRemains ? __input[input] : __EOS);
0876:
0877:                scan = offset;
0878:                maxScan = __program.length;
0879:
0880:                while (scan < maxScan /*&& scan > 0*/) {
0881:                    next = OpCode._getNext(__program, scan);
0882:
0883:                    switch (op = __program[scan]) {
0884:
0885:                    case OpCode._BOL:
0886:                        if (input == __bol ? __previousChar == '\n'
0887:                                : (__multiline
0888:                                        && (inputRemains || input < __eol) && __input[input - 1] == '\n'))
0889:                            break;
0890:                        return false;
0891:
0892:                    case OpCode._MBOL:
0893:                        if (input == __bol ? __previousChar == '\n'
0894:                                : ((inputRemains || input < __eol) && __input[input - 1] == '\n'))
0895:                            break;
0896:                        return false;
0897:
0898:                    case OpCode._SBOL:
0899:                        if (input == __bol && __previousChar == '\n')
0900:                            break;
0901:                        return false;
0902:
0903:                    case OpCode._GBOL:
0904:                        if (input == __bol)
0905:                            break;
0906:                        return true;
0907:
0908:                    case OpCode._EOL:
0909:                        if ((inputRemains || input < __eol) && nextChar != '\n')
0910:                            return false;
0911:                        if (!__multiline && __eol - input > 1)
0912:                            return false;
0913:                        break;
0914:
0915:                    case OpCode._MEOL:
0916:                        if ((inputRemains || input < __eol) && nextChar != '\n')
0917:                            return false;
0918:                        break;
0919:
0920:                    case OpCode._SEOL:
0921:                        if ((inputRemains || input < __eol) && nextChar != '\n')
0922:                            return false;
0923:                        if (__eol - input > 1)
0924:                            return false;
0925:                        break;
0926:
0927:                    case OpCode._SANY:
0928:                        if (!inputRemains && input >= __eol)
0929:                            return false;
0930:                        inputRemains = (++input < __endOffset);
0931:                        nextChar = (inputRemains ? __input[input] : __EOS);
0932:                        break;
0933:
0934:                    case OpCode._ANY:
0935:                        if ((!inputRemains && input >= __eol)
0936:                                || nextChar == '\n')
0937:                            return false;
0938:                        inputRemains = (++input < __endOffset);
0939:                        nextChar = (inputRemains ? __input[input] : __EOS);
0940:                        break;
0941:
0942:                    case OpCode._EXACTLY:
0943:                        current = OpCode._getOperand(scan);
0944:                        line = __program[current++];
0945:
0946:                        if (__program[current] != nextChar)
0947:                            return false;
0948:                        if (__eol - input < line)
0949:                            return false;
0950:
0951:                        if (line > 1
0952:                                && !__compare(__program, current, __input,
0953:                                        input, line))
0954:                            return false;
0955:
0956:                        input += line;
0957:                        inputRemains = (input < __endOffset);
0958:                        nextChar = (inputRemains ? __input[input] : __EOS);
0959:                        break;
0960:
0961:                    case OpCode._ANYOF:
0962:                        current = OpCode._getOperand(scan);
0963:
0964:                        if (nextChar == __EOS && inputRemains)
0965:                            nextChar = __input[input];
0966:
0967:                        if (nextChar >= 256
0968:                                || (__program[current + (nextChar >> 4)] & (1 << (nextChar & 0xf))) != 0)
0969:                            return false;
0970:
0971:                        if (!inputRemains && input >= __eol)
0972:                            return false;
0973:
0974:                        inputRemains = (++input < __endOffset);
0975:                        nextChar = (inputRemains ? __input[input] : __EOS);
0976:                        break;
0977:
0978:                    case OpCode._ANYOFUN:
0979:                    case OpCode._NANYOFUN:
0980:                        current = OpCode._getOperand(scan);
0981:
0982:                        if (nextChar == __EOS && inputRemains)
0983:                            nextChar = __input[input];
0984:
0985:                        if (!__matchUnicodeClass(nextChar, __program, current,
0986:                                op))
0987:                            return false;
0988:
0989:                        if (!inputRemains && input >= __eol)
0990:                            return false;
0991:
0992:                        inputRemains = (++input < __endOffset);
0993:                        nextChar = (inputRemains ? __input[input] : __EOS);
0994:                        break;
0995:
0996:                    case OpCode._ALNUM:
0997:                        if (!inputRemains)
0998:                            return false;
0999:                        if (!OpCode._isWordCharacter(nextChar))
1000:                            return false;
1001:                        inputRemains = (++input < __endOffset);
1002:                        nextChar = (inputRemains ? __input[input] : __EOS);
1003:                        break;
1004:
1005:                    case OpCode._NALNUM:
1006:                        if (!inputRemains && input >= __eol)
1007:                            return false;
