[
https://issues.apache.org/jira/browse/LUCENE-5293?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13801023#comment-13801023
]
Paul Elschot commented on LUCENE-5293:
--------------------------------------
bq. Instead of keeping the oversized encoder, maybe the encoder could be
resized to its actual size when finished encoding? This is what the other
(wah8/pfor) sets do.
I suppose you mean by recomputing the actually needed sizes for the long arrays
of the encoder after all encoding is done, and then reallocating them?
That would certainly be possible. I'll have a look at wah8/pfor for this.
bq. Maybe pfor/wah8 would still be useful here since they efficiently encode
almost full sets?
The problem is that none of the compressing sets is as good as FBS at advancing
far in almost full sets.
I'm working on this now, another patch is slowly on its way.
> Also use EliasFanoDocIdSet in CachingWrapperFilter
> --------------------------------------------------
>
> Key: LUCENE-5293
> URL: https://issues.apache.org/jira/browse/LUCENE-5293
> Project: Lucene - Core
> Issue Type: Improvement
> Components: core/search
> Reporter: Paul Elschot
> Priority: Minor
> Attachments: LUCENE-5293.patch, LUCENE-5293.patch, LUCENE-5293.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.1#6144)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
