jpountz commented on PR #14353:
URL: https://github.com/apache/lucene/pull/14353#issuecomment-2721927266
Thanks @iverase !
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
T
jpountz merged PR #14353:
URL: https://github.com/apache/lucene/pull/14353
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@lucene.apa
jpountz opened a new pull request, #14353:
URL: https://github.com/apache/lucene/pull/14353
When a `PointRangeQuery` matches for than 50% of points and the field is
dense and single-valued, it internally computes the set of docs that don't
match the query, which should be faster since fewer
jpountz commented on PR #14353:
URL: https://github.com/apache/lucene/pull/14353#issuecomment-2721296892
Maybe I spoke too soon, there seems to be a tiny speedup:
```
TaskQPS baseline StdDevQPS
my_modified_version StdDevPct diff
jpountz commented on PR #14353:
URL: https://github.com/apache/lucene/pull/14353#issuecomment-2721270141
Nightly benchmarks don't show a change since their points are not clustered
by doc ID so they don't call `visit(DocIdSetIterator)`.
--
This is an automated message from the Apache Git
