X-Git-Url: http://git.phpeclipse.com
diff --git a/net.sourceforge.phpeclipse/src/net/sourceforge/phpeclipse/phpeditor/php/PHPPartitionScanner.java b/net.sourceforge.phpeclipse/src/net/sourceforge/phpeclipse/phpeditor/php/PHPPartitionScanner.java
index bebcc1f..54eaa7d 100644
--- a/net.sourceforge.phpeclipse/src/net/sourceforge/phpeclipse/phpeditor/php/PHPPartitionScanner.java
+++ b/net.sourceforge.phpeclipse/src/net/sourceforge/phpeclipse/phpeditor/php/PHPPartitionScanner.java
@@ -1,554 +1,377 @@
-/**********************************************************************
-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
-**********************************************************************/
+/**
+ * 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
+ * Created on 05.03.2003
+ *
+ * @author Stefan Langer (musk)
+ * @version $Revision: 1.15 $
+ */
package net.sourceforge.phpeclipse.phpeditor.php;
-import java.io.CharArrayWriter;
-import java.util.ArrayList;
-import java.util.List;
+import java.util.*;
-import org.eclipse.jface.text.rules.ICharacterScanner;
-import org.eclipse.jface.text.rules.IPredicateRule;
-import org.eclipse.jface.text.rules.IToken;
-import org.eclipse.jface.text.rules.IWordDetector;
-import org.eclipse.jface.text.rules.MultiLineRule;
-import org.eclipse.jface.text.rules.RuleBasedPartitionScanner;
-import org.eclipse.jface.text.rules.Token;
-import org.eclipse.jface.text.rules.WordRule;
+import org.eclipse.jface.text.*;
+import org.eclipse.jface.text.rules.*;
/**
- * This scanner recognizes the JavaDoc comments and Java multi line comments.
+ *
*/
-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";
-// public final static String HTML = "__html";
-
- 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);
+public class PHPPartitionScanner implements IPartitionTokenScanner {
+ private static final boolean DEBUG = false;
+ private IDocument fDocument = null;
+ private int fOffset = -1;
+ private String fContentType = IPHPPartitionScannerConstants.HTML;
+ private String fPrevContentType;
+
+ private boolean partitionBorder = false;
+ private int fTokenOffset;
+ private int fEnd = -1;
+ private int fLength;
+ private Map tokens = new HashMap();
- 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 PHPPartitionScanner() {
+ this.tokens.put(
+ IPHPPartitionScannerConstants.PHP,
+ new Token(IPHPPartitionScannerConstants.PHP));
+ this.tokens.put(
+ IPHPPartitionScannerConstants.PHP_MULTILINE_COMMENT,
+ new Token(IPHPPartitionScannerConstants.PHP_MULTILINE_COMMENT));
+ this.tokens.put(
+ IPHPPartitionScannerConstants.HTML,
+ new Token(IPHPPartitionScannerConstants.HTML));
+ this.tokens.put(
+ IPHPPartitionScannerConstants.HTML_MULTILINE_COMMENT,
+ new Token(IPHPPartitionScannerConstants.HTML_MULTILINE_COMMENT));
+ this.tokens.put(
+ IDocument.DEFAULT_CONTENT_TYPE,
+ new Token(IDocument.DEFAULT_CONTENT_TYPE));
+ }
- public PHPMultiLineRule(String startSequence, String endSequence, IToken token) {
- super(startSequence, endSequence, token);
+ private IToken getToken(String type) {
+ fLength = fOffset - fTokenOffset;
+ IToken token = (IToken) this.tokens.get(type);
+ Assert.isNotNull(token, "Token for type \"" + type + "\" not found!");
+ if (DEBUG) {
+ System.out.println(
+ "Partition: fTokenOffset="
+ + fTokenOffset
+ + " fContentType="
+ + fContentType
+ + " fLength="
+ + fLength);
}
+ return token;
+ }
- public PHPMultiLineRule(String startSequence, String endSequence, IToken token, char escapeCharacter) {
- super(startSequence, endSequence, token, escapeCharacter);
+/* (non-Javadoc)
+ * @see org.eclipse.jface.text.rules.IPartitionTokenScanner#setPartialRange(org.eclipse.jface.text.IDocument, int, int, java.lang.String, int)
+ */
+public void setPartialRange(
+ IDocument document,
+ int offset,
+ int length,
+ String contentType,
+ int partitionOffset)
+{
+ this.setRange(document, offset, length);
+ if (DEBUG)
+ {
+ System.out.println(
+ "PartialRange: contentType="
+ + contentType
+ + " partitionOffset="
+ + partitionOffset);
}
-
- 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 (lineCommentMode && (c == '\n')) {
- lineCommentMode = false;
- // read until end of line
- } else if ((!stringMode) && (c == '#')) {
- // read until end of line
- lineCommentMode = true;
- continue;
- } else if ((!stringMode) && (!multiLineCommentMode) && (c == '/')) {
- c2 = scanner.read();
- if (c2 == '/') {
- lineCommentMode = true;
- continue;
- } else if (c2 == '*') {
- multiLineCommentMode = true;
- continue;
- } else {
- scanner.unread();
- }
- } else if (c == '*' && multiLineCommentMode) {
- c2 = scanner.read();
- if (c2 == '/') {
- multiLineCommentMode = false;
- continue;
- } else {
- scanner.unread();
- }
- } else if (c == '\\' && stringMode) {
- c2 = scanner.read();
- if (c2 == '"') {
- continue;
- } else {
- scanner.unread();
- }
- } else if ((!lineCommentMode) && (!multiLineCommentMode) && (c == '"')) {
- if (stringMode) {
- stringMode = false;
- } else {
- stringMode = true;
- }
- continue;
+
+ if(offset == partitionOffset)
+ {
+ try
+ {
+ fContentType = fDocument.getContentType(offset);
}
- if (lineCommentMode || multiLineCommentMode || stringMode) {
- 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;
- }
+ catch (BadLocationException e)
+ {
+ //should never happen
}
- }
- scanner.unread();
- return false;
- }
- }
-
-// 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)) {
-// // null 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);
+ }
+ else if (this.tokens.containsKey(contentType))
+ fContentType = contentType;
+ // TODO Calculate previouse contenttype
+ if (partitionOffset > -1)
+ {
+ partitionBorder = false;
+ fTokenOffset = partitionOffset;
}
+}
- /**
- * Evaluates this rules without considering any column constraints. Resumes
- * detection, i.e. look sonly for the end sequence required by this rule if the
- * resume flag is set.
