Removed unuse builder from project
[phpeclipse.git] / net.sourceforge.phpeclipse / src / net / sourceforge / phpeclipse / phpeditor / php / PHPPartitionScanner.java
index c15249b..514e802 100644 (file)
 /**********************************************************************
-Copyright (c) 2000, 2002 IBM Corp. and others.
-All rights reserved. This program and the accompanying materials
-are made available under the terms of the Common Public License v1.0
-which accompanies this distribution, and is available at
-http://www.eclipse.org/legal/cpl-v10.html
-
-Contributors:
-    IBM Corporation - Initial implementation
-    Klaus Hartlage - www.eclipseproject.de
-**********************************************************************/
+ Copyright (c) 2002  Widespace, OU  and others.
+ All rights reserved.   This program and the accompanying materials
+ are made available under the terms of the Common Public License v1.0
+ which accompanies this distribution, and is available at
+ http://solareclipse.sourceforge.net/legal/cpl-v10.html
+
+ Contributors:
+ Igor Malinin - initial contribution
+
+ $Id: PHPPartitionScanner.java,v 1.34 2006-10-21 23:18:32 pombredanne Exp $
+ **********************************************************************/
 package net.sourceforge.phpeclipse.phpeditor.php;
 
-import java.util.ArrayList;
-import java.util.List;
+import java.util.HashMap;
+import java.util.Map;
 
+import net.sourceforge.phpdt.internal.compiler.parser.Scanner;
+import net.sourceforge.phpeclipse.ui.text.rules.AbstractPartitioner;
+
+import org.eclipse.jface.text.Assert;
+import org.eclipse.jface.text.BadLocationException;
+import org.eclipse.jface.text.IDocument;
 import org.eclipse.jface.text.rules.ICharacterScanner;
-import org.eclipse.jface.text.rules.IPredicateRule;
+import org.eclipse.jface.text.rules.IPartitionTokenScanner;
 import org.eclipse.jface.text.rules.IToken;
-import org.eclipse.jface.text.rules.MultiLineRule;
-import org.eclipse.jface.text.rules.RuleBasedPartitionScanner;
 import org.eclipse.jface.text.rules.Token;
 
 /**
- * This scanner recognizes the JavaDoc comments and Java multi line comments.
+ * 
+ * 
+ * @author Igor Malinin
  */
-public class PHPPartitionScanner extends RuleBasedPartitionScanner {
-
-  private final static String SKIP = "__skip"; //$NON-NLS-1$
-  public final static String HTML_MULTILINE_COMMENT = "__html_multiline_comment"; //$NON-NLS-1$
-  //   public final static String JAVA_DOC= "__java_javadoc"; //$NON-NLS-1$
-  public final static String PHP = "__php"; //$NON-NLS-1$
-  //public final static String HTML = "__html"; //$NON-NLS-1$
-
-  public final static IToken php = new Token(PHP);
-  //public final static IToken html = new Token(HTML);
-  public final static IToken comment = new Token(HTML_MULTILINE_COMMENT);
-
-  protected final static char[] php0EndSequence = { '<', '?' };
-  protected final static char[] php1EndSequence = { '<', '?', 'p', 'h', 'p' };
-  protected final static char[] php2EndSequence = { '<', '?', 'P', 'H', 'P' };
-
-  private StringBuffer test;
-
-  public class PHPMultiLineRule extends MultiLineRule {
-
-    public PHPMultiLineRule(String startSequence, String endSequence, IToken token) {
-      super(startSequence, endSequence, token);
-    }
-
-    public PHPMultiLineRule(String startSequence, String endSequence, IToken token, char escapeCharacter) {
-      super(startSequence, endSequence, token, escapeCharacter);
-    }
-
-    protected boolean endSequenceDetected(ICharacterScanner scanner) {
-      int c;
-      int c2;
-
-      boolean lineCommentMode = false;
-      boolean multiLineCommentMode = false;
-      boolean stringMode = false;
-
-      char[][] delimiters = scanner.getLegalLineDelimiters();
-      while ((c = scanner.read()) != ICharacterScanner.EOF) {
-        if (c == '#') {
-          // read until end of line
-          while ((c = scanner.read()) != ICharacterScanner.EOF) {
-            if (fEndSequence.length > 0 && c == fEndSequence[0]) {
-              // Check if the specified end sequence has been found.