1008:                        if (OpCode._isWordCharacter(nextChar))
1009:                            return false;
1010:                        inputRemains = (++input < __endOffset);
1011:                        nextChar = (inputRemains ? __input[input] : __EOS);
1012:                        break;
1013:
1014:                    case OpCode._NBOUND:
1015:                    case OpCode._BOUND:
1016:                        boolean a,
1017:                        b;
1018:
1019:                        if (input == __bol)
1020:                            a = OpCode._isWordCharacter(__previousChar);
1021:                        else
1022:                            a = OpCode._isWordCharacter(__input[input - 1]);
1023:
1024:                        b = OpCode._isWordCharacter(nextChar);
1025:
1026:                        if ((a == b) == (__program[scan] == OpCode._BOUND))
1027:                            return false;
1028:                        break;
1029:
1030:                    case OpCode._SPACE:
1031:                        if (!inputRemains && input >= __eol)
1032:                            return false;
1033:                        if (!Character.isWhitespace(nextChar))
1034:                            return false;
1035:                        inputRemains = (++input < __endOffset);
1036:                        nextChar = (inputRemains ? __input[input] : __EOS);
1037:                        break;
1038:
1039:                    case OpCode._NSPACE:
1040:                        if (!inputRemains)
1041:                            return false;
1042:                        if (Character.isWhitespace(nextChar))
1043:                            return false;
1044:                        inputRemains = (++input < __endOffset);
1045:                        nextChar = (inputRemains ? __input[input] : __EOS);
1046:                        break;
1047:
1048:                    case OpCode._DIGIT:
1049:                        if (!Character.isDigit(nextChar))
1050:                            return false;
1051:                        inputRemains = (++input < __endOffset);
1052:                        nextChar = (inputRemains ? __input[input] : __EOS);
1053:                        break;
1054:
1055:                    case OpCode._NDIGIT:
1056:                        if (!inputRemains && input >= __eol)
1057:                            return false;
1058:                        if (Character.isDigit(nextChar))
1059:                            return false;
1060:                        inputRemains = (++input < __endOffset);
1061:                        nextChar = (inputRemains ? __input[input] : __EOS);
1062:                        break;
1063:
1064:                    case OpCode._REF:
1065:                        arg = OpCode._getArg1(__program, scan);
1066:                        current = __beginMatchOffsets[arg];
1067:
1068:                        if (current == OpCode._NULL_OFFSET)
1069:                            return false;
1070:
1071:                        if (__endMatchOffsets[arg] == OpCode._NULL_OFFSET)
1072:                            return false;
1073:
1074:                        if (current == __endMatchOffsets[arg])
1075:                            break;
1076:
1077:                        if (__input[current] != nextChar)
1078:                            return false;
1079:
1080:                        line = __endMatchOffsets[arg] - current;
1081:
1082:                        if (input + line > __eol)
1083:                            return false;
1084:
1085:                        if (line > 1
1086:                                && !__compare(__input, current, __input, input,
1087:                                        line))
1088:                            return false;
1089:
1090:                        input += line;
1091:                        inputRemains = (input < __endOffset);
1092:                        nextChar = (inputRemains ? __input[input] : __EOS);
1093:                        break;
1094:
1095:                    case OpCode._NOTHING:
1096:                        break;
1097:
1098:                    case OpCode._BACK:
1099:                        break;
1100:
1101:                    case OpCode._OPEN:
1102:                        arg = OpCode._getArg1(__program, scan);
1103:                        __beginMatchOffsets[arg] = input;
1104:
1105:                        if (arg > __expSize)
1106:                            __expSize = arg;
1107:                        break;
1108:
1109:                    case OpCode._CLOSE:
1110:                        arg = OpCode._getArg1(__program, scan);
1111:                        __endMatchOffsets[arg] = input;
1112:
1113:                        if (arg > __lastParen)
1114:                            __lastParen = arg;
1115:                        break;
1116:
1117:                    case OpCode._CURLYX:
1118:                        rep = new Perl5Repetition();
1119:                        rep._lastRepetition = __currentRep;
1120:                        __currentRep = rep;
1121:
1122:                        rep._parenFloor = __lastParen;
1123:                        rep._numInstances = -1;
1124:                        rep._min = OpCode._getArg1(__program, scan);
1125:                        rep._max = OpCode._getArg2(__program, scan);
1126:                        rep._scan = OpCode._getNextOperator(scan) + 2;
1127:                        rep._next = next;
1128:                        rep._minMod = minMod;
1129:                        // Must initialize to -1 because if we initialize to 0 and are
1130:                        // at the beginning of the input the OpCode._WHILEM case will
1131:                        // not work right.
1132:                        rep._lastLocation = -1;
1133:                        __inputOffset = input;
1134:
1135:                        // use minMod as temporary
1136:                        minMod = __match(OpCode._getPrevOperator(next));
1137:
1138:                        // leave scope call not pertinent?
