net.sourceforge.phpeclipse/src/net/sourceforge/phpdt/internal/compiler/parser/Scanner.java

   1 /***********************************************************************************************************************************
   2  * Copyright (c) 2000, 2001, 2002 International Business Machines Corp. and others. All rights reserved. This program and the
   3  * accompanying materials are made available under the terms of the Common Public License v0.5 which accompanies this distribution,
   4  * and is available at http://www.eclipse.org/legal/cpl-v05.html
   5  *
   6  * Contributors: IBM Corporation - initial API and implementation
   7  **********************************************************************************************************************************/
   8 package net.sourceforge.phpdt.internal.compiler.parser;
   9
  10 import java.util.ArrayList;
  11 import java.util.Iterator;
  12 import java.util.List;
  13
  14 import net.sourceforge.phpdt.core.compiler.CharOperation;
  15 import net.sourceforge.phpdt.core.compiler.IScanner;
  16 import net.sourceforge.phpdt.core.compiler.ITerminalSymbols;
  17 import net.sourceforge.phpdt.core.compiler.InvalidInputException;
  18 import net.sourceforge.phpdt.internal.compiler.ast.StringLiteral;
  19 import net.sourceforge.phpdt.internal.compiler.env.ICompilationUnit;
  20
  21 public class Scanner implements IScanner, ITerminalSymbols {
  22         /*
  23          * APIs ares - getNextToken() which return the current type of the token (this
  24          * value is not memorized by the scanner) - getCurrentTokenSource() which
  25          * provides with the token "REAL" source (aka all unicode have been
  26          * transformed into a correct char) - sourceStart gives the position into the
  27          * stream - currentPosition-1 gives the sourceEnd position into the stream
  28          */
  29         // 1.4 feature
  30 //      private boolean assertMode;
  31
  32         public boolean useAssertAsAnIndentifier = false;
  33
  34         // flag indicating if processed source contains occurrences of keyword assert
  35         public boolean containsAssertKeyword = false;
  36
  37         public boolean recordLineSeparator;
  38
  39         public boolean ignorePHPOneLiner = false;
  40
  41         public boolean phpMode = false;
  42
  43         public boolean phpExpressionTag = false;
  44
  45         // public Stack encapsedStringStack = null;
  46
  47         public char currentCharacter;
  48
  49         public int startPosition;
  50
  51         public int currentPosition;
  52
  53         public int initialPosition, eofPosition;
  54
  55         // after this position eof are generated instead of real token from the
  56         // source
  57         public boolean tokenizeComments;
  58
  59         public boolean tokenizeWhiteSpace;
  60
  61         public boolean tokenizeStrings;
  62
  63         // source should be viewed as a window (aka a part)
  64         // of a entire very large stream
  65         public char source[];
  66
  67         // unicode support
  68         public char[] withoutUnicodeBuffer;
  69
  70         public int withoutUnicodePtr;
  71
  72         // when == 0 ==> no unicode in the current token
  73         public boolean unicodeAsBackSlash = false;
  74
  75         public boolean scanningFloatLiteral = false;
  76
  77         // support for /** comments
  78         public int[] commentStops = new int[10];
  79
  80         public int[] commentStarts = new int[10];
  81
  82         public int commentPtr = -1; // no comment test with commentPtr value -1
  83
  84         protected int lastCommentLinePosition = -1;
  85
  86         // diet parsing support - jump over some method body when requested
  87         public boolean diet = false;
  88
  89         // support for the poor-line-debuggers ....
  90         // remember the position of the cr/lf
  91         public int[] lineEnds = new int[250];
  92
  93         public int linePtr = -1;
  94
  95         public boolean wasAcr = false;
  96
  97         public static final String END_OF_SOURCE = "End_Of_Source"; //$NON-NLS-1$
  98
  99         public static final String INVALID_HEXA = "Invalid_Hexa_Literal"; //$NON-NLS-1$
 100
 101         public static final String INVALID_OCTAL = "Invalid_Octal_Literal"; //$NON-NLS-1$
 102
 103         public static final String INVALID_CHARACTER_CONSTANT = "Invalid_Character_Constant"; //$NON-NLS-1$
 104
 105         public static final String INVALID_ESCAPE = "Invalid_Escape"; //$NON-NLS-1$
 106
 107         public static final String INVALID_INPUT = "Invalid_Input"; //$NON-NLS-1$
 108
 109         public static final String INVALID_UNICODE_ESCAPE = "Invalid_Unicode_Escape"; //$NON-NLS-1$
 110
 111         public static final String INVALID_FLOAT = "Invalid_Float_Literal"; //$NON-NLS-1$
 112
 113         public static final String NULL_SOURCE_STRING = "Null_Source_String"; //$NON-NLS-1$
 114
 115         public static final String UNTERMINATED_STRING = "Unterminated_String"; //$NON-NLS-1$
 116
 117         public static final String UNTERMINATED_COMMENT = "Unterminated_Comment"; //$NON-NLS-1$
 118
 119         public static final String INVALID_CHAR_IN_STRING = "Invalid_Char_In_String"; //$NON-NLS-1$
 120
 121         // ----------------optimized identifier managment------------------
 122         static final char[] charArray_a = new char[] { 'a' }, charArray_b = new char[] { 'b' }, charArray_c = new char[] { 'c' },
 123                         charArray_d = new char[] { 'd' }, charArray_e = new char[] { 'e' }, charArray_f = new char[] { 'f' },
 124                         charArray_g = new char[] { 'g' }, charArray_h = new char[] { 'h' }, charArray_i = new char[] { 'i' },
 125                         charArray_j = new char[] { 'j' }, charArray_k = new char[] { 'k' }, charArray_l = new char[] { 'l' },
 126                         charArray_m = new char[] { 'm' }, charArray_n = new char[] { 'n' }, charArray_o = new char[] { 'o' },
 127                         charArray_p = new char[] { 'p' }, charArray_q = new char[] { 'q' }, charArray_r = new char[] { 'r' },
 128                         charArray_s = new char[] { 's' }, charArray_t = new char[] { 't' }, charArray_u = new char[] { 'u' },
 129                         charArray_v = new char[] { 'v' }, charArray_w = new char[] { 'w' }, charArray_x = new char[] { 'x' },
 130                         charArray_y = new char[] { 'y' }, charArray_z = new char[] { 'z' };
 131
 132         static final char[] charArray_va = new char[] { '$', 'a' }, charArray_vb = new char[] { '$', 'b' }, charArray_vc = new char[] {
 133                         '$', 'c' }, charArray_vd = new char[] { '$', 'd' }, charArray_ve = new char[] { '$', 'e' }, charArray_vf = new char[] { '$',
 134                         'f' }, charArray_vg = new char[] { '$', 'g' }, charArray_vh = new char[] { '$', 'h' },
 135                         charArray_vi = new char[] { '$', 'i' }, charArray_vj = new char[] { '$', 'j' }, charArray_vk = new char[] { '$', 'k' },
 136                         charArray_vl = new char[] { '$', 'l' }, charArray_vm = new char[] { '$', 'm' }, charArray_vn = new char[] { '$', 'n' },
 137                         charArray_vo = new char[] { '$', 'o' }, charArray_vp = new char[] { '$', 'p' }, charArray_vq = new char[] { '$', 'q' },
 138                         charArray_vr = new char[] { '$', 'r' }, charArray_vs = new char[] { '$', 's' }, charArray_vt = new char[] { '$', 't' },
 139                         charArray_vu = new char[] { '$', 'u' }, charArray_vv = new char[] { '$', 'v' }, charArray_vw = new char[] { '$', 'w' },
 140                         charArray_vx = new char[] { '$', 'x' }, charArray_vy = new char[] { '$', 'y' }, charArray_vz = new char[] { '$', 'z' };
 141
 142         public final static int MAX_OBVIOUS = 256;
 143
 144         static final int[] ObviousIdentCharNatures = new int[MAX_OBVIOUS];
 145
 146         public final static int C_DOLLAR = 8;
 147
 148         public final static int C_LETTER = 4;
 149
 150         public final static int C_DIGIT = 3;
 151
 152         public final static int C_SEPARATOR = 2;
 153
 154         public final static int C_SPACE = 1;
 155         static {
 156                 for (int i = '0'; i <= '9'; i++)
 157                         ObviousIdentCharNatures[i] = C_DIGIT;
 158
 159                 for (int i = 'a'; i <= 'z'; i++)
 160                         ObviousIdentCharNatures[i] = C_LETTER;
 161                 for (int i = 'A'; i <= 'Z'; i++)
 162                         ObviousIdentCharNatures[i] = C_LETTER;
 163                 ObviousIdentCharNatures['_'] = C_LETTER;
 164                 for (int i = 127; i <= 255; i++)
 165                         ObviousIdentCharNatures[i] = C_LETTER;
 166
 167                 ObviousIdentCharNatures['$'] = C_DOLLAR;
 168
 169                 ObviousIdentCharNatures[10] = C_SPACE; // \ u000a: LINE FEED
 170                 ObviousIdentCharNatures[12] = C_SPACE; // \ u000c: FORM FEED
 171                 ObviousIdentCharNatures[13] = C_SPACE; // \ u000d: CARRIAGE RETURN
 172                 ObviousIdentCharNatures[32] = C_SPACE; // \ u0020: SPACE
 173                 ObviousIdentCharNatures[9] = C_SPACE; // \ u0009: HORIZONTAL TABULATION
 174
 175                 ObviousIdentCharNatures['.'] = C_SEPARATOR;
 176                 ObviousIdentCharNatures[':'] = C_SEPARATOR;
 177                 ObviousIdentCharNatures[';'] = C_SEPARATOR;
 178                 ObviousIdentCharNatures[','] = C_SEPARATOR;
 179                 ObviousIdentCharNatures['['] = C_SEPARATOR;
 180                 ObviousIdentCharNatures[']'] = C_SEPARATOR;
 181                 ObviousIdentCharNatures['('] = C_SEPARATOR;
 182                 ObviousIdentCharNatures[')'] = C_SEPARATOR;
 183                 ObviousIdentCharNatures['{'] = C_SEPARATOR;
 184                 ObviousIdentCharNatures['}'] = C_SEPARATOR;
 185                 ObviousIdentCharNatures['+'] = C_SEPARATOR;
 186                 ObviousIdentCharNatures['-'] = C_SEPARATOR;
 187                 ObviousIdentCharNatures['*'] = C_SEPARATOR;
 188                 ObviousIdentCharNatures['/'] = C_SEPARATOR;
 189                 ObviousIdentCharNatures['='] = C_SEPARATOR;
 190                 ObviousIdentCharNatures['&'] = C_SEPARATOR;
 191                 ObviousIdentCharNatures['|'] = C_SEPARATOR;
 192                 ObviousIdentCharNatures['?'] = C_SEPARATOR;
 193                 ObviousIdentCharNatures['<'] = C_SEPARATOR;
 194                 ObviousIdentCharNatures['>'] = C_SEPARATOR;
 195                 ObviousIdentCharNatures['!'] = C_SEPARATOR;
 196                 ObviousIdentCharNatures['%'] = C_SEPARATOR;
 197                 ObviousIdentCharNatures['^'] = C_SEPARATOR;
 198                 ObviousIdentCharNatures['~'] = C_SEPARATOR;
 199                 ObviousIdentCharNatures['"'] = C_SEPARATOR;
 200                 ObviousIdentCharNatures['\''] = C_SEPARATOR;
 201         }
 202         static final char[] initCharArray = new char[] { '\u0000', '\u0000', '\u0000', '\u0000', '\u0000', '\u0000' };
 203
 204         static final int TableSize = 30, InternalTableSize = 6;
 205
 206         // 30*6 = 180 entries
 207         public static final int OptimizedLength = 6;
 208
 209         public/* static */
 210         final char[][][][] charArray_length = new char[OptimizedLength][TableSize][InternalTableSize][];
 211
 212         // support for detecting non-externalized string literals
 213         int currentLineNr = -1;
 214
 215         int previousLineNr = -1;
 216
 217         NLSLine currentLine = null;
 218
 219         List lines = new ArrayList();
 220
 221         public static final String TAG_PREFIX = "//$NON-NLS-"; //$NON-NLS-1$
 222
 223         public static final int TAG_PREFIX_LENGTH = TAG_PREFIX.length();
 224
 225         public static final String TAG_POSTFIX = "$"; //$NON-NLS-1$
 226
 227         public static final int TAG_POSTFIX_LENGTH = TAG_POSTFIX.length();
 228
 229         public StringLiteral[] nonNLSStrings = null;
 230
 231         public boolean checkNonExternalizedStringLiterals = true;
 232
 233         public boolean wasNonExternalizedStringLiteral = false;
 234
 235         /* static */{
 236                 for (int i = 0; i < 6; i++) {
 237                         for (int j = 0; j < TableSize; j++) {
 238                                 for (int k = 0; k < InternalTableSize; k++) {
 239                                         charArray_length[i][j][k] = initCharArray;
 240                                 }
 241                         }
 242                 }
 243         }
 244
 245         static int newEntry2 = 0, newEntry3 = 0, newEntry4 = 0, newEntry5 = 0, newEntry6 = 0;
 246
 247         public static final int RoundBracket = 0;
 248
 249         public static final int SquareBracket = 1;
 250
 251         public static final int CurlyBracket = 2;
 252
 253         public static final int BracketKinds = 3;
 254
 255         // task tag support
 256         public char[][] foundTaskTags = null;
 257
 258         public char[][] foundTaskMessages;
 259
 260         public char[][] foundTaskPriorities = null;
 261
 262         public int[][] foundTaskPositions;
 263
 264         public int foundTaskCount = 0;
 265
 266         public char[][] taskTags = null;
 267
 268         public char[][] taskPriorities = null;
 269
 270         public boolean isTaskCaseSensitive = true;
 271
 272         public static final boolean DEBUG = false;
 273
 274         public static final boolean TRACE = false;
 275
 276         public ICompilationUnit compilationUnit = null;
 277
 278         /**
 279          * Determines if the specified character is permissible as the first character
 280          * in a PHP identifier or variable
 281          *
 282          * The '$' character for PHP variables is regarded as a correct first
 283          * character !
 284          *
 285          */
 286         public static boolean isPHPIdentOrVarStart(char ch) {
 287                 if (ch < MAX_OBVIOUS) {
 288                         return ObviousIdentCharNatures[ch]==C_LETTER || ObviousIdentCharNatures[ch]==C_DOLLAR;
 289                 }
 290                 return false;
 291                 //return Character.isLetter(ch) || (ch == '$') || (ch == '_') || (0x7F <= ch && ch <= 0xFF);
 292         }
 293
 294         /**
 295          * Determines if the specified character is permissible as the first character
 296          * in a PHP identifier.
 297          *
 298          * The '$' character for PHP variables isn't regarded as the first character !
 299          */
 300         public static boolean isPHPIdentifierStart(char ch) {
 301                 if (ch < MAX_OBVIOUS) {
 302                         return ObviousIdentCharNatures[ch]==C_LETTER;
 303                 }
 304                 return false;
 305 //              return Character.isLetter(ch) || (ch == '_') || (0x7F <= ch && ch <= 0xFF);
 306         }
 307
 308         /**
 309          * Determines if the specified character may be part of a PHP identifier as
 310          * other than the first character
 311          */
 312         public static boolean isPHPIdentifierPart(char ch) {
 313                 if (ch < MAX_OBVIOUS) {
 314                         return ObviousIdentCharNatures[ch]==C_LETTER || ObviousIdentCharNatures[ch]==C_DIGIT;
 315                 }
 316                 return false;
 317 //              return Character.isLetterOrDigit(ch) || (ch == '_') || (0x7F <= ch && ch <= 0xFF);
 318         }
 319
 320         public static boolean isSQLIdentifierPart(char ch) {
 321                 if (ch < MAX_OBVIOUS) {
 322                         return ObviousIdentCharNatures[ch]==C_LETTER || ObviousIdentCharNatures[ch]==C_DIGIT;
 323                 }
 324                 return false;
 325         }
 326
 327         public final boolean atEnd() {
 328                 // This code is not relevant if source is
 329                 // Only a part of the real stream input
 330                 return source.length == currentPosition;
 331         }
 332
 333         public char[] getCurrentIdentifierSource() {
 334                 // return the token REAL source (aka unicodes are precomputed)
 335                 char[] result;
 336                 // if (withoutUnicodePtr != 0)
 337                 // //0 is used as a fast test flag so the real first char is in position 1
 338                 // System.arraycopy(
 339                 // withoutUnicodeBuffer,
 340                 // 1,
 341                 // result = new char[withoutUnicodePtr],
 342                 // 0,
 343                 // withoutUnicodePtr);
 344                 // else {
 345                 int length = currentPosition - startPosition;
 346                 switch (length) { // see OptimizedLength
 347                 case 1:
 348                         return optimizedCurrentTokenSource1();
 349                 case 2:
 350                         return optimizedCurrentTokenSource2();
 351                 case 3:
 352                         return optimizedCurrentTokenSource3();
 353                 case 4:
 354                         return optimizedCurrentTokenSource4();
 355                 case 5:
 356                         return optimizedCurrentTokenSource5();
 357                 case 6:
 358                         return optimizedCurrentTokenSource6();
 359                 }
 360                 // no optimization
 361                 System.arraycopy(source, startPosition, result = new char[length], 0, length);
 362                 // }
 363                 return result;
 364         }
 365
 366         public int getCurrentTokenEndPosition() {
 367                 return this.currentPosition - 1;
 368         }
 369
 370         public final char[] getCurrentTokenSource() {
 371                 // Return the token REAL source (aka unicodes are precomputed)
 372                 char[] result;
 373                 // if (withoutUnicodePtr != 0)
 374                 // // 0 is used as a fast test flag so the real first char is in position 1
 375                 // System.arraycopy(
 376                 // withoutUnicodeBuffer,
 377                 // 1,
 378                 // result = new char[withoutUnicodePtr],
 379                 // 0,
 380                 // withoutUnicodePtr);
 381                 // else {
 382                 int length;
 383                 System.arraycopy(source, startPosition, result = new char[length = currentPosition - startPosition], 0, length);
 384                 // }
 385                 return result;
 386         }
 387
 388         public final char[] getCurrentTokenSource(int startPos) {
 389                 // Return the token REAL source (aka unicodes are precomputed)
 390                 char[] result;
 391                 // if (withoutUnicodePtr != 0)
 392                 // // 0 is used as a fast test flag so the real first char is in position 1
 393                 // System.arraycopy(
 394                 // withoutUnicodeBuffer,
 395                 // 1,
 396                 // result = new char[withoutUnicodePtr],
 397                 // 0,
 398                 // withoutUnicodePtr);
 399                 // else {
 400                 int length;
 401                 System.arraycopy(source, startPos, result = new char[length = currentPosition - startPos], 0, length);
 402                 // }
 403                 return result;
 404         }
 405
 406         public final char[] getCurrentTokenSourceString() {
 407                 // return the token REAL source (aka unicodes are precomputed).
 408                 // REMOVE the two " that are at the beginning and the end.
 409                 char[] result;
 410                 if (withoutUnicodePtr != 0)
 411                         // 0 is used as a fast test flag so the real first char is in position 1
 412                         System.arraycopy(withoutUnicodeBuffer, 2,
 413                         // 2 is 1 (real start) + 1 (to jump over the ")
 414                                         result = new char[withoutUnicodePtr - 2], 0, withoutUnicodePtr - 2);
 415                 else {
 416                         int length;
 417                         System.arraycopy(source, startPosition + 1, result = new char[length = currentPosition - startPosition - 2], 0, length);
 418                 }
 419                 return result;
 420         }
 421
 422         public final char[] getRawTokenSourceEnd() {
 423                 int length = this.eofPosition - this.currentPosition - 1;
 424                 char[] sourceEnd = new char[length];
 425                 System.arraycopy(this.source, this.currentPosition, sourceEnd, 0, length);
 426                 return sourceEnd;
 427         }
 428
 429         public int getCurrentTokenStartPosition() {
 430                 return this.startPosition;
 431         }
 432
 433         public final String getCurrentStringLiteral() {
 434                 char[] result = getCurrentStringLiteralSource();
 435                 return new String(result);
 436         }
 437
 438         public final char[] getCurrentStringLiteralSource() {
 439                 // Return the token REAL source (aka unicodes are precomputed)
 440                 if (startPosition + 1 >= currentPosition) {
 441                         return new char[0];
 442                 }
 443                 char[] result;
 444                 int length;
 445                 System.arraycopy(source, startPosition + 1, result = new char[length = currentPosition - startPosition - 2], 0, length);
 446                 // }
 447                 return result;
 448         }
 449
 450         public final char[] getCurrentStringLiteralSource(int startPos) {
 451                 // Return the token REAL source (aka unicodes are precomputed)
 452                 char[] result;
 453                 int length;
 454                 System.arraycopy(source, startPos + 1, result = new char[length = currentPosition - startPos - 2], 0, length);
 455                 // }
 456                 return result;
 457         }
 458
 459         /*
 460          * Search the source position corresponding to the end of a given line number
 461          *
 462          * Line numbers are 1-based, and relative to the scanner initialPosition.
 463          * Character positions are 0-based.
 464          *
 465          * In case the given line number is inconsistent, answers -1.
 466          */
 467         public final int getLineEnd(int lineNumber) {
 468                 if (lineEnds == null)
 469                         return -1;
 470                 if (lineNumber >= lineEnds.length)
 471                         return -1;
 472                 if (lineNumber <= 0)
 473                         return -1;
 474                 if (lineNumber == lineEnds.length - 1)
 475                         return eofPosition;
 476                 return lineEnds[lineNumber - 1];
 477                 // next line start one character behind the lineEnd of the previous line
 478         }
 479
 480         /**
 481          * Search the source position corresponding to the beginning of a given line
 482          * number
 483          *
 484          * Line numbers are 1-based, and relative to the scanner initialPosition.
 485          * Character positions are 0-based.
 486          *
 487          * e.g. getLineStart(1) --> 0 i.e. first line starts at character 0.
 488          *
 489          * In case the given line number is inconsistent, answers -1.
 490          */
 491         public final int getLineStart(int lineNumber) {
 492                 if (lineEnds == null)
 493                         return -1;
 494                 if (lineNumber >= lineEnds.length)
 495                         return -1;
 496                 if (lineNumber <= 0)
 497                         return -1;
 498                 if (lineNumber == 1)
 499                         return initialPosition;
 500                 return lineEnds[lineNumber - 2] + 1;
 501                 // next line start one character behind the lineEnd of the previous line
 502         }
 503
 504         public final boolean getNextChar(char testedChar) {
 505                 // BOOLEAN
 506                 // handle the case of unicode.
 507                 // when a unicode appears then we must use a buffer that holds char
 508                 // internal values
 509                 // At the end of this method currentCharacter holds the new visited char
 510                 // and currentPosition points right next after it
 511                 // Both previous lines are true if the currentCharacter is == to the
 512                 // testedChar
 513                 // On false, no side effect has occured.
