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 bef873a..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 @@ -12,6 +12,7 @@ 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; @@ -32,6 +33,7 @@ public class Scanner implements IScanner, ITerminalSymbols { public boolean containsAssertKeyword = false; public boolean recordLineSeparator; public boolean phpMode = false; + public Stack encapsedStringStack = null; public char currentCharacter; public int startPosition; public int currentPosition; @@ -40,6 +42,7 @@ public class Scanner implements IScanner, ITerminalSymbols { // 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[]; @@ -131,12 +134,8 @@ public class Scanner implements IScanner, ITerminalSymbols { public char[][] taskTags = null; public char[][] taskPriorities = null; public static final boolean DEBUG = false; - public Scanner() { - this(false, false); - } - public Scanner(boolean tokenizeComments, boolean tokenizeWhiteSpace) { - this(tokenizeComments, tokenizeWhiteSpace, false); - } + public static final boolean TRACE = false; + /** * Determines if the specified character is permissible as the first * character in a PHP identifier @@ -675,66 +674,424 @@ 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++]) == '\\') @@ -765,10 +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--; @@ -781,7 +1269,7 @@ public class Scanner implements IScanner, ITerminalSymbols { // ---------Identify the next token------------- switch (currentCharacter) { case '(' : - return TokenNameLPAREN; + return getCastOrParen(); case ')' : return TokenNameRPAREN; case '{' : @@ -797,6 +1285,8 @@ public class Scanner implements IScanner, ITerminalSymbols { case ',' : return TokenNameCOMMA; case '.' : + if (getNextChar('=')) + return TokenNameDOT_EQUAL; if (getNextCharAsDigit()) return scanNumber(true); return TokenNameDOT; @@ -842,56 +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 '>' : @@ -941,7 +1444,11 @@ 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 ':' : @@ -950,389 +1457,30 @@ public class Scanner implements IScanner, ITerminalSymbols { 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 @@ -1362,7 +1510,8 @@ 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); @@ -1379,7 +1528,7 @@ public class Scanner implements IScanner, ITerminalSymbols { if (getNextChar('>')) { startPosition = currentPosition - 2; phpMode = false; - return TokenNameStopPHP; + return TokenNameINLINE_HTML; } } //get the next char @@ -1433,7 +1582,8 @@ public class Scanner implements IScanner, ITerminalSymbols { endPositionForLineComment = currentPosition - 1; } recordComment(false); - if ((currentCharacter == '\r') + if (this.taskTags != null) checkTaskTag(this.startPosition, this.currentPosition); + if ((currentCharacter == '\r') || (currentCharacter == '\n')) { checkNonExternalizeString(); if (recordLineSeparator) { @@ -1546,8 +1696,6 @@ public class Scanner implements IScanner, ITerminalSymbols { } break; } - if (getNextChar('=')) - return TokenNameDIVIDE_EQUAL; return TokenNameDIVIDE; } case '\u001a' : @@ -1558,13 +1706,19 @@ public class Scanner implements IScanner, ITerminalSymbols { 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); @@ -1579,6 +1733,63 @@ 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) { + //