- *
- * @param scanner the character scanner to be used
- * @param resume true if detection should be resumed, false 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 {
+ /* (non-Javadoc)
+ * @see org.eclipse.jface.text.rules.ITokenScanner#getTokenLength()
+ */
+ public int getTokenLength() {
+ return fLength;
+ }
- int c = scanner.read();
- // if (c == fStartSequence[0]) {
- // if (sequenceDetected(scanner, fStartSequence, false)) {
- if (endSequenceDetected(scanner))
- return fToken;
- // }
- // }
- }
+ /* (non-Javadoc)
+ * @see org.eclipse.jface.text.rules.ITokenScanner#getTokenOffset()
+ */
+ public int getTokenOffset() {
+ return fTokenOffset;
+ }
- scanner.unread();
- return Token.UNDEFINED;
+/* (non-Javadoc)
+ * @see org.eclipse.jface.text.rules.ITokenScanner#nextToken()
+ */
+public IToken nextToken()
+{
+ int c;
+
+ // check if we are not allready at the end of the
+ // file
+ if ((c = read()) == ICharacterScanner.EOF)
+ {
+ partitionBorder = false;
+ return Token.EOF;
}
+ else
+ unread();
- /*
- * @see IRule#evaluate
- */
- public IToken evaluate(ICharacterScanner scanner) {
- return evaluate(scanner, false);
+ if (partitionBorder)
+ {
+ fTokenOffset = fOffset;
+ partitionBorder = 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 true 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 true 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)) {
- // 0)
+ {
+ fOffset -= 5;
+ IToken token = getToken(fContentType);
+ // save previouse contenttype
+ fPrevContentType = fContentType;
+
+ fContentType = IPHPPartitionScannerConstants.PHP;
+
+ return token;
+ }
+ else
+ fContentType = IPHPPartitionScannerConstants.PHP;
+
+ // remember offset of this partition
+ fTokenOffset = fOffset - 5;
+ }
+ else if (
+ fContentType
+ != IPHPPartitionScannerConstants.PHP_MULTILINE_COMMENT
+ && checkPattern(new char[] { '?' }, false))
+ {
+ if (fContentType != IPHPPartitionScannerConstants.PHP
+ && fOffset - 2 > 0)
+ {
+ fOffset -= 2;
+ IToken token = getToken(fContentType);
+ // save previouse contenttype
+ fPrevContentType =
+ fContentType;
+ fContentType = IPHPPartitionScannerConstants.PHP;
+ return token;
+ }
+ else
+ fContentType = IPHPPartitionScannerConstants.PHP;
+ // remember offset of this partition
+ fTokenOffset = fOffset - 2;
+ }
+ else if (checkPattern(new char[] { '!', '-', '-' }))
+ { // return previouse partition
+ if (
+ fContentType
+ != IPHPPartitionScannerConstants.HTML_MULTILINE_COMMENT
+ && fOffset - 4 > 0)
+ {
+ fOffset -= 4;
+ IToken token = getToken(fContentType);
+ fContentType =
+ IPHPPartitionScannerConstants
+ .HTML_MULTILINE_COMMENT;
+ return token;
+ }
+ else
+ fContentType =
+ IPHPPartitionScannerConstants
+ .HTML_MULTILINE_COMMENT;
+ fTokenOffset = fOffset - 4;
+ }
+ break; case '?' :
+ if (fContentType == IPHPPartitionScannerConstants.PHP)
+ {
+ if ((c = read()) == '>')
+ { // TODO Actually calculate the previouse contenttype from the document
+ if (
+ fPrevContentType
+ != null)
+ fContentType = fPrevContentType; else
+ fContentType =
+ IPHPPartitionScannerConstants.HTML;
+ partitionBorder = true;
+ return getToken(
+ IPHPPartitionScannerConstants.PHP);
+ }
+ else if (c != ICharacterScanner.EOF)
+ unread(); }
+ break; case '-' :
+ if (fContentType
+ == IPHPPartitionScannerConstants.HTML_MULTILINE_COMMENT
+ && checkPattern(new char[] { '-', '>' }))
+ {
+ fContentType = IPHPPartitionScannerConstants.HTML;
+ partitionBorder = true;
+ return getToken(
+ IPHPPartitionScannerConstants
+ .HTML_MULTILINE_COMMENT);
+ }
+ break; case '/' :
+ if ((c = read()) == '*')
+ { // MULTINE COMMENT JAVASCRIPT, CSS, PHP
+ if (
+ fContentType == IPHPPartitionScannerConstants.PHP
