-/**
- * 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: 1.18 $
- */
+/**********************************************************************
+ 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 1.29 2005-05-15 23:24:41 axelcl Exp $
+ **********************************************************************/
package net.sourceforge.phpeclipse.phpeditor.php;
-import java.util.*;
+import java.util.HashMap;
+import java.util.Map;
+
+import net.sourceforge.phpeclipse.ui.text.rules.AbstractPartitioner;
-import org.eclipse.jface.text.*;
-import org.eclipse.jface.text.rules.*;
+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.IPartitionTokenScanner;
+import org.eclipse.jface.text.rules.IToken;
+import org.eclipse.jface.text.rules.Token;
/**
*
+ *
+ * @author Igor Malinin
*/
-public class PHPPartitionScanner implements IPartitionTokenScanner
-{
- private static final boolean DEBUG = false;
- private boolean fInString = false;
- private boolean fInDoubString = false;
- private IDocument fDocument = null;
- private int fOffset = -1;
- private String fContentType = IPHPPartitionScannerConstants.HTML;
- private String fPrevContentType = IPHPPartitionScannerConstants.HTML;
- private boolean partitionBorder = false;
- private int fTokenOffset;
- private int fEnd = -1;
- private int fLength;
- private int fCurrentLength;
- private Map tokens = new HashMap();
-
- 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 class PHPPartitionScanner implements IPartitionTokenScanner {
+ public static final String PHP_SCRIPTING_AREA = "__php_scripting_area ";
- private IToken getToken(String type)
- {
- fLength = fCurrentLength;
- if (DEBUG)
- {
-
- try
- {
- if (fLength <= 0)
- {
- int line = fDocument.getLineOfOffset(fOffset);
- System.err.println(
- "Error at "
- + line
- + " offset:"
- + String.valueOf(
- fOffset - fDocument.getLineOffset(line)));
- }
- }
- catch (BadLocationException e)
- { // should never happen
- // TODO Write stacktrace to log
- e.printStackTrace();
- }
- }
- Assert.isTrue(fLength > 0, "Partition length <= 0!");
- fCurrentLength = 0;
- // String can never cross partition borders so reset string detection
- fInString = false;
- fInDoubString = false;
- 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="
- + type
- + " fLength="
- + fLength);
- }
- return token;
- }
+ public static final int STATE_DEFAULT = 0;
- /* (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)
- {
- if (DEBUG)
- {
- System.out.println(
- "PartialRange: contentType="
- + contentType
- + " partitionOffset="
- + partitionOffset);
- }
+ // public static final int STATE_TAG = 1;
+ // public static final int STATE_SCRIPT = 2;
- try
- {
- if (partitionOffset > -1)
- {
- partitionBorder = false;
- // because of strings we have to parse the whole partition
- this.setRange(
- document,
- partitionOffset,
- offset - partitionOffset + length);
- // sometimes we get a wrong partition so we retrieve the partition
- // directly from the document
- fContentType = fDocument.getContentType(partitionOffset);
- }
- else
- this.setRange(document, offset, length);
+ private IDocument document;
- }
- catch (BadLocationException e)
- {
- // should never happen
- // TODO print stack trace to log
- // fall back just scan the whole document again
- this.setRange(document, 0, fDocument.getLength());
- }
+ private int begin;
- }
+ private int end;
- /* (non-Javadoc)
- * @see org.eclipse.jface.text.rules.ITokenScanner#getTokenLength()
- */
- public int getTokenLength()
- {
- return fLength;
- }
+ private int offset;
- /* (non-Javadoc)
- * @see org.eclipse.jface.text.rules.ITokenScanner#getTokenOffset()
- */
- public int getTokenOffset()
- {
- return fTokenOffset;
- }
+ private int length;
+
+ private int position;
+
+ private int state;
+
+ private Map tokens = new HashMap();
+
+ public PHPPartitionScanner() {
+ }
- /* (non-Javadoc)
- * @see org.eclipse.jface.text.rules.ITokenScanner#nextToken()
+ /*
+ * @see org.eclipse.jface.text.rules.ITokenScanner#nextToken()
+ */
+ public IToken nextToken() {
+ offset += length;
+
