iverase commented on code in PR #13668:
URL: https://github.com/apache/lucene/pull/13668#discussion_r1722783860
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDConfig.java:
##
@@ -31,48 +48,23 @@ public final class BKDConfig {
/** Maximum number of index dimensions */
epotyom commented on PR #13657:
URL: https://github.com/apache/lucene/pull/13657#issuecomment-2298104290
> @epotyom
> > I've noticed by accident that creating two identical lists of
collectors, before and after createWeight, makes regression disappear
>
> Hmm, that is strange. I w
gsmiller commented on PR #13672:
URL: https://github.com/apache/lucene/pull/13672#issuecomment-2297951310
It just occurred to me we can’t use DocValuesRangeIterator directly here as
it could produce false positive matches. I’ll write a test to capture this and
iterate on the PR.
--
This
rmuir commented on code in PR #13672:
URL: https://github.com/apache/lucene/pull/13672#discussion_r1722557603
##
lucene/core/src/test/org/apache/lucene/search/TestDocValuesRewriteMethod.java:
##
@@ -61,14 +61,19 @@ public void setUp() throws Exception {
.setMaxB
rmuir commented on code in PR #13672:
URL: https://github.com/apache/lucene/pull/13672#discussion_r1722556851
##
lucene/core/src/java/org/apache/lucene/document/DocValuesRangeIterator.java:
##
@@ -24,8 +24,10 @@
/**
* Wrapper around a {@link TwoPhaseIterator} for a doc-values
gsmiller opened a new pull request, #13672:
URL: https://github.com/apache/lucene/pull/13672
### Description
`DocValuesRewriteMethod` (used under-the-hood in `MultiTermQuery`) is
another spot where we can leverage the new doc value skip lists when present to
potentially skip over doc
gsmiller commented on PR #13568:
URL: https://github.com/apache/lucene/pull/13568#issuecomment-2297766370
Thanks @javanna for raising this. I agree with trying to figure out how to
shrink this API surface area in `IndexSearcher` back down prior to the 10.0
release. I don't have any immediat
gsmiller opened a new issue, #13671:
URL: https://github.com/apache/lucene/issues/13671
### Description
GH#13568 introduced a new public method to `IndexSearcher` along with a new
concept of a `CollectorOwner`. This was done to allow drill-sideways to use the
new sandbox faceting fun
javanna commented on PR #13568:
URL: https://github.com/apache/lucene/pull/13568#issuecomment-2297566037
Hey all, sorry about my late comment, I noticed the `IndexSearcher` changes,
the additional search method that takes a `CollectorOwner` and the new
`CollectorOwner` abstraction. It looks
gsmiller commented on PR #13657:
URL: https://github.com/apache/lucene/pull/13657#issuecomment-2297489039
@epotyom
--
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 unsub
gsmiller merged PR #13669:
URL: https://github.com/apache/lucene/pull/13669
--
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.ap
gsmiller commented on PR #13669:
URL: https://github.com/apache/lucene/pull/13669#issuecomment-2297481110
Looks good, thanks! I'm OK with merging this, but I want to also take a
moment to ask that we're somewhat cautious with racking up lots of small
changes like this that cannot be back-po
vigyasharma commented on PR #13525:
URL: https://github.com/apache/lucene/pull/13525#issuecomment-2297380564
> This PR has not had activity in the past 2 weeks, labeling it as stale...
Just to update on some activity here:
I'm working on parent block join benchmarks in `luceneutil`
gsmiller commented on PR #13658:
URL: https://github.com/apache/lucene/pull/13658#issuecomment-2297082943
Good question @jpountz. If I'm looking at this properly, it seems to not be
getting inlined in my benchmark run. (I grabbed your patch from #13670 and
loaded one of the jfr files produc
uschindler commented on code in PR #13668:
URL: https://github.com/apache/lucene/pull/13668#discussion_r1722023350
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDConfig.java:
##
@@ -31,48 +48,23 @@ public final class BKDConfig {
/** Maximum number of index dimensions *
mikemccand commented on PR #12862:
URL: https://github.com/apache/lucene/pull/12862#issuecomment-2296974933
Hmm it looks like this change was nearly ready to be merged? But code has
since drifted and the PR now has some rust (conflicts)?