-              if (sequenceDetected(scanner, fEndSequence, true))
-                return true;
-            } else if (c == '\n') {
-              break;
-            }
-          }
-          continue;
-        } else if (c == '/' && (c = scanner.read()) != ICharacterScanner.EOF) {
-          if (c == '/') {
-            // read until end of line
-            while ((c = scanner.read()) != ICharacterScanner.EOF) {
-              if (fEndSequence.length > 0 && c == fEndSequence[0]) {
-                // Check if the specified end sequence has been found.
-                if (sequenceDetected(scanner, fEndSequence, true))
-                  return true;
-              } else if (c == '\n') {
-                break;
-              }
-            }
-            continue;
-          } else if (c == '*') {
-            // multi-line comment
-            while ((c = scanner.read()) != ICharacterScanner.EOF) {
-              if (c == '*' && (c = scanner.read()) != ICharacterScanner.EOF) {
-                if (c == '/') {
-                  break;
-                }
-                scanner.unread();
-              }
-            }
-
-            continue;
-          } else {
-            scanner.unread();
-          }
-        } else if (c == '"') {
-          // string mode
-          while ((c = scanner.read()) != ICharacterScanner.EOF) {
-            if (c == '\\') {
-              c = scanner.read();
-            } else if (c == '"') {
-              break;
-            }
-          }
-          continue;
-        } else if (c == '\'') {
-          // string mode
-          while ((c = scanner.read()) != ICharacterScanner.EOF) {
-            if (c == '\\') {
-              c = scanner.read();
-            } else if (c == '\'') {
-              break;
-            }
-          }
-          continue;
-        }
-
-        if (c == fEscapeCharacter) {
-          // Skip the escaped character.
-          scanner.read();
-        } else if (fEndSequence.length > 0 && c == fEndSequence[0]) {
-          // Check if the specified end sequence has been found.
-          if (sequenceDetected(scanner, fEndSequence, true))
-            return true;
-        } else if (fBreaksOnEOL) {
-          // Check for end of line since it can be used to terminate the pattern.
-          for (int i = 0; i < delimiters.length; i++) {
-            if (c == delimiters[i][0] && sequenceDetected(scanner, delimiters[i], false))
-              return true;
-          }
-        }
-      } 
-      boolean phpMode = false;
-      if (c == ICharacterScanner.EOF) {
-        phpMode = true;
-      }
-      scanner.unread();
-      return phpMode;
-    }
-  }
-
-  //  public class HTMLMultiLineRule extends MultiLineRule {
-  //
-  //    public HTMLMultiLineRule(String startSequence, String endSequence, IToken token) {
-  //      super(startSequence, endSequence, token);
-  //    }
-  //
-  //    public HTMLMultiLineRule(String startSequence, String endSequence, IToken token, char escapeCharacter) {
-  //      super(startSequence, endSequence, token, escapeCharacter);
-  //    }
-  //
-  //    protected boolean endSequenceDetected(ICharacterScanner scanner) {
-  //      int c;
-  //
-  //      char[][] delimiters = scanner.getLegalLineDelimiters();
-  //      while ((c = scanner.read()) != ICharacterScanner.EOF) {
-  //        if (c == '<') {
-  //          //       scanner.unread();
-  //          if (sequenceDetected(scanner, php2EndSequence, true)) {
-  //            // <?PHP
-  //            scanner.unread();
-  //            scanner.unread();
-  //            scanner.unread();
-  //            scanner.unread();
-  //            scanner.unread();
-  //            return true;
-  //          }
-  //          if (sequenceDetected(scanner, php1EndSequence, true)) {
-  //            // <?php
-  //            scanner.unread();
-  //            scanner.unread();
-  //            scanner.unread();
-  //            scanner.unread();
-  //            scanner.unread();
-  //            return true;
-  //          }
-  //          if (sequenceDetected(scanner, php0EndSequence, true)) {
-  //            // <?