 514                 // ALL getNextChar.... ARE OPTIMIZED COPIES
 515                 int temp = currentPosition;
 516                 try {
 517                         currentCharacter = source[currentPosition++];
 518                         // if (((currentCharacter = source[currentPosition++]) == '\\')
 519                         // && (source[currentPosition] == 'u')) {
 520                         // //-------------unicode traitement ------------
 521                         // int c1, c2, c3, c4;
 522                         // int unicodeSize = 6;
 523                         // currentPosition++;
 524                         // while (source[currentPosition] == 'u') {
 525                         // currentPosition++;
 526                         // unicodeSize++;
 527                         // }
 528                         //
 529                         // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
 530                         // || c1 < 0)
 531                         // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
 532                         // || c2 < 0)
 533                         // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
 534                         // || c3 < 0)
 535                         // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
 536                         // || c4 < 0)) {
 537                         // currentPosition = temp;
 538                         // return false;
 539                         // }
 540                         //
 541                         // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
 542                         // if (currentCharacter != testedChar) {
 543                         // currentPosition = temp;
 544                         // return false;
 545                         // }
 546                         // unicodeAsBackSlash = currentCharacter == '\\';
 547                         //
 548                         // //need the unicode buffer
 549                         // if (withoutUnicodePtr == 0) {
 550                         // //buffer all the entries that have been left aside....
 551                         // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
 552                         // System.arraycopy(
 553                         // source,
 554                         // startPosition,
 555                         // withoutUnicodeBuffer,
 556                         // 1,
 557                         // withoutUnicodePtr);
 558                         // }
 559                         // //fill the buffer with the char
 560                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 561                         // return true;
 562                         //
 563                         // } //-------------end unicode traitement--------------
 564                         // else {
 565                         if (currentCharacter != testedChar) {
 566                                 currentPosition = temp;
 567                                 return false;
 568                         }
 569                         unicodeAsBackSlash = false;
 570                         // if (withoutUnicodePtr != 0)
 571                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 572                         return true;
 573                         // }
 574                 } catch (IndexOutOfBoundsException e) {
 575                         unicodeAsBackSlash = false;
 576                         currentPosition = temp;
 577                         return false;
 578                 }
 579         }
 580
 581         public final int getNextChar(char testedChar1, char testedChar2) {
 582                 // INT 0 : testChar1 \\\\///\\\\ 1 : testedChar2 \\\\///\\\\ -1 : others
 583                 // test can be done with (x==0) for the first and (x>0) for the second
 584                 // handle the case of unicode.
 585                 // when a unicode appears then we must use a buffer that holds char
 586                 // internal values
 587                 // At the end of this method currentCharacter holds the new visited char
 588                 // and currentPosition points right next after it
 589                 // Both previous lines are true if the currentCharacter is == to the
 590                 // testedChar1/2
 591                 // On false, no side effect has occured.
 592                 // ALL getNextChar.... ARE OPTIMIZED COPIES
 593                 int temp = currentPosition;
 594                 try {
 595                         int result;
 596                         currentCharacter = source[currentPosition++];
 597                         // if (((currentCharacter = source[currentPosition++]) == '\\')
 598                         // && (source[currentPosition] == 'u')) {
 599                         // //-------------unicode traitement ------------
 600                         // int c1, c2, c3, c4;
 601                         // int unicodeSize = 6;
 602                         // currentPosition++;
 603                         // while (source[currentPosition] == 'u') {
 604                         // currentPosition++;
 605                         // unicodeSize++;
 606                         // }
 607                         //
 608                         // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
 609                         // || c1 < 0)
 610                         // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
 611                         // || c2 < 0)
 612                         // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
 613                         // || c3 < 0)
 614                         // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
 615                         // || c4 < 0)) {
 616                         // currentPosition = temp;
 617                         // return 2;
 618                         // }
 619                         //
 620                         // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
 621                         // if (currentCharacter == testedChar1)
 622                         // result = 0;
 623                         // else if (currentCharacter == testedChar2)
 624                         // result = 1;
 625                         // else {
 626                         // currentPosition = temp;
 627                         // return -1;
 628                         // }
 629                         //
 630                         // //need the unicode buffer
 631                         // if (withoutUnicodePtr == 0) {
 632                         // //buffer all the entries that have been left aside....
 633                         // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
 634                         // System.arraycopy(
 635                         // source,
 636                         // startPosition,
 637                         // withoutUnicodeBuffer,
 638                         // 1,
 639                         // withoutUnicodePtr);
 640                         // }
 641                         // //fill the buffer with the char
 642                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 643                         // return result;
 644                         // } //-------------end unicode traitement--------------
 645                         // else {
 646                         if (currentCharacter == testedChar1)
 647                                 result = 0;
 648                         else if (currentCharacter == testedChar2)
 649                                 result = 1;
 650                         else {
 651                                 currentPosition = temp;
 652                                 return -1;
 653                         }
 654                         // if (withoutUnicodePtr != 0)
 655                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 656                         return result;
 657                         // }
 658                 } catch (IndexOutOfBoundsException e) {
 659                         currentPosition = temp;
 660                         return -1;
 661                 }
 662         }
 663
 664         public final boolean getNextCharAsDigit() {
 665                 // BOOLEAN
 666                 // handle the case of unicode.
 667                 // when a unicode appears then we must use a buffer that holds char
 668                 // internal values
 669                 // At the end of this method currentCharacter holds the new visited char
 670                 // and currentPosition points right next after it
 671                 // Both previous lines are true if the currentCharacter is a digit
 672                 // On false, no side effect has occured.
 673                 // ALL getNextChar.... ARE OPTIMIZED COPIES
 674                 int temp = currentPosition;
 675                 try {
 676                         currentCharacter = source[currentPosition++];
 677                         // if (((currentCharacter = source[currentPosition++]) == '\\')
 678                         // && (source[currentPosition] == 'u')) {
 679                         // //-------------unicode traitement ------------
 680                         // int c1, c2, c3, c4;
 681                         // int unicodeSize = 6;
 682                         // currentPosition++;
 683                         // while (source[currentPosition] == 'u') {
 684                         // currentPosition++;
 685                         // unicodeSize++;
 686                         // }
 687                         //
 688                         // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
 689                         // || c1 < 0)
 690                         // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
 691                         // || c2 < 0)
 692                         // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
 693                         // || c3 < 0)
 694                         // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
 695                         // || c4 < 0)) {
 696                         // currentPosition = temp;
 697                         // return false;
 698                         // }
 699                         //
 700                         // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
 701                         // if (!Character.isDigit(currentCharacter)) {
 702                         // currentPosition = temp;
 703                         // return false;
 704                         // }
 705                         //
 706                         // //need the unicode buffer
 707                         // if (withoutUnicodePtr == 0) {
 708                         // //buffer all the entries that have been left aside....
 709                         // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
 710                         // System.arraycopy(
 711                         // source,
 712                         // startPosition,
 713                         // withoutUnicodeBuffer,
 714                         // 1,
 715                         // withoutUnicodePtr);
 716                         // }
 717                         // //fill the buffer with the char
 718                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 719                         // return true;
 720                         // } //-------------end unicode traitement--------------
 721                         // else {
 722                         if (!Character.isDigit(currentCharacter)) {
 723                                 currentPosition = temp;
 724                                 return false;
 725                         }
 726                         // if (withoutUnicodePtr != 0)
 727                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 728                         return true;
 729                         // }
 730                 } catch (IndexOutOfBoundsException e) {
 731                         currentPosition = temp;
 732                         return false;
 733                 }
 734         }
 735
 736         public final boolean getNextCharAsDigit(int radix) {
 737                 // BOOLEAN
 738                 // handle the case of unicode.
 739                 // when a unicode appears then we must use a buffer that holds char
 740                 // internal values
 741                 // At the end of this method currentCharacter holds the new visited char
 742                 // and currentPosition points right next after it
 743                 // Both previous lines are true if the currentCharacter is a digit base on
 744                 // radix
 745                 // On false, no side effect has occured.
 746                 // ALL getNextChar.... ARE OPTIMIZED COPIES
 747                 int temp = currentPosition;
 748                 try {
 749                         currentCharacter = source[currentPosition++];
 750                         // if (((currentCharacter = source[currentPosition++]) == '\\')
 751                         // && (source[currentPosition] == 'u')) {
 752                         // //-------------unicode traitement ------------
 753                         // int c1, c2, c3, c4;
 754                         // int unicodeSize = 6;
 755                         // currentPosition++;
 756                         // while (source[currentPosition] == 'u') {
 757                         // currentPosition++;
 758                         // unicodeSize++;
 759                         // }
 760                         //
 761                         // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
 762                         // || c1 < 0)
 763                         // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
 764                         // || c2 < 0)
 765                         // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
 766                         // || c3 < 0)
 767                         // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
 768                         // || c4 < 0)) {
 769                         // currentPosition = temp;
 770                         // return false;
 771                         // }
 772                         //
 773                         // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
 774                         // if (Character.digit(currentCharacter, radix) == -1) {
 775                         // currentPosition = temp;
 776                         // return false;
 777                         // }
 778                         //
 779                         // //need the unicode buffer
 780                         // if (withoutUnicodePtr == 0) {
 781                         // //buffer all the entries that have been left aside....
 782                         // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
 783                         // System.arraycopy(
 784                         // source,
 785                         // startPosition,
 786                         // withoutUnicodeBuffer,
 787                         // 1,
 788                         // withoutUnicodePtr);
 789                         // }
 790                         // //fill the buffer with the char
 791                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 792                         // return true;
 793                         // } //-------------end unicode traitement--------------
 794                         // else {
 795                         if (Character.digit(currentCharacter, radix) == -1) {
 796                                 currentPosition = temp;
 797                                 return false;
 798                         }
 799                         // if (withoutUnicodePtr != 0)
 800                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 801                         return true;
 802                         // }
 803                 } catch (IndexOutOfBoundsException e) {
 804                         currentPosition = temp;
 805                         return false;
 806                 }
 807         }
 808
 809         public boolean getNextCharAsJavaIdentifierPart() {
 810                 // BOOLEAN
 811                 // handle the case of unicode.
 812                 // when a unicode appears then we must use a buffer that holds char
 813                 // internal values
 814                 // At the end of this method currentCharacter holds the new visited char
 815                 // and currentPosition points right next after it
 816                 // Both previous lines are true if the currentCharacter is a
 817                 // JavaIdentifierPart
 818                 // On false, no side effect has occured.
 819                 // ALL getNextChar.... ARE OPTIMIZED COPIES
 820                 int temp = currentPosition;
 821                 try {
 822                         currentCharacter = source[currentPosition++];
 823                         // if (((currentCharacter = source[currentPosition++]) == '\\')
 824                         // && (source[currentPosition] == 'u')) {
 825                         // //-------------unicode traitement ------------
 826                         // int c1, c2, c3, c4;
 827                         // int unicodeSize = 6;
 828                         // currentPosition++;
 829                         // while (source[currentPosition] == 'u') {
 830                         // currentPosition++;
 831                         // unicodeSize++;
 832                         // }
 833                         //
 834                         // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
 835                         // || c1 < 0)
 836                         // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
 837                         // || c2 < 0)
 838                         // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
 839                         // || c3 < 0)
 840                         // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
 841                         // || c4 < 0)) {
 842                         // currentPosition = temp;
 843                         // return false;
 844                         // }
 845                         //
 846                         // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
 847                         // if (!isPHPIdentifierPart(currentCharacter)) {
 848                         // currentPosition = temp;
 849                         // return false;
 850                         // }
 851                         //
 852                         // //need the unicode buffer
 853                         // if (withoutUnicodePtr == 0) {
 854                         // //buffer all the entries that have been left aside....
 855                         // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
 856                         // System.arraycopy(
 857                         // source,
 858                         // startPosition,
 859                         // withoutUnicodeBuffer,
 860                         // 1,
 861                         // withoutUnicodePtr);
 862                         // }
 863                         // //fill the buffer with the char
 864                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 865                         // return true;
 866                         // } //-------------end unicode traitement--------------
 867                         // else {
 868                         if (!isPHPIdentifierPart(currentCharacter)) {
 869                                 currentPosition = temp;
 870                                 return false;
 871                         }
 872                         // if (withoutUnicodePtr != 0)
 873                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
 874                         return true;
 875                         // }
 876                 } catch (IndexOutOfBoundsException e) {
 877                         currentPosition = temp;
 878                         return false;
 879                 }
 880         }
 881
 882         public int getCastOrParen() {
 883                 int tempPosition = currentPosition;
 884                 char tempCharacter = currentCharacter;
 885                 int tempToken = TokenNameLPAREN;
 886                 boolean found = false;
 887                 StringBuffer buf = new StringBuffer();
 888                 try {
 889                         do {
 890                                 currentCharacter = source[currentPosition++];
 891                         } while (currentCharacter == ' ' || currentCharacter == '\t');
 892                         while (ObviousIdentCharNatures[currentCharacter]==C_LETTER) {
 893                                 //      while((currentCharacter >= 'a' && currentCharacter <= 'z') || (currentCharacter >= 'A' && currentCharacter <= 'Z')) {
 894                                 buf.append(currentCharacter);
 895                                 currentCharacter = source[currentPosition++];
 896                         }
 897                         if (buf.length() >= 3 && buf.length() <= 7) {
 898                                 char[] data = buf.toString().toCharArray();
 899                                 int index = 0;
 900                                 switch (data.length) {
 901                                 case 3:
 902                                         // int
 903                                         if ((data[index] == 'i') && (data[++index] == 'n') && (data[++index] == 't')) {
 904                                                 found = true;
 905                                                 tempToken = TokenNameintCAST;
 906                                         }
 907                                         break;
 908                                 case 4:
 909                                         // bool real
 910                                         if ((data[index] == 'b') && (data[++index] == 'o') && (data[++index] == 'o') && (data[++index] == 'l')) {
 911                                                 found = true;
 912                                                 tempToken = TokenNameboolCAST;
 913                                         } else {
 914                                                 index = 0;
 915                                                 if ((data[index] == 'r') && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'l')) {
 916                                                         found = true;
 917                                                         tempToken = TokenNamedoubleCAST;
 918                                                 }
 919                                         }
 920                                         break;
 921                                 case 5:
 922                                         // array unset float
 923                                         if ((data[index] == 'a') && (data[++index] == 'r') && (data[++index] == 'r') && (data[++index] == 'a')
 924                                                         && (data[++index] == 'y')) {
 925                                                 found = true;
 926                                                 tempToken = TokenNamearrayCAST;
 927                                         } else {
 928                                                 index = 0;
 929                                                 if ((data[index] == 'u') && (data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 'e')
 930                                                                 && (data[++index] == 't')) {
 931                                                         found = true;
 932                                                         tempToken = TokenNameunsetCAST;
 933                                                 } else {
 934                                                         index = 0;
 935                                                         if ((data[index] == 'f') && (data[++index] == 'l') && (data[++index] == 'o') && (data[++index] == 'a')
 936                                                                         && (data[++index] == 't')) {
 937                                                                 found = true;
 938                                                                 tempToken = TokenNamedoubleCAST;
 939                                                         }
 940                                                 }
 941                                         }
 942                                         break;
 943                                 case 6:
 944                                         // object string double
 945                                         if ((data[index] == 'o') && (data[++index] == 'b') && (data[++index] == 'j') && (data[++index] == 'e')
 946                                                         && (data[++index] == 'c') && (data[++index] == 't')) {
 947                                                 found = true;
 948                                                 tempToken = TokenNameobjectCAST;
 949                                         } else {
 950                                                 index = 0;
 951                                                 if ((data[index] == 's') && (data[++index] == 't') && (data[++index] == 'r') && (data[++index] == 'i')
 952                                                                 && (data[++index] == 'n') && (data[++index] == 'g')) {
 953                                                         found = true;
 954                                                         tempToken = TokenNamestringCAST;
 955                                                 } else {
 956                                                         index = 0;
 957                                                         if ((data[index] == 'd') && (data[++index] == 'o') && (data[++index] == 'u') && (data[++index] == 'b')
 958                                                                         && (data[++index] == 'l') && (data[++index] == 'e')) {
 959                                                                 found = true;
 960                                                                 tempToken = TokenNamedoubleCAST;
 961                                                         }
 962                                                 }
 963                                         }
 964                                         break;
 965                                 case 7:
 966                                         // boolean integer
 967                                         if ((data[index] == 'b') && (data[++index] == 'o') && (data[++index] == 'o') && (data[++index] == 'l')
 968                                                         && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'n')) {
 969                                                 found = true;
 970                                                 tempToken = TokenNameboolCAST;
 971                                         } else {
 972                                                 index = 0;
 973                                                 if ((data[index] == 'i') && (data[++index] == 'n') && (data[++index] == 't') && (data[++index] == 'e')
 974                                                                 && (data[++index] == 'g') && (data[++index] == 'e') && (data[++index] == 'r')) {
 975                                                         found = true;
 976                                                         tempToken = TokenNameintCAST;
 977                                                 }
 978                                         }
 979                                         break;
 980                                 }
 981                                 if (found) {
 982                                         while (currentCharacter == ' ' || currentCharacter == '\t') {
 983                                                 currentCharacter = source[currentPosition++];
 984                                         }
 985                                         if (currentCharacter == ')') {
 986                                                 return tempToken;
 987                                         }
 988                                 }
 989                         }
 990                 } catch (IndexOutOfBoundsException e) {
 991                 }
 992                 currentCharacter = tempCharacter;
 993                 currentPosition = tempPosition;
 994                 return TokenNameLPAREN;
 995         }
 996
 997         public void consumeStringInterpolated() throws InvalidInputException {
 998                 try {
 999                         // consume next character
1000                         unicodeAsBackSlash = false;
1001                         currentCharacter = source[currentPosition++];
1002                         // if (((currentCharacter = source[currentPosition++]) == '\\')
1003                         // && (source[currentPosition] == 'u')) {
1004                         // getNextUnicodeChar();
1005                         // } else {
1006                         // if (withoutUnicodePtr != 0) {
1007                         // withoutUnicodeBuffer[++withoutUnicodePtr] =
1008                         // currentCharacter;
1009                         // }
1010                         // }
1011                         while (currentCharacter != '`') {
1012                                 /** ** in PHP \r and \n are valid in string literals *** */
1013                                 // if ((currentCharacter == '\n')
1014                                 // || (currentCharacter == '\r')) {
1015                                 // // relocate if finding another quote fairly close: thus unicode
1016                                 // '/u000D' will be fully consumed
1017                                 // for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1018                                 // if (currentPosition + lookAhead == source.length)
1019                                 // break;
1020                                 // if (source[currentPosition + lookAhead] == '\n')
1021                                 // break;
1022                                 // if (source[currentPosition + lookAhead] == '\"') {
1023                                 // currentPosition += lookAhead + 1;
1024                                 // break;
1025                                 // }
1026                                 // }
1027                                 // throw new InvalidInputException(INVALID_CHAR_IN_STRING);
1028                                 // }
1029                                 if (currentCharacter == '\\') {
1030                                         int escapeSize = currentPosition;
1031                                         boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
1032                                         // scanEscapeCharacter make a side effect on this value and we need
1033                                         // the previous value few lines down this one
1034                                         scanDoubleQuotedEscapeCharacter();
1035                                         escapeSize = currentPosition - escapeSize;
1036                                         if (withoutUnicodePtr == 0) {
1037                                                 // buffer all the entries that have been left aside....
1038                                                 withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition;
1039                                                 System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr);
1040                                                 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1041                                         } else { // overwrite the / in the buffer
1042                                                 withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
1043                                                 if (backSlashAsUnicodeInString) { // there are TWO \ in the stream
1044                                                         // where only one is correct
1045                                                         withoutUnicodePtr--;
1046                                                 }
1047                                         }
1048                                 } else if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1049                                         if (recordLineSeparator) {
1050                                                 pushLineSeparator();
1051                                         }
1052                                 }
1053                                 // consume next character
1054                                 unicodeAsBackSlash = false;
1055                                 currentCharacter = source[currentPosition++];
1056                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
1057                                 // && (source[currentPosition] == 'u')) {
1058                                 // getNextUnicodeChar();
1059                                 // } else {
1060                                 if (withoutUnicodePtr != 0) {
1061                                         withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1062                                 }
1063                                 // }
1064                         }
1065                 } catch (IndexOutOfBoundsException e) {
1066                         // reset end position for error reporting
1067                         currentPosition -= 2;
1068                         throw new InvalidInputException(UNTERMINATED_STRING);
1069                 } catch (InvalidInputException e) {
1070                         if (e.getMessage().equals(INVALID_ESCAPE)) {
1071                                 // relocate if finding another quote fairly close: thus unicode
1072                                 // '/u000D' will be fully consumed
1073                                 for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1074                                         if (currentPosition + lookAhead == source.length)
1075                                                 break;
1076                                         if (source[currentPosition + lookAhead] == '\n')
1077                                                 break;
1078                                         if (source[currentPosition + lookAhead] == '`') {
1079                                                 currentPosition += lookAhead + 1;
1080                                                 break;
1081                                         }
1082                                 }
1083                         }
1084                         throw e; // rethrow
1085                 }
1086                 if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags
1087                         // //$NON-NLS-?$ where ? is an
1088                         // int.
1089                         if (currentLine == null) {
1090                                 currentLine = new NLSLine();
1091                                 lines.add(currentLine);
1092                         }
1093                         currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1));
1094                 }
1095         }
1096
1097         public void consumeStringConstant() throws InvalidInputException {
1098                 try {
1099                         // consume next character
1100                         unicodeAsBackSlash = false;
1101                         currentCharacter = source[currentPosition++];
1102                         // if (((currentCharacter = source[currentPosition++]) == '\\')
1103                         // && (source[currentPosition] == 'u')) {
1104                         // getNextUnicodeChar();
1105                         // } else {
1106                         // if (withoutUnicodePtr != 0) {
1107                         // withoutUnicodeBuffer[++withoutUnicodePtr] =
1108                         // currentCharacter;
1109                         // }
1110                         // }
1111                         while (currentCharacter != '\'') {
1112                                 /** ** in PHP \r and \n are valid in string literals *** */
1113                                 // if ((currentCharacter == '\n')
1114                                 // || (currentCharacter == '\r')) {
1115                                 // // relocate if finding another quote fairly close: thus unicode
1116                                 // '/u000D' will be fully consumed
1117                                 // for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1118                                 // if (currentPosition + lookAhead == source.length)
1119                                 // break;
1120                                 // if (source[currentPosition + lookAhead] == '\n')
1121                                 // break;
1122                                 // if (source[currentPosition + lookAhead] == '\"') {
1123                                 // currentPosition += lookAhead + 1;
1124                                 // break;
1125                                 // }
1126                                 // }
1127                                 // throw new InvalidInputException(INVALID_CHAR_IN_STRING);
1128                                 // }
1129                                 if (currentCharacter == '\\') {
1130                                         int escapeSize = currentPosition;
1131                                         boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
1132                                         // scanEscapeCharacter make a side effect on this value and we need
1133                                         // the previous value few lines down this one
1134                                         scanSingleQuotedEscapeCharacter();
1135                                         escapeSize = currentPosition - escapeSize;
1136                                         if (withoutUnicodePtr == 0) {
1137                                                 // buffer all the entries that have been left aside....