--
This is an automated message from the Apache Gi
jpountz commented on PR #13658:
URL: https://github.com/apache/lucene/pull/13658#issuecomment-2296948223
Thanks for looking @gsmiller ! Are the splitLongs calls also not inlined for
you as well? I submitted a PR to make nightly reports tell us about which
methods get inlined / compiled / in
uschindler commented on code in PR #13668:
URL: https://github.com/apache/lucene/pull/13668#discussion_r1722023350
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDConfig.java:
##
@@ -31,48 +48,23 @@ public final class BKDConfig {
/** Maximum number of index dimensions *
epotyom commented on PR #13657:
URL: https://github.com/apache/lucene/pull/13657#issuecomment-2296904888
Just for tracking, repeating this comment
https://github.com/apache/lucene/pull/13656#issuecomment-2296813725 here:
> The biggest difference in the profiler seems to be that we spe
gsmiller commented on PR #13658:
URL: https://github.com/apache/lucene/pull/13658#issuecomment-2296882889
I ran `wikimedium10m` and can (sort of) see the regression on AndHighMed
(but only that task). Results below, but I see `2.4% with a 0.058 p-value. This
is on an AWS m4.12xlarge x86-bas
ChrisHegarty commented on issue #13659:
URL: https://github.com/apache/lucene/issues/13659#issuecomment-2296727766
> I mean it would be nice if we didn't generate these degenerate Component
in the first place? But this will work
I had the same thought, but then backed away quickly! ;
iverase commented on code in PR #13668:
URL: https://github.com/apache/lucene/pull/13668#discussion_r1721868595
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDConfig.java:
##
@@ -31,48 +48,23 @@ public final class BKDConfig {
/** Maximum number of index dimensions */
mrhbj opened a new pull request, #13669:
URL: https://github.com/apache/lucene/pull/13669
### Description
--
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 uns
mrhbj closed pull request #13666: Make the code look simpler
URL: https://github.com/apache/lucene/pull/13666
--
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-
mikemccand commented on PR #13656:
URL: https://github.com/apache/lucene/pull/13656#issuecomment-2296609036
> Have you followed what @epotyom and I found so far in #13657?
Yes! And it's completely insane... it might be time to turn on
`-XX:+PrintAssembly` and watch how hotspot is spe
ChrisHegarty commented on PR #13660:
URL: https://github.com/apache/lucene/pull/13660#issuecomment-2296590719
Belated LGTM.
--
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.
uschindler commented on code in PR #13668:
URL: https://github.com/apache/lucene/pull/13668#discussion_r1721755168
##
lucene/core/src/java/org/apache/lucene/util/bkd/BKDConfig.java:
##
@@ -31,48 +48,23 @@ public final class BKDConfig {
/** Maximum number of index dimensions *
mrhbj commented on PR #13666:
URL: https://github.com/apache/lucene/pull/13666#issuecomment-2296511386
you are right,i will change my commit messave,thanks
---Original---
From: "Michael ***@***.***>
Date: Mon, Aug 19, 2024 20:32 PM
To: ***@***.***>;
Cc: ***@***.
benwtrent closed issue #13659:
TestHnswFloatVectorGraph.testRandomReadWriteAndMerge fails with
java.lang.IndexOutOfBoundsException
URL: https://github.com/apache/lucene/issues/13659
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHu
benwtrent merged PR #13660:
URL: https://github.com/apache/lucene/pull/13660
--
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.a
msokolov commented on PR #13666:
URL: https://github.com/apache/lucene/pull/13666#issuecomment-2296463007
This looks good to me, but perhaps you could make the commit message more
descriptive of the actual change. If I read "make the code look simpler" I have
misgivings! Maybe mention the s
ChrisHegarty opened a new issue, #13667:
URL: https://github.com/apache/lucene/issues/13667
### Description
Fails intermittently on _main_, with the following stacktrace:
```
TestTieredMergePolicy > testSimulateUpdates FAILED
java.lang.AssertionError: mergeFactor=2 min
vsop-479 commented on PR #13596:
URL: https://github.com/apache/lucene/pull/13596#issuecomment-2296249587
@jpountz
Please take a look when you get a chance.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL a
mrhbj opened a new pull request, #13666:
URL: https://github.com/apache/lucene/pull/13666
### Description
--
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 uns
mrhbj closed pull request #13665: Make the code look simpler
URL: https://github.com/apache/lucene/pull/13665
--
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-
mrhbj opened a new pull request, #13665:
URL: https://github.com/apache/lucene/pull/13665
### Description
--
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 uns
36 matches
Mail list logo
