[
https://issues.apache.org/jira/browse/SOLR-2759?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13103650#comment-13103650
]
Chris Male commented on SOLR-2759:
----------------------------------
Actually, we only need the 1 Map which will contain the no-stopword filtering
Analyzers, since we can just retrieve the other directly from the Schema (since
it doesn't need to be altered).
> Remove the need for EDismax's ExtendedAnalyzer
> ----------------------------------------------
>
> Key: SOLR-2759
> URL: https://issues.apache.org/jira/browse/SOLR-2759
> Project: Solr
> Issue Type: Improvement
> Components: search
> Reporter: Chris Male
>
> ExtendedDisMaxQParserPlugin.ExtendedAnalyzer cannot be ported over to being
> reusable due to its support for deciding at runtime whether stopwords should
> be filtered or not.
> One way to resolve this is to maintain 2 lazy-initialized Map<String,
> Analyzer> in the ExtendedSolrQueryParser, one with the stopword filtering
> Analyzers for fields, the other with no-stopword filtering Analyzers. Then
> ExtendedSolrQueryParser can override the
> QueryParserBase.newFieldQuery(Analyzer, String, String, boolean) and
> substitute in the appropriate Analyzer based on the runtime behavior desired
> at that time.
> This will mean an ever so slight increase in memory consumption due to the 2
> maps, but will allow for reusability which will reduce memory consumption and
> increase query throughput.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