+ /*
+ * switch (state) { case STATE_TAG: return nextTagToken(); }
*/
- 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();
- if (partitionBorder)
- {
- fTokenOffset = fOffset;
- partitionBorder = false;
- }
+ switch (read()) {
+ case ICharacterScanner.EOF:
+ state = STATE_DEFAULT;
+ return getToken(null);
+
+ case '<':
+ switch (read()) {
+ case ICharacterScanner.EOF:
+ state = STATE_DEFAULT;
+ return getToken(null);
- while ((c = read()) != ICharacterScanner.EOF)
- {
- switch (c)
- {
- case '<' :
- if (!isInString(IPHPPartitionScannerConstants.PHP)
- && fContentType
- != IPHPPartitionScannerConstants.PHP_MULTILINE_COMMENT
- && checkPattern(new char[] { '?', 'p', 'h', 'p' }, true))
- {
- if (fContentType != IPHPPartitionScannerConstants.PHP
- && fCurrentLength > 5)
- {
- unread(5);
- IToken token = getToken(fContentType);
- // save previouse contenttype
- //TODO build stack for previouse contenttype
- fPrevContentType = fContentType;
-
- fContentType = IPHPPartitionScannerConstants.PHP;
-
- return token;
- }
- else
- fContentType = IPHPPartitionScannerConstants.PHP;
-
- // remember offset of this partition
- fTokenOffset = fOffset - 5;
- fCurrentLength = 5;
- }
- else if (
- !isInString(IPHPPartitionScannerConstants.PHP)
- && fContentType
- != IPHPPartitionScannerConstants
- .PHP_MULTILINE_COMMENT
- && checkPattern(new char[] { '?' }, false))
- {
- if (fContentType != IPHPPartitionScannerConstants.PHP
- && fCurrentLength > 2)
- {
- unread(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;
- fCurrentLength = 2;
- }
- else if (
- !isInString(IPHPPartitionScannerConstants.PHP)
- && checkPattern(new char[] { '!', '-', '-' }))
- { // return previouse partition
- if (fContentType
- != IPHPPartitionScannerConstants
- .HTML_MULTILINE_COMMENT
- && fCurrentLength > 4)
- {
- unread(4);
- IToken token = getToken(fContentType);
- fContentType =
- IPHPPartitionScannerConstants
- .HTML_MULTILINE_COMMENT;
- return token;
- }
- else
- fContentType =
- IPHPPartitionScannerConstants
- .HTML_MULTILINE_COMMENT;
-
- fTokenOffset = fOffset - 4;
- fCurrentLength = 4;
- }
- break;
- case '?' :
- if (!isInString(IPHPPartitionScannerConstants.PHP)
- && fContentType == IPHPPartitionScannerConstants.PHP)
- {
- if ((c = read()) == '>')
- {
- if (fPrevContentType != null)
- fContentType = fPrevContentType;
- else
- fContentType =
- IPHPPartitionScannerConstants.HTML;
- partitionBorder = true;
- return getToken(IPHPPartitionScannerConstants.PHP);
- }
- else if (c != ICharacterScanner.EOF)
- unread();
- }
- break;
- case '-' :
- if (!isInString(IPHPPartitionScannerConstants.PHP)
- && fContentType
- == IPHPPartitionScannerConstants
- .HTML_MULTILINE_COMMENT
- && checkPattern(new char[] { '-', '>' }))
- {
- fContentType = IPHPPartitionScannerConstants.HTML;
- partitionBorder = true;
- return getToken(
- IPHPPartitionScannerConstants
- .HTML_MULTILINE_COMMENT);
- }
- break;
- case '/' :
- if (!isInString(IPHPPartitionScannerConstants.PHP) && (c = read()) == '*')
- { // MULTINE COMMENT JAVASCRIPT, CSS, PHP
- if (fContentType == IPHPPartitionScannerConstants.PHP
- && fCurrentLength > 2)
- {
- unread(2);
- IToken token = getToken(fContentType);
- fContentType =
- IPHPPartitionScannerConstants
- .PHP_MULTILINE_COMMENT;
- return token;
- }
- else if (
- fContentType
- == IPHPPartitionScannerConstants
- .PHP_MULTILINE_COMMENT)
- {
-
- fTokenOffset = fOffset - 2;
- fCurrentLength = 2;
- }
-
- }
- else if (!isInString(IPHPPartitionScannerConstants.PHP) && c != ICharacterScanner.EOF)
- unread();
- break;
- case '*' :
- if (!isInString(IPHPPartitionScannerConstants.PHP) && (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 (!isInString(IPHPPartitionScannerConstants.PHP) && c != ICharacterScanner.EOF)
- unread();
- break;
- case '\'' :
- if (!fInDoubString)
- fInString = !fInString;
- break;
- case '"' :
- // toggle String mode
- if (!fInString)
- fInDoubString = !fInDoubString;
- break;
- }
- } // end of file reached but we have to return the
- // last partition.
- return getToken(fContentType);
+ case '?': // <?
