[
https://issues.apache.org/jira/browse/LUCENE-6093?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14299732#comment-14299732
]
ASF subversion and git services commented on LUCENE-6093:
---------------------------------------------------------
Commit 1656173 from [~mikemccand] in branch 'dev/trunk'
[ https://svn.apache.org/r1656173 ]
LUCENE-6093: don't throw NPE when BlendedInfixSuggester.lookup doesn't have a
prefix token
> BlendedInfixSuggester throws NullPointerException if there were discarded
> trailing characters in the query
> ----------------------------------------------------------------------------------------------------------
>
> Key: LUCENE-6093
> URL: https://issues.apache.org/jira/browse/LUCENE-6093
> Project: Lucene - Core
> Issue Type: Bug
> Components: modules/spellchecker
> Affects Versions: 4.10.2
> Environment: Any environment
> Reporter: Andrei Filimonov
> Priority: Critical
> Fix For: 5.0, Trunk, 5.1
>
> Attachments: LUCENE-6093.patch
>
>
> BlendedInfixSuggester throws NullPointerException if there were discarded
> trailing characters (e.g. whitespace or special character) in the query.
> The problem seems to be in the createCoefficient method that fails to check
> if prefixToken parameter is null. AnalyzingInfixSuggester sets prefixToken to
> null in the described case and passes it to BlendedInfixSuggester. On the
> side not even if BlendedInfixSuggester is changed to handle this creates a
> problem to calculate the weights as prefixToken is null and cannot be used. I
> would be better to have AnalyzingInfixSuggester to always set prefixToken to
> lastToken.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
