[
https://issues.apache.org/jira/browse/LUCENE-6250?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14330241#comment-14330241
]
Paul Elschot edited comment on LUCENE-6250 at 2/21/15 1:44 PM:
---------------------------------------------------------------
After taking a closer look I think SpanScorer should be changed first to be as
lazy as possible with the underlying Spans.
For example there is a spans.next() call in the SpanScorer constructor, and
that is just not a good idea when doing approximations.
Also I'm adding a Spans.nextAtCurrentDoc() method to allow the span scoring for
a single document to stay at the current document.
This nextAtCurrentDoc() is easy to implement on TermSpans. For the other Span
implementations in core this will take me some time.
was (Author: [email protected]):
After taking a closer look I think SpanScorer should be changed first to be as
lazy as possible with the underlying Spans.
For example there is a spans.next() call in the SpansScorer constructor, and
that is just not a good idea when doing approximations.
Also I'm adding a Spans.nextAtCurrentDoc() method to allow the span scoring for
a single document to stay at the current document.
This nextAtCurrentDoc() is easy to implement on TermSpans. For the other Span
implementations in core this will take me some time.
> Approximations on Spans
> -----------------------
>
> Key: LUCENE-6250
> URL: https://issues.apache.org/jira/browse/LUCENE-6250
> Project: Lucene - Core
> Issue Type: Improvement
> Reporter: Paul Elschot
> Attachments: ApproxSpans-20150216a.patch
>
>
> Approximate spans using existing conjunction/disjunction approximations
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
