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
6 * Contributors: IBM Corporation - initial API and implementation
7 **********************************************************************************************************************************/
8 package net.sourceforge.phpdt.internal.compiler.parser;
10 import java.util.ArrayList;
11 import java.util.Iterator;
12 import java.util.List;
13 import java.util.Stack;
15 import net.sourceforge.phpdt.core.compiler.CharOperation;
16 import net.sourceforge.phpdt.core.compiler.IScanner;
17 import net.sourceforge.phpdt.core.compiler.ITerminalSymbols;
18 import net.sourceforge.phpdt.core.compiler.InvalidInputException;
19 import net.sourceforge.phpdt.internal.compiler.env.ICompilationUnit;
20 import net.sourceforge.phpeclipse.internal.compiler.ast.StringLiteral;
22 public class Scanner implements IScanner, ITerminalSymbols {
24 * APIs ares - getNextToken() which return the current type of the token (this value is not memorized by the scanner) -
25 * getCurrentTokenSource() which provides with the token "REAL" source (aka all unicode have been transformed into a correct char) -
26 * sourceStart gives the position into the stream - currentPosition-1 gives the sourceEnd position into the stream
29 private boolean assertMode;
31 public boolean useAssertAsAnIndentifier = false;
33 //flag indicating if processed source contains occurrences of keyword assert
34 public boolean containsAssertKeyword = false;
36 public boolean recordLineSeparator;
38 public boolean ignorePHPOneLiner = false;
40 public boolean phpMode = false;
42 public Stack encapsedStringStack = null;
44 public char currentCharacter;
46 public int startPosition;
48 public int currentPosition;
50 public int initialPosition, eofPosition;
52 // after this position eof are generated instead of real token from the
54 public boolean tokenizeComments;
56 public boolean tokenizeWhiteSpace;
58 public boolean tokenizeStrings;
60 //source should be viewed as a window (aka a part)
61 //of a entire very large stream
65 public char[] withoutUnicodeBuffer;
67 public int withoutUnicodePtr;
69 //when == 0 ==> no unicode in the current token
70 public boolean unicodeAsBackSlash = false;
72 public boolean scanningFloatLiteral = false;
74 //support for /** comments
75 public int[] commentStops = new int[10];
77 public int[] commentStarts = new int[10];
79 public int commentPtr = -1; // no comment test with commentPtr value -1
81 protected int lastCommentLinePosition = -1;
83 //diet parsing support - jump over some method body when requested
84 public boolean diet = false;
86 //support for the poor-line-debuggers ....
87 //remember the position of the cr/lf
88 public int[] lineEnds = new int[250];
90 public int linePtr = -1;
92 public boolean wasAcr = false;
94 public static final String END_OF_SOURCE = "End_Of_Source"; //$NON-NLS-1$
96 public static final String INVALID_HEXA = "Invalid_Hexa_Literal"; //$NON-NLS-1$
98 public static final String INVALID_OCTAL = "Invalid_Octal_Literal"; //$NON-NLS-1$
100 public static final String INVALID_CHARACTER_CONSTANT = "Invalid_Character_Constant"; //$NON-NLS-1$
102 public static final String INVALID_ESCAPE = "Invalid_Escape"; //$NON-NLS-1$
104 public static final String INVALID_INPUT = "Invalid_Input"; //$NON-NLS-1$
106 public static final String INVALID_UNICODE_ESCAPE = "Invalid_Unicode_Escape"; //$NON-NLS-1$
108 public static final String INVALID_FLOAT = "Invalid_Float_Literal"; //$NON-NLS-1$
110 public static final String NULL_SOURCE_STRING = "Null_Source_String"; //$NON-NLS-1$
112 public static final String UNTERMINATED_STRING = "Unterminated_String"; //$NON-NLS-1$
114 public static final String UNTERMINATED_COMMENT = "Unterminated_Comment"; //$NON-NLS-1$
116 public static final String INVALID_CHAR_IN_STRING = "Invalid_Char_In_String"; //$NON-NLS-1$
118 //----------------optimized identifier managment------------------
119 static final char[] charArray_a = new char[] { 'a' }, charArray_b = new char[] { 'b' }, charArray_c = new char[] { 'c' },
120 charArray_d = new char[] { 'd' }, charArray_e = new char[] { 'e' }, charArray_f = new char[] { 'f' },
121 charArray_g = new char[] { 'g' }, charArray_h = new char[] { 'h' }, charArray_i = new char[] { 'i' },
122 charArray_j = new char[] { 'j' }, charArray_k = new char[] { 'k' }, charArray_l = new char[] { 'l' },
123 charArray_m = new char[] { 'm' }, charArray_n = new char[] { 'n' }, charArray_o = new char[] { 'o' },
124 charArray_p = new char[] { 'p' }, charArray_q = new char[] { 'q' }, charArray_r = new char[] { 'r' },
125 charArray_s = new char[] { 's' }, charArray_t = new char[] { 't' }, charArray_u = new char[] { 'u' },
126 charArray_v = new char[] { 'v' }, charArray_w = new char[] { 'w' }, charArray_x = new char[] { 'x' },
127 charArray_y = new char[] { 'y' }, charArray_z = new char[] { 'z' };
129 static final char[] charArray_va = new char[] { '$', 'a' }, charArray_vb = new char[] { '$', 'b' }, charArray_vc = new char[] {
131 'c' }, charArray_vd = new char[] { '$', 'd' }, charArray_ve = new char[] { '$', 'e' },
132 charArray_vf = new char[] { '$', 'f' }, charArray_vg = new char[] { '$', 'g' }, charArray_vh = new char[] { '$', 'h' },
133 charArray_vi = new char[] { '$', 'i' }, charArray_vj = new char[] { '$', 'j' }, charArray_vk = new char[] { '$', 'k' },
134 charArray_vl = new char[] { '$', 'l' }, charArray_vm = new char[] { '$', 'm' }, charArray_vn = new char[] { '$', 'n' },
135 charArray_vo = new char[] { '$', 'o' }, charArray_vp = new char[] { '$', 'p' }, charArray_vq = new char[] { '$', 'q' },
136 charArray_vr = new char[] { '$', 'r' }, charArray_vs = new char[] { '$', 's' }, charArray_vt = new char[] { '$', 't' },
137 charArray_vu = new char[] { '$', 'u' }, charArray_vv = new char[] { '$', 'v' }, charArray_vw = new char[] { '$', 'w' },
138 charArray_vx = new char[] { '$', 'x' }, charArray_vy = new char[] { '$', 'y' }, charArray_vz = new char[] { '$', 'z' };
140 static final char[] initCharArray = new char[] { '\u0000', '\u0000', '\u0000', '\u0000', '\u0000', '\u0000' };
142 static final int TableSize = 30, InternalTableSize = 6;
145 public static final int OptimizedLength = 6;
148 final char[][][][] charArray_length = new char[OptimizedLength][TableSize][InternalTableSize][];
150 // support for detecting non-externalized string literals
151 int currentLineNr = -1;
153 int previousLineNr = -1;
155 NLSLine currentLine = null;
157 List lines = new ArrayList();
159 public static final String TAG_PREFIX = "//$NON-NLS-"; //$NON-NLS-1$
161 public static final int TAG_PREFIX_LENGTH = TAG_PREFIX.length();
163 public static final String TAG_POSTFIX = "$"; //$NON-NLS-1$
165 public static final int TAG_POSTFIX_LENGTH = TAG_POSTFIX.length();
167 public StringLiteral[] nonNLSStrings = null;
169 public boolean checkNonExternalizedStringLiterals = true;
171 public boolean wasNonExternalizedStringLiteral = false;
173 for (int i = 0; i < 6; i++) {
174 for (int j = 0; j < TableSize; j++) {
175 for (int k = 0; k < InternalTableSize; k++) {
176 charArray_length[i][j][k] = initCharArray;
182 static int newEntry2 = 0, newEntry3 = 0, newEntry4 = 0, newEntry5 = 0, newEntry6 = 0;
184 public static final int RoundBracket = 0;
186 public static final int SquareBracket = 1;
188 public static final int CurlyBracket = 2;
190 public static final int BracketKinds = 3;
193 public char[][] foundTaskTags = null;
195 public char[][] foundTaskMessages;
197 public char[][] foundTaskPriorities = null;
199 public int[][] foundTaskPositions;
201 public int foundTaskCount = 0;
203 public char[][] taskTags = null;
205 public char[][] taskPriorities = null;
207 public boolean isTaskCaseSensitive = true;
209 public static final boolean DEBUG = false;
211 public static final boolean TRACE = false;
213 public ICompilationUnit compilationUnit = null;
216 * Determines if the specified character is permissible as the first character in a PHP identifier or variable
218 * The '$' character for PHP variables is regarded as a correct first character !
221 public static boolean isPHPIdentOrVarStart(char ch) {
222 return Character.isLetter(ch) || (ch == '$') || (ch == '_') || (0x7F <= ch && ch <= 0xFF);
226 * Determines if the specified character is permissible as the first character in a PHP identifier.
228 * The '$' character for PHP variables isn't regarded as the first character !
230 public static boolean isPHPIdentifierStart(char ch) {
231 return Character.isLetter(ch) || (ch == '_') || (0x7F <= ch && ch <= 0xFF);
235 * Determines if the specified character may be part of a PHP identifier as other than the first character
237 public static boolean isPHPIdentifierPart(char ch) {
238 return Character.isLetterOrDigit(ch) || (ch == '_') || (0x7F <= ch && ch <= 0xFF);
241 public final boolean atEnd() {
242 // This code is not relevant if source is
243 // Only a part of the real stream input
244 return source.length == currentPosition;
247 public char[] getCurrentIdentifierSource() {
248 //return the token REAL source (aka unicodes are precomputed)
250 // if (withoutUnicodePtr != 0)
251 // //0 is used as a fast test flag so the real first char is in position 1
253 // withoutUnicodeBuffer,
255 // result = new char[withoutUnicodePtr],
257 // withoutUnicodePtr);
259 int length = currentPosition - startPosition;
260 switch (length) { // see OptimizedLength
262 return optimizedCurrentTokenSource1();
264 return optimizedCurrentTokenSource2();
266 return optimizedCurrentTokenSource3();
268 return optimizedCurrentTokenSource4();
270 return optimizedCurrentTokenSource5();
272 return optimizedCurrentTokenSource6();
275 System.arraycopy(source, startPosition, result = new char[length], 0, length);
280 public int getCurrentTokenEndPosition() {
281 return this.currentPosition - 1;
284 public final char[] getCurrentTokenSource() {
285 // Return the token REAL source (aka unicodes are precomputed)
287 // if (withoutUnicodePtr != 0)
288 // // 0 is used as a fast test flag so the real first char is in position 1
290 // withoutUnicodeBuffer,
292 // result = new char[withoutUnicodePtr],
294 // withoutUnicodePtr);
297 System.arraycopy(source, startPosition, result = new char[length = currentPosition - startPosition], 0, length);
302 public final char[] getCurrentTokenSource(int startPos) {
303 // Return the token REAL source (aka unicodes are precomputed)
305 // if (withoutUnicodePtr != 0)
306 // // 0 is used as a fast test flag so the real first char is in position 1
308 // withoutUnicodeBuffer,
310 // result = new char[withoutUnicodePtr],
312 // withoutUnicodePtr);
315 System.arraycopy(source, startPos, result = new char[length = currentPosition - startPos], 0, length);
320 public final char[] getCurrentTokenSourceString() {
321 //return the token REAL source (aka unicodes are precomputed).
322 //REMOVE the two " that are at the beginning and the end.
324 if (withoutUnicodePtr != 0)
325 //0 is used as a fast test flag so the real first char is in position 1
326 System.arraycopy(withoutUnicodeBuffer, 2,
327 //2 is 1 (real start) + 1 (to jump over the ")
328 result = new char[withoutUnicodePtr - 2], 0, withoutUnicodePtr - 2);
331 System.arraycopy(source, startPosition + 1, result = new char[length = currentPosition - startPosition - 2], 0, length);
336 public final char[] getRawTokenSourceEnd() {
337 int length = this.eofPosition - this.currentPosition - 1;
338 char[] sourceEnd = new char[length];
339 System.arraycopy(this.source, this.currentPosition, sourceEnd, 0, length);
343 public int getCurrentTokenStartPosition() {
344 return this.startPosition;
347 public final char[] getCurrentStringLiteralSource() {
348 // Return the token REAL source (aka unicodes are precomputed)
349 if (startPosition + 1 >= currentPosition) {
354 System.arraycopy(source, startPosition + 1, result = new char[length = currentPosition - startPosition - 2], 0, length);
359 public final char[] getCurrentStringLiteralSource(int startPos) {
360 // Return the token REAL source (aka unicodes are precomputed)
363 System.arraycopy(source, startPos + 1, result = new char[length = currentPosition - startPos - 2], 0, length);
369 * Search the source position corresponding to the end of a given line number
371 * Line numbers are 1-based, and relative to the scanner initialPosition. Character positions are 0-based.
373 * In case the given line number is inconsistent, answers -1.
375 public final int getLineEnd(int lineNumber) {
376 if (lineEnds == null)
378 if (lineNumber >= lineEnds.length)
382 if (lineNumber == lineEnds.length - 1)
384 return lineEnds[lineNumber - 1];
385 // next line start one character behind the lineEnd of the previous line
389 * Search the source position corresponding to the beginning of a given line number
391 * Line numbers are 1-based, and relative to the scanner initialPosition. Character positions are 0-based.
393 * e.g. getLineStart(1) --> 0 i.e. first line starts at character 0.
395 * In case the given line number is inconsistent, answers -1.
397 public final int getLineStart(int lineNumber) {
398 if (lineEnds == null)
400 if (lineNumber >= lineEnds.length)
405 return initialPosition;
406 return lineEnds[lineNumber - 2] + 1;
407 // next line start one character behind the lineEnd of the previous line
410 public final boolean getNextChar(char testedChar) {
412 //handle the case of unicode.
413 //when a unicode appears then we must use a buffer that holds char
415 //At the end of this method currentCharacter holds the new visited char
416 //and currentPosition points right next after it
417 //Both previous lines are true if the currentCharacter is == to the
419 //On false, no side effect has occured.
420 //ALL getNextChar.... ARE OPTIMIZED COPIES
421 int temp = currentPosition;
423 currentCharacter = source[currentPosition++];
424 // if (((currentCharacter = source[currentPosition++]) == '\\')
425 // && (source[currentPosition] == 'u')) {
426 // //-------------unicode traitement ------------
427 // int c1, c2, c3, c4;
428 // int unicodeSize = 6;
429 // currentPosition++;
430 // while (source[currentPosition] == 'u') {
431 // currentPosition++;
435 // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
437 // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
439 // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
441 // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
443 // currentPosition = temp;
447 // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
448 // if (currentCharacter != testedChar) {
449 // currentPosition = temp;
452 // unicodeAsBackSlash = currentCharacter == '\\';
454 // //need the unicode buffer
455 // if (withoutUnicodePtr == 0) {
456 // //buffer all the entries that have been left aside....
457 // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
461 // withoutUnicodeBuffer,
463 // withoutUnicodePtr);
465 // //fill the buffer with the char
466 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
469 // } //-------------end unicode traitement--------------
471 if (currentCharacter != testedChar) {
472 currentPosition = temp;
475 unicodeAsBackSlash = false;
476 // if (withoutUnicodePtr != 0)
477 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
480 } catch (IndexOutOfBoundsException e) {
481 unicodeAsBackSlash = false;
482 currentPosition = temp;
487 public final int getNextChar(char testedChar1, char testedChar2) {
488 //INT 0 : testChar1 \\\\///\\\\ 1 : testedChar2 \\\\///\\\\ -1 : others
489 //test can be done with (x==0) for the first and (x>0) for the second
490 //handle the case of unicode.
491 //when a unicode appears then we must use a buffer that holds char
493 //At the end of this method currentCharacter holds the new visited char
494 //and currentPosition points right next after it
495 //Both previous lines are true if the currentCharacter is == to the
497 //On false, no side effect has occured.
