Mailing-List: contact dev-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@lucene.apache.org
Message-ID: <17497366.28721289492714031.JavaMail.jira@thor>
Date: Thu, 11 Nov 2010 11:25:14 -0500 (EST)
From: "Robert Muir (JIRA)" <jira@apache.org>
To: dev@lucene.apache.org
Subject: [jira] Commented: (LUCENE-2754) add spanquery support for all
 multitermqueries
In-Reply-To: <7553028.25941289483173470.JavaMail.jira@thor>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/LUCENE-2754?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12931070#action_12931070 ] 

Robert Muir commented on LUCENE-2754:
-------------------------------------

bq. What about the third BQ rewrite mode (is it needed or is scoring different (we have two rerwite modes for fuzzy)?

In this case, i thought not because it would be constant score queries, not spanqueries.
I think a constant score spanquery would be useless :)

{quote}
So the rewrite methods can only be used for real SpanQueries, how about moving those rewrite methods out of MTQ? I know, then we must make some classes more public... But I was already thinking about the code mess in MTQ. Maybe we should remove those modes and put into separate classes at all.
{quote}

I agree completely... its tricky to scroll through the massive .java file and find what we want. 
I would also prefer in this case for the Span-rewrite methods to be in the actual o.a.l.search.span package...


> add spanquery support for all multitermqueries
> ----------------------------------------------
>
>                 Key: LUCENE-2754
>                 URL: https://issues.apache.org/jira/browse/LUCENE-2754
>             Project: Lucene - Java
>          Issue Type: New Feature
>          Components: Search
>            Reporter: Robert Muir
>             Fix For: 4.0
>
>         Attachments: LUCENE-2754.patch
>
>
> I set fix version: 4.0, but possibly we could do this for 3.x too
> Currently, we have a special SpanRegexQuery in contrib, and issues like LUCENE-522 open for SpanFuzzyQuery.
> The SpanRegexQuery in contrib is a little messy additionally.
> For any arbitrary MultiTermQueries to work as a SpanQuery, there are only 3 requirements:
> # The un-rewritten query must extend SpanQuery so it can be included in Span clauses
> # The rewritten query should be SpanOrQuery instead of BooleanQuery
> # The rewritten term clauses should be SpanTermQueries.
> Instead of having logic like this for each query, i suggest adding two rewrite methods:
> * ScoringSpanBoolean rewrite
> * TopTermsSpanBoolean rewrite
> as a start i wrote these up, and added a SpanMultiTermQueryWrapper that can be used to wrap any multitermquery this way.
> there are a few kinks, but I think the MTQ policeman can probably help get through them.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org