rmuir opened a new pull request, #14318:
URL: https://github.com/apache/lucene/pull/14318
These scripts are important, but have zero static analysis, type-checking,
or formatting.
Add simple make-based build with CI hook for this.
Format check is not yet enabled, we should "mak
rmuir commented on PR #14318:
URL: https://github.com/apache/lucene/pull/14318#issuecomment-2692256799
If we are ok with it, I would want to backport these changes and cleanups to
10.x branch, too, so that things start to become easier.
--
This is an automated message from the Apache Git
ChrisHegarty opened a new pull request, #14320:
URL: https://github.com/apache/lucene/pull/14320
This commit changes DirectIOIndexInput::seek so that it repositions the
indexInput's position without a read, when the new position is within the
bounds of the current buffer.
Prior to th
ChrisHegarty commented on issue #14315:
URL: https://github.com/apache/lucene/issues/14315#issuecomment-2692428359
You found the bug @rmuir - when seeking, we're always reading from the file
even when the new position is within the bounds of the current buffer. I raised
this small PR #14320
rmuir commented on issue #14315:
URL: https://github.com/apache/lucene/issues/14315#issuecomment-2691557207
When I run this seed on my system I see:
```
:lucene:misc:test (SUCCESS): 63 test(s), 2 skipped
The slowest tests (exceeding 500 ms) during this run:
4.86s TestDirectIODi
gaoj0017 commented on PR #14078:
URL: https://github.com/apache/lucene/pull/14078#issuecomment-2692580673
Thank you for acknowledging that our extended RaBitQ method proposes the
idea of exploring different scalar quantization parameters on a per-vector
basis for the first time and OSQ adop
mkhludnev merged PR #14317:
URL: https://github.com/apache/lucene/pull/14317
--
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
rmuir commented on issue #14315:
URL: https://github.com/apache/lucene/issues/14315#issuecomment-2692076156
From the strace output it seems to me like, we may not be buffering
properly? so we keep re-reading the entire buffer over and over? maybe we
aren't cloning our buffer on slice() and
mkhludnev merged PR #14276:
URL: https://github.com/apache/lucene/pull/14276
--
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
dweiss commented on issue #14316:
URL: https://github.com/apache/lucene/issues/14316#issuecomment-2692160562
It's in .github/workflows/run-nightly-smoketester.yml, but the python
scripts may also require a manual review.
--
This is an automated message from the Apache Git Service.
To resp
Brijeshthummar02 commented on issue #14316:
URL: https://github.com/apache/lucene/issues/14316#issuecomment-2692137269
@dweiss i can fix it can you provide path to the yml file.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
dweiss closed issue #14316: Python release checker scripts need an update to
use Java 23
URL: https://github.com/apache/lucene/issues/14316
--
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 specif
dweiss closed issue #14305: develocity build scans fail to upload sometimes
URL: https://github.com/apache/lucene/issues/14305
--
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 #14318:
URL: https://github.com/apache/lucene/pull/14318#issuecomment-2692327772
I will followup with changes to iterate. mainly want to detect common
mistakes as a start, then it won't be so terrifying to change or review these
files
--
This is an automated message
rmuir merged PR #14318:
URL: https://github.com/apache/lucene/pull/14318
--
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 issue #14315:
URL: https://github.com/apache/lucene/issues/14315#issuecomment-2691142322
I'd expect that BaseDirectoryTestCase is too intense for direct-io. This
test has a lot of randomization like what you are seeing, and none of the IO
will be cached on linux.
I
rmuir opened a new pull request, #14319:
URL: https://github.com/apache/lucene/pull/14319
Fix basic errors from linter and get undefined variable analysis working
through the type-checker.
This will detect common problems such as typos, instead of at runtime. The
changes are straight
rmuir commented on issue #14315:
URL: https://github.com/apache/lucene/issues/14315#issuecomment-2691684579
easy to reproduce on linux if you want to dive in more:
terminal 1: for monitoring whole system and locating PID
```console
$ sudo perf top --sort overhead,overhead_us,over
18 matches
Mail list logo