498 //ALL getNextChar.... ARE OPTIMIZED COPIES
499 int temp = currentPosition;
502 currentCharacter = source[currentPosition++];
503 // if (((currentCharacter = source[currentPosition++]) == '\\')
504 // && (source[currentPosition] == 'u')) {
505 // //-------------unicode traitement ------------
506 // int c1, c2, c3, c4;
507 // int unicodeSize = 6;
508 // currentPosition++;
509 // while (source[currentPosition] == 'u') {
510 // currentPosition++;
514 // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
516 // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
518 // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
520 // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
522 // currentPosition = temp;
526 // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
527 // if (currentCharacter == testedChar1)
529 // else if (currentCharacter == testedChar2)
532 // currentPosition = temp;
536 // //need the unicode buffer
537 // if (withoutUnicodePtr == 0) {
538 // //buffer all the entries that have been left aside....
539 // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
543 // withoutUnicodeBuffer,
545 // withoutUnicodePtr);
547 // //fill the buffer with the char
548 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
550 // } //-------------end unicode traitement--------------
552 if (currentCharacter == testedChar1)
554 else if (currentCharacter == testedChar2)
557 currentPosition = temp;
560 // if (withoutUnicodePtr != 0)
561 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
564 } catch (IndexOutOfBoundsException e) {
565 currentPosition = temp;
570 public final boolean getNextCharAsDigit() {
572 //handle the case of unicode.
573 //when a unicode appears then we must use a buffer that holds char
575 //At the end of this method currentCharacter holds the new visited char
576 //and currentPosition points right next after it
577 //Both previous lines are true if the currentCharacter is a digit
578 //On false, no side effect has occured.
579 //ALL getNextChar.... ARE OPTIMIZED COPIES
580 int temp = currentPosition;
582 currentCharacter = source[currentPosition++];
583 // if (((currentCharacter = source[currentPosition++]) == '\\')
584 // && (source[currentPosition] == 'u')) {
585 // //-------------unicode traitement ------------
586 // int c1, c2, c3, c4;
587 // int unicodeSize = 6;
588 // currentPosition++;
589 // while (source[currentPosition] == 'u') {
590 // currentPosition++;
594 // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
596 // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
598 // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
600 // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
602 // currentPosition = temp;
606 // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
607 // if (!Character.isDigit(currentCharacter)) {
608 // currentPosition = temp;
612 // //need the unicode buffer
613 // if (withoutUnicodePtr == 0) {
614 // //buffer all the entries that have been left aside....
615 // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
619 // withoutUnicodeBuffer,
621 // withoutUnicodePtr);
623 // //fill the buffer with the char
624 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
626 // } //-------------end unicode traitement--------------
628 if (!Character.isDigit(currentCharacter)) {
629 currentPosition = temp;
632 // if (withoutUnicodePtr != 0)
633 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
636 } catch (IndexOutOfBoundsException e) {
637 currentPosition = temp;
642 public final boolean getNextCharAsDigit(int radix) {
644 //handle the case of unicode.
645 //when a unicode appears then we must use a buffer that holds char
647 //At the end of this method currentCharacter holds the new visited char
648 //and currentPosition points right next after it
649 //Both previous lines are true if the currentCharacter is a digit base on
651 //On false, no side effect has occured.
652 //ALL getNextChar.... ARE OPTIMIZED COPIES
653 int temp = currentPosition;
655 currentCharacter = source[currentPosition++];
656 // if (((currentCharacter = source[currentPosition++]) == '\\')
657 // && (source[currentPosition] == 'u')) {
658 // //-------------unicode traitement ------------
659 // int c1, c2, c3, c4;
660 // int unicodeSize = 6;
661 // currentPosition++;
662 // while (source[currentPosition] == 'u') {
663 // currentPosition++;
667 // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
669 // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
671 // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
673 // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
675 // currentPosition = temp;
679 // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
680 // if (Character.digit(currentCharacter, radix) == -1) {
681 // currentPosition = temp;
685 // //need the unicode buffer
686 // if (withoutUnicodePtr == 0) {
687 // //buffer all the entries that have been left aside....
688 // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
692 // withoutUnicodeBuffer,
694 // withoutUnicodePtr);
696 // //fill the buffer with the char
697 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
699 // } //-------------end unicode traitement--------------
701 if (Character.digit(currentCharacter, radix) == -1) {
702 currentPosition = temp;
705 // if (withoutUnicodePtr != 0)
706 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
709 } catch (IndexOutOfBoundsException e) {
710 currentPosition = temp;
715 public boolean getNextCharAsJavaIdentifierPart() {
717 //handle the case of unicode.
718 //when a unicode appears then we must use a buffer that holds char
720 //At the end of this method currentCharacter holds the new visited char
721 //and currentPosition points right next after it
722 //Both previous lines are true if the currentCharacter is a
723 // JavaIdentifierPart
724 //On false, no side effect has occured.
725 //ALL getNextChar.... ARE OPTIMIZED COPIES
726 int temp = currentPosition;
728 currentCharacter = source[currentPosition++];
729 // if (((currentCharacter = source[currentPosition++]) == '\\')
730 // && (source[currentPosition] == 'u')) {
731 // //-------------unicode traitement ------------
732 // int c1, c2, c3, c4;
733 // int unicodeSize = 6;
734 // currentPosition++;
735 // while (source[currentPosition] == 'u') {
736 // currentPosition++;
740 // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
742 // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
744 // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
746 // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
748 // currentPosition = temp;
752 // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
753 // if (!isPHPIdentifierPart(currentCharacter)) {
754 // currentPosition = temp;
758 // //need the unicode buffer
759 // if (withoutUnicodePtr == 0) {
760 // //buffer all the entries that have been left aside....
761 // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
765 // withoutUnicodeBuffer,
767 // withoutUnicodePtr);
769 // //fill the buffer with the char
770 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
772 // } //-------------end unicode traitement--------------
774 if (!isPHPIdentifierPart(currentCharacter)) {
775 currentPosition = temp;
778 // if (withoutUnicodePtr != 0)
779 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
782 } catch (IndexOutOfBoundsException e) {
783 currentPosition = temp;
788 public int getCastOrParen() {
789 int tempPosition = currentPosition;
790 char tempCharacter = currentCharacter;
791 int tempToken = TokenNameLPAREN;
792 boolean found = false;
793 StringBuffer buf = new StringBuffer();
796 currentCharacter = source[currentPosition++];
797 } while (currentCharacter == ' ' || currentCharacter == '\t');
798 while ((currentCharacter >= 'a' && currentCharacter <= 'z') || (currentCharacter >= 'A' && currentCharacter <= 'Z')) {
799 buf.append(currentCharacter);
800 currentCharacter = source[currentPosition++];
802 if (buf.length() >= 3 && buf.length() <= 7) {
803 char[] data = buf.toString().toCharArray();
805 switch (data.length) {
808 if ((data[index] == 'i') && (data[++index] == 'n') && (data[++index] == 't')) {
810 tempToken = TokenNameintCAST;
815 if ((data[index] == 'b') && (data[++index] == 'o') && (data[++index] == 'o') && (data[++index] == 'l')) {
817 tempToken = TokenNameboolCAST;
820 if ((data[index] == 'r') && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'l')) {
822 tempToken = TokenNamedoubleCAST;
828 if ((data[index] == 'a') && (data[++index] == 'r') && (data[++index] == 'r') && (data[++index] == 'a')
829 && (data[++index] == 'y')) {
831 tempToken = TokenNamearrayCAST;
834 if ((data[index] == 'u') && (data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 'e')
835 && (data[++index] == 't')) {
837 tempToken = TokenNameunsetCAST;
840 if ((data[index] == 'f') && (data[++index] == 'l') && (data[++index] == 'o') && (data[++index] == 'a')
841 && (data[++index] == 't')) {
843 tempToken = TokenNamedoubleCAST;
849 // object string double
850 if ((data[index] == 'o') && (data[++index] == 'b') && (data[++index] == 'j') && (data[++index] == 'e')
851 && (data[++index] == 'c') && (data[++index] == 't')) {
853 tempToken = TokenNameobjectCAST;
856 if ((data[index] == 's') && (data[++index] == 't') && (data[++index] == 'r') && (data[++index] == 'i')
857 && (data[++index] == 'n') && (data[++index] == 'g')) {
859 tempToken = TokenNamestringCAST;
862 if ((data[index] == 'd') && (data[++index] == 'o') && (data[++index] == 'u') && (data[++index] == 'b')
863 && (data[++index] == 'l') && (data[++index] == 'e')) {
865 tempToken = TokenNamedoubleCAST;
872 if ((data[index] == 'b') && (data[++index] == 'o') && (data[++index] == 'o') && (data[++index] == 'l')
873 && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'n')) {
875 tempToken = TokenNameboolCAST;
878 if ((data[index] == 'i') && (data[++index] == 'n') && (data[++index] == 't') && (data[++index] == 'e')
879 && (data[++index] == 'g') && (data[++index] == 'e') && (data[++index] == 'r')) {
881 tempToken = TokenNameintCAST;
887 while (currentCharacter == ' ' || currentCharacter == '\t') {
888 currentCharacter = source[currentPosition++];
890 if (currentCharacter == ')') {
895 } catch (IndexOutOfBoundsException e) {
897 currentCharacter = tempCharacter;
898 currentPosition = tempPosition;
899 return TokenNameLPAREN;
902 public void consumeStringInterpolated() throws InvalidInputException {
904 // consume next character
905 unicodeAsBackSlash = false;
906 currentCharacter = source[currentPosition++];
907 // if (((currentCharacter = source[currentPosition++]) == '\\')
908 // && (source[currentPosition] == 'u')) {
909 // getNextUnicodeChar();
911 // if (withoutUnicodePtr != 0) {
912 // withoutUnicodeBuffer[++withoutUnicodePtr] =
916 while (currentCharacter != '`') {
917 /** ** in PHP \r and \n are valid in string literals *** */
918 // if ((currentCharacter == '\n')
919 // || (currentCharacter == '\r')) {
920 // // relocate if finding another quote fairly close: thus unicode
921 // '/u000D' will be fully consumed
922 // for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
923 // if (currentPosition + lookAhead == source.length)
925 // if (source[currentPosition + lookAhead] == '\n')
927 // if (source[currentPosition + lookAhead] == '\"') {
928 // currentPosition += lookAhead + 1;
932 // throw new InvalidInputException(INVALID_CHAR_IN_STRING);
934 if (currentCharacter == '\\') {
935 int escapeSize = currentPosition;
936 boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
937 //scanEscapeCharacter make a side effect on this value and we need
938 // the previous value few lines down this one
939 scanDoubleQuotedEscapeCharacter();
940 escapeSize = currentPosition - escapeSize;
941 if (withoutUnicodePtr == 0) {
942 //buffer all the entries that have been left aside....
943 withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition;
944 System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr);
945 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
946 } else { //overwrite the / in the buffer
947 withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
948 if (backSlashAsUnicodeInString) { //there are TWO \ in the stream
949 // where only one is correct
953 } else if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
954 if (recordLineSeparator) {
958 // consume next character
959 unicodeAsBackSlash = false;
960 currentCharacter = source[currentPosition++];
961 // if (((currentCharacter = source[currentPosition++]) == '\\')
962 // && (source[currentPosition] == 'u')) {
963 // getNextUnicodeChar();
965 if (withoutUnicodePtr != 0) {
966 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
970 } catch (IndexOutOfBoundsException e) {
971 // reset end position for error reporting
972 currentPosition -= 2;
973 throw new InvalidInputException(UNTERMINATED_STRING);
974 } catch (InvalidInputException e) {
975 if (e.getMessage().equals(INVALID_ESCAPE)) {
976 // relocate if finding another quote fairly close: thus unicode
977 // '/u000D' will be fully consumed
978 for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
979 if (currentPosition + lookAhead == source.length)
981 if (source[currentPosition + lookAhead] == '\n')
983 if (source[currentPosition + lookAhead] == '`') {
984 currentPosition += lookAhead + 1;
991 if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags
992 // //$NON-NLS-?$ where ? is an
994 if (currentLine == null) {
995 currentLine = new NLSLine();
996 lines.add(currentLine);
998 currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1));
1002 public void consumeStringConstant() throws InvalidInputException {
1004 // consume next character
1005 unicodeAsBackSlash = false;
1006 currentCharacter = source[currentPosition++];
1007 // if (((currentCharacter = source[currentPosition++]) == '\\')
1008 // && (source[currentPosition] == 'u')) {
1009 // getNextUnicodeChar();
1011 // if (withoutUnicodePtr != 0) {
1012 // withoutUnicodeBuffer[++withoutUnicodePtr] =
1013 // currentCharacter;
1016 while (currentCharacter != '\'') {
1017 /** ** in PHP \r and \n are valid in string literals *** */
1018 // if ((currentCharacter == '\n')
1019 // || (currentCharacter == '\r')) {
1020 // // relocate if finding another quote fairly close: thus unicode
1021 // '/u000D' will be fully consumed
1022 // for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1023 // if (currentPosition + lookAhead == source.length)
1025 // if (source[currentPosition + lookAhead] == '\n')
1027 // if (source[currentPosition + lookAhead] == '\"') {
1028 // currentPosition += lookAhead + 1;
1032 // throw new InvalidInputException(INVALID_CHAR_IN_STRING);
1034 if (currentCharacter == '\\') {
1035 int escapeSize = currentPosition;
1036 boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
1037 //scanEscapeCharacter make a side effect on this value and we need
1038 // the previous value few lines down this one
1039 scanSingleQuotedEscapeCharacter();
1040 escapeSize = currentPosition - escapeSize;
1041 if (withoutUnicodePtr == 0) {
1042 //buffer all the entries that have been left aside....
1043 withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition;
1044 System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr);
1045 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1046 } else { //overwrite the / in the buffer
1047 withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
1048 if (backSlashAsUnicodeInString) { //there are TWO \ in the stream
1049 // where only one is correct
1050 withoutUnicodePtr--;
1053 } else if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1054 if (recordLineSeparator) {
1055 pushLineSeparator();
1058 // consume next character
1059 unicodeAsBackSlash = false;
1060 currentCharacter = source[currentPosition++];
1061 // if (((currentCharacter = source[currentPosition++]) == '\\')
1062 // && (source[currentPosition] == 'u')) {
1063 // getNextUnicodeChar();
1065 if (withoutUnicodePtr != 0) {
1066 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1070 } catch (IndexOutOfBoundsException e) {
1071 // reset end position for error reporting
1072 currentPosition -= 2;
1073 throw new InvalidInputException(UNTERMINATED_STRING);
1074 } catch (InvalidInputException e) {
1075 if (e.getMessage().equals(INVALID_ESCAPE)) {
1076 // relocate if finding another quote fairly close: thus unicode
1077 // '/u000D' will be fully consumed
1078 for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1079 if (currentPosition + lookAhead == source.length)
1081 if (source[currentPosition + lookAhead] == '\n')
1083 if (source[currentPosition + lookAhead] == '\'') {
1084 currentPosition += lookAhead + 1;
1091 if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags
1092 // //$NON-NLS-?$ where ? is an
1094 if (currentLine == null) {
1095 currentLine = new NLSLine();
1096 lines.add(currentLine);
1098 currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1));
1102 public void consumeStringLiteral() throws InvalidInputException {
1104 // consume next character
1105 unicodeAsBackSlash = false;
1106 currentCharacter = source[currentPosition++];
1107 // if (((currentCharacter = source[currentPosition++]) == '\\')
1108 // && (source[currentPosition] == 'u')) {
1109 // getNextUnicodeChar();
1111 // if (withoutUnicodePtr != 0) {
1112 // withoutUnicodeBuffer[++withoutUnicodePtr] =
1113 // currentCharacter;
1116 while (currentCharacter != '"') {
1117 /** ** in PHP \r and \n are valid in string literals *** */
1118 // if ((currentCharacter == '\n')
1119 // || (currentCharacter == '\r')) {
1120 // // relocate if finding another quote fairly close: thus unicode
1121 // '/u000D' will be fully consumed
1122 // for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1123 // if (currentPosition + lookAhead == source.length)
1125 // if (source[currentPosition + lookAhead] == '\n')
1127 // if (source[currentPosition + lookAhead] == '\"') {
1128 // currentPosition += lookAhead + 1;
1132 // throw new InvalidInputException(INVALID_CHAR_IN_STRING);
1134 if (currentCharacter == '\\') {
1135 int escapeSize = currentPosition;
1136 boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
1137 //scanEscapeCharacter make a side effect on this value and we need
1138 // the previous value few lines down this one
1139 scanDoubleQuotedEscapeCharacter();
1140 escapeSize = currentPosition - escapeSize;
1141 if (withoutUnicodePtr == 0) {
1142 //buffer all the entries that have been left aside....
