mocobeta commented on code in PR #917:
URL: https://github.com/apache/lucene/pull/917#discussion_r878647487
##
.github/workflows/distribution.yml:
##
@@ -19,8 +19,9 @@ jobs:
runs-on: ${{ matrix.os }}
strategy:
matrix:
-# we run the distribution tests on
mocobeta commented on code in PR #917:
URL: https://github.com/apache/lucene/pull/917#discussion_r878647263
##
.github/workflows/distribution.yml:
##
@@ -19,8 +19,9 @@ jobs:
runs-on: ${{ matrix.os }}
strategy:
matrix:
-# we run the distribution tests on
mocobeta commented on PR #913:
URL: https://github.com/apache/lucene/pull/913#issuecomment-1133536049
> Looks like some other issue blocking the forked process.
Yes I think it could completely hang for some reason.
--
This is an automated message from the Apache Git Service.
To resp
dweiss commented on PR #913:
URL: https://github.com/apache/lucene/pull/913#issuecomment-1133533455
> I'm planning to disable the test on windows VM - it looks too slow in
GitHub Actions.
I'm not sure it's the slowness anymore. Looks like some other issue blocking
the forked process.
mocobeta commented on PR #913:
URL: https://github.com/apache/lucene/pull/913#issuecomment-1133532803
Also, it should run on headless JVM; we are running it on CI servers without
displays.
https://github.com/apache/lucene/blob/71a9acb2e2aa55257021eefce1e5d8d390bc7048/lucene/luke/src/java
mocobeta commented on PR #893:
URL: https://github.com/apache/lucene/pull/893#issuecomment-1133530902
We increased the timeout to 120 seconds, but the test still occasionally
fails on Windows. I opened #917 to disable it on windows for now.
--
This is an automated message from the Apache
mocobeta opened a new pull request, #917:
URL: https://github.com/apache/lucene/pull/917
The test occasionally is too slow and fails on Windows VM.
We can't increase the timeout infinitely, I would disable it on GitHub
Actions for now (we still run it on Jenkins Servers).
--
This is a
mocobeta commented on PR #913:
URL: https://github.com/apache/lucene/pull/913#issuecomment-1133529000
> hm this "luke-can-be-launched" test failed for me locally too. I guess it
needs a JVM with a head?
It's not a part of the mandatory test suite but you can run it headlessly
(if you
rmuir commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878641687
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+ p
mocobeta opened a new pull request, #916:
URL: https://github.com/apache/lucene/pull/916
Some part of the pull request template has been a bit obsoleted, and there
are overlaps between the contribution guide and the template.
I noticed the majority of developers (including me...) dele
[
https://issues.apache.org/jira/browse/LUCENE-10312?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540341#comment-17540341
]
ASF subversion and git services commented on LUCENE-10312:
--
Co
mocobeta merged PR #908:
URL: https://github.com/apache/lucene/pull/908
--
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.apache
mocobeta commented on code in PR #912:
URL: https://github.com/apache/lucene/pull/912#discussion_r878635805
##
lucene/core/src/java/org/apache/lucene/store/MMapDirectory.java:
##
@@ -232,55 +224,62 @@ public IndexInput openInput(String name, IOContext
context) throws IOExceptio
gsmiller commented on PR #915:
URL: https://github.com/apache/lucene/pull/915#issuecomment-1133483198
Ran benchmarks to make sure this didn't regress performance in any
meaningful way. I'm not seeing any regressions:
```
TaskQPS baseline StdDevQPS c
gsmiller opened a new pull request, #915:
URL: https://github.com/apache/lucene/pull/915
# Description
The facets module has had a fair amount of recent development, some
resulting in a fair amount of copy/paste duplication. This attempts to clean it
up a bit and simplify/refactor so
Yuti-G commented on PR #777:
URL: https://github.com/apache/lucene/pull/777#issuecomment-1133481903
> This looks great! Thanks @Yuti-G! It would be nice if we could create a
common abstract class to hold some of the common logic between this and the
non-concurrent implementation. Seems like
Yuti-G opened a new pull request, #914:
URL: https://github.com/apache/lucene/pull/914
# Description
This new API returns all children for a specified path and allows users to
sort in their desired criteria
# Solution
* Added getAllChildren function in the Facets class.
[
https://issues.apache.org/jira/browse/LUCENE-10370?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dawid Weiss resolved LUCENE-10370.
