Mikep86 commented on code in PR #13697:
URL: https://github.com/apache/lucene/pull/13697#discussion_r1735379356
##
lucene/join/src/java/org/apache/lucene/search/join/ToParentBlockJoinQuery.java:
##
@@ -440,6 +500,83 @@ private String formatScoreExplanation(int matches, int
star
Mikep86 commented on code in PR #13697:
URL: https://github.com/apache/lucene/pull/13697#discussion_r1735376518
##
lucene/join/src/test/org/apache/lucene/search/join/TestBlockJoinBulkScorer.java:
##
Review Comment:
I plan on adding another test to check that the min competi
Mikep86 commented on code in PR #13697:
URL: https://github.com/apache/lucene/pull/13697#discussion_r1735374017
##
lucene/join/src/test/org/apache/lucene/search/join/TestBlockJoinBulkScorer.java:
##
@@ -0,0 +1,264 @@
+package org.apache.lucene.search.join;
+
+import com.carrotse
Mikep86 commented on code in PR #13697:
URL: https://github.com/apache/lucene/pull/13697#discussion_r1735373453
##
lucene/join/src/test/org/apache/lucene/search/join/TestBlockJoinBulkScorer.java:
##
@@ -0,0 +1,264 @@
+package org.apache.lucene.search.join;
+
+import com.carrotse
Mikep86 commented on code in PR #13697:
URL: https://github.com/apache/lucene/pull/13697#discussion_r1735372582
##
lucene/join/src/test/org/apache/lucene/search/join/TestBlockJoinBulkScorer.java:
##
@@ -0,0 +1,264 @@
+package org.apache.lucene.search.join;
+
+import com.carrotse
Mikep86 opened a new pull request, #13697:
URL: https://github.com/apache/lucene/pull/13697
(no comment)
--
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-m
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2316262279
> > after having a shower in my German bathroom
>
> I'm sure all of us can't stop picturing this German bathroom, Uwe! :)
https://x.com/UweSays/status/915396168506449
dweiss commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2316189280
This is a regression from adding incremental builds. These tasks have side
effects and should always execute.
--
This is an automated message from the Apache Git Service.
To respo
dweiss commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2316191913
> after having a shower in my German bathroom
I'm sure all of us can't stop picturing this German bathroom, Uwe! :)
--
This is an automated message from the Apache Git Servic
uschindler merged PR #13696:
URL: https://github.com/apache/lucene/pull/13696
--
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 closed issue #13695: Gradle build sometimes gives spurious
"unreferenced license file" warnings
URL: https://github.com/apache/lucene/issues/13695
--
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 t
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315998879
@mikemccand: Beer is ready and after having a shower in my German bathroom
the hot day can end!
--
This is an automated message from the Apache Git Service.
To respond to the m
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315983388
Here is the PR: #13696
I added a comment to the top of this file and a TODO to explain that all
tasks in this file must always run and therefore cannot define any outputs.
uschindler opened a new pull request, #13696:
URL: https://github.com/apache/lucene/pull/13696
Revert changes by #12150 in jar-checks.gradle, because tasks in this file
shares internal state between tasks without using files. Because of this all
tasks here must always execute together, so t
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315959316
It's fast. I see no delay.
--
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 spe
mikemccand commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315958533
+1 to revert that part of the optimization (no incremental building of
license checking) for now?
How slow is the full check?
--
This is an automated message from the Ap
uschindler commented on PR #12150:
URL: https://github.com/apache/lucene/pull/12150#issuecomment-2315957699
Unfortunately the changes applied here for the file `jar-checks.gradle` (the
new build outputs) cause random build warnings, see
https://github.com/apache/lucene/issues/13695.
mikemccand commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315956877
Wow, thank you for digging so deep @uschindler! Maybe your fridge has beers
again!
--
This is an automated message from the Apache Git Service.
To respond to the message, plea
mikemccand commented on PR #13521:
URL: https://github.com/apache/lucene/pull/13521#issuecomment-2315947436
> @mikemccand -- did you mean to close this PR?
Ugh, no I did not! Sorry, I'll reopen!!
--
This is an automated message from the Apache Git Service.
To respond to the message
expani opened a new pull request, #13521:
URL: https://github.com/apache/lucene/pull/13521
### Background
Lucene uses 3 different ways of storing the docIds in KDD file of a BKD Tree
based index if the docIds in a leaf block are not sorted :
- If the difference b/w min and max
msfroh commented on PR #13521:
URL: https://github.com/apache/lucene/pull/13521#issuecomment-2315946796
@mikemccand -- did you mean to close this PR?
If @expani adds the unrolled version (`Bit21With3StepsEncoder`) to
`DocIdsWriter`, the change should be ready to go, IMO.
Even o
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315945766
If I comment out those 2 lines, it works:
https://github.com/apache/lucene/blob/d55b92bae01be46656c4d373d8cb1d081b280469/gradle/validation/jar-checks.gradle#L201-L202
So
mikemccand commented on PR #13521:
URL: https://github.com/apache/lucene/pull/13521#issuecomment-2315942081
> We can replace it with Bit21With2StepsEncoder in future when the
performance is comparable to x86.
I wonder what mechanism we could use to remind ourselves when performance of
mikemccand commented on PR #13521:
URL: https://github.com/apache/lucene/pull/13521#issuecomment-2315935834
> @jpountz IMO We should use `Bit21With3StepsEncoder` in DocIdsWriter as
using `Bit21With2StepsEncoder` might lead to performance regression for
workloads in aarch64 platforms.