1139:                        __currentRep = rep._lastRepetition;
1140:                        return minMod;
1141:
1142:                    case OpCode._WHILEM:
1143:                        rep = __currentRep;
1144:
1145:                        arg = rep._numInstances + 1;
1146:                        __inputOffset = input;
1147:
1148:                        if (input == rep._lastLocation) {
1149:                            __currentRep = rep._lastRepetition;
1150:                            line = __currentRep._numInstances;
1151:                            if (__match(rep._next))
1152:                                return true;
1153:                            __currentRep._numInstances = line;
1154:                            __currentRep = rep;
1155:                            return false;
1156:                        }
1157:
1158:                        if (arg < rep._min) {
1159:                            rep._numInstances = arg;
1160:                            rep._lastLocation = input;
1161:                            if (__match(rep._scan))
1162:                                return true;
1163:                            rep._numInstances = arg - 1;
1164:                            return false;
1165:                        }
1166:
1167:                        if (rep._minMod) {
1168:                            __currentRep = rep._lastRepetition;
1169:                            line = __currentRep._numInstances;
1170:                            if (__match(rep._next))
1171:                                return true;
1172:                            __currentRep._numInstances = line;
1173:                            __currentRep = rep;
1174:
1175:                            if (arg >= rep._max)
1176:                                return false;
1177:
1178:                            __inputOffset = input;
1179:                            rep._numInstances = arg;
1180:                            rep._lastLocation = input;
1181:
1182:                            if (__match(rep._scan))
1183:                                return true;
1184:
1185:                            rep._numInstances = arg - 1;
1186:                            return false;
1187:                        }
1188:
1189:                        if (arg < rep._max) {
1190:                            __pushState(rep._parenFloor);
1191:                            rep._numInstances = arg;
1192:                            rep._lastLocation = input;
1193:                            if (__match(rep._scan))
1194:                                return true;
1195:                            __popState();
1196:                            __inputOffset = input;
1197:                        }
1198:
1199:                        __currentRep = rep._lastRepetition;
1200:                        line = __currentRep._numInstances;
1201:                        if (__match(rep._next))
1202:                            return true;
1203:
1204:                        rep._numInstances = line;
1205:                        __currentRep = rep;
1206:                        rep._numInstances = arg - 1;
1207:                        return false;
1208:
1209:                    case OpCode._BRANCH:
1210:                        if (__program[next] != OpCode._BRANCH)
1211:                            next = OpCode._getNextOperator(scan);
1212:                        else {
1213:                            int lastParen;
1214:
1215:                            lastParen = __lastParen;
1216:
1217:                            do {
1218:
1219:                                __inputOffset = input;
1220:
1221:                                if (__match(OpCode._getNextOperator(scan)))
1222:                                    return true;
1223:
1224:                                for (arg = __lastParen; arg > lastParen; --arg)
1225:                                    //__endMatchOffsets[arg] = 0;
1226:                                    __endMatchOffsets[arg] = OpCode._NULL_OFFSET;
1227:                                __lastParen = arg;
1228:
1229:                                scan = OpCode._getNext(__program, scan);
1230:                            } while (scan != OpCode._NULL_OFFSET
1231:                                    && __program[scan] == OpCode._BRANCH);
1232:                            return false;
1233:                        }
1234:
1235:                        break;
1236:
1237:                    case OpCode._MINMOD:
1238:                        minMod = true;
1239:                        break;
1240:
1241:                    case OpCode._CURLY:
1242:                    case OpCode._STAR:
1243:                    case OpCode._PLUS:
1244:                        if (op == OpCode._CURLY) {
1245:                            line = OpCode._getArg1(__program, scan);
1246:                            arg = OpCode._getArg2(__program, scan);
1247:                            scan = OpCode._getNextOperator(scan) + 2;
1248:                        } else if (op == OpCode._STAR) {
1249:                            line = 0;
1250:                            arg = Character.MAX_VALUE;
1251:                            scan = OpCode._getNextOperator(scan);
1252:                        } else {
1253:                            line = 1;
1254:                            arg = Character.MAX_VALUE;
1255:                            scan = OpCode._getNextOperator(scan);
1256:                        }
1257:
1258:                        if (__program[next] == OpCode._EXACTLY) {
1259:                            nextChar = __program[OpCode._getOperand(next) + 1];
1260:                            current = 0;
1261:                        } else {
1262:                            nextChar = __EOS;
1263:                            current = -1000;
1264:                        }
1265:                        __inputOffset = input;
1266:
1267:                        if (minMod) {
1268:                            minMod = false;
1269:
1270:                            if (line > 0 && __repeat(scan, line) < line)
1271:                                return false;
1272:
1273:                            while (arg >= line
1274:                                    || (arg == Character.MAX_VALUE && line > 0)) {
1275:                                // there may be a bug here with respect to
1276:                                // __inputOffset >= __endOffset, but it seems to be right for
1277:                                // now.  the issue is with __inputOffset being reset later.
1278:                                // is this test really supposed to happen here?
1279:                                if (current == -1000
1280:                                        || __inputOffset >= __endOffset
1281:                                        || __input[__inputOffset] == nextChar) {
1282:                                    if (__match(next))
1283:                                        return true;
1284:                                }
1285:
1286:                                __inputOffset = input + line;
1287:
1288:                                if (__repeat(scan, 1) != 0) {
1289:                                    ++line;
1290:                                    __inputOffset = input + line;
1291:                                } else
1292:                                    return false;
1293:                            }
1294:
1295:                        } else {
1296:                            arg = __repeat(scan, arg);
1297:
1298:                            if (line < arg
1299:                                    && OpCode._opType[__program[next]] == OpCode._EOL
1300:                                    && ((!__multiline && __program[next] != OpCode._MEOL) || __program[next] == OpCode._SEOL))
1301:                                line = arg;
1302:
1303:                            while (arg >= line) {
1304:                                // there may be a bug here with respect to
1305:                                // __inputOffset >= __endOffset, but it seems to be right for
1306:                                // now.  the issue is with __inputOffset being reset later.
1307:                                // is this test really supposed to happen here?