1138                                                 withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition;
1139                                                 System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr);
1140                                                 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1141                                         } else { // overwrite the / in the buffer
1142                                                 withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
1143                                                 if (backSlashAsUnicodeInString) { // there are TWO \ in the stream
1144                                                         // where only one is correct
1145                                                         withoutUnicodePtr--;
1146                                                 }
1147                                         }
1148                                 } else if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1149                                         if (recordLineSeparator) {
1150                                                 pushLineSeparator();
1151                                         }
1152                                 }
1153                                 // consume next character
1154                                 unicodeAsBackSlash = false;
1155                                 currentCharacter = source[currentPosition++];
1156                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
1157                                 // && (source[currentPosition] == 'u')) {
1158                                 // getNextUnicodeChar();
1159                                 // } else {
1160                                 if (withoutUnicodePtr != 0) {
1161                                         withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1162                                 }
1163                                 // }
1164                         }
1165                 } catch (IndexOutOfBoundsException e) {
1166                         // reset end position for error reporting
1167                         currentPosition -= 2;
1168                         throw new InvalidInputException(UNTERMINATED_STRING);
1169                 } catch (InvalidInputException e) {
1170                         if (e.getMessage().equals(INVALID_ESCAPE)) {
1171                                 // relocate if finding another quote fairly close: thus unicode
1172                                 // '/u000D' will be fully consumed
1173                                 for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1174                                         if (currentPosition + lookAhead == source.length)
1175                                                 break;
1176                                         if (source[currentPosition + lookAhead] == '\n')
1177                                                 break;
1178                                         if (source[currentPosition + lookAhead] == '\'') {
1179                                                 currentPosition += lookAhead + 1;
1180                                                 break;
1181                                         }
1182                                 }
1183                         }
1184                         throw e; // rethrow
1185                 }
1186                 if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags
1187                         // //$NON-NLS-?$ where ? is an
1188                         // int.
1189                         if (currentLine == null) {
1190                                 currentLine = new NLSLine();
1191                                 lines.add(currentLine);
1192                         }
1193                         currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1));
1194                 }
1195         }
1196
1197         public void consumeStringLiteral() throws InvalidInputException {
1198                 try {
1199                         boolean openDollarBrace = false;
1200                         // consume next character
1201                         unicodeAsBackSlash = false;
1202                         currentCharacter = source[currentPosition++];
1203                         while (currentCharacter != '"' || openDollarBrace) {
1204                                 /** ** in PHP \r and \n are valid in string literals *** */
1205                                 if (currentCharacter == '\\') {
1206                                         int escapeSize = currentPosition;
1207                                         boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
1208                                         // scanEscapeCharacter make a side effect on this value and we need
1209                                         // the previous value few lines down this one
1210                                         scanDoubleQuotedEscapeCharacter();
1211                                         escapeSize = currentPosition - escapeSize;
1212                                         if (withoutUnicodePtr == 0) {
1213                                                 // buffer all the entries that have been left aside....
1214                                                 withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition;
1215                                                 System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr);
1216                                                 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1217                                         } else { // overwrite the / in the buffer
1218                                                 withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
1219                                                 if (backSlashAsUnicodeInString) { // there are TWO \ in the stream
1220                                                         // where only one is correct
1221                                                         withoutUnicodePtr--;
1222                                                 }
1223                                         }
1224                                 } else if (currentCharacter == '$' && source[currentPosition] == '{') {
1225                                         openDollarBrace = true;
1226                                 } else if (currentCharacter == '{' && source[currentPosition] == '$') {
1227                                         openDollarBrace = true;
1228                                 } else if (currentCharacter == '}') {
1229                                         openDollarBrace = false;
1230                                 } else if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1231                                         if (recordLineSeparator) {
1232                                                 pushLineSeparator();
1233                                         }
1234                                 }
1235                                 // consume next character
1236                                 unicodeAsBackSlash = false;
1237                                 currentCharacter = source[currentPosition++];
1238                                 if (withoutUnicodePtr != 0) {
1239                                         withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1240                                 }
1241                         }
1242                 } catch (IndexOutOfBoundsException e) {
1243                         // reset end position for error reporting
1244                         currentPosition -= 2;
1245                         throw new InvalidInputException(UNTERMINATED_STRING);
1246                 } catch (InvalidInputException e) {
1247                         if (e.getMessage().equals(INVALID_ESCAPE)) {
1248                                 // relocate if finding another quote fairly close: thus unicode
1249                                 // '/u000D' will be fully consumed
1250                                 for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1251                                         if (currentPosition + lookAhead == source.length)
1252                                                 break;
1253                                         if (source[currentPosition + lookAhead] == '\n')
1254                                                 break;
1255                                         if (source[currentPosition + lookAhead] == '\"') {
1256                                                 currentPosition += lookAhead + 1;
1257                                                 break;
1258                                         }
1259                                 }
1260                         }
1261                         throw e; // rethrow
1262                 }
1263                 if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags
1264                         // //$NON-NLS-?$ where ? is an
1265                         // int.
1266                         if (currentLine == null) {
1267                                 currentLine = new NLSLine();
1268                                 lines.add(currentLine);
1269                         }
1270                         currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1));
1271                 }
1272         }
1273
1274         public int getNextToken() throws InvalidInputException {
1275                 phpExpressionTag = false;
1276                 if (!phpMode) {
1277                         return getInlinedHTMLToken(currentPosition);
1278                 }
1279                 if (phpMode) {
1280                         this.wasAcr = false;
1281                         if (diet) {
1282                                 jumpOverMethodBody();
1283                                 diet = false;
1284                                 return currentPosition > source.length ? TokenNameEOF : TokenNameRBRACE;
1285                         }
1286                         try {
1287                                 while (true) {
1288                                         withoutUnicodePtr = 0;
1289                                         // start with a new token
1290                                         char encapsedChar = ' ';
1291                                         // if (!encapsedStringStack.isEmpty()) {
1292                                         // encapsedChar = ((Character)
1293                                         // encapsedStringStack.peek()).charValue();
1294                                         // }
1295                                         // if (encapsedChar != '$' && encapsedChar != ' ') {
1296                                         // currentCharacter = source[currentPosition++];
1297                                         // if (currentCharacter == encapsedChar) {
1298                                         // switch (currentCharacter) {
1299                                         // case '`':
1300                                         // return TokenNameEncapsedString0;
1301                                         // case '\'':
1302                                         // return TokenNameEncapsedString1;
1303                                         // case '"':
1304                                         // return TokenNameEncapsedString2;
1305                                         // }
1306                                         // }
1307                                         // while (currentCharacter != encapsedChar) {
1308                                         // /** ** in PHP \r and \n are valid in string literals *** */
1309                                         // switch (currentCharacter) {
1310                                         // case '\\':
1311                                         // int escapeSize = currentPosition;
1312                                         // boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
1313                                         // //scanEscapeCharacter make a side effect on this value and
1314                                         // // we need the previous value few lines down this one
1315                                         // scanDoubleQuotedEscapeCharacter();
1316                                         // escapeSize = currentPosition - escapeSize;
1317                                         // if (withoutUnicodePtr == 0) {
1318                                         // //buffer all the entries that have been left aside....
1319                                         // withoutUnicodePtr = currentPosition - escapeSize - 1 -
1320                                         // startPosition;
1321                                         // System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1,
1322                                         // withoutUnicodePtr);
1323                                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1324                                         // } else { //overwrite the / in the buffer
1325                                         // withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
1326                                         // if (backSlashAsUnicodeInString) { //there are TWO \ in
1327                                         // withoutUnicodePtr--;
1328                                         // }
1329                                         // }
1330                                         // break;
1331                                         // case '\r':
1332                                         // case '\n':
1333                                         // if (recordLineSeparator) {
1334                                         // pushLineSeparator();
1335                                         // }
1336                                         // break;
1337                                         // case '$':
1338                                         // if (isPHPIdentifierStart(source[currentPosition]) ||
1339                                         // source[currentPosition] == '{') {
1340                                         // currentPosition--;
1341                                         // encapsedStringStack.push(new Character('$'));
1342                                         // return TokenNameSTRING;
1343                                         // }
1344                                         // break;
1345                                         // case '{':
1346                                         // if (source[currentPosition] == '$') { // CURLY_OPEN
1347                                         // currentPosition--;
1348                                         // encapsedStringStack.push(new Character('$'));
1349                                         // return TokenNameSTRING;
1350                                         // }
1351                                         // }
1352                                         // // consume next character
1353                                         // unicodeAsBackSlash = false;
1354                                         // currentCharacter = source[currentPosition++];
1355                                         // if (withoutUnicodePtr != 0) {
1356                                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1357                                         // }
1358                                         // // }
1359                                         // } // end while
1360                                         // currentPosition--;
1361                                         // return TokenNameSTRING;
1362                                         // }
1363                                         // ---------Consume white space and handles startPosition---------
1364                                         int whiteStart = currentPosition;
1365                                         startPosition = currentPosition;
1366                                         currentCharacter = source[currentPosition++];
1367                                         // if (encapsedChar == '$') {
1368                                         // switch (currentCharacter) {
1369                                         // case '\\':
1370                                         // currentCharacter = source[currentPosition++];
1371                                         // return TokenNameSTRING;
1372                                         // case '{':
1373                                         // if (encapsedChar == '$') {
1374                                         // if (getNextChar('$'))
1375                                         // return TokenNameLBRACE_DOLLAR;
1376                                         // }
1377                                         // return TokenNameLBRACE;
1378                                         // case '}':
1379                                         // return TokenNameRBRACE;
1380                                         // case '[':
1381                                         // return TokenNameLBRACKET;
1382                                         // case ']':
1383                                         // return TokenNameRBRACKET;
1384                                         // case '\'':
1385                                         // if (tokenizeStrings) {
1386                                         // consumeStringConstant();
1387                                         // return TokenNameStringSingleQuote;
1388                                         // }
1389                                         // return TokenNameEncapsedString1;
1390                                         // case '"':
1391                                         // return TokenNameEncapsedString2;
1392                                         // case '`':
1393                                         // if (tokenizeStrings) {
1394                                         // consumeStringInterpolated();
1395                                         // return TokenNameStringInterpolated;
1396                                         // }
1397                                         // return TokenNameEncapsedString0;
1398                                         // case '-':
1399                                         // if (getNextChar('>'))
1400                                         // return TokenNameMINUS_GREATER;
1401                                         // return TokenNameSTRING;
1402                                         // default:
1403                                         // if (currentCharacter == '$') {
1404                                         // int oldPosition = currentPosition;
1405                                         // try {
1406                                         // currentCharacter = source[currentPosition++];
1407                                         // if (currentCharacter == '{') {
1408                                         // return TokenNameDOLLAR_LBRACE;
1409                                         // }
1410                                         // if (isPHPIdentifierStart(currentCharacter)) {
1411                                         // return scanIdentifierOrKeyword(true);
1412                                         // } else {
1413                                         // currentPosition = oldPosition;
1414                                         // return TokenNameSTRING;
1415                                         // }
1416                                         // } catch (IndexOutOfBoundsException e) {
1417                                         // currentPosition = oldPosition;
1418                                         // return TokenNameSTRING;
1419                                         // }
1420                                         // }
1421                                         // if (isPHPIdentifierStart(currentCharacter))
1422                                         // return scanIdentifierOrKeyword(false);
1423                                         // if (Character.isDigit(currentCharacter))
1424                                         // return scanNumber(false);
1425                                         // return TokenNameERROR;
1426                                         // }
1427                                         // }
1428                                         // boolean isWhiteSpace;
1429
1430                                         while ((currentCharacter == ' ') || Character.isWhitespace(currentCharacter)) {
1431                                                 startPosition = currentPosition;
1432                                                 currentCharacter = source[currentPosition++];
1433                                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
1434                                                 // && (source[currentPosition] == 'u')) {
1435                                                 // isWhiteSpace = jumpOverUnicodeWhiteSpace();
1436                                                 // } else {
1437                                                 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1438                                                         checkNonExternalizeString();
1439                                                         if (recordLineSeparator) {
1440                                                                 pushLineSeparator();
1441                                                         } else {
1442                                                                 currentLine = null;
1443                                                         }
1444                                                 }
1445                                                 // isWhiteSpace = (currentCharacter == ' ')
1446                                                 // || Character.isWhitespace(currentCharacter);
1447                                                 // }
1448                                         }
1449                                         if (tokenizeWhiteSpace && (whiteStart != currentPosition - 1)) {
1450                                                 // reposition scanner in case we are interested by spaces as tokens
1451                                                 currentPosition--;
1452                                                 startPosition = whiteStart;
1453                                                 return TokenNameWHITESPACE;
1454                                         }
1455                                         // little trick to get out in the middle of a source compuation
1456                                         if (currentPosition > eofPosition)
1457                                                 return TokenNameEOF;
1458                                         // ---------Identify the next token-------------
1459                                         switch (currentCharacter) {
1460                                         case '(':
1461                                                 return getCastOrParen();
1462                                         case ')':
1463                                                 return TokenNameRPAREN;
1464                                         case '{':
1465                                                 return TokenNameLBRACE;
1466                                         case '}':
1467                                                 return TokenNameRBRACE;
1468                                         case '[':
1469                                                 return TokenNameLBRACKET;
1470                                         case ']':
1471                                                 return TokenNameRBRACKET;
1472                                         case ';':
1473                                                 return TokenNameSEMICOLON;
1474                                         case ',':
1475                                                 return TokenNameCOMMA;
1476                                         case '.':
1477                                                 if (getNextChar('='))
1478                                                         return TokenNameDOT_EQUAL;
1479                                                 if (getNextCharAsDigit())
1480                                                         return scanNumber(true);
1481                                                 return TokenNameDOT;
1482                                         case '+': {
1483                                                 int test;
1484                                                 if ((test = getNextChar('+', '=')) == 0)
1485                                                         return TokenNamePLUS_PLUS;
1486                                                 if (test > 0)
1487                                                         return TokenNamePLUS_EQUAL;
1488                                                 return TokenNamePLUS;
1489                                         }
1490                                         case '-': {
1491                                                 int test;
1492                                                 if ((test = getNextChar('-', '=')) == 0)
1493                                                         return TokenNameMINUS_MINUS;
1494                                                 if (test > 0)
1495                                                         return TokenNameMINUS_EQUAL;
1496                                                 if (getNextChar('>'))
1497                                                         return TokenNameMINUS_GREATER;
1498                                                 return TokenNameMINUS;
1499                                         }
1500                                         case '~':
1501                                                 if (getNextChar('='))
1502                                                         return TokenNameTWIDDLE_EQUAL;
1503                                                 return TokenNameTWIDDLE;
1504                                         case '!':
1505                                                 if (getNextChar('=')) {
1506                                                         if (getNextChar('=')) {
1507                                                                 return TokenNameNOT_EQUAL_EQUAL;
1508                                                         }
1509                                                         return TokenNameNOT_EQUAL;
1510                                                 }
1511                                                 return TokenNameNOT;
1512                                         case '*':
1513                                                 if (getNextChar('='))
1514                                                         return TokenNameMULTIPLY_EQUAL;
1515                                                 return TokenNameMULTIPLY;
1516                                         case '%':
1517                                                 if (getNextChar('='))
1518                                                         return TokenNameREMAINDER_EQUAL;
1519                                                 return TokenNameREMAINDER;
1520                                         case '<': {
1521                                                 int oldPosition = currentPosition;
1522                                                 try {
1523                                                         currentCharacter = source[currentPosition++];
1524                                                 } catch (IndexOutOfBoundsException e) {
1525                                                         currentPosition = oldPosition;
1526                                                         return TokenNameLESS;
1527                                                 }
1528                                                 switch (currentCharacter) {
1529                                                 case '=':
1530                                                         return TokenNameLESS_EQUAL;
1531                                                 case '>':
1532                                                         return TokenNameNOT_EQUAL;
1533                                                 case '<':
1534                                                         if (getNextChar('='))
1535                                                                 return TokenNameLEFT_SHIFT_EQUAL;
1536                                                         if (getNextChar('<')) {
1537                                                                 currentCharacter = source[currentPosition++];
1538                                                                 while (Character.isWhitespace(currentCharacter)) {
1539                                                                         currentCharacter = source[currentPosition++];
1540                                                                 }
1541                                                                 int heredocStart = currentPosition - 1;
1542                                                                 int heredocLength = 0;
1543                                                                 if (isPHPIdentifierStart(currentCharacter)) {
1544                                                                         currentCharacter = source[currentPosition++];
1545                                                                 } else {
1546                                                                         return TokenNameERROR;
1547                                                                 }
1548                                                                 while (isPHPIdentifierPart(currentCharacter)) {
1549                                                                         currentCharacter = source[currentPosition++];
1550                                                                 }
1551                                                                 heredocLength = currentPosition - heredocStart - 1;
1552                                                                 // heredoc end-tag determination
1553                                                                 boolean endTag = true;
1554                                                                 char ch;
1555                                                                 do {
1556                                                                         ch = source[currentPosition++];
1557                                                                         if (ch == '\r' || ch == '\n') {
1558                                                                                 if (recordLineSeparator) {
1559                                                                                         pushLineSeparator();
1560                                                                                 } else {
1561                                                                                         currentLine = null;
1562                                                                                 }
1563                                                                                 for (int i = 0; i < heredocLength; i++) {
1564                                                                                         if (source[currentPosition + i] != source[heredocStart + i]) {
1565                                                                                                 endTag = false;
1566                                                                                                 break;
1567                                                                                         }
1568                                                                                 }
1569                                                                                 if (endTag) {
1570                                                                                         currentPosition += heredocLength - 1;
1571                                                                                         currentCharacter = source[currentPosition++];
1572                                                                                         break; // do...while loop
1573                                                                                 } else {
1574                                                                                         endTag = true;
1575                                                                                 }
1576                                                                         }
1577                                                                 } while (true);
1578                                                                 return TokenNameHEREDOC;
1579                                                         }
1580                                                         return TokenNameLEFT_SHIFT;
1581                                                 }
1582                                                 currentPosition = oldPosition;
1583                                                 return TokenNameLESS;
1584                                         }
1585                                         case '>': {
1586                                                 int test;
1587                                                 if ((test = getNextChar('=', '>')) == 0)
1588                                                         return TokenNameGREATER_EQUAL;
1589                                                 if (test > 0) {
1590                                                         if ((test = getNextChar('=', '>')) == 0)
1591                                                                 return TokenNameRIGHT_SHIFT_EQUAL;
1592                                                         return TokenNameRIGHT_SHIFT;
1593                                                 }
1594                                                 return TokenNameGREATER;
1595                                         }
1596                                         case '=':
1597                                                 if (getNextChar('=')) {
1598                                                         if (getNextChar('=')) {
1599                                                                 return TokenNameEQUAL_EQUAL_EQUAL;
1600                                                         }
1601                                                         return TokenNameEQUAL_EQUAL;
1602                                                 }
1603                                                 if (getNextChar('>'))
1604                                                         return TokenNameEQUAL_GREATER;
1605                                                 return TokenNameEQUAL;
1606                                         case '&': {
1607                                                 int test;
1608                                                 if ((test = getNextChar('&', '=')) == 0)
1609                                                         return TokenNameAND_AND;
1610                                                 if (test > 0)
1611                                                         return TokenNameAND_EQUAL;
1612                                                 return TokenNameAND;
1613                                         }
1614                                         case '|': {
1615                                                 int test;
1616                                                 if ((test = getNextChar('|', '=')) == 0)
1617                                                         return TokenNameOR_OR;
1618                                                 if (test > 0)
1619                                                         return TokenNameOR_EQUAL;
1620                                                 return TokenNameOR;
1621                                         }
1622                                         case '^':
1623                                                 if (getNextChar('='))
1624                                                         return TokenNameXOR_EQUAL;
1625                                                 return TokenNameXOR;
1626                                         case '?':
1627                                                 if (getNextChar('>')) {
1628                                                         phpMode = false;
1629                                                         if (currentPosition == source.length) {
1630                                                                 phpMode = true;
1631                                                                 return TokenNameINLINE_HTML;
1632                                                         }
1633                                                         return getInlinedHTMLToken(currentPosition - 2);
1634                                                 }
1635                                                 return TokenNameQUESTION;
1636                                         case ':':
1637                                                 if (getNextChar(':'))
1638                                                         return TokenNamePAAMAYIM_NEKUDOTAYIM;
1639                                                 return TokenNameCOLON;
1640                                         case '@':
1641                                                 return TokenNameAT;
1642                                         case '\'':
1643                                                 consumeStringConstant();
1644                                                 return TokenNameStringSingleQuote;
1645                                         case '"':
1646                                                 // if (tokenizeStrings) {
1647                                                 consumeStringLiteral();
1648                                                 return TokenNameStringDoubleQuote;
1649                                         // }
1650                                         // return TokenNameEncapsedString2;
1651                                         case '`':
1652                                                 // if (tokenizeStrings) {
1653                                                 consumeStringInterpolated();