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315935860
OK, I found the bug. The comment explains it:
https://github.com/apache/lucene/blob/d55b92bae01be46656c4d373d8cb1d081b280469/gradle/validation/jar-checks.gradle#L138-L140
mikemccand closed pull request #13521: Introduce new encoding of BPV 21 for
DocIdsWriter used in BKD Tree
URL: https://github.com/apache/lucene/pull/13521
--
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 g
msfroh commented on PR #13521:
URL: https://github.com/apache/lucene/pull/13521#issuecomment-2315907724
Okay -- I was able to speed up the SIMD implementation a fair bit. Honestly,
my main stupid mistake was that I hadn't declared `LONG_SPECIES` as `static
final`, which probably prevented s
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315903858
OK, I debugged a bit:
```
>gradlew :lucene:checkDanglingLicenseFiles
Starting a Gradle Daemon (subsequent builds will be faster)
> Task :errorProneSkipped
WA
risdenk commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315812110
Solr has the same issue and it was reported here
https://issues.apache.org/jira/browse/SOLR-17142 - there was a mailing list
post where I found the culprit but I haven't had a chanc
jpountz merged PR #13692:
URL: https://github.com/apache/lucene/pull/13692
--
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
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315672592
P.S.: I did not open an issue because the last time I was confronted with it
(about 1 year ago), I was at the mentag stage of "damn fcking gradle - I need
more beers", but my fri
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315667317
Maybe it can be solved by making the "ext" property a build output and let
all tasks depend on it.
--
This is an automated message from the Apache Git Service.
To respond to th
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315658433
Let's ask @dweiss, I gave up on this, as my Gradle-Foo is too limited,
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to G
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315656656
I think to fix this, the `collectJarInfos` must write that information to a
monitored file in the build outputs.
All pother tasks using that information must read the outpu
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315647367
> Just curious - what is the purpose of the check? I guess it sounds
self-explanatory: if there is a license file thatisn't referenced (where?) we
would presumably want to clean
msokolov commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315623970
Just curious - what is the purpose of the check? I guess it sounds
self-explanatory: if there is a license file thatisn't referenced (where?) we
would presumably want to clean it u
uschindler commented on issue #13695:
URL: https://github.com/apache/lucene/issues/13695#issuecomment-2315617744
I alos noticed this several times. I also reported an issue to Dawid, but
the cause for this is very strage. It depends on the order of execution.
I used to know the cause
mikemccand opened a new issue, #13695:
URL: https://github.com/apache/lucene/issues/13695
### Description
On `main` today, with Java 21 on a recent Arch Linux, if I run `./gradlew
clean` then `./gradlew check`, all is good. It passes.
But if I run `./gradlew check` again, I so
mikemccand merged PR #13694:
URL: https://github.com/apache/lucene/pull/13694
--
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.
mikemccand commented on code in PR #13694:
URL: https://github.com/apache/lucene/pull/13694#discussion_r1734770769
##
build-tools/build-infra/src/main/java/org/apache/lucene/gradle/ProfileResults.java:
##
@@ -156,7 +156,12 @@ private static String pad(String input) {
/** Pr
gsmiller commented on issue #13671:
URL: https://github.com/apache/lucene/issues/13671#issuecomment-2315408209
@epotyom right makes sense. I should have been a little more specific. When
I said "new method" I was thinking along the same lines of changing the
recently-added method to not nee
gsmiller commented on code in PR #13692:
URL: https://github.com/apache/lucene/pull/13692#discussion_r1734731228
##
lucene/core/src/java/org/apache/lucene/codecs/lucene912/Lucene912PostingsReader.java:
##
@@ -212,13 +213,74 @@ static void prefixSum(long[] buffer, int count, long
jpountz commented on code in PR #13692:
URL: https://github.com/apache/lucene/pull/13692#discussion_r1734653387
##
lucene/core/src/java/org/apache/lucene/codecs/lucene912/Lucene912PostingsReader.java:
##
@@ -212,13 +213,74 @@ static void prefixSum(long[] buffer, int count, long
msokolov commented on code in PR #13694:
URL: https://github.com/apache/lucene/pull/13694#discussion_r1734623994
##
build-tools/build-infra/src/main/java/org/apache/lucene/gradle/ProfileResults.java:
##
@@ -156,7 +156,12 @@ private static String pad(String input) {
/** Proc
mikemccand opened a new pull request, #13694:
URL: https://github.com/apache/lucene/pull/13694
I was trying to debug whether the distance metric I had requested using
`luceneutil`'s `knnPerfTest.py` was properly working its way all the way down
that stack, and found a few `toString` methods
mikemccand commented on issue #13671:
URL: https://github.com/apache/lucene/issues/13671#issuecomment-2315164006
I added `blocker` label to this and removed `type:enhancement` (it seemed
super weird to have an enhancement blocker!).
--
This is an automated message from the Apache Git Serv
mikemccand commented on issue #13519:
URL: https://github.com/apache/lucene/issues/13519#issuecomment-2315154624
I just tested KNN recall using [`knnPerfTest.py` from
`luceneutil`](https://github.com/mikemccand/luceneutil/blob/main/src/python/knnPerfTest.py)
on 4, 7, 8 bit quantization, and
stefanvodita merged PR #13690:
URL: https://github.com/apache/lucene/pull/13690
--
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...@lucen
epotyom commented on issue #13671:
URL: https://github.com/apache/lucene/issues/13671#issuecomment-2314681028
@gsmiller , there is new method
```java
public void DrillSideways#search(
DrillDownQuery query,
CollectorOwner drillDownCollectorOwner,
List> drillSidewaysCol
49 matches
Mail list logo