1143 withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition;
1144 System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr);
1145 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1146 } else { //overwrite the / in the buffer
1147 withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
1148 if (backSlashAsUnicodeInString) { //there are TWO \ in the stream
1149 // where only one is correct
1150 withoutUnicodePtr--;
1153 } else if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1154 if (recordLineSeparator) {
1155 pushLineSeparator();
1158 // consume next character
1159 unicodeAsBackSlash = false;
1160 currentCharacter = source[currentPosition++];
1161 // if (((currentCharacter = source[currentPosition++]) == '\\')
1162 // && (source[currentPosition] == 'u')) {
1163 // getNextUnicodeChar();
1165 if (withoutUnicodePtr != 0) {
1166 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1170 } catch (IndexOutOfBoundsException e) {
1171 // reset end position for error reporting
1172 currentPosition -= 2;
1173 throw new InvalidInputException(UNTERMINATED_STRING);
1174 } catch (InvalidInputException e) {
1175 if (e.getMessage().equals(INVALID_ESCAPE)) {
1176 // relocate if finding another quote fairly close: thus unicode
1177 // '/u000D' will be fully consumed
1178 for (int lookAhead = 0; lookAhead < 50; lookAhead++) {
1179 if (currentPosition + lookAhead == source.length)
1181 if (source[currentPosition + lookAhead] == '\n')
1183 if (source[currentPosition + lookAhead] == '\"') {
1184 currentPosition += lookAhead + 1;
1191 if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags
1192 // //$NON-NLS-?$ where ? is an
1194 if (currentLine == null) {
1195 currentLine = new NLSLine();
1196 lines.add(currentLine);
1198 currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1));
1202 public int getNextToken() throws InvalidInputException {
1204 return getInlinedHTML(currentPosition);
1207 this.wasAcr = false;
1209 jumpOverMethodBody();
1211 return currentPosition > source.length ? TokenNameEOF : TokenNameRBRACE;
1215 withoutUnicodePtr = 0;
1216 //start with a new token
1217 char encapsedChar = ' ';
1218 if (!encapsedStringStack.isEmpty()) {
1219 encapsedChar = ((Character) encapsedStringStack.peek()).charValue();
1221 if (encapsedChar != '$' && encapsedChar != ' ') {
1222 currentCharacter = source[currentPosition++];
1223 if (currentCharacter == encapsedChar) {
1224 switch (currentCharacter) {
1226 return TokenNameEncapsedString0;
1228 return TokenNameEncapsedString1;
1230 return TokenNameEncapsedString2;
1233 while (currentCharacter != encapsedChar) {
1234 /** ** in PHP \r and \n are valid in string literals *** */
1235 switch (currentCharacter) {
1237 int escapeSize = currentPosition;
1238 boolean backSlashAsUnicodeInString = unicodeAsBackSlash;
1239 //scanEscapeCharacter make a side effect on this value and
1240 // we need the previous value few lines down this one
1241 scanDoubleQuotedEscapeCharacter();
1242 escapeSize = currentPosition - escapeSize;
1243 if (withoutUnicodePtr == 0) {
1244 //buffer all the entries that have been left aside....
1245 withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition;
1246 System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr);
1247 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1248 } else { //overwrite the / in the buffer
1249 withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter;
1250 if (backSlashAsUnicodeInString) { //there are TWO \ in
1251 withoutUnicodePtr--;
1257 if (recordLineSeparator) {
1258 pushLineSeparator();
1262 if (isPHPIdentifierStart(source[currentPosition]) || source[currentPosition] == '{') {
1264 encapsedStringStack.push(new Character('$'));
1265 return TokenNameSTRING;
1269 if (source[currentPosition] == '$') { // CURLY_OPEN
1271 encapsedStringStack.push(new Character('$'));
1272 return TokenNameSTRING;
1275 // consume next character
1276 unicodeAsBackSlash = false;
1277 currentCharacter = source[currentPosition++];
1278 if (withoutUnicodePtr != 0) {
1279 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
1284 return TokenNameSTRING;
1286 // ---------Consume white space and handles startPosition---------
1287 int whiteStart = currentPosition;
1288 startPosition = currentPosition;
1289 currentCharacter = source[currentPosition++];
1290 if (encapsedChar == '$') {
1291 switch (currentCharacter) {
1293 currentCharacter = source[currentPosition++];
1294 return TokenNameSTRING;
1296 if (encapsedChar == '$') {
1297 if (getNextChar('$'))
1298 return TokenNameLBRACE_DOLLAR;
1300 return TokenNameLBRACE;
1302 return TokenNameRBRACE;
1304 return TokenNameLBRACKET;
1306 return TokenNameRBRACKET;
1308 if (tokenizeStrings) {
1309 consumeStringConstant();
1310 return TokenNameStringSingleQuote;
1312 return TokenNameEncapsedString1;
1314 return TokenNameEncapsedString2;
1316 if (tokenizeStrings) {
1317 consumeStringInterpolated();
1318 return TokenNameStringInterpolated;
1320 return TokenNameEncapsedString0;
1322 if (getNextChar('>'))
1323 return TokenNameMINUS_GREATER;
1324 return TokenNameSTRING;
1326 if (currentCharacter == '$') {
1327 int oldPosition = currentPosition;
1329 currentCharacter = source[currentPosition++];
1330 if (currentCharacter == '{') {
1331 return TokenNameDOLLAR_LBRACE;
1333 if (isPHPIdentifierStart(currentCharacter)) {
1334 return scanIdentifierOrKeyword(true);
1336 currentPosition = oldPosition;
1337 return TokenNameSTRING;
1339 } catch (IndexOutOfBoundsException e) {
1340 currentPosition = oldPosition;
1341 return TokenNameSTRING;
1344 if (isPHPIdentifierStart(currentCharacter))
1345 return scanIdentifierOrKeyword(false);
1346 if (Character.isDigit(currentCharacter))
1347 return scanNumber(false);
1348 return TokenNameERROR;
1351 // boolean isWhiteSpace;
1353 while ((currentCharacter == ' ') || Character.isWhitespace(currentCharacter)) {
1354 startPosition = currentPosition;
1355 currentCharacter = source[currentPosition++];
1356 // if (((currentCharacter = source[currentPosition++]) == '\\')
1357 // && (source[currentPosition] == 'u')) {
1358 // isWhiteSpace = jumpOverUnicodeWhiteSpace();
1360 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1361 checkNonExternalizeString();
1362 if (recordLineSeparator) {
1363 pushLineSeparator();
1368 // isWhiteSpace = (currentCharacter == ' ')
1369 // || Character.isWhitespace(currentCharacter);
1372 if (tokenizeWhiteSpace && (whiteStart != currentPosition - 1)) {
1373 // reposition scanner in case we are interested by spaces as tokens
1375 startPosition = whiteStart;
1376 return TokenNameWHITESPACE;
1378 //little trick to get out in the middle of a source compuation
1379 if (currentPosition > eofPosition)
1380 return TokenNameEOF;
1381 // ---------Identify the next token-------------
1382 switch (currentCharacter) {
1384 return getCastOrParen();
1386 return TokenNameRPAREN;
1388 return TokenNameLBRACE;
1390 return TokenNameRBRACE;
1392 return TokenNameLBRACKET;
1394 return TokenNameRBRACKET;
1396 return TokenNameSEMICOLON;
1398 return TokenNameCOMMA;
1400 if (getNextChar('='))
1401 return TokenNameDOT_EQUAL;
1402 if (getNextCharAsDigit())
1403 return scanNumber(true);
1404 return TokenNameDOT;
1407 if ((test = getNextChar('+', '=')) == 0)
1408 return TokenNamePLUS_PLUS;
1410 return TokenNamePLUS_EQUAL;
1411 return TokenNamePLUS;
1415 if ((test = getNextChar('-', '=')) == 0)
1416 return TokenNameMINUS_MINUS;
1418 return TokenNameMINUS_EQUAL;
1419 if (getNextChar('>'))
1420 return TokenNameMINUS_GREATER;
1421 return TokenNameMINUS;
1424 if (getNextChar('='))
1425 return TokenNameTWIDDLE_EQUAL;
1426 return TokenNameTWIDDLE;
1428 if (getNextChar('=')) {
1429 if (getNextChar('=')) {
1430 return TokenNameNOT_EQUAL_EQUAL;
1432 return TokenNameNOT_EQUAL;
1434 return TokenNameNOT;
1436 if (getNextChar('='))
1437 return TokenNameMULTIPLY_EQUAL;
1438 return TokenNameMULTIPLY;
1440 if (getNextChar('='))
1441 return TokenNameREMAINDER_EQUAL;
1442 return TokenNameREMAINDER;
1444 int oldPosition = currentPosition;
1446 currentCharacter = source[currentPosition++];
1447 } catch (IndexOutOfBoundsException e) {
1448 currentPosition = oldPosition;
1449 return TokenNameLESS;
1451 switch (currentCharacter) {
1453 return TokenNameLESS_EQUAL;
1455 return TokenNameNOT_EQUAL;
1457 if (getNextChar('='))
1458 return TokenNameLEFT_SHIFT_EQUAL;
1459 if (getNextChar('<')) {
1460 currentCharacter = source[currentPosition++];
1461 while (Character.isWhitespace(currentCharacter)) {
1462 currentCharacter = source[currentPosition++];
1464 int heredocStart = currentPosition - 1;
1465 int heredocLength = 0;
1466 if (isPHPIdentifierStart(currentCharacter)) {
1467 currentCharacter = source[currentPosition++];
1469 return TokenNameERROR;
1471 while (isPHPIdentifierPart(currentCharacter)) {
1472 currentCharacter = source[currentPosition++];
1474 heredocLength = currentPosition - heredocStart - 1;
1475 // heredoc end-tag determination
1476 boolean endTag = true;
1479 ch = source[currentPosition++];
1480 if (ch == '\r' || ch == '\n') {
1481 if (recordLineSeparator) {
1482 pushLineSeparator();
1486 for (int i = 0; i < heredocLength; i++) {
1487 if (source[currentPosition + i] != source[heredocStart + i]) {
1493 currentPosition += heredocLength - 1;
1494 currentCharacter = source[currentPosition++];
1495 break; // do...while loop
1501 return TokenNameHEREDOC;
1503 return TokenNameLEFT_SHIFT;
1505 currentPosition = oldPosition;
1506 return TokenNameLESS;
1510 if ((test = getNextChar('=', '>')) == 0)
1511 return TokenNameGREATER_EQUAL;
1513 if ((test = getNextChar('=', '>')) == 0)
1514 return TokenNameRIGHT_SHIFT_EQUAL;
1515 return TokenNameRIGHT_SHIFT;
1517 return TokenNameGREATER;
1520 if (getNextChar('=')) {
1521 if (getNextChar('=')) {
1522 return TokenNameEQUAL_EQUAL_EQUAL;
1524 return TokenNameEQUAL_EQUAL;
1526 if (getNextChar('>'))
1527 return TokenNameEQUAL_GREATER;
1528 return TokenNameEQUAL;
1531 if ((test = getNextChar('&', '=')) == 0)
1532 return TokenNameAND_AND;
1534 return TokenNameAND_EQUAL;
1535 return TokenNameAND;
1539 if ((test = getNextChar('|', '=')) == 0)
1540 return TokenNameOR_OR;
1542 return TokenNameOR_EQUAL;
1546 if (getNextChar('='))
1547 return TokenNameXOR_EQUAL;
1548 return TokenNameXOR;
1550 if (getNextChar('>')) {
1552 if (currentPosition == source.length) {
1554 return TokenNameINLINE_HTML;
1556 return getInlinedHTML(currentPosition - 2);
1558 return TokenNameQUESTION;
1560 if (getNextChar(':'))
1561 return TokenNamePAAMAYIM_NEKUDOTAYIM;
1562 return TokenNameCOLON;
1566 consumeStringConstant();
1567 return TokenNameStringSingleQuote;
1569 if (tokenizeStrings) {
1570 consumeStringLiteral();
1571 return TokenNameStringDoubleQuote;
1573 return TokenNameEncapsedString2;
1575 if (tokenizeStrings) {
1576 consumeStringInterpolated();
1577 return TokenNameStringInterpolated;
1579 return TokenNameEncapsedString0;
1582 char startChar = currentCharacter;
1583 if (getNextChar('=') && startChar == '/') {
1584 return TokenNameDIVIDE_EQUAL;
1587 if ((startChar == '#') || (test = getNextChar('/', '*')) == 0) {
1589 this.lastCommentLinePosition = this.currentPosition;
1590 int endPositionForLineComment = 0;
1591 try { //get the next char
1592 currentCharacter = source[currentPosition++];
1593 // if (((currentCharacter = source[currentPosition++])
1595 // && (source[currentPosition] == 'u')) {
1596 // //-------------unicode traitement ------------
1597 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
1598 // currentPosition++;
1599 // while (source[currentPosition] == 'u') {
1600 // currentPosition++;
1603 // Character.getNumericValue(source[currentPosition++]))
1607 // Character.getNumericValue(source[currentPosition++]))
1611 // Character.getNumericValue(source[currentPosition++]))
1615 // Character.getNumericValue(source[currentPosition++]))
1619 // InvalidInputException(INVALID_UNICODE_ESCAPE);
1621 // currentCharacter =
1622 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
1625 //handle the \\u case manually into comment
1626 // if (currentCharacter == '\\') {
1627 // if (source[currentPosition] == '\\')
1628 // currentPosition++;
1629 // } //jump over the \\
1630 boolean isUnicode = false;
1631 while (currentCharacter != '\r' && currentCharacter != '\n') {
1632 this.lastCommentLinePosition = this.currentPosition;
1633 if (currentCharacter == '?') {
1634 if (getNextChar('>')) {
1635 startPosition = currentPosition - 2;
1637 return TokenNameINLINE_HTML;
1642 currentCharacter = source[currentPosition++];
1643 // if (((currentCharacter = source[currentPosition++])
1645 // && (source[currentPosition] == 'u')) {
1646 // isUnicode = true;
1647 // //-------------unicode traitement ------------
1648 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
1649 // currentPosition++;
1650 // while (source[currentPosition] == 'u') {
1651 // currentPosition++;
1654 // Character.getNumericValue(source[currentPosition++]))
1658 // Character.getNumericValue(
1659 // source[currentPosition++]))
1663 // Character.getNumericValue(
1664 // source[currentPosition++]))
1668 // Character.getNumericValue(
1669 // source[currentPosition++]))
1673 // InvalidInputException(INVALID_UNICODE_ESCAPE);
1675 // currentCharacter =
1676 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
1679 //handle the \\u case manually into comment
1680 // if (currentCharacter == '\\') {
1681 // if (source[currentPosition] == '\\')
1682 // currentPosition++;
1683 // } //jump over the \\
1686 endPositionForLineComment = currentPosition - 6;
1688 endPositionForLineComment = currentPosition - 1;
1690 // recordComment(false);
1691 recordComment(TokenNameCOMMENT_LINE);
1692 if (this.taskTags != null)
1693 checkTaskTag(this.startPosition, this.currentPosition);
1694 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1695 checkNonExternalizeString();
1696 if (recordLineSeparator) {
1698 pushUnicodeLineSeparator();
1700 pushLineSeparator();
1706 if (tokenizeComments) {
1708 currentPosition = endPositionForLineComment;
1709 // reset one character behind
1711 return TokenNameCOMMENT_LINE;
1713 } catch (IndexOutOfBoundsException e) { //an eof will them
1715 if (tokenizeComments) {
1717 // reset one character behind
1718 return TokenNameCOMMENT_LINE;
1724 //traditional and annotation comment
1725 boolean isJavadoc = false, star = false;
1726 // consume next character
1727 unicodeAsBackSlash = false;
1728 currentCharacter = source[currentPosition++];
1729 // if (((currentCharacter = source[currentPosition++]) ==
1731 // && (source[currentPosition] == 'u')) {
1732 // getNextUnicodeChar();
1734 // if (withoutUnicodePtr != 0) {
1735 // withoutUnicodeBuffer[++withoutUnicodePtr] =
1736 // currentCharacter;
1739 if (currentCharacter == '*') {
1743 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1744 checkNonExternalizeString();
1745 if (recordLineSeparator) {
1746 pushLineSeparator();
1751 try { //get the next char
1752 currentCharacter = source[currentPosition++];
1753 // if (((currentCharacter = source[currentPosition++])
1755 // && (source[currentPosition] == 'u')) {
1756 // //-------------unicode traitement ------------
1757 // getNextUnicodeChar();
1759 //handle the \\u case manually into comment
1760 // if (currentCharacter == '\\') {
1761 // if (source[currentPosition] == '\\')
1762 // currentPosition++;
1763 // //jump over the \\
1765 // empty comment is not a javadoc /**/
1766 if (currentCharacter == '/') {
1769 //loop until end of comment */
1770 while ((currentCharacter != '/') || (!star)) {
1771 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1772 checkNonExternalizeString();
1773 if (recordLineSeparator) {
1774 pushLineSeparator();
1779 star = currentCharacter == '*';
1781 currentCharacter = source[currentPosition++];
1782 // if (((currentCharacter = source[currentPosition++])
1784 // && (source[currentPosition] == 'u')) {
1785 // //-------------unicode traitement ------------
1786 // getNextUnicodeChar();
1788 //handle the \\u case manually into comment
1789 // if (currentCharacter == '\\') {
1790 // if (source[currentPosition] == '\\')
1791 // currentPosition++;
1792 // } //jump over the \\
1794 //recordComment(isJavadoc);
1796 recordComment(TokenNameCOMMENT_PHPDOC);
1798 recordComment(TokenNameCOMMENT_BLOCK);
1801 if (tokenizeComments) {
1803 return TokenNameCOMMENT_PHPDOC;
1804 return TokenNameCOMMENT_BLOCK;
1807 if (this.taskTags != null) {
1808 checkTaskTag(this.startPosition, this.currentPosition);
1810 } catch (IndexOutOfBoundsException e) {
1811 // reset end position for error reporting
1812 currentPosition -= 2;
1813 throw new InvalidInputException(UNTERMINATED_COMMENT);
1817 return TokenNameDIVIDE;
1821 return TokenNameEOF;
1822 //the atEnd may not be <currentPosition == source.length> if
1823 // source is only some part of a real (external) stream
1824 throw new InvalidInputException("Ctrl-Z"); //$NON-NLS-1$
1826 if (currentCharacter == '$') {
1827 int oldPosition = currentPosition;
1829 currentCharacter = source[currentPosition++];
1830 if (isPHPIdentifierStart(currentCharacter)) {
1831 return scanIdentifierOrKeyword(true);
1833 currentPosition = oldPosition;
1834 return TokenNameDOLLAR;
1836 } catch (IndexOutOfBoundsException e) {
1837 currentPosition = oldPosition;
1838 return TokenNameDOLLAR;
1841 if (isPHPIdentifierStart(currentCharacter))
1842 return scanIdentifierOrKeyword(false);
1843 if (Character.isDigit(currentCharacter))
1844 return scanNumber(false);
1845 return TokenNameERROR;
1848 } //-----------------end switch while try--------------------
1849 catch (IndexOutOfBoundsException e) {
1852 return TokenNameEOF;
1855 private int getInlinedHTML(int start) throws InvalidInputException {
1856 int token = getInlinedHTMLToken(start);
1857 if (token == TokenNameINLINE_HTML) {
1858 // Stack stack = new Stack();
1859 // // scan html for errors
1860 // Source inlinedHTMLSource = new Source(new String(source, startPosition, currentPosition - startPosition));
1861 // int lastPHPEndPos=0;
1862 // for (Iterator i=inlinedHTMLSource.getNextTagIterator(0); i.hasNext();) {
1863 // Tag tag=(Tag)i.next();
1865 // if (tag instanceof StartTag) {
1866 // StartTag startTag=(StartTag)tag;
1867 // // System.out.println("startTag: "+tag);
1868 // if (startTag.isServerTag()) {
1869 // // TODO : what to do with a server tag ?