--
Resolution: Fixed
> Fix classpath/module path of tests forking their own Java (TestNRTReplica
msokolov commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878580201
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+
[
https://issues.apache.org/jira/browse/LUCENE-10370?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dawid Weiss updated LUCENE-10370:
-
Fix Version/s: 9.3
> Fix classpath/module path of tests forking their own Java (TestNRTReplicat
[
https://issues.apache.org/jira/browse/LUCENE-10370?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540315#comment-17540315
]
ASF subversion and git services commented on LUCENE-10370:
--
Co
[
https://issues.apache.org/jira/browse/LUCENE-10370?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540314#comment-17540314
]
ASF subversion and git services commented on LUCENE-10370:
--
Co
dweiss merged PR #909:
URL: https://github.com/apache/lucene/pull/909
--
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.apache.o
dweiss commented on PR #909:
URL: https://github.com/apache/lucene/pull/909#issuecomment-1133434751
> If we make it an assumption, it makes the results look a bit nicer?
Depends. It can go unnoticed then because it superficially looks just as if
the test was ignored. I'd rather have i
[
https://issues.apache.org/jira/browse/LUCENE-10544?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540309#comment-17540309
]
Greg Miller edited comment on LUCENE-10544 at 5/20/22 10:03 PM:
-
[
https://issues.apache.org/jira/browse/LUCENE-10544?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540309#comment-17540309
]
Greg Miller commented on LUCENE-10544:
--
As I understood Adrien's suggestion, I thi
jtibshirani commented on code in PR #910:
URL: https://github.com/apache/lucene/pull/910#discussion_r878568768
##
lucene/sandbox/src/test/org/apache/lucene/sandbox/search/TestCombinedFieldQuery.java:
##
@@ -589,4 +589,52 @@ public SimScorer scorer(
return new BM25Similari
msokolov commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878557225
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+
rmuir commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878562615
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+ p
madrob opened a new pull request, #2660:
URL: https://github.com/apache/lucene-solr/pull/2660
Depends on #2641 will commit both at the same time
--
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 th
msokolov commented on PR #913:
URL: https://github.com/apache/lucene/pull/913#issuecomment-1133402474
hm this "luke-can-be-launched" test failed for me locally too. I guess it
needs a JVM with a head?
--
This is an automated message from the Apache Git Service.
To respond to the message,
msokolov commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878557225
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+
rmuir commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878555057
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+ p
msokolov commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878553114
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+
rmuir commented on code in PR #913:
URL: https://github.com/apache/lucene/pull/913#discussion_r878550812
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -213,4 +213,21 @@ public static void add(float[] u, float[] v) {
u[i] += v[i];
}
}
+
+ p
msokolov commented on PR #909:
URL: https://github.com/apache/lucene/pull/909#issuecomment-1133380108
If we make it an assumption, it makes the results look a bit nicer?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
jtibshirani commented on PR #833:
URL: https://github.com/apache/lucene/pull/833#issuecomment-1133369782
Thanks @mocobeta, it is so nice to have benchmarks for
ExitableDirectoryReader now. It looks like the impact is not significant for
vector search.
--
This is an automated message from
msokolov opened a new pull request, #913:
URL: https://github.com/apache/lucene/pull/913
This patch revisits the idea of storing vectors in 8-bit precision. It
incorporates some @rcmuir and @jpountz suggestions from
https://github.com/apache/lucene/pull/899 which was the first attempt. The
msokolov commented on code in PR #899:
URL: https://github.com/apache/lucene/pull/899#discussion_r878531565
##
lucene/core/src/java/org/apache/lucene/codecs/lucene92/ExpandingRandomAccessVectorValues.java:
##
@@ -0,0 +1,57 @@
+package org.apache.lucene.codecs.lucene92;
+
+import
msokolov closed pull request #899: Lucene 10577
URL: https://github.com/apache/lucene/pull/899
--
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-un
[
https://issues.apache.org/jira/browse/LUCENE-10574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540283#comment-17540283
]
Robert Muir commented on LUCENE-10574:
--
The benchmark was originally some code wri
[
https://issues.apache.org/jira/browse/LUCENE-9625?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540284#comment-17540284
]
Michael Sokolov commented on LUCENE-9625:
-
I realized maybe this deserves a bett
[
https://issues.apache.org/jira/browse/LUCENE-10585?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Greg Miller reassigned LUCENE-10585:
Assignee: Greg Miller
> Cleanup copy/paste code in facets, particularly in SSDV
> --
Greg Miller created LUCENE-10585:
Summary: Cleanup copy/paste code in facets, particularly in SSDV
Key: LUCENE-10585
URL: https://issues.apache.org/jira/browse/LUCENE-10585
Project: Lucene - Core
[
https://issues.apache.org/jira/browse/LUCENE-10584?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Greg Miller reassigned LUCENE-10584:
Assignee: Greg Miller
> SSDV facets should support hierarchical paths in #getSpecificVal
[
https://issues.apache.org/jira/browse/LUCENE-10574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540280#comment-17540280
]
Michael Sokolov commented on LUCENE-10574:
--
Very nice! I confess I don't reall
Greg Miller created LUCENE-10584:
Summary: SSDV facets should support hierarchical paths in
#getSpecificValue
Key: LUCENE-10584
URL: https://issues.apache.org/jira/browse/LUCENE-10584
Project: Lucene
mikemccand commented on PR #633:
URL: https://github.com/apache/lucene/pull/633#issuecomment-1133249474
Hmm I beasted the PR on current `main` branch using `./gradlew
:lucene:core:test --tests "org.apache.lucene.index.TestAddIndexes"
-Dtests.iters=500 `.
