[
https://issues.apache.org/jira/browse/LUCENE-7026?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15147162#comment-15147162
]
Michael McCandless commented on LUCENE-7026:
--------------------------------------------
Are you sharing your own index with the index that {{AnalyzingInfixSuggester}}
creates? I.e. are you indexing your own context fields directly using your own
{{IndexWriter}}?
You are not supposed to do that. Instead, give {{AnalyzingInfixSuggester}} its
own private index and let it do all indexing into that.
> AnalyzingInfixSuggester doesn't support Phrase Queries for contexts
> -------------------------------------------------------------------
>
> Key: LUCENE-7026
> URL: https://issues.apache.org/jira/browse/LUCENE-7026
> Project: Lucene - Core
> Issue Type: Bug
> Components: core/index
> Affects Versions: 5.3.1
> Reporter: Kamaldeep Singh
>
> AnalyzingInfixSuggester uses Term Queries for filtering out the contexts. If
> the StandardAnalyzer is used while indexing , since all the contexts are
> stored as StringField and analyzed , spaces in them are removed and words
> converted to tokens. So contexts having spaces in their terms can't be
> queried , And we also can't pass a PhraseQuery as a BooleanQuery to lookup()
> since Phrase Queries won't work on string fields where positions of tokens
> are not stored.
> Option left is to use a different analyzer while indexing
> Analyzing Infix Suggester should store contexts as TextField instead of
> StringField so they can be queried.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