-  //            scanner.unread();
-  //            scanner.unread();
-  //            return true;
-  //          }
-  //          //      scanner.read();
-  //        }
-  //
-  //      }
-  //      scanner.unread();
-  //      return false;
-  //    }
-  //
-  //    protected IToken doEvaluate(ICharacterScanner scanner, boolean resume) {
-  //
-  //      if (resume) {
-  //
-  //        if (endSequenceDetected(scanner))
-  //          return fToken;
-  //
-  //      } else {
-  //
-  //        int c = scanner.read();
-  //        //     if (c == fStartSequence[0]) {
-  //        //       if (sequenceDetected(scanner, fStartSequence, false)) {
-  //        if (endSequenceDetected(scanner))
-  //          return fToken;
-  //        //       }
-  //        //     }
-  //      }
-  //
-  //      scanner.unread();
-  //      return Token.UNDEFINED;
-  //    }
-  //
-  //    public IToken evaluate(ICharacterScanner scanner, boolean resume) {
-  //      if (fColumn == UNDEFINED)
-  //        return doEvaluate(scanner, resume);
-  //
-  //      int c = scanner.read();
-  //      scanner.unread();
-  //      //    if (c == fStartSequence[0])
-  //      return (fColumn == scanner.getColumn() ? doEvaluate(scanner, resume) : Token.UNDEFINED);
-  //      //    else
-  //      //      return Token.UNDEFINED;
-  //    }
-  //  }
-
-//  public class HTMLPatternRule implements IPredicateRule {
-//
-//    protected static final int UNDEFINED = -1;
-//
-//    /** The token to be returned on success */
-//    protected IToken fToken;
-//
-//    /** The pattern's column constrain */
-//    protected int fColumn = UNDEFINED;
-//    /** The pattern's escape character */
-//    protected char fEscapeCharacter;
-//    /** Indicates whether end of line termines the pattern */
-//    protected boolean fBreaksOnEOL;
-//
-//    /**
-//     * Creates a rule for the given starting and ending sequence.
-//     * When these sequences are detected the rule will return the specified token.
-//     * Alternatively, the sequence can also be ended by the end of the line.
-//     * Any character which follows the given escapeCharacter will be ignored.
-//     *
-//     * @param startSequence the pattern's start sequence
-//     * @param endSequence the pattern's end sequence, <code>null</code> is a legal value
-//     * @param token the token which will be returned on success
-//     * @param escapeCharacter any character following this one will be ignored
-//     * @param indicates whether the end of the line also termines the pattern
-//     */
-//    public HTMLPatternRule(IToken token) {
-//      fToken = token;
-//      fEscapeCharacter = (char) 0;
-//      fBreaksOnEOL = false;
-//    }
-//
-//    /**
-//     * Sets a column constraint for this rule. If set, the rule's token
-//     * will only be returned if the pattern is detected starting at the 
-//     * specified column. If the column is smaller then 0, the column
-//     * constraint is considered removed.
-//     *
-//     * @param column the column in which the pattern starts
-//     */
-//    public void setColumnConstraint(int column) {
-//      if (column < 0)
-//        column = UNDEFINED;
-//      fColumn = column;
-//    }
-//
-//    /**
-//     * Evaluates this rules without considering any column constraints.
-//     *
-//     * @param scanner the character scanner to be used
-//     * @return the token resulting from this evaluation
-//     */
-//    protected IToken doEvaluate(ICharacterScanner scanner) {
-//      return doEvaluate(scanner, false);
-//    }
-//
-//    /**
-//     * Evaluates this rules without considering any column constraints. Resumes
-//     * detection, i.e. look sonly for the end sequence required by this rule if the
-//     * <code>resume</code> flag is set.