1308:                                if (current == -1000
1309:                                        || __inputOffset >= __endOffset
1310:                                        || __input[__inputOffset] == nextChar) {
1311:                                    if (__match(next))
1312:                                        return true;
1313:                                }
1314:
1315:                                --arg;
1316:                                __inputOffset = input + arg;
1317:                            }
1318:                        }
1319:
1320:                        return false;
1321:
1322:                    case OpCode._SUCCEED:
1323:                    case OpCode._END:
1324:                        __inputOffset = input;
1325:                        // This enforces the rule that two consecutive matches cannot have
1326:                        // the same end offset.
1327:                        if (__inputOffset == __lastMatchInputEndOffset)
1328:                            return false;
1329:                        return true;
1330:
1331:                    case OpCode._IFMATCH:
1332:                        __inputOffset = input;
1333:                        scan = OpCode._getNextOperator(scan);
1334:                        if (!__match(scan))
1335:                            return false;
1336:                        break;
1337:
1338:                    case OpCode._UNLESSM:
1339:                        __inputOffset = input;
1340:                        scan = OpCode._getNextOperator(scan);
1341:                        if (__match(scan))
1342:                            return false;
1343:                        break;
1344:
1345:                    default:
1346:                        // todo: Need to throw an exception here.
1347:
1348:                    } // end switch
1349:
1350:                    //scan = (next > 0 ? next : 0);
1351:                    scan = next;
1352:                } // end while scan
1353:
1354:                return false;
1355:            }
1356:
1357:            /**
1358:             * Set whether or not subsequent calls to {@link #matches matches()}
1359:             * or {@link #contains contains()} should treat the input as
1360:             * consisting of multiple lines.  The default behavior is for 
1361:             * input to be treated as consisting of multiple lines.  This method
1362:             * should only be called if the Perl5Pattern used for a match was
1363:             * compiled without either of the Perl5Compiler.MULTILINE_MASK or
1364:             * Perl5Compiler.SINGLELINE_MASK flags, and you want to alter the
1365:             * behavior of how the <b>^</b>, <b>$</b>, and <b>.</b> metacharacters are
1366:             * interpreted on the fly.  The compilation options used when compiling
1367:             * a pattern ALWAYS override the behavior specified by setMultiline().  See
1368:             * {@link Perl5Compiler} for more details.
1369:             * <p>
1370:             * @param multiline  If set to true treats the input as consisting of
1371:             *        multiple lines with respect to the <b>^</b> and <b>$</b> 
1372:             *        metacharacters.  If set to false treats the input as consisting
1373:             *        of a single line with respect to the <b>^</b> and <b>$</b> 
1374:             *        metacharacters.
1375:             */
1376:            public void setMultiline(boolean multiline) {
1377:                __multiline = multiline;
1378:            }
1379:
1380:            /**
1381:             * @return True if the matcher is treating input as consisting of multiple
1382:             *         lines with respect to the <b>^</b> and <b>$</b> metacharacters,
1383:             *         false otherwise.
1384:             */
1385:            public boolean isMultiline() {
1386:                return __multiline;
1387:            }
1388:
1389:            char[] _toLower(char[] input) {
1390:                int current;
1391:                char[] inp;
1392:                // todo:
1393:                // Certainly not the best way to do case insensitive matching.
1394:                // Must definitely change this in some way, but for now we
1395:                // do what Perl does and make a copy of the input, converting
1396:                // it all to lowercase.  This is truly better handled in the
1397:                // compilation phase.
1398:                inp = new char[input.length];
1399:                System.arraycopy(input, 0, inp, 0, input.length);
1400:                input = inp;
1401:
1402:                // todo: Need to inline toLowerCase()
1403:                for (current = 0; current < input.length; current++)
1404:                    if (Character.isUpperCase(input[current]))
1405:                        input[current] = Character.toLowerCase(input[current]);
1406:
1407:                return input;
1408:            }
1409:
1410:            /**
1411:             * Determines if a prefix of a string (represented as a char[])
1412:             * matches a given pattern, starting from a given offset into the string.
1413:             * If a prefix of the string matches the pattern, a MatchResult instance
1414:             * representing the match is made accesible via
1415:             * {@link #getMatch()}.
1416:             * <p>
1417:             * This method is useful for certain common token identification tasks
1418:             * that are made more difficult without this functionality.
1419:             * <p>
1420:             * @param input  The char[] to test for a prefix match.
1421:             * @param pattern  The Pattern to be matched.
1422:             * @param offset The offset at which to start searching for the prefix.
1423:             * @return True if input matches pattern, false otherwise.
1424:             */
1425:            public boolean matchesPrefix(char[] input, Pattern pattern,
1426:                    int offset) {
1427:                Perl5Pattern expression;
1428:
1429:                expression = (Perl5Pattern) pattern;
1430:                __originalInput = input;
1431:                if (expression._isCaseInsensitive)
1432:                    input = _toLower(input);
1433:
1434:                __initInterpreterGlobals(expression, input, 0, input.length,
1435:                        offset);
1436:
1437:                __lastSuccess = __tryExpression(offset);
1438:                __lastMatchResult = null;
1439:
1440:                return __lastSuccess;
1441:            }
1442:
1443:            /**
1444:             * Determines if a prefix of a string (represented as a char[])
1445:             * matches a given pattern.
1446:             * If a prefix of the string matches the pattern, a MatchResult instance
1447:             * representing the match is made accesible via
1448:             * {@link #getMatch()}.