1871 // // do whatever with HTML start tag
1872 // // use startTag.getElement() to find the element corresponding
1873 // // to this start tag which may be useful if you implement code
1875 // stack.push(startTag);
1878 // EndTag endTag=(EndTag)tag;
1879 // StartTag stag = (StartTag) stack.peek();
1880 //// System.out.println("endTag: "+tag);
1881 // // do whatever with HTML end tag.
1890 * @throws InvalidInputException
1892 private int getInlinedHTMLToken(int start) throws InvalidInputException {
1893 if (currentPosition > source.length) {
1894 currentPosition = source.length;
1895 return TokenNameEOF;
1897 startPosition = start;
1900 currentCharacter = source[currentPosition++];
1901 if (currentCharacter == '<') {
1902 if (getNextChar('?')) {
1903 currentCharacter = source[currentPosition++];
1904 if ((currentCharacter != 'P') && (currentCharacter != 'p')) {
1906 // (currentCharacter == ' ') || Character.isWhitespace(currentCharacter)) {
1908 if (ignorePHPOneLiner) {
1909 if (lookAheadLinePHPTag() == TokenNameINLINE_HTML) {
1911 return TokenNameINLINE_HTML;
1915 return TokenNameINLINE_HTML;
1918 // boolean phpStart = (currentCharacter == 'P') || (currentCharacter == 'p');
1920 int test = getNextChar('H', 'h');
1922 test = getNextChar('P', 'p');
1925 if (ignorePHPOneLiner) {
1926 if (lookAheadLinePHPTag() == TokenNameINLINE_HTML) {
1928 return TokenNameINLINE_HTML;
1932 return TokenNameINLINE_HTML;
1940 if ((currentCharacter == '\r') || (currentCharacter == '\n')) {
1941 if (recordLineSeparator) {
1942 pushLineSeparator();
1947 } //-----------------while--------------------
1949 return TokenNameINLINE_HTML;
1950 } //-----------------try--------------------
1951 catch (IndexOutOfBoundsException e) {
1952 startPosition = start;
1956 return TokenNameINLINE_HTML;
1962 private int lookAheadLinePHPTag() {
1963 // check if the PHP is only in this line (for CodeFormatter)
1964 int currentPositionInLine = currentPosition;
1965 char previousCharInLine = ' ';
1966 char currentCharInLine = ' ';
1967 boolean singleQuotedStringActive = false;
1968 boolean doubleQuotedStringActive = false;
1971 // look ahead in this line
1973 previousCharInLine = currentCharInLine;
1974 currentCharInLine = source[currentPositionInLine++];
1975 switch (currentCharInLine) {
1977 if (previousCharInLine == '?') {
1978 // update the scanner's current Position in the source
1979 currentPosition = currentPositionInLine;
1980 // use as "dummy" token
1981 return TokenNameEOF;
1985 if (doubleQuotedStringActive) {
1986 if (previousCharInLine != '\\') {
1987 doubleQuotedStringActive = false;
1990 if (!singleQuotedStringActive) {
1991 doubleQuotedStringActive = true;
1996 if (singleQuotedStringActive) {
1997 if (previousCharInLine != '\\') {
1998 singleQuotedStringActive = false;
2001 if (!doubleQuotedStringActive) {
2002 singleQuotedStringActive = true;
2008 return TokenNameINLINE_HTML;
2010 if (!singleQuotedStringActive && !doubleQuotedStringActive) {
2012 return TokenNameINLINE_HTML;
2016 if (previousCharInLine == '/' && !singleQuotedStringActive && !doubleQuotedStringActive) {
2018 return TokenNameINLINE_HTML;
2022 if (previousCharInLine == '/' && !singleQuotedStringActive && !doubleQuotedStringActive) {
2024 return TokenNameINLINE_HTML;
2029 } catch (IndexOutOfBoundsException e) {
2031 currentPosition = currentPositionInLine;
2032 return TokenNameINLINE_HTML;
2036 // public final void getNextUnicodeChar()
2037 // throws IndexOutOfBoundsException, InvalidInputException {
2039 // //handle the case of unicode.
2040 // //when a unicode appears then we must use a buffer that holds char
2042 // //At the end of this method currentCharacter holds the new visited char
2043 // //and currentPosition points right next after it
2045 // //ALL getNextChar.... ARE OPTIMIZED COPIES
2047 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0, unicodeSize = 6;
2048 // currentPosition++;
2049 // while (source[currentPosition] == 'u') {
2050 // currentPosition++;
2054 // if ((c1 = Character.getNumericValue(source[currentPosition++])) > 15
2056 // || (c2 = Character.getNumericValue(source[currentPosition++])) > 15
2058 // || (c3 = Character.getNumericValue(source[currentPosition++])) > 15
2060 // || (c4 = Character.getNumericValue(source[currentPosition++])) > 15
2062 // throw new InvalidInputException(INVALID_UNICODE_ESCAPE);
2064 // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2065 // //need the unicode buffer
2066 // if (withoutUnicodePtr == 0) {
2067 // //buffer all the entries that have been left aside....
2068 // withoutUnicodePtr = currentPosition - unicodeSize - startPosition;
2069 // System.arraycopy(
2072 // withoutUnicodeBuffer,
2074 // withoutUnicodePtr);
2076 // //fill the buffer with the char
2077 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2079 // unicodeAsBackSlash = currentCharacter == '\\';
2082 * Tokenize a method body, assuming that curly brackets are properly balanced.
2084 public final void jumpOverMethodBody() {
2085 this.wasAcr = false;
2088 while (true) { //loop for jumping over comments
2089 // ---------Consume white space and handles startPosition---------
2090 boolean isWhiteSpace;
2092 startPosition = currentPosition;
2093 currentCharacter = source[currentPosition++];
2094 // if (((currentCharacter = source[currentPosition++]) == '\\')
2095 // && (source[currentPosition] == 'u')) {
2096 // isWhiteSpace = jumpOverUnicodeWhiteSpace();
2098 if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2099 pushLineSeparator();
2100 isWhiteSpace = Character.isWhitespace(currentCharacter);
2102 } while (isWhiteSpace);
2103 // -------consume token until } is found---------
2104 switch (currentCharacter) {
2115 test = getNextChar('\\');
2118 scanDoubleQuotedEscapeCharacter();
2119 } catch (InvalidInputException ex) {
2123 // try { // consume next character
2124 unicodeAsBackSlash = false;
2125 currentCharacter = source[currentPosition++];
2126 // if (((currentCharacter = source[currentPosition++]) == '\\')
2127 // && (source[currentPosition] == 'u')) {
2128 // getNextUnicodeChar();
2130 if (withoutUnicodePtr != 0) {
2131 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2134 // } catch (InvalidInputException ex) {
2142 // try { // consume next character
2143 unicodeAsBackSlash = false;
2144 currentCharacter = source[currentPosition++];
2145 // if (((currentCharacter = source[currentPosition++]) == '\\')
2146 // && (source[currentPosition] == 'u')) {
2147 // getNextUnicodeChar();
2149 if (withoutUnicodePtr != 0) {
2150 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2153 // } catch (InvalidInputException ex) {
2155 while (currentCharacter != '"') {
2156 if (currentCharacter == '\r') {
2157 if (source[currentPosition] == '\n')
2160 // the string cannot go further that the line
2162 if (currentCharacter == '\n') {
2164 // the string cannot go further that the line
2166 if (currentCharacter == '\\') {
2168 scanDoubleQuotedEscapeCharacter();
2169 } catch (InvalidInputException ex) {
2173 // try { // consume next character
2174 unicodeAsBackSlash = false;
2175 currentCharacter = source[currentPosition++];
2176 // if (((currentCharacter = source[currentPosition++]) == '\\')
2177 // && (source[currentPosition] == 'u')) {
2178 // getNextUnicodeChar();
2180 if (withoutUnicodePtr != 0) {
2181 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2184 // } catch (InvalidInputException ex) {
2187 } catch (IndexOutOfBoundsException e) {
2193 if ((test = getNextChar('/', '*')) == 0) {
2197 currentCharacter = source[currentPosition++];
2198 // if (((currentCharacter = source[currentPosition++]) ==
2200 // && (source[currentPosition] == 'u')) {
2201 // //-------------unicode traitement ------------
2202 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2203 // currentPosition++;
2204 // while (source[currentPosition] == 'u') {
2205 // currentPosition++;
2208 // Character.getNumericValue(source[currentPosition++]))
2212 // Character.getNumericValue(source[currentPosition++]))
2216 // Character.getNumericValue(source[currentPosition++]))
2220 // Character.getNumericValue(source[currentPosition++]))
2223 // //error don't care of the value
2224 // currentCharacter = 'A';
2225 // } //something different from \n and \r
2227 // currentCharacter =
2228 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2231 while (currentCharacter != '\r' && currentCharacter != '\n') {
2233 currentCharacter = source[currentPosition++];
2234 // if (((currentCharacter = source[currentPosition++])
2236 // && (source[currentPosition] == 'u')) {
2237 // //-------------unicode traitement ------------
2238 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2239 // currentPosition++;
2240 // while (source[currentPosition] == 'u') {
2241 // currentPosition++;
2244 // Character.getNumericValue(source[currentPosition++]))
2248 // Character.getNumericValue(source[currentPosition++]))
2252 // Character.getNumericValue(source[currentPosition++]))
2256 // Character.getNumericValue(source[currentPosition++]))
2259 // //error don't care of the value
2260 // currentCharacter = 'A';
2261 // } //something different from \n and \r
2263 // currentCharacter =
2264 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2268 if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2269 pushLineSeparator();
2270 } catch (IndexOutOfBoundsException e) {
2271 } //an eof will them be generated
2275 //traditional and annotation comment
2276 boolean star = false;
2277 // try { // consume next character
2278 unicodeAsBackSlash = false;
2279 currentCharacter = source[currentPosition++];
2280 // if (((currentCharacter = source[currentPosition++]) == '\\')
2281 // && (source[currentPosition] == 'u')) {
2282 // getNextUnicodeChar();
2284 if (withoutUnicodePtr != 0) {
2285 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2288 // } catch (InvalidInputException ex) {
2290 if (currentCharacter == '*') {
2293 if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2294 pushLineSeparator();
2295 try { //get the next char
2296 currentCharacter = source[currentPosition++];
2297 // if (((currentCharacter = source[currentPosition++]) ==
2299 // && (source[currentPosition] == 'u')) {
2300 // //-------------unicode traitement ------------
2301 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2302 // currentPosition++;
2303 // while (source[currentPosition] == 'u') {
2304 // currentPosition++;
2307 // Character.getNumericValue(source[currentPosition++]))
2311 // Character.getNumericValue(source[currentPosition++]))
2315 // Character.getNumericValue(source[currentPosition++]))
2319 // Character.getNumericValue(source[currentPosition++]))
2322 // //error don't care of the value
2323 // currentCharacter = 'A';
2324 // } //something different from * and /
2326 // currentCharacter =
2327 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2330 //loop until end of comment */
2331 while ((currentCharacter != '/') || (!star)) {
2332 if (recordLineSeparator && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2333 pushLineSeparator();
2334 star = currentCharacter == '*';
2336 currentCharacter = source[currentPosition++];
2337 // if (((currentCharacter = source[currentPosition++])
2339 // && (source[currentPosition] == 'u')) {
2340 // //-------------unicode traitement ------------
2341 // int c1 = 0, c2 = 0, c3 = 0, c4 = 0;
2342 // currentPosition++;
2343 // while (source[currentPosition] == 'u') {
2344 // currentPosition++;
2347 // Character.getNumericValue(source[currentPosition++]))
2351 // Character.getNumericValue(source[currentPosition++]))
2355 // Character.getNumericValue(source[currentPosition++]))
2359 // Character.getNumericValue(source[currentPosition++]))
2362 // //error don't care of the value
2363 // currentCharacter = 'A';
2364 // } //something different from * and /
2366 // currentCharacter =
2367 // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2371 } catch (IndexOutOfBoundsException e) {
2379 if (isPHPIdentifierStart(currentCharacter) || currentCharacter == '$') {
2381 scanIdentifierOrKeyword((currentCharacter == '$'));
2382 } catch (InvalidInputException ex) {
2387 if (Character.isDigit(currentCharacter)) {
2390 } catch (InvalidInputException ex) {
2397 //-----------------end switch while try--------------------
2398 } catch (IndexOutOfBoundsException e) {
2399 } catch (InvalidInputException e) {
2404 // public final boolean jumpOverUnicodeWhiteSpace()
2405 // throws InvalidInputException {
2407 // //handle the case of unicode. Jump over the next whiteSpace
2408 // //making startPosition pointing on the next available char
2409 // //On false, the currentCharacter is filled up with a potential
2413 // this.wasAcr = false;
2414 // int c1, c2, c3, c4;
2415 // int unicodeSize = 6;
2416 // currentPosition++;
2417 // while (source[currentPosition] == 'u') {
2418 // currentPosition++;
2422 // if (((c1 = Character.getNumericValue(source[currentPosition++])) > 15
2424 // || ((c2 = Character.getNumericValue(source[currentPosition++])) > 15
2426 // || ((c3 = Character.getNumericValue(source[currentPosition++])) > 15
2428 // || ((c4 = Character.getNumericValue(source[currentPosition++])) > 15
2430 // throw new InvalidInputException(INVALID_UNICODE_ESCAPE);
2433 // currentCharacter = (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4);
2434 // if (recordLineSeparator
2435 // && ((currentCharacter == '\r') || (currentCharacter == '\n')))
2436 // pushLineSeparator();
2437 // if (Character.isWhitespace(currentCharacter))
2440 // //buffer the new char which is not a white space
2441 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2442 // //withoutUnicodePtr == 1 is true here
2444 // } catch (IndexOutOfBoundsException e) {
2445 // throw new InvalidInputException(INVALID_UNICODE_ESCAPE);
2448 public final int[] getLineEnds() {
2449 //return a bounded copy of this.lineEnds
2451 System.arraycopy(lineEnds, 0, copy = new int[linePtr + 1], 0, linePtr + 1);
2455 public char[] getSource() {
2459 public static boolean isIdentifierOrKeyword(int token) {