-//     *
-//     * @param scanner the character scanner to be used
-//     * @param resume <code>true</code> if detection should be resumed, <code>false</code> otherwise
-//     * @return the token resulting from this evaluation
-//     * @since 2.0
-//     */
-//    protected IToken doEvaluate(ICharacterScanner scanner, boolean resume) {
-//
-//      if (resume) {
-//
-//        if (endSequenceDetected(scanner))
-//          return fToken;
-//
-//      } else {
-//
-//        int c = scanner.read();
-//        //      if (c == fStartSequence[0]) {
-//        //        if (sequenceDetected(scanner, fStartSequence, false)) {
-//        if (endSequenceDetected(scanner))
-//          return fToken;
-//        //        }
-//        //      }
-//      }
-//
-//      scanner.unread();
-//      return Token.UNDEFINED;
-//    }
-//
-//    /*
-//     * @see IRule#evaluate
-//     */
-//    public IToken evaluate(ICharacterScanner scanner) {
-//      return evaluate(scanner, false);
-//    }
-//
-//    /**
-//     * Returns whether the end sequence was detected. As the pattern can be considered 
-//     * ended by a line delimiter, the result of this method is <code>true</code> if the 
-//     * rule breaks on the end  of the line, or if the EOF character is read.
-//     *
-//     * @param scanner the character scanner to be used
-//     * @return <code>true</code> if the end sequence has been detected
-//     */
-//    protected boolean endSequenceDetected(ICharacterScanner scanner) {
-//      int c;
-//
-//      char[][] delimiters = scanner.getLegalLineDelimiters();
-//      while ((c = scanner.read()) != ICharacterScanner.EOF) {
-//        if (c == '<') {
-//          //       scanner.unread();
-//          if (sequenceDetected(scanner, php2EndSequence, true)) {
-//            // <?PHP
-//            scanner.unread();
-//            scanner.unread();
-//            scanner.unread();
-//            scanner.unread();
-//            scanner.unread();
-//            return true;
-//          }
-//          if (sequenceDetected(scanner, php1EndSequence, true)) {
-//            // <?php
-//            scanner.unread();
-//            scanner.unread();
-//            scanner.unread();
-//            scanner.unread();
-//            scanner.unread();
-//            return true;
-//          }
-//          if (sequenceDetected(scanner, php0EndSequence, true)) {
-//            // <?
-//            scanner.unread();
-//            scanner.unread();
-//            return true;
-//          }
-//          //      scanner.read();
-//        }
-//
-//      }
-//      scanner.unread();
-//      return false;
-//    }
-//
-//    /**
-//     * Returns whether the next characters to be read by the character scanner
-//     * are an exact match with the given sequence. No escape characters are allowed 
-//     * within the sequence. If specified the sequence is considered to be found
-//     * when reading the EOF character.
-//     *
-//     * @param scanner the character scanner to be used
-//     * @param sequence the sequence to be detected
-//     * @param eofAllowed indicated whether EOF terminates the pattern
-//     * @return <code>true</code> if the given sequence has been detected
-//     */
-//    protected boolean sequenceDetected(ICharacterScanner scanner, char[] sequence, boolean eofAllowed) {
-//      for (int i = 1; i < sequence.length; i++) {
-//        int c = scanner.read();
-//        if (c == ICharacterScanner.EOF && eofAllowed) {
-//          return true;
-//        } else if (c != sequence[i]) {
-//          // Non-matching character detected, rewind the scanner back to the start.
-//          scanner.unread();
-//          for (int j = i - 1; j > 0; j--)
-//            scanner.unread();
-//          return false;
-//        }
-//      }
-//
-//      return true;
-//    }
-//
-//    /*
-//     * @see IPredicateRule#evaluate(ICharacterScanner, boolean)
-//     * @since 2.0
-//     */
-//    public IToken evaluate(ICharacterScanner scanner, boolean resume) {
-//      if (fColumn == UNDEFINED)
-//        return doEvaluate(scanner, resume);
-//
-//      int c = scanner.read();
-//      scanner.unread();
-//      //    if (c == fStartSequence[0])
-//      return (fColumn == scanner.getColumn() ? doEvaluate(scanner, resume) : Token.UNDEFINED);
-//      //    else
-//      //      return Token.UNDEFINED;
-//    }
-//
-//    /*
-//     * @see IPredicateRule#getSuccessToken()
-//     * @since 2.0
-//     */
-//    public IToken getSuccessToken() {
-//      return fToken;
-//    }
-//  }
-  /**
-   * Detector for empty comments.
