X-Git-Url: http://secure.phpeclipse.com diff --git a/net.sourceforge.phpeclipse/src/net/sourceforge/phpdt/internal/compiler/parser/Scanner.java b/net.sourceforge.phpeclipse/src/net/sourceforge/phpdt/internal/compiler/parser/Scanner.java index a9520dd..ea1f9c0 100644 --- a/net.sourceforge.phpeclipse/src/net/sourceforge/phpdt/internal/compiler/parser/Scanner.java +++ b/net.sourceforge.phpeclipse/src/net/sourceforge/phpdt/internal/compiler/parser/Scanner.java @@ -9,72 +9,62 @@ * IBM Corporation - initial API and implementation ******************************************************************************/ package net.sourceforge.phpdt.internal.compiler.parser; - import java.util.ArrayList; import java.util.Iterator; import java.util.List; - +import java.util.Stack; import net.sourceforge.phpdt.core.compiler.CharOperation; import net.sourceforge.phpdt.core.compiler.IScanner; import net.sourceforge.phpdt.core.compiler.ITerminalSymbols; import net.sourceforge.phpdt.core.compiler.InvalidInputException; import net.sourceforge.phpdt.internal.compiler.ast.StringLiteral; - public class Scanner implements IScanner, ITerminalSymbols { - /* - * APIs ares - getNextToken() which return the current type of the token (this value is not memorized by the scanner) - - * getCurrentTokenSource() which provides with the token "REAL" source (aka all unicode have been transformed into a correct - * char) - sourceStart gives the position into the stream - currentPosition-1 gives the sourceEnd position into the stream + * APIs ares - getNextToken() which return the current type of the token + * (this value is not memorized by the scanner) - getCurrentTokenSource() + * which provides with the token "REAL" source (aka all unicode have been + * transformed into a correct char) - sourceStart gives the position into the + * stream - currentPosition-1 gives the sourceEnd position into the stream */ - // 1.4 feature private boolean assertMode; public boolean useAssertAsAnIndentifier = false; //flag indicating if processed source contains occurrences of keyword assert public boolean containsAssertKeyword = false; - public boolean recordLineSeparator; public boolean phpMode = false; - + public Stack encapsedStringStack = null; public char currentCharacter; public int startPosition; public int currentPosition; public int initialPosition, eofPosition; - // after this position eof are generated instead of real token from the source - + // after this position eof are generated instead of real token from the + // source public boolean tokenizeComments; public boolean tokenizeWhiteSpace; - + public boolean tokenizeStrings; //source should be viewed as a window (aka a part) //of a entire very large stream public char source[]; - //unicode support public char[] withoutUnicodeBuffer; public int withoutUnicodePtr; //when == 0 ==> no unicode in the current token public boolean unicodeAsBackSlash = false; - public boolean scanningFloatLiteral = false; - //support for /** comments //public char[][] comments = new char[10][]; public int[] commentStops = new int[10]; public int[] commentStarts = new int[10]; public int commentPtr = -1; // no comment test with commentPtr value -1 - //diet parsing support - jump over some method body when requested public boolean diet = false; - //support for the poor-line-debuggers .... //remember the position of the cr/lf public int[] lineEnds = new int[250]; public int linePtr = -1; public boolean wasAcr = false; - public static final String END_OF_SOURCE = "End_Of_Source"; //$NON-NLS-1$ - public static final String INVALID_HEXA = "Invalid_Hexa_Literal"; //$NON-NLS-1$ public static final String INVALID_OCTAL = "Invalid_Octal_Literal"; //$NON-NLS-1$ public static final String INVALID_CHARACTER_CONSTANT = "Invalid_Character_Constant"; //$NON-NLS-1$ @@ -82,45 +72,31 @@ public class Scanner implements IScanner, ITerminalSymbols { public static final String INVALID_INPUT = "Invalid_Input"; //$NON-NLS-1$ public static final String INVALID_UNICODE_ESCAPE = "Invalid_Unicode_Escape"; //$NON-NLS-1$ public static final String INVALID_FLOAT = "Invalid_Float_Literal"; //$NON-NLS-1$ - public static final String NULL_SOURCE_STRING = "Null_Source_String"; //$NON-NLS-1$ public static final String UNTERMINATED_STRING = "Unterminated_String"; //$NON-NLS-1$ public static final String UNTERMINATED_COMMENT = "Unterminated_Comment"; //$NON-NLS-1$ public static final String INVALID_CHAR_IN_STRING = "Invalid_Char_In_String"; //$NON-NLS-1$ - //----------------optimized identifier managment------------------ - static final char[] charArray_a = new char[] { 'a' }, - charArray_b = new char[] { 'b' }, - charArray_c = new char[] { 'c' }, - charArray_d = new char[] { 'd' }, - charArray_e = new char[] { 'e' }, - charArray_f = new char[] { 'f' }, - charArray_g = new char[] { 'g' }, - charArray_h = new char[] { 'h' }, - charArray_i = new char[] { 'i' }, - charArray_j = new char[] { 'j' }, - charArray_k = new char[] { 'k' }, - charArray_l = new char[] { 'l' }, - charArray_m = new char[] { 'm' }, - charArray_n = new char[] { 'n' }, - charArray_o = new char[] { 'o' }, - charArray_p = new char[] { 'p' }, - charArray_q = new char[] { 'q' }, - charArray_r = new char[] { 'r' }, - charArray_s = new char[] { 's' }, - charArray_t = new char[] { 't' }, - charArray_u = new char[] { 'u' }, - charArray_v = new char[] { 'v' }, - charArray_w = new char[] { 'w' }, - charArray_x = new char[] { 'x' }, - charArray_y = new char[] { 'y' }, - charArray_z = new char[] { 'z' }; - - static final char[] initCharArray = new char[] { '\u0000', '\u0000', '\u0000', '\u0000', '\u0000', '\u0000' }; + static final char[] charArray_a = new char[]{'a'}, + charArray_b = new char[]{'b'}, charArray_c = new char[]{'c'}, + charArray_d = new char[]{'d'}, charArray_e = new char[]{'e'}, + charArray_f = new char[]{'f'}, charArray_g = new char[]{'g'}, + charArray_h = new char[]{'h'}, charArray_i = new char[]{'i'}, + charArray_j = new char[]{'j'}, charArray_k = new char[]{'k'}, + charArray_l = new char[]{'l'}, charArray_m = new char[]{'m'}, + charArray_n = new char[]{'n'}, charArray_o = new char[]{'o'}, + charArray_p = new char[]{'p'}, charArray_q = new char[]{'q'}, + charArray_r = new char[]{'r'}, charArray_s = new char[]{'s'}, + charArray_t = new char[]{'t'}, charArray_u = new char[]{'u'}, + charArray_v = new char[]{'v'}, charArray_w = new char[]{'w'}, + charArray_x = new char[]{'x'}, charArray_y = new char[]{'y'}, + charArray_z = new char[]{'z'}; + static final char[] initCharArray = new char[]{'\u0000', '\u0000', '\u0000', + '\u0000', '\u0000', '\u0000'}; static final int TableSize = 30, InternalTableSize = 6; //30*6 = 180 entries public static final int OptimizedLength = 6; - public /* static */ + public/* static */ final char[][][][] charArray_length = new char[OptimizedLength][TableSize][InternalTableSize][]; // support for detecting non-externalized string literals int currentLineNr = -1; @@ -134,8 +110,7 @@ public class Scanner implements IScanner, ITerminalSymbols { public StringLiteral[] nonNLSStrings = null; public boolean checkNonExternalizedStringLiterals = true; public boolean wasNonExternalizedStringLiteral = false; - - /* static */ { + /* static */{ for (int i = 0; i < 6; i++) { for (int j = 0; j < TableSize; j++) { for (int k = 0; k < InternalTableSize; k++) { @@ -144,13 +119,12 @@ public class Scanner implements IScanner, ITerminalSymbols { } } } - static int newEntry2 = 0, newEntry3 = 0, newEntry4 = 0, newEntry5 = 0, newEntry6 = 0; - + static int newEntry2 = 0, newEntry3 = 0, newEntry4 = 0, newEntry5 = 0, + newEntry6 = 0; public static final int RoundBracket = 0; public static final int SquareBracket = 1; public static final int CurlyBracket = 2; public static final int BracketKinds = 3; - // task tag support public char[][] foundTaskTags = null; public char[][] foundTaskMessages; @@ -159,39 +133,31 @@ public class Scanner implements IScanner, ITerminalSymbols { public int foundTaskCount = 0; public char[][] taskTags = null; public char[][] taskPriorities = null; - - public static final boolean DEBUG = true; - - public Scanner() { - this(false, false); - } - public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace) { - this(tokenizeComments, tokenizeWhiteSpace, false); - } + public static final boolean DEBUG = false; + public static final boolean TRACE = false; /** - * Determines if the specified character is permissible as the first character in a PHP identifier + * Determines if the specified character is permissible as the first + * character in a PHP identifier */ public static boolean isPHPIdentifierStart(char ch) { return Character.isLetter(ch) || (ch == '_') || (0x7F <= ch && ch <= 0xFF); } - /** - * Determines if the specified character may be part of a PHP identifier as other than the first character + * Determines if the specified character may be part of a PHP identifier as + * other than the first character */ public static boolean isPHPIdentifierPart(char ch) { - return Character.isLetterOrDigit(ch) || (ch == '_') || (0x7F <= ch && ch <= 0xFF); + return Character.isLetterOrDigit(ch) || (ch == '_') + || (0x7F <= ch && ch <= 0xFF); } - public final boolean atEnd() { // This code is not relevant if source is // Only a part of the real stream input - return source.length == currentPosition; } public char[] getCurrentIdentifierSource() { //return the token REAL source (aka unicodes are precomputed) - char[] result; // if (withoutUnicodePtr != 0) // //0 is used as a fast test flag so the real first char is in position 1 @@ -218,17 +184,16 @@ public class Scanner implements IScanner, ITerminalSymbols { return optimizedCurrentTokenSource6(); } //no optimization - System.arraycopy(source, startPosition, result = new char[length], 0, length); + System.arraycopy(source, startPosition, result = new char[length], 0, + length); // } return result; } public int getCurrentTokenEndPosition() { return this.currentPosition - 1; } - public final char[] getCurrentTokenSource() { // Return the token REAL source (aka unicodes are precomputed) - char[] result; // if (withoutUnicodePtr != 0) // // 0 is used as a fast test flag so the real first char is in position 1 @@ -240,14 +205,13 @@ public class Scanner implements IScanner, ITerminalSymbols { // withoutUnicodePtr); // else { int length; - System.arraycopy(source, startPosition, result = new char[length = currentPosition - startPosition], 0, length); + System.arraycopy(source, startPosition, + result = new char[length = currentPosition - startPosition], 0, length); // } return result; } - public final char[] getCurrentTokenSource(int startPos) { // Return the token REAL source (aka unicodes are precomputed) - char[] result; // if (withoutUnicodePtr != 0) // // 0 is used as a fast test flag so the real first char is in position 1 @@ -259,81 +223,79 @@ public class Scanner implements IScanner, ITerminalSymbols { // withoutUnicodePtr); // else { int length; - System.arraycopy(source, startPos, result = new char[length = currentPosition - startPos], 0, length); + System.arraycopy(source, startPos, + result = new char[length = currentPosition - startPos], 0, length); // } return result; } - public final char[] getCurrentTokenSourceString() { //return the token REAL source (aka unicodes are precomputed). //REMOVE the two " that are at the beginning and the end. - char[] result; if (withoutUnicodePtr != 0) //0 is used as a fast test flag so the real first char is in position 1 - System.arraycopy(withoutUnicodeBuffer, 2, + System.arraycopy(withoutUnicodeBuffer, 2, //2 is 1 (real start) + 1 (to jump over the ") - result = new char[withoutUnicodePtr - 2], 0, withoutUnicodePtr - 2); + result = new char[withoutUnicodePtr - 2], 0, withoutUnicodePtr - 2); else { int length; - System.arraycopy(source, startPosition + 1, result = new char[length = currentPosition - startPosition - 2], 0, length); + System.arraycopy(source, startPosition + 1, + result = new char[length = currentPosition - startPosition - 2], 0, + length); } return result; } public int getCurrentTokenStartPosition() { return this.startPosition; } - public final char[] getCurrentStringLiteralSource() { // Return the token REAL source (aka unicodes are precomputed) - char[] result; - int length; - System.arraycopy(source, startPosition + 1, result = new char[length = currentPosition - startPosition - 2], 0, length); + System.arraycopy(source, startPosition + 1, + result = new char[length = currentPosition - startPosition - 2], 0, + length); // } return result; } - /* * Search the source position corresponding to the end of a given line number * - * Line numbers are 1-based, and relative to the scanner initialPosition. Character positions are 0-based. + * Line numbers are 1-based, and relative to the scanner initialPosition. + * Character positions are 0-based. * * In case the given line number is inconsistent, answers -1. */ public final int getLineEnd(int lineNumber) { - if (lineEnds == null) return -1; if (lineNumber >= lineEnds.length) return -1; if (lineNumber <= 0) return -1; - if (lineNumber == lineEnds.length - 1) return eofPosition; return lineEnds[lineNumber - 1]; // next line start one character behind the lineEnd of the previous line } /** - * Search the source position corresponding to the beginning of a given line number + * Search the source position corresponding to the beginning of a given line + * number * - * Line numbers are 1-based, and relative to the scanner initialPosition. Character positions are 0-based. + * Line numbers are 1-based, and relative to the scanner initialPosition. + * Character positions are 0-based. * * e.g. getLineStart(1) --> 0 i.e. first line starts at character 0. * * In case the given line number is inconsistent, answers -1. */ public final int getLineStart(int lineNumber) { - if (lineEnds == null) return -1; if (lineNumber >= lineEnds.length) return -1; if (lineNumber <= 0) return -1; - if (lineNumber == 1) return initialPosition; return lineEnds[lineNumber - 2] + 1; @@ -342,14 +304,14 @@ public class Scanner implements IScanner, ITerminalSymbols { public final boolean getNextChar(char testedChar) { //BOOLEAN //handle the case of unicode. - //when a unicode appears then we must use a buffer that holds char internal values + //when a unicode appears then we must use a buffer that holds char + // internal values //At the end of this method currentCharacter holds the new visited char //and currentPosition points right next after it - //Both previous lines are true if the currentCharacter is == to the testedChar + //Both previous lines are true if the currentCharacter is == to the + // testedChar //On false, no side effect has occured. - //ALL getNextChar.... ARE OPTIMIZED COPIES - int temp = currentPosition; try { currentCharacter = source[currentPosition++]; @@ -419,14 +381,14 @@ public class Scanner implements IScanner, ITerminalSymbols { //INT 0 : testChar1 \\\\///\\\\ 1 : testedChar2 \\\\///\\\\ -1 : others //test can be done with (x==0) for the first and (x>0) for the second //handle the case of unicode. - //when a unicode appears then we must use a buffer that holds char internal values + //when a unicode appears then we must use a buffer that holds char + // internal values //At the end of this method currentCharacter holds the new visited char //and currentPosition points right next after it - //Both previous lines are true if the currentCharacter is == to the testedChar1/2 + //Both previous lines are true if the currentCharacter is == to the + // testedChar1/2 //On false, no side effect has occured. - //ALL getNextChar.... ARE OPTIMIZED COPIES - int temp = currentPosition; try { int result; @@ -488,7 +450,6 @@ public class Scanner implements IScanner, ITerminalSymbols { currentPosition = temp; return -1; } - // if (withoutUnicodePtr != 0) // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; return result; @@ -501,14 +462,13 @@ public class Scanner implements IScanner, ITerminalSymbols { public final boolean getNextCharAsDigit() { //BOOLEAN //handle the case of unicode. - //when a unicode appears then we must use a buffer that holds char internal values + //when a unicode appears then we must use a buffer that holds char + // internal values //At the end of this method currentCharacter holds the new visited char //and currentPosition points right next after it //Both previous lines are true if the currentCharacter is a digit //On false, no side effect has occured. - //ALL getNextChar.... ARE OPTIMIZED COPIES - int temp = currentPosition; try { currentCharacter = source[currentPosition++]; @@ -573,14 +533,14 @@ public class Scanner implements IScanner, ITerminalSymbols { public final boolean getNextCharAsDigit(int radix) { //BOOLEAN //handle the case of unicode. - //when a unicode appears then we must use a buffer that holds char internal values + //when a unicode appears then we must use a buffer that holds char + // internal values //At the end of this method currentCharacter holds the new visited char //and currentPosition points right next after it - //Both previous lines are true if the currentCharacter is a digit base on radix + //Both previous lines are true if the currentCharacter is a digit base on + // radix //On false, no side effect has occured. - //ALL getNextChar.... ARE OPTIMIZED COPIES - int temp = currentPosition; try { currentCharacter = source[currentPosition++]; @@ -645,14 +605,14 @@ public class Scanner implements IScanner, ITerminalSymbols { public boolean getNextCharAsJavaIdentifierPart() { //BOOLEAN //handle the case of unicode. - //when a unicode appears then we must use a buffer that holds char internal values + //when a unicode appears then we must use a buffer that holds char + // internal values //At the end of this method currentCharacter holds the new visited char //and currentPosition points right next after it - //Both previous lines are true if the currentCharacter is a JavaIdentifierPart + //Both previous lines are true if the currentCharacter is a + // JavaIdentifierPart //On false, no side effect has occured. - //ALL getNextChar.... ARE OPTIMIZED COPIES - int temp = currentPosition; try { currentCharacter = source[currentPosition++]; @@ -705,7 +665,6 @@ public class Scanner implements IScanner, ITerminalSymbols { currentPosition = temp; return false; } - // if (withoutUnicodePtr != 0) // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; return true; @@ -715,67 +674,425 @@ public class Scanner implements IScanner, ITerminalSymbols { return false; } } - - public int getNextToken() throws InvalidInputException { - int htmlPosition = currentPosition; + public int getCastOrParen() { + int tempPosition = currentPosition; + char tempCharacter = currentCharacter; + int tempToken = TokenNameLPAREN; + boolean found = false; + StringBuffer buf = new StringBuffer(); try { - while (!phpMode) { + do { currentCharacter = source[currentPosition++]; - if (currentCharacter == '<') { - if (getNextChar('?')) { - currentCharacter = source[currentPosition++]; - if ((currentCharacter == ' ') || Character.isWhitespace(currentCharacter)) { - // = 'a' && currentCharacter <= 'z') + || (currentCharacter >= 'A' && currentCharacter <= 'Z')) { + buf.append(currentCharacter); + currentCharacter = source[currentPosition++]; + } + if (buf.length() >= 3 && buf.length() <= 7) { + char[] data = buf.toString().toCharArray(); + int index = 0; + switch (data.length) { + case 3 : + // int + if ((data[index] == 'i') && (data[++index] == 'n') + && (data[++index] == 't')) { + found = true; + tempToken = TokenNameintCAST; + } + break; + case 4 : + // bool real + if ((data[index] == 'b') && (data[++index] == 'o') + && (data[++index] == 'o') && (data[++index] == 'l')) { + found = true; + tempToken = TokenNameboolCAST; + } else { + index = 0; + if ((data[index] == 'r') && (data[++index] == 'e') + && (data[++index] == 'a') && (data[++index] == 'l')) { + found = true; + tempToken = TokenNamedoubleCAST; } + } + break; + case 5 : + // array unset float + if ((data[index] == 'a') && (data[++index] == 'r') + && (data[++index] == 'r') && (data[++index] == 'a') + && (data[++index] == 'y')) { + found = true; + tempToken = TokenNamearrayCAST; } else { - boolean phpStart = (currentCharacter == 'P') || (currentCharacter == 'p'); - if (phpStart) { - int test = getNextChar('H', 'h'); - if (test >= 0) { - test = getNextChar('P', 'p'); - if (test >= 0) { - // source.length ? TokenNameEOF : TokenNameRBRACE; } try { - while (true) { //loop for jumping over comments + while (true) { withoutUnicodePtr = 0; - //start with a new token (even comment written with unicode ) - + //start with a new token + char encapsedChar = ' '; + if (!encapsedStringStack.isEmpty()) { + encapsedChar = ((Character) encapsedStringStack.peek()).charValue(); + } + if (encapsedChar != '$' && encapsedChar != ' ') { + currentCharacter = source[currentPosition++]; + if (currentCharacter == encapsedChar) { + switch (currentCharacter) { + case '`' : + return TokenNameEncapsedString0; + case '\'' : + return TokenNameEncapsedString1; + case '"' : + return TokenNameEncapsedString2; + } + } + while (currentCharacter != encapsedChar) { + /** ** in PHP \r and \n are valid in string literals *** */ + switch (currentCharacter) { + case '\\' : + int escapeSize = currentPosition; + boolean backSlashAsUnicodeInString = unicodeAsBackSlash; + //scanEscapeCharacter make a side effect on this value and + // we need the previous value few lines down this one + scanDoubleQuotedEscapeCharacter(); + escapeSize = currentPosition - escapeSize; + if (withoutUnicodePtr == 0) { + //buffer all the entries that have been left aside.... + withoutUnicodePtr = currentPosition - escapeSize - 1 + - startPosition; + System.arraycopy(source, startPosition, + withoutUnicodeBuffer, 1, withoutUnicodePtr); + withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; + } else { //overwrite the / in the buffer + withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter; + if (backSlashAsUnicodeInString) { //there are TWO \ in + withoutUnicodePtr--; + } + } + break; + case '$' : + if (isPHPIdentifierStart(source[currentPosition]) + || source[currentPosition] == '{') { + currentPosition--; + encapsedStringStack.push(new Character('$')); + return TokenNameSTRING; + } + break; + case '{' : + if (source[currentPosition] == '$') { // CURLY_OPEN + currentPosition--; + encapsedStringStack.push(new Character('$')); + return TokenNameSTRING; + } + } + // consume next character + unicodeAsBackSlash = false; + currentCharacter = source[currentPosition++]; + if (withoutUnicodePtr != 0) { + withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; + } + // } + } // end while + currentPosition--; + return TokenNameSTRING; + } // ---------Consume white space and handles startPosition--------- int whiteStart = currentPosition; - boolean isWhiteSpace; - do { + startPosition = currentPosition; + currentCharacter = source[currentPosition++]; + if (encapsedChar == '$') { + switch (currentCharacter) { + case '\\' : + currentCharacter = source[currentPosition++]; + return TokenNameSTRING; + case '{' : + if (encapsedChar == '$') { + if (getNextChar('$')) + return TokenNameCURLY_OPEN; + } + return TokenNameLBRACE; + case '}' : + return TokenNameRBRACE; + case '[' : + return TokenNameLBRACKET; + case ']' : + return TokenNameRBRACKET; + case '\'' : + if (tokenizeStrings) { + consumeStringConstant(); + return TokenNameStringConstant; + } + return TokenNameEncapsedString1; + case '"' : + return TokenNameEncapsedString2; + case '`' : + if (tokenizeStrings) { + consumeStringInterpolated(); + return TokenNameStringInterpolated; + } + return TokenNameEncapsedString0; + case '-' : + if (getNextChar('>')) + return TokenNameMINUS_GREATER; + return TokenNameSTRING; + default : + if (currentCharacter == '$') { + int oldPosition = currentPosition; + try { + currentCharacter = source[currentPosition++]; + if (currentCharacter == '{') { + return TokenNameDOLLAR_LBRACE; + } + if (isPHPIdentifierStart(currentCharacter)) { + return scanIdentifierOrKeyword(true); + } else { + currentPosition = oldPosition; + return TokenNameSTRING; + } + } catch (IndexOutOfBoundsException e) { + currentPosition = oldPosition; + return TokenNameSTRING; + } + } + if (isPHPIdentifierStart(currentCharacter)) + return scanIdentifierOrKeyword(false); + if (Character.isDigit(currentCharacter)) + return scanNumber(false); + return TokenNameERROR; + } + } + // boolean isWhiteSpace; + + while ((currentCharacter == ' ') + || Character.isWhitespace(currentCharacter)) { startPosition = currentPosition; currentCharacter = source[currentPosition++]; // if (((currentCharacter = source[currentPosition++]) == '\\') @@ -806,9 +1253,10 @@ public class Scanner implements IScanner, ITerminalSymbols { currentLine = null; } } - isWhiteSpace = (currentCharacter == ' ') || Character.isWhitespace(currentCharacter); + // isWhiteSpace = (currentCharacter == ' ') + // || Character.isWhitespace(currentCharacter); // } - } while (isWhiteSpace); + } if (tokenizeWhiteSpace && (whiteStart != currentPosition - 1)) { // reposition scanner in case we are interested by spaces as tokens currentPosition--; @@ -818,12 +1266,10 @@ public class Scanner implements IScanner, ITerminalSymbols { //little trick to get out in the middle of a source compuation if (currentPosition > eofPosition) return TokenNameEOF; - // ---------Identify the next token------------- - switch (currentCharacter) { case '(' : - return TokenNameLPAREN; + return getCastOrParen(); case ')' : return TokenNameRPAREN; case '{' : @@ -838,8 +1284,9 @@ public class Scanner implements IScanner, ITerminalSymbols { return TokenNameSEMICOLON; case ',' : return TokenNameCOMMA; - case '.' : + if (getNextChar('=')) + return TokenNameDOT_EQUAL; if (getNextCharAsDigit()) return scanNumber(true); return TokenNameDOT; @@ -861,7 +1308,6 @@ public class Scanner implements IScanner, ITerminalSymbols { return