[
https://issues.apache.org/jira/browse/LUCENE-1476?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12668797#action_12668797
]
Michael McCandless commented on LUCENE-1476:
--------------------------------------------
bq. Presumably you spliced the improved nextSetBit into Jason's patch, correct?
Actually I used your entire patch on its own.
bq. Didn't you have a close-to-ideal patch using sorted ints that performed
well up to 10% deletions? What did that look like?
I thought I did -- but it was rather hacked up (I "fixed" SegmentReader to do
always do an up-front conversion into int[] deletedDocs). I'll re-test it to
try to repro my initial rough results.
bq. I also suspect that when there are many deletions, the sheer number of
method calls to perform the deletions iteration is a burden. The iterator has
to compete with an inline-able method from a final class (BitVector).
Right, for a highish %tg deletion it seems likely that random-access will win.
> BitVector implement DocIdSet, IndexReader returns DocIdSet deleted docs
> -----------------------------------------------------------------------
>
> Key: LUCENE-1476
> URL: https://issues.apache.org/jira/browse/LUCENE-1476
> Project: Lucene - Java
> Issue Type: Improvement
> Components: Index
> Affects Versions: 2.4
> Reporter: Jason Rutherglen
> Priority: Trivial
> Attachments: LUCENE-1476.patch, LUCENE-1476.patch, LUCENE-1476.patch,
> LUCENE-1476.patch, LUCENE-1476.patch, quasi_iterator_deletions.diff,
> quasi_iterator_deletions_r2.diff, quasi_iterator_deletions_r3.diff,
> searchdeletes.alg, sortBench2.py, sortCollate2.py, TestDeletesDocIdSet.java
>
> Original Estimate: 12h
> Remaining Estimate: 12h
>
> Update BitVector to implement DocIdSet. Expose deleted docs DocIdSet from
> IndexReader.
--
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: [email protected]
For additional commands, e-mail: [email protected]