1449:             * <p>
1450:             * This method is useful for certain common token identification tasks
1451:             * that are made more difficult without this functionality.
1452:             * <p>
1453:             * @param input  The char[] to test for a prefix match.
1454:             * @param pattern  The Pattern to be matched.
1455:             * @return True if input matches pattern, false otherwise.
1456:             */
1457:            public boolean matchesPrefix(char[] input, Pattern pattern) {
1458:                return matchesPrefix(input, pattern, 0);
1459:            }
1460:
1461:            /**
1462:             * Determines if a prefix of a string matches a given pattern.
1463:             * If a prefix of the string matches the pattern, a MatchResult instance
1464:             * representing the match is made accesible via
1465:             * {@link #getMatch()}.
1466:             * <p>
1467:             * This method is useful for certain common token identification tasks
1468:             * that are made more difficult without this functionality.
1469:             * <p>
1470:             * @param input  The String to test for a prefix match.
1471:             * @param pattern  The Pattern to be matched.
1472:             * @return True if input matches pattern, false otherwise.
1473:             */
1474:            public boolean matchesPrefix(String input, Pattern pattern) {
1475:                return matchesPrefix(input.toCharArray(), pattern, 0);
1476:            }
1477:
1478:            /**
1479:             * Determines if a prefix of a PatternMatcherInput instance
1480:             * matches a given pattern.  If there is a match, a MatchResult instance
1481:             * representing the match is made accesible via
1482:             * {@link #getMatch()}.  Unlike the
1483:             * {@link #contains(PatternMatcherInput, Pattern)}
1484:             * method, the current offset of the PatternMatcherInput argument
1485:             * is not updated.  However, unlike the
1486:             * {@link #matches matches(PatternMatcherInput, Pattern)} method,
1487:             * matchesPrefix() will start its search from the current offset
1488:             * rather than the begin offset of the PatternMatcherInput.
1489:             * <p>
1490:             * This method is useful for certain common token identification tasks
1491:             * that are made more difficult without this functionality.
1492:             * <p>
1493:             * @param input  The PatternMatcherInput to test for a prefix match.
1494:             * @param pattern  The Pattern to be matched.
1495:             * @return True if input matches pattern, false otherwise.
1496:             */
1497:            public boolean matchesPrefix(PatternMatcherInput input,
1498:                    Pattern pattern) {
1499:                char[] inp;
1500:                Perl5Pattern expression;
1501:
1502:                expression = (Perl5Pattern) pattern;
1503:
1504:                __originalInput = input._originalBuffer;
1505:                if (expression._isCaseInsensitive) {
1506:                    if (input._toLowerBuffer == null)
1507:                        input._toLowerBuffer = _toLower(__originalInput);
1508:                    inp = input._toLowerBuffer;
1509:                } else
1510:                    inp = __originalInput;
1511:
1512:                __initInterpreterGlobals(expression, inp, input._beginOffset,
1513:                        input._endOffset, input._currentOffset);
1514:                __lastSuccess = __tryExpression(input._currentOffset);
1515:                __lastMatchResult = null;
1516:
1517:                return __lastSuccess;
1518:            }
1519:
1520:            /**
1521:             * Determines if a string (represented as a char[]) exactly 
1522:             * matches a given pattern.  If
1523:             * there is an exact match, a MatchResult instance
1524:             * representing the match is made accesible via
1525:             * {@link #getMatch()}.  The pattern must be
1526:             * a Perl5Pattern instance, otherwise a ClassCastException will
1527:             * be thrown.  You are not required to, and indeed should NOT try to
1528:             * (for performance reasons), catch a ClassCastException because it
1529:             * will never be thrown as long as you use a Perl5Pattern as the pattern
1530:             * parameter.
1531:             * <p>
1532:             * <b>Note:</b> matches() is not the same as sticking a ^ in front of
1533:             * your expression and a $ at the end of your expression in Perl5
1534:             * and using the =~ operator, even though in many cases it will be
1535:             * equivalent.  matches() literally looks for an exact match according
1536:             * to the rules of Perl5 expression matching.  Therefore, if you have
1537:             * a pattern <em>foo|foot</em> and are matching the input <em>foot</em>
1538:             * it will not produce an exact match.  But <em>foot|foo</em> will
1539:             * produce an exact match for either <em>foot</em> or <em>foo</em>.
1540:             * Remember, Perl5 regular expressions do not match the longest
1541:             * possible match.  From the perlre manpage:
1542:             * <blockquote>
1543:             *   Alternatives are tried from left to right, so the first
1544:             *   alternative found for which the entire expression matches,
1545:             *   is the one that is chosen. This means that alternatives
1546:             *   are not necessarily greedy. For example: when matching
1547:             *   foo|foot against "barefoot", only the "foo" part will
1548:             *   match, as that is the first alternative tried, and it
1549:             *   successfully matches the target string.
1550:             * </blockquote>
1551:             * <p>
1552:             * @param input  The char[] to test for an exact match.
1553:             * @param pattern  The Perl5Pattern to be matched.
1554:             * @return True if input matches pattern, false otherwise.
1555:             * @exception ClassCastException If a Pattern instance other than a
1556:             *         Perl5Pattern is passed as the pattern parameter.