1654                                                 return TokenNameStringInterpolated;
1655                                         // }
1656                                         // return TokenNameEncapsedString0;
1657                                         case '#':
1658                                         case '/': {
1659                                                 char startChar = currentCharacter;
1660                                                 if (getNextChar('=') && startChar == '/') {
1661                                                         return TokenNameDIVIDE_EQUAL;
1662                                                 }
1663                                                 int test;
1664                                                 if ((startChar == '#') || (test = getNextChar('/', '*')) == 0) {
1665                                                         // line comment
1666                                                         this.lastCommentLinePosition = this.currentPosition;
1667                                                         int endPositionForLineComment = 0;
1668                                                         try { // get the next char
1669                                                                 currentCharacter = source[currentPosition++];
1670                                                                 // if (((currentCharacter = source[currentPosition++])
1671                                                                 // == '\\')
1672                                                                 // && (source[currentPosition] == 'u')) {
1673                                                                 // //-------------unicode traitement ------------
1674                                                                 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
1675                                                                 // currentPosition++;
1676                                                                 // while (source[currentPosition] == 'u') {
1677                                                                 // currentPosition++;
1678                                                                 // }
1679                                                                 // if ((c1 =
1680                                                                 // Character.getNumericValue(source[currentPosition++]))
1681                                                                 // > 15
1682                                                                 // || c1 < 0
1683                                                                 // || (c2 =
1684                                                                 // Character.getNumericValue(source[currentPosition++]))
1685                                                                 // > 15
1686                                                                 // || c2 < 0
1687                                                                 // || (c3 =
1688                                                                 // Character.getNumericValue(source[currentPosition++]))
1689                                                                 // > 15
1690                                                                 // || c3 < 0
1691                                                                 // || (c4 =
1692                                                                 // Character.getNumericValue(source[currentPosition++]))
1693                                                                 // > 15
1694                                                                 // || c4 < 0) {
1695                                                                 // throw new
1696                                                                 // InvalidInputException(INVALID_UNICODE_ESCAPE);
1697                                                                 // } else {
1698                                                                 // currentCharacter =
1699                                                                 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
1700                                                                 // }
1701                                                                 // }
1702                                                                 // handle the \\u case manually into comment
1703                                                                 // if (currentCharacter == '\\') {
1704                                                                 // if (source[currentPosition] == '\\')
1705                                                                 // currentPosition++;
1706                                                                 // } //jump over the \\
1707                                                                 boolean isUnicode = false;
1708                                                                 while (currentCharacter != '\r' && currentCharacter != '\n') {
1709                                                                         this.lastCommentLinePosition = this.currentPosition;
1710                                                                         if (currentCharacter == '?') {
1711                                                                                 if (getNextChar('>')) {
1712                                                                                         // ?> breaks line comments
1713                                                                                         startPosition = currentPosition - 2;
1714                                                                                         phpMode = false;
1715                                                                                         return TokenNameINLINE_HTML;
1716                                                                                 }
1717                                                                         }
1718                                                                         // get the next char
1719                                                                         isUnicode = false;
1720                                                                         currentCharacter = source[currentPosition++];
1721                                                                         // if (((currentCharacter = source[currentPosition++])
1722                                                                         // == '\\')
1723                                                                         // && (source[currentPosition] == 'u')) {
1724                                                                         // isUnicode = true;
1725                                                                         // //-------------unicode traitement ------------
1726                                                                         // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
1727                                                                         // currentPosition++;
1728                                                                         // while (source[currentPosition] == 'u') {
1729                                                                         // currentPosition++;
1730                                                                         // }
1731                                                                         // if ((c1 =
1732                                                                         // Character.getNumericValue(source[currentPosition++]))
1733                                                                         // > 15
1734                                                                         // || c1 < 0
1735                                                                         // || (c2 =
1736                                                                         // Character.getNumericValue(
1737                                                                         // source[currentPosition++]))
1738                                                                         // > 15
1739                                                                         // || c2 < 0
1740                                                                         // || (c3 =
1741                                                                         // Character.getNumericValue(
1742                                                                         // source[currentPosition++]))
1743                                                                         // > 15
1744                                                                         // || c3 < 0
1745                                                                         // || (c4 =
1746                                                                         // Character.getNumericValue(
1747                                                                         // source[currentPosition++]))
1748                                                                         // > 15
1749                                                                         // || c4 < 0) {
1750                                                                         // throw new
1751                                                                         // InvalidInputException(INVALID_UNICODE_ESCAPE);
1752                                                                         // } else {
1753                                                                         // currentCharacter =
1754                                                                         // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
1755                                                                         // }
1756                                                                         // }
1757                                                                         // handle the \\u case manually into comment
1758                                                                         // if (currentCharacter == '\\') {
1759                                                                         // if (source[currentPosition] == '\\')
1760                                                                         // currentPosition++;
1761                                                                         // } //jump over the \\
1762                                                                 }
1763                                                                 if (isUnicode) {
1764                                                                         endPositionForLineComment = currentPosition - 6;
1765                                                                 } else {
1766                                                                         endPositionForLineComment = currentPosition - 1;
1767                                                                 }
1768                                                                 // recordComment(false);
1769                                                                 recordComment(TokenNameCOMMENT_LINE);
1770                                                                 if (this.taskTags != null)
1771                                                                         checkTaskTag(this.startPosition, this.currentPosition);
1772                                                                 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1773                                                                         checkNonExternalizeString();
1774                                                                         if (recordLineSeparator) {
1775                                                                                 if (isUnicode) {
1776                                                                                         pushUnicodeLineSeparator();
1777                                                                                 } else {
1778                                                                                         pushLineSeparator();
1779                                                                                 }
1780                                                                         } else {
1781                                                                                 currentLine = null;
1782                                                                         }
1783                                                                 }
1784                                                                 if (tokenizeComments) {
1785                                                                         if (!isUnicode) {
1786                                                                                 currentPosition = endPositionForLineComment;
1787                                                                                 // reset one character behind
1788                                                                         }
1789                                                                         return TokenNameCOMMENT_LINE;
1790                                                                 }
1791                                                         } catch (IndexOutOfBoundsException e) { // an eof will them
1792                                                                 // be generated
1793                                                                 if (tokenizeComments) {
1794                                                                         currentPosition--;
1795                                                                         // reset one character behind
1796                                                                         return TokenNameCOMMENT_LINE;
1797                                                                 }
1798                                                         }
1799                                                         break;
1800                                                 }
1801                                                 if (test > 0) {
1802                                                         // traditional and annotation comment
1803                                                         boolean isJavadoc = false, star = false;
1804                                                         // consume next character
1805                                                         unicodeAsBackSlash = false;
1806                                                         currentCharacter = source[currentPosition++];
1807                                                         // if (((currentCharacter = source[currentPosition++]) ==
1808                                                         // '\\')
1809                                                         // && (source[currentPosition] == 'u')) {
1810                                                         // getNextUnicodeChar();
1811                                                         // } else {
1812                                                         // if (withoutUnicodePtr != 0) {
1813                                                         // withoutUnicodeBuffer[++withoutUnicodePtr] =
1814                                                         // currentCharacter;
1815                                                         // }
1816                                                         // }
1817                                                         if (currentCharacter == '*') {
1818                                                                 isJavadoc = true;
1819                                                                 star = true;
1820                                                         }
1821                                                         if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1822                                                                 checkNonExternalizeString();
1823                                                                 if (recordLineSeparator) {
1824                                                                         pushLineSeparator();
1825                                                                 } else {
1826                                                                         currentLine = null;
1827                                                                 }
1828                                                         }
1829                                                         try { // get the next char
1830                                                                 currentCharacter = source[currentPosition++];
1831                                                                 // if (((currentCharacter = source[currentPosition++])
1832                                                                 // == '\\')
1833                                                                 // && (source[currentPosition] == 'u')) {
1834                                                                 // //-------------unicode traitement ------------
1835                                                                 // getNextUnicodeChar();
1836                                                                 // }
1837                                                                 // handle the \\u case manually into comment
1838                                                                 // if (currentCharacter == '\\') {
1839                                                                 // if (source[currentPosition] == '\\')
1840                                                                 // currentPosition++;
1841                                                                 // //jump over the \\
1842                                                                 // }
1843                                                                 // empty comment is not a javadoc /**/
1844                                                                 if (currentCharacter == '/') {
1845                                                                         isJavadoc = false;
1846                                                                 }
1847                                                                 // loop until end of comment */
1848                                                                 while ((currentCharacter != '/') || (!star)) {
1849                                                                         if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1850                                                                                 checkNonExternalizeString();
1851                                                                                 if (recordLineSeparator) {
1852                                                                                         pushLineSeparator();
1853                                                                                 } else {
1854                                                                                         currentLine = null;
1855                                                                                 }
1856                                                                         }
1857                                                                         star = currentCharacter == '*';
1858                                                                         // get next char
1859                                                                         currentCharacter = source[currentPosition++];
1860                                                                         // if (((currentCharacter = source[currentPosition++])
1861                                                                         // == '\\')
1862                                                                         // && (source[currentPosition] == 'u')) {
1863                                                                         // //-------------unicode traitement ------------
1864                                                                         // getNextUnicodeChar();
1865                                                                         // }
1866                                                                         // handle the \\u case manually into comment
1867                                                                         // if (currentCharacter == '\\') {
1868                                                                         // if (source[currentPosition] == '\\')
1869                                                                         // currentPosition++;
1870                                                                         // } //jump over the \\
1871                                                                 }
1872                                                                 // recordComment(isJavadoc);
1873                                                                 if (isJavadoc) {
1874                                                                         recordComment(TokenNameCOMMENT_PHPDOC);
1875                                                                 } else {
1876                                                                         recordComment(TokenNameCOMMENT_BLOCK);
1877                                                                 }
1878
1879                                                                 if (tokenizeComments) {
1880                                                                         if (isJavadoc)
1881                                                                                 return TokenNameCOMMENT_PHPDOC;
1882                                                                         return TokenNameCOMMENT_BLOCK;
1883                                                                 }
1884
1885                                                                 if (this.taskTags != null) {
1886                                                                         checkTaskTag(this.startPosition, this.currentPosition);
1887                                                                 }
1888                                                         } catch (IndexOutOfBoundsException e) {
1889                                                                 // reset end position for error reporting
1890                                                                 currentPosition -= 2;
1891                                                                 throw new InvalidInputException(UNTERMINATED_COMMENT);
1892                                                         }
1893                                                         break;
1894                                                 }
1895                                                 return TokenNameDIVIDE;
1896                                         }
1897                                         case '\u001a':
1898                                                 if (atEnd())
1899                                                         return TokenNameEOF;
1900                                                 // the atEnd may not be <currentPosition == source.length> if
1901                                                 // source is only some part of a real (external) stream
1902                                                 throw new InvalidInputException("Ctrl-Z"); //$NON-NLS-1$
1903                                         default:
1904                                                 if (currentCharacter == '$') {
1905                                                         int oldPosition = currentPosition;
1906                                                         try {
1907                                                                 currentCharacter = source[currentPosition++];
1908                                                                 if (isPHPIdentifierStart(currentCharacter)) {
1909                                                                         return scanIdentifierOrKeyword(true);
1910                                                                 } else {
1911                                                                         currentPosition = oldPosition;
1912                                                                         return TokenNameDOLLAR;
1913                                                                 }
1914                                                         } catch (IndexOutOfBoundsException e) {
1915                                                                 currentPosition = oldPosition;
1916                                                                 return TokenNameDOLLAR;
1917                                                         }
1918                                                 }
1919                                                 if (isPHPIdentifierStart(currentCharacter))
1920                                                         return scanIdentifierOrKeyword(false);
1921                                                 if (Character.isDigit(currentCharacter))
1922                                                         return scanNumber(false);
1923                                                 return TokenNameERROR;
1924                                         }
1925                                 }
1926                         } // -----------------end switch while try--------------------
1927                         catch (IndexOutOfBoundsException e) {
1928                         }
1929                 }
1930                 return TokenNameEOF;
1931         }
1932
1933         /**
1934          * @return
1935          * @throws InvalidInputException
1936          */
1937         private int getInlinedHTMLToken(int start) throws InvalidInputException {
1938                 if (currentPosition > source.length) {
1939                         currentPosition = source.length;
1940                         return TokenNameEOF;
1941                 }
1942                 startPosition = start;
1943                 try {
1944                         while (!phpMode) {
1945                                 currentCharacter = source[currentPosition++];
1946                                 if (currentCharacter == '<') {
1947                                         if (getNextChar('?')) {
1948                                                 currentCharacter = source[currentPosition++];
1949                                                 if ((currentCharacter != 'P') && (currentCharacter != 'p')) {
1950                                                         if (currentCharacter != '=') { // <?=
1951                                                                 currentPosition--;
1952                                                         } else {
1953                                                                 phpExpressionTag = true;
1954                                                         }
1955                                                         // <?
1956                                                         if (ignorePHPOneLiner) { // for CodeFormatter
1957                                                                 if (lookAheadLinePHPTag() == TokenNameINLINE_HTML) {
1958                                                                         phpMode = true;
1959                                                                         return TokenNameINLINE_HTML;
1960                                                                 }
1961                                                         } else {
1962                                                                 phpMode = true;
1963                                                                 return TokenNameINLINE_HTML;
1964                                                         }
1965                                                 } else {
1966                                                         // boolean phpStart = (currentCharacter == 'P') ||
1967                                                         // (currentCharacter == 'p');
1968                                                         // if (phpStart) {
1969                                                         int test = getNextChar('H', 'h');
1970                                                         if (test >= 0) {
1971                                                                 test = getNextChar('P', 'p');
1972                                                                 if (test >= 0) {
1973                                                                         // <?PHP <?php
1974                                                                         if (ignorePHPOneLiner) {
1975                                                                                 if (lookAheadLinePHPTag() == TokenNameINLINE_HTML) {
1976                                                                                         phpMode = true;
1977                                                                                         return TokenNameINLINE_HTML;
1978                                                                                 }
1979                                                                         } else {
1980                                                                                 phpMode = true;
1981                                                                                 return TokenNameINLINE_HTML;
1982                                                                         }
1983                                                                 }
1984                                                         }
1985                                                         // }
1986                                                 }
1987                                         }
1988                                 }
1989                                 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1990                                         if (recordLineSeparator) {
1991                                                 pushLineSeparator();
1992                                         } else {
1993                                                 currentLine = null;
1994                                         }
1995                                 }
1996                         } // -----------------while--------------------
1997                         phpMode = true;
1998                         return TokenNameINLINE_HTML;
1999                 } // -----------------try--------------------
2000                 catch (IndexOutOfBoundsException e) {
2001                         startPosition = start;
2002                         currentPosition--;
2003                 }
2004                 phpMode = true;
2005                 return TokenNameINLINE_HTML;
2006         }
2007
2008         /**
2009          * @return
2010          */
2011         private int lookAheadLinePHPTag() {
2012                 // check if the PHP is only in this line (for CodeFormatter)
2013                 int currentPositionInLine = currentPosition;
2014                 char previousCharInLine = ' ';
2015                 char currentCharInLine = ' ';
2016                 boolean singleQuotedStringActive = false;
2017                 boolean doubleQuotedStringActive = false;
2018
2019                 try {
2020                         // look ahead in this line
2021                         while (true) {
2022                                 previousCharInLine = currentCharInLine;
2023                                 currentCharInLine = source[currentPositionInLine++];
2024                                 switch (currentCharInLine) {
2025                                 case '>':
2026                                         if (previousCharInLine == '?') {
2027                                                 // update the scanner's current Position in the source
2028                                                 currentPosition = currentPositionInLine;
2029                                                 // use as "dummy" token
2030                                                 return TokenNameEOF;
2031                                         }
2032                                         break;
2033                                 case '\\':
2034                                         if (doubleQuotedStringActive) {
2035                                                 // ignore escaped characters in double quoted strings
2036                                                 previousCharInLine = currentCharInLine;
2037                                                 currentCharInLine = source[currentPositionInLine++];
2038                                         }
2039                                 case '\"':
2040                                         if (doubleQuotedStringActive) {
2041                                                 doubleQuotedStringActive = false;
2042                                         } else {
2043                                                 if (!singleQuotedStringActive) {
2044                                                         doubleQuotedStringActive = true;
2045                                                 }
2046                                         }
2047                                         break;
2048                                 case '\'':
2049                                         if (singleQuotedStringActive) {
2050                                                 if (previousCharInLine != '\\') {
2051                                                         singleQuotedStringActive = false;
2052                                                 }
2053                                         } else {
2054                                                 if (!doubleQuotedStringActive) {
2055                                                         singleQuotedStringActive = true;
2056                                                 }
2057                                         }
2058                                         break;
2059                                 case '\n':
2060                                         phpMode = true;
2061                                         return TokenNameINLINE_HTML;
2062                                 case '#':
2063                                         if (!singleQuotedStringActive && !doubleQuotedStringActive) {
2064                                                 phpMode = true;
2065                                                 return TokenNameINLINE_HTML;
2066                                         }
2067                                         break;
2068                                 case '/':
2069                                         if (previousCharInLine == '/' && !singleQuotedStringActive && !doubleQuotedStringActive) {
2070                                                 phpMode = true;
2071                                                 return TokenNameINLINE_HTML;
2072                                         }
2073                                         break;
2074                                 case '*':
2075                                         if (previousCharInLine == '/' && !singleQuotedStringActive && !doubleQuotedStringActive) {
2076                                                 phpMode = true;
2077                                                 return TokenNameINLINE_HTML;
2078                                         }
2079                                         break;
2080                                 }
2081                         }
2082                 } catch (IndexOutOfBoundsException e) {
2083                         phpMode = true;
2084                         currentPosition = currentPositionInLine;
2085                         return TokenNameINLINE_HTML;
2086                 }
2087         }
2088
2089         // public final void getNextUnicodeChar()
2090         // throws IndexOutOfBoundsException, InvalidInputException {
2091         // //VOID
2092         // //handle the case of unicode.
2093         // //when a unicode appears then we must use a buffer that holds char
2094         // internal values
2095         // //At the end of this method currentCharacter holds the new visited char
2096         // //and currentPosition points right next after it
2097         //
2098         // //ALL getNextChar.... ARE OPTIMIZED COPIES
2099         //
2100         // int c1 = 0, c2 = 0, c3 = 0, c4 = 0, unicodeSize = 6;
2101         // currentPosition++;
2102         // while (source[currentPosition] == 'u') {
2103         // currentPosition++;
2104         // unicodeSize++;
2105         // }
2106         //
2107         // if ((c1 = Character.getNumericValue(source[currentPosition++])) > 15
2108         // || c1 < 0
2109         // || (c2 = Character.getNumericValue(source[currentPosition++])) > 15
2110         // || c2 < 0
2111         // || (c3 = Character.getNumericValue(source[currentPosition++])) > 15
2112         // || c3 < 0
2113         // || (c4 = Character.getNumericValue(source[currentPosition++])) > 15
2114         // || c4 < 0) {
2115         // throw new InvalidInputException(INVALID_UNICODE_ESCAPE);
2116         // } else {
2117         // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2118         // //need the unicode buffer
2119         // if (withoutUnicodePtr == 0) {
2120         // //buffer all the entries that have been left aside....
2121         // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
2122         // System.arraycopy(
2123         // source,
2124         // startPosition,
2125         // withoutUnicodeBuffer,
2126         // 1,
2127         // withoutUnicodePtr);
2128         // }
2129         // //fill the buffer with the char
2130         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2131         // }
2132         // unicodeAsBackSlash = currentCharacter == '\\';
2133         // }
2134         /*
2135          * Tokenize a method body, assuming that curly brackets are properly balanced.
