From 10951a717f43cbcc8462b5b9d851d2bd2b78ed24 Mon Sep 17 00:00:00 2001
From: eg3p <eg3p>
Date: Sun, 24 Jun 2007 18:28:47 +0000
Subject: [PATCH] improved indexing

---
 archive/copy-to-solr-webapp/Readme.txt        | 21 ++++++++++++-------
 archive/copy-to-solr-webapp/schema.xml        |  3 ++-
 archive/styles/mergeMetadataAndData.xsl       |  2 +-
 ...arizeTranscript.xsl => solarizeTitles.xsl} | 10 ++++++---
 lucene-thdl-build.xml                         |  6 +++++-
 5 files changed, 28 insertions(+), 14 deletions(-)
 rename archive/styles/{solarizeTranscript.xsl => solarizeTitles.xsl} (90%)

diff --git a/archive/copy-to-solr-webapp/Readme.txt b/archive/copy-to-solr-webapp/Readme.txt
index 2307f3c..db40fad 100644
--- a/archive/copy-to-solr-webapp/Readme.txt
+++ b/archive/copy-to-solr-webapp/Readme.txt
@@ -2,14 +2,19 @@ The files in this directory must be copied to your solr webapp
 before you can use lucene-thdl-build.xml to post, commit, or 
 delete documents from your solr server.
 
-schema.xml & solrconfig.xml: 
-    Copy these files to your solr/conf directory,
-    replacing the existing files with these names
+First, run the task solr-prepare-for-copy-to-solr-webapp.
 
- 
-lucene-thdl.jar :
-    Create an up to date copy of this file by
-    running the task lucene-thdl-jar, then copy
-    to your solr/lib directory. If solr/lib does
+Then, copy files to SOLR as follows:
+
+*.xml: 
+    Copy to your solr/conf directory, replacing any
+    existing files..
+    
+*.xsl
+    Copy to the solr/conf/xslt directory.
+
+*.jar
+    Copy to your solr/lib directory. If solr/lib does
     not exist, then create it.
 
+
diff --git a/archive/copy-to-solr-webapp/schema.xml b/archive/copy-to-solr-webapp/schema.xml
index f498187..c848a22 100644
--- a/archive/copy-to-solr-webapp/schema.xml
+++ b/archive/copy-to-solr-webapp/schema.xml
@@ -188,6 +188,7 @@
    
    <!-- transcript documents -->
    <field name="id" type="string" indexed="true" stored="true"/>
+   <field name="numericRef" type="string" indexed="true" stored="false"/>
    <field name="form_bo" type="text_tibetan" indexed="true" stored="true"/> 
 
    <!-- Dynamic field definitions.  If a field name is not found, dynamicFields
@@ -244,6 +245,6 @@
       A custom similarity may be specified here, but the default is fine
       for most applications.  -->
  <!-- <similarity class="org.apache.lucene.search.DefaultSimilarity"/> -->
-
+ 
 </schema>
 
diff --git a/archive/styles/mergeMetadataAndData.xsl b/archive/styles/mergeMetadataAndData.xsl
index 78de4c4..3b403df 100644
--- a/archive/styles/mergeMetadataAndData.xsl
+++ b/archive/styles/mergeMetadataAndData.xsl
@@ -17,7 +17,7 @@
 	<xsl:template match="/">
         <xsl:for-each select="//transcript">
             <xsl:variable name="filename" select="."/>
-            <xsl:result-document href="Transcript_{../@id}.xml" format="unicode.transcript.with.metadata">
+            <xsl:result-document href="{../@id}.xml" format="unicode.transcript.with.metadata">
                 <xsl:element name="TITLE">
                     <xsl:attribute name="id">
                         <xsl:value-of select="../@id"/>
diff --git a/archive/styles/solarizeTranscript.xsl b/archive/styles/solarizeTitles.xsl
similarity index 90%
rename from archive/styles/solarizeTranscript.xsl
rename to archive/styles/solarizeTitles.xsl
index 6c45ce3..1d76c93 100644
--- a/archive/styles/solarizeTranscript.xsl
+++ b/archive/styles/solarizeTitles.xsl
@@ -28,7 +28,8 @@
     <xsl:template match="METADATA">
         <xsl:param name="title.id" select="''"/>
         <doc>
-            <field name="id"><xsl:value-of select="$title.id"/></field>
+            <field name="id"><xsl:value-of select="$title.id"/></field> <!-- has 't' prefix -->
+            <field name="numericRef"><xsl:value-of select="substring($title.id,2)"/></field> <!-- strips prefix for searching -->
             <field name="thdlType_s"><xsl:value-of select="$TITLE_TYPE"/></field>
             <field name="belongsTo_idlist"><xsl:value-of select="belongsTo"/></field>
             <field name="speechType_s"><xsl:value-of select="speechType"/></field>
@@ -49,7 +50,8 @@
         </doc>
         <xsl:for-each select="video">
             <doc>
-                <field name="id"><xsl:value-of select="concat(@id)"/></field>
+                <field name="id"><xsl:value-of select="@id"/></field> <!-- has 'v' prefix -->
+                <field name="numericRef"><xsl:value-of select="substring(@id,2)"/></field> <!-- strips prefix for searching -->
                 <field name="title_idref"><xsl:value-of select="$title.id"/></field>
                 <field name="thdlType_s"><xsl:value-of select="$VIDEO_TYPE"/></field>
                 <field name="mediaType_s"><xsl:value-of select="mediaDescription"/></field>
@@ -103,8 +105,10 @@
     <xsl:template match="S">
         <xsl:param name="title.id" select="''"/>
         <xsl:param name="belongs.to" select="''"/>
+        <xsl:variable name="s.id" select="concat($title.id, '_', @id)"/>
         <doc>
-            <field name="id"><xsl:value-of select="concat($title.id, '_', @id)"/></field>
+            <field name="id"><xsl:value-of select="$s.id"/></field> <!-- has 't' prefix -->
+            <field name="numericRef"><xsl:value-of select="substring($s.id,2)"/></field> <!-- strips prefix for searching -->
             <field name="title_idref"><xsl:value-of select="$title.id"/></field>
             <field name="thdlType_s"><xsl:value-of select="$TRANSCRIPT_FRAGMENT_TYPE"/></field>
             <field name="belongsTo_idlist"><xsl:value-of select="$belongs.to"/></field>
diff --git a/lucene-thdl-build.xml b/lucene-thdl-build.xml
index 8a2a6f9..4912f3f 100644
--- a/lucene-thdl-build.xml
+++ b/lucene-thdl-build.xml
@@ -100,7 +100,7 @@
     <!-- solr tasks -->
     <target name="solr-prepare-titles">
         <mkdir dir="${solr.titles}"/>
-        <xslt basedir="${transcripts}" includes="*.xml" destdir="${solr.titles}" extension=".xml" style="${styles}/solarizeTranscript.xsl"/>
+        <xslt basedir="${transcripts}" includes="*.xml" destdir="${solr.titles}" extension=".xml" style="${styles}/solarizeTitles.xsl"/>
     </target>
     
     <target name="solr-prepare-nontitles">
@@ -141,6 +141,10 @@
             <jvmarg value="-Dcommit=yes"/>
 		</java>
     </target>
+    
+    <target name="solr-prepare-for-copy-to-solr-webapp" depends="lucene-thdl-jar">
+        <!--<copy file="${styles}/solarizeConstantsForImport.xsl" todir="${copy.to.solr.webapp}"/>-->
+    </target>
         
         <!-- insert whether or not media exists for segment -->
   <!--      <java classname="net.sf.saxon.Transform" fork="yes">