TokenNameMINUS_EQUAL; if (getNextChar('>')) return TokenNameMINUS_GREATER; - return TokenNameMINUS; } case '~' : @@ -886,59 +1332,69 @@ public class Scanner implements IScanner, ITerminalSymbols { return TokenNameREMAINDER; case '<' : { - int test; - if ((test = getNextChar('=', '<')) == 0) - return TokenNameLESS_EQUAL; - if (test > 0) { - if (getNextChar('=')) - return TokenNameLEFT_SHIFT_EQUAL; - if (getNextChar('<')) { - int heredocStart = currentPosition; - int heredocLength = 0; - currentCharacter = source[currentPosition++]; - if (isPHPIdentifierStart(currentCharacter)) { - currentCharacter = source[currentPosition++]; - } else { - return TokenNameERROR; - } - while (isPHPIdentifierPart(currentCharacter)) { + int oldPosition = currentPosition; + try { + currentCharacter = source[currentPosition++]; + } catch (IndexOutOfBoundsException e) { + currentPosition = oldPosition; + return TokenNameLESS; + } + switch (currentCharacter) { + case '=' : + return TokenNameLESS_EQUAL; + case '>' : + return TokenNameNOT_EQUAL; + case '<' : + if (getNextChar('=')) + return TokenNameLEFT_SHIFT_EQUAL; + if (getNextChar('<')) { currentCharacter = source[currentPosition++]; - } - - heredocLength = currentPosition - heredocStart - 1; - - // heredoc end-tag determination - boolean endTag = true; - char ch; - do { - ch = source[currentPosition++]; - if (ch == '\r' || ch == '\n') { - if (recordLineSeparator) { - pushLineSeparator(); - } else { - currentLine = null; - } - for (int i = 0; i < heredocLength; i++) { - if (source[currentPosition + i] != source[heredocStart + i]) { - endTag = false; - break; + while (Character.isWhitespace(currentCharacter)) { + currentCharacter = source[currentPosition++]; + } + int heredocStart = currentPosition - 1; + int heredocLength = 0; + if (isPHPIdentifierStart(currentCharacter)) { + currentCharacter = source[currentPosition++]; + } else { + return TokenNameERROR; + } + while (isPHPIdentifierPart(currentCharacter)) { + currentCharacter = source[currentPosition++]; + } + heredocLength = currentPosition - heredocStart - 1; + // heredoc end-tag determination + boolean endTag = true; + char ch; + do { + ch = source[currentPosition++]; + if (ch == '\r' || ch == '\n') { + if (recordLineSeparator) { + pushLineSeparator(); + } else { + currentLine = null; + } + for (int i = 0; i < heredocLength; i++) { + if (source[currentPosition + i] != source[heredocStart + + i]) { + endTag = false; + break; + } + } + if (endTag) { + currentPosition += heredocLength - 1; + currentCharacter = source[currentPosition++]; + break; // do...while loop + } else { + endTag = true; } } - if (endTag) { - currentPosition += heredocLength - 1; - currentCharacter = source[currentPosition++]; - break; // do...while loop - } else { - endTag = true; - } - } - - } while (true); - - return TokenNameHEREDOC; - } - return TokenNameLEFT_SHIFT; + } while (true); + return TokenNameHEREDOC; + } + return TokenNameLEFT_SHIFT; } + currentPosition = oldPosition; return TokenNameLESS; } case '>' : @@ -988,367 +1444,43 @@ public class Scanner implements IScanner, ITerminalSymbols { case '?' : if (getNextChar('>')) { phpMode = false; - return TokenNameStopPHP; + if (currentPosition == source.length) { + phpMode = true; + return TokenNameINLINE_HTML; + } + return getInlinedHTML(currentPosition - 2); } return TokenNameQUESTION; case ':' : if (getNextChar(':')) - return TokenNameCOLON_COLON; + return TokenNamePAAMAYIM_NEKUDOTAYIM; return TokenNameCOLON; case '@' : return TokenNameAT; - // case '\'' : - // { - // int test; - // if ((test = getNextChar('\n', '\r')) == 0) { - // throw new InvalidInputException(INVALID_CHARACTER_CONSTANT); - // } - // if (test > 0) { - // // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - // for (int lookAhead = 0; - // lookAhead < 3; - // lookAhead++) { - // if (currentPosition + lookAhead - // == source.length) - // break; - // if (source[currentPosition + lookAhead] - // == '\n') - // break; - // if (source[currentPosition + lookAhead] - // == '\'') { - // currentPosition += lookAhead + 1; - // break; - // } - // } - // throw new InvalidInputException(INVALID_CHARACTER_CONSTANT); - // } - // } - // if (getNextChar('\'')) { - // // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - // for (int lookAhead = 0; - // lookAhead < 3; - // lookAhead++) { - // if (currentPosition + lookAhead - // == source.length) - // break; - // if (source[currentPosition + lookAhead] - // == '\n') - // break; - // if (source[currentPosition + lookAhead] - // == '\'') { - // currentPosition += lookAhead + 1; - // break; - // } - // } - // throw new InvalidInputException(INVALID_CHARACTER_CONSTANT); - // } - // if (getNextChar('\\')) - // scanEscapeCharacter(); - // else { // consume next character - // unicodeAsBackSlash = false; - // if (((currentCharacter = source[currentPosition++]) - // == '\\') - // && (source[currentPosition] == 'u')) { - // getNextUnicodeChar(); - // } else { - // if (withoutUnicodePtr != 0) { - // withoutUnicodeBuffer[++withoutUnicodePtr] = - // currentCharacter; - // } - // } - // } - // // if (getNextChar('\'')) - // // return TokenNameCharacterLiteral; - // // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - // for (int lookAhead = 0; lookAhead < 20; lookAhead++) { - // if (currentPosition + lookAhead == source.length) - // break; - // if (source[currentPosition + lookAhead] == '\n') - // break; - // if (source[currentPosition + lookAhead] == '\'') { - // currentPosition += lookAhead + 1; - // break; - // } - // } - // throw new InvalidInputException(INVALID_CHARACTER_CONSTANT); case '\'' : - try { - // consume next character - unicodeAsBackSlash = false; - currentCharacter = source[currentPosition++]; - // if (((currentCharacter = source[currentPosition++]) == '\\') - // && (source[currentPosition] == 'u')) { - // getNextUnicodeChar(); - // } else { - // if (withoutUnicodePtr != 0) { - // withoutUnicodeBuffer[++withoutUnicodePtr] = - // currentCharacter; - // } - // } - - while (currentCharacter != '\'') { - - /** ** in PHP \r and \n are valid in string literals *** */ - // if ((currentCharacter == '\n') - // || (currentCharacter == '\r')) { - // // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - // for (int lookAhead = 0; lookAhead < 50; lookAhead++) { - // if (currentPosition + lookAhead == source.length) - // break; - // if (source[currentPosition + lookAhead] == '\n') - // break; - // if (source[currentPosition + lookAhead] == '\"') { - // currentPosition += lookAhead + 1; - // break; - // } - // } - // throw new InvalidInputException(INVALID_CHAR_IN_STRING); - // } - if (currentCharacter == '\\') { - int escapeSize = currentPosition; - boolean backSlashAsUnicodeInString = unicodeAsBackSlash; - //scanEscapeCharacter make a side effect on this value and we need the previous value few lines down this one - scanSingleQuotedEscapeCharacter(); - escapeSize = currentPosition - escapeSize; - if (withoutUnicodePtr == 0) { - //buffer all the entries that have been left aside.... - withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition; - System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr); - withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; - } else { //overwrite the / in the buffer - withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter; - if (backSlashAsUnicodeInString) { //there are TWO \ in the stream where only one is correct - withoutUnicodePtr--; - } - } - } - // consume next character - unicodeAsBackSlash = false; - currentCharacter = source[currentPosition++]; - // if (((currentCharacter = source[currentPosition++]) == '\\') - // && (source[currentPosition] == 'u')) { - // getNextUnicodeChar(); - // } else { - if (withoutUnicodePtr != 0) { - withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; - } - // } - - } - } catch (IndexOutOfBoundsException e) { - throw new InvalidInputException(UNTERMINATED_STRING); - } catch (InvalidInputException e) { - if (e.getMessage().equals(INVALID_ESCAPE)) { - // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - for (int lookAhead = 0; lookAhead < 50; lookAhead++) { - if (currentPosition + lookAhead == source.length) - break; - if (source[currentPosition + lookAhead] == '\n') - break; - if (source[currentPosition + lookAhead] == '\'') { - currentPosition += lookAhead + 1; - break; - } - } - - } - throw e; // rethrow - } - if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags //$NON-NLS-?$ where ? is an int. - if (currentLine == null) { - currentLine = new NLSLine(); - lines.add(currentLine); - } - currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1)); - } + consumeStringConstant(); return TokenNameStringConstant; case '"' : - try { - // consume next character - unicodeAsBackSlash = false; - currentCharacter = source[currentPosition++]; - // if (((currentCharacter = source[currentPosition++]) == '\\') - // && (source[currentPosition] == 'u')) { - // getNextUnicodeChar(); - // } else { - // if (withoutUnicodePtr != 0) { - // withoutUnicodeBuffer[++withoutUnicodePtr] = - // currentCharacter; - // } - // } - - while (currentCharacter != '"') { - - /** ** in PHP \r and \n are valid in string literals *** */ - // if ((currentCharacter == '\n') - // || (currentCharacter == '\r')) { - // // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - // for (int lookAhead = 0; lookAhead < 50; lookAhead++) { - // if (currentPosition + lookAhead == source.length) - // break; - // if (source[currentPosition + lookAhead] == '\n') - // break; - // if (source[currentPosition + lookAhead] == '\"') { - // currentPosition += lookAhead + 1; - // break; - // } - // } - // throw new InvalidInputException(INVALID_CHAR_IN_STRING); - // } - if (currentCharacter == '\\') { - int escapeSize = currentPosition; - boolean backSlashAsUnicodeInString = unicodeAsBackSlash; - //scanEscapeCharacter make a side effect on this value and we need the previous value few lines down this one - scanDoubleQuotedEscapeCharacter(); - escapeSize = currentPosition - escapeSize; - if (withoutUnicodePtr == 0) { - //buffer all the entries that have been left aside.... - withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition; - System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr); - withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; - } else { //overwrite the / in the buffer - withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter; - if (backSlashAsUnicodeInString) { //there are TWO \ in the stream where only one is correct - withoutUnicodePtr--; - } - } - } - // consume next character - unicodeAsBackSlash = false; - currentCharacter = source[currentPosition++]; - // if (((currentCharacter = source[currentPosition++]) == '\\') - // && (source[currentPosition] == 'u')) { - // getNextUnicodeChar(); - // } else { - if (withoutUnicodePtr != 0) { - withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; - } - // } - - } - } catch (IndexOutOfBoundsException e) { - throw new InvalidInputException(UNTERMINATED_STRING); - } catch (InvalidInputException e) { - if (e.getMessage().equals(INVALID_ESCAPE)) { - // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - for (int lookAhead = 0; lookAhead < 50; lookAhead++) { - if (currentPosition + lookAhead == source.length) - break; - if (source[currentPosition + lookAhead] == '\n') - break; - if (source[currentPosition + lookAhead] == '\"') { - currentPosition += lookAhead + 1; - break; - } - } - - } - throw e; // rethrow + if (tokenizeStrings) { + consumeStringLiteral(); + return TokenNameStringLiteral; } - if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags //$NON-NLS-?