1557:             */
1558:            public boolean matches(char[] input, Pattern pattern) {
1559:                Perl5Pattern expression;
1560:
1561:                expression = (Perl5Pattern) pattern;
1562:                __originalInput = input;
1563:                if (expression._isCaseInsensitive)
1564:                    input = _toLower(input);
1565:
1566:                __initInterpreterGlobals(expression, input, 0, input.length, 0);
1567:                __lastSuccess = (__tryExpression(0) && __endMatchOffsets[0] == input.length);
1568:                __lastMatchResult = null;
1569:
1570:                return __lastSuccess;
1571:            }
1572:
1573:            /**
1574:             * Determines if a string exactly matches a given pattern.  If
1575:             * there is an exact match, a MatchResult instance
1576:             * representing the match is made accesible via
1577:             * {@link #getMatch()}.  The pattern must be
1578:             * a Perl5Pattern instance, otherwise a ClassCastException will
1579:             * be thrown.  You are not required to, and indeed should NOT try to
1580:             * (for performance reasons), catch a ClassCastException because it
1581:             * will never be thrown as long as you use a Perl5Pattern as the pattern
1582:             * parameter.
1583:             * <p>
1584:             * <b>Note:</b> matches() is not the same as sticking a ^ in front of
1585:             * your expression and a $ at the end of your expression in Perl5
1586:             * and using the =~ operator, even though in many cases it will be
1587:             * equivalent.  matches() literally looks for an exact match according
1588:             * to the rules of Perl5 expression matching.  Therefore, if you have
1589:             * a pattern <em>foo|foot</em> and are matching the input <em>foot</em>
1590:             * it will not produce an exact match.  But <em>foot|foo</em> will
1591:             * produce an exact match for either <em>foot</em> or <em>foo</em>.
1592:             * Remember, Perl5 regular expressions do not match the longest
1593:             * possible match.  From the perlre manpage:
1594:             * <blockquote>
1595:             *   Alternatives are tried from left to right, so the first
1596:             *   alternative found for which the entire expression matches,
1597:             *   is the one that is chosen. This means that alternatives
1598:             *   are not necessarily greedy. For example: when matching
1599:             *   foo|foot against "barefoot", only the "foo" part will
1600:             *   match, as that is the first alternative tried, and it
1601:             *   successfully matches the target string.
1602:             * </blockquote>
1603:             * <p>
1604:             * @param input  The String to test for an exact match.
1605:             * @param pattern  The Perl5Pattern to be matched.
1606:             * @return True if input matches pattern, false otherwise.
1607:             * @exception ClassCastException If a Pattern instance other than a
1608:             *         Perl5Pattern is passed as the pattern parameter.
1609:             */
1610:            public boolean matches(String input, Pattern pattern) {
1611:                return matches(input.toCharArray(), pattern);
1612:            }
1613:
1614:            /**
1615:             * Determines if the contents of a PatternMatcherInput instance
1616:             * exactly matches a given pattern.  If
1617:             * there is an exact match, a MatchResult instance
1618:             * representing the match is made accesible via
1619:             * {@link #getMatch()}.  Unlike the
1620:             * {@link #contains(PatternMatcherInput, Pattern)}
1621:             * method, the current offset of the PatternMatcherInput argument
1622:             * is not updated.  You should remember that the region between
1623:             * the begin (NOT the current) and end offsets of the PatternMatcherInput
1624:             * will be tested for an exact match.
1625:             * <p>
1626:             * The pattern must be a Perl5Pattern instance, otherwise a
1627:             * ClassCastException will be thrown.  You are not required to, and 
1628:             * indeed should NOT try to (for performance reasons), catch a
1629:             * ClassCastException because it will never be thrown as long as you use
1630:             * a Perl5Pattern as the pattern parameter.
1631:             * <p>
1632:             * <b>Note:</b> matches() is not the same as sticking a ^ in front of
1633:             * your expression and a $ at the end of your expression in Perl5
1634:             * and using the =~ operator, even though in many cases it will be
1635:             * equivalent.  matches() literally looks for an exact match according
1636:             * to the rules of Perl5 expression matching.  Therefore, if you have
1637:             * a pattern <em>foo|foot</em> and are matching the input <em>foot</em>
1638:             * it will not produce an exact match.  But <em>foot|foo</em> will
1639:             * produce an exact match for either <em>foot</em> or <em>foo</em>.
1640:             * Remember, Perl5 regular expressions do not match the longest
1641:             * possible match.  From the perlre manpage:
1642:             * <blockquote>
1643:             *   Alternatives are tried from left to right, so the first
1644:             *   alternative found for which the entire expression matches,
1645:             *   is the one that is chosen. This means that alternatives
1646:             *   are not necessarily greedy. For example: when matching
1647:             *   foo|foot against "barefoot", only the "foo" part will
1648:             *   match, as that is the first alternative tried, and it
1649:             *   successfully matches the target string.
1650:             * </blockquote>
1651:             * <p>
1652:             * @param input  The PatternMatcherInput to test for a match.
1653:             * @param pattern  The Perl5Pattern to be matched.
1654:             * @return True if input matches pattern, false otherwise.
1655:             * @exception ClassCastException If a Pattern instance other than a
1656:             *         Perl5Pattern is passed as the pattern parameter.