The first time, it seemed to
Thomas Hoffmann created LUCENE-10583:
Summary: Deadlock with MMapDirectory while waitForMerges
Key: LUCENE-10583
URL: https://issues.apache.org/jira/browse/LUCENE-10583
Project: Lucene - Core
uschindler opened a new pull request, #912:
URL: https://github.com/apache/lucene/pull/912
**INFO: This is a followup of #518: It's the same code base, but with API
changes from JDK 19 applied**
This is just a draft PR for a first insight on memory mapping improvements
in JDK 19+.
uschindler closed pull request #911: Initial rewrite of MMapDirectory for
JDK-19 preview Panama APIs (>= JDK-19-ea+23)
URL: https://github.com/apache/lucene/pull/911
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
mikemccand commented on PR #633:
URL: https://github.com/apache/lucene/pull/633#issuecomment-1133136314
I think this is ready!
What a great change, giving `MergePolicy` purview over how `addIndexes` does
its merging too. I'm also beasting `TestAddIndexes` and will report back.
rmuir commented on PR #909:
URL: https://github.com/apache/lucene/pull/909#issuecomment-1133125938
i'm suspicious if some of these tests would even run directly from the IDE
anyway. even if you fallback to classpath, TestIndexWriterOnJRECrash might not
work from IDEA. I installed idea the o
uschindler opened a new pull request, #911:
URL: https://github.com/apache/lucene/pull/911
**INFO: This is a followup of #518: It's the same code base, but with API
changes from JDK 19 applied**
This is just a draft PR for a first insight on memory mapping improvements
in JDK 19+.
dweiss commented on PR #909:
URL: https://github.com/apache/lucene/pull/909#issuecomment-1133086554
@uschindler Are you ok with the patch as is? Should I rework it into an
assumption error? Should I fall back to classpath if nothing is available (I'm
not sold on this one)?
--
This is an
mocobeta commented on PR #833:
URL: https://github.com/apache/lucene/pull/833#issuecomment-1133038858
For your information, luceneutil now supports ExitableDirectoryReader
(https://github.com/mikemccand/luceneutil/pull/172) and I ran the benchmark
with vector values.
Data: 500k English W
ywelsch opened a new pull request, #910:
URL: https://github.com/apache/lucene/pull/910
CombinedFieldQuery does not properly combine distributed collection
statistics, resulting in an IllegalArgumentException during searches.
Originally surfaced in this Elasticsearch issue:
https://g
Yannick Welsch created LUCENE-10582:
---
Summary: CombinedFieldQuery fails with distributed field statistics
Key: LUCENE-10582
URL: https://issues.apache.org/jira/browse/LUCENE-10582
Project: Lucene - C
[
https://issues.apache.org/jira/browse/LUCENE-10579?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robert Muir resolved LUCENE-10579.
--
Resolution: Fixed
> fix smoketester backwards-check to not parse stdout
> ---
[
https://issues.apache.org/jira/browse/LUCENE-10579?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540069#comment-17540069
]
ASF subversion and git services commented on LUCENE-10579:
--
Co
[
https://issues.apache.org/jira/browse/LUCENE-10579?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540070#comment-17540070
]
ASF subversion and git services commented on LUCENE-10579:
--
Co
rmuir merged PR #903:
URL: https://github.com/apache/lucene/pull/903
--
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.apache.or
[
https://issues.apache.org/jira/browse/LUCENE-9625?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540066#comment-17540066
]
Balmukund Mandal commented on LUCENE-9625:
--
Thank you Michael
> Benchmark KNN
dweiss commented on PR #909:
URL: https://github.com/apache/lucene/pull/909#issuecomment-1132763931
I though about it but... Nah. I don't think it makes sense. You can run
tests via gradle from IDEs as well if you have to.