$ where ? is an int. - if (currentLine == null) { - currentLine = new NLSLine(); - lines.add(currentLine); - } - currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1)); - } - return TokenNameStringLiteral; + return TokenNameEncapsedString2; case '`' : - try { - // consume next character - unicodeAsBackSlash = false; - currentCharacter = source[currentPosition++]; - // if (((currentCharacter = source[currentPosition++]) == '\\') - // && (source[currentPosition] == 'u')) { - // getNextUnicodeChar(); - // } else { - // if (withoutUnicodePtr != 0) { - // withoutUnicodeBuffer[++withoutUnicodePtr] = - // currentCharacter; - // } - // } - - while (currentCharacter != '`') { - - /** ** in PHP \r and \n are valid in string literals *** */ - // if ((currentCharacter == '\n') - // || (currentCharacter == '\r')) { - // // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - // for (int lookAhead = 0; lookAhead < 50; lookAhead++) { - // if (currentPosition + lookAhead == source.length) - // break; - // if (source[currentPosition + lookAhead] == '\n') - // break; - // if (source[currentPosition + lookAhead] == '\"') { - // currentPosition += lookAhead + 1; - // break; - // } - // } - // throw new InvalidInputException(INVALID_CHAR_IN_STRING); - // } - if (currentCharacter == '\\') { - int escapeSize = currentPosition; - boolean backSlashAsUnicodeInString = unicodeAsBackSlash; - //scanEscapeCharacter make a side effect on this value and we need the previous value few lines down this one - scanDoubleQuotedEscapeCharacter(); - escapeSize = currentPosition - escapeSize; - if (withoutUnicodePtr == 0) { - //buffer all the entries that have been left aside.... - withoutUnicodePtr = currentPosition - escapeSize - 1 - startPosition; - System.arraycopy(source, startPosition, withoutUnicodeBuffer, 1, withoutUnicodePtr); - withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; - } else { //overwrite the / in the buffer - withoutUnicodeBuffer[withoutUnicodePtr] = currentCharacter; - if (backSlashAsUnicodeInString) { //there are TWO \ in the stream where only one is correct - withoutUnicodePtr--; - } - } - } - // consume next character - unicodeAsBackSlash = false; - currentCharacter = source[currentPosition++]; - // if (((currentCharacter = source[currentPosition++]) == '\\') - // && (source[currentPosition] == 'u')) { - // getNextUnicodeChar(); - // } else { - if (withoutUnicodePtr != 0) { - withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; - } - // } - - } - } catch (IndexOutOfBoundsException e) { - throw new InvalidInputException(UNTERMINATED_STRING); - } catch (InvalidInputException e) { - if (e.getMessage().equals(INVALID_ESCAPE)) { - // relocate if finding another quote fairly close: thus unicode '/u000D' will be fully consumed - for (int lookAhead = 0; lookAhead < 50; lookAhead++) { - if (currentPosition + lookAhead == source.length) - break; - if (source[currentPosition + lookAhead] == '\n') - break; - if (source[currentPosition + lookAhead] == '`') { - currentPosition += lookAhead + 1; - break; - } - } - - } - throw e; // rethrow + if (tokenizeStrings) { + consumeStringInterpolated(); + return TokenNameStringInterpolated; } - if (checkNonExternalizedStringLiterals) { // check for presence of NLS tags //$NON-NLS-?$ where ? is an int. - if (currentLine == null) { - currentLine = new NLSLine(); - lines.add(currentLine); - } - currentLine.add(new StringLiteral(getCurrentTokenSourceString(), startPosition, currentPosition - 1)); - } - return TokenNameStringInterpolated; + return TokenNameEncapsedString0; case '#' : case '/' : { + char startChar = currentCharacter; + if (getNextChar('=')) { + return TokenNameDIVIDE_EQUAL; + } int test; - if ((currentCharacter == '#') || (test = getNextChar('/', '*')) == 0) { + if ((startChar == '#') || (test = getNextChar('/', '*')) == 0) { //line comment int endPositionForLineComment = 0; try { //get the next char @@ -1378,13 +1510,13 @@ public class Scanner implements IScanner, ITerminalSymbols { // Character.getNumericValue(source[currentPosition++])) // > 15 // || c4 < 0) { - // throw new InvalidInputException(INVALID_UNICODE_ESCAPE); + // throw new + // InvalidInputException(INVALID_UNICODE_ESCAPE); // } else { // currentCharacter = // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4); // } // } - //handle the \\u case manually into comment // if (currentCharacter == '\\') { // if (source[currentPosition] == '\\') @@ -1396,10 +1528,9 @@ public class Scanner implements IScanner, ITerminalSymbols { if (getNextChar('>')) { startPosition = currentPosition - 2; phpMode = false; - return TokenNameStopPHP; + return TokenNameINLINE_HTML; } } - //get the next char isUnicode = false; currentCharacter = source[currentPosition++]; @@ -1432,7 +1563,8 @@ public class Scanner implements IScanner, ITerminalSymbols { // source[currentPosition++])) // > 15 // || c4 < 0) { - // throw new InvalidInputException(INVALID_UNICODE_ESCAPE); + // throw new + // InvalidInputException(INVALID_UNICODE_ESCAPE); // } else { // currentCharacter = // (char) (((c1 * 16 + c2) * 16 + c3) * 16 + c4); @@ -1450,7 +1582,9 @@ public class Scanner implements IScanner, ITerminalSymbols { endPositionForLineComment = currentPosition - 1; } recordComment(false); - if ((currentCharacter == '\r') || (currentCharacter == '\n')) { + if (this.taskTags != null) checkTaskTag(this.startPosition, this.currentPosition); + if ((currentCharacter == '\r') + || (currentCharacter == '\n')) { checkNonExternalizeString(); if (recordLineSeparator) { if (isUnicode) { @@ -1469,7 +1603,8 @@ public class Scanner implements IScanner, ITerminalSymbols { } return TokenNameCOMMENT_LINE; } - } catch (IndexOutOfBoundsException e) { //an eof will them be generated + } catch (IndexOutOfBoundsException e) { //an eof will them + // be generated if (tokenizeComments) { currentPosition--; // reset one character behind @@ -1484,7 +1619,8 @@ public class Scanner implements IScanner, ITerminalSymbols { // consume next character unicodeAsBackSlash = false; currentCharacter = source[currentPosition++]; - // if (((currentCharacter = source[currentPosition++]) == '\\') + // if (((currentCharacter = source[currentPosition++]) == + // '\\') // && (source[currentPosition] == 'u')) { // getNextUnicodeChar(); // } else { @@ -1493,7 +1629,6 @@ public class Scanner implements IScanner, ITerminalSymbols { // currentCharacter; // } // } - if (currentCharacter == '*') { isJavadoc = true; star = true; @@ -1526,7 +1661,8 @@ public class Scanner implements IScanner, ITerminalSymbols { } //loop until end of comment */ while ((currentCharacter != '/') || (!star)) { - if ((currentCharacter == '\r') || (currentCharacter == '\n')) { + if ((currentCharacter == '\r') + || (currentCharacter == '\n')) { checkNonExternalizeString(); if (recordLineSeparator) { pushLineSeparator(); @@ -1560,25 +1696,29 @@ public class Scanner implements IScanner, ITerminalSymbols { } break; } - if (getNextChar('=')) - return TokenNameDIVIDE_EQUAL; return TokenNameDIVIDE; } case '\u001a' : if (atEnd()) return TokenNameEOF; - //the atEnd may not be if source is only some part of a real (external) stream + //the atEnd may not be if + // source is only some part of a real (external) stream throw new InvalidInputException("Ctrl-Z"); //$NON-NLS-1$ - default : if (currentCharacter == '$') { - while ((currentCharacter = source[currentPosition++]) == '$') { + int oldPosition = currentPosition; + try { + currentCharacter = source[currentPosition++]; + if (isPHPIdentifierStart(currentCharacter)) { + return scanIdentifierOrKeyword(true); + } else { + currentPosition = oldPosition; + return TokenNameDOLLAR; + } + } catch (IndexOutOfBoundsException e) { + currentPosition = oldPosition; + return TokenNameDOLLAR; } - if (currentCharacter == '{') - return TokenNameDOLLAR_LBRACE; - if (isPHPIdentifierStart(currentCharacter)) - return scanIdentifierOrKeyword(true); - return TokenNameERROR; } if (isPHPIdentifierStart(currentCharacter)) return scanIdentifierOrKeyword(false); @@ -1593,12 +1733,69 @@ public class Scanner implements IScanner, ITerminalSymbols { } return TokenNameEOF; } - + /** + * @return @throws + * InvalidInputException + */ + private int getInlinedHTML(int start) throws InvalidInputException { + // int htmlPosition = start; + if (currentPosition > source.length) { + currentPosition = source.length; + return TokenNameEOF; + } + startPosition = start; + try { + while (!phpMode) { + currentCharacter = source[currentPosition++]; + if (currentCharacter == '<') { + if (getNextChar('?')) { + currentCharacter = source[currentPosition++]; + if ((currentCharacter == ' ') + || Character.isWhitespace(currentCharacter)) { + // = 0) { + test = getNextChar('P', 'p'); + if (test >= 0) { + // = InternalTableSize) max = 0; char[] r; - table[max] = (r = new char[] { c0, c1 }); + table[max] = (r = new char[]{c0, c1}); newEntry2 = max; return r; } - final char[] optimizedCurrentTokenSource3() { //try to return the same char[] build only once - char c0, c1, c2; - int hash = - (((c0 = source[startPosition]) << 12) + ((c1 = source[startPosition + 1]) << 6) + (c2 = source[startPosition + 2])) + int hash = (((c0 = source[startPosition]) << 12) + + ((c1 = source[startPosition + 1]) << 6) + (c2 = source[startPosition + 2])) % TableSize; char[][] table = charArray_length[1][hash]; int i = newEntry3; @@ -2133,26 +2328,23 @@ public class Scanner implements IScanner, ITerminalSymbols { if (++max >= InternalTableSize) max = 0; char[] r; - table[max] = (r = new char[] { c0, c1, c2 }); + table[max] = (r = new char[]{c0, c1, c2}); newEntry3 = max; return r; } - final char[] optimizedCurrentTokenSource4() { //try to return the same char[] build only once - char c0, c1, c2, c3; - long hash = - ((((long) (c0 = source[startPosition])) << 18) + long hash = ((((long) (c0 = source[startPosition])) << 18) + ((c1 = source[startPosition + 1]) << 12) - + ((c2 = source[startPosition + 2]) << 6) - + (c3 = source[startPosition + 3])) + + ((c2 = source[startPosition + 2]) << 6) + (c3 = source[startPosition + 3])) % TableSize; char[][] table = charArray_length[2][(int) hash]; int i = newEntry4; while (++i < InternalTableSize) { char[] charArray = table[i]; - if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3])) + if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) + && (c3 == charArray[3])) return charArray; } //---------other side--------- @@ -2160,35 +2352,32 @@ public class Scanner implements IScanner, ITerminalSymbols { int max = newEntry4; while (++i <= max) { char[] charArray = table[i]; - if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3])) + if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) + && (c3 == charArray[3])) return charArray; } //--------add the entry------- if (++max >= InternalTableSize) max = 0; char[] r; - table[max] = (r = new char[] { c0, c1, c2, c3 }); + table[max] = (r = new char[]{c0, c1, c2, c3}); newEntry4 = max; return r; - } - final char[] optimizedCurrentTokenSource5() { //try to return the same char[] build only once - char c0, c1, c2, c3, c4; - long hash = - ((((long) (c0 = source[startPosition])) << 24) + long hash = ((((long) (c0 = source[startPosition])) << 24) + (((long) (c1 = source[startPosition + 1])) << 18) + ((c2 = source[startPosition + 2]) << 12) - + ((c3 = source[startPosition + 3]) << 6) - + (c4 = source[startPosition + 4])) + + ((c3 = source[startPosition + 3]) << 6) + (c4 = source[startPosition + 4])) % TableSize; char[][] table = charArray_length[3][(int) hash]; int i = newEntry5; while (++i < InternalTableSize) { char[] charArray = table[i]; - if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4])) + if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) + && (c3 == charArray[3]) && (c4 == charArray[4])) return charArray; } //---------other side--------- @@ -2196,41 +2385,34 @@ public class Scanner implements IScanner, ITerminalSymbols { int max = newEntry5; while (++i <= max) { char[] charArray = table[i]; - if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) && (c3 == charArray[3]) && (c4 == charArray[4])) + if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) + && (c3 == charArray[3]) && (c4 == charArray[4])) return charArray; } //--------add the entry------- if (++max >= InternalTableSize) max = 0; char[] r; - table[max] = (r = new char[] { c0, c1, c2, c3, c4 }); + table[max] = (r = new char[]{c0, c1, c2, c3, c4}); newEntry5 = max; return r; - } - final char[] optimizedCurrentTokenSource6() { //try to return the same char[] build only once - char c0, c1, c2, c3, c4, c5; - long hash = - ((((long) (c0 = source[startPosition])) << 32) + long hash = ((((long) (c0 = source[startPosition])) << 32) + (((long) (c1 = source[startPosition + 1])) << 24) + (((long) (c2 = source[startPosition + 2])) << 18) + ((c3 = source[startPosition + 3]) << 12) - + ((c4 = source[startPosition + 4]) << 6) - + (c5 = source[startPosition + 5])) + + ((c4 = source[startPosition + 4]) << 6) + (c5 = source[startPosition + 5])) % TableSize; char[][] table = charArray_length[4][(int) hash]; int i = newEntry6; while (++i < InternalTableSize) { char[] charArray = table[i]; - if ((c0 == charArray[0]) - && (c1 == charArray[1]) - && (c2 == charArray[2]) - && (c3 == charArray[3]) - && (c4 == charArray[4]) - && (c5 == charArray[5])) + if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) + && (c3 == charArray[3]) && (c4 == charArray[4]) + && (c5 == charArray[5])) return charArray; } //---------other side--------- @@ -2238,33 +2420,27 @@ public class Scanner implements IScanner, ITerminalSymbols { int max = newEntry6; while (++i <= max) { char[] charArray = table[i]; - if ((c0 == charArray[0]) - && (c1 == charArray[1]) - && (c2 == charArray[2]) - && (c3 == charArray[3]) - && (c4 == charArray[4]) - && (c5 == charArray[5])) + if ((c0 == charArray[0]) && (c1 == charArray[1]) && (c2 == charArray[2]) + && (c3 == charArray[3]) && (c4 == charArray[4]) + && (c5 == charArray[5])) return charArray; } //--------add the entry------- if (++max >= InternalTableSize) max = 0; char[] r; - table[max] = (r = new char[] { c0, c1, c2, c3, c4, c5 }); + table[max] = (r = new char[]{c0, c1, c2, c3, c4, c5}); newEntry6 = max; return r; } - public final void pushLineSeparator() throws InvalidInputException { //see comment on isLineDelimiter(char) for the use of '\n' and '\r' final int INCREMENT = 250; - if (this.checkNonExternalizedStringLiterals) { // reinitialize the current line for non externalize strings purpose currentLine = null; } //currentCharacter is at position currentPosition-1 - // cr 000D if (currentCharacter == '\r') { int separatorPos = currentPosition - 1; @@ -2323,17 +2499,13 @@ public class Scanner implements IScanner, ITerminalSymbols { } public final void pushUnicodeLineSeparator() { // isUnicode means that the \r or \n has been read as a unicode character - //see comment on isLineDelimiter(char) for the use of '\n' and '\r' - final int INCREMENT = 250; //currentCharacter is at position currentPosition-1 - if (this.checkNonExternalizedStringLiterals) { // reinitialize the current line for non externalize strings purpose currentLine = null; } - // cr 000D if (currentCharacter == '\r') { int separatorPos = currentPosition - 6; @@ -2387,10 +2559,11 @@ public class Scanner implements IScanner, ITerminalSymbols { } } public final void recordComment(boolean isJavadoc) { - // a new annotation comment is recorded try { - commentStops[++commentPtr] = isJavadoc ? currentPosition : -currentPosition; + commentStops[++commentPtr] = isJavadoc + ? currentPosition + : -currentPosition; } catch (IndexOutOfBoundsException e) { int oldStackLength = commentStops.length; int[] oldStack = commentStops; @@ -2402,23 +2575,20 @@ public class Scanner implements IScanner, ITerminalSymbols { commentStarts = new int[oldStackLength + 30]; System.arraycopy(old, 0, commentStarts, 0, oldStackLength); } - //the buffer is of a correct size here commentStarts[commentPtr] = startPosition; } public void resetTo(int begin, int end) { //reset the scanner to a given position where it may rescan again - diet = false; initialPosition = startPosition = currentPosition = begin; eofPosition = end < Integer.MAX_VALUE ? end + 1 : end; commentPtr = -1; // reset comment stack } - - public final void scanSingleQuotedEscapeCharacter() throws InvalidInputException { + public final void scanSingleQuotedEscapeCharacter() + throws InvalidInputException { // the string with "\\u" is a legal string of two chars \ and u //thus we use a direct access to the source (for regular cases). - // if (unicodeAsBackSlash) { // // consume next character // unicodeAsBackSlash = false; @@ -2444,11 +2614,10 @@ public class Scanner implements IScanner, ITerminalSymbols { currentPosition--; } } - - public final void scanDoubleQuotedEscapeCharacter() throws InvalidInputException { + public final void scanDoubleQuotedEscapeCharacter() + throws InvalidInputException { // the string with "\\u" is a legal string of two chars \ and u //thus we use a direct access to the source (for regular cases). - // if (unicodeAsBackSlash) { // // consume next character // unicodeAsBackSlash = false; @@ -2472,9 +2641,9 @@ public class Scanner implements IScanner, ITerminalSymbols { case 'n' : currentCharacter = '\n'; break; - // case 'f' : - // currentCharacter = '\f'; - // break; + // case 'f' : + // currentCharacter = '\f'; + // break; case 'r' : currentCharacter = '\r'; break; @@ -2495,7 +2664,6 @@ public class Scanner implements IScanner, ITerminalSymbols { // OctalDigit // OctalDigit OctalDigit // ZeroToThree OctalDigit OctalDigit - int number = Character.getNumericValue(currentCharacter); if (number >= 0 && number <= 7) { boolean zeroToThreeNot = number > 3; @@ -2503,22 +2671,27 @@ public class Scanner implements IScanner, ITerminalSymbols { int digit = Character.getNumericValue(currentCharacter); if (digit >= 0 && digit <= 7) { number = (number * 8) + digit; - if (Character.isDigit(currentCharacter = source[currentPosition++])) { - if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit Digit --> ignore last character + if (Character + .isDigit(currentCharacter = source[currentPosition++])) { + if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit + // Digit --> ignore last character currentPosition--; } else { digit = Character.getNumericValue(currentCharacter); if (digit >= 0 && digit <= 7) { // has read \ZeroToThree OctalDigit OctalDigit number = (number * 8) + digit; - } else { // has read \ZeroToThree OctalDigit NonOctalDigit --> ignore last character + } else { // has read \ZeroToThree OctalDigit NonOctalDigit + // --> ignore last character currentPosition--; } } - } else { // has read \OctalDigit NonDigit--> ignore last character + } else { // has read \OctalDigit NonDigit--> ignore last + // character currentPosition--; } - } else { // has read \OctalDigit NonOctalDigit--> ignore last character + } else { // has read \OctalDigit NonOctalDigit--> ignore last + // character currentPosition--; } } else { // has read \OctalDigit --> ignore last character @@ -2528,43 +2701,36 @@ public class Scanner implements IScanner, ITerminalSymbols { throw new InvalidInputException(INVALID_ESCAPE); currentCharacter = (char) number; } - //else - // throw new InvalidInputException(INVALID_ESCAPE); + //else + // throw new InvalidInputException(INVALID_ESCAPE); } } - // public int scanIdentifierOrKeyword() throws InvalidInputException { // return scanIdentifierOrKeyword( false ); // } - - public int scanIdentifierOrKeyword(boolean isVariable) throws InvalidInputException { + public int scanIdentifierOrKeyword(boolean isVariable) + throws InvalidInputException { //test keywords - //first dispatch on the first char. //then the length. If there are several //keywors with the same length AND the same first char, then do another //disptach on the second char :-)...cool....but fast ! - useAssertAsAnIndentifier = false; - while (getNextCharAsJavaIdentifierPart()) { }; - if (isVariable) { - if (new String(getCurrentTokenSource()).equals("$this")) { - return TokenNamethis; - } + // if (new String(getCurrentTokenSource()).equals("$this")) { + // return TokenNamethis; + // } return TokenNameVariable; } int index, length; char[] data; char firstLetter; // if (withoutUnicodePtr == 0) - //quick test on length == 1 but not on length > 12 while most identifier //have a length which is <= 12...but there are lots of identifier with //only one char.... - // { if ((length = currentPosition - startPosition) == 1) return TokenNameIdentifier; @@ -2585,86 +2751,140 @@ public class Scanner implements IScanner, ITerminalSymbols { // } // index = 1; // } - firstLetter = data[index]; switch (firstLetter) { - - case 'a' : // as and array abstract + case '_' : + switch (length) { + case 8 : + //__FILE__ + if ((data[++index] == '_') && (data[++index] == 'f') + && (data[++index] == 'i') && (data[++index] == 'l') + && (data[++index] == 'e') && (data[++index] == '_') + && (data[++index] == '_')) + return TokenNameFILE; + index = 0; //__LINE__ + if ((data[++index] == '_') && (data[++index] == 'l') + && (data[++index] == 'i') && (data[++index] == 'n') + && (data[++index] == 'e') && (data[++index] == '_') + && (data[++index] == '_')) + return TokenNameLINE; + break; + case 9 : + //__CLASS__ + if ((data[++index] == '_') && (data[++index] == 'c') + && (data[++index] == 'l') && (data[++index] == 'a') + && (data[++index] == 's') && (data[++index] == 's') + && (data[++index] == '_') && (data[++index] == '_')) + return TokenNameCLASS_C; + break; + case 11 : + //__METHOD__ + if ((data[++index] == '_') && (data[++index] == 'm') + && (data[++index] == 'e') && (data[++index] == 't') + && (data[++index] == 'h') && (data[++index] == 'o') + && (data[++index] == 'd') && (data[++index] == '_') + && (data[++index] == '_')) + return TokenNameMETHOD_C; + break; + case 12 : + //__FUNCTION__ + if ((data[++index] == '_') && (data[++index] == 'f') + && (data[++index] == 'u') && (data[++index] == 'n') + && (data[++index] == 'c') && (data[++index] == 't') + && (data[++index] == 'i') && (data[++index] == 'o') + && (data[++index] == 'n') && (data[++index] == '_') + && (data[++index] == '_')) + return TokenNameFUNC_C; + break; + } + return TokenNameIdentifier; + case 'a' : + // as and array abstract switch (length) { - case 2 : //as + case 2 : + //as if ((data[++index] == 's')) { return TokenNameas; } else { return TokenNameIdentifier; } - case 3 : //and + case 3 : + //and if ((data[++index] == 'n') && (data[++index] == 'd')) { - return TokenNameAND; + return TokenNameand; } else { return TokenNameIdentifier; } - case 5 : // array - if ((data[++index] == 'r') && (data[++index] == 'r') && (data[++index] == 'a') && (data[++index] == 'y')) + case 5 : + // array + if ((data[++index] == 'r') && (data[++index] == 'r') + && (data[++index] == 'a') && (data[++index] == 'y')) return TokenNamearray; else return TokenNameIdentifier; case 8 : - if ((data[++index] == 'b') - && (data[++index] == 's') - && (data[++index] == 't') - && (data[++index] == 'r') - && (data[++index] == 'a') - && (data[++index] == 'c') - && (data[++index] == 't')) + if ((data[++index] == 'b') && (data[++index] == 's') + && (data[++index] == 't') && (data[++index] == 'r') + && (data[++index] == 'a') && (data[++index] == 'c') + && (data[++index] == 't')) return TokenNameabstract; else return TokenNameIdentifier; default : return TokenNameIdentifier; } - case 'b' : //break + case 'b' : + //break switch (length) { case 5 : - if ((data[++index] == 'r') && (data[++index] == 'e') && (data[++index] == 'a') && (data[++index] == 'k')) + if ((data[++index] == 'r') && (data[++index] == 'e') + && (data[++index] == 'a') && (data[++index] == 'k')) return TokenNamebreak; else return TokenNameIdentifier; default : return TokenNameIdentifier; } - - case 'c' : //case catch class const continue + case 'c' : + //case catch class clone const continue switch (length) { case 4 : - if ((data[++index] == 'a') && (data[++index] == 's') && (data[++index] == 'e')) + if ((data[++index] == 'a') && (data[++index] == 's') + && (data[++index] == 'e')) return TokenNamecase; else return TokenNameIdentifier; case 5 : - if ((data[++index] == 'a') && (data[++index] == 't') && (data[++index] == 'c') && (data[++index] == 'h')) + if ((data[++index] == 'a') && (data[++index] == 't') + && (data[++index] == 'c') && (data[++index] == 'h')) return TokenNamecatch; - if ((data[index] == 'l') && (data[++index] == 'a') && (data[++index] == 's') && (data[++index] == 's')) + index = 0; + if ((data[++index] == 'l') && (data[++index] == 'a') + && (data[++index] == 's') && (data[++index] == 's')) return TokenNameclass; - if ((data[index] == 'o') && (data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 't')) + index = 0; + if ((data[++index] == 'l') && (data[++index] == 'o') + && (data[++index] == 'n') && (data[++index] == 'e')) + return TokenNameclone; + index = 0; + if ((data[++index] == 'o') && (data[++index] == 'n') + && (data[++index] == 's') && (data[++index] == 't')) return TokenNameconst; else return TokenNameIdentifier; case 8 : - if ((data[++index] == 'o') - && (data[++index] == 'n') - && (data[++index] == 't') - && (data[++index] == 'i') - && (data[++index] == 'n') - && (data[++index] == 'u') - && (data[++index] == 'e')) + if ((data[++index] == 'o') && (data[++index] == 'n') + && (data[++index] == 't') && (data[++index] == 'i') + && (data[++index] == 'n') && (data[++index] == 'u') + && (data[++index] == 'e')) return TokenNamecontinue; else return TokenNameIdentifier; default : return TokenNameIdentifier; } - - case 'd' : // declare default do die + case 'd' : + // declare default do die // TODO delete define ==> no keyword ! switch (length) { case 2 : @@ -2672,135 +2892,117 @@ public class Scanner implements IScanner, ITerminalSymbols { return TokenNamedo; else return TokenNameIdentifier; -// case 6 : -// if ((data[++index] == 'e') -// && (data[++index] == 'f') -// && (data[++index] == 'i') -// && (data[++index] == 'n') -// && (data[++index] == 'e')) -// return TokenNamedefine; -// else -// return TokenNameIdentifier; + // case 6 : + // if ((data[++index] == 'e') + // && (data[++index] == 'f') + // && (data[++index] == 'i') + // && (data[++index] == 'n') + // && (data[++index] == 'e')) + // return TokenNamedefine; + // else + // return TokenNameIdentifier; case 7 : - if ((data[++index] == 'e') - && (data[++index] == 'c') - && (data[++index] == 'l') - && (data[++index] == 'a') - && (data[++index] == 'r') - && (data[++index] == 'e')) + if ((data[++index] == 'e') && (data[++index] == 'c') + && (data[++index] == 'l') && (data[++index] == 'a') + && (data[++index] == 'r') && (data[++index] == 'e')) return TokenNamedeclare; index = 0; - if ((data[++index] == 'e') - && (data[++index] == 'f') - && (data[++index] == 'a') - && (data[++index] == 'u') - && (data[++index] == 'l') - && (data[++index] == 't')) + if ((data[++index] == 'e') && (data[++index] == 'f') + && (data[++index] == 'a') && (data[++index] == 'u') + && (data[++index] == 'l') && (data[++index] == 't')) return TokenNamedefault; else return TokenNameIdentifier; default : return TokenNameIdentifier; } - case 'e' : //echo else exit elseif extends eval + case 'e' : + //echo else exit elseif extends eval switch (length) { case 4 : - if ((data[++index] == 