2460 return (token == TokenNameIdentifier) || (token > TokenNameKEYWORD);
2463 final char[] optimizedCurrentTokenSource1() {
2464 //return always the same char[] build only once
2465 //optimization at no speed cost of 99.5 % of the singleCharIdentifier
2466 char charOne = source[startPosition];
2521 return new char[] { charOne };
2525 final char[] optimizedCurrentTokenSource2() {
2527 c0 = source[startPosition];
2528 c1 = source[startPosition + 1];
2530 //return always the same char[] build only once
2531 //optimization at no speed cost of 99.5 % of the singleCharIdentifier
2534 return charArray_va;
2536 return charArray_vb;
2538 return charArray_vc;
2540 return charArray_vd;
2542 return charArray_ve;
2544 return charArray_vf;
2546 return charArray_vg;
2548 return charArray_vh;
2550 return charArray_vi;
2552 return charArray_vj;
2554 return charArray_vk;
2556 return charArray_vl;
2558 return charArray_vm;
2560 return charArray_vn;
2562 return charArray_vo;
2564 return charArray_vp;
2566 return charArray_vq;
2568 return charArray_vr;
2570 return charArray_vs;
2572 return charArray_vt;
2574 return charArray_vu;
2576 return charArray_vv;
2578 return charArray_vw;
2580 return charArray_vx;
2582 return charArray_vy;
2584 return charArray_vz;
2587 //try to return the same char[] build only once
2588 int hash = ((c0 << 6) + c1) % TableSize;
2589 char[][] table = charArray_length[0][hash];
2591 while (++i < InternalTableSize) {
2592 char[] charArray = table[i];
2593 if ((c0 == charArray[0]) && (c1 == charArray[1]))
2596 //---------other side---------
2598 int max = newEntry2;
2599 while (++i <= max) {
2600 char[] charArray = table[i];
2601 if ((c0 == charArray[0]) && (c1 == charArray[1]))
2604 //--------add the entry-------
2605 if (++max >= InternalTableSize)
2608 table[max] = (r = new char[] { c0, c1 });
2613 final char[] optimizedCurrentTokenSource3() {
2614 //try to return the same char[] build only once
2616 int hash = (((c0 = source[startPosition]) << 12) + ((c1 = source[startPosition + 1]) << 6) + (c2 = source[startPosition + 2]))
2618 char[][] table = charArray_length[1][hash];
2620 while (++i < InternalTableSize) {
2621 char[] charArray = table[i];
2622 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]))
2625 //---------other side---------
2627 int max = newEntry3;
2628 while (++i <= max) {
2629 char[] charArray = table[i];
2630 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]))
2633 //--------add the entry-------
2634 if (++max >= InternalTableSize)
2637 table[max] = (r = new char[] { c0, c1, c2 });
2642 final char[] optimizedCurrentTokenSource4() {
2643 //try to return the same char[] build only once
2644 char c0, c1, c2, c3;
2645 long hash = ((((long) (c0 = source[startPosition])) << 18) + ((c1 = source[startPosition + 1]) << 12)
2646 + ((c2 = source[startPosition + 2]) << 6) + (c3 = source[startPosition + 3]))
2648 char[][] table = charArray_length[2][(int) hash];
2650 while (++i < InternalTableSize) {
2651 char[] charArray = table[i];
2652 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]))
2655 //---------other side---------
2657 int max = newEntry4;
2658 while (++i <= max) {
2659 char[] charArray = table[i];
2660 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]))
2663 //--------add the entry-------
2664 if (++max >= InternalTableSize)
2667 table[max] = (r = new char[] { c0, c1, c2, c3 });
2672 final char[] optimizedCurrentTokenSource5() {
2673 //try to return the same char[] build only once
2674 char c0, c1, c2, c3, c4;
2675 long hash = ((((long) (c0 = source[startPosition])) << 24) + (((long) (c1 = source[startPosition + 1])) << 18)
2676 + ((c2 = source[startPosition + 2]) << 12) + ((c3 = source[startPosition + 3]) << 6) + (c4 = source[startPosition + 4]))
2678 char[][] table = charArray_length[3][(int) hash];
2680 while (++i < InternalTableSize) {
2681 char[] charArray = table[i];
2682 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4]))
2685 //---------other side---------
2687 int max = newEntry5;
2688 while (++i <= max) {
2689 char[] charArray = table[i];
2690 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4]))
2693 //--------add the entry-------
2694 if (++max >= InternalTableSize)
2697 table[max] = (r = new char[] { c0, c1, c2, c3, c4 });
2702 final char[] optimizedCurrentTokenSource6() {
2703 //try to return the same char[] build only once
2704 char c0, c1, c2, c3, c4, c5;
2705 long hash = ((((long) (c0 = source[startPosition])) << 32) + (((long) (c1 = source[startPosition + 1])) << 24)
2706 + (((long) (c2 = source[startPosition + 2])) << 18) + ((c3 = source[startPosition + 3]) << 12)
2707 + ((c4 = source[startPosition + 4]) << 6) + (c5 = source[startPosition + 5]))
2709 char[][] table = charArray_length[4][(int) hash];
2711 while (++i < InternalTableSize) {
2712 char[] charArray = table[i];
2713 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4])
2714 && (c5 == charArray[5]))
2717 //---------other side---------
2719 int max = newEntry6;
2720 while (++i <= max) {
2721 char[] charArray = table[i];
2722 if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4])
2723 && (c5 == charArray[5]))
2726 //--------add the entry-------
2727 if (++max >= InternalTableSize)
2730 table[max] = (r = new char[] { c0, c1, c2, c3, c4, c5 });
2735 public final void pushLineSeparator() throws InvalidInputException {
2736 //see comment on isLineDelimiter(char) for the use of '\n' and '\r'
2737 final int INCREMENT = 250;
2738 if (this.checkNonExternalizedStringLiterals) {
2739 // reinitialize the current line for non externalize strings purpose
2742 //currentCharacter is at position currentPosition-1
2744 if (currentCharacter == '\r') {
2745 int separatorPos = currentPosition - 1;
2746 if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2748 //System.out.println("CR-" + separatorPos);
2750 lineEnds[++linePtr] = separatorPos;
2751 } catch (IndexOutOfBoundsException e) {
2752 //linePtr value is correct
2753 int oldLength = lineEnds.length;
2754 int[] old = lineEnds;
2755 lineEnds = new int[oldLength + INCREMENT];
2756 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2757 lineEnds[linePtr] = separatorPos;
2759 // look-ahead for merged cr+lf
2761 if (source[currentPosition] == '\n') {
2762 //System.out.println("look-ahead LF-" + currentPosition);
2763 lineEnds[linePtr] = currentPosition;
2769 } catch (IndexOutOfBoundsException e) {
2774 if (currentCharacter == '\n') {
2775 //must merge eventual cr followed by lf
2776 if (wasAcr && (lineEnds[linePtr] == (currentPosition - 2))) {
2777 //System.out.println("merge LF-" + (currentPosition - 1));
2778 lineEnds[linePtr] = currentPosition - 1;
2780 int separatorPos = currentPosition - 1;
2781 if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2783 // System.out.println("LF-" + separatorPos);
2785 lineEnds[++linePtr] = separatorPos;
2786 } catch (IndexOutOfBoundsException e) {
2787 //linePtr value is correct
2788 int oldLength = lineEnds.length;
2789 int[] old = lineEnds;
2790 lineEnds = new int[oldLength + INCREMENT];
2791 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2792 lineEnds[linePtr] = separatorPos;
2800 public final void pushUnicodeLineSeparator() {
2801 // isUnicode means that the \r or \n has been read as a unicode character
2802 //see comment on isLineDelimiter(char) for the use of '\n' and '\r'
2803 final int INCREMENT = 250;
2804 //currentCharacter is at position currentPosition-1
2805 if (this.checkNonExternalizedStringLiterals) {
2806 // reinitialize the current line for non externalize strings purpose
2810 if (currentCharacter == '\r') {
2811 int separatorPos = currentPosition - 6;
2812 if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2814 //System.out.println("CR-" + separatorPos);
2816 lineEnds[++linePtr] = separatorPos;
2817 } catch (IndexOutOfBoundsException e) {
2818 //linePtr value is correct
2819 int oldLength = lineEnds.length;
2820 int[] old = lineEnds;
2821 lineEnds = new int[oldLength + INCREMENT];
2822 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2823 lineEnds[linePtr] = separatorPos;
2825 // look-ahead for merged cr+lf
2826 if (source[currentPosition] == '\n') {
2827 //System.out.println("look-ahead LF-" + currentPosition);
2828 lineEnds[linePtr] = currentPosition;
2836 if (currentCharacter == '\n') {
2837 //must merge eventual cr followed by lf
2838 if (wasAcr && (lineEnds[linePtr] == (currentPosition - 7))) {
2839 //System.out.println("merge LF-" + (currentPosition - 1));
2840 lineEnds[linePtr] = currentPosition - 6;
2842 int separatorPos = currentPosition - 6;
2843 if ((linePtr > 0) && (lineEnds[linePtr] >= separatorPos))
2845 // System.out.println("LF-" + separatorPos);
2847 lineEnds[++linePtr] = separatorPos;
2848 } catch (IndexOutOfBoundsException e) {
2849 //linePtr value is correct
2850 int oldLength = lineEnds.length;
2851 int[] old = lineEnds;
2852 lineEnds = new int[oldLength + INCREMENT];
2853 System.arraycopy(old, 0, lineEnds, 0, oldLength);
2854 lineEnds[linePtr] = separatorPos;
2862 public void recordComment(int token) {
2864 int stopPosition = this.currentPosition;
2866 case TokenNameCOMMENT_LINE:
2867 stopPosition = -this.lastCommentLinePosition;
2869 case TokenNameCOMMENT_BLOCK:
2870 stopPosition = -this.currentPosition;
2874 // a new comment is recorded
2875 int length = this.commentStops.length;
2876 if (++this.commentPtr >= length) {
2877 System.arraycopy(this.commentStops, 0, this.commentStops = new int[length + 30], 0, length);
2878 //grows the positions buffers too
2879 System.arraycopy(this.commentStarts, 0, this.commentStarts = new int[length + 30], 0, length);
2881 this.commentStops[this.commentPtr] = stopPosition;
2882 this.commentStarts[this.commentPtr] = this.startPosition;
2885 // public final void recordComment(boolean isJavadoc) {
2886 // // a new annotation comment is recorded
2888 // commentStops[++commentPtr] = isJavadoc
2889 // ? currentPosition
2890 // : -currentPosition;
2891 // } catch (IndexOutOfBoundsException e) {
2892 // int oldStackLength = commentStops.length;
2893 // int[] oldStack = commentStops;
2894 // commentStops = new int[oldStackLength + 30];
2895 // System.arraycopy(oldStack, 0, commentStops, 0, oldStackLength);
2896 // commentStops[commentPtr] = isJavadoc ? currentPosition : -currentPosition;
2897 // //grows the positions buffers too
2898 // int[] old = commentStarts;
2899 // commentStarts = new int[oldStackLength + 30];
2900 // System.arraycopy(old, 0, commentStarts, 0, oldStackLength);
2902 // //the buffer is of a correct size here
2903 // commentStarts[commentPtr] = startPosition;
2905 public void resetTo(int begin, int end) {
2906 //reset the scanner to a given position where it may rescan again
2908 initialPosition = startPosition = currentPosition = begin;
2909 eofPosition = end < Integer.MAX_VALUE ? end + 1 : end;
2910 commentPtr = -1; // reset comment stack
2913 public final void scanSingleQuotedEscapeCharacter() throws InvalidInputException {
2914 // the string with "\\u" is a legal string of two chars \ and u
2915 //thus we use a direct access to the source (for regular cases).
2916 // if (unicodeAsBackSlash) {
2917 // // consume next character
2918 // unicodeAsBackSlash = false;
2919 // if (((currentCharacter = source[currentPosition++]) == '\\')
2920 // && (source[currentPosition] == 'u')) {
2921 // getNextUnicodeChar();
2923 // if (withoutUnicodePtr != 0) {
2924 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2928 currentCharacter = source[currentPosition++];
2929 switch (currentCharacter) {
2931 currentCharacter = '\'';
2934 currentCharacter = '\\';
2937 currentCharacter = '\\';
2942 public final void scanDoubleQuotedEscapeCharacter() throws InvalidInputException {
2943 // the string with "\\u" is a legal string of two chars \ and u
2944 //thus we use a direct access to the source (for regular cases).
2945 // if (unicodeAsBackSlash) {
2946 // // consume next character
2947 // unicodeAsBackSlash = false;
2948 // if (((currentCharacter = source[currentPosition++]) == '\\')
2949 // && (source[currentPosition] == 'u')) {
2950 // getNextUnicodeChar();
2952 // if (withoutUnicodePtr != 0) {
2953 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
2957 currentCharacter = source[currentPosition++];
2958 switch (currentCharacter) {
2960 // currentCharacter = '\b';
2963 currentCharacter = '\t';
2966 currentCharacter = '\n';
2969 // currentCharacter = '\f';
2972 currentCharacter = '\r';
2975 currentCharacter = '\"';
2978 currentCharacter = '\'';
2981 currentCharacter = '\\';
2984 currentCharacter = '$';
2987 // -----------octal escape--------------
2989 // OctalDigit OctalDigit
2990 // ZeroToThree OctalDigit OctalDigit
2991 int number = Character.getNumericValue(currentCharacter);
2992 if (number >= 0 && number <= 7) {
2993 boolean zeroToThreeNot = number > 3;
2994 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
2995 int digit = Character.getNumericValue(currentCharacter);
2996 if (digit >= 0 && digit <= 7) {
2997 number = (number * 8) + digit;
2998 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
2999 if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit
3000 // Digit --> ignore last character
3003 digit = Character.getNumericValue(currentCharacter);
3004 if (digit >= 0 && digit <= 7) {
3005 // has read \ZeroToThree OctalDigit OctalDigit
3006 number = (number * 8) + digit;
3007 } else { // has read \ZeroToThree OctalDigit NonOctalDigit
3008 // --> ignore last character
3012 } else { // has read \OctalDigit NonDigit--> ignore last
3016 } else { // has read \OctalDigit NonOctalDigit--> ignore last
3020 } else { // has read \OctalDigit --> ignore last character
3024 throw new InvalidInputException(INVALID_ESCAPE);
3025 currentCharacter = (char) number;
3028 // throw new InvalidInputException(INVALID_ESCAPE);
3032 // public int scanIdentifierOrKeyword() throws InvalidInputException {
3033 // return scanIdentifierOrKeyword( false );
3035 public int scanIdentifierOrKeyword(boolean isVariable) throws InvalidInputException {
3037 //first dispatch on the first char.