2136          */
2137         public final void jumpOverMethodBody() {
2138                 this.wasAcr = false;
2139                 int found = 1;
2140                 try {
2141                         while (true) { // loop for jumping over comments
2142                                 // ---------Consume white space and handles startPosition---------
2143                                 boolean isWhiteSpace;
2144                                 do {
2145                                         startPosition = currentPosition;
2146                                         currentCharacter = source[currentPosition++];
2147                                         // if (((currentCharacter = source[currentPosition++]) == '\\')
2148                                         // && (source[currentPosition] == 'u')) {
2149                                         // isWhiteSpace = jumpOverUnicodeWhiteSpace();
2150                                         // } else {
2151                                         if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2152                                                 pushLineSeparator();
2153                                         isWhiteSpace = Character.isWhitespace(currentCharacter);
2154                                         // }
2155                                 } while (isWhiteSpace);
2156                                 // -------consume token until } is found---------
2157                                 switch (currentCharacter) {
2158                                 case '{':
2159                                         found++;
2160                                         break;
2161                                 case '}':
2162                                         found--;
2163                                         if (found == 0)
2164                                                 return;
2165                                         break;
2166                                 case '\'': {
2167                                         boolean test;
2168                                         test = getNextChar('\\');
2169                                         if (test) {
2170                                                 try {
2171                                                         scanDoubleQuotedEscapeCharacter();
2172                                                 } catch (InvalidInputException ex) {
2173                                                 }
2174                                                 ;
2175                                         } else {
2176                                                 // try { // consume next character
2177                                                 unicodeAsBackSlash = false;
2178                                                 currentCharacter = source[currentPosition++];
2179                                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
2180                                                 // && (source[currentPosition] == 'u')) {
2181                                                 // getNextUnicodeChar();
2182                                                 // } else {
2183                                                 if (withoutUnicodePtr != 0) {
2184                                                         withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2185                                                 }
2186                                                 // }
2187                                                 // } catch (InvalidInputException ex) {
2188                                                 // };
2189                                         }
2190                                         getNextChar('\'');
2191                                         break;
2192                                 }
2193                                 case '"':
2194                                         try {
2195                                                 // try { // consume next character
2196                                                 unicodeAsBackSlash = false;
2197                                                 currentCharacter = source[currentPosition++];
2198                                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
2199                                                 // && (source[currentPosition] == 'u')) {
2200                                                 // getNextUnicodeChar();
2201                                                 // } else {
2202                                                 if (withoutUnicodePtr != 0) {
2203                                                         withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2204                                                 }
2205                                                 // }
2206                                                 // } catch (InvalidInputException ex) {
2207                                                 // };
2208                                                 while (currentCharacter != '"') {
2209                                                         if (currentCharacter == '\r') {
2210                                                                 if (source[currentPosition] == '\n')
2211                                                                         currentPosition++;
2212                                                                 break;
2213                                                                 // the string cannot go further that the line
2214                                                         }
2215                                                         if (currentCharacter == '\n') {
2216                                                                 break;
2217                                                                 // the string cannot go further that the line
2218                                                         }
2219                                                         if (currentCharacter == '\\') {
2220                                                                 try {
2221                                                                         scanDoubleQuotedEscapeCharacter();
2222                                                                 } catch (InvalidInputException ex) {
2223                                                                 }
2224                                                                 ;
2225                                                         }
2226                                                         // try { // consume next character
2227                                                         unicodeAsBackSlash = false;
2228                                                         currentCharacter = source[currentPosition++];
2229                                                         // if (((currentCharacter = source[currentPosition++]) == '\\')
2230                                                         // && (source[currentPosition] == 'u')) {
2231                                                         // getNextUnicodeChar();
2232                                                         // } else {
2233                                                         if (withoutUnicodePtr != 0) {
2234                                                                 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2235                                                         }
2236                                                         // }
2237                                                         // } catch (InvalidInputException ex) {
2238                                                         // };
2239                                                 }
2240                                         } catch (IndexOutOfBoundsException e) {
2241                                                 return;
2242                                         }
2243                                         break;
2244                                 case '/': {
2245                                         int test;
2246                                         if ((test = getNextChar('/', '*')) == 0) {
2247                                                 // line comment
2248                                                 try {
2249                                                         // get the next char
2250                                                         currentCharacter = source[currentPosition++];
2251                                                         // if (((currentCharacter = source[currentPosition++]) ==
2252                                                         // '\\')
2253                                                         // && (source[currentPosition] == 'u')) {
2254                                                         // //-------------unicode traitement ------------
2255                                                         // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2256                                                         // currentPosition++;
2257                                                         // while (source[currentPosition] == 'u') {
2258                                                         // currentPosition++;
2259                                                         // }
2260                                                         // if ((c1 =
2261                                                         // Character.getNumericValue(source[currentPosition++]))
2262                                                         // > 15
2263                                                         // || c1 < 0
2264                                                         // || (c2 =
2265                                                         // Character.getNumericValue(source[currentPosition++]))
2266                                                         // > 15
2267                                                         // || c2 < 0
2268                                                         // || (c3 =
2269                                                         // Character.getNumericValue(source[currentPosition++]))
2270                                                         // > 15
2271                                                         // || c3 < 0
2272                                                         // || (c4 =
2273                                                         // Character.getNumericValue(source[currentPosition++]))
2274                                                         // > 15
2275                                                         // || c4 < 0) {
2276                                                         // //error don't care of the value
2277                                                         // currentCharacter = 'A';
2278                                                         // } //something different from \n and \r
2279                                                         // else {
2280                                                         // currentCharacter =
2281                                                         // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2282                                                         // }
2283                                                         // }
2284                                                         while (currentCharacter != '\r' && currentCharacter != '\n') {
2285                                                                 // get the next char
2286                                                                 currentCharacter = source[currentPosition++];
2287                                                                 // if (((currentCharacter = source[currentPosition++])
2288                                                                 // == '\\')
2289                                                                 // && (source[currentPosition] == 'u')) {
2290                                                                 // //-------------unicode traitement ------------
2291                                                                 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2292                                                                 // currentPosition++;
2293                                                                 // while (source[currentPosition] == 'u') {
2294                                                                 // currentPosition++;
2295                                                                 // }
2296                                                                 // if ((c1 =
2297                                                                 // Character.getNumericValue(source[currentPosition++]))
2298                                                                 // > 15
2299                                                                 // || c1 < 0
2300                                                                 // || (c2 =
2301                                                                 // Character.getNumericValue(source[currentPosition++]))
2302                                                                 // > 15
2303                                                                 // || c2 < 0
2304                                                                 // || (c3 =
2305                                                                 // Character.getNumericValue(source[currentPosition++]))
2306                                                                 // > 15
2307                                                                 // || c3 < 0
2308                                                                 // || (c4 =
2309                                                                 // Character.getNumericValue(source[currentPosition++]))
2310                                                                 // > 15
2311                                                                 // || c4 < 0) {
2312                                                                 // //error don't care of the value
2313                                                                 // currentCharacter = 'A';
2314                                                                 // } //something different from \n and \r
2315                                                                 // else {
2316                                                                 // currentCharacter =
2317                                                                 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2318                                                                 // }
2319                                                                 // }
2320                                                         }
2321                                                         if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2322                                                                 pushLineSeparator();
2323                                                 } catch (IndexOutOfBoundsException e) {
2324                                                 } // an eof will them be generated
2325                                                 break;
2326                                         }
2327                                         if (test > 0) {
2328                                                 // traditional and annotation comment
2329                                                 boolean star = false;
2330                                                 // try { // consume next character
2331                                                 unicodeAsBackSlash = false;
2332                                                 currentCharacter = source[currentPosition++];
2333                                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
2334                                                 // && (source[currentPosition] == 'u')) {
2335                                                 // getNextUnicodeChar();
2336                                                 // } else {
2337                                                 if (withoutUnicodePtr != 0) {
2338                                                         withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2339                                                 }
2340                                                 // };
2341                                                 // } catch (InvalidInputException ex) {
2342                                                 // };
2343                                                 if (currentCharacter == '*') {
2344                                                         star = true;
2345                                                 }
2346                                                 if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2347                                                         pushLineSeparator();
2348                                                 try { // get the next char
2349                                                         currentCharacter = source[currentPosition++];
2350                                                         // if (((currentCharacter = source[currentPosition++]) ==
2351                                                         // '\\')
2352                                                         // && (source[currentPosition] == 'u')) {
2353                                                         // //-------------unicode traitement ------------
2354                                                         // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2355                                                         // currentPosition++;
2356                                                         // while (source[currentPosition] == 'u') {
2357                                                         // currentPosition++;
2358                                                         // }
2359                                                         // if ((c1 =
2360                                                         // Character.getNumericValue(source[currentPosition++]))
2361                                                         // > 15
2362                                                         // || c1 < 0
2363                                                         // || (c2 =
2364                                                         // Character.getNumericValue(source[currentPosition++]))
2365                                                         // > 15
2366                                                         // || c2 < 0
2367                                                         // || (c3 =
2368                                                         // Character.getNumericValue(source[currentPosition++]))
2369                                                         // > 15
2370                                                         // || c3 < 0
2371                                                         // || (c4 =
2372                                                         // Character.getNumericValue(source[currentPosition++]))
2373                                                         // > 15
2374                                                         // || c4 < 0) {
2375                                                         // //error don't care of the value
2376                                                         // currentCharacter = 'A';
2377                                                         // } //something different from * and /
2378                                                         // else {
2379                                                         // currentCharacter =
2380                                                         // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2381                                                         // }
2382                                                         // }
2383                                                         // loop until end of comment */
2384                                                         while ((currentCharacter != '/') || (!star)) {
2385                                                                 if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2386                                                                         pushLineSeparator();
2387                                                                 star = currentCharacter == '*';
2388                                                                 // get next char
2389                                                                 currentCharacter = source[currentPosition++];
2390                                                                 // if (((currentCharacter = source[currentPosition++])
2391                                                                 // == '\\')
2392                                                                 // && (source[currentPosition] == 'u')) {
2393                                                                 // //-------------unicode traitement ------------
2394                                                                 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2395                                                                 // currentPosition++;
2396                                                                 // while (source[currentPosition] == 'u') {
2397                                                                 // currentPosition++;
2398                                                                 // }
2399                                                                 // if ((c1 =
2400                                                                 // Character.getNumericValue(source[currentPosition++]))
2401                                                                 // > 15
2402                                                                 // || c1 < 0
2403                                                                 // || (c2 =
2404                                                                 // Character.getNumericValue(source[currentPosition++]))
2405                                                                 // > 15
2406                                                                 // || c2 < 0
2407                                                                 // || (c3 =
2408                                                                 // Character.getNumericValue(source[currentPosition++]))
2409                                                                 // > 15
2410                                                                 // || c3 < 0
2411                                                                 // || (c4 =
2412                                                                 // Character.getNumericValue(source[currentPosition++]))
2413                                                                 // > 15
2414                                                                 // || c4 < 0) {
2415                                                                 // //error don't care of the value
2416                                                                 // currentCharacter = 'A';
2417                                                                 // } //something different from * and /
2418                                                                 // else {
2419                                                                 // currentCharacter =
2420                                                                 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2421                                                                 // }
2422                                                                 // }
2423                                                         }
2424                                                 } catch (IndexOutOfBoundsException e) {
2425                                                         return;
2426                                                 }
2427                                                 break;
2428                                         }
2429                                         break;
2430                                 }
2431                                 default:
2432                                         if (isPHPIdentOrVarStart(currentCharacter) ) {
2433                                                 try {
2434                                                         scanIdentifierOrKeyword((currentCharacter == '$'));
2435                                                 } catch (InvalidInputException ex) {
2436                                                 }
2437                                                 ;
2438                                                 break;
2439                                         }
2440                                 if ( ObviousIdentCharNatures[currentCharacter]==C_DIGIT) {
2441 //                                      if (Character.isDigit(currentCharacter)) {
2442                                                 try {
2443                                                         scanNumber(false);
2444                                                 } catch (InvalidInputException ex) {
2445                                                 }
2446                                                 ;
2447                                                 break;
2448                                         }
2449                                 }
2450                         }
2451                         // -----------------end switch while try--------------------
2452                 } catch (IndexOutOfBoundsException e) {
2453                 } catch (InvalidInputException e) {
2454                 }
2455                 return;
2456         }
2457
2458         // public final boolean jumpOverUnicodeWhiteSpace()
2459         // throws InvalidInputException {
2460         // //BOOLEAN
2461         // //handle the case of unicode. Jump over the next whiteSpace
2462         // //making startPosition pointing on the next available char
2463         // //On false, the currentCharacter is filled up with a potential
2464         // //correct char
2465         //
2466         // try {
2467         // this.wasAcr = false;
2468         // int c1, c2, c3, c4;
2469         // int unicodeSize = 6;
2470         // currentPosition++;
2471         // while (source[currentPosition] == 'u') {
2472         // currentPosition++;
2473         // unicodeSize++;
2474         // }
2475         //
2476         // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
2477         // || c1 < 0)
2478         // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
2479         // || c2 < 0)
2480         // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
2481         // || c3 < 0)
2482         // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
2483         // || c4 < 0)) {
2484         // throw new InvalidInputException(INVALID_UNICODE_ESCAPE);
2485         // }
2486         //
2487         // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2488         // if (recordLineSeparator
2489         // && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2490         // pushLineSeparator();
2491         // if (Character.isWhitespace(currentCharacter))
2492         // return true;
2493         //
2494         // //buffer the new char which is not a white space
2495         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2496         // //withoutUnicodePtr == 1 is true here
2497         // return false;
2498         // } catch (IndexOutOfBoundsException e) {
2499         // throw new InvalidInputException(INVALID_UNICODE_ESCAPE);
2500         // }
2501         // }
2502         public final int[] getLineEnds() {
2503                 // return a bounded copy of this.lineEnds
2504                 int[] copy;
2505                 System.arraycopy(lineEnds, 0, copy = new int[linePtr + 1], 0, linePtr + 1);
2506                 return copy;
2507         }
2508
2509         public char[] getSource() {
2510                 return this.source;
2511         }
2512
2513         public static boolean isIdentifierOrKeyword(int token) {
2514                 return (token == TokenNameIdentifier) || (token > TokenNameKEYWORD);
2515         }
2516
2517         final char[] optimizedCurrentTokenSource1() {
2518                 // return always the same char[] build only once
2519                 // optimization at no speed cost of 99.5 % of the singleCharIdentifier
2520                 char charOne = source[startPosition];
2521                 switch (charOne) {
2522                 case 'a':
2523                         return charArray_a;
2524                 case 'b':
2525                         return charArray_b;
2526                 case 'c':
2527                         return charArray_c;
2528                 case 'd':
2529                         return charArray_d;
2530                 case 'e':
2531                         return charArray_e;
2532                 case 'f':
2533                         return charArray_f;
2534                 case 'g':
2535                         return charArray_g;
2536                 case 'h':
2537                         return charArray_h;
2538                 case 'i':
2539                         return charArray_i;
2540                 case 'j':
2541                         return charArray_j;
2542                 case 'k':
2543                         return charArray_k;
2544                 case 'l':
2545                         return charArray_l;
2546                 case 'm':
2547                         return charArray_m;
2548                 case 'n':
2549                         return charArray_n;
2550                 case 'o':
2551                         return charArray_o;
2552                 case 'p':
2553                         return charArray_p;
2554                 case 'q':
2555                         return charArray_q;
2556                 case 'r':
2557                         return charArray_r;
2558                 case 's':
2559                         return charArray_s;
2560                 case 't':
2561                         return charArray_t;
2562                 case 'u':
2563                         return charArray_u;
2564                 case 'v':
2565                         return charArray_v;
2566                 case 'w':
2567                         return charArray_w;
2568                 case 'x':
2569                         return charArray_x;
2570                 case 'y':
2571                         return charArray_y;
2572                 case 'z':
2573                         return charArray_z;
2574                 default:
2575                         return new char[] { charOne };
2576                 }
2577         }
2578
2579         final char[] optimizedCurrentTokenSource2() {
2580                 char c0, c1;
2581                 c0 = source[startPosition];
2582                 c1 = source[startPosition + 1];
2583                 if (c0 == '$') {
2584                         // return always the same char[] build only once
2585                         // optimization at no speed cost of 99.5 % of the singleCharIdentifier
2586                         switch (c1) {
2587                         case 'a':
2588                                 return charArray_va;
2589                         case 'b':
2590                                 return charArray_vb;
2591                         case 'c':
2592                                 return charArray_vc;
2593                         case 'd':
2594                                 return charArray_vd;
2595                         case 'e':
2596                                 return charArray_ve;
2597                         case 'f':
2598                                 return charArray_vf;
2599                         case 'g':
2600                                 return charArray_vg;
2601                         case 'h':
2602                                 return charArray_vh;
2603                         case 'i':
2604                                 return charArray_vi;
2605                         case 'j':
2606                                 return charArray_vj;
2607                         case 'k':
2608                                 return charArray_vk;
2609                         case 'l':
2610                                 return charArray_vl;
2611                         case 'm':
2612                                 return charArray_vm;
2613                         case 'n':
2614                                 return charArray_vn;
2615                         case 'o':
2616                                 return charArray_vo;
2617                         case 'p':
2618                                 return charArray_vp;
2619                         case 'q':
2620                                 return charArray_vq;
2621                         case 'r':
2622                                 return charArray_vr;
2623                         case 's':
2624                                 return charArray_vs;
2625                         case 't':
2626                                 return charArray_vt;
2627                         case 'u':
2628                                 return charArray_vu;
2629                         case 'v':
2630                                 return charArray_vv;
2631                         case 'w':
2632                                 return charArray_vw;
2633                         case 'x':
2634                                 return charArray_vx;
2635                         case 'y':
2636                                 return charArray_vy;
2637                         case 'z':
2638                                 return charArray_vz;
2639                         }
2640                 }
2641                 // try to return the same char[] build only once
2642                 int hash = ((c0 << 6) + c1) % TableSize;
2643                 char[][] table = charArray_length[0][hash];
2644                 int i = newEntry2;
2645                 while (++i < InternalTableSize) {
2646                         char[] charArray = table[i];
2647                         if ((c0 == charArray[0]) && (c1 == charArray[1]))
2648                                 return charArray;
2649                 }
2650                 // ---------other side---------
2651                 i = -1;
2652                 int max = newEntry2;
2653                 while (++i <= max) {
2654                         char[] charArray = table[i];
2655                         if ((c0 == charArray[0]) && (c1 == charArray[1]))
2656                                 return charArray;
2657                 }
2658                 // --------add the entry-------
2659                 if (++max >= InternalTableSize)
2660                         max = 0;
2661                 char[] r;
2662                 table[max] = (r = new char[] { c0, c1 });
2663                 newEntry2 = max;
2664                 return r;
2665         }
2666
2667         final char[] optimizedCurrentTokenSource3() {
2668                 // try to return the same char[] build only once
2669                 char c0, c1, c2;
2670                 int hash = (((c0 = source[startPosition]) << 12) + ((c1 = source[startPosition + 1]) << 6) + (c2 = source[startPosition + 2]))
2671                                 % TableSize;
2672                 char[][] table = charArray_length[1][hash];
2673                 int i = newEntry3;
2674                 while (++i < InternalTableSize) {
2675                         char[] charArray = table[i];
2676                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]))
2677                                 return charArray;
2678                 }
2679                 // ---------other side---------
2680                 i = -1;
2681                 int max = newEntry3;
2682                 while (++i <= max) {
2683                         char[] charArray = table[i];
2684                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]))
2685                                 return charArray;
2686                 }
2687                 // --------add the entry-------
2688                 if (++max >= InternalTableSize)
2689                         max = 0;
2690                 char[] r;
2691                 table[max] = (r = new char[] { c0, c1, c2 });
2692                 newEntry3 = max;
2693                 return r;
2694         }
2695
2696         final char[] optimizedCurrentTokenSource4() {
2697                 // try to return the same char[] build only once
2698                 char c0, c1, c2, c3;
2699                 long hash = ((((long) (c0 = source[startPosition])) << 18) + ((c1 = source[startPosition + 1]) << 12)
2700                                 + ((c2 = source[startPosition + 2]) << 6) + (c3 = source[startPosition + 3]))
2701                                 % TableSize;
2702                 char[][] table = charArray_length[2][(int) hash];
2703                 int i = newEntry4;
2704                 while (++i < InternalTableSize) {
2705                         char[] charArray = table[i];
2706                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]))
2707                                 return charArray;
2708                 }
2709                 // ---------other side---------
2710                 i = -1;
2711                 int max = newEntry4;
2712                 while (++i <= max) {
2713                         char[] charArray = table[i];
2714                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]))
2715                                 return charArray;
2716                 }
2717                 // --------add the entry-------
2718                 if (++max >= InternalTableSize)
2719                         max = 0;
2720                 char[] r;
2721                 table[max] = (r = new char[] { c0, c1, c2, c3 });
2722                 newEntry4 = max;
2723                 return r;
2724         }
2725
2726         final char[] optimizedCurrentTokenSource5() {
2727                 // try to return the same char[] build only once
2728                 char c0, c1, c2, c3, c4;
2729                 long hash = ((((long) (c0 = source[startPosition])) << 24) + (((long) (c1 = source[startPosition + 1])) << 18)
2730                                 + ((c2 = source[startPosition + 2]) << 12) + ((c3 = source[startPosition + 3]) << 6) + (c4 = source[startPosition + 4]))
2731                                 % TableSize;
2732                 char[][] table = charArray_length[3][(int) hash];
2733                 int i = newEntry5;
2734                 while (++i < InternalTableSize) {
2735                         char[] charArray = table[i];
2736                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4]))
2737                                 return charArray;
2738                 }
2739                 // ---------other side---------
2740                 i = -1;
2741                 int max = newEntry5;
2742                 while (++i <= max) {
2743                         char[] charArray = table[i];
2744                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4]))
2745                                 return charArray;
2746                 }
2747                 // --------add the entry-------
2748                 if (++max >= InternalTableSize)
2749                         max = 0;
2750                 char[] r;
2751                 table[max] = (r = new char[] { c0, c1, c2, c3, c4 });
2752                 newEntry5 = max;
2753                 return r;
2754         }
2755
2756         final char[] optimizedCurrentTokenSource6() {
2757                 // try to return the same char[] build only once
2758                 char c0, c1, c2, c3, c4, c5;
2759                 long hash = ((((long) (c0 = source[startPosition])) << 32) + (((long) (c1 = source[startPosition + 1])) << 24)
2760                                 + (((long) (c2 = source[startPosition + 2])) << 18) + ((c3 = source[startPosition + 3]) << 12)
2761                                 + ((c4 = source[startPosition + 4]) << 6) + (c5 = source[startPosition + 5]))
2762                                 % TableSize;
2763                 char[][] table = charArray_length[4][(int) hash];
2764                 int i = newEntry6;
2765                 while (++i < InternalTableSize) {
2766                         char[] charArray = table[i];
2767                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4])
2768                                         && (c5 == charArray[5]))
2769                                 return charArray;
2770                 }
2771                 // ---------other side---------
2772                 i = -1;
2773                 int max = newEntry6;
2774                 while (++i <= max) {
2775                         char[] charArray = table[i];
2776                         if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4])
2777                                         && (c5 == charArray[5]))
2778                                 return charArray;
2779                 }
2780                 // --------add the entry-------
2781                 if (++max >= InternalTableSize)
2782                         max = 0;
2783                 char[] r;
2784                 table[max] = (r = new char[] { c0, c1, c2, c3, c4, c5 });
2785                 newEntry6 = max;
2786                 return r;
2787         }
2788
2789         public final void pushLineSeparator() throws InvalidInputException {
2790                 // see comment on isLineDelimiter(char) for the use of '\n' and '\r'
2791                 final int INCREMENT = 250;
2792                 if (this.checkNonExternalizedStringLiterals) {
2793                         // reinitialize the current line for non externalize strings purpose
2794                         currentLine = null;
2795                 }
2796                 // currentCharacter is at position currentPosition-1
2797                 // cr 000D
2798                 if (currentCharacter == '\r') {
2799                         int separatorPos = currentPosition - 1;
2800                         if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2801                                 return;
2802                         // System.out.println("CR-" + separatorPos);
2803                         try {
2804                                 lineEnds[++linePtr] = separatorPos;
2805                         } catch (IndexOutOfBoundsException e) {
2806                                 // linePtr value is correct
2807                                 int oldLength = lineEnds.length;
2808                                 int[] old = lineEnds;
2809                                 lineEnds = new int[oldLength + INCREMENT];
2810                                 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2811                                 lineEnds[linePtr] = separatorPos;
2812                         }
2813                         // look-ahead for merged cr+lf
2814                         try {
2815                                 if (source[currentPosition] == '\n') {
2816                                         // System.out.println("look-ahead LF-" + currentPosition);
2817                                         lineEnds[linePtr] = currentPosition;
2818                                         currentPosition++;
2819                                         wasAcr = false;
2820                                 } else {
2821                                         wasAcr = true;
2822                                 }
2823                         } catch (IndexOutOfBoundsException e) {
2824                                 wasAcr = true;
2825                         }
2826                 } else {
2827                         // lf 000A
2828                         if (currentCharacter == '\n') {
2829                                 // must merge eventual cr followed by lf
2830                                 if (wasAcr && (lineEnds[linePtr] == (currentPosition - 2))) {
2831                                         // System.out.println("merge LF-" + (currentPosition - 1));
2832                                         lineEnds[linePtr] = currentPosition - 1;
2833                                 } else {
2834                                         int separatorPos = currentPosition - 1;
2835                                         if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2836                                                 return;
2837                                         // System.out.println("LF-" + separatorPos);
2838                                         try {
2839                                                 lineEnds[++linePtr] = separatorPos;
2840                                         } catch (IndexOutOfBoundsException e) {
2841                                                 // linePtr value is correct
2842                                                 int oldLength = lineEnds.length;
2843                                                 int[] old = lineEnds;
2844                                                 lineEnds = new int[oldLength + INCREMENT];
2845                                                 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2846                                                 lineEnds[linePtr] = separatorPos;
2847                                         }
2848                                 }
2849                                 wasAcr = false;
2850                         }
2851                 }
2852         }
2853
2854         public final void pushUnicodeLineSeparator() {
2855                 // isUnicode means that the \r or \n has been read as a unicode character
2856                 // see comment on isLineDelimiter(char) for the use of '\n' and '\r'
2857                 final int INCREMENT = 250;
2858                 // currentCharacter is at position currentPosition-1
2859                 if (this.checkNonExternalizedStringLiterals) {
2860                         // reinitialize the current line for non externalize strings purpose
2861                         currentLine = null;
2862                 }
2863                 // cr 000D
2864                 if (currentCharacter == '\r') {
2865                         int separatorPos = currentPosition - 6;
2866                         if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2867                                 return;
2868                         // System.out.println("CR-" + separatorPos);
2869                         try {
2870                                 lineEnds[++linePtr] = separatorPos;
2871                         } catch (IndexOutOfBoundsException e) {
2872                                 // linePtr value is correct
2873                                 int oldLength = lineEnds.length;
2874                                 int[] old = lineEnds;
2875                                 lineEnds = new int[oldLength + INCREMENT];
2876                                 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2877                                 lineEnds[linePtr] = separatorPos;
2878                         }
2879                         // look-ahead for merged cr+lf
2880                         if (source[currentPosition] == '\n') {
2881                                 // System.out.println("look-ahead LF-" + currentPosition);
2882                                 lineEnds[linePtr] = currentPosition;
2883                                 currentPosition++;
2884                                 wasAcr = false;
2885                         } else {
2886                                 wasAcr = true;
2887                         }
2888                 } else {
2889                         // lf 000A
2890                         if (currentCharacter == '\n') {
2891                                 // must merge eventual cr followed by lf
2892                                 if (wasAcr && (lineEnds[linePtr] == (currentPosition - 7))) {
2893                                         // System.out.println("merge LF-" + (currentPosition - 1));
2894                                         lineEnds[linePtr] = currentPosition - 6;
2895                                 } else {
2896                                         int separatorPos = currentPosition - 6;
2897                                         if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2898                                                 return;
2899                                         // System.out.println("LF-" + separatorPos);
2900                                         try {
2901                                                 lineEnds[++linePtr] = separatorPos;
2902                                         } catch (IndexOutOfBoundsException e) {
2903                                                 // linePtr value is correct
2904                                                 int oldLength = lineEnds.length;
2905                                                 int[] old = lineEnds;
2906                                                 lineEnds = new int[oldLength + INCREMENT];
2907                                                 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2908                                                 lineEnds[linePtr] = separatorPos;
2909                                         }
2910                                 }
2911                                 wasAcr = false;
2912                         }
2913                 }
2914         }
2915
2916         public void recordComment(int token) {
2917                 // compute position
2918                 int stopPosition = this.currentPosition;
2919                 switch (token) {
2920                 case TokenNameCOMMENT_LINE:
2921                         stopPosition = -this.lastCommentLinePosition;
2922                         break;
2923                 case TokenNameCOMMENT_BLOCK:
2924                         stopPosition = -this.currentPosition;
2925                         break;
2926                 }
2927
2928                 // a new comment is recorded
2929                 int length = this.commentStops.length;
2930                 if (++this.commentPtr >= length) {
2931                         System.arraycopy(this.commentStops, 0, this.commentStops = new int[length + 30], 0, length);
2932                         // grows the positions buffers too
2933                         System.arraycopy(this.commentStarts, 0, this.commentStarts = new int[length + 30], 0, length);
2934                 }
2935                 this.commentStops[this.commentPtr] = stopPosition;
2936                 this.commentStarts[this.commentPtr] = this.startPosition;
2937         }
2938
2939         // public final void recordComment(boolean isJavadoc) {
2940         // // a new annotation comment is recorded
2941         // try {
2942         // commentStops[++commentPtr] = isJavadoc
2943         // ? currentPosition
2944         // : -currentPosition;
2945         // } catch (IndexOutOfBoundsException e) {
2946         // int oldStackLength = commentStops.length;
2947         // int[] oldStack = commentStops;
2948         // commentStops = new int[oldStackLength + 30];
2949         // System.arraycopy(oldStack, 0, commentStops, 0, oldStackLength);
2950         // commentStops[commentPtr] = isJavadoc ? currentPosition : -currentPosition;
2951         // //grows the positions buffers too
2952         // int[] old = commentStarts;
2953         // commentStarts = new int[oldStackLength + 30];
2954         // System.arraycopy(old, 0, commentStarts, 0, oldStackLength);
2955         // }
2956         // //the buffer is of a correct size here
2957         // commentStarts[commentPtr] = startPosition;
2958         // }
2959         public void resetTo(int begin, int end) {
2960                 // reset the scanner to a given position where it may rescan again
2961                 diet = false;
2962                 initialPosition = startPosition = currentPosition = begin;
2963                 eofPosition = end < Integer.MAX_VALUE ? end + 1 : end;
2964                 commentPtr = -1; // reset comment stack
2965         }
2966
2967         public final void scanSingleQuotedEscapeCharacter() throws InvalidInputException {
2968                 // the string with "\\u" is a legal string of two chars \ and u
2969                 // thus we use a direct access to the source (for regular cases).