'c') && (data[++index] == 'h') && (data[++index] == 'o')) + if ((data[++index] == 'c') && (data[++index] == 'h') + && (data[++index] == 'o')) return TokenNameecho; - else if ((data[index] == 'l') && (data[++index] == 's') && (data[++index] == 'e')) + else if ((data[index] == 'l') && (data[++index] == 's') + && (data[++index] == 'e')) return TokenNameelse; - else if ((data[index] == 'x') && (data[++index] == 'i') && (data[++index] == 't')) + else if ((data[index] == 'x') && (data[++index] == 'i') + && (data[++index] == 't')) return TokenNameexit; - else if ((data[index] == 'v') && (data[++index] == 'a') && (data[++index] == 'l')) + else if ((data[index] == 'v') && (data[++index] == 'a') + && (data[++index] == 'l')) return TokenNameeval; else return TokenNameIdentifier; - case 5 : // endif empty - if ((data[++index] == 'n') && (data[++index] == 'd') && (data[++index] == 'i') && (data[++index] == 'f')) + case 5 : + // endif empty + if ((data[++index] == 'n') && (data[++index] == 'd') + && (data[++index] == 'i') && (data[++index] == 'f')) return TokenNameendif; - if ((data[index] == 'm') && (data[++index] == 'p') && (data[++index] == 't') && (data[++index] == 'y')) + if ((data[index] == 'm') && (data[++index] == 'p') + && (data[++index] == 't') && (data[++index] == 'y')) return TokenNameempty; else return TokenNameIdentifier; - case 6 : // endfor - if ((data[++index] == 'n') - && (data[++index] == 'd') - && (data[++index] == 'f') - && (data[++index] == 'o') - && (data[++index] == 'r')) + case 6 : + // endfor + if ((data[++index] == 'n') && (data[++index] == 'd') + && (data[++index] == 'f') && (data[++index] == 'o') + && (data[++index] == 'r')) return TokenNameendfor; - else if ( - (data[index] == 'l') - && (data[++index] == 's') - && (data[++index] == 'e') - && (data[++index] == 'i') + else if ((data[index] == 'l') && (data[++index] == 's') + && (data[++index] == 'e') && (data[++index] == 'i') && (data[++index] == 'f')) return TokenNameelseif; else return TokenNameIdentifier; case 7 : - if ((data[++index] == 'x') - && (data[++index] == 't') - && (data[++index] == 'e') - && (data[++index] == 'n') - && (data[++index] == 'd') - && (data[++index] == 's')) + if ((data[++index] == 'x') && (data[++index] == 't') + && (data[++index] == 'e') && (data[++index] == 'n') + && (data[++index] == 'd') && (data[++index] == 's')) return TokenNameextends; else return TokenNameIdentifier; - case 8 : // endwhile - if ((data[++index] == 'n') - && (data[++index] == 'd') - && (data[++index] == 'w') - && (data[++index] == 'h') - && (data[++index] == 'i') - && (data[++index] == 'l') - && (data[++index] == 'e')) + case 8 : + // endwhile + if ((data[++index] == 'n') && (data[++index] == 'd') + && (data[++index] == 'w') && (data[++index] == 'h') + && (data[++index] == 'i') && (data[++index] == 'l') + && (data[++index] == 'e')) return TokenNameendwhile; else return TokenNameIdentifier; - case 9 : // endswitch - if ((data[++index] == 'n') - && (data[++index] == 'd') - && (data[++index] == 's') - && (data[++index] == 'w') - && (data[++index] == 'i') - && (data[++index] == 't') - && (data[++index] == 'c') - && (data[++index] == 'h')) + case 9 : + // endswitch + if ((data[++index] == 'n') && (data[++index] == 'd') + && (data[++index] == 's') && (data[++index] == 'w') + && (data[++index] == 'i') && (data[++index] == 't') + && (data[++index] == 'c') && (data[++index] == 'h')) return TokenNameendswitch; else return TokenNameIdentifier; - case 10 : // enddeclare - if ((data[++index] == 'n') - && (data[++index] == 'd') - && (data[++index] == 'd') - && (data[++index] == 'e') - && (data[++index] == 'c') - && (data[++index] == 'l') - && (data[++index] == 'a') - && (data[++index] == 'r') - && (data[++index] == 'e')) + case 10 : + // enddeclare + if ((data[++index] == 'n') && (data[++index] == 'd') + && (data[++index] == 'd') && (data[++index] == 'e') + && (data[++index] == 'c') && (data[++index] == 'l') + && (data[++index] == 'a') && (data[++index] == 'r') + && (data[++index] == 'e')) return TokenNameendforeach; index = 0; if ((data[++index] == 'n') // endforeach - && (data[++index] == 'd') - && (data[++index] == 'f') - && (data[++index] == 'o') - && (data[++index] == 'r') - && (data[++index] == 'e') - && (data[++index] == 'a') - && (data[++index] == 'c') - && (data[++index] == 'h')) + && (data[++index] == 'd') && (data[++index] == 'f') + && (data[++index] == 'o') && (data[++index] == 'r') + && (data[++index] == 'e') && (data[++index] == 'a') + && (data[++index] == 'c') && (data[++index] == 'h')) return TokenNameendforeach; else return TokenNameIdentifier; - default : return TokenNameIdentifier; } - - case 'f' : //for false final function + case 'f' : + //for false final function switch (length) { case 3 : if ((data[++index] == 'o') && (data[++index] == 'r')) @@ -2808,157 +3010,139 @@ public class Scanner implements IScanner, ITerminalSymbols { else return TokenNameIdentifier; case 5 : - if ((data[++index] == 'a') && (data[++index] == 'l') && (data[++index] == 's') && (data[++index] == 'e')) - return TokenNamefalse; - if ((data[index] == 'i') && (data[++index] == 'n') && (data[++index] == 'a') && (data[++index] == 'l')) + // if ((data[++index] == 'a') && (data[++index] == 'l') + // && (data[++index] == 's') && (data[++index] == 'e')) + // return TokenNamefalse; + if ((data[++index] == 'i') && (data[++index] == 'n') + && (data[++index] == 'a') && (data[++index] == 'l')) return TokenNamefinal; else return TokenNameIdentifier; - case 7 : // foreach - if ((data[++index] == 'o') - && (data[++index] == 'r') - && (data[++index] == 'e') - && (data[++index] == 'a') - && (data[++index] == 'c') - && (data[++index] == 'h')) + case 7 : + // foreach + if ((data[++index] == 'o') && (data[++index] == 'r') + && (data[++index] == 'e') && (data[++index] == 'a') + && (data[++index] == 'c') && (data[++index] == 'h')) return TokenNameforeach; else return TokenNameIdentifier; - case 8 : // function - if ((data[++index] == 'u') - && (data[++index] == 'n') - && (data[++index] == 'c') - && (data[++index] == 't') - && (data[++index] == 'i') - && (data[++index] == 'o') - && (data[++index] == 'n')) + case 8 : + // function + if ((data[++index] == 'u') && (data[++index] == 'n') + && (data[++index] == 'c') && (data[++index] == 't') + && (data[++index] == 'i') && (data[++index] == 'o') + && (data[++index] == 'n')) return TokenNamefunction; else return TokenNameIdentifier; default : return TokenNameIdentifier; } - case 'g' : //global + case 'g' : + //global if (length == 6) { - if ((data[++index] == 'l') - && (data[++index] == 'o') - && (data[++index] == 'b') - && (data[++index] == 'a') - && (data[++index] == 'l')) { + if ((data[++index] == 'l') && (data[++index] == 'o') + && (data[++index] == 'b') && (data[++index] == 'a') + && (data[++index] == 'l')) { return TokenNameglobal; } } return TokenNameIdentifier; - - case 'i' : //if int isset include include_once instanceof interface implements + case 'i' : + //if int isset include include_once instanceof interface implements switch (length) { case 2 : if (data[++index] == 'f') return TokenNameif; else return TokenNameIdentifier; - // case 3 : - // if ((data[++index] == 'n') && (data[++index] == 't')) - // return TokenNameint; - // else - // return TokenNameIdentifier; + // case 3 : + // if ((data[++index] == 'n') && (data[++index] == 't')) + // return TokenNameint; + // else + // return TokenNameIdentifier; case 5 : - if ((data[++index] == 's') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 't')) + if ((data[++index] == 's') && (data[++index] == 's') + && (data[++index] == 'e') && (data[++index] == 't')) return TokenNameisset; else return TokenNameIdentifier; case 7 : - if ((data[++index] == 'n') - && (data[++index] == 'c') - && (data[++index] == 'l') - && (data[++index] == 'u') - && (data[++index] == 'd') - && (data[++index] == 'e')) + if ((data[++index] == 'n') && (data[++index] == 'c') + && (data[++index] == 'l') && (data[++index] == 'u') + && (data[++index] == 'd') && (data[++index] == 'e')) return TokenNameinclude; else return TokenNameIdentifier; - case 9 : // interface - if ((data[++index] == 'n') - && (data[++index] == 't') - && (data[++index] == 'e') - && (data[++index] == 'r') - && (data[++index] == 'f') - && (data[++index] == 'a') - && (data[++index] == 'c') - && (data[++index] == 'e')) + case 9 : + // interface + if ((data[++index] == 'n') && (data[++index] == 't') + && (data[++index] == 'e') && (data[++index] == 'r') + && (data[++index] == 'f') && (data[++index] == 'a') + && (data[++index] == 'c') && (data[++index] == 'e')) return TokenNameinterface; else return TokenNameIdentifier; - case 10 : // instanceof - if ((data[++index] == 'n') - && (data[++index] == 's') - && (data[++index] == 't') - && (data[++index] == 'a') - && (data[++index] == 'n') - && (data[++index] == 'c') - && (data[++index] == 'e') - && (data[++index] == 'o') - && (data[++index] == 'f')) + case 10 : + // instanceof + if ((data[++index] == 'n') && (data[++index] == 's') + && (data[++index] == 't') && (data[++index] == 'a') + && (data[++index] == 'n') && (data[++index] == 'c') + && (data[++index] == 'e') && (data[++index] == 'o') + && (data[++index] == 'f')) return TokenNameinstanceof; - if ((data[index] == 'm') - && (data[++index] == 'p') - && (data[++index] == 'l') - && (data[++index] == 'e') - && (data[++index] == 'm') - && (data[++index] == 'e') - && (data[++index] == 'n') - && (data[++index] == 't') - && (data[++index] == 's')) + if ((data[index] == 'm') && (data[++index] == 'p') + && (data[++index] == 'l') && (data[++index] == 'e') + && (data[++index] == 'm') && (data[++index] == 'e') + && (data[++index] == 'n') && (data[++index] == 't') + && (data[++index] == 's')) return TokenNameimplements; else return TokenNameIdentifier; case 12 : - if ((data[++index] == 'n') - && (data[++index] == 'c') - && (data[++index] == 'l') - && (data[++index] == 'u') - && (data[++index] == 'd') - && (data[++index] == 'e') - && (data[++index] == '_') - && (data[++index] == 'o') - && (data[++index] == 'n') - && (data[++index] == 'c') - && (data[++index] == 'e')) + if ((data[++index] == 'n') && (data[++index] == 'c') + && (data[++index] == 'l') && (data[++index] == 'u') + && (data[++index] == 'd') && (data[++index] == 'e') + && (data[++index] == '_') && (data[++index] == 'o') + && (data[++index] == 'n') && (data[++index] == 'c') + && (data[++index] == 'e')) return TokenNameinclude_once; else return TokenNameIdentifier; default : return TokenNameIdentifier; } - - case 'l' : //list + case 'l' : + //list if (length == 4) { - if ((data[++index] == 'i') && (data[++index] == 's') && (data[++index] == 't')) { + if ((data[++index] == 'i') && (data[++index] == 's') + && (data[++index] == 't')) { return TokenNamelist; } } return TokenNameIdentifier; - - case 'n' : // new null + case 'n' : + // new null switch (length) { case 3 : if ((data[++index] == 'e') && (data[++index] == 'w')) return TokenNamenew; else return TokenNameIdentifier; - case 4 : - if ((data[++index] == 'u') && (data[++index] == 'l') && (data[++index] == 'l')) - return TokenNamenull; - else - return TokenNameIdentifier; - + // case 4 : + // if ((data[++index] == 'u') && (data[++index] == 'l') + // && (data[++index] == 'l')) + // return TokenNamenull; + // else + // return TokenNameIdentifier; default : return TokenNameIdentifier; } - case 'o' : // or old_function + case 'o' : + // or old_function if (length == 2) { if (data[++index] == 'r') { - return TokenNameOR; + return TokenNameor; } } // if (length == 12) { @@ -2977,95 +3161,76 @@ public class Scanner implements IScanner, ITerminalSymbols { // } // } return TokenNameIdentifier; - - case 'p' : // print public private protected + case 'p' : + // print public private protected switch (length) { case 5 : - if ((data[++index] == 'r') && (data[++index] == 'i') && (data[++index] == 'n') && (data[++index] == 't')) { + if ((data[++index] == 'r') && (data[++index] == 'i') + && (data[++index] == 'n') && (data[++index] == 't')) { return TokenNameprint; } else return TokenNameIdentifier; case 6 : - if ((data[++index] == 'u') - && (data[++index] == 'b') - && (data[++index] == 'l') - && (data[++index] == 'i') - && (data[++index] == 'c')) { + if ((data[++index] == 'u') && (data[++index] == 'b') + && (data[++index] == 'l') && (data[++index] == 'i') + && (data[++index] == 'c')) { return TokenNamepublic; } else return TokenNameIdentifier; case 7 : - if ((data[++index] == 'r') - && (data[++index] == 'i') - && (data[++index] == 'v') - && (data[++index] == 'a') - && (data[++index] == 't') - && (data[++index] == 'e')) { + if ((data[++index] == 'r') && (data[++index] == 'i') + && (data[++index] == 'v') && (data[++index] == 'a') + && (data[++index] == 't') && (data[++index] == 'e')) { return TokenNameprivate; } else return TokenNameIdentifier; case 9 : - if ((data[++index] == 'r') - && (data[++index] == 'o') - && (data[++index] == 't') - && (data[++index] == 'e') - && (data[++index] == 'c') - && (data[++index] == 