From: gn64_jp <gn64@rec10.org>
Date: Sun, 17 Nov 2013 22:42:49 +0000 (+0900)
Subject: solr : Use JapaneseTokenizerFactory(Kuromoji) instead of CJK-Bigram Filter.
X-Git-Url: http://git.osdn.net/view?p=libre10%2Flibre10.git;a=commitdiff_plain;h=9a509ba3cde28f69836a6f2a03811d6ece58505e

solr : Use JapaneseTokenizerFactory(Kuromoji) instead of CJK-Bigram Filter.
---

diff --git a/solr-4.5.1/libre10/solr/libre10/conf/schema.xml b/solr-4.5.1/libre10/solr/libre10/conf/schema.xml
index dfa5961..c7b26c4 100644
--- a/solr-4.5.1/libre10/solr/libre10/conf/schema.xml
+++ b/solr-4.5.1/libre10/solr/libre10/conf/schema.xml
@@ -16,6 +16,17 @@
 			<filter class="solr.NGramFilterFactory" minGramSize="2" maxGramSize="15"/>-->
 		</analyzer>
 	</fieldType>
+    <fieldType name="text_ja" class="solr.TextField">
+        <analyzer>
+            <tokenizer class="solr.JapaneseTokenizerFactory" mode="extended"/>
+            <filter class="solr.JapaneseBaseFormFilterFactory"/>
+            <filter class="solr.JapanesePartOfSpeechStopFilterFactory" tags="lang/stoptags_ja.txt" enablePositionIncrements="true"/>
+            <filter class="solr.CJKWidthFilterFactory" />
+            <filter class="solr.StopFilterFactory" ignoreCase="true" words="lang/stopwords_ja.txt" enablePositionIncrements="true" />
+            <filter class="solr.JapaneseKatakanaStemFilterFactory" minimumLength="2"/>
+            <filter class="solr.LowerCaseFilterFactory"/>
+		</analyzer>
+    </fieldType>
 	<fieldType name="long" class="solr.LongField" omitNorms="true"/>
 	<fieldType name="uuid" class="solr.UUIDField" indexed="true" />
 
@@ -28,7 +39,7 @@
 <field name="title_group_id" type="string" indexed="true" stored="true" required="true" />
 <field name="page" type="long" indexed="true" stored="true" required="true" />
 <field name="pagemax" type="long" indexed="true" stored="true" required="true" />
-<field name="text" type="text_cjk" indexed="true" stored="true" required="true" termVectors="true" termPositions="true" termOffsets="true" />
+<field name="text" type="text_ja" indexed="true" stored="true" required="true" termVectors="true" termPositions="true" termOffsets="true" />
 <!--<field name="text" type="text_cjk" indexed="true" stored="true" required="true" />-->
 <field name="genre" type="string" indexed="true" stored="true" required="false" multiValued="true" />
 <field name="path_id" type="string" indexed="false" stored="true" required="false" />