-   */
-//  static class EmptyCommentDetector implements IWordDetector {
-//
-//    /* (non-Javadoc)
-//    * Method declared on IWordDetector
-//     */
-//    public boolean isWordStart(char c) {
-//      return (c == '/');
-//    }
-//
-//    /* (non-Javadoc)
-//    * Method declared on IWordDetector
-//     */
-//    public boolean isWordPart(char c) {
-//      return (c == '*' || c == '/');
-//    }
-//  };
-
-  /**
-   * 
-   */
-//  static class WordPredicateRule extends WordRule implements IPredicateRule {
-//
-//    private IToken fSuccessToken;
-//
-//    public WordPredicateRule(IToken successToken) {
-//      super(new EmptyCommentDetector());
-//      fSuccessToken = successToken;
-//      addWord("/**/", fSuccessToken);
-//    }
-//
-//    /*
-//     * @see org.eclipse.jface.text.rules.IPredicateRule#evaluate(ICharacterScanner, boolean)
-//     */
-//    public IToken evaluate(ICharacterScanner scanner, boolean resume) {
-//      return super.evaluate(scanner);
-//    }
-//
-//    /*
-//     * @see org.eclipse.jface.text.rules.IPredicateRule#getSuccessToken()
-//     */
-//    public IToken getSuccessToken() {
-//      return fSuccessToken;
-//    }
-//  };
-
-  /**
-   * Creates the partitioner and sets up the appropriate rules.
-   */
-  public PHPPartitionScanner() {
-    super();
-
-    //    IToken php = new Token(PHP);
-    //    IToken html = new Token(HTML);
-    //    IToken comment = new Token(HTML_MULTILINE_COMMENT);
-
-    List rules = new ArrayList();
-
-    // Add rule for single line comments.
-    // rules.add(new EndOfLineRule("//", Token.UNDEFINED));
-
-    // Add rule for strings and character constants.
-    //         rules.add(new SingleLineRule("\"", "\"", Token.UNDEFINED, '\\'));
-    // rules.add(new SingleLineRule("'", "'", Token.UNDEFINED, '\\')); 
-
-    // Add special case word rule.
-    //    rules.add(new WordPredicateRule(comment));
-
-    // Add rules for multi-line comments and javadoc.
-    //rules.add(new MultiLineRule("/**", "*/", javaDoc));
-    //  rules.add(new HTMLMultiLineRule("<", "<?", html));
-
-    rules.add(new MultiLineRule("<!--", "-->", comment));
-    rules.add(new PHPMultiLineRule("<?\r", "?>", php));
-    rules.add(new PHPMultiLineRule("<?\n", "?>", php));
-    rules.add(new PHPMultiLineRule("<?\t", "?>", php));
-    rules.add(new PHPMultiLineRule("<? ", "?>", php));
-    rules.add(new PHPMultiLineRule("<?php", "?>", php));
-    rules.add(new PHPMultiLineRule("<?PHP", "?>", php));
-
-    //    rules.add(new HTMLPatternRule(html)); // "<", "<?",
-    //Add rule for processing instructions
-
-    IPredicateRule[] result = new IPredicateRule[rules.size()];
-    rules.toArray(result);
-    setPredicateRules(result);
-   // setDefaultReturnToken(html);
-  }
-
-  //    public IToken nextToken() {
-  //      
-  //      if (fContentType == null || fRules == null)
-  //        return getNextToken();
-  //      
-  //      fTokenOffset= fOffset;
-  //      fColumn= UNDEFINED;
-  //      boolean resume= (fPartitionOffset < fOffset);
-  //          
-  //      IPredicateRule rule;
-  //      IToken token;
-  //      
-  //      for (int i= 0; i < fRules.length; i++) {
-  //        rule= (IPredicateRule) fRules[i];
-  //        token= rule.getSuccessToken();
-  //        if (fContentType.equals(token.getData())) {
-  //          if (resume)
-  //            fTokenOffset= fPartitionOffset;
-  //          token= rule.evaluate(this, resume);
-  //          if (!token.isUndefined()) {
-  //            fContentType= null;
-  //            return token;
-  //          }
-  //        }
-  //      }
-  //      
-  //      fContentType= null;
-  //      return getNextToken();
-  //    }
-  //    
-  //    public IToken getNextToken() {
-  //      
-  //      IToken token;
-  //      
-  //      while (true) {
-  //        
-  //        fTokenOffset= fOffset;
-  //        fColumn= UNDEFINED;
-  //        
-  //        if (fRules != null) {
-  //          for (int i= 0; i < fRules.length; i++) {
-  //            token= (fRules[i].evaluate(this));
-  //            if (!token.isUndefined())
-  //              return token;
-  //          }
-  //        }
-  //        
-  //        if (read() == EOF)
-  //          return Token.EOF;
-  //        else
-  //          return fDefaultReturnToken;
-  //      }
-  //    }
-}
+public class PHPPartitionScanner implements IPartitionTokenScanner {
+       public static final String PHP_SCRIPTING_AREA = "__php_scripting_area ";
+
+       public static final int STATE_DEFAULT = 0;
+
+       // public static final int STATE_TAG = 1;
+       // public static final int STATE_SCRIPT = 2;
+
+       private IDocument document;
+
+       // private int begin;
+
+       private int end;
+
+       private int offset;
+
