Lucene: Status=Open

[Save this search]

Author relation

Updated ago

Comment count

Reaction count

Review Requested

User

Last comment user

Type

Filters: Status (Open), Issue type, Author relation, Created, Updated, Updated ago, Comment count, Reaction count, Review Requested, Mentioned, Reviewed, Commented, User, Last comment user, Draft, Component, Type, Labels, Commits?, Reporter, Assignee

#16660 PR: Skip blocks without decompressing them when reading selected stored fields.

1.7 minutes ago 0 comments 0 votes 0 watches vsop-479

Description

#16540 PR: Store doc values disi data into a separate file

29.6 minutes ago 5 comments 0 votes 0 watches costin, github-actions[bot], romseygeek, sgup432

Description This moves out the doc values disi data (used to store sparse data) into a separate ... Closes - https://github.com/apache/lucene/issues/16298 <!-- If this is your first contribution to Lucene, please make sure you have reviewed the ...

romseygeek 29.6 minutes ago:	Thanks @sgup432, this looks great. Let's move the CHANGES entry to 10.6, and I will merge and backport.
romseygeek 31.2 minutes ago:	This can go into 10.6

#16030 PR: Implement data-blind scalar quantization

4.6 hours ago 27 comments 0 votes 0 watches github-actions[bot], kaivalnp, mccullocht, mikemccand, romseygeek

Add an option to the quantization format to enable or disable centering (enabled by default). When centering is disabled we also stop writing the float vectors which can lead to significant ... Special handling is included during merges -- we check that all of the input is in the same ...

mccullocht 1.5 days ago:	It's not part of the automated indexing flow but IIUC there is already support for truncating the ... At 4 or 8 bits this is almost always pretty reasonable.
mccullocht 1.5 days ago:	This file has been deleted.

#16659 PR: ci: use faster arm runners when we don't care

6.3 hours ago 2 comments 0 votes 0 watches rmuir

For many of these jobs the execution time might not matter, but the lower queue time gives a less ... This switches over remainder of CI jobs to arm runner, although the labeller job i switched to slim ... This doesn't change build matrix, instead just jobs that were only running on x86 (since e.g. they ...

rmuir 6.3 hours ago:	The labeller job won't cut over on a PR, since its a pull_request_target. change has to actually ... I switched over the new intranet mirror job (was a bit behind). So that just leaves the two on ubuntu-latest: * explicitly tested in the build matrix * faiss ...
rmuir 7 hours ago:	This is kinda what I mean, see all the ubuntu-latest jobs queued, yet everything else done. Good to use the faster computer if you really don't care. <img width="993" height="440" alt="Screen_Shot_2026-09-09_at_22 36 17" src="https://github.com/ ...

#16567 PR: Add ivfaster: a segment-lifecycle-aware IVF vector codec for the sandbox

7.1 hours ago 3 comments 0 votes 0 watches RKSPD, github-actions[bot], gsmiller, msokolov

IVFaster: a segment-lifecycle-aware IVF vector codec (sandbox) It's no secret that vector search ... Segments are immutable and merge constantly, and both properties are cheap for an inverted index ... There have been attempts to improve this behavior, namely IncrementalHnswGraphMerger, which adopts ...

gsmiller 12.5 days ago:	Disclaimer: Rikhil and I work together and started a conversation about this earlier today, but I ... A property that really excites me about IVF is that docIDs can be traversed in monotonically ... I find this compelling since it means docs could potentially be evaluated doc-at-a-time along with ...
RKSPD 13.6 days ago:	> exciting! the numbers you shared look compelling. Can you address the failing checks?

#16657 PR: Fix hashCode/equals contract violation in LRUQueryCache QueryCacheKey

7.7 hours ago 0 comments 0 votes 0 watches github-actions[bot], sgup432

Description LRUQueryCache was recently refactored(releasing in lucene 11.0) to partition the ... I found out a bug in this QueryCacheKey where this compares queries by value in equals() but hashes ... For example: we cache new TermQuery("color","red"), then look it up with another new TermQuery(" ...

#16658 PR: ci: switch to arm runner for codeql jobs

7.8 hours ago 1 comments 0 votes 0 watches rmuir

arm runners are usually a bit faster and its the same cost. For example, for our test jobs, the avg runtime is 4:50 vs x86 average of 6:42 codeql Analyze( ... Let's cut it over now that arm is supported.

rmuir 7.8 hours ago:

Shaves 2 minutes off the run it seems

«
1
2
3
4
5
»

2,656 issues

[15.9 msec search, 16.9 msec total]