The only thing I was wondering would be whether to throw an assert
[
https://issues.apache.org/jira/browse/LUCENE-10544?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540056#comment-17540056
]
Deepika Sharma commented on LUCENE-10544:
-
For adding timeout feature In {{Inde
uschindler commented on PR #909:
URL: https://github.com/apache/lucene/pull/909#issuecomment-1132736446
Could we not fallback to the classpath code if the forkoptions file system
property is not there?
This would allow to run tests in most jvms. Maybe print a warning then.
--
This
rmuir commented on code in PR #892:
URL: https://github.com/apache/lucene/pull/892#discussion_r877982698
##
lucene/core/src/java/org/apache/lucene/codecs/lucene90/compressing/Lucene90CompressingStoredFieldsWriter.java:
##
@@ -553,14 +554,20 @@ private void copyChunks(
final
rmuir commented on PR #892:
URL: https://github.com/apache/lucene/pull/892#issuecomment-1132725061
Maybe one way to negate risk is to make the same changes to term vectors. It
effectively increases the testing of such changes and at least makes all this
bulk merge logic more consistent betw
[
https://issues.apache.org/jira/browse/LUCENE-10370?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dawid Weiss updated LUCENE-10370:
-
Description:
TestNRTReplication fails because it assumes classpath can just be copied to a
sub
rmuir commented on PR #892:
URL: https://github.com/apache/lucene/pull/892#issuecomment-1132717121
This adds a little more complexity to the scary bulk merge logic, though. is
it worth it?
change is somewhat terrifying. can we do something to make it less so?
--
This is an automat
dweiss commented on PR #901:
URL: https://github.com/apache/lucene/pull/901#issuecomment-1132716807
I filed a PR fixing those "forking" tests (not just the ones above, others
as well). Rewriting the nrt tests completely so that they don't fork at all is
a more complex task and probably dese
dweiss opened a new pull request, #909:
URL: https://github.com/apache/lucene/pull/909
This is putting some lipstick on the pig. Forking those JVMs is really an
awkward thing to do. I understand some tests are specifically testing JVM
crashes, etc., but, uh, those process builders/ pipes ar
[
https://issues.apache.org/jira/browse/LUCENE-10370?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dawid Weiss reassigned LUCENE-10370:
Assignee: Dawid Weiss
> Fix classpath/module path of tests forking their own Java (TestN
jpountz commented on code in PR #897:
URL: https://github.com/apache/lucene/pull/897#discussion_r877887683
##
lucene/core/src/java/org/apache/lucene/document/LatLonPoint.java:
##
@@ -362,4 +377,72 @@ public static Query newDistanceFeatureQuery(
}
return query;
}
+
+
jpountz commented on PR #892:
URL: https://github.com/apache/lucene/pull/892#issuecomment-1132621070
Quadratic merging is gone but this change still looks helpful so I updated
the title/description.
--
This is an automated message from the Apache Git Service.
To respond to the message, pl
jpountz commented on PR #907:
URL: https://github.com/apache/lucene/pull/907#issuecomment-1132617984
Thanks for tackling this! Your are going in the right direction. Could we
drop most of the `if (terms != Terms.EMPTY)` checks and handle them like any
non-null `Terms` instance?
--
This i
[
https://issues.apache.org/jira/browse/LUCENE-10574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17539982#comment-17539982
]
ASF subversion and git services commented on LUCENE-10574:
--
Co
[
https://issues.apache.org/jira/browse/LUCENE-10574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17539980#comment-17539980
]
ASF subversion and git services commented on LUCENE-10574:
--
Co
jpountz merged PR #905:
URL: https://github.com/apache/lucene/pull/905
--
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.apache.
[
https://issues.apache.org/jira/browse/LUCENE-10581?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17539976#comment-17539976
]
Adrien Grand commented on LUCENE-10581:
---
StoredFieldsBenchmark on the entire geon
Adrien Grand created LUCENE-10581:
-
Summary: Optimize stored fields merges on the first segment
Key: LUCENE-10581
URL: https://issues.apache.org/jira/browse/LUCENE-10581
Project: Lucene - Core
[
https://issues.apache.org/jira/browse/LUCENE-10574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17539959#comment-17539959
]
Adrien Grand commented on LUCENE-10574:
---
There was a good speedup on the stored f
82 matches
Mail list logo
