[
https://issues.apache.org/jira/browse/SOLR-1279?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12731125#action_12731125
]
Robert Muir commented on SOLR-1279:
-----------------------------------
Sergey, have you looked at SOLR-1266?
By using the new stemEnglishPossessive=0 option, I think you can get the same
behavior with WordDelimiterFilter, if you use preserveOriginal=1 along with
catenateWords=1
> ApostropheTokenizer
> -------------------
>
> Key: SOLR-1279
> URL: https://issues.apache.org/jira/browse/SOLR-1279
> Project: Solr
> Issue Type: New Feature
> Components: Analysis
> Reporter: Sergey Borisov
> Priority: Minor
> Fix For: 1.4
>
> Attachments: ApostropheTokenizer.zip
>
>
> ApostropheTokenizer creates extra tokens during the analysis stage for the
> fields containing apostrophes. The reason for adding this is to ensure that
> documents that differ only by apostrophe have the same relevancy score.
> For example, if the document contains string "McDonald's", it will be
> tokenized as "McDonald's McDonalds". This way when the search is performed
> against "McDonald's" or "McDonalds" will produce similar score.
> This code handles up to two apostrophes in a token.
> To use this tokenizer add the following line in schema.xml
> <analyzer type="index">
> <filter class="org.apache.lucene.analysis.ApostropheTokenFactory"/>
> ...
> </analyzer>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
