atris commented on issue #13675:
URL: https://github.com/apache/lucene/issues/13675#issuecomment-2412918581
I have recently been interested in this direction and plan on spending non
trivial amount of time on this over the next few weeks. Assuming we haven't
started dev on this, I am assign
atris commented on issue #13868:
URL: https://github.com/apache/lucene/issues/13868#issuecomment-2412914455
@jpountz Interestingly, I have been spending some time on making this work -
if its ok, I will self assign this issue?
--
This is an automated message from the Apache Git Service.
T
vsop-479 commented on PR #12868:
URL: https://github.com/apache/lucene/pull/12868#issuecomment-2412698518
BTW, why `StringHelper` is a abstract class? Can we make it final?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
HoustonPutman commented on issue #13760:
URL: https://github.com/apache/lucene/issues/13760#issuecomment-2412577287
@timgrein , I've posted a PR for my idea that @stefanvodita mentioned. If
you have the JMH benchmark, I'd love to test it out on mine as well.
--
This is an automated messag
github-actions[bot] commented on PR #13834:
URL: https://github.com/apache/lucene/pull/13834#issuecomment-2412576598
This PR has not had activity in the past 2 weeks, labeling it as stale. If
the PR is waiting for review, notify the d...@lucene.apache.org list. Thank you
for your contributi
HoustonPutman opened a new pull request, #13914:
URL: https://github.com/apache/lucene/pull/13914
Resolves #13760
### Description
This is using a similar approach to how Solr used to compute multiple
percentiles at a single time. Basically utilize the quick select metho
mdmarshmallow opened a new issue, #13913:
URL: https://github.com/apache/lucene/issues/13913
### Description
This is related to https://github.com/apache/lucene/issues/13883. The idea
is to allow users to specify the RAM usage once and it will be automatically
spread across N IndexWr
msokolov commented on PR #13910:
URL: https://github.com/apache/lucene/pull/13910#issuecomment-2412421807
I wonder whether we should backport the fixes to the
`Lucene90HnswVectorsReader`? I tend to think we ought to, although the usage
might be tiny to nonexistent
--
This is an automated
javanna commented on PR #13874:
URL: https://github.com/apache/lucene/pull/13874#issuecomment-2412365840
I did find above a diff that is exactly the same as the PR I opened to
update AddBackcompatindices :) (#13911) . I also opened a PR against main to
cherry-pick this change and make the
javanna opened a new pull request, #13911:
URL: https://github.com/apache/lucene/pull/13911
We updated TestGenerateBwcIndices to create int7 HNSW indices instead of
int8 with #13874. The corresponding python code part of the release wizard
needs to be updated accordingly.
--
This
javanna commented on PR #13874:
URL: https://github.com/apache/lucene/pull/13874#issuecomment-2412274304
Heya, I am working on generating the backwards indices after releasing
Lucene 10, and here are my observations:
- I think that we need to forward port this change to main as well?
uschindler commented on PR #13906:
URL: https://github.com/apache/lucene/pull/13906#issuecomment-2411866037
P.P.S.: Just some background: The code was copied from ByteBufferIndexInput
where the clones were necessary. With MemorySegment we no longer create
`dup()`s, as MemorySegment is state
original-brownbear commented on PR #13906:
URL: https://github.com/apache/lucene/pull/13906#issuecomment-2411864517
> One thing that is a bit awkward to me is that it makes clones cheaper than
slices, so e.g. refactoring TermsEnum#postings to work on a slice that contains
just the postings
msokolov opened a new pull request, #13910:
URL: https://github.com/apache/lucene/pull/13910
While exploring some recall-related failures in another PR I went looking
for a unit test that checks HNSW/KNN recall and couldn't find any. I think we
used to have one but maybe we removed it becau
uschindler commented on code in PR #13906:
URL: https://github.com/apache/lucene/pull/13906#discussion_r1799835249
##
lucene/core/src/java21/org/apache/lucene/store/MemorySegmentIndexInput.java:
##
@@ -563,6 +563,13 @@ public final MemorySegmentIndexInput slice(String
sliceDesc
uschindler commented on code in PR #13906:
URL: https://github.com/apache/lucene/pull/13906#discussion_r1799826647
##
lucene/core/src/java21/org/apache/lucene/store/MemorySegmentIndexInput.java:
##
@@ -563,6 +563,13 @@ public final MemorySegmentIndexInput slice(String
sliceDesc
zkendall commented on code in PR #2686:
URL: https://github.com/apache/lucene-solr/pull/2686#discussion_r1799751372
##
solr/core/src/java/org/apache/solr/update/SolrIndexSplitter.java:
##
@@ -130,9 +131,26 @@ public SolrIndexSplitter(SplitIndexCommand cmd) {
}
routeFie
jpountz commented on code in PR #13906:
URL: https://github.com/apache/lucene/pull/13906#discussion_r1799801328
##
lucene/core/src/java21/org/apache/lucene/store/MemorySegmentIndexInput.java:
##
@@ -563,6 +563,13 @@ public final MemorySegmentIndexInput slice(String
sliceDescrip
dsmiley commented on code in PR #2686:
URL: https://github.com/apache/lucene-solr/pull/2686#discussion_r1799803611
##
solr/core/src/java/org/apache/solr/update/SolrIndexSplitter.java:
##
@@ -130,9 +131,26 @@ public SolrIndexSplitter(SplitIndexCommand cmd) {
}
routeFiel
stefanvodita commented on issue #13898:
URL: https://github.com/apache/lucene/issues/13898#issuecomment-2411705472
@prudhvigodithi - I opened #13898 inspired by your PR, tested it on my fork,
and it's working correctly.
