Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
36,196 Commits 27 Branches 467 Tags 751 MiB
Commit Graph

36196 Commits

Author SHA1 Message Date
Julie Tibshirani 187f843e2a
LUCENE-10577: Add vectors format unit test and fix toString (#998) 
We forgot to add this unit test when introducing the new 9.3 vectors format.
This commit adds the test and fixes issues it uncovered in toString.
 2022-07-02 19:00:29 +02:00
Uwe Schindler fb617e29d0 Remove deprecations in main (#978) 2022-07-01 16:32:50 +02:00
Uwe Schindler 8a70988e63
Remove/deprecate obsolete constants in oal.util.Constants; remove code which is no longer executed after Java 9 (#978) 2022-07-01 16:24:18 +02:00
Greg Miller 5f2a4998a0
LUCENE-10603: Migrate remaining SSDV iteration to use docValueCount in production code (#995) 2022-06-30 14:01:14 -07:00
Greg Miller e05b3ec7de Add missing CHANGES entries for GH#983 and GH#984 2022-06-29 13:08:01 -07:00
Greg Miller bede1c3e8c
Switch Float/IntTaxonomyFacets to primitive list data structures in getAllChildren (#984) 2022-06-29 13:01:59 -07:00
Greg Miller 5c3d92d69d
Some refactoring/cleanup of AbstractSortedSetDocValueFacetCounts (#983) 2022-06-29 13:01:47 -07:00
Michael Sokolov e078bc1cd9 fix minor style issues in test 2022-06-29 11:09:46 -04:00
Michael Sokolov 95de554b65 CHANGES entry for LUCENE-10151 2022-06-29 10:35:09 -04:00
Deepika0510 af05550ebf
LUCENE-10151: Adding Timeout Support to IndexSearcher (#927) 
Authored-by: Deepika Sharma <dpshrma@amazon.com>
 2022-06-29 10:32:12 -04:00
Tomoko Uchida 64321114e1
LUCENE-10557: temprarily enable github issue (#988) 2022-06-29 13:27:53 +09:00
Dawid Weiss 5328a70a29
Update randomizedtesting to 2.8.0, hppc to 0.9.1, morfologik to 2.1.9. (#991) 2022-06-28 15:42:07 +02:00
Alessandro Benedetti 8cf694fed2
LUCENE-10593: VectorSimilarityFunction reverse removal (#926) 
* Vector Similarity Function reverse property removed

* NeighborQueue tie-breaking fixed (node id + node score encoding)