+       private int length;
+
+       private int position;
+
+       // private int state;
+
+       private Map tokens = new HashMap();
+
+       public PHPPartitionScanner() {
+       }
+
+       /*
+        * @see org.eclipse.jface.text.rules.ITokenScanner#nextToken()
+        */
+       public IToken nextToken() {
+               offset += length;
+
+               /*
+                * switch (state) { case STATE_TAG: return nextTagToken(); }
+                */
+
+               switch (read()) {
+               case ICharacterScanner.EOF:
+                       // state = STATE_DEFAULT;
+                       return getToken(null);
+
+               case '<':
+                       switch (read()) {
+                       case ICharacterScanner.EOF:
+                               // state = STATE_DEFAULT;
+                               return getToken(null);
+
+                       case '?': // <?
+                               // int ch = read();
+                               //
+                               // switch (ch) {
+                               // case ICharacterScanner.EOF:
+                               // state = STATE_DEFAULT;
+                               // return getToken(PHP_SCRIPTING_AREA);
+                               // }
+                               return scanUntilPHPEndToken(PHP_SCRIPTING_AREA);
+                       }
+
+                       unread();
+               }
+
+               loop: while (true) {
+                       switch (read()) {
+                       case ICharacterScanner.EOF:
+                               // state = STATE_DEFAULT;
+                               return getToken(null);
+
+                       case '<':
+                               switch (read()) {
+                               case ICharacterScanner.EOF:
+                                       // state = STATE_DEFAULT;
+                                       return getToken(null);
+
+                               case '?':
+                                       unread();
+                                       break;
+
+                               case '<':
+                                       unread();
+
+                               default:
+                                       continue loop;
+                               }
+
+                               unread();
+
+                               // state = STATE_DEFAULT;
+                               return getToken(null);
+                       }
+               }
+       }
+
+       private IToken scanUntilPHPEndToken(String token) {
+               int ch = read();
+               while (true) {
+                       switch (ch) {
+                       case ICharacterScanner.EOF:
+                               // state = STATE_DEFAULT;
+                               return getToken(token);
+                       case '"': // double quoted string
+                               // read until end of double quoted string
+                               if (!readUntilEscapedDQ()) {
+                                       // state = STATE_DEFAULT;
+                                       return getToken(token);
+                               }
+                               break;
+                       case '<': // heredoc string
+                               ch = read();
+                               switch (ch) {
+                               case ICharacterScanner.EOF:
+                                       break;
+                               case '<':
+                                       ch = read();
+                                       switch (ch) {
+                                       case ICharacterScanner.EOF:
+                                               break;
+                                       case '<':
+                                               // read until end of heredoc string
+                                               if (!readUntilEscapedHEREDOC()) {
+                                                       // state = STATE_DEFAULT;
+                                                       return getToken(token);
+                                               }
+                                       }
+                               }
+                               break;
+                       case '\'': // single quoted string
+                               // read until end of single quoted string
+                               if (!readUntilEscapedSQ()) {
+                                       // state = STATE_DEFAULT;
+                                       return getToken(token);
+                               }
+                               break;
+                       case '/': // comment start?