@javanna - I like that idea! Maybe we can do that as a follow-up.
stefanvodita opened a new pull request, #13909:
URL: https://github.com/apache/lucene/pull/13909
### Description
This runs along the checks we already have for PR creation/update and warns
us if there is no CHANGES.txt entry.
--
This is an automated message from the Apache Git
stefanvodita commented on PR #13908:
URL: https://github.com/apache/lucene/pull/13908#issuecomment-2411670171
Testing a Github workflow, not intended to merge.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL ab
stefanvodita closed pull request #13908: Test changelog verifier
URL: https://github.com/apache/lucene/pull/13908
--
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
stefanvodita opened a new pull request, #13908:
URL: https://github.com/apache/lucene/pull/13908
test
--
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
zkendall commented on code in PR #2686:
URL: https://github.com/apache/lucene-solr/pull/2686#discussion_r1799751372
##
solr/core/src/java/org/apache/solr/update/SolrIndexSplitter.java:
##
@@ -130,9 +131,26 @@ public SolrIndexSplitter(SplitIndexCommand cmd) {
}
routeFie
zkendall commented on code in PR #2686:
URL: https://github.com/apache/lucene-solr/pull/2686#discussion_r1799751372
##
solr/core/src/java/org/apache/solr/update/SolrIndexSplitter.java:
##
@@ -130,9 +131,26 @@ public SolrIndexSplitter(SplitIndexCommand cmd) {
}
routeFie
zkendall commented on code in PR #2686:
URL: https://github.com/apache/lucene-solr/pull/2686#discussion_r1799751372
##
solr/core/src/java/org/apache/solr/update/SolrIndexSplitter.java:
##
@@ -130,9 +131,26 @@ public SolrIndexSplitter(SplitIndexCommand cmd) {
}
routeFie
zkendall commented on code in PR #2686:
URL: https://github.com/apache/lucene-solr/pull/2686#discussion_r1799751372
##
solr/core/src/java/org/apache/solr/update/SolrIndexSplitter.java:
##
@@ -130,9 +131,26 @@ public SolrIndexSplitter(SplitIndexCommand cmd) {
}
routeFie
timgrein commented on issue #13760:
URL: https://github.com/apache/lucene/issues/13760#issuecomment-2411631641
Cool, I've found some performance improvements (~10-15%), which can be
reproduced through a new `jmh` benchmark I've added. I'll open a PR the next
few days and tag you :)
--
Th
prudhvigodithi commented on PR #13902:
URL: https://github.com/apache/lucene/pull/13902#issuecomment-2411600165
Thanks for the review and approval @mkhludnev @jpountz.
@getsaurabh02 @dblock
--
This is an automated message from the Apache Git Service.
To respond to the message, please lo
prudhvigodithi commented on issue #12686:
URL: https://github.com/apache/lucene/issues/12686#issuecomment-2411600774
Hey @harshavamsi the PR to support the `IndexOrDocValuesQuery` for query
highlighting is now merged.
Thanks.
@getsaurabh02 @msfroh
--
This is an automated message fro
original-brownbear merged PR #13901:
URL: https://github.com/apache/lucene/pull/13901
--
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...
original-brownbear commented on PR #13901:
URL: https://github.com/apache/lucene/pull/13901#issuecomment-241157
Thanks Adrien!