1657:             */
1658:            public boolean matches(PatternMatcherInput input, Pattern pattern) {
1659:                char[] inp;
1660:                Perl5Pattern expression;
1661:
1662:                expression = (Perl5Pattern) pattern;
1663:
1664:                __originalInput = input._originalBuffer;
1665:                if (expression._isCaseInsensitive) {
1666:                    if (input._toLowerBuffer == null)
1667:                        input._toLowerBuffer = _toLower(__originalInput);
1668:                    inp = input._toLowerBuffer;
1669:                } else
1670:                    inp = __originalInput;
1671:
1672:                __initInterpreterGlobals(expression, inp, input._beginOffset,
1673:                        input._endOffset, input._beginOffset);
1674:
1675:                __lastMatchResult = null;
1676:
1677:                if (__tryExpression(input._beginOffset)) {
1678:                    if (__endMatchOffsets[0] == input._endOffset
1679:                            || input.length() == 0
1680:                            || input._beginOffset == input._endOffset) {
1681:                        __lastSuccess = true;
1682:                        return true;
1683:                    }
1684:                }
1685:
1686:                __lastSuccess = false;
1687:
1688:                return false;
1689:            }
1690:
1691:            /**
1692:             * Determines if a string contains a pattern.  If the pattern is
1693:             * matched by some substring of the input, a MatchResult instance
1694:             * representing the <b> first </b> such match is made acessible via 
1695:             * {@link #getMatch()}.  If you want to access
1696:             * subsequent matches you should either use a PatternMatcherInput object
1697:             * or use the offset information in the MatchResult to create a substring
1698:             * representing the remaining input.  Using the MatchResult offset 
1699:             * information is the recommended method of obtaining the parts of the
1700:             * string preceeding the match and following the match.
1701:             * <p>
1702:             * The pattern must be a Perl5Pattern instance, otherwise a
1703:             * ClassCastException will be thrown.  You are not required to, and 
1704:             * indeed should NOT try to (for performance reasons), catch a
1705:             * ClassCastException because it will never be thrown as long as you use
1706:             * a Perl5Pattern as the pattern parameter.
1707:             * <p>
1708:             * @param input  The String to test for a match.
1709:             * @param pattern  The Perl5Pattern to be matched.
1710:             * @return True if the input contains a pattern match, false otherwise.
1711:             * @exception ClassCastException If a Pattern instance other than a
1712:             *         Perl5Pattern is passed as the pattern parameter.
1713:             */
1714:            public boolean contains(String input, Pattern pattern) {
1715:                return contains(input.toCharArray(), pattern);
1716:            }
1717:
1718:            /**
1719:             * Determines if a string (represented as a char[]) contains a pattern.
1720:             * If the pattern is
1721:             * matched by some substring of the input, a MatchResult instance
1722:             * representing the <b> first </b> such match is made acessible via 
1723:             * {@link #getMatch()}.  If you want to access
1724:             * subsequent matches you should either use a PatternMatcherInput object
1725:             * or use the offset information in the MatchResult to create a substring
1726:             * representing the remaining input.  Using the MatchResult offset 
1727:             * information is the recommended method of obtaining the parts of the
1728:             * string preceeding the match and following the match.
1729:             * <p>
1730:             * The pattern must be a Perl5Pattern instance, otherwise a
1731:             * ClassCastException will be thrown.  You are not required to, and 
1732:             * indeed should NOT try to (for performance reasons), catch a
1733:             * ClassCastException because it will never be thrown as long as you use
1734:             * a Perl5Pattern as the pattern parameter.
1735:             * <p>
1736:             * @param input  The char[] to test for a match.
1737:             * @param pattern  The Perl5Pattern to be matched.
1738:             * @return True if the input contains a pattern match, false otherwise.
1739:             * @exception ClassCastException If a Pattern instance other than a
1740:             *         Perl5Pattern is passed as the pattern parameter.
1741:             */
1742:            public boolean contains(char[] input, Pattern pattern) {
1743:                Perl5Pattern expression;
1744:
1745:                expression = (Perl5Pattern) pattern;
1746:
1747:                __originalInput = input;
1748:
1749:                if (expression._isCaseInsensitive)
1750:                    input = _toLower(input);
1751:
1752:                return __interpret(expression, input, 0, input.length, 0);
1753:            }
1754:
1755:            private static final int __DEFAULT_LAST_MATCH_END_OFFSET = -100;
1756:            private int __lastMatchInputEndOffset = __DEFAULT_LAST_MATCH_END_OFFSET;
1757:
1758:            /**
1759:             * Determines if the contents of a PatternMatcherInput, starting from the
1760:             * current offset of the input contains a pattern.
1761:             * If a pattern match is found, a MatchResult
1762:             * instance representing the <b>first</b> such match is made acessible via 
1763:             * {@link #getMatch()}.  The current offset of the
1764:             * PatternMatcherInput is set to the offset corresponding to the end
1765:             * of the match, so that a subsequent call to this method will continue
1766:             * searching where the last call left off.  You should remember that the
1767:             * region between the begin and end offsets of the PatternMatcherInput are
1768:             * considered the input to be searched, and that the current offset
1769:             * of the PatternMatcherInput reflects where a search will start from.
1770:             * Matches extending beyond the end offset of the PatternMatcherInput
1771:             * will not be matched.  In other words, a match must occur entirely
1772:             * between the begin and end offsets of the input.  See
1773:             * {@link PatternMatcherInput} for more details.
