[
https://issues.apache.org/jira/browse/LUCENE-7836?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16015561#comment-16015561
]
Michael McCandless commented on LUCENE-7836:
--------------------------------------------
I suspect the problem is "Hello World" was indexed as 2 tokens, since you're
indexing with whitespace tokenization, which means the regexp will never match
a single token.
If you indexed "Hello World" as a single token then the regexp should match.
> Multiple Token Regex Search Not working
> ---------------------------------------
>
> Key: LUCENE-7836
> URL: https://issues.apache.org/jira/browse/LUCENE-7836
> Project: Lucene - Core
> Issue Type: Bug
> Environment: Linux(Centos)
> Reporter: Parmeet Singh Sachdeva
> Priority: Blocker
>
> I am able to search for a regex query like H[a-z]llo but I am not able to
> search for a regex query like H[a-z]llo Wor[a-z]d even though I have "Hello
> World" in my source tree. I am not able to search multi-word regex queries.
> I am using OpenGrok which is, in turn, using
> org.apache.lucene.search.RegexpQuery class of Lucene 6.5.0 which extends
> AutomatonQuery.
> I have indexed the data based on whitespace tokenization.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
