NPE when SynonymFilter use tokenizerFactory and no-dictionary jar

[GoogleCodeExporter]

NPE occur that no-dic.jar use dictionaryDir if SynonymFilter speccified 
tokenizerFactory attr.

It is because a cause of NPE is not called JapaneseTokenizerFactory.inform 
method.
The inform method set a dictionaryDir. 
no-dic.jar does not include dictionary files.

Example schema.xml setting:
--
    <fieldType name="text_ja" class="solr.TextField" positionIncrementGap="100" autoGeneratePhraseQueries="false">
      <analyzer>
        <tokenizer class="solr.JapaneseTokenizerFactory" compositePOS="compositePOS.txt" dictionaryDir="dictionary/naist-chasen"/>

      <filter class="solr.SynonymFilterFactory" synonyms="synonyms.txt" ignoreCase="true" expand="true" tokenizerFactory="solr.JapaneseTokenizerFactory" compositePOS="compositePOS.txt" dictionaryDir="dictionary/naist-chasen"/>
      </analyzer>
    </fieldType>

Original issue reported on code.google.com by johtani on 16 Feb 2012 at 12:07

[GoogleCodeExporter]

I find issue in solr.
https://issues.apache.org/jira/browse/SOLR-2909

I think that corresponding by issue of Solr is probably right. 
However, I am thinking now because I want to offer a solution in a user of 
lucene-gosen.

Original comment by johtani on 20 Feb 2012 at 6:26

[GoogleCodeExporter]

Solr patch was released.

Original comment by johtani on 22 Feb 2012 at 8:37

• Added labels: Priority-Medium
• Removed labels: Priority-High

[GoogleCodeExporter]

use solr > 3.6.0 . 

Original comment by johtani on 29 May 2012 at 7:45

[GoogleCodeExporter]

Original comment by johtani on 4 Jul 2012 at 12:35

• Changed state: WontFix