1774:             * <p>
1775:             * As a side effect, if a match is found, the PatternMatcherInput match
1776:             * offset information is updated.  See the
1777:             * {@link PatternMatcherInput#setMatchOffsets(int, int)}
1778:             * method for more details.
1779:             * <p>
1780:             * The pattern must be a Perl5Pattern instance, otherwise a
1781:             * ClassCastException will be thrown.  You are not required to, and 
1782:             * indeed should NOT try to (for performance reasons), catch a
1783:             * ClassCastException because it will never be thrown as long as you use
1784:             * a Perl5Pattern as the pattern parameter.
1785:             * <p>
1786:             * This method is usually used in a loop as follows:
1787:             * <blockquote><pre>
1788:             * PatternMatcher matcher;
1789:             * PatternCompiler compiler;
1790:             * Pattern pattern;
1791:             * PatternMatcherInput input;
1792:             * MatchResult result;
1793:             *
1794:             * compiler = new Perl5Compiler();
1795:             * matcher  = new Perl5Matcher();
1796:             *
1797:             * try {
1798:             *   pattern = compiler.compile(somePatternString);
1799:             * } catch(MalformedPatternException e) {
1800:             *   System.err.println("Bad pattern.");
1801:             *   System.err.println(e.getMessage());
1802:             *   return;
1803:             * }
1804:             *
1805:             * input   = new PatternMatcherInput(someStringInput);
1806:             *
1807:             * while(matcher.contains(input, pattern)) {
1808:             *   result = matcher.getMatch();  
1809:             *   // Perform whatever processing on the result you want.
1810:             * }
1811:             *
1812:             * </pre></blockquote>
1813:             * <p>
1814:             * @param input  The PatternMatcherInput to test for a match.
1815:             * @param pattern  The Pattern to be matched.
1816:             * @return True if the input contains a pattern match, false otherwise.
1817:             * @exception ClassCastException If a Pattern instance other than a
1818:             *         Perl5Pattern is passed as the pattern parameter.
1819:             */
1820:            public boolean contains(PatternMatcherInput input, Pattern pattern) {
1821:                char[] inp;
1822:                Perl5Pattern expression;
1823:                boolean matchFound;
1824:
1825:                //if(input.length() > 0) {
1826:                // We want to allow a null string to match at the end of the input
1827:                // which is why we don't check endOfInput.  Not sure if this is a
1828:                // safe thing to do or not.
1829:                if (input._currentOffset > input._endOffset)
1830:                    return false;
1831:                //}
1832:                /* else 
1833:                  if(input._endOfInput())
1834:                return false;
1835:                 */
1836:                expression = (Perl5Pattern) pattern;
1837:                __originalInput = input._originalBuffer;
1838:
1839:                // Todo:
1840:                // Really should only reduce to lowercase that part of the
1841:                // input that is necessary, instead of the whole thing.
1842:                // Adjust MatchResult offsets accordingly.  Actually, pass an adjustment
1843:                // value to __interpret.
1844:                __originalInput = input._originalBuffer;
1845:                if (expression._isCaseInsensitive) {
1846:                    if (input._toLowerBuffer == null)
1847:                        input._toLowerBuffer = _toLower(__originalInput);
1848:                    inp = input._toLowerBuffer;
1849:                } else
1850:                    inp = __originalInput;
1851:
1852:                __lastMatchInputEndOffset = input.getMatchEndOffset();
1853:
1854:                matchFound = __interpret(expression, inp, input._beginOffset,
1855:                        input._endOffset, input._currentOffset);
1856:
1857:                if (matchFound) {
1858:                    input.setCurrentOffset(__endMatchOffsets[0]);
1859:                    input.setMatchOffsets(__beginMatchOffsets[0],
1860:                            __endMatchOffsets[0]);
1861:                } else {
1862:                    input.setCurrentOffset(input._endOffset + 1);
1863:                }
1864:
1865:                // Restore so it doesn't interfere with other unrelated matches.
1866:                __lastMatchInputEndOffset = __DEFAULT_LAST_MATCH_END_OFFSET;
1867:
1868:                return matchFound;
1869:            }
1870:
1871:            /**
1872:             * Fetches the last match found by a call to a matches() or contains()
1873:             * method.  If you plan on modifying the original search input, you
1874:             * must call this method BEFORE you modify the original search input,
1875:             * as a lazy evaluation technique is used to create the MatchResult.
1876:             * This reduces the cost of pattern matching when you don't care about
1877:             * the actual match and only care if the pattern occurs in the input.
1878:             * Otherwise, a MatchResult would be created for every match found,
1879:             * whether or not the MatchResult was later used by a call to getMatch().
1880:             * <p>
1881:             * @return A MatchResult instance containing the pattern match found
1882:             *         by the last call to any one of the matches() or contains()
1883:             *         methods.  If no match was found by the last call, returns
1884:             *         null. 
1885:             */
1886:            public MatchResult getMatch() {
1887:                if (!__lastSuccess)
1888:                    return null;
1889:
1890:                if (__lastMatchResult == null)
1891:                    __setLastMatchResult();
1892:
1893:                return __lastMatchResult;
1894:            }
1895:
1896:        }
www.java2java.com | Contact Us
Copyright 2009 - 12 Demo Source and Support. All rights reserved.
All other trademarks are property of their respective owners.