[
https://issues.apache.org/jira/browse/LUCENE-6083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14233428#comment-14233428
]
Paul Elschot commented on LUCENE-6083:
--------------------------------------
Jason,
1) About FilterSpans: I actually started this from SpanNotQuery, so there may
be even more boilerplate to get rid of.
2) b) can also be had by manipulating one of the input spans, and SpanNotQuery
allows passing pre/post offsets to be added before the comparisons. Maybe it is
better to allow some offsets like SpanNotQuery.
2) c), or something close that can be had now by using SpanNearQuery, and
negative pre/post offsets also go in this direction.
ContainedSpanMatchFinder would also work.
I'm not sure in which direction to generalize.
In the current patch the toContained() and toContaining() methods only use
skipTo() on one of the input spans. It is more efficient to also use skipTo()
on the other one, so I'll add that first.
> Span containing/contained queries
> ---------------------------------
>
> Key: LUCENE-6083
> URL: https://issues.apache.org/jira/browse/LUCENE-6083
> Project: Lucene - Core
> Issue Type: Improvement
> Components: core/search
> Reporter: Paul Elschot
> Priority: Minor
> Attachments: LUCENE-6083.patch
>
>
> SpanContainingQuery reducing a spans to where it is containing another spans.
> SpanContainedQuery reducing a spans to where it is contained in another spans.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
