[
https://issues.apache.org/jira/browse/LUCENE-1476?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12669025#action_12669025
]
Michael McCandless commented on LUCENE-1476:
--------------------------------------------
{quote}
This seems like something we can port to Java and get into
contrib/benchmark. Particularly automatically creating the indexes.
{quote}
That would be great! Maybe it's not so much work... contrib/benchmark can
already do the "rounds" to enumerate through the different combinations of
query, index, etc.
Automatically creating the indices is trickier; you'd want to 1) create the
first index, 2) copy it and do X%tg deletions, 3) repeat 2 for different values
of X.
Also tricky is gathering a series of results and saving it somehow so that you
can then make a baseline for other runs.
Or we could keep using Python (pickling is so easy ;)). It's nice to have a
"scripting" layer on top of contrib/benchmark.
> 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: hacked-deliterator.patch, 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]