't') - && (data[++index] == 'e') - && (data[++index] == 'd')) { + if ((data[++index] == 'r') && (data[++index] == 'o') + && (data[++index] == 't') && (data[++index] == 'e') + && (data[++index] == 'c') && (data[++index] == 't') + && (data[++index] == 'e') && (data[++index] == 'd')) { return TokenNameprotected; } else return TokenNameIdentifier; } return TokenNameIdentifier; - case 'r' : //return require require_once + case 'r' : + //return require require_once if (length == 6) { - if ((data[++index] == 'e') - && (data[++index] == 't') - && (data[++index] == 'u') - && (data[++index] == 'r') - && (data[++index] == 'n')) { + if ((data[++index] == 'e') && (data[++index] == 't') + && (data[++index] == 'u') && (data[++index] == 'r') + && (data[++index] == 'n')) { return TokenNamereturn; } } else if (length == 7) { - if ((data[++index] == 'e') - && (data[++index] == 'q') - && (data[++index] == 'u') - && (data[++index] == 'i') - && (data[++index] == 'r') - && (data[++index] == 'e')) { + if ((data[++index] == 'e') && (data[++index] == 'q') + && (data[++index] == 'u') && (data[++index] == 'i') + && (data[++index] == 'r') && (data[++index] == 'e')) { return TokenNamerequire; } } else if (length == 12) { - if ((data[++index] == 'e') - && (data[++index] == 'q') - && (data[++index] == 'u') - && (data[++index] == 'i') - && (data[++index] == 'r') - && (data[++index] == 'e') - && (data[++index] == '_') - && (data[++index] == 'o') - && (data[++index] == 'n') - && (data[++index] == 'c') - && (data[++index] == 'e')) { + if ((data[++index] == 'e') && (data[++index] == 'q') + && (data[++index] == 'u') && (data[++index] == 'i') + && (data[++index] == 'r') && (data[++index] == 'e') + && (data[++index] == '_') && (data[++index] == 'o') + && (data[++index] == 'n') && (data[++index] == 'c') + && (data[++index] == 'e')) { return TokenNamerequire_once; } } else return TokenNameIdentifier; - - case 's' : //static switch + case 's' : + //static switch switch (length) { case 6 : if (data[++index] == 't') - if ((data[++index] == 'a') && (data[++index] == 't') && (data[++index] == 'i') && (data[++index] == 'c')) { + if ((data[++index] == 'a') && (data[++index] == 't') + && (data[++index] == 'i') && (data[++index] == 'c')) { return TokenNamestatic; } else return TokenNameIdentifier; - else if ( - (data[index] == 'w') - && (data[++index] == 'i') - && (data[++index] == 't') - && (data[++index] == 'c') + else if ((data[index] == 'w') && (data[++index] == 'i') + && (data[++index] == 't') && (data[++index] == 'c') && (data[++index] == 'h')) return TokenNameswitch; else @@ -3073,29 +3238,31 @@ public class Scanner implements IScanner, ITerminalSymbols { default : return TokenNameIdentifier; } - - case 't' : // try true throw + case 't' : + // try true throw switch (length) { case 3 : if ((data[++index] == 'r') && (data[++index] == 'y')) return TokenNametry; else return TokenNameIdentifier; - case 4 : - if ((data[++index] == 'r') && (data[++index] == 'u') && (data[++index] == 'e')) - return TokenNametrue; - else - return TokenNameIdentifier; + // case 4 : + // if ((data[++index] == 'r') && (data[++index] == 'u') + // && (data[++index] == 'e')) + // return TokenNametrue; + // else + // return TokenNameIdentifier; case 5 : - if ((data[++index] == 'h') && (data[++index] == 'r') && (data[++index] == 'o') && (data[++index] == 'w')) + if ((data[++index] == 'h') && (data[++index] == 'r') + && (data[++index] == 'o') && (data[++index] == 'w')) return TokenNamethrow; else return TokenNameIdentifier; - default : return TokenNameIdentifier; } - case 'u' : //use unset + case 'u' : + //use unset switch (length) { case 3 : if ((data[++index] == 's') && (data[++index] == 'e')) @@ -3103,49 +3270,51 @@ public class Scanner implements IScanner, ITerminalSymbols { else return TokenNameIdentifier; case 5 : - if ((data[++index] == 'n') && (data[++index] == 's') && (data[++index] == 'e') && (data[++index] == 't')) + if ((data[++index] == 'n') && (data[++index] == 's') + && (data[++index] == 'e') && (data[++index] == 't')) return TokenNameunset; else return TokenNameIdentifier; default : return TokenNameIdentifier; } - case 'v' : //var + case 'v' : + //var switch (length) { case 3 : if ((data[++index] == 'a') && (data[++index] == 'r')) return TokenNamevar; else return TokenNameIdentifier; - default : return TokenNameIdentifier; } - - case 'w' : //while + case 'w' : + //while switch (length) { case 5 : - if ((data[++index] == 'h') && (data[++index] == 'i') && (data[++index] == 'l') && (data[++index] == 'e')) + if ((data[++index] == 'h') && (data[++index] == 'i') + && (data[++index] == 'l') && (data[++index] == 'e')) return TokenNamewhile; else return TokenNameIdentifier; - //case 6:if ( (data[++index] =='i') && (data[++index]=='d') && (data[++index]=='e') && (data[++index]=='f')&& - // (data[++index]=='p')) - //return TokenNamewidefp ; - //else - //return TokenNameIdentifier; + //case 6:if ( (data[++index] =='i') && (data[++index]=='d') && + // (data[++index]=='e') && (data[++index]=='f')&& + // (data[++index]=='p')) + //return TokenNamewidefp ; + //else + //return TokenNameIdentifier; default : return TokenNameIdentifier; } - - case 'x' : //xor + case 'x' : + //xor switch (length) { case 3 : if ((data[++index] == 'o') && (data[++index] == 'r')) - return TokenNameXOR; + return TokenNamexor; else return TokenNameIdentifier; - default : return TokenNameIdentifier; } @@ -3154,11 +3323,9 @@ public class Scanner implements IScanner, ITerminalSymbols { } } public int scanNumber(boolean dotPrefix) throws InvalidInputException { - //when entering this method the currentCharacter is the firt //digit of the number , i.e. it may be preceeded by a . when //dotPrefix is true - boolean floating = dotPrefix; if ((!dotPrefix) && (currentCharacter == '0')) { if (getNextChar('x', 'X') >= 0) { //----------hexa----------------- @@ -3184,14 +3351,13 @@ public class Scanner implements IScanner, ITerminalSymbols { // else return TokenNameIntegerLiteral; } - //there is x or X in the number - //potential octal ! ... some one may write 000099.0 ! thus 00100 < 00078.0 is true !!!!! crazy language + //potential octal ! ... some one may write 000099.0 ! thus 00100 < + // 00078.0 is true !!!!! crazy language if (getNextCharAsDigit()) { //-------------potential octal----------------- while (getNextCharAsDigit()) { }; - // if (getNextChar('l', 'L') >= 0) { // return TokenNameLongLiteral; // } @@ -3199,7 +3365,6 @@ public class Scanner implements IScanner, ITerminalSymbols { // if (getNextChar('f', 'F') >= 0) { // return TokenNameFloatingPointLiteral; // } - if (getNextChar('d', 'D') >= 0) { return TokenNameDoubleLiteral; } else { //make the distinction between octal and float .... @@ -3218,7 +3383,6 @@ public class Scanner implements IScanner, ITerminalSymbols { // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; // } // } - if ((currentCharacter == '-') || (currentCharacter == '+')) { // consume next character unicodeAsBackSlash = false; @@ -3250,21 +3414,16 @@ public class Scanner implements IScanner, ITerminalSymbols { /* carry on */ } } - while (getNextCharAsDigit()) { }; - // if ((!dotPrefix) && (getNextChar('l', 'L') >= 0)) // return TokenNameLongLiteral; - if ((!dotPrefix) && (getNextChar('.'))) { //decimal part that can be empty while (getNextCharAsDigit()) { }; floating = true; } - //if floating is true both exponant and suffix may be optional - if (getNextChar('e', 'E') >= 0) { floating = true; // consume next character @@ -3278,8 +3437,9 @@ public class Scanner implements IScanner, ITerminalSymbols { // withoutUnicodeBuffer[++withoutUnicodePtr] = currentCharacter; // } // } - - if ((currentCharacter == '-') || (currentCharacter == '+')) { // consume next character + if ((currentCharacter == '-') || (currentCharacter == '+')) { // consume + // next + // character unicodeAsBackSlash = false; currentCharacter = source[currentPosition++]; // if (((currentCharacter = source[currentPosition++]) == '\\') @@ -3296,14 +3456,11 @@ public class Scanner implements IScanner, ITerminalSymbols { while (getNextCharAsDigit()) { }; } - if (getNextChar('d', 'D') >= 0) return TokenNameDoubleLiteral; // if (getNextChar('f', 'F') >= 0) // return TokenNameFloatingPointLiteral; - //the long flag has been tested before - return floating ? TokenNameDoubleLiteral : TokenNameIntegerLiteral; } /** @@ -3311,7 +3468,6 @@ public class Scanner implements IScanner, ITerminalSymbols { * */ public final int getLineNumber(int position) { - if (lineEnds == null) return 1; int length = linePtr + 1; @@ -3334,14 +3490,11 @@ public class Scanner implements IScanner, ITerminalSymbols { } return m + 2; } - public void setPHPMode(boolean mode) { phpMode = mode; } - public final void setSource(char[] source) { //the source-buffer is set to sourceString - if (source == null) { this.source = new char[0]; } else { @@ -3351,18 +3504,15 @@ public class Scanner implements IScanner, ITerminalSymbols { initialPosition = currentPosition = 0; containsAssertKeyword = false; withoutUnicodeBuffer = new char[this.source.length]; - + encapsedStringStack = new Stack(); } - public String toString() { if (startPosition == source.length) return "EOF\n\n" + new String(source); //$NON-NLS-1$ if (currentPosition > source.length) return "behind the EOF :-( ....\n\n" + new String(source); //$NON-NLS-1$ - char front[] = new char[startPosition]; System.arraycopy(source, 0, front, 0, startPosition); - int middleLength = (currentPosition - 1) - startPosition + 1; char middle[]; if (middleLength > -1) { @@ -3371,27 +3521,30 @@ public class Scanner implements IScanner, ITerminalSymbols { } else { middle = new char[0]; } - char end[] = new char[source.length - (currentPosition - 1)]; - System.arraycopy(source, (currentPosition - 1) + 1, end, 0, source.length - (currentPosition - 1) - 1); - - return new String(front) + "\n===============================\nStarts here -->" //$NON-NLS-1$ - + new String(middle) + "<-- Ends here\n===============================\n" //$NON-NLS-1$ - + new String(end); + System.arraycopy(source, (currentPosition - 1) + 1, end, 0, source.length + - (currentPosition - 1) - 1); + return new String(front) + + "\n===============================\nStarts here -->" //$NON-NLS-1$ + + new String(middle) + + "<-- Ends here\n===============================\n" //$NON-NLS-1$ + + new String(end); } public final String toStringAction(int act) { - switch (act) { case TokenNameERROR : - return "ScannerError"; // + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ - case TokenNameStopPHP : - return "StopPHP(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ + return "ScannerError"; // + new String(getCurrentTokenSource()) + ")"; + // //$NON-NLS-1$ + case TokenNameINLINE_HTML : + return "Inline-HTML(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameIdentifier : return "Identifier(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameVariable : return "Variable(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameabstract : return "abstract"; //$NON-NLS-1$ + case TokenNameand : + return "AND"; //$NON-NLS-1$ case TokenNamearray : return "array"; //$NON-NLS-1$ case TokenNameas : @@ -3402,12 +3555,20 @@ public class Scanner implements IScanner, ITerminalSymbols { return "case"; //$NON-NLS-1$ case TokenNameclass : return "class"; //$NON-NLS-1$ + case TokenNamecatch : + return "catch"; //$NON-NLS-1$ + case TokenNameclone : + //$NON-NLS-1$ + return "clone"; + case TokenNameconst : + //$NON-NLS-1$ + return "const"; case TokenNamecontinue : return "continue"; //$NON-NLS-1$ case TokenNamedefault : return "default"; //$NON-NLS-1$ -// case TokenNamedefine : -// return "define"; //$NON-NLS-1$ + // case TokenNamedefine : + // return "define"; //$NON-NLS-1$ case TokenNamedo : return "do"; //$NON-NLS-1$ case TokenNameecho : @@ -3426,10 +3587,12 @@ public class Scanner implements IScanner, ITerminalSymbols { return "endswitch"; //$NON-NLS-1$ case TokenNameendwhile : return "endwhile"; //$NON-NLS-1$ + case TokenNameexit: + return "exit"; case TokenNameextends : return "extends"; //$NON-NLS-1$ - case TokenNamefalse : - return "false"; //$NON-NLS-1$ + // case TokenNamefalse : + // return "false"; //$NON-NLS-1$ case TokenNamefinal : return "final"; //$NON-NLS-1$ case TokenNamefor : @@ -3448,14 +3611,20 @@ public class Scanner implements IScanner, ITerminalSymbols { return "include"; //$NON-NLS-1$ case TokenNameinclude_once : return "include_once"; //$NON-NLS-1$ + case TokenNameinstanceof : + return "instanceof"; //$NON-NLS-1$ case TokenNameinterface : return "interface"; //$NON-NLS-1$ + case TokenNameisset : + return "isset"; //$NON-NLS-1$ case TokenNamelist : return "list"; //$NON-NLS-1$ case TokenNamenew : return "new"; //$NON-NLS-1$ - case TokenNamenull : - return "null"; //$NON-NLS-1$ + // case TokenNamenull : + // return "null"; //$NON-NLS-1$ + case TokenNameor : + return "OR"; //$NON-NLS-1$ case TokenNameprint : return "print"; //$NON-NLS-1$ case TokenNameprivate : @@ -3474,29 +3643,39 @@ public class Scanner implements IScanner, ITerminalSymbols { return "static"; //$NON-NLS-1$ case TokenNameswitch : return "switch"; //$NON-NLS-1$ - case TokenNametrue : - return "true"; //$NON-NLS-1$ + // case TokenNametrue : + // return "true"; //$NON-NLS-1$ case TokenNameunset : return "unset"; //$NON-NLS-1$ case TokenNamevar : return "var"; //$NON-NLS-1$ case TokenNamewhile : return "while"; //$NON-NLS-1$ - case TokenNamethis : - return "$this"; //$NON-NLS-1$ + case TokenNamexor : + return "XOR"; //$NON-NLS-1$ + // case TokenNamethis : + // return "$this"; //$NON-NLS-1$ case TokenNameIntegerLiteral : return "Integer(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameDoubleLiteral : return "Double(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameStringLiteral : - return "String(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ + return "StringLiteral(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameStringConstant : return "StringConstant(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameStringInterpolated : - return "StringInterpolated(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ + return "StringInterpolated(" + new String(getCurrentTokenSource()) + + ")"; //$NON-NLS-1$ //$NON-NLS-2$ + case TokenNameEncapsedString0 : + return "`"; //$NON-NLS-1$ + case TokenNameEncapsedString1 : + return "\'"; //$NON-NLS-1$ + case TokenNameEncapsedString2 : + return "\""; //$NON-NLS-1$ + case TokenNameSTRING : + return "STRING(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ //$NON-NLS-2$ case TokenNameHEREDOC : return "HEREDOC(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ - case TokenNamePLUS_PLUS : return "++"; //$NON-NLS-1$ case TokenNameMINUS_MINUS : @@ -3535,6 +3714,8 @@ public class Scanner implements IScanner, ITerminalSymbols { return "^="; //$NON-NLS-1$ case TokenNameREMAINDER_EQUAL : return "%="; //$NON-NLS-1$ + case TokenNameDOT_EQUAL : + return ".="; //$NON-NLS-1$ case TokenNameLEFT_SHIFT_EQUAL : return "<<="; //$NON-NLS-1$ case TokenNameRIGHT_SHIFT_EQUAL : @@ -3597,6 +3778,8 @@ public class Scanner implements IScanner, ITerminalSymbols { return "="; //$NON-NLS-1$ case TokenNameAT : return "@"; + case TokenNameDOLLAR : + return "$"; case TokenNameDOLLAR_LBRACE : return "${"; case TokenNameEOF : @@ -3609,35 +3792,71 @@ public class Scanner implements IScanner, ITerminalSymbols { return "COMMENT_BLOCK(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ case TokenNameCOMMENT_PHPDOC : return "COMMENT_PHPDOC(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ - case TokenNameHTML : - return "HTML(" + new String(getCurrentTokenSource()) + ")"; //$NON-NLS-1$ + // case TokenNameHTML : + // return "HTML(" + new String(getCurrentTokenSource()) + ")"; + // //$NON-NLS-1$ + case TokenNameFILE : + return "__FILE__"; //$NON-NLS-1$ + case TokenNameLINE : + return "__LINE__"; //$NON-NLS-1$ + case TokenNameCLASS_C : + return "__CLASS__"; //$NON-NLS-1$ + case TokenNameMETHOD_C : + return "__METHOD__"; //$NON-NLS-1$ + case TokenNameFUNC_C : + return "__FUNCTION__"; //$NON-NLS-1 + case TokenNameboolCAST : + return "( bool )"; //$NON-NLS-1$ + case TokenNameintCAST : + return "( int )"; //$NON-NLS-1$ + case TokenNamedoubleCAST : + return "( double )"; //$NON-NLS-1$ + case TokenNameobjectCAST : + return "( object )"; //$NON-NLS-1$ + case TokenNamestringCAST : + return "( string )"; //$NON-NLS-1$ default : - return "not-a-token(" + (new Integer(act)) + ") " + new String(getCurrentTokenSource()); //$NON-NLS-1$ + return "not-a-token(" + (new Integer(act)) + ") " + + new String(getCurrentTokenSource()); //$NON-NLS-1$ } } - - public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, boolean checkNonExternalizedStringLiterals) { - this(tokenizeComments, tokenizeWhiteSpace, checkNonExternalizedStringLiterals, false); + + public Scanner() { + this(false, false); } - - public Scanner( - boolean tokenizeComments, - boolean tokenizeWhiteSpace, - boolean checkNonExternalizedStringLiterals, - boolean assertMode) { + public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace) { + this(tokenizeComments, tokenizeWhiteSpace, false); + } + public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, + boolean checkNonExternalizedStringLiterals) { + this(tokenizeComments, tokenizeWhiteSpace, + checkNonExternalizedStringLiterals, false); + } + public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, + boolean checkNonExternalizedStringLiterals, boolean assertMode) { + this(tokenizeComments, tokenizeWhiteSpace, + checkNonExternalizedStringLiterals, assertMode, false, null, null); + } + public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace, + boolean checkNonExternalizedStringLiterals, boolean assertMode, + boolean tokenizeStrings, + char[][] taskTags, + char[][] taskPriorities) { this.eofPosition = Integer.MAX_VALUE; this.tokenizeComments = tokenizeComments; this.tokenizeWhiteSpace = tokenizeWhiteSpace; + this.tokenizeStrings = tokenizeStrings; this.checkNonExternalizedStringLiterals = checkNonExternalizedStringLiterals; this.assertMode = assertMode; + this.encapsedStringStack = null; + this.taskTags = taskTags; + this.taskPriorities = taskPriorities; } - private void checkNonExternalizeString() throws InvalidInputException { if (currentLine == null) return; parseTags(currentLine); } - private void parseTags(NLSLine line) throws InvalidInputException { String s = new String(getCurrentTokenSource()); int pos = s.indexOf(TAG_PREFIX); @@ -3658,7 +3877,6 @@ public class Scanner implements IScanner, ITerminalSymbols { } pos = s.indexOf(TAG_PREFIX, start); } - this.nonNLSStrings = new StringLiteral[lineLength]; int nonNLSCounter = 0; for (Iterator iterator = line.iterator(); iterator.hasNext();) { @@ -3674,19 +3892,20 @@ public class Scanner implements IScanner, ITerminalSymbols { } this.wasNonExternalizedStringLiteral = true; if (nonNLSCounter != lineLength) { - System.arraycopy(this.nonNLSStrings, 0, (this.nonNLSStrings = new StringLiteral[nonNLSCounter]), 0, nonNLSCounter); + System.arraycopy(this.nonNLSStrings, 0, + (this.nonNLSStrings = new StringLiteral[nonNLSCounter]), 0, + nonNLSCounter); } currentLine = null; } - public final void scanEscapeCharacter() throws InvalidInputException { // the string with "\\u" is a legal string of two chars \ and u //thus we use a direct access to the source (for regular cases). - if (unicodeAsBackSlash) { // consume next character unicodeAsBackSlash = false; - // if (((currentCharacter = source[currentPosition++]) == '\\') && (source[currentPosition] == 'u')) { + // if (((currentCharacter = source[currentPosition++]) == '\\') && + // (source[currentPosition] == 'u')) { // getNextUnicodeChar(); // } else { if (withoutUnicodePtr != 0) { @@ -3725,7 +3944,6 @@ public class Scanner implements IScanner, ITerminalSymbols { // OctalDigit // OctalDigit OctalDigit // ZeroToThree OctalDigit OctalDigit - int number = Character.getNumericValue(currentCharacter); if (number >= 0 && number <= 7) { boolean zeroToThreeNot = number > 3; @@ -3733,21 +3951,27 @@ public class Scanner implements IScanner, ITerminalSymbols { int digit = Character.getNumericValue(currentCharacter); if (digit >= 0 && digit <= 7) { number = (number * 8) + digit; - if (Character.isDigit(currentCharacter = source[currentPosition++])) { - if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit Digit --> ignore last character + if (Character + .isDigit(currentCharacter = source[currentPosition++])) { + if (zeroToThreeNot) { // has read \NotZeroToThree OctalDigit + // Digit --> ignore last character currentPosition--; } else { digit = Character.getNumericValue(currentCharacter); - if (digit >= 0 && digit <= 7) { // has read \ZeroToThree OctalDigit OctalDigit + if (digit >= 0 && digit <= 7) { // has read \ZeroToThree + // OctalDigit OctalDigit number = (number * 8) + digit; - } else { // has read \ZeroToThree OctalDigit NonOctalDigit --> ignore last character + } else { // has read \ZeroToThree OctalDigit NonOctalDigit + // --> ignore last character currentPosition--; } } - } else { // has read \OctalDigit NonDigit--> ignore last character + } else { // has read \OctalDigit NonDigit--> ignore last + // character currentPosition--; } - } else { // has read \OctalDigit NonOctalDigit--> ignore last character + } else { // has read \OctalDigit NonOctalDigit--> ignore last + // character currentPosition--; } } else { // has read \OctalDigit --> ignore last character @@ -3760,83 +3984,76 @@ public class Scanner implements IScanner, ITerminalSymbols { throw new InvalidInputException(INVALID_ESCAPE); } } - // chech presence of task: tags public void checkTaskTag(int commentStart, int commentEnd) { - // only look for newer task: tags - if (this.foundTaskCount > 0 && this.foundTaskPositions[this.foundTaskCount - 1][0] >= commentStart) { + if (this.foundTaskCount > 0 + && this.foundTaskPositions[this.foundTaskCount - 1][0] >= commentStart) { return; } int foundTaskIndex = this.foundTaskCount; - nextChar : for (int i = commentStart; i < commentEnd && i < this.eofPosition; i++) { - + nextChar : for (int i = commentStart; i < commentEnd + && i < this.eofPosition; i++) { char[] tag = null; char[] priority = null; - // check for tag occurrence nextTag : for (int itag = 0; itag < this.taskTags.length; itag++) { tag = this.taskTags[itag]; - priority = this.taskPriorities != null && itag < this.taskPriorities.length ? this.taskPriorities[itag] : null; + priority = this.taskPriorities != null + && itag < this.taskPriorities.length + ? this.taskPriorities[itag] + : null; int tagLength = tag.length; for (int t = 0; t < tagLength; t++) { if (this.source[i + t] != tag[t]) continue nextTag; } - if (this.foundTaskTags == null) { this.foundTaskTags = new char[5][]; this.foundTaskMessages = new char[5][]; this.foundTaskPriorities = new char[5][]; this.foundTaskPositions = new int[5][]; } else if (this.foundTaskCount == this.foundTaskTags.length) { - System.arraycopy(this.foundTaskTags, 0, this.foundTaskTags = new char[this.foundTaskCount * 2][], 0, this.foundTaskCount); - System.arraycopy( - this.foundTaskMessages, - 0, - this.foundTaskMessages = new char[this.foundTaskCount * 2][], - 0, - this.foundTaskCount); - System.arraycopy( - this.foundTaskPriorities, - 0, - this.foundTaskPriorities = new char[this.foundTaskCount * 2][], - 0, - this.foundTaskCount); - System.arraycopy( - this.foundTaskPositions, - 0, - this.foundTaskPositions = new int[this.foundTaskCount * 2][], - 0, - this.foundTaskCount); + System.arraycopy(this.foundTaskTags, 0, + this.foundTaskTags = new char[this.foundTaskCount * 2][], 0, + this.foundTaskCount); + System.arraycopy(this.foundTaskMessages, 0, + this.foundTaskMessages = new char[this.foundTaskCount * 2][], 0, + this.foundTaskCount); + System.arraycopy(this.foundTaskPriorities, 0, + this.foundTaskPriorities = new char[this.foundTaskCount * 2][], + 0, this.foundTaskCount); + System.arraycopy(this.foundTaskPositions, 0, + this.foundTaskPositions = new int[this.foundTaskCount * 2][], 0, + this.foundTaskCount); } this.foundTaskTags[this.foundTaskCount] = tag; this.foundTaskPriorities[this.foundTaskCount] = priority; - this.foundTaskPositions[this.foundTaskCount] = new int[] { i, i + tagLength - 1 }; + this.foundTaskPositions[this.foundTaskCount] = new int[]{i, + i + tagLength - 1}; this.foundTaskMessages[this.foundTaskCount] = CharOperation.NO_CHAR; this.foundTaskCount++; - i += tagLength - 1; // will be incremented when looping } } - for (int i = foundTaskIndex; i < this.foundTaskCount; i++) { // retrieve message start and end positions - int msgStart = this.foundTaskPositions[i][0] + this.foundTaskTags[i].length; - int max_value = i + 1 < this.foundTaskCount ? this.foundTaskPositions[i + 1][0] - 1 : commentEnd - 1; + int msgStart = this.foundTaskPositions[i][0] + + this.foundTaskTags[i].length; + int max_value = i + 1 < this.foundTaskCount + ? this.foundTaskPositions[i + 1][0] - 1 + : commentEnd - 1; // at most beginning of next task if (max_value < msgStart) max_value = msgStart; // would only occur if tag is before EOF. int end = -1; char c; - for (int j = msgStart; j < max_value; j++) { if ((c = this.source[j]) == '\n' || c == '\r') { end = j - 1; break; } } - if (end == -1) { for (int j = max_value; j > msgStart; j--) { if ((c = this.source[j]) == '*') { @@ -3847,26 +4064,20 @@ public class Scanner implements IScanner, ITerminalSymbols { if (end == -1) end = max_value; } - if (msgStart == end) continue; // empty - // trim the message while (CharOperation.isWhitespace(source[end]) && msgStart <= end) end--; while (CharOperation.isWhitespace(source[msgStart]) && msgStart <= end) msgStart++; - // update the end position of the task this.foundTaskPositions[i][1] = end; - // get the message source final int messageLength = end - msgStart + 1; char[] message = new char[messageLength]; - System.arraycopy(source, msgStart, message, 0, messageLength); this.foundTaskMessages[i] = message; } } - }