3038 //then the length. If there are several
3039 //keywors with the same length AND the same first char, then do another
3040 //disptach on the second char :-)...cool....but fast !
3041 useAssertAsAnIndentifier = false;
3042 while (getNextCharAsJavaIdentifierPart()) {
3046 // if (new String(getCurrentTokenSource()).equals("$this")) {
3047 // return TokenNamethis;
3049 return TokenNameVariable;
3054 // if (withoutUnicodePtr == 0)
3055 //quick test on length == 1 but not on length > 12 while most identifier
3056 //have a length which is <= 12...but there are lots of identifier with
3059 if ((length = currentPosition - startPosition) == 1)
3060 return TokenNameIdentifier;
3062 data = new char[length];
3063 index = startPosition;
3064 for (int i = 0; i < length; i++) {
3065 data[i] = Character.toLowerCase(source[index + i]);
3069 // if ((length = withoutUnicodePtr) == 1)
3070 // return TokenNameIdentifier;
3071 // // data = withoutUnicodeBuffer;
3072 // data = new char[withoutUnicodeBuffer.length];
3073 // for (int i = 0; i < withoutUnicodeBuffer.length; i++) {
3074 // data[i] = Character.toLowerCase(withoutUnicodeBuffer[i]);
3078 firstLetter = data[index];
3079 switch (firstLetter) {
3084 if ((data[++index] == '_') && (data[++index] == 'f') && (data[++index] == 'i') && (data[++index] == 'l')
3085 && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == '_'))
3086 return TokenNameFILE;
3087 index = 0; //__LINE__
3088 if ((data[++index] == '_') && (data[++index] == 'l') && (data[++index] == 'i') && (data[++index] == 'n')
3089 && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == '_'))
3090 return TokenNameLINE;
3094 if ((data[++index] == '_') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'a')
3095 && (data[++index] == 's') && (data[++index] == 's') && (data[++index] == '_') && (data[++index] == '_'))
3096 return TokenNameCLASS_C;
3100 if ((data[++index] == '_') && (data[++index] == 'm') && (data[++index] == 'e') && (data[++index] == 't')
3101 && (data[++index] == 'h') && (data[++index] == 'o') && (data[++index] == 'd') && (data[++index] == '_')
3102 && (data[++index] == '_'))
3103 return TokenNameMETHOD_C;
3107 if ((data[++index] == '_') && (data[++index] == 'f') && (data[++index] == 'u') && (data[++index] == 'n')
3108 && (data[++index] == 'c') && (data[++index] == 't') && (data[++index] == 'i') && (data[++index] == 'o')
3109 && (data[++index] == 'n') && (data[++index] == '_') && (data[++index] == '_'))
3110 return TokenNameFUNC_C;
3113 return TokenNameIdentifier;
3115 // as and array abstract
3119 if ((data[++index] == 's')) {
3122 return TokenNameIdentifier;
3126 if ((data[++index] == 'n') && (data[++index] == 'd')) {
3127 return TokenNameand;
3129 return TokenNameIdentifier;
3133 if ((data[++index] == 'r') && (data[++index] == 'r') && (data[++index] == 'a') && (data[++index] == 'y'))
3134 return TokenNamearray;
3136 return TokenNameIdentifier;
3138 if ((data[++index] == 'b') && (data[++index] == 's') && (data[++index] == 't') && (data[++index] == 'r')
3139 && (data[++index] == 'a') && (data[++index] == 'c') && (data[++index] == 't'))
3140 return TokenNameabstract;
3142 return TokenNameIdentifier;
3144 return TokenNameIdentifier;
3150 if ((data[++index] == 'r') && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'k'))
3151 return TokenNamebreak;
3153 return TokenNameIdentifier;
3155 return TokenNameIdentifier;
3158 //case catch class clone const continue
3161 if ((data[++index] == 'a') && (data[++index] == 's') && (data[++index] == 'e'))
3162 return TokenNamecase;
3164 return TokenNameIdentifier;
3166 if ((data[++index] == 'a') && (data[++index] == 't') && (data[++index] == 'c') && (data[++index] == 'h'))
3167 return TokenNamecatch;
3169 if ((data[++index] == 'l') && (data[++index] == 'a') && (data[++index] == 's') && (data[++index] == 's'))
3170 return TokenNameclass;
3172 if ((data[++index] == 'l') && (data[++index] == 'o') && (data[++index] == 'n') && (data[++index] == 'e'))
3173 return TokenNameclone;
3175 if ((data[++index] == 'o') && (data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 't'))
3176 return TokenNameconst;
3178 return TokenNameIdentifier;
3180 if ((data[++index] == 'o') && (data[++index] == 'n') && (data[++index] == 't') && (data[++index] == 'i')
3181 && (data[++index] == 'n') && (data[++index] == 'u') && (data[++index] == 'e'))
3182 return TokenNamecontinue;
3184 return TokenNameIdentifier;
3186 return TokenNameIdentifier;
3189 // declare default do die
3190 // TODO delete define ==> no keyword !
3193 if ((data[++index] == 'o'))
3196 return TokenNameIdentifier;
3198 // if ((data[++index] == 'e')
3199 // && (data[++index] == 'f')
3200 // && (data[++index] == 'i')
3201 // && (data[++index] == 'n')
3202 // && (data[++index] == 'e'))
3203 // return TokenNamedefine;
3205 // return TokenNameIdentifier;
3207 if ((data[++index] == 'e') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'a')
3208 && (data[++index] == 'r') && (data[++index] == 'e'))
3209 return TokenNamedeclare;
3211 if ((data[++index] == 'e') && (data[++index] == 'f') && (data[++index] == 'a') && (data[++index] == 'u')
3212 && (data[++index] == 'l') && (data[++index] == 't'))
3213 return TokenNamedefault;
3215 return TokenNameIdentifier;
3217 return TokenNameIdentifier;
3220 //echo else exit elseif extends eval
3223 if ((data[++index] == 'c') && (data[++index] == 'h') && (data[++index] == 'o'))
3224 return TokenNameecho;
3225 else if ((data[index] == 'l') && (data[++index] == 's') && (data[++index] == 'e'))
3226 return TokenNameelse;
3227 else if ((data[index] == 'x') && (data[++index] == 'i') && (data[++index] == 't'))
3228 return TokenNameexit;
3229 else if ((data[index] == 'v') && (data[++index] == 'a') && (data[++index] == 'l'))
3230 return TokenNameeval;
3232 return TokenNameIdentifier;
3235 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'i') && (data[++index] == 'f'))
3236 return TokenNameendif;
3237 if ((data[index] == 'm') && (data[++index] == 'p') && (data[++index] == 't') && (data[++index] == 'y'))
3238 return TokenNameempty;
3240 return TokenNameIdentifier;
3243 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'f') && (data[++index] == 'o')
3244 && (data[++index] == 'r'))
3245 return TokenNameendfor;
3246 else if ((data[index] == 'l') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 'i')
3247 && (data[++index] == 'f'))
3248 return TokenNameelseif;
3250 return TokenNameIdentifier;
3252 if ((data[++index] == 'x') && (data[++index] == 't') && (data[++index] == 'e') && (data[++index] == 'n')
3253 && (data[++index] == 'd') && (data[++index] == 's'))
3254 return TokenNameextends;
3256 return TokenNameIdentifier;
3259 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'w') && (data[++index] == 'h')
3260 && (data[++index] == 'i') && (data[++index] == 'l') && (data[++index] == 'e'))
3261 return TokenNameendwhile;
3263 return TokenNameIdentifier;
3266 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 's') && (data[++index] == 'w')
3267 && (data[++index] == 'i') && (data[++index] == 't') && (data[++index] == 'c') && (data[++index] == 'h'))
3268 return TokenNameendswitch;
3270 return TokenNameIdentifier;
3273 if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'd') && (data[++index] == 'e')
3274 && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'a') && (data[++index] == 'r')
3275 && (data[++index] == 'e'))
3276 return TokenNameenddeclare;
3278 if ((data[++index] == 'n') // endforeach
3279 && (data[++index] == 'd') && (data[++index] == 'f') && (data[++index] == 'o') && (data[++index] == 'r')
3280 && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'c') && (data[++index] == 'h'))
3281 return TokenNameendforeach;
3283 return TokenNameIdentifier;
3285 return TokenNameIdentifier;
3288 //for false final function
3291 if ((data[++index] == 'o') && (data[++index] == 'r'))
3292 return TokenNamefor;
3294 return TokenNameIdentifier;
3296 // if ((data[++index] == 'a') && (data[++index] == 'l')
3297 // && (data[++index] == 's') && (data[++index] == 'e'))
3298 // return TokenNamefalse;
3299 if ((data[++index] == 'i') && (data[++index] == 'n') && (data[++index] == 'a') && (data[++index] == 'l'))
3300 return TokenNamefinal;
3302 return TokenNameIdentifier;
3305 if ((data[++index] == 'o') && (data[++index] == 'r') && (data[++index] == 'e') && (data[++index] == 'a')
3306 && (data[++index] == 'c') && (data[++index] == 'h'))
3307 return TokenNameforeach;
3309 return TokenNameIdentifier;
3312 if ((data[++index] == 'u') && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 't')
3313 && (data[++index] == 'i') && (data[++index] == 'o') && (data[++index] == 'n'))
3314 return TokenNamefunction;
3316 return TokenNameIdentifier;
3318 return TokenNameIdentifier;
3323 if ((data[++index] == 'l') && (data[++index] == 'o') && (data[++index] == 'b') && (data[++index] == 'a')
3324 && (data[++index] == 'l')) {
3325 return TokenNameglobal;
3328 return TokenNameIdentifier;
3330 //if int isset include include_once instanceof interface implements
3333 if (data[++index] == 'f')
3336 return TokenNameIdentifier;
3338 // if ((data[++index] == 'n') && (data[++index] == 't'))
3339 // return TokenNameint;
3341 // return TokenNameIdentifier;
3343 if ((data[++index] == 's') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 't'))
3344 return TokenNameisset;
3346 return TokenNameIdentifier;
3348 if ((data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'u')
3349 && (data[++index] == 'd') && (data[++index] == 'e'))
3350 return TokenNameinclude;
3352 return TokenNameIdentifier;
3355 if ((data[++index] == 'n') && (data[++index] == 't') && (data[++index] == 'e') && (data[++index] == 'r')
3356 && (data[++index] == 'f') && (data[++index] == 'a') && (data[++index] == 'c') && (data[++index] == 'e'))
3357 return TokenNameinterface;
3359 return TokenNameIdentifier;
3362 if ((data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 't') && (data[++index] == 'a')
3363 && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'e') && (data[++index] == 'o')
3364 && (data[++index] == 'f'))
3365 return TokenNameinstanceof;
3366 if ((data[index] == 'm') && (data[++index] == 'p') && (data[++index] == 'l') && (data[++index] == 'e')
3367 && (data[++index] == 'm') && (data[++index] == 'e') && (data[++index] == 'n') && (data[++index] == 't')
3368 && (data[++index] == 's'))
3369 return TokenNameimplements;
3371 return TokenNameIdentifier;
3373 if ((data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'l') && (data[++index] == 'u')
3374 && (data[++index] == 'd') && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == 'o')
3375 && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'e'))
3376 return TokenNameinclude_once;
3378 return TokenNameIdentifier;
3380 return TokenNameIdentifier;
3385 if ((data[++index] == 'i') && (data[++index] == 's') && (data[++index] == 't')) {
3386 return TokenNamelist;
3389 return TokenNameIdentifier;
3394 if ((data[++index] == 'e') && (data[++index] == 'w'))
3395 return TokenNamenew;
3397 return TokenNameIdentifier;
3399 // if ((data[++index] == 'u') && (data[++index] == 'l')
3400 // && (data[++index] == 'l'))
3401 // return TokenNamenull;
3403 // return TokenNameIdentifier;
3405 return TokenNameIdentifier;
3410 if (data[++index] == 'r') {
3414 // if (length == 12) {
3415 // if ((data[++index] == 'l')
3416 // && (data[++index] == 'd')
3417 // && (data[++index] == '_')
3418 // && (data[++index] == 'f')
3419 // && (data[++index] == 'u')
3420 // && (data[++index] == 'n')
3421 // && (data[++index] == 'c')
3422 // && (data[++index] == 't')
3423 // && (data[++index] == 'i')
3424 // && (data[++index] == 'o')
3425 // && (data[++index] == 'n')) {
3426 // return TokenNameold_function;
3429 return TokenNameIdentifier;
3431 // print public private protected
3434 if ((data[++index] == 'r') && (data[++index] == 'i') && (data[++index] == 'n') && (data[++index] == 't')) {
3435 return TokenNameprint;
3437 return TokenNameIdentifier;
3439 if ((data[++index] == 'u') && (data[++index] == 'b') && (data[++index] == 'l') && (data[++index] == 'i')
3440 && (data[++index] == 'c')) {
3441 return TokenNamepublic;
3443 return TokenNameIdentifier;
3445 if ((data[++index] == 'r') && (data[++index] == 'i') && (data[++index] == 'v') && (data[++index] == 'a')
3446 && (data[++index] == 't') && (data[++index] == 'e')) {
3447 return TokenNameprivate;
3449 return TokenNameIdentifier;
3451 if ((data[++index] == 'r') && (data[++index] == 'o') && (data[++index] == 't') && (data[++index] == 'e')
3452 && (data[++index] == 'c') && (data[++index] == 't') && (data[++index] == 'e') && (data[++index] == 'd')) {
3453 return TokenNameprotected;
3455 return TokenNameIdentifier;
3457 return TokenNameIdentifier;
3459 //return require require_once
3461 if ((data[++index] == 'e') && (data[++index] == 't') && (data[++index] == 'u') && (data[++index] == 'r')
3462 && (data[++index] == 'n')) {
3463 return TokenNamereturn;
3465 } else if (length == 7) {
3466 if ((data[++index] == 'e') && (data[++index] == 'q') && (data[++index] == 'u') && (data[++index] == 'i')
3467 && (data[++index] == 'r') && (data[++index] == 'e')) {
3468 return TokenNamerequire;
3470 } else if (length == 12) {
3471 if ((data[++index] == 'e') && (data[++index] == 'q') && (data[++index] == 'u') && (data[++index] == 'i')
3472 && (data[++index] == 'r') && (data[++index] == 'e') && (data[++index] == '_') && (data[++index] == 'o')
3473 && (data[++index] == 'n') && (data[++index] == 'c') && (data[++index] == 'e')) {
3474 return TokenNamerequire_once;
3477 return TokenNameIdentifier;
3482 if (data[++index] == 't')
3483 if ((data[++index] == 'a') && (data[++index] == 't') && (data[++index] == 'i') && (data[++index] == 'c')) {
3484 return TokenNamestatic;
3486 return TokenNameIdentifier;
3487 else if ((data[index] == 'w') && (data[++index] == 'i') && (data[++index] == 't') && (data[++index] == 'c')
3488 && (data[++index] == 'h'))
3489 return TokenNameswitch;
3491 return TokenNameIdentifier;
3493 return TokenNameIdentifier;
3499 if ((data[++index] == 'r') && (data[++index] == 'y'))
3500 return TokenNametry;
3502 return TokenNameIdentifier;
3504 // if ((data[++index] == 'r') && (data[++index] == 'u')
3505 // && (data[++index] == 'e'))
3506 // return TokenNametrue;
3508 // return TokenNameIdentifier;
3510 if ((data[++index] == 'h') && (data[++index] == 'r') && (data[++index] == 'o') && (data[++index] == 'w'))
3511 return TokenNamethrow;
3513 return TokenNameIdentifier;
3515 return TokenNameIdentifier;
3521 if ((data[++index] == 's') && (data[++index] == 'e'))
3522 return TokenNameuse;
3524 return TokenNameIdentifier;
3526 if ((data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 't'))
3527 return TokenNameunset;
3529 return TokenNameIdentifier;
3531 return TokenNameIdentifier;
3537 if ((data[++index] == 'a') && (data[++index] == 'r'))
3538 return TokenNamevar;
3540 return TokenNameIdentifier;
3542 return TokenNameIdentifier;
3548 if ((data[++index] == 'h') && (data[++index] == 'i') && (data[++index] == 'l') && (data[++index] == 'e'))
3549 return TokenNamewhile;
3551 return TokenNameIdentifier;
3552 //case 6:if ( (data[++index] =='i') && (data[++index]=='d') &&
3553 // (data[++index]=='e') && (data[++index]=='f')&&
3554 // (data[++index]=='p'))
3555 //return TokenNamewidefp ;
3557 //return TokenNameIdentifier;
3559 return TokenNameIdentifier;
3565 if ((data[++index] == 'o') && (data[++index] == 'r'))
3566 return TokenNamexor;
3568 return TokenNameIdentifier;
3570 return TokenNameIdentifier;
3573 return TokenNameIdentifier;
3577 public int scanNumber(boolean dotPrefix) throws InvalidInputException {
3578 //when entering this method the currentCharacter is the firt
3579 //digit of the number , i.e. it may be preceeded by a . when
3581 boolean floating = dotPrefix;
3582 if ((!dotPrefix) && (currentCharacter == '0')) {
3583 if (getNextChar('x', 'X') >= 0) { //----------hexa-----------------
3584 //force the first char of the hexa number do exist...