2970                 // if (unicodeAsBackSlash) {
2971                 // // consume next character
2972                 // unicodeAsBackSlash = false;
2973                 // if (((currentCharacter = source[currentPosition++]) == '\\')
2974                 // && (source[currentPosition] == 'u')) {
2975                 // getNextUnicodeChar();
2976                 // } else {
2977                 // if (withoutUnicodePtr != 0) {
2978                 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2979                 // }
2980                 // }
2981                 // } else
2982                 currentCharacter = source[currentPosition++];
2983                 switch (currentCharacter) {
2984                 case '\'':
2985                         currentCharacter = '\'';
2986                         break;
2987                 case '\\':
2988                         currentCharacter = '\\';
2989                         break;
2990                 default:
2991                         currentCharacter = '\\';
2992                         currentPosition--;
2993                 }
2994         }
2995
2996         public final void scanDoubleQuotedEscapeCharacter() throws InvalidInputException {
2997                 currentCharacter = source[currentPosition++];
2998                 switch (currentCharacter) {
2999                 // case 'b' :
3000                 // currentCharacter = '\b';
3001                 // break;
3002                 case 't':
3003                         currentCharacter = '\t';
3004                         break;
3005                 case 'n':
3006                         currentCharacter = '\n';
3007                         break;
3008                 // case 'f' :
3009                 // currentCharacter = '\f';
3010                 // break;
3011                 case 'r':
3012                         currentCharacter = '\r';
3013                         break;
3014                 case '\"':
3015                         currentCharacter = '\"';
3016                         break;
3017                 case '\'':
3018                         currentCharacter = '\'';
3019                         break;
3020                 case '\\':
3021                         currentCharacter = '\\';
3022                         break;
3023                 case '$':
3024                         currentCharacter = '$';
3025                         break;
3026                 default:
3027                         // -----------octal escape--------------
3028                         // OctalDigit
3029                         // OctalDigit OctalDigit
3030                         // ZeroToThree OctalDigit OctalDigit
3031                         int number = Character.getNumericValue(currentCharacter);
3032                         if (number >= 0 && number <= 7) {
3033                                 boolean zeroToThreeNot = number > 3;
3034                                 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
3035                                         int digit = Character.getNumericValue(currentCharacter);
3036                                         if (digit >= 0 && digit <= 7) {
3037                                                 number = (number * 8) + digit;
3038                                                 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
3039                                                         if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit
3040                                                                 // Digit --> ignore last character
3041                                                                 currentPosition--;
3042                                                         } else {
3043                                                                 digit = Character.getNumericValue(currentCharacter);
3044                                                                 if (digit >= 0 && digit <= 7) {
3045                                                                         // has read \ZeroToThree OctalDigit OctalDigit
3046                                                                         number = (number * 8) + digit;
3047                                                                 } else { // has read \ZeroToThree OctalDigit NonOctalDigit
3048                                                                         // --> ignore last character
3049                                                                         currentPosition--;
3050                                                                 }
3051                                                         }
3052                                                 } else { // has read \OctalDigit NonDigit--> ignore last
3053                                                         // character
3054                                                         currentPosition--;
3055                                                 }
3056                                         } else { // has read \OctalDigit NonOctalDigit--> ignore last
3057                                                 // character
3058                                                 currentPosition--;
3059                                         }
3060                                 } else { // has read \OctalDigit --> ignore last character
3061                                         currentPosition--;
3062                                 }
3063                                 if (number > 255)
3064                                         throw new InvalidInputException(INVALID_ESCAPE);
3065                                 currentCharacter = (char) number;
3066                         }
3067                 // else
3068                 // throw new InvalidInputException(INVALID_ESCAPE);
3069                 }
3070         }
3071
3072         // public int scanIdentifierOrKeyword() throws InvalidInputException {
3073         // return scanIdentifierOrKeyword( false );
3074         // }
3075         public int scanIdentifierOrKeyword(boolean isVariable) throws InvalidInputException {
3076                 // test keywords
3077                 // first dispatch on the first char.
3078                 // then the length. If there are several
3079                 // keywors with the same length AND the same first char, then do another
3080                 // disptach on the second char :-)...cool....but fast !
3081                 useAssertAsAnIndentifier = false;
3082                 while (getNextCharAsJavaIdentifierPart()) {
3083                 }
3084                 ;
3085                 if (isVariable) {
3086                         // if (new String(getCurrentTokenSource()).equals("$this")) {
3087                         // return TokenNamethis;
3088                         // }
3089                         return TokenNameVariable;
3090                 }
3091                 int index, length;
3092                 char[] data;
3093                 char firstLetter;
3094                 // if (withoutUnicodePtr == 0)
3095                 // quick test on length == 1 but not on length > 12 while most identifier
3096                 // have a length which is <= 12...but there are lots of identifier with
3097                 // only one char....
3098                 // {
3099                 if ((length = currentPosition - startPosition) == 1)
3100                         return TokenNameIdentifier;
3101                 // data = source;
3102                 data = new char[length];
3103                 index = startPosition;
3104                 for (int i = 0; i < length; i++) {
3105                         data[i] = Character.toLowerCase(source[index + i]);
3106                 }
3107                 index = 0;
3108                 // } else {
3109                 // if ((length = withoutUnicodePtr) == 1)
3110                 // return TokenNameIdentifier;
3111                 // // data = withoutUnicodeBuffer;
3112                 // data = new char[withoutUnicodeBuffer.length];
3113                 // for (int i = 0; i < withoutUnicodeBuffer.length; i++) {
3114                 // data[i] = Character.toLowerCase(withoutUnicodeBuffer[i]);
3115                 // }
3116                 // index = 1;
3117                 // }
3118                 firstLetter = data[index];
3119                 switch (firstLetter) {
3120                 case '_':
3121                         switch (length) {
3122                         case 8:
3123                                 // __FILE__
3124                                 if ((data[++index] == '_') && (data[++index] == 'f') && (data[++index] == 'i') && (data[++index] == 'l')
3125                                                 && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == '_'))
3126                                         return TokenNameFILE;
3127                                 index = 0; // __LINE__
3128                                 if ((data[++index] == '_') && (data[++index] == 'l') && (data[++index] == 'i') && (data[++index] == 'n')
3129                                                 && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == '_'))
3130                                         return TokenNameLINE;
3131                                 break;
3132                         case 9:
3133                                 // __CLASS__
3134                                 if ((data[++index] == '_') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'a')
3135                                                 && (data[++index] == 's') && (data[++index] == 's') && (data[++index] == '_') && (data[++index] == '_'))
3136                                         return TokenNameCLASS_C;
3137                                 break;
3138                         case 11:
3139                                 // __METHOD__
3140                                 if ((data[++index] == '_') && (data[++index] == 'm') && (data[++index] == 'e') && (data[++index] == 't')
3141                                                 && (data[++index] == 'h') && (data[++index] == 'o') && (data[++index] == 'd') && (data[++index] == '_')
3142                                                 && (data[++index] == '_'))
3143                                         return TokenNameMETHOD_C;
3144                                 break;
3145                         case 12:
3146                                 // __FUNCTION__
3147                                 if ((data[++index] == '_') && (data[++index] == 'f') && (data[++index] == 'u') && (data[++index] == 'n')
3148                                                 && (data[++index] == 'c') && (data[++index] == 't') && (data[++index] == 'i') && (data[++index] == 'o')
3149                                                 && (data[++index] == 'n') && (data[++index] == '_') && (data[++index] == '_'))
3150                                         return TokenNameFUNC_C;
3151                                 break;
3152                         }
3153                         return TokenNameIdentifier;
3154                 case 'a':
3155                         // as and array abstract
3156                         switch (length) {
3157                         case 2:
3158                                 // as
3159                                 if ((data[++index] == 's')) {
3160                                         return TokenNameas;
3161                                 } else {
3162                                         return TokenNameIdentifier;
3163                                 }
3164                         case 3:
3165                                 // and
3166                                 if ((data[++index] == 'n') && (data[++index] == 'd')) {
3167                                         return TokenNameand;
3168                                 } else {
3169                                         return TokenNameIdentifier;
3170                                 }
3171                         case 5:
3172                                 // array
3173                                 if ((data[++index] == 'r') && (data[++index] == 'r') && (data[++index] == 'a') && (data[++index] == 'y'))
3174                                         return TokenNamearray;
3175                                 else
3176                                         return TokenNameIdentifier;
3177                         case 8:
3178                                 if ((data[++index] == 'b') && (data[++index] == 's') && (data[++index] == 't') && (data[++index] == 'r')
3179                                                 && (data[++index] == 'a') && (data[++index] == 'c') && (data[++index] == 't'))
3180                                         return TokenNameabstract;
3181                                 else
3182                                         return TokenNameIdentifier;
3183                         default:
3184                                 return TokenNameIdentifier;
3185                         }
3186                 case 'b':
3187                         // break
3188                         switch (length) {
3189                         case 5:
3190                                 if ((data[++index] == 'r') && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'k'))
3191                                         return TokenNamebreak;
3192                                 else
3193                                         return TokenNameIdentifier;
3194                         default:
3195                                 return TokenNameIdentifier;
3196                         }
3197                 case 'c':
3198                         // case catch class clone const continue
3199                         switch (length) {
3200                         case 4:
3201                                 if ((data[++index] == 'a') && (data[++index] == 's') && (data[++index] == 'e'))
3202                                         return TokenNamecase;
3203                                 else
3204                                         return TokenNameIdentifier;
3205                         case 5:
3206                                 if ((data[++index] == 'a') && (data[++index] == 't') && (data[++index] == 'c') && (data[++index] == 'h'))
3207                                         return TokenNamecatch;
3208                                 index = 0;
3209                                 if ((data[++index] == 'l') && (data[++index] == 'a') && (data[++index] == 's') && (data[++index] == 's'))
3210                                         return TokenNameclass;
3211                                 index = 0;
3212                                 if ((data[++index] == 'l') && (data[++index] == 'o') && (data[++index] == 'n') && (data[++index] == 'e'))
3213                                         return TokenNameclone;
3214                                 index = 0;
3215                                 if ((data[++index] == 'o') && (data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 't'))
3216                                         return TokenNameconst;
3217                                 else
3218                                         return TokenNameIdentifier;
3219                         case 8:
3220                                 if ((data[++index] == 'o') && (data[++index] == 'n') && (data[++index] == 't') && (data[++index] == 'i')
3221                                                 && (data[++index] == 'n') && (data[++index] == 'u') && (data[++index] == 'e'))
3222                                         return TokenNamecontinue;
3223                                 else
3224                                         return TokenNameIdentifier;
3225                         default:
3226                                 return TokenNameIdentifier;
3227                         }
3228                 case 'd':
3229                         // declare default do die
3230                         // TODO delete define ==> no keyword !
3231                         switch (length) {
3232                         case 2:
3233                                 if ((data[++index] == 'o'))
3234                                         return TokenNamedo;
3235                                 else
3236                                         return TokenNameIdentifier;
3237                         // case 6 :
3238                         // if ((data[++index] == 'e')
3239                         // && (data[++index] == 'f')
3240                         // && (data[++index] == 'i')
3241                         // && (data[++index] == 'n')
3242                         // && (data[++index] == 'e'))
3243                         // return TokenNamedefine;
3244                         // else
3245                         // return TokenNameIdentifier;
3246                         case 7:
3247                                 if ((data[++index] == 'e') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'a')
3248                                                 && (data[++index] == 'r') && (data[++index] == 'e'))
3249                                         return TokenNamedeclare;
3250                                 index = 0;
3251                                 if ((data[++index] == 'e') && (data[++index] == 'f') && (data[++index] == 'a') && (data[++index] == 'u')
3252                                                 && (data[++index] == 'l') && (data[++index] == 't'))
3253                                         return TokenNamedefault;
3254                                 else
3255                                         return TokenNameIdentifier;
3256                         default:
3257                                 return TokenNameIdentifier;
3258                         }
3259                 case 'e':
3260                         // echo else exit elseif extends eval
3261                         switch (length) {
3262                         case 4:
3263                                 if ((data[++index] == 'c') && (data[++index] == 'h') && (data[++index] == 'o'))
3264                                         return TokenNameecho;
3265                                 else if ((data[index] == 'l') && (data[++index] == 's') && (data[++index] == 'e'))
3266                                         return TokenNameelse;
3267                                 else if ((data[index] == 'x') && (data[++index] == 'i') && (data[++index] == 't'))
3268                                         return TokenNameexit;
3269                                 else if ((data[index] == 'v') && (data[++index] == 'a') && (data[++index] == 'l'))
3270                                         return TokenNameeval;
3271                                 else
3272                                         return TokenNameIdentifier;
3273                         case 5:
3274                                 // endif empty
3275                                 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'i') && (data[++index] == 'f'))
3276                                         return TokenNameendif;
3277                                 if ((data[index] == 'm') && (data[++index] == 'p') && (data[++index] == 't') && (data[++index] == 'y'))
3278                                         return TokenNameempty;
3279                                 else
3280                                         return TokenNameIdentifier;
3281                         case 6:
3282                                 // endfor
3283                                 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'f') && (data[++index] == 'o')
3284                                                 && (data[++index] == 'r'))
3285                                         return TokenNameendfor;
3286                                 else if ((data[index] == 'l') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 'i')
3287                                                 && (data[++index] == 'f'))
3288                                         return TokenNameelseif;
3289                                 else
3290                                         return TokenNameIdentifier;
3291                         case 7:
3292                                 if ((data[++index] == 'x') && (data[++index] == 't') && (data[++index] == 'e') && (data[++index] == 'n')
3293                                                 && (data[++index] == 'd') && (data[++index] == 's'))
3294                                         return TokenNameextends;
3295                                 else
3296                                         return TokenNameIdentifier;
3297                         case 8:
3298                                 // endwhile
3299                                 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'w') && (data[++index] == 'h')
3300                                                 && (data[++index] == 'i') && (data[++index] == 'l') && (data[++index] == 'e'))
3301                                         return TokenNameendwhile;
3302                                 else
3303                                         return TokenNameIdentifier;
3304                         case 9:
3305                                 // endswitch
3306                                 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 's') && (data[++index] == 'w')
3307                                                 && (data[++index] == 'i') && (data[++index] == 't') && (data[++index] == 'c') && (data[++index] == 'h'))
3308                                         return TokenNameendswitch;
3309                                 else
3310                                         return TokenNameIdentifier;
3311                         case 10:
3312                                 // enddeclare
3313                                 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'd') && (data[++index] == 'e')
3314                                                 && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'a') && (data[++index] == 'r')
3315                                                 && (data[++index] == 'e'))
3316                                         return TokenNameenddeclare;
3317                                 index = 0;
3318                                 if ((data[++index] == 'n') // endforeach
3319                                                 && (data[++index] == 'd') && (data[++index] == 'f') && (data[++index] == 'o') && (data[++index] == 'r')
3320                                                 && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'c') && (data[++index] == 'h'))
3321                                         return TokenNameendforeach;
3322                                 else
3323                                         return TokenNameIdentifier;
3324                         default:
3325                                 return TokenNameIdentifier;
3326                         }
3327                 case 'f':
3328                         // for false final function
3329                         switch (length) {
3330                         case 3:
3331                                 if ((data[++index] == 'o') && (data[++index] == 'r'))
3332                                         return TokenNamefor;
3333                                 else
3334                                         return TokenNameIdentifier;
3335                         case 5:
3336                                 // if ((data[++index] == 'a') && (data[++index] == 'l')
3337                                 // && (data[++index] == 's') && (data[++index] == 'e'))
3338                                 // return TokenNamefalse;
3339                                 if ((data[++index] == 'i') && (data[++index] == 'n') && (data[++index] == 'a') && (data[++index] == 'l'))
3340                                         return TokenNamefinal;
3341                                 else
3342                                         return TokenNameIdentifier;
3343                         case 7:
3344                                 // foreach
3345                                 if ((data[++index] == 'o') && (data[++index] == 'r') && (data[++index] == 'e') && (data[++index] == 'a')
3346                                                 && (data[++index] == 'c') && (data[++index] == 'h'))
3347                                         return TokenNameforeach;
3348                                 else
3349                                         return TokenNameIdentifier;
3350                         case 8:
3351                                 // function
3352                                 if ((data[++index] == 'u') && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 't')
3353                                                 && (data[++index] == 'i') && (data[++index] == 'o') && (data[++index] == 'n'))
3354                                         return TokenNamefunction;
3355                                 else
3356                                         return TokenNameIdentifier;
3357                         default:
3358                                 return TokenNameIdentifier;
3359                         }
3360                 case 'g':
3361                         // global
3362                         if (length == 6) {
3363                                 if ((data[++index] == 'l') && (data[++index] == 'o') && (data[++index] == 'b') && (data[++index] == 'a')
3364                                                 && (data[++index] == 'l')) {
3365                                         return TokenNameglobal;
3366                                 }
3367                         }
3368                         return TokenNameIdentifier;
3369                 case 'i':
3370                         // if int isset include include_once instanceof interface implements
3371                         switch (length) {
3372                         case 2:
3373                                 if (data[++index] == 'f')
3374                                         return TokenNameif;
3375                                 else
3376                                         return TokenNameIdentifier;
3377                         // case 3 :
3378                         // if ((data[++index] == 'n') && (data[++index] == 't'))
3379                         // return TokenNameint;
3380                         // else
3381                         // return TokenNameIdentifier;
3382                         case 5:
3383                                 if ((data[++index] == 's') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 't'))
3384                                         return TokenNameisset;
3385                                 else
3386                                         return TokenNameIdentifier;
3387                         case 7:
3388                                 if ((data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'u')
3389                                                 && (data[++index] == 'd') && (data[++index] == 'e'))
3390                                         return TokenNameinclude;
3391                                 else
3392                                         return TokenNameIdentifier;
3393                         case 9:
3394                                 // interface
3395                                 if ((data[++index] == 'n') && (data[++index] == 't') && (data[++index] == 'e') && (data[++index] == 'r')
3396                                                 && (data[++index] == 'f') && (data[++index] == 'a') && (data[++index] == 'c') && (data[++index] == 'e'))
3397                                         return TokenNameinterface;
3398                                 else
3399                                         return TokenNameIdentifier;
3400                         case 10:
3401                                 // instanceof
3402                                 if ((data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 't') && (data[++index] == 'a')
3403                                                 && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'e') && (data[++index] == 'o')
3404                                                 && (data[++index] == 'f'))
3405                                         return TokenNameinstanceof;
3406                                 if ((data[index] == 'm') && (data[++index] == 'p') && (data[++index] == 'l') && (data[++index] == 'e')
3407                                                 && (data[++index] == 'm') && (data[++index] == 'e') && (data[++index] == 'n') && (data[++index] == 't')
3408                                                 && (data[++index] == 's'))
3409                                         return TokenNameimplements;
3410                                 else
3411                                         return TokenNameIdentifier;
3412                         case 12:
3413                                 if ((data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'u')
3414                                                 && (data[++index] == 'd') && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == 'o')
3415                                                 && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'e'))
3416                                         return TokenNameinclude_once;
3417                                 else
3418                                         return TokenNameIdentifier;
3419                         default:
3420                                 return TokenNameIdentifier;
3421                         }
3422                 case 'l':
3423                         // list
3424                         if (length == 4) {
3425                                 if ((data[++index] == 'i') && (data[++index] == 's') && (data[++index] == 't')) {
3426                                         return TokenNamelist;
3427                                 }
3428                         }
3429                         return TokenNameIdentifier;
3430                 case 'n':
3431                         // new null
3432                         switch (length) {
3433                         case 3:
3434                                 if ((data[++index] == 'e') && (data[++index] == 'w'))
3435                                         return TokenNamenew;
3436                                 else
3437                                         return TokenNameIdentifier;
3438                         // case 4 :
3439                         // if ((data[++index] == 'u') && (data[++index] == 'l')
3440                         // && (data[++index] == 'l'))
3441                         // return TokenNamenull;
3442                         // else
3443                         // return TokenNameIdentifier;
3444                         default:
3445                                 return TokenNameIdentifier;
3446                         }
3447                 case 'o':
3448                         // or old_function
3449                         if (length == 2) {
3450                                 if (data[++index] == 'r') {
3451                                         return TokenNameor;
3452                                 }
3453                         }
