xlawrence 2005/11/09 17:52:21 CET
Modified files:
core/src/java/org/jahia/services/htmlparser
NekoHtmlParser.java
Log:
Cleanup
Revision Changes Path
1.5 +29 -52
jahia/core/src/java/org/jahia/services/htmlparser/NekoHtmlParser.java
http://jahia.mine.nu:8080/cgi-bin/cvsweb.cgi/jahia/core/src/java/org/jahia/services/htmlparser/NekoHtmlParser.java.diff?r1=1.4&r2=1.5&f=h
Index: NekoHtmlParser.java
===================================================================
RCS file:
/home/cvs/repository/jahia/core/src/java/org/jahia/services/htmlparser/NekoHtmlParser.java,v
retrieving revision 1.4
retrieving revision 1.5
diff -u -r1.4 -r1.5
--- NekoHtmlParser.java 9 Nov 2005 11:00:36 -0000 1.4
+++ NekoHtmlParser.java 9 Nov 2005 16:52:21 -0000 1.5
@@ -1,22 +1,20 @@
package org.jahia.services.htmlparser;
-import java.io.ByteArrayInputStream;
-import java.io.ByteArrayOutputStream;
-import java.io.StringReader;
-import java.io.StringWriter;
-import java.util.Vector;
-import javax.xml.transform.OutputKeys;
-import javax.xml.transform.Transformer;
-import javax.xml.transform.TransformerFactory;
-import javax.xml.transform.dom.DOMSource;
-import javax.xml.transform.stream.StreamResult;
import org.apache.xalan.templates.OutputProperties;
import org.cyberneko.html.parsers.DOMParser;
import org.jahia.utils.JahiaTools;
-import org.jahia.utils.fileparsers.CharsetDetection;
import org.w3c.dom.Document;
import org.xml.sax.InputSource;
+import javax.xml.transform.OutputKeys;
+import javax.xml.transform.Transformer;
+import javax.xml.transform.TransformerFactory;
+import javax.xml.transform.dom.DOMSource;
+import javax.xml.transform.stream.StreamResult;
+import java.io.StringReader;
+import java.io.StringWriter;
+import java.util.Vector;
+
/**
*
* <p>Title: Html Parser default implementation based on Neko Html Parser</p>
@@ -56,54 +54,18 @@
* Parses and generates a clean html document, remove unwanted markups,..
* Using default settings
*
- * @param inputString
- * @param DOMVisitors
- * @return
- */
- public String parse(final String inputString, final Vector DOMVisitors){
- return parse(inputString,-1,DOMVisitors);
- }
-
- /**
- * Parses and generates a clean html document, remove unwanted markups,..
- * Using settings as defined for a given site
- *
- * @param inputString
- * @param DOMVisitors
- * @param siteKey
- * @return
- */
- public String parse(final String inputString, final Vector DOMVisitors,
- final int siteId){
- if ( inputString == null || inputString.trim().equals("") ){
- return inputString;
- }
- return parse(inputString,siteId,DOMVisitors);
- }
-
- /**
- * Parses and generates a clean html document, remove unwanted markups,..
- * Using settings as defined for a given site
- *
* @param input
- * @param siteKey
* @param DOMVisitors
- * @return
*/
- public static String parse( final String input,
- final int siteId,
- final Vector DOMVisitors){
-
- if ( input == null || "".equals(input.trim())){
+ public String parse(final String input, final Vector DOMVisitors){
+ if ( input == null || "".equals(input.trim())){
return input;
}
- String result = new String(input);
- result = JahiaTools.replacePattern(result, "&", AMPERSAND);
-
- int size = 0;
+ String result;
+ int size;
try {
- InputSource in = new InputSource(new StringReader(input));
+ final InputSource in = new InputSource(new StringReader(input));
domParser.parse(in);
Document doc = domParser.getDocument();
@@ -138,4 +100,19 @@
}
return result;
}
+
+ /**
+ * Parses and generates a clean html document, remove unwanted markups,..
+ * Using settings as defined for a given site
+ *
+ * @param inputString
+ * @param DOMVisitors
+ */
+ public String parse(final String inputString, final Vector DOMVisitors,
+ final int siteId){
+ if ( inputString == null || inputString.trim().equals("") ){
+ return inputString;
+ }
+ return parse(inputString, DOMVisitors);
+ }
}
