aboutsummaryrefslogtreecommitdiffstats
path: root/searchlib/src/vespa/searchlib/attribute
Commit message (Expand)AuthorAgeFilesLines
* Add noexceptHenning Baldersheim2024-04-252-17/+12
* Share code to calculate FlowStats based on attribute::HitEstimate.Geir Storli2024-04-232-48/+18
* Share code to calculate FlowStats based on IDirectPostingStore::LookupResult.Geir Storli2024-04-233-30/+34
* Use explicit and do not expose nbostream in headerfile.Henning Baldersheim2024-04-221-0/+1
* Merge pull request #30976 from vespa-engine/vekterli/wire-fuzzy-prefix-match-...Henning Baldersheim2024-04-223-10/+18
|\
| * Wire fuzzy prefix matching support through the query stackTor Brede Vekterli2024-04-193-10/+18
* | Use better non-strict cost estimates for btree and disk index iterators.Geir Storli2024-04-192-5/+7
|/
* Adjust strict cost of intermediate / complex leaf blueprints.Geir Storli2024-04-151-2/+2
* Re-adjust cost model params for IN/weightedSet/dotProduct.