+ && fOffset - 2 > 0)
+ {
+ fOffset -= 2;
+ IToken token = getToken(fContentType);
+ fContentType =
+ IPHPPartitionScannerConstants
+ .PHP_MULTILINE_COMMENT;
+ return token;
+ }
+ else if (
+ fContentType
+ == IPHPPartitionScannerConstants
+ .PHP_MULTILINE_COMMENT)
+ {
+
+ fTokenOffset = fOffset - 2; }
+
+ }
+ else if (c != ICharacterScanner.EOF)
+ unread(); break; case '*' :
+ if ((c = read()) == '/')
+ {
+ if (fContentType
+ == IPHPPartitionScannerConstants
+ .PHP_MULTILINE_COMMENT)
+ {
+ fContentType =
+ IPHPPartitionScannerConstants.PHP;
+ partitionBorder = true;
+ return getToken(
+ IPHPPartitionScannerConstants
+ .PHP_MULTILINE_COMMENT);
+ }
+ else if (
+ fContentType
+ == IPHPPartitionScannerConstants
+ .CSS_MULTILINE_COMMENT)
+ {
+ }
+ else if (
+ fContentType
+ == IPHPPartitionScannerConstants
+ .JS_MULTILINE_COMMENT)
+ {
+ }
+ }
+ else if (c != ICharacterScanner.EOF)
+ unread(); break; }
+ } // end of file reached but we have to return the
+ // last partition.
+ return getToken(fContentType);
}
-
- }
- 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 true 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;
+ /* (non-Javadoc)
+ * @see org.eclipse.jface.text.rules.ITokenScanner#setRange(org.eclipse.jface.text.IDocument, int, int)
+ */
+ public void setRange(IDocument document, int offset, int length) {
+ if (DEBUG) {
+ System.out.println("SET RANGE: offset=" + offset + " length=" + length);
}
- /*
- * @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;
- }
+ fDocument = document;
+ fOffset = offset;
+ fTokenOffset = offset;
+ fLength = 0;
+ fEnd = fOffset + length;
+ //partitionBorder = false;
+ }
- /*
- * @see IPredicateRule#getSuccessToken()
- * @since 2.0
- */
- public IToken getSuccessToken() {
- return fToken;
+ private int read() {
+ try {
+ if (fOffset < fEnd) {
+ return fDocument.getChar(fOffset++);
+ }
+ return ICharacterScanner.EOF;
+ } catch (BadLocationException e) {
+ // should never happen
+ // TODO write stacktrace to log
+ fOffset = fEnd;
+ return ICharacterScanner.EOF;
}
}
- /**
- * Detector for empty comments.
- */
- static class EmptyCommentDetector implements IWordDetector {
- /* (non-Javadoc)
- * Method declared on IWordDetector
- */
- public boolean isWordStart(char c) {
- return (c == '/');
- }
+ private void unread() {
+ --fOffset;
+ }
- /* (non-Javadoc)
- * Method declared on IWordDetector
- */
- public boolean isWordPart(char c) {
- return (c == '*' || c == '/');
- }
- };
+ private boolean checkPattern(char[] pattern) {
+ return checkPattern(pattern, false);
+ }
/**
- *
+ * Check if next character sequence read from document is equals to
+ * the provided pattern. Pattern is read from left to right until the
+ * first character read doesn't match. If this happens all read characters are
+ * unread.
+ * @param pattern The pattern to check.
+ * @return true if pattern is equals else returns false.
*/
- 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;
+ private boolean checkPattern(char[] pattern, boolean ignoreCase) {
+ int prevOffset = fOffset;
+ for (int i = 0; i < pattern.length; i++) {
+ int c = read();
+
+ if (c == ICharacterScanner.EOF
+ || !letterEquals(c, pattern[i], ignoreCase)) {
+ fOffset = prevOffset;
+ return false;
+ }
}
- };
-
- /**
- * 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("<", "", comment));
- rules.add(new PHPMultiLineRule("", php));
- rules.add(new PHPMultiLineRule("", php));
- rules.add(new PHPMultiLineRule("", php));
- rules.add(new PHPMultiLineRule("", php));
- rules.add(new PHPMultiLineRule("", php));
- rules.add(new PHPMultiLineRule("", php));
+ return true;
+ }
-// rules.add(new HTMLPatternRule(html)); // "<", "