3585 // consume next character
3586 unicodeAsBackSlash = false;
3587 currentCharacter = source[currentPosition++];
3588 // if (((currentCharacter = source[currentPosition++]) == '\\')
3589 // && (source[currentPosition] == 'u')) {
3590 // getNextUnicodeChar();
3592 // if (withoutUnicodePtr != 0) {
3593 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3596 if (Character.digit(currentCharacter, 16) == -1)
3597 throw new InvalidInputException(INVALID_HEXA);
3599 while (getNextCharAsDigit(16)) {
3602 // if (getNextChar('l', 'L') >= 0)
3603 // return TokenNameLongLiteral;
3605 return TokenNameIntegerLiteral;
3607 //there is x or X in the number
3608 //potential octal ! ... some one may write 000099.0 ! thus 00100 <
3609 // 00078.0 is true !!!!! crazy language
3610 if (getNextCharAsDigit()) {
3611 //-------------potential octal-----------------
3612 while (getNextCharAsDigit()) {
3615 // if (getNextChar('l', 'L') >= 0) {
3616 // return TokenNameLongLiteral;
3619 // if (getNextChar('f', 'F') >= 0) {
3620 // return TokenNameFloatingPointLiteral;
3622 if (getNextChar('d', 'D') >= 0) {
3623 return TokenNameDoubleLiteral;
3624 } else { //make the distinction between octal and float ....
3625 if (getNextChar('.')) { //bingo ! ....
3626 while (getNextCharAsDigit()) {
3629 if (getNextChar('e', 'E') >= 0) {
3630 // consume next character
3631 unicodeAsBackSlash = false;
3632 currentCharacter = source[currentPosition++];
3633 // if (((currentCharacter = source[currentPosition++]) == '\\')
3634 // && (source[currentPosition] == 'u')) {
3635 // getNextUnicodeChar();
3637 // if (withoutUnicodePtr != 0) {
3638 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3641 if ((currentCharacter == '-') || (currentCharacter == '+')) {
3642 // consume next character
3643 unicodeAsBackSlash = false;
3644 currentCharacter = source[currentPosition++];
3645 // if (((currentCharacter = source[currentPosition++]) == '\\')
3646 // && (source[currentPosition] == 'u')) {
3647 // getNextUnicodeChar();
3649 // if (withoutUnicodePtr != 0) {
3650 // withoutUnicodeBuffer[++withoutUnicodePtr] =
3651 // currentCharacter;
3655 if (!Character.isDigit(currentCharacter))
3656 throw new InvalidInputException(INVALID_FLOAT);
3657 while (getNextCharAsDigit()) {
3661 // if (getNextChar('f', 'F') >= 0)
3662 // return TokenNameFloatingPointLiteral;
3663 getNextChar('d', 'D'); //jump over potential d or D
3664 return TokenNameDoubleLiteral;
3666 return TokenNameIntegerLiteral;
3673 while (getNextCharAsDigit()) {
3676 // if ((!dotPrefix) && (getNextChar('l', 'L') >= 0))
3677 // return TokenNameLongLiteral;
3678 if ((!dotPrefix) && (getNextChar('.'))) { //decimal part that can be empty
3679 while (getNextCharAsDigit()) {
3684 //if floating is true both exponant and suffix may be optional
3685 if (getNextChar('e', 'E') >= 0) {
3687 // consume next character
3688 unicodeAsBackSlash = false;
3689 currentCharacter = source[currentPosition++];
3690 // if (((currentCharacter = source[currentPosition++]) == '\\')
3691 // && (source[currentPosition] == 'u')) {
3692 // getNextUnicodeChar();
3694 // if (withoutUnicodePtr != 0) {
3695 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3698 if ((currentCharacter == '-') || (currentCharacter == '+')) { // consume
3701 unicodeAsBackSlash = false;
3702 currentCharacter = source[currentPosition++];
3703 // if (((currentCharacter = source[currentPosition++]) == '\\')
3704 // && (source[currentPosition] == 'u')) {
3705 // getNextUnicodeChar();
3707 // if (withoutUnicodePtr != 0) {
3708 // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
3712 if (!Character.isDigit(currentCharacter))
3713 throw new InvalidInputException(INVALID_FLOAT);
3714 while (getNextCharAsDigit()) {
3718 if (getNextChar('d', 'D') >= 0)
3719 return TokenNameDoubleLiteral;
3720 // if (getNextChar('f', 'F') >= 0)
3721 // return TokenNameFloatingPointLiteral;
3722 //the long flag has been tested before
3723 return floating ? TokenNameDoubleLiteral : TokenNameIntegerLiteral;
3727 * Search the line number corresponding to a specific position
3730 public final int getLineNumber(int position) {
3731 if (lineEnds == null)
3733 int length = linePtr + 1;
3736 int g = 0, d = length - 1;
3740 if (position < lineEnds[m]) {
3742 } else if (position > lineEnds[m]) {
3748 if (position < lineEnds[m]) {
3754 public void setPHPMode(boolean mode) {
3758 public final void setSource(char[] source) {
3759 setSource(null, source);
3762 public final void setSource(ICompilationUnit compilationUnit, char[] source) {
3763 //the source-buffer is set to sourceString
3764 this.compilationUnit = compilationUnit;
3765 if (source == null) {
3766 this.source = new char[0];
3768 this.source = source;
3771 initialPosition = currentPosition = 0;
3772 containsAssertKeyword = false;
3773 withoutUnicodeBuffer = new char[this.source.length];
3774 encapsedStringStack = new Stack();
3777 public String toString() {
3778 if (startPosition == source.length)
3779 return "EOF\n\n" + new String(source); //$NON-NLS-1$
3780 if (currentPosition > source.length)
3781 return "behind the EOF :-( ....\n\n" + new String(source); //$NON-NLS-1$
3782 char front[] = new char[startPosition];
3783 System.arraycopy(source, 0, front, 0, startPosition);
3784 int middleLength = (currentPosition - 1) - startPosition + 1;
3786 if (middleLength > -1) {
3787 middle = new char[middleLength];
3788 System.arraycopy(source, startPosition, middle, 0, middleLength);
3790 middle = new char[0];
3792 char end[] = new char[source.length - (currentPosition - 1)];
3793 System.arraycopy(source, (currentPosition - 1) + 1, end, 0, source.length - (currentPosition - 1) - 1);
3794 return new String(front) + "\n===============================\nStarts here -->" //$NON-NLS-1$
3795 + new String(middle) + "<-- Ends here\n===============================\n" //$NON-NLS-1$
3799 public final String toStringAction(int act) {
3801 case TokenNameERROR:
3802 return "ScannerError"; // + new String(getCurrentTokenSource()) + ")";
3804 case TokenNameINLINE_HTML:
3805 return "Inline-HTML(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3806 case TokenNameIdentifier:
3807 return "Identifier(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3808 case TokenNameVariable:
3809 return "Variable(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3810 case TokenNameabstract:
3811 return "abstract"; //$NON-NLS-1$
3813 return "AND"; //$NON-NLS-1$
3814 case TokenNamearray:
3815 return "array"; //$NON-NLS-1$
3817 return "as"; //$NON-NLS-1$
3818 case TokenNamebreak:
3819 return "break"; //$NON-NLS-1$
3821 return "case"; //$NON-NLS-1$
3822 case TokenNameclass:
3823 return "class"; //$NON-NLS-1$
3824 case TokenNamecatch:
3825 return "catch"; //$NON-NLS-1$
3826 case TokenNameclone:
3829 case TokenNameconst:
3832 case TokenNamecontinue:
3833 return "continue"; //$NON-NLS-1$
3834 case TokenNamedefault:
3835 return "default"; //$NON-NLS-1$
3836 // case TokenNamedefine :
3837 // return "define"; //$NON-NLS-1$
3839 return "do"; //$NON-NLS-1$
3841 return "echo"; //$NON-NLS-1$
3843 return "else"; //$NON-NLS-1$
3844 case TokenNameelseif:
3845 return "elseif"; //$NON-NLS-1$
3846 case TokenNameendfor:
3847 return "endfor"; //$NON-NLS-1$
3848 case TokenNameendforeach:
3849 return "endforeach"; //$NON-NLS-1$
3850 case TokenNameendif:
3851 return "endif"; //$NON-NLS-1$
3852 case TokenNameendswitch:
3853 return "endswitch"; //$NON-NLS-1$
3854 case TokenNameendwhile:
3855 return "endwhile"; //$NON-NLS-1$
3858 case TokenNameextends:
3859 return "extends"; //$NON-NLS-1$
3860 // case TokenNamefalse :
3861 // return "false"; //$NON-NLS-1$
3862 case TokenNamefinal:
3863 return "final"; //$NON-NLS-1$
3865 return "for"; //$NON-NLS-1$
3866 case TokenNameforeach:
3867 return "foreach"; //$NON-NLS-1$
3868 case TokenNamefunction:
3869 return "function"; //$NON-NLS-1$
3870 case TokenNameglobal:
3871 return "global"; //$NON-NLS-1$
3873 return "if"; //$NON-NLS-1$
3874 case TokenNameimplements:
3875 return "implements"; //$NON-NLS-1$
3876 case TokenNameinclude:
3877 return "include"; //$NON-NLS-1$
3878 case TokenNameinclude_once:
3879 return "include_once"; //$NON-NLS-1$
3880 case TokenNameinstanceof:
3881 return "instanceof"; //$NON-NLS-1$
3882 case TokenNameinterface:
3883 return "interface"; //$NON-NLS-1$
3884 case TokenNameisset:
3885 return "isset"; //$NON-NLS-1$
3887 return "list"; //$NON-NLS-1$
3889 return "new"; //$NON-NLS-1$
3890 // case TokenNamenull :
3891 // return "null"; //$NON-NLS-1$
3893 return "OR"; //$NON-NLS-1$
3894 case TokenNameprint:
3895 return "print"; //$NON-NLS-1$
3896 case TokenNameprivate:
3897 return "private"; //$NON-NLS-1$
3898 case TokenNameprotected:
3899 return "protected"; //$NON-NLS-1$
3900 case TokenNamepublic:
3901 return "public"; //$NON-NLS-1$
3902 case TokenNamerequire:
3903 return "require"; //$NON-NLS-1$
3904 case TokenNamerequire_once:
3905 return "require_once"; //$NON-NLS-1$
3906 case TokenNamereturn:
3907 return "return"; //$NON-NLS-1$
3908 case TokenNamestatic:
3909 return "static"; //$NON-NLS-1$
3910 case TokenNameswitch:
3911 return "switch"; //$NON-NLS-1$
3912 // case TokenNametrue :
3913 // return "true"; //$NON-NLS-1$
3914 case TokenNameunset:
3915 return "unset"; //$NON-NLS-1$
3917 return "var"; //$NON-NLS-1$
3918 case TokenNamewhile:
3919 return "while"; //$NON-NLS-1$
3921 return "XOR"; //$NON-NLS-1$
3922 // case TokenNamethis :
3923 // return "$this"; //$NON-NLS-1$
3924 case TokenNameIntegerLiteral:
3925 return "Integer(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3926 case TokenNameDoubleLiteral:
3927 return "Double(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3928 case TokenNameStringDoubleQuote:
3929 return "StringLiteral(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3930 case TokenNameStringSingleQuote:
3931 return "StringConstant(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3932 case TokenNameStringInterpolated:
3933 return "StringInterpolated(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3934 case TokenNameEncapsedString0:
3935 return "`"; //$NON-NLS-1$
3936 case TokenNameEncapsedString1:
3937 return "\'"; //$NON-NLS-1$
3938 case TokenNameEncapsedString2:
3939 return "\""; //$NON-NLS-1$
3940 case TokenNameSTRING:
3941 return "STRING(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$
3942 case TokenNameHEREDOC:
3943 return "HEREDOC(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
3944 case TokenNamePLUS_PLUS:
3945 return "++"; //$NON-NLS-1$
3946 case TokenNameMINUS_MINUS:
3947 return "--"; //$NON-NLS-1$
3948 case TokenNameEQUAL_EQUAL:
3949 return "=="; //$NON-NLS-1$
3950 case TokenNameEQUAL_EQUAL_EQUAL:
3951 return "==="; //$NON-NLS-1$
3952 case TokenNameEQUAL_GREATER:
3953 return "=>"; //$NON-NLS-1$
3954 case TokenNameLESS_EQUAL:
3955 return "<="; //$NON-NLS-1$
3956 case TokenNameGREATER_EQUAL:
3957 return ">="; //$NON-NLS-1$
3958 case TokenNameNOT_EQUAL:
3959 return "!="; //$NON-NLS-1$
3960 case TokenNameNOT_EQUAL_EQUAL:
3961 return "!=="; //$NON-NLS-1$
3962 case TokenNameLEFT_SHIFT:
3963 return "<<"; //$NON-NLS-1$
3964 case TokenNameRIGHT_SHIFT:
3965 return ">>"; //$NON-NLS-1$
3966 case TokenNamePLUS_EQUAL:
3967 return "+="; //$NON-NLS-1$
3968 case TokenNameMINUS_EQUAL:
3969 return "-="; //$NON-NLS-1$
3970 case TokenNameMULTIPLY_EQUAL:
3971 return "*="; //$NON-NLS-1$
3972 case TokenNameDIVIDE_EQUAL:
3973 return "/="; //$NON-NLS-1$
3974 case TokenNameAND_EQUAL:
3975 return "&="; //$NON-NLS-1$
3976 case TokenNameOR_EQUAL:
3977 return "|="; //$NON-NLS-1$
3978 case TokenNameXOR_EQUAL:
3979 return "^="; //$NON-NLS-1$
3980 case TokenNameREMAINDER_EQUAL:
3981 return "%="; //$NON-NLS-1$
3982 case TokenNameDOT_EQUAL:
3983 return ".="; //$NON-NLS-1$
3984 case TokenNameLEFT_SHIFT_EQUAL:
3985 return "<<="; //$NON-NLS-1$
3986 case TokenNameRIGHT_SHIFT_EQUAL:
3987 return ">>="; //$NON-NLS-1$
3988 case TokenNameOR_OR:
3989 return "||"; //$NON-NLS-1$
3990 case TokenNameAND_AND:
3991 return "&&"; //$NON-NLS-1$
3993 return "+"; //$NON-NLS-1$
3994 case TokenNameMINUS:
3995 return "-"; //$NON-NLS-1$
3996 case TokenNameMINUS_GREATER:
3999 return "!"; //$NON-NLS-1$
4000 case TokenNameREMAINDER:
4001 return "%"; //$NON-NLS-1$
4003 return "^"; //$NON-NLS-1$
4005 return "&"; //$NON-NLS-1$
4006 case TokenNameMULTIPLY:
4007 return "*"; //$NON-NLS-1$
4009 return "|"; //$NON-NLS-1$
4010 case TokenNameTWIDDLE:
4011 return "~"; //$NON-NLS-1$
4012 case TokenNameTWIDDLE_EQUAL:
4013 return "~="; //$NON-NLS-1$
4014 case TokenNameDIVIDE:
4015 return "/"; //$NON-NLS-1$
4016 case TokenNameGREATER:
4017 return ">"; //$NON-NLS-1$
4019 return "<"; //$NON-NLS-1$
4020 case TokenNameLPAREN:
4021 return "("; //$NON-NLS-1$
4022 case TokenNameRPAREN:
4023 return ")"; //$NON-NLS-1$
4024 case TokenNameLBRACE:
4025 return "{"; //$NON-NLS-1$
4026 case TokenNameRBRACE:
4027 return "}"; //$NON-NLS-1$
4028 case TokenNameLBRACKET:
4029 return "["; //$NON-NLS-1$
4030 case TokenNameRBRACKET:
4031 return "]"; //$NON-NLS-1$
4032 case TokenNameSEMICOLON:
4033 return ";"; //$NON-NLS-1$
4034 case TokenNameQUESTION:
4035 return "?"; //$NON-NLS-1$
4036 case TokenNameCOLON:
4037 return ":"; //$NON-NLS-1$
4038 case TokenNameCOMMA:
4039 return ","; //$NON-NLS-1$
4041 return "."; //$NON-NLS-1$
4042 case TokenNameEQUAL:
4043 return "="; //$NON-NLS-1$
4046 case TokenNameDOLLAR:
4048 case TokenNameDOLLAR_LBRACE:
4050 case TokenNameLBRACE_DOLLAR:
4053 return "EOF"; //$NON-NLS-1$
4054 case TokenNameWHITESPACE:
4055 return "WHITESPACE(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4056 case TokenNameCOMMENT_LINE:
4057 return "COMMENT_LINE(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4058 case TokenNameCOMMENT_BLOCK:
4059 return "COMMENT_BLOCK(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4060 case TokenNameCOMMENT_PHPDOC:
4061 return "COMMENT_PHPDOC(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$
4062 // case TokenNameHTML :
4063 // return "HTML(" + new String(getCurrentTokenSource()) + ")";
4066 return "__FILE__"; //$NON-NLS-1$
4068 return "__LINE__"; //$NON-NLS-1$
4069 case TokenNameCLASS_C:
4070 return "__CLASS__"; //$NON-NLS-1$
4071 case TokenNameMETHOD_C:
4072 return "__METHOD__"; //$NON-NLS-1$
4073 case TokenNameFUNC_C:
4074 return "__FUNCTION__"; //$NON-NLS-1
4075 case TokenNameboolCAST:
4076 return "( bool )"; //$NON-NLS-1$
4077 case TokenNameintCAST:
4078 return "( int )"; //$NON-NLS-1$
4079 case TokenNamedoubleCAST:
4080 return "( double )"; //$NON-NLS-1$
4081 case TokenNameobjectCAST:
4082 return "( object )"; //$NON-NLS-1$
4083 case TokenNamestringCAST:
4084 return "( string )"; //$NON-NLS-1$
4086 return "not-a-token(" + (new Integer(act)) + ") " + new String(getCurrentTokenSource()); //$NON-NLS-1$
4094 public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace) {
4095 this(tokenizeComments, tokenizeWhiteSpace, false);
4098 public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, boolean checkNonExternalizedStringLiterals) {
4099 this(tokenizeComments, tokenizeWhiteSpace, checkNonExternalizedStringLiterals, false);
4102 public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, boolean checkNonExternalizedStringLiterals,
4103 boolean assertMode) {
4104 this(tokenizeComments, tokenizeWhiteSpace, checkNonExternalizedStringLiterals, assertMode, false, null, null, true);
4107 public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, boolean checkNonExternalizedStringLiterals,
4108 boolean assertMode, boolean tokenizeStrings, char[][] taskTags, char[][] taskPriorities, boolean isTaskCaseSensitive) {
4109 this.eofPosition = Integer.MAX_VALUE;
4110 this.tokenizeComments = tokenizeComments;
4111 this.tokenizeWhiteSpace = tokenizeWhiteSpace;
4112 this.tokenizeStrings = tokenizeStrings;
4113 this.checkNonExternalizedStringLiterals = checkNonExternalizedStringLiterals;
4114 this.assertMode = assertMode;
4115 this.encapsedStringStack = null;
4116 this.taskTags = taskTags;
4117 this.taskPriorities = taskPriorities;
4120 private void checkNonExternalizeString() throws InvalidInputException {
4121 if (currentLine == null)
4123 parseTags(currentLine);
4126 private void parseTags(NLSLine line) throws InvalidInputException {
4127 String s = new String(getCurrentTokenSource());
4128 int pos = s.indexOf(TAG_PREFIX);
4129 int lineLength = line.size();
4131 int start = pos + TAG_PREFIX_LENGTH;
4132 int end = s.indexOf(TAG_POSTFIX, start);
4133 String index = s.substring(start, end);
4136 i = Integer.parseInt(index) - 1;
4137 // Tags are one based not zero based.
