jpountz commented on issue #14399:
URL: https://github.com/apache/lucene/issues/14399#issuecomment-2869526058
I don't see an obvious solution either. My preference would be to remove the
cache and make this decision on a per-segment basis, but this would require
moving some methods around,
markrmiller commented on issue #13797:
URL: https://github.com/apache/lucene/issues/13797#issuecomment-2869364967
Yeah, that would be preferable to a bunch of versioned code that essentially
all reads the same format. I suppose I was thinking of using the config for
"what is the minimum ver
rmuir commented on issue #13797:
URL: https://github.com/apache/lucene/issues/13797#issuecomment-2869305562
I think there may be confusion around:
1. minimum created version <-- reflects lucene version that first created
the index
2. minimum version of any segment <-- reflects what act
vigyasharma commented on issue #13797:
URL: https://github.com/apache/lucene/issues/13797#issuecomment-2869271879
Thanks for the example Mark, that was helpful. I think the proposal makes
sense. It makes upgrades across versions easier without adding significant
backward compatibility overh
vigyasharma commented on PR #13976:
URL: https://github.com/apache/lucene/pull/13976#issuecomment-2869263962
Bumping up since the PR went stale. Should I finalize the changes in #14486 ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on t
vigyasharma opened a new issue, #14643:
URL: https://github.com/apache/lucene/issues/14643
I've been running benchmarks on the KNN parent-join query to get comparison
numbers for multi-vectors (https://github.com/apache/lucene/pull/14173). I see
a pretty notable difference in recall when me
uschindler closed pull request #14642: Expressions: Avoid creating a dynamic
constant with BSM if MethodHandle descriptor can be used
URL: https://github.com/apache/lucene/pull/14642
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHu
uschindler commented on PR #14642:
URL: https://github.com/apache/lucene/pull/14642#issuecomment-2869229169
I just added the improved test (with a really private method/class) to main:
04126613864
--
This is an automated message from the Apache Git Service.
To respond to the message, pl
uschindler closed pull request #14642: Expressions: Avoid creating a dynamic
constant with BSM if MethodHandle descriptor can be used
URL: https://github.com/apache/lucene/pull/14642
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHu
uschindler commented on PR #14642:
URL: https://github.com/apache/lucene/pull/14642#issuecomment-2869223919
I am not really happy with that PR because it adds complexity for nonsense:
The improvements are not really there because we move the MethodHandle access
checks and cracking to the co
vigyasharma commented on PR #14443:
URL: https://github.com/apache/lucene/pull/14443#issuecomment-2869220781
Thanks for the review, @mikemccand . I've updated the PR with changes.
I did a rebase on `main` since it had moved quite a bit. There was no
conflict for the files here, but s
uschindler commented on PR #14642:
URL: https://github.com/apache/lucene/pull/14642#issuecomment-2869217876
This code is more complex, but does the right thing. Only if the
MethodHandle is direct and works witho our own lookup, we translate the method
call to a invokestatic directly.
uschindler commented on PR #14642:
URL: https://github.com/apache/lucene/pull/14642#issuecomment-2869214522
There is a problem with that commit, because it does not work with private
lookup. Theres an additional check needed.
The problem is that `describeConstable()` uses a full-trust
vigyasharma commented on code in PR #14443:
URL: https://github.com/apache/lucene/pull/14443#discussion_r2083302934
##
lucene/core/src/java/org/apache/lucene/search/SearcherManager.java:
##
@@ -131,17 +133,32 @@ public SearcherManager(DirectoryReader reader,
SearcherFactory sea
vigyasharma commented on code in PR #14443:
URL: https://github.com/apache/lucene/pull/14443#discussion_r2083301651
##
lucene/core/src/java/org/apache/lucene/search/RefreshCommitSupplier.java:
##
@@ -0,0 +1,39 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
uschindler opened a new pull request, #14642:
URL: https://github.com/apache/lucene/pull/14642
This is an optimization to avoid using a bootstrap when creting the
MethodHandle constant. This won't be backported as it's not easy possible to do
with ASM.
Basically, if the `MethodHandle
dweiss commented on issue #14617:
URL: https://github.com/apache/lucene/issues/14617#issuecomment-2869139385
Cannot be reproduced and happens only occasionally though. Strange.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub a
uschindler merged PR #14641:
URL: https://github.com/apache/lucene/pull/14641
--
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.
uschindler opened a new pull request, #14641:
URL: https://github.com/apache/lucene/pull/14641
nothing to add
--
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
rmuir commented on issue #14617:
URL: https://github.com/apache/lucene/issues/14617#issuecomment-2869067023
> Caused by:
[com.thoughtworks.xstream.io](http://com.thoughtworks.xstream.io/).StreamException:
Invalid character 0xd83a in XML stream
probably caused by an unpaired surrogate
msokolov opened a new pull request, #14640:
URL: https://github.com/apache/lucene/pull/14640
This fixes the test failure and seems correct to me. My only question is
whether `TotalHits.Relation.EQUAL_TO` is a reliable indicator that we did not
`earlyTerminate()`
--
This is an automat
msokolov opened a new issue, #14639:
URL: https://github.com/apache/lucene/issues/14639
### Description
org.apache.lucene.search.TestSeededKnnFloatVectorQuery.testTimeout test
failures demonstrate that when we early terminate (timeout) query execution,
the optimistic strategy still t
rmuir commented on PR #14633:
URL: https://github.com/apache/lucene/pull/14633#issuecomment-2869000472
>
> I thought it did not work with multi-catch, but that it works here. Of
course you could make a more horrible hack, but some people don't like this in
the Lucene community (its know
rmuir merged PR #14356:
URL: https://github.com/apache/lucene/pull/14356
--
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.apach
rmuir commented on PR #14356:
URL: https://github.com/apache/lucene/pull/14356#issuecomment-2868997493
Thank you @renatoh !
--
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.
rmuir commented on PR #13270:
URL: https://github.com/apache/lucene/pull/13270#issuecomment-2868962747
outdated PR: we're now on commons-compress 1.27.1 with updates assisted by
dependabot
--
This is an automated message from the Apache Git Service.
To respond to the message, please log o
rmuir closed pull request #13270: update commons-compress from 1.19 to 1.21
URL: https://github.com/apache/lucene/pull/13270
--
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
rmuir closed pull request #13269: upgrade commons codec from 1.13.0 to 1.16.0
URL: https://github.com/apache/lucene/pull/13269
--
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
rmuir commented on PR #13269:
URL: https://github.com/apache/lucene/pull/13269#issuecomment-2868962207
outdated PR: we're now on commons-codec 1.18.0 with update PRs assisted with
dependabot
--
This is an automated message from the Apache Git Service.
To respond to the message, please log
29 matches
Mail list logo