--
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 com
prudhvigodithi commented on issue #13896:
URL: https://github.com/apache/lucene/issues/13896#issuecomment-2411592601
Thanks @stefanvodita, I would like to have the maintainers of the repo to
take the initial stab at this metrics dashboard (come up the server, app,
choosing the database and
jpountz opened a new pull request, #13907:
URL: https://github.com/apache/lucene/pull/13907
This commit tries to save calls to `madvise` which are not necessary, either
because they map to the OS' default, or because the advice would be overridden
later on anyway. I have not noticed specifi
original-brownbear commented on code in PR #13901:
URL: https://github.com/apache/lucene/pull/13901#discussion_r1799704584
##
lucene/core/src/java/org/apache/lucene/codecs/lucene912/Lucene912PostingsReader.java:
##
@@ -429,9 +403,44 @@ public PostingsEnum reset(IntBlockTermState
original-brownbear opened a new pull request, #13906:
URL: https://github.com/apache/lucene/pull/13906
No need to slice when it's a clone that is to be used with random access, we
already enforce thread access rules anyway. Also, no point in copying the
memory segment instance via noop slic
dnhatn merged PR #13903:
URL: https://github.com/apache/lucene/pull/13903
--
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.apac
stefanvodita commented on issue #13896:
URL: https://github.com/apache/lucene/issues/13896#issuecomment-2411304207
That sounds like a plan! I'd be thrilled if you were driving this
@prudhvigodithi - thank you for offering your help.
It would also be great to get feedback from others on wh
jpountz merged PR #12868:
URL: https://github.com/apache/lucene/pull/12868
--
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
easyice opened a new pull request, #13905:
URL: https://github.com/apache/lucene/pull/13905
Clean up unused variable `MergePointVisitor#sortedPackedPoints`
--
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
jpountz commented on PR #13800:
URL: https://github.com/apache/lucene/pull/13800#issuecomment-243632
Here's the fix for conjunctions: https://github.com/apache/lucene/pull/13904.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Gi
jpountz commented on code in PR #13901:
URL: https://github.com/apache/lucene/pull/13901#discussion_r1799409756
##
lucene/core/src/java/org/apache/lucene/codecs/lucene912/Lucene912PostingsReader.java:
##
@@ -429,9 +403,44 @@ public PostingsEnum reset(IntBlockTermState termState,
jpountz merged PR #13902:
URL: https://github.com/apache/lucene/pull/13902
--
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
bugmakerr commented on PR #13763:
URL: https://github.com/apache/lucene/pull/13763#issuecomment-2410922885
hi @jpountz ,since we have moved to lucene 10, should we merge this and add
back #13686?
--
This is an automated message from the Apache Git Service.
To respond to the message, p
uschindler commented on PR #13894:
URL: https://github.com/apache/lucene/pull/13894#issuecomment-2410803086
I was a bit afraid that this might have been caused by missing "prefetch()"
implementation, but for MemorySegmentIndexInput also the random access slice
uses the correct implementatio
uschindler commented on PR #13894:
URL: https://github.com/apache/lucene/pull/13894#issuecomment-2410749511
I dont think this is a real slowdown caused by the commit here. It is more
caused by the Hotspot optimizer misinterpreting something. We should get the
assembly code from the benchmar
shubhamvishu commented on PR #12868:
URL: https://github.com/apache/lucene/pull/12868#issuecomment-2410748497
@jpountz I simplified the expression now. Let me know if the change looks
good? Thanks!
--
This is an automated message from the Apache Git Service.
To respond to the message, ple
shubhamvishu commented on code in PR #12868:
URL: https://github.com/apache/lucene/pull/12868#discussion_r1799172633
##
lucene/codecs/src/java/org/apache/lucene/codecs/bloom/FuzzySet.java:
##
@@ -150,9 +149,10 @@ private FuzzySet(FixedBitSet filter, int bloomSize, int
hashCount
jpountz commented on PR #13904:
URL: https://github.com/apache/lucene/pull/13904#issuecomment-2410663119
I confirmed that this makes things better when all clauses are
`ConstantScoreQuery`s, luceneutil doesn't report a slowdown:
```
TaskQPS baseline
original-brownbear commented on PR #13894:
URL: https://github.com/apache/lucene/pull/13894#issuecomment-2410525472
Looks to me like a few inlining decisions changed (judging by the profiling
and the relative weight of inlined versions vs non-inlined versions of some
method). I'm not even e
jpountz opened a new pull request, #13904:
URL: https://github.com/apache/lucene/pull/13904
`BlockMaxConjunctionBulkScorer` only checks if it can early exit based on
impacts once per window, and windows are computed using impact blocks of the
leading clause. So this logic is defeated if the
javanna commented on code in PR #13891:
URL: https://github.com/apache/lucene/pull/13891#discussion_r1798961357
##
lucene/CHANGES.txt:
##
@@ -47,6 +52,9 @@ API Changes
the entire segment should be scored. Subclasses that override the method
should instead override its replac
javanna commented on issue #13898:
URL: https://github.com/apache/lucene/issues/13898#issuecomment-2410385807
Yes to this! It would be great to combine this with setting the milestone :)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on t
jpountz commented on PR #13894:
URL: https://github.com/apache/lucene/pull/13894#issuecomment-2410306410
Looks like there was a slowdown on some taxo facets tasks (which use binary
doc values to store the taxonomy). `OrHighMedDayTaxoFacets`,
`AndHighHighDayTaxoFacets` and `MedTermDayTaxoFac
55 matches
Mail list logo