3454                         // if (length == 12) {
3455                         // if ((data[++index] == 'l')
3456                         // && (data[++index] == 'd')
3457                         // && (data[++index] == '_')
3458                         // && (data[++index] == 'f')
3459                         // && (data[++index] == 'u')
3460                         // && (data[++index] == 'n')
3461                         // && (data[++index] == 'c')
3462                         // && (data[++index] == 't')
3463                         // && (data[++index] == 'i')
3464                         // && (data[++index] == 'o')
3465                         // && (data[++index] == 'n')) {
3466                         // return TokenNameold_function;
3467                         // }
3468                         // }
3469                         return TokenNameIdentifier;
3470                 case 'p':
3471                         // print public private protected
3472                         switch (length) {
3473                         case 5:
3474                                 if ((data[++index] == 'r') && (data[++index] == 'i') && (data[++index] == 'n') && (data[++index] == 't')) {
3475                                         return TokenNameprint;
3476                                 } else
3477                                         return TokenNameIdentifier;
3478                         case 6:
3479                                 if ((data[++index] == 'u') && (data[++index] == 'b') && (data[++index] == 'l') && (data[++index] == 'i')
3480                                                 && (data[++index] == 'c')) {
3481                                         return TokenNamepublic;
3482                                 } else
3483                                         return TokenNameIdentifier;
3484                         case 7:
3485                                 if ((data[++index] == 'r') && (data[++index] == 'i') && (data[++index] == 'v') && (data[++index] == 'a')
3486                                                 && (data[++index] == 't') && (data[++index] == 'e')) {
3487                                         return TokenNameprivate;
3488                                 } else
3489                                         return TokenNameIdentifier;
3490                         case 9:
3491                                 if ((data[++index] == 'r') && (data[++index] == 'o') && (data[++index] == 't') && (data[++index] == 'e')
3492                                                 && (data[++index] == 'c') && (data[++index] == 't') && (data[++index] == 'e') && (data[++index] == 'd')) {
3493                                         return TokenNameprotected;
3494                                 } else
3495                                         return TokenNameIdentifier;
3496                         }
3497                         return TokenNameIdentifier;
3498                 case 'r':
3499                         // return require require_once
3500                         if (length == 6) {
3501                                 if ((data[++index] == 'e') && (data[++index] == 't') && (data[++index] == 'u') && (data[++index] == 'r')
3502                                                 && (data[++index] == 'n')) {
3503                                         return TokenNamereturn;
3504                                 }
3505                         } else if (length == 7) {
3506                                 if ((data[++index] == 'e') && (data[++index] == 'q') && (data[++index] == 'u') && (data[++index] == 'i')
3507                                                 && (data[++index] == 'r') && (data[++index] == 'e')) {
3508                                         return TokenNamerequire;
3509                                 }
3510                         } else if (length == 12) {
3511                                 if ((data[++index] == 'e') && (data[++index] == 'q') && (data[++index] == 'u') && (data[++index] == 'i')
3512                                                 && (data[++index] == 'r') && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == 'o')
3513                                                 && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'e')) {
3514                                         return TokenNamerequire_once;
3515                                 }
3516                         } else
3517                                 return TokenNameIdentifier;
3518                 case 's':
3519                         // static switch
3520                         switch (length) {
3521                         case 6:
3522                                 if (data[++index] == 't')
3523                                         if ((data[++index] == 'a') && (data[++index] == 't') && (data[++index] == 'i') && (data[++index] == 'c')) {
3524                                                 return TokenNamestatic;
3525                                         } else
3526                                                 return TokenNameIdentifier;
3527                                 else if ((data[index] == 'w') && (data[++index] == 'i') && (data[++index] == 't') && (data[++index] == 'c')
3528                                                 && (data[++index] == 'h'))
3529                                         return TokenNameswitch;
3530                                 else
3531                                         return TokenNameIdentifier;
3532                         default:
3533                                 return TokenNameIdentifier;
3534                         }
3535                 case 't':
3536                         // try true throw
3537                         switch (length) {
3538                         case 3:
3539                                 if ((data[++index] == 'r') && (data[++index] == 'y'))
3540                                         return TokenNametry;
3541                                 else
3542                                         return TokenNameIdentifier;
3543                         // case 4 :
3544                         // if ((data[++index] == 'r') && (data[++index] == 'u')
3545                         // && (data[++index] == 'e'))
3546                         // return TokenNametrue;
3547                         // else
3548                         // return TokenNameIdentifier;
3549                         case 5:
3550                                 if ((data[++index] == 'h') && (data[++index] == 'r') && (data[++index] == 'o') && (data[++index] == 'w'))
3551                                         return TokenNamethrow;
3552                                 else
3553                                         return TokenNameIdentifier;
3554                         default:
3555                                 return TokenNameIdentifier;
3556                         }
3557                 case 'u':
3558                         // use unset
3559                         switch (length) {
3560                         case 3:
3561                                 if ((data[++index] == 's') && (data[++index] == 'e'))
3562                                         return TokenNameuse;
3563                                 else
3564                                         return TokenNameIdentifier;
3565                         case 5:
3566                                 if ((data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 't'))
3567                                         return TokenNameunset;
3568                                 else
3569                                         return TokenNameIdentifier;
3570                         default:
3571                                 return TokenNameIdentifier;
3572                         }
3573                 case 'v':
3574                         // var
3575                         switch (length) {
3576                         case 3:
3577                                 if ((data[++index] == 'a') && (data[++index] == 'r'))
3578                                         return TokenNamevar;
3579                                 else
3580                                         return TokenNameIdentifier;
3581                         default:
3582                                 return TokenNameIdentifier;
3583                         }
3584                 case 'w':
3585                         // while
3586                         switch (length) {
3587                         case 5:
3588                                 if ((data[++index] == 'h') && (data[++index] == 'i') && (data[++index] == 'l') && (data[++index] == 'e'))
3589                                         return TokenNamewhile;
3590                                 else
3591                                         return TokenNameIdentifier;
3592                         // case 6:if ( (data[++index] =='i') && (data[++index]=='d') &&
3593                         // (data[++index]=='e') && (data[++index]=='f')&&
3594                         // (data[++index]=='p'))
3595                         // return TokenNamewidefp ;
3596                         // else
3597                         // return TokenNameIdentifier;
3598                         default:
3599                                 return TokenNameIdentifier;
3600                         }
3601                 case 'x':
3602                         // xor
3603                         switch (length) {
3604                         case 3:
3605                                 if ((data[++index] == 'o') && (data[++index] == 'r'))
3606                                         return TokenNamexor;
3607                                 else
3608                                         return TokenNameIdentifier;
3609                         default:
3610                                 return TokenNameIdentifier;
3611                         }
3612                 default:
3613                         return TokenNameIdentifier;
3614                 }
3615         }
3616
3617         public int scanNumber(boolean dotPrefix) throws InvalidInputException {
3618                 // when entering this method the currentCharacter is the firt
3619                 // digit of the number , i.e. it may be preceeded by a . when
3620                 // dotPrefix is true
3621                 boolean floating = dotPrefix;
3622                 if ((!dotPrefix) && (currentCharacter == '0')) {
3623                         if (getNextChar('x', 'X') >= 0) { // ----------hexa-----------------
3624                                 // force the first char of the hexa number do exist...
3625                                 // consume next character
3626                                 unicodeAsBackSlash = false;
3627                                 currentCharacter = source[currentPosition++];
3628                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
3629                                 // && (source[currentPosition] == 'u')) {
3630                                 // getNextUnicodeChar();
3631                                 // } else {
3632                                 // if (withoutUnicodePtr != 0) {
3633                                 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3634                                 // }
3635                                 // }
3636                                 if (Character.digit(currentCharacter, 16) == -1)
3637                                         throw new InvalidInputException(INVALID_HEXA);
3638                                 // ---end forcing--
3639                                 while (getNextCharAsDigit(16)) {
3640                                 }
3641                                 ;
3642                                 // if (getNextChar('l', 'L') >= 0)
3643                                 // return TokenNameLongLiteral;
3644                                 // else
3645                                 return TokenNameIntegerLiteral;
3646                         }
3647                         // there is x or X in the number
3648                         // potential octal ! ... some one may write 000099.0 ! thus 00100 <
3649                         // 00078.0 is true !!!!! crazy language
3650                         if (getNextCharAsDigit()) {
3651                                 // -------------potential octal-----------------
3652                                 while (getNextCharAsDigit()) {
3653                                 }
3654                                 ;
3655                                 // if (getNextChar('l', 'L') >= 0) {
3656                                 // return TokenNameLongLiteral;
3657                                 // }
3658                                 //
3659                                 // if (getNextChar('f', 'F') >= 0) {
3660                                 // return TokenNameFloatingPointLiteral;
3661                                 // }
3662                                 if (getNextChar('d', 'D') >= 0) {
3663                                         return TokenNameDoubleLiteral;
3664                                 } else { // make the distinction between octal and float ....
3665                                         if (getNextChar('.')) { // bingo ! ....
3666                                                 while (getNextCharAsDigit()) {
3667                                                 }
3668                                                 ;
3669                                                 if (getNextChar('e', 'E') >= 0) {
3670                                                         // consume next character
3671                                                         unicodeAsBackSlash = false;
3672                                                         currentCharacter = source[currentPosition++];
3673                                                         // if (((currentCharacter = source[currentPosition++]) == '\\')
3674                                                         // && (source[currentPosition] == 'u')) {
3675                                                         // getNextUnicodeChar();
3676                                                         // } else {
3677                                                         // if (withoutUnicodePtr != 0) {
3678                                                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3679                                                         // }
3680                                                         // }
3681                                                         if ((currentCharacter == '-') || (currentCharacter == '+')) {
3682                                                                 // consume next character
3683                                                                 unicodeAsBackSlash = false;
3684                                                                 currentCharacter = source[currentPosition++];
3685                                                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
3686                                                                 // && (source[currentPosition] == 'u')) {
3687                                                                 // getNextUnicodeChar();
3688                                                                 // } else {
3689                                                                 // if (withoutUnicodePtr != 0) {
3690                                                                 // withoutUnicodeBuffer[++withoutUnicodePtr] =
3691                                                                 // currentCharacter;
3692                                                                 // }
3693                                                                 // }
3694                                                         }
3695                                                         if (!Character.isDigit(currentCharacter))
3696                                                                 throw new InvalidInputException(INVALID_FLOAT);
3697                                                         while (getNextCharAsDigit()) {
3698                                                         }
3699                                                         ;
3700                                                 }
3701                                                 // if (getNextChar('f', 'F') >= 0)
3702                                                 // return TokenNameFloatingPointLiteral;
3703                                                 getNextChar('d', 'D'); // jump over potential d or D
3704                                                 return TokenNameDoubleLiteral;
3705                                         } else {
3706                                                 return TokenNameIntegerLiteral;
3707                                         }
3708                                 }
3709                         } else {
3710                                 /* carry on */
3711                         }
3712                 }
3713                 while (getNextCharAsDigit()) {
3714                 }
3715                 ;
3716                 // if ((!dotPrefix) && (getNextChar('l', 'L') >= 0))
3717                 // return TokenNameLongLiteral;
3718                 if ((!dotPrefix) && (getNextChar('.'))) { // decimal part that can be empty
3719                         while (getNextCharAsDigit()) {
3720                         }
3721                         ;
3722                         floating = true;
3723                 }
3724                 // if floating is true both exponant and suffix may be optional
3725                 if (getNextChar('e', 'E') >= 0) {
3726                         floating = true;
3727                         // consume next character
3728                         unicodeAsBackSlash = false;
3729                         currentCharacter = source[currentPosition++];
3730                         // if (((currentCharacter = source[currentPosition++]) == '\\')
3731                         // && (source[currentPosition] == 'u')) {
3732                         // getNextUnicodeChar();
3733                         // } else {
3734                         // if (withoutUnicodePtr != 0) {
3735                         // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3736                         // }
3737                         // }
3738                         if ((currentCharacter == '-') || (currentCharacter == '+')) { // consume
3739                                 // next
3740                                 // character
3741                                 unicodeAsBackSlash = false;
3742                                 currentCharacter = source[currentPosition++];
3743                                 // if (((currentCharacter = source[currentPosition++]) == '\\')
3744                                 // && (source[currentPosition] == 'u')) {
3745                                 // getNextUnicodeChar();
3746                                 // } else {
3747                                 // if (withoutUnicodePtr != 0) {
3748                                 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3749                                 // }
3750                                 // }
3751                         }
3752                         if (!Character.isDigit(currentCharacter))
3753                                 throw new InvalidInputException(INVALID_FLOAT);
3754                         while (getNextCharAsDigit()) {
3755                         }
3756                         ;
3757                 }
3758                 if (getNextChar('d', 'D') >= 0)
3759                         return TokenNameDoubleLiteral;
3760                 // if (getNextChar('f', 'F') >= 0)
3761                 // return TokenNameFloatingPointLiteral;
3762                 // the long flag has been tested before
3763                 return floating ? TokenNameDoubleLiteral : TokenNameIntegerLiteral;
3764         }
3765
3766         /**
3767          * Search the line number corresponding to a specific position
3768          *
3769          */
3770         public final int getLineNumber(int position) {
3771                 if (lineEnds == null)
3772                         return 1;
3773                 int length = linePtr + 1;
3774                 if (length == 0)
3775                         return 1;
3776                 int g = 0, d = length - 1;
3777                 int m = 0;
3778                 while (g <= d) {
3779                         m = (g + d) / 2;
3780                         if (position < lineEnds[m]) {
3781                                 d = m - 1;
3782                         } else if (position > lineEnds[m]) {
3783                                 g = m + 1;
3784                         } else {
3785                                 return m + 1;
3786                         }
3787                 }
3788                 if (position < lineEnds[m]) {
3789                         return m + 1;
3790                 }
3791                 return m + 2;
3792         }
3793
3794         public void setPHPMode(boolean mode) {
3795                 phpMode = mode;
3796         }
3797
3798         public final void setSource(char[] source) {
3799                 setSource(null, source);
3800         }
3801
3802         public final void setSource(ICompilationUnit compilationUnit, char[] source) {
3803                 // the source-buffer is set to sourceString
3804                 this.compilationUnit = compilationUnit;
3805                 if (source == null) {
3806                         this.source = new char[0];
3807                 } else {
3808                         this.source = source;
3809                 }
3810                 startPosition = -1;
3811                 initialPosition = currentPosition = 0;
3812                 containsAssertKeyword = false;
3813                 withoutUnicodeBuffer = new char[this.source.length];
3814                 // encapsedStringStack = new Stack();
3815         }
3816
3817         public String toString() {
3818                 if (startPosition == source.length)
3819                         return "EOF\n\n" + new String(source); //$NON-NLS-1$
3820                 if (currentPosition > source.length)
3821                         return "behind the EOF :-( ....\n\n" + new String(source); //$NON-NLS-1$
3822                 char front[] = new char[startPosition];
3823                 System.arraycopy(source, 0, front, 0, startPosition);
3824                 int middleLength = (currentPosition - 1) - startPosition + 1;
3825                 char middle[];
3826                 if (middleLength > -1) {
3827                         middle = new char[middleLength];
3828                         System.arraycopy(source, startPosition, middle, 0, middleLength);
3829                 } else {
3830                         middle = new char[0];
3831                 }
3832                 char end[] = new char[source.length - (currentPosition - 1)];
3833                 System.arraycopy(source, (currentPosition - 1) + 1, end, 0, source.length - (currentPosition - 1) - 1);
3834                 return new String(front) + "\n===============================\nStarts here -->" //$NON-NLS-1$
3835                                 + new String(middle) + "<-- Ends here\n===============================\n" //$NON-NLS-1$
3836                                 + new String(end);
3837         }
3838
3839         public final String toStringAction(int act) {
3840                 switch (act) {
3841                 case TokenNameERROR:
3842                         return "ScannerError"; // + new String(getCurrentTokenSource()) + ")";
3843                 // //$NON-NLS-1$
3844                 case TokenNameINLINE_HTML:
3845                         return "Inline-HTML(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3846                 case TokenNameIdentifier:
3847                         return "Identifier(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3848                 case TokenNameVariable:
3849                         return "Variable(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3850                 case TokenNameabstract:
3851                         return "abstract"; //$NON-NLS-1$
3852                 case TokenNameand:
3853                         return "AND"; //$NON-NLS-1$
3854                 case TokenNamearray:
3855                         return "array"; //$NON-NLS-1$
3856                 case TokenNameas:
3857                         return "as"; //$NON-NLS-1$
3858                 case TokenNamebreak:
3859                         return "break"; //$NON-NLS-1$
3860                 case TokenNamecase:
3861                         return "case"; //$NON-NLS-1$
3862                 case TokenNameclass:
3863                         return "class"; //$NON-NLS-1$
3864                 case TokenNamecatch:
3865                         return "catch"; //$NON-NLS-1$
3866                 case TokenNameclone:
3867                         //$NON-NLS-1$
3868                         return "clone";
3869                 case TokenNameconst:
3870                         //$NON-NLS-1$
3871                         return "const";
3872                 case TokenNamecontinue:
3873                         return "continue"; //$NON-NLS-1$
3874                 case TokenNamedefault:
3875                         return "default"; //$NON-NLS-1$
3876                 // case TokenNamedefine :
3877                 // return "define"; //$NON-NLS-1$
3878                 case TokenNamedo:
3879                         return "do"; //$NON-NLS-1$
3880                 case TokenNameecho:
3881                         return "echo"; //$NON-NLS-1$
3882                 case TokenNameelse:
3883                         return "else"; //$NON-NLS-1$
3884                 case TokenNameelseif:
3885                         return "elseif"; //$NON-NLS-1$
3886                 case TokenNameendfor:
3887                         return "endfor"; //$NON-NLS-1$
3888                 case TokenNameendforeach:
3889                         return "endforeach"; //$NON-NLS-1$
3890                 case TokenNameendif:
3891                         return "endif"; //$NON-NLS-1$
3892                 case TokenNameendswitch:
3893                         return "endswitch"; //$NON-NLS-1$
3894                 case TokenNameendwhile:
3895                         return "endwhile"; //$NON-NLS-1$
3896                 case TokenNameexit:
3897                         return "exit";
3898                 case TokenNameextends:
3899                         return "extends"; //$NON-NLS-1$
3900                 // case TokenNamefalse :
3901                 // return "false"; //$NON-NLS-1$
3902                 case TokenNamefinal:
3903                         return "final"; //$NON-NLS-1$
3904                 case TokenNamefor:
3905                         return "for"; //$NON-NLS-1$
3906                 case TokenNameforeach:
3907                         return "foreach"; //$NON-NLS-1$
3908                 case TokenNamefunction:
3909                         return "function"; //$NON-NLS-1$
3910                 case TokenNameglobal:
3911                         return "global"; //$NON-NLS-1$
3912                 case TokenNameif:
3913                         return "if"; //$NON-NLS-1$
3914                 case TokenNameimplements:
3915                         return "implements"; //$NON-NLS-1$
3916                 case TokenNameinclude:
3917                         return "include"; //$NON-NLS-1$
3918                 case TokenNameinclude_once:
3919                         return "include_once"; //$NON-NLS-1$
3920                 case TokenNameinstanceof:
3921                         return "instanceof"; //$NON-NLS-1$
3922                 case TokenNameinterface:
3923                         return "interface"; //$NON-NLS-1$
3924                 case TokenNameisset:
3925                         return "isset"; //$NON-NLS-1$
3926                 case TokenNamelist:
3927                         return "list"; //$NON-NLS-1$
3928                 case TokenNamenew:
3929                         return "new"; //$NON-NLS-1$
3930                 // case TokenNamenull :
3931                 // return "null"; //$NON-NLS-1$
3932                 case TokenNameor:
3933                         return "OR"; //$NON-NLS-1$
3934                 case TokenNameprint:
3935                         return "print"; //$NON-NLS-1$
3936                 case TokenNameprivate:
3937                         return "private"; //$NON-NLS-1$
3938                 case TokenNameprotected:
3939                         return "protected"; //$NON-NLS-1$
3940                 case TokenNamepublic:
3941                         return "public"; //$NON-NLS-1$
3942                 case TokenNamerequire:
3943                         return "require"; //$NON-NLS-1$
3944                 case TokenNamerequire_once:
3945                         return "require_once"; //$NON-NLS-1$
3946                 case TokenNamereturn:
3947                         return "return"; //$NON-NLS-1$
3948                 case TokenNamestatic:
3949                         return "static"; //$NON-NLS-1$
3950                 case TokenNameswitch:
3951                         return "switch"; //$NON-NLS-1$
3952                 // case TokenNametrue :
3953                 // return "true"; //$NON-NLS-1$
3954                 case TokenNameunset:
3955                         return "unset"; //$NON-NLS-1$
3956                 case TokenNamevar:
3957                         return "var"; //$NON-NLS-1$
3958                 case TokenNamewhile:
3959                         return "while"; //$NON-NLS-1$
3960                 case TokenNamexor:
3961                         return "XOR"; //$NON-NLS-1$
3962                 // case TokenNamethis :
3963                 // return "$this"; //$NON-NLS-1$
3964                 case TokenNameIntegerLiteral:
3965                         return "Integer(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3966                 case TokenNameDoubleLiteral:
3967                         return "Double(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3968                 case TokenNameStringDoubleQuote:
3969                         return "StringLiteral(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3970                 case TokenNameStringSingleQuote:
3971                         return "StringConstant(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3972                 case TokenNameStringInterpolated:
3973                         return "StringInterpolated(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3974                 case TokenNameEncapsedString0:
3975                         return "`"; //$NON-NLS-1$
3976                         // case TokenNameEncapsedString1:
3977                         // return "\'"; //$NON-NLS-1$
3978                         // case TokenNameEncapsedString2:
3979                         // return "\""; //$NON-NLS-1$
3980                 case TokenNameSTRING:
