From 00b73276cdfa385732f9a54861f6a40c0395b4d4 Mon Sep 17 00:00:00 2001
From: Kasun Gajasinghe <kasunbg@gmail.com>
Date: Mon, 3 Oct 2011 19:13:27 +0000
Subject: [PATCH] Webhelp ant script changes - HTML transformation support for
 WebHelp - Uses Tagsoup for parsing the bad html. tagsoup-1.2.1.jar is added
 to trunk/xsl-webhelpindexer/lib/

---
 xsl/webhelp/build.xml       | 17 ++++++++++++++---
 xsl/webhelp/xsl/webhelp.xsl |  2 --
 2 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/xsl/webhelp/build.xml b/xsl/webhelp/build.xml
index 502af2173..7b0990a30 100644
--- a/xsl/webhelp/build.xml
+++ b/xsl/webhelp/build.xml
@@ -10,6 +10,7 @@
         <pathelement location="${extensions.dir}/webhelpindexer.jar"/>
 	<pathelement location="${extensions.dir}/lucene-analyzers-3.0.0.jar"/>
 	<pathelement location="${extensions.dir}/lucene-core-3.0.0.jar"/>
+	<pathelement location="${extensions.dir}/tagsoup-1.2.1.jar"/>
     </path>
 
   <condition property="perform-validation-dtd">
@@ -87,13 +88,25 @@
 
 	<echo>Indexing html files in ${output-dir}/content</echo>
 
-       <java classname="com.nexwave.nquindexer.IndexerMain"> 
+       <java classname="com.nexwave.nquindexer.IndexerMain" fork="true"> 
          <sysproperty key="htmlDir" value="${output-dir}/content"/>
          <sysproperty key="indexerLanguage" value="${webhelp.indexer.language}"/>
          <sysproperty key="htmlExtension" value="${html.extension}"/>
          <sysproperty key="doStem" value="${enable.stemming}"/>
          <sysproperty key="tocFile" value="${toc.file}"/>
 
+	 <!--TagSoup SAX Parser for parsing even the bad html contents. see 
+	      http://sourceforge.net/tracker/?func=detail&aid=3401185&group_id=21935&atid=373750-->
+         <sysproperty key="org.xml.sax.driver" value="org.ccil.cowan.tagsoup.Parser"/>
+         <sysproperty key="javax.xml.parsers.SAXParserFactory" value="org.ccil.cowan.tagsoup.jaxp.SAXFactoryImpl"/>
+
+         <!-- Uncomment the following if Xerces is your preference as the SAX XML Parser. Note that the indexing will fail with Xerces
+	      if the html files are not XML-conformance -->
+  	 <!--sysproperty key="org.xml.sax.driver" value="org.apache.xerces.parsers.SAXParser"/>	      
+         <sysproperty key="javax.xml.parsers.SAXParserFactory" value="org.apache.xerces.jaxp.SAXParserFactoryImpl"/-->
+
+         <!-- Debug the indexer on port 5005 via remote-debug -->
+         <!--jvmarg line="-Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=y,address=5005"/-->
          <classpath>
 	    <path refid="classpath"/>
 	    <pathelement location="/usr/share/java/xercesImpl.jar"/>  	
@@ -104,8 +117,6 @@
 	  </classpath>
        </java>	
  
-	<!--indexertask htmldir="${output-dir}/content" indexerLanguage="${webhelp.indexer.language}"/-->
-
 	<delete>
 	  <fileset dir="${output-dir}/content/search" includes="*.props"/>
 	</delete>
diff --git a/xsl/webhelp/xsl/webhelp.xsl b/xsl/webhelp/xsl/webhelp.xsl
index 6df507082..acdf3830c 100644
--- a/xsl/webhelp/xsl/webhelp.xsl
+++ b/xsl/webhelp/xsl/webhelp.xsl
@@ -8,7 +8,6 @@
 
     <xsl:import href="../../xhtml/chunk.xsl"/>
 
-
     <xsl:output
             method="html"
             encoding="utf-8"
@@ -20,7 +19,6 @@
             <xsl:otherwise>html</xsl:otherwise>
         </xsl:choose>
     </xsl:param>
-    
 
     <!-- webhelp-specific params! -->
     <!-- To be added to normal params file someday -->
-- 
2.40.0