4138 } catch (NumberFormatException e) {
4139 i = -1; // we don't want to consider this as a valid NLS tag
4141 if (line.exists(i)) {
4144 pos = s.indexOf(TAG_PREFIX, start);
4146 this.nonNLSStrings = new StringLiteral[lineLength];
4147 int nonNLSCounter = 0;
4148 for (Iterator iterator = line.iterator(); iterator.hasNext();) {
4149 StringLiteral literal = (StringLiteral) iterator.next();
4150 if (literal != null) {
4151 this.nonNLSStrings[nonNLSCounter++] = literal;
4154 if (nonNLSCounter == 0) {
4155 this.nonNLSStrings = null;
4159 this.wasNonExternalizedStringLiteral = true;
4160 if (nonNLSCounter != lineLength) {
4161 System.arraycopy(this.nonNLSStrings, 0, (this.nonNLSStrings = new StringLiteral[nonNLSCounter]), 0, nonNLSCounter);
4166 public final void scanEscapeCharacter() throws InvalidInputException {
4167 // the string with "\\u" is a legal string of two chars \ and u
4168 //thus we use a direct access to the source (for regular cases).
4169 if (unicodeAsBackSlash) {
4170 // consume next character
4171 unicodeAsBackSlash = false;
4172 // if (((currentCharacter = source[currentPosition++]) == '\\') &&
4173 // (source[currentPosition] == 'u')) {
4174 // getNextUnicodeChar();
4176 if (withoutUnicodePtr != 0) {
4177 withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter;
4181 currentCharacter = source[currentPosition++];
4182 switch (currentCharacter) {
4184 currentCharacter = '\b';
4187 currentCharacter = '\t';
4190 currentCharacter = '\n';
4193 currentCharacter = '\f';
4196 currentCharacter = '\r';
4199 currentCharacter = '\"';
4202 currentCharacter = '\'';
4205 currentCharacter = '\\';
4208 // -----------octal escape--------------
4210 // OctalDigit OctalDigit
4211 // ZeroToThree OctalDigit OctalDigit
4212 int number = Character.getNumericValue(currentCharacter);
4213 if (number >= 0 && number <= 7) {
4214 boolean zeroToThreeNot = number > 3;
4215 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
4216 int digit = Character.getNumericValue(currentCharacter);
4217 if (digit >= 0 && digit <= 7) {
4218 number = (number * 8) + digit;
4219 if (Character.isDigit(currentCharacter = source[currentPosition++])) {
4220 if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit
4221 // Digit --> ignore last character
4224 digit = Character.getNumericValue(currentCharacter);
4225 if (digit >= 0 && digit <= 7) { // has read \ZeroToThree
4226 // OctalDigit OctalDigit
4227 number = (number * 8) + digit;
4228 } else { // has read \ZeroToThree OctalDigit NonOctalDigit
4229 // --> ignore last character
4233 } else { // has read \OctalDigit NonDigit--> ignore last
4237 } else { // has read \OctalDigit NonOctalDigit--> ignore last
4241 } else { // has read \OctalDigit --> ignore last character
4245 throw new InvalidInputException(INVALID_ESCAPE);
4246 currentCharacter = (char) number;
4248 throw new InvalidInputException(INVALID_ESCAPE);
4252 //chech presence of task: tags
4253 //TODO (frederic) see if we need to take unicode characters into account...
4254 public void checkTaskTag(int commentStart, int commentEnd) {
4255 char[] src = this.source;
4257 // only look for newer task: tags
4258 if (this.foundTaskCount > 0 && this.foundTaskPositions[this.foundTaskCount - 1][0] >= commentStart) {
4261 int foundTaskIndex = this.foundTaskCount;
4262 char previous = src[commentStart + 1]; // should be '*' or '/'
4263 nextChar: for (int i = commentStart + 2; i < commentEnd && i < this.eofPosition; i++) {
4265 char[] priority = null;
4266 // check for tag occurrence only if not ambiguous with javadoc tag
4267 if (previous != '@') {
4268 nextTag: for (int itag = 0; itag < this.taskTags.length; itag++) {
4269 tag = this.taskTags[itag];
4270 int tagLength = tag.length;
4274 // ensure tag is not leaded with letter if tag starts with a letter
4275 if (Scanner.isPHPIdentifierStart(tag[0])) {
4276 if (Scanner.isPHPIdentifierPart(previous)) {
4281 for (int t = 0; t < tagLength; t++) {
4284 if (x >= this.eofPosition || x >= commentEnd)
4286 if ((sc = src[i + t]) != (tc = tag[t])) { // case sensitive check
4287 if (this.isTaskCaseSensitive || (Character.toLowerCase(sc) != Character.toLowerCase(tc))) { // case insensitive check
4292 // ensure tag is not followed with letter if tag finishes with a letter
4293 if (i + tagLength < commentEnd && Scanner.isPHPIdentifierPart(src[i + tagLength - 1])) {
4294 if (Scanner.isPHPIdentifierPart(src[i + tagLength]))
4297 if (this.foundTaskTags == null) {
4298 this.foundTaskTags = new char[5][];
4299 this.foundTaskMessages = new char[5][];
4300 this.foundTaskPriorities = new char[5][];
4301 this.foundTaskPositions = new int[5][];
4302 } else if (this.foundTaskCount == this.foundTaskTags.length) {
4303 System.arraycopy(this.foundTaskTags, 0, this.foundTaskTags = new char[this.foundTaskCount * 2][], 0,
4304 this.foundTaskCount);
4305 System.arraycopy(this.foundTaskMessages, 0, this.foundTaskMessages = new char[this.foundTaskCount * 2][], 0,
4306 this.foundTaskCount);
4307 System.arraycopy(this.foundTaskPriorities, 0, this.foundTaskPriorities = new char[this.foundTaskCount * 2][], 0,
4308 this.foundTaskCount);
4309 System.arraycopy(this.foundTaskPositions, 0, this.foundTaskPositions = new int[this.foundTaskCount * 2][], 0,
4310 this.foundTaskCount);
4313 priority = this.taskPriorities != null && itag < this.taskPriorities.length ? this.taskPriorities[itag] : null;
4315 this.foundTaskTags[this.foundTaskCount] = tag;
4316 this.foundTaskPriorities[this.foundTaskCount] = priority;
4317 this.foundTaskPositions[this.foundTaskCount] = new int[] { i, i + tagLength - 1 };
4318 this.foundTaskMessages[this.foundTaskCount] = CharOperation.NO_CHAR;
4319 this.foundTaskCount++;
4320 i += tagLength - 1; // will be incremented when looping
4326 for (int i = foundTaskIndex; i < this.foundTaskCount; i++) {
4327 // retrieve message start and end positions
4328 int msgStart = this.foundTaskPositions[i][0] + this.foundTaskTags[i].length;
4329 int max_value = i + 1 < this.foundTaskCount ? this.foundTaskPositions[i + 1][0] - 1 : commentEnd - 1;
4330 // at most beginning of next task
4331 if (max_value < msgStart) {
4332 max_value = msgStart; // would only occur if tag is before EOF.
4336 for (int j = msgStart; j < max_value; j++) {
4337 if ((c = src[j]) == '\n' || c == '\r') {
4343 for (int j = max_value; j > msgStart; j--) {
4344 if ((c = src[j]) == '*') {
4352 if (msgStart == end)
4355 while (CharOperation.isWhitespace(src[end]) && msgStart <= end)
4357 while (CharOperation.isWhitespace(src[msgStart]) && msgStart <= end)
4359 // update the end position of the task
4360 this.foundTaskPositions[i][1] = end;
4361 // get the message source
4362 final int messageLength = end - msgStart + 1;
4363 char[] message = new char[messageLength];
4364 System.arraycopy(src, msgStart, message, 0, messageLength);
4365 this.foundTaskMessages[i] = message;
4369 // chech presence of task: tags
4370 // public void checkTaskTag(int commentStart, int commentEnd) {
4371 // // only look for newer task: tags
4372 // if (this.foundTaskCount > 0 && this.foundTaskPositions[this.foundTaskCount - 1][0] >= commentStart) {
4375 // int foundTaskIndex = this.foundTaskCount;
4376 // nextChar: for (int i = commentStart; i < commentEnd && i < this.eofPosition; i++) {
4377 // char[] tag = null;
4378 // char[] priority = null;
4379 // // check for tag occurrence
4380 // nextTag: for (int itag = 0; itag < this.taskTags.length; itag++) {
4381 // tag = this.taskTags[itag];
4382 // priority = this.taskPriorities != null && itag < this.taskPriorities.length ? this.taskPriorities[itag] : null;
4383 // int tagLength = tag.length;
4384 // for (int t = 0; t < tagLength; t++) {
4385 // if (this.source[i + t] != tag[t])
4386 // continue nextTag;
4388 // if (this.foundTaskTags == null) {
4389 // this.foundTaskTags = new char[5][];
4390 // this.foundTaskMessages = new char[5][];
4391 // this.foundTaskPriorities = new char[5][];
4392 // this.foundTaskPositions = new int[5][];
4393 // } else if (this.foundTaskCount == this.foundTaskTags.length) {
4394 // System.arraycopy(this.foundTaskTags, 0, this.foundTaskTags = new char[this.foundTaskCount * 2][], 0, this.foundTaskCount);
4395 // System.arraycopy(this.foundTaskMessages, 0, this.foundTaskMessages = new char[this.foundTaskCount * 2][], 0,
4396 // this.foundTaskCount);
4397 // System.arraycopy(this.foundTaskPriorities, 0, this.foundTaskPriorities = new char[this.foundTaskCount * 2][], 0,
4398 // this.foundTaskCount);
4399 // System.arraycopy(this.foundTaskPositions, 0, this.foundTaskPositions = new int[this.foundTaskCount * 2][], 0,
4400 // this.foundTaskCount);
4402 // this.foundTaskTags[this.foundTaskCount] = tag;
4403 // this.foundTaskPriorities[this.foundTaskCount] = priority;
4404 // this.foundTaskPositions[this.foundTaskCount] = new int[] { i, i + tagLength - 1 };
4405 // this.foundTaskMessages[this.foundTaskCount] = CharOperation.NO_CHAR;
4406 // this.foundTaskCount++;
4407 // i += tagLength - 1; // will be incremented when looping
4410 // for (int i = foundTaskIndex; i < this.foundTaskCount; i++) {
4411 // // retrieve message start and end positions
4412 // int msgStart = this.foundTaskPositions[i][0] + this.foundTaskTags[i].length;
4413 // int max_value = i + 1 < this.foundTaskCount ? this.foundTaskPositions[i + 1][0] - 1 : commentEnd - 1;
4414 // // at most beginning of next task
4415 // if (max_value < msgStart)
4416 // max_value = msgStart; // would only occur if tag is before EOF.
4419 // for (int j = msgStart; j < max_value; j++) {
4420 // if ((c = this.source[j]) == '\n' || c == '\r') {
4426 // for (int j = max_value; j > msgStart; j--) {
4427 // if ((c = this.source[j]) == '*') {
4435 // if (msgStart == end)
4436 // continue; // empty
4437 // // trim the message
4438 // while (CharOperation.isWhitespace(source[end]) && msgStart <= end)
4440 // while (CharOperation.isWhitespace(source[msgStart]) && msgStart <= end)
4442 // // update the end position of the task
4443 // this.foundTaskPositions[i][1] = end;
4444 // // get the message source
4445 // final int messageLength = end - msgStart + 1;
4446 // char[] message = new char[messageLength];
4447 // System.arraycopy(source, msgStart, message, 0, messageLength);
4448 // this.foundTaskMessages[i] = message;