3981                         return "STRING_DQ(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3982                 case TokenNameHEREDOC:
3983                         return "HEREDOC(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
3984                 case TokenNamePLUS_PLUS:
3985                         return "++"; //$NON-NLS-1$
3986                 case TokenNameMINUS_MINUS:
3987                         return "--"; //$NON-NLS-1$
3988                 case TokenNameEQUAL_EQUAL:
3989                         return "=="; //$NON-NLS-1$
3990                 case TokenNameEQUAL_EQUAL_EQUAL:
3991                         return "==="; //$NON-NLS-1$
3992                 case TokenNameEQUAL_GREATER:
3993                         return "=>"; //$NON-NLS-1$
3994                 case TokenNameLESS_EQUAL:
3995                         return "<="; //$NON-NLS-1$
3996                 case TokenNameGREATER_EQUAL:
3997                         return ">="; //$NON-NLS-1$
3998                 case TokenNameNOT_EQUAL:
3999                         return "!="; //$NON-NLS-1$
4000                 case TokenNameNOT_EQUAL_EQUAL:
4001                         return "!=="; //$NON-NLS-1$
4002                 case TokenNameLEFT_SHIFT:
4003                         return "<<"; //$NON-NLS-1$
4004                 case TokenNameRIGHT_SHIFT:
4005                         return ">>"; //$NON-NLS-1$
4006                 case TokenNamePLUS_EQUAL:
4007                         return "+="; //$NON-NLS-1$
4008                 case TokenNameMINUS_EQUAL:
4009                         return "-="; //$NON-NLS-1$
4010                 case TokenNameMULTIPLY_EQUAL:
4011                         return "*="; //$NON-NLS-1$
4012                 case TokenNameDIVIDE_EQUAL:
4013                         return "/="; //$NON-NLS-1$
4014                 case TokenNameAND_EQUAL:
4015                         return "&="; //$NON-NLS-1$
4016                 case TokenNameOR_EQUAL:
4017                         return "|="; //$NON-NLS-1$
4018                 case TokenNameXOR_EQUAL:
4019                         return "^="; //$NON-NLS-1$
4020                 case TokenNameREMAINDER_EQUAL:
4021                         return "%="; //$NON-NLS-1$
4022                 case TokenNameDOT_EQUAL:
4023                         return ".="; //$NON-NLS-1$
4024                 case TokenNameLEFT_SHIFT_EQUAL:
4025                         return "<<="; //$NON-NLS-1$
4026                 case TokenNameRIGHT_SHIFT_EQUAL:
4027                         return ">>="; //$NON-NLS-1$
4028                 case TokenNameOR_OR:
4029                         return "||"; //$NON-NLS-1$
4030                 case TokenNameAND_AND:
4031                         return "&&"; //$NON-NLS-1$
4032                 case TokenNamePLUS:
4033                         return "+"; //$NON-NLS-1$
4034                 case TokenNameMINUS:
4035                         return "-"; //$NON-NLS-1$
4036                 case TokenNameMINUS_GREATER:
4037                         return "->";
4038                 case TokenNameNOT:
4039                         return "!"; //$NON-NLS-1$
4040                 case TokenNameREMAINDER:
4041                         return "%"; //$NON-NLS-1$
4042                 case TokenNameXOR:
4043                         return "^"; //$NON-NLS-1$
4044                 case TokenNameAND:
4045                         return "&"; //$NON-NLS-1$
4046                 case TokenNameMULTIPLY:
4047                         return "*"; //$NON-NLS-1$
4048                 case TokenNameOR:
4049                         return "|"; //$NON-NLS-1$
4050                 case TokenNameTWIDDLE:
4051                         return "~"; //$NON-NLS-1$
4052                 case TokenNameTWIDDLE_EQUAL:
4053                         return "~="; //$NON-NLS-1$
4054                 case TokenNameDIVIDE:
4055                         return "/"; //$NON-NLS-1$
4056                 case TokenNameGREATER:
4057                         return ">"; //$NON-NLS-1$
4058                 case TokenNameLESS:
4059                         return "<"; //$NON-NLS-1$
4060                 case TokenNameLPAREN:
4061                         return "("; //$NON-NLS-1$
4062                 case TokenNameRPAREN:
4063                         return ")"; //$NON-NLS-1$
4064                 case TokenNameLBRACE:
4065                         return "{"; //$NON-NLS-1$
4066                 case TokenNameRBRACE:
4067                         return "}"; //$NON-NLS-1$
4068                 case TokenNameLBRACKET:
4069                         return "["; //$NON-NLS-1$
4070                 case TokenNameRBRACKET:
4071                         return "]"; //$NON-NLS-1$
4072                 case TokenNameSEMICOLON:
4073                         return ";"; //$NON-NLS-1$
4074                 case TokenNameQUESTION:
4075                         return "?"; //$NON-NLS-1$
4076                 case TokenNameCOLON:
4077                         return ":"; //$NON-NLS-1$
4078                 case TokenNameCOMMA:
4079                         return ","; //$NON-NLS-1$
4080                 case TokenNameDOT:
4081                         return "."; //$NON-NLS-1$
4082                 case TokenNameEQUAL:
4083                         return "="; //$NON-NLS-1$
4084                 case TokenNameAT:
4085                         return "@";
4086                 case TokenNameDOLLAR:
4087                         return "$";
4088                 case TokenNameDOLLAR_LBRACE:
4089                         return "${";
4090                 case TokenNameLBRACE_DOLLAR:
4091                         return "{$";
4092                 case TokenNameEOF:
4093                         return "EOF"; //$NON-NLS-1$
4094                 case TokenNameWHITESPACE:
4095                         return "WHITESPACE(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4096                 case TokenNameCOMMENT_LINE:
4097                         return "COMMENT_LINE(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4098                 case TokenNameCOMMENT_BLOCK:
4099                         return "COMMENT_BLOCK(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4100                 case TokenNameCOMMENT_PHPDOC:
4101                         return "COMMENT_PHPDOC(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4102                 // case TokenNameHTML :
4103                 // return "HTML(" + new String(getCurrentTokenSource()) + ")";
4104                 // //$NON-NLS-1$
4105                 case TokenNameFILE:
4106                         return "__FILE__"; //$NON-NLS-1$
4107                 case TokenNameLINE:
4108                         return "__LINE__"; //$NON-NLS-1$
4109                 case TokenNameCLASS_C:
4110                         return "__CLASS__"; //$NON-NLS-1$
4111                 case TokenNameMETHOD_C:
4112                         return "__METHOD__"; //$NON-NLS-1$
4113                 case TokenNameFUNC_C:
4114                         return "__FUNCTION__"; //$NON-NLS-1
4115                 case TokenNameboolCAST:
4116                         return "( bool )"; //$NON-NLS-1$
4117                 case TokenNameintCAST:
4118                         return "( int )"; //$NON-NLS-1$
4119                 case TokenNamedoubleCAST:
4120                         return "( double )"; //$NON-NLS-1$
4121                 case TokenNameobjectCAST:
4122                         return "( object )"; //$NON-NLS-1$
4123                 case TokenNamestringCAST:
4124                         return "( string )"; //$NON-NLS-1$
4125                 default:
4126                         return "not-a-token(" + (new Integer(act)) + ") " + new String(getCurrentTokenSource()); //$NON-NLS-1$
4127                 }
4128         }
4129
4130         public Scanner() {
4131                 this(false, false);
4132         }
4133
4134         public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace) {
4135                 this(tokenizeComments, tokenizeWhiteSpace, false);
4136         }
4137
4138         public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, boolean checkNonExternalizedStringLiterals) {
4139                 this(tokenizeComments, tokenizeWhiteSpace, checkNonExternalizedStringLiterals, false);
4140         }
4141
4142         public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, boolean checkNonExternalizedStringLiterals,
4143                         boolean assertMode) {
4144                 this(tokenizeComments, tokenizeWhiteSpace, checkNonExternalizedStringLiterals, assertMode, false, null, null, true);
4145         }
4146
4147         public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, boolean checkNonExternalizedStringLiterals,
4148                         boolean assertMode, boolean tokenizeStrings, char[][] taskTags, char[][] taskPriorities, boolean isTaskCaseSensitive) {
4149                 this.eofPosition = Integer.MAX_VALUE;
4150                 this.tokenizeComments = tokenizeComments;
4151                 this.tokenizeWhiteSpace = tokenizeWhiteSpace;
4152                 this.tokenizeStrings = tokenizeStrings;
4153                 this.checkNonExternalizedStringLiterals = checkNonExternalizedStringLiterals;
4154 //              this.assertMode = assertMode;
4155                 // this.encapsedStringStack = null;
4156                 this.taskTags = taskTags;
4157                 this.taskPriorities = taskPriorities;
4158         }
4159
4160         private void checkNonExternalizeString() throws InvalidInputException {
4161                 if (currentLine == null)
4162                         return;
4163                 parseTags(currentLine);
4164         }
4165
4166         private void parseTags(NLSLine line) throws InvalidInputException {
4167                 String s = new String(getCurrentTokenSource());
4168                 int pos = s.indexOf(TAG_PREFIX);
4169                 int lineLength = line.size();
4170                 while (pos != -1) {
4171                         int start = pos + TAG_PREFIX_LENGTH;
4172                         int end = s.indexOf(TAG_POSTFIX, start);
4173                         String index = s.substring(start, end);
4174                         int i = 0;
4175                         try {
4176                                 i = Integer.parseInt(index) - 1;
4177                                 // Tags are one based not zero based.
4178                         } catch (NumberFormatException e) {
4179                                 i = -1; // we don't want to consider this as a valid NLS tag
4180                         }
4181                         if (line.exists(i)) {
4182                                 line.set(i, null);
4183                         }
4184                         pos = s.indexOf(TAG_PREFIX, start);
4185                 }
4186                 this.nonNLSStrings = new StringLiteral[lineLength];
4187                 int nonNLSCounter = 0;
4188                 for (Iterator iterator = line.iterator(); iterator.hasNext();) {
4189                         StringLiteral literal = (StringLiteral) iterator.next();
4190                         if (literal != null) {
4191                                 this.nonNLSStrings[nonNLSCounter++] = literal;
4192                         }
4193                 }
4194                 if (nonNLSCounter == 0) {
4195                         this.nonNLSStrings = null;
4196                         currentLine = null;
4197                         return;
4198                 }
4199                 this.wasNonExternalizedStringLiteral = true;
4200                 if (nonNLSCounter != lineLength) {
4201                         System.arraycopy(this.nonNLSStrings, 0, (this.nonNLSStrings = new StringLiteral[nonNLSCounter]), 0, nonNLSCounter);
4202                 }
4203                 currentLine = null;
4204         }
4205
4206         public final void scanEscapeCharacter() throws InvalidInputException {
4207                 // the string with "\\u" is a legal string of two chars \ and u
4208                 // thus we use a direct access to the source (for regular cases).
4209                 if (unicodeAsBackSlash) {
4210                         // consume next character
4211                         unicodeAsBackSlash = false;
4212                         // if (((currentCharacter = source[currentPosition++]) == '\\') &&
4213                         // (source[currentPosition] == 'u')) {
4214                         // getNextUnicodeChar();
4215                         // } else {
4216                         if (withoutUnicodePtr != 0) {
4217                                 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
4218                                 // }
4219                         }
4220                 } else
4221                         currentCharacter = source[currentPosition++];
4222                 switch (currentCharacter) {
4223                 case 'b':
4224                         currentCharacter = '\b';
4225                         break;
4226                 case 't':
4227                         currentCharacter = '\t';
4228                         break;
4229                 case 'n':
4230                         currentCharacter = '\n';
4231                         break;
4232                 case 'f':
4233                         currentCharacter = '\f';
4234                         break;
4235                 case 'r':
4236                         currentCharacter = '\r';
4237                         break;
4238                 case '\"':
4239                         currentCharacter = '\"';
4240                         break;
4241                 case '\'':
4242                         currentCharacter = '\'';
4243                         break;
4244                 case '\\':
4245                         currentCharacter = '\\';
4246                         break;
4247                 default:
4248                         // -----------octal escape--------------
4249                         // OctalDigit
4250                         // OctalDigit OctalDigit
4251                         // ZeroToThree OctalDigit OctalDigit
4252                         int number = Character.getNumericValue(currentCharacter);
4253                         if (number >= 0 && number <= 7) {
4254                                 boolean zeroToThreeNot = number > 3;
4255                                 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
4256                                         int digit = Character.getNumericValue(currentCharacter);
4257                                         if (digit >= 0 && digit <= 7) {
4258                                                 number = (number * 8) + digit;
4259                                                 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
4260                                                         if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit
4261                                                                 // Digit --> ignore last character
4262                                                                 currentPosition--;
4263                                                         } else {
4264                                                                 digit = Character.getNumericValue(currentCharacter);
4265                                                                 if (digit >= 0 && digit <= 7) { // has read \ZeroToThree
4266                                                                         // OctalDigit OctalDigit
4267                                                                         number = (number * 8) + digit;
4268                                                                 } else { // has read \ZeroToThree OctalDigit NonOctalDigit
4269                                                                         // --> ignore last character
4270                                                                         currentPosition--;
4271                                                                 }
4272                                                         }
4273                                                 } else { // has read \OctalDigit NonDigit--> ignore last
4274                                                         // character
4275                                                         currentPosition--;
4276                                                 }
4277                                         } else { // has read \OctalDigit NonOctalDigit--> ignore last
4278                                                 // character
4279                                                 currentPosition--;
4280                                         }
4281                                 } else { // has read \OctalDigit --> ignore last character
4282                                         currentPosition--;
4283                                 }
4284                                 if (number > 255)
4285                                         throw new InvalidInputException(INVALID_ESCAPE);
4286                                 currentCharacter = (char) number;
4287                         } else
4288                                 throw new InvalidInputException(INVALID_ESCAPE);
4289                 }
4290         }
4291
4292         // chech presence of task: tags
4293         // TODO (frederic) see if we need to take unicode characters into account...
4294         public void checkTaskTag(int commentStart, int commentEnd) {
4295                 char[] src = this.source;
4296
4297                 // only look for newer task: tags
4298                 if (this.foundTaskCount > 0 && this.foundTaskPositions[this.foundTaskCount - 1][0] >= commentStart) {
4299                         return;
4300                 }
4301                 int foundTaskIndex = this.foundTaskCount;
4302                 char previous = src[commentStart + 1]; // should be '*' or '/'
4303                 nextChar: for (int i = commentStart + 2; i < commentEnd && i < this.eofPosition; i++) {
4304                         char[] tag = null;
4305                         char[] priority = null;
4306                         // check for tag occurrence only if not ambiguous with javadoc tag
4307                         if (previous != '@') {
4308                                 nextTag: for (int itag = 0; itag < this.taskTags.length; itag++) {
4309                                         tag = this.taskTags[itag];
4310                                         int tagLength = tag.length;
4311                                         if (tagLength == 0)
4312                                                 continue nextTag;
4313
4314                                         // ensure tag is not leaded with letter if tag starts with a letter
4315                                         if (Scanner.isPHPIdentifierStart(tag[0])) {
4316                                                 if (Scanner.isPHPIdentifierPart(previous)) {
4317                                                         continue nextTag;
4318                                                 }
4319                                         }
4320
4321                                         for (int t = 0; t < tagLength; t++) {
4322                                                 char sc, tc;
4323                                                 int x = i + t;
4324                                                 if (x >= this.eofPosition || x >= commentEnd)
4325                                                         continue nextTag;
4326                                                 if ((sc = src[i + t]) != (tc = tag[t])) { // case sensitive check
4327                                                         if (this.isTaskCaseSensitive || (Character.toLowerCase(sc) != Character.toLowerCase(tc))) { // case
4328                                                                                                                                                                                                                                                                                                                                                                                                                                         // insensitive
4329                                                                                                                                                                                                                                                                                                                                                                                                                                         // check
4330                                                                 continue nextTag;
4331                                                         }
4332                                                 }
4333                                         }
4334                                         // ensure tag is not followed with letter if tag finishes with a
4335                                         // letter
4336                                         if (i + tagLength < commentEnd && Scanner.isPHPIdentifierPart(src[i + tagLength - 1])) {
4337                                                 if (Scanner.isPHPIdentifierPart(src[i + tagLength]))
4338                                                         continue nextTag;
4339                                         }
4340                                         if (this.foundTaskTags == null) {
4341                                                 this.foundTaskTags = new char[5][];
4342                                                 this.foundTaskMessages = new char[5][];
4343                                                 this.foundTaskPriorities = new char[5][];
4344                                                 this.foundTaskPositions = new int[5][];
4345                                         } else if (this.foundTaskCount == this.foundTaskTags.length) {
4346                                                 System.arraycopy(this.foundTaskTags, 0, this.foundTaskTags = new char[this.foundTaskCount * 2][], 0,
4347                                                                 this.foundTaskCount);
4348                                                 System.arraycopy(this.foundTaskMessages, 0, this.foundTaskMessages = new char[this.foundTaskCount * 2][], 0,
4349                                                                 this.foundTaskCount);
4350                                                 System.arraycopy(this.foundTaskPriorities, 0, this.foundTaskPriorities = new char[this.foundTaskCount * 2][], 0,
4351                                                                 this.foundTaskCount);
4352                                                 System.arraycopy(this.foundTaskPositions, 0, this.foundTaskPositions = new int[this.foundTaskCount * 2][], 0,
4353                                                                 this.foundTaskCount);
4354                                         }
4355
4356                                         priority = this.taskPriorities != null && itag < this.taskPriorities.length ? this.taskPriorities[itag] : null;
4357
4358                                         this.foundTaskTags[this.foundTaskCount] = tag;
4359                                         this.foundTaskPriorities[this.foundTaskCount] = priority;
4360                                         this.foundTaskPositions[this.foundTaskCount] = new int[] { i, i + tagLength - 1 };
4361                                         this.foundTaskMessages[this.foundTaskCount] = CharOperation.NO_CHAR;
4362                                         this.foundTaskCount++;
4363                                         i += tagLength - 1; // will be incremented when looping
4364                                         break nextTag;
4365                                 }
4366                         }
4367                         previous = src[i];
4368                 }
4369                 for (int i = foundTaskIndex; i < this.foundTaskCount; i++) {
4370                         // retrieve message start and end positions
4371                         int msgStart = this.foundTaskPositions[i][0] + this.foundTaskTags[i].length;
4372                         int max_value = i + 1 < this.foundTaskCount ? this.foundTaskPositions[i + 1][0] - 1 : commentEnd - 1;
4373                         // at most beginning of next task
4374                         if (max_value < msgStart) {
4375                                 max_value = msgStart; // would only occur if tag is before EOF.
4376                         }
4377                         int end = -1;
4378                         char c;
4379                         for (int j = msgStart; j < max_value; j++) {
4380                                 if ((c = src[j]) == '\n' || c == '\r') {
4381                                         end = j - 1;
4382                                         break;
4383                                 }
4384                         }
4385                         if (end == -1) {
4386                                 for (int j = max_value; j > msgStart; j--) {
4387                                         if ((c = src[j]) == '*') {
4388                                                 end = j - 1;
4389                                                 break;
4390                                         }
4391                                 }
4392                                 if (end == -1)
4393                                         end = max_value;
4394                         }
4395                         if (msgStart == end)
4396                                 continue; // empty
4397                         // trim the message
4398                         while (CharOperation.isWhitespace(src[end]) && msgStart <= end)
4399                                 end--;
4400                         while (CharOperation.isWhitespace(src[msgStart]) && msgStart <= end)
4401                                 msgStart++;
4402                         // update the end position of the task
4403                         this.foundTaskPositions[i][1] = end;
4404                         // get the message source
4405                         final int messageLength = end - msgStart + 1;
4406                         char[] message = new char[messageLength];
4407                         System.arraycopy(src, msgStart, message, 0, messageLength);
4408                         this.foundTaskMessages[i] = message;
4409                 }
4410         }
4411
4412         // chech presence of task: tags
4413         // public void checkTaskTag(int commentStart, int commentEnd) {
4414         // // only look for newer task: tags
4415         // if (this.foundTaskCount > 0 && this.foundTaskPositions[this.foundTaskCount
4416         // - 1][0] >= commentStart) {
4417         // return;
4418         // }
4419         // int foundTaskIndex = this.foundTaskCount;
4420         // nextChar: for (int i = commentStart; i < commentEnd && i <
4421         // this.eofPosition; i++) {
4422         // char[] tag = null;
4423         // char[] priority = null;
4424         // // check for tag occurrence
4425         // nextTag: for (int itag = 0; itag < this.taskTags.length; itag++) {
4426         // tag = this.taskTags[itag];
4427         // priority = this.taskPriorities != null && itag < this.taskPriorities.length
4428         // ? this.taskPriorities[itag] : null;
4429         // int tagLength = tag.length;
4430         // for (int t = 0; t < tagLength; t++) {
4431         // if (this.source[i + t] != tag[t])
4432         // continue nextTag;
4433         // }
4434         // if (this.foundTaskTags == null) {
4435         // this.foundTaskTags = new char[5][];
4436         // this.foundTaskMessages = new char[5][];
4437         // this.foundTaskPriorities = new char[5][];
4438         // this.foundTaskPositions = new int[5][];
4439         // } else if (this.foundTaskCount == this.foundTaskTags.length) {
4440         // System.arraycopy(this.foundTaskTags, 0, this.foundTaskTags = new
4441         // char[this.foundTaskCount * 2][], 0, this.foundTaskCount);
4442         // System.arraycopy(this.foundTaskMessages, 0, this.foundTaskMessages = new
4443         // char[this.foundTaskCount * 2][], 0,
4444         // this.foundTaskCount);
4445         // System.arraycopy(this.foundTaskPriorities, 0, this.foundTaskPriorities =
4446         // new char[this.foundTaskCount * 2][], 0,
4447         // this.foundTaskCount);
4448         // System.arraycopy(this.foundTaskPositions, 0, this.foundTaskPositions = new
4449         // int[this.foundTaskCount * 2][], 0,
4450         // this.foundTaskCount);
4451         // }
4452         // this.foundTaskTags[this.foundTaskCount] = tag;
4453         // this.foundTaskPriorities[this.foundTaskCount] = priority;
4454         // this.foundTaskPositions[this.foundTaskCount] = new int[] { i, i + tagLength
4455         // - 1 };
4456         // this.foundTaskMessages[this.foundTaskCount] = CharOperation.NO_CHAR;
4457         // this.foundTaskCount++;
4458         // i += tagLength - 1; // will be incremented when looping
4459         // }
4460         // }
4461         // for (int i = foundTaskIndex; i < this.foundTaskCount; i++) {
4462         // // retrieve message start and end positions
4463         // int msgStart = this.foundTaskPositions[i][0] +
4464         // this.foundTaskTags[i].length;
4465         // int max_value = i + 1 < this.foundTaskCount ? this.foundTaskPositions[i +
4466         // 1][0] - 1 : commentEnd - 1;
4467         // // at most beginning of next task
4468         // if (max_value < msgStart)
4469         // max_value = msgStart; // would only occur if tag is before EOF.
4470         // int end = -1;
4471         // char c;
4472         // for (int j = msgStart; j < max_value; j++) {
4473         // if ((c = this.source[j]) == '\n' || c == '\r') {
4474         // end = j - 1;
4475         // break;
4476         // }
4477         // }
4478         // if (end == -1) {
4479         // for (int j = max_value; j > msgStart; j--) {
4480         // if ((c = this.source[j]) == '*') {
4481         // end = j - 1;
4482         // break;
4483         // }
4484         // }
4485         // if (end == -1)
4486         // end = max_value;
4487         // }
4488         // if (msgStart == end)
4489         // continue; // empty
4490         // // trim the message
4491         // while (CharOperation.isWhitespace(source[end]) && msgStart <= end)
4492         // end--;
4493         // while (CharOperation.isWhitespace(source[msgStart]) && msgStart <= end)
4494         // msgStart++;
4495         // // update the end position of the task
4496         // this.foundTaskPositions[i][1] = end;
4497         // // get the message source
4498         // final int messageLength = end - msgStart + 1;
4499         // char[] message = new char[messageLength];
4500         // System.arraycopy(source, msgStart, message, 0, messageLength);
4501         // this.foundTaskMessages[i] = message;
4502         // }
4503         // }
4504 }