+                               ch = read();
+                               switch (ch) {
+                               case ICharacterScanner.EOF:
+                                       break;
+                               case '/':
+                                       // read until end of line
+                                       if (!readSingleLine()) {
+                                               // state = STATE_DEFAULT;
+                                               return getToken(token);
+                                       }
+                                       break;
+                               case '*':
+                                       // read until end of comment
+                                       if (!readMultiLineComment()) {
+                                               // state = STATE_DEFAULT;
+                                               return getToken(token);
+                                       }
+                                       break;
+                               default:
+                                       continue;
+                               }
+                               break;
+                       case '#': // line comment
+                               // read until end of line
+                               if (!readSingleLine()) {
+                                       // state = STATE_DEFAULT;
+                                       return getToken(token);
+                               }
+                               break;
+                       case '?':
+                               ch = read();
+                               switch (ch) {
+                               case ICharacterScanner.EOF:
+                               case '>':
+                                       // state = STATE_DEFAULT;
+                                       return getToken(token);
+
+                               case '?':
+                                       continue;
+                               default:
+                                       continue;
+                               }
+                       }
+
+                       ch = read();
+               }
+       }
+
+       private IToken getToken(String type) {
+               length = position - offset;
+
+               if (length == 0) {
+                       return Token.EOF;
+               }
+
+               // if (length<0) {
+               // try {
+               // System.out.println("Length<0:"+document.get(offset,5)+""+length);
+               // } catch (BadLocationException e) {
+               // e.printStackTrace();
+               // }
+               // }
+
+               if (type == null) {
+                       return Token.UNDEFINED;
+               }
+
+               IToken token = (IToken) tokens.get(type);
+               if (token == null) {
+                       token = new Token(type);
+                       tokens.put(type, token);
+               }
+
+               return token;
+       }
+
+       private int read() {
+               if (position >= end) {
+                       return ICharacterScanner.EOF;
+               }
+
+               try {
+                       return document.getChar(position++);
+               } catch (BadLocationException e) {
+                       --position;
+                       return ICharacterScanner.EOF;
+               }
+       }
+
+       private boolean readUntilEscapedDQ() {
+               // search last double quoted character
+               try {
+                       char ch;
+                       while (true) {
+                               if (position >= end) {
+                                       return false;
+                               }
+                               ch = document.getChar(position++);
+                               if (ch == '\\') {
+                                       if (position >= end) {
+                                               return false;
+                                       }
+                                       ch = document.getChar(position++); // ignore escaped
+                                                                                                               // character
+                               } else if (ch == '"') {
+                                       return true;
+                               }
+                       }
+               } catch (BadLocationException e) {
+                       --position;
+               }
+               return false;
+       }
+
+       private boolean readUntilEscapedSQ() {
+               // search last single quoted character
+               try {
+                       char ch;
+                       while (true) {
+                               if (position >= end) {
+                                       return false;
+                               }
+                               ch = document.getChar(position++);
+                               if (ch == '\\') {
+                                       if (position >= end) {
+                                               return false;
+                                       }
+                                       ch = document.getChar(position++); // ignore escaped
+                                                                                                               // character
+                               } else if (ch == '\'') {
+                                       return true;
+                               }
+                       }
+               } catch (BadLocationException e) {
+                       --position;
+               }
+               return false;
+       }
+
+       private boolean readUntilEscapedHEREDOC() {
+               // search until heredoc ends
+               try {
+                       char ch;
+                       StringBuffer buf = new StringBuffer();