+// int ch = read();
+//
+// switch (ch) {
+// case ICharacterScanner.EOF:
+// state = STATE_DEFAULT;
+// return getToken(PHP_SCRIPTING_AREA);
+// }
+ return scanUntilPHPEndToken(PHP_SCRIPTING_AREA);
+ }
+
+ unread();
}
- /* (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);
+
+ 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;
}
- fDocument = document;
- fOffset = offset;
- fTokenOffset = offset;
- fCurrentLength = 0;
- fLength = 0;
- fEnd = fOffset + length;
- fInString = false;
- fInDoubString = false;
- fContentType = IPHPPartitionScannerConstants.HTML;
-// String[] prev = getPartitionStack(offset);
+ unread();
+
+ state = STATE_DEFAULT;
+ return getToken(null);
+ }
}
+ }
- private int read()
- {
- try
- {
- if (fOffset < fEnd)
- {
- fCurrentLength++;
- return fDocument.getChar(fOffset++);
- }
- return ICharacterScanner.EOF;
+ 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 '\'': // single quoted string
+ // read until end of single quoted string
+ if (!readUntilEscapedSQ()) {
+ state = STATE_DEFAULT;
+ return getToken(token);
}
- catch (BadLocationException e)
- {
- // should never happen
- // TODO write stacktrace to log
- fOffset = fEnd;
- return ICharacterScanner.EOF;
+ 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;
- private void unread()
- {
- --fOffset;
- --fCurrentLength;
+ 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();
+// }
+// }
- private void unread(int num)
- {
- fOffset -= num;
- fCurrentLength -= num;
+ if (type == null) {
+ return Token.UNDEFINED;
}
- private boolean checkPattern(char[] pattern)
- {
- return checkPattern(pattern, false);
+ IToken token = (IToken) tokens.get(type);
+ if (token == null) {
+ token = new Token(type);
+ tokens.put(type, token);
}
- /**
- * 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 <code>true</code> if pattern is equals else returns <code>false</code>.
- */
- private boolean checkPattern(char[] pattern, boolean ignoreCase)
- {
- int prevOffset = fOffset;
- int prevLength = fCurrentLength;
- for (int i = 0; i < pattern.length; i++)
- {
- int c = read();
-
- if (c == ICharacterScanner.EOF
- || !letterEquals(c, pattern[i], ignoreCase))
- {
- fOffset = prevOffset;
- fCurrentLength = prevLength;
- return false;
- }
- }
+ return token;
+ }
- return true;
+ private int read() {
+ if (position >= end) {
+ return ICharacterScanner.EOF;
}
- private boolean letterEquals(int test, char letter, boolean ignoreCase)
- {
- if (test == letter)
- return true;
- else if (
- ignoreCase
- && Character.isLowerCase(letter)
- && test == Character.toUpperCase(letter))
- return true;
- else if (
- ignoreCase
- && Character.isUpperCase(letter)
- && test == Character.toLowerCase(letter))
- return true;
+ try {
+ return document.getChar(position++);
+ } catch (BadLocationException e) {
+ --position;
+ return ICharacterScanner.EOF;
+ }
+ }
- return false;
+ 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;
}
-
- /**
- * Checks wether the offset is in a <code>String</code> and the specified
- * contenttype is the current content type.
- * Strings are delimited, mutual exclusive, by a " or by a '.
- *
- * @param contentType The contenttype to check.
- * @return <code>true</code> if the current offset is in a string else
- * returns false.
- */
- private boolean isInString(String contentType)
- {
- if(fContentType == contentType)
- return (fInString || fInDoubString);
- else
- return false;
+ 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;
}
-
- /**
- * Returns the previouse partition stack for the given offset.
- *
- * @param offset The offset to return the previouse partitionstack for.
- *
- * @return The stack as a string array.
- */
- private String[] getPartitionStack(int offset)
- {
- ArrayList types = new ArrayList();
- int tmpOffset = 0;
- try
- {
- ITypedRegion region = fDocument.getPartition(offset);
- tmpOffset = region.getOffset();
- while(tmpOffset-1 > 0)
- {
- region = fDocument.getPartition(tmpOffset-1);
- tmpOffset = region.getOffset();
- types.add(0, region.getType());
- }
+ return false;
+ }
+
+ private boolean readSingleLine() {
+ try {
+ do {
+ if (position >= end) {
+ return false;
}
- catch (BadLocationException e)
- {
- if(DEBUG)
- {
- e.printStackTrace();
- }
+ } 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;
}
-
- String[] retVal = new String[types.size()];
-
- retVal = (String[])types.toArray(retVal);
- return retVal;
+ 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