* NeighborQueue readability refactor

* BoundChecker removal (now it's only in backward-codecs)
 2022-06-28 15:33:11 +02:00
Lu Xugang 3e74ebbc0d
Add entry (#990) 2022-06-28 12:03:18 +08:00
Lu Xugang d8fb47b674
LUCENE-10623: Error implementation of docValueCount for SortingSortedSetDocValues (#967) 2022-06-28 11:49:47 +08:00
Julie Tibshirani 7b58088bd5
Fix FieldExistsQuery rewrite when all docs have vectors (#986) 
Before we were checking the number of vectors in the segment against the total
number of documents in IndexReader. This meant FieldExistsQuery would not
rewrite to MatchAllDocsQuery when there were multiple segments.
 2022-06-27 09:59:53 +02:00
Kaival Parikh 03846b468e
LUCENE-10606: For KnnVectorQuery, optimize case where filter is backed by BitSetIterator (#951) 
Instead of collecting hit-by-hit using a `LeafCollector`, we break down the
search by instantiating a weight, creating scorers, and checking the underlying
iterator. If it is backed by a `BitSet`, we directly update the reference (as
we won't be editing the `Bits`). Else we can create a new `BitSet` from the
iterator using `BitSet.of`.
 2022-06-27 08:52:52 +02:00
Shai Erera 9338909373
Fix typos and minor refactoring to FacetConfig (#982) 2022-06-27 07:34:21 +03:00
Shai Erera ef486ea11a Use TestUtil.nextInt() instead of random().nextInt() 2022-06-25 18:13:38 +03:00
Marc D'mello f6bb9d218c
LUCENE-10274: Add FacetSets faceting capabilities (#841) 
Co-authored-by: Marc D'Mello <dmellomd@amazon.com>
Co-authored-by: Shai Erera <serera@gmail.com>
Co-authored-by: Greg Miller <gsmiller@gmail.com>
 2022-06-25 17:27:11 +03:00
Uwe Schindler eafc6420f3
Exclude Lucene's own JAR files from classpath entries in Eclipse config (#976) 2022-06-24 19:42:42 +02:00
Christine Poerschke 9199d48e7e
Remove outdated comment in UnifiedHighlighter.get(Formatter|Scorer) javadoc. (#820) 2022-06-24 18:00:28 +01:00
Luca Cavanna 2ccec3deb7
Rework TestElevationComparator (#980) 
The test used to leave hanging threads behind following a failure. Also one method was executing two different tests. I split the existing method into two and I am now leveraging setup and teardown to properly close all the resources both when the tests succeed as well as whey they fail.
 2022-06-24 18:11:51 +02:00
Uwe Schindler 831ae2465b
Remove the deflater hack introduced because of JDK-8252739 (#977) 
This bug was fixed in JDK-16 so it does not apply to main branch (Java 17) anymore.
 2022-06-24 14:23:10 +02:00
Adrien Grand 4c1ae2a332
LUCENE-10620: Pass the Weight to Collectors. (#964) 
This allows `Collector`s to use `Weight#count` when appropriate.
 2022-06-23 17:56:15 +02:00
tang donghai 2da9951f23
LUCENE-10607: Fix potential integer overflow in maxArcs computions (#970) 2022-06-22 22:23:36 +02:00
Gautam Worah bc72a53efc
Index only the docs for FacetField posting list (#922) 2022-06-22 12:09:56 -07:00
Mike Drob 3ca108fb73
Add back-compat indices for 8.11.2 2022-06-22 10:29:58 -07:00
Mike Drob 05b1465d36
Sync CHANGES for 8.11.2 2022-06-21 22:59:01 -07:00
Mike Drob 4eb53df424
Add bugfix version 8.11.2 2022-06-21 22:17:51 -07:00
Yuting Gan bdcb4b3716
LUCENE-10550: Add getAllChildren functionality to facets (#914) 2022-06-21 16:48:50 -07:00
Mike Drob c51c86c5a1
DOAP 8.11.2 2022-06-21 14:23:33 -07:00
Greg Miller 8f459eb0f9
LUCENE-10603: Mark SortedSetDocValues#NO_MORE_ORDS deprecated (#969) 2022-06-21 09:05:15 -07:00
Adrien Grand 68f77dbf53
Handle more cases in `BooleanWeight#count`. (#961) 
As suggested by @zhaih on #950, we could support more cases in
`BooleanWeight#count`. This PR adds support for these cases specifically:
 - Pure disjunctions where only one clause has a non-zero count.
 - Pure disjunctions where one clause matches all docs.
 - Negations where positive clauses match all docs (pure negation).
 - Negations where positive clauses match no docs.
 - Negations where negative clauses match no docs.
 - Negations where negative clauses match all docs.
 2022-06-21 17:29:43 +02:00
Adrien Grand adcf58fe87 LUCENE-10507: Fix test failure. 2022-06-21 09:15:10 +02:00
JoeHF 08a9dfddeb
LUCENE-10618: Implement BooleanQuery rewrite rules based for minimumShouldMatch (#965) 2022-06-20 14:04:38 +02:00
Greg Miller 6ba759df86
LUCENE-10603: Change iteration methodology for SSDV ordinals in the faceting module (#954) 2022-06-17 10:37:24 -07:00
Adrien Grand b180a8a97e LUCENE-10078: Fix more TestIndexWriterWithThreads failures. 2022-06-17 17:54:26 +02:00
Luca Cavanna e78b2f2fe0
Make it more likely to perform concurrent search in tests (#959) 
1) Replace default useThreads value: rarely() -> randomBoolean()
2) apply lower slices thresholds more frequently: randomBoolean() -> frequently
3) lower maxDocsPerSlice and maxSegmentsPerSlice threshold when applied
4) apply lower maxSegments and maxSegmentsPerSlice also when wrapWithAssertions is true
 2022-06-17 17:27:08 +02:00
Julie Tibshirani 450ee81154 LUCENE-10611: Tweak the CHANGES description 2022-06-16 12:04:47 -07:00
Kaival Parikh 6df6cb093c
LUCENE-10611: Fix Heap Error in HnswGraphSearcher (#958) 
The HNSW graph search does not consider that visitedLimit may be reached in the
upper levels of graph search itself

This occurs when the pre-filter is too restrictive (and its count sets the
visitedLimit). So instead of switching over to exactSearch, it tries to pop
from an empty heap and throws an error.

We can check if results are incomplete after searching in upper levels, and
break out accordingly. This way it won't throw heap errors, and gracefully
switch to exactSearch instead
 2022-06-16 12:01:54 -07:00
Lu Xugang 78b7b17f93
LUCENE-10600: SortedSetDocValues#docValueCount should be an int, not long (#960) 2022-06-16 12:22:05 +08:00
Greg Miller 0b5e0bfa4f
LUCENE-10584: Properly support #getSpecificValue for hierarchical dims in SSDV faceting (#929) 2022-06-15 11:31:23 -07:00
Adrien Grand d52122605d LUCENE-10608: Ignore conjunctions that have a non-zero minShouldMatch from the count logic. 2022-06-15 10:13:22 +02:00
Adrien Grand 7bbc1bf28d LUCENE-10078, LUCENE-10617: Tentative fix for TestMergeSchedulerExternal. 2022-06-15 09:45:47 +02:00
Adrien Grand 3be18bc62e LUCENE-10078: Fix TestIndexWriterOutOfFileDescriptors failures due to merge-on-refresh. 2022-06-15 09:20:21 +02:00
Julie Tibshirani 89dbe651cc
Make sure KnnVectorQuery applies search boost (#956) 
Before, the rewritten query DocAndScoreQuery ignored the boost.
 2022-06-14 08:34:02 -07:00
Adrien Grand 83461601ad
LUCENE-10608: Implement Weight#count on pure conjunctions. (#950) 2022-06-14 09:41:38 +02:00
Lu Xugang 7504b0a258
LUCENE-10598: Use count to record docValueCount similar to SortedNumericDocValues did (#942) 2022-06-14 13:03:19 +08:00
Dawid Weiss 76d418676e LUCENE-10613: Clean up outdated NOTICE.txt information concerning morfologik 2022-06-13 19:50:54 +02:00