+                       char[] heredocIdent;
+                       if (position >= end) {
+                               return false;
+                       }
+                       ch = document.getChar(position++);
+                       if (!Scanner.isPHPIdentifierStart(ch)) {
+                               return false;
+                       }
+                       while (Scanner.isPHPIdentifierPart(ch)) {
+                               buf.append(ch);
+                               if (position >= end) {
+                                       return false;
+                               }
+                               ch = document.getChar(position++);
+                       }
+                       heredocIdent = buf.toString().toCharArray();
+                       while (true) {
+                               if (position >= end) {
+                                       return false;
+                               }
+                               ch = document.getChar(position++);
+                               if (ch == '\n') { // heredoc could end after a newline
+                                       int pos = 0;
+                                       while (true) {
+                                               if (position >= end) {
+                                                       return false;
+                                               }
+                                               if (pos == heredocIdent.length) {
+                                                       return true;
+                                               }
+                                               ch = document.getChar(position++); // ignore escaped
+                                                                                                                       // character
+                                               if (ch != heredocIdent[pos]) {
+                                                       break;
+                                               }
+                                               pos++;
+                                       }
+                               }
+                       }
+               } catch (BadLocationException e) {
+                       --position;
+               }
+               return false;
+       }
+
+       private boolean readSingleLine() {
+               try {
+                       do {
+                               if (position >= end) {
+                                       return false;
+                               }
+                       } while (document.getChar(position++) != '\n');
+                       return true;
+               } catch (BadLocationException e) {
+                       --position;
+               }
+               return false;
+       }
+
+       private boolean readMultiLineComment() {
+               try {
+                       char ch;
+                       while (true) {
+                               if (position >= end) {
+                                       return false;
+                               }
+                               ch = document.getChar(position++);
+                               if (ch == '*') {
+                                       if (position >= end) {
+                                               return false;
+                                       }
+                                       if (document.getChar(position) == '/') {
+                                               position++;
+                                               return true;
+                                       }
+                               }
+                       }
+               } catch (BadLocationException e) {
+                       --position;
+               }
+               return false;
+       }
+
+       private void unread() {
+               --position;
+       }
+
+       /*
+        * @see org.eclipse.jface.text.rules.ITokenScanner#getTokenOffset()
+        */
+       public int getTokenOffset() {
+               if (AbstractPartitioner.DEBUG) {
+                       Assert.isTrue(offset >= 0, Integer.toString(offset));
+               }
+               return offset;
+       }
+
+       /*
+        * @see org.eclipse.jface.text.rules.ITokenScanner#getTokenLength()
+        */
+       public int getTokenLength() {
+               return length;
+       }
+
+       /*
+        * @see org.eclipse.jface.text.rules.ITokenScanner#setRange(IDocument, int,
+        *      int)
+        */
+       public void setRange(IDocument document, int offset, int length) {
+               this.document = document;
+               // this.begin = offset;
+               this.end = offset + length;
+
+               this.offset = offset;
+               this.position = offset;
+               this.length = 0;
+       }
+
+       /*
+        * @see org.eclipse.jface.text.rules.IPartitionTokenScanner
+        */
+       public void setPartialRange(IDocument document, int offset, int length,
+                       String contentType, int partitionOffset) {
+               // state = STATE_DEFAULT;
+               if (partitionOffset > -1) {
+                       int delta = offset - partitionOffset;
+                       if (delta > 0) {
+                               setRange(document, partitionOffset, length + delta);
+                               return;
+                       }
+               }
+               setRange(document, partitionOffset, length);
+       }
+
+       // private boolean isContinuationPartition(IDocument document, int offset) {
+       // try {
+       // String type = document.getContentType(offset - 1);
+       //
+       // if (type != IDocument.DEFAULT_CONTENT_TYPE) {
+       // return true;
+       // }
+       // } catch (BadLocationException e) {}
+       //
+       // return false;
+       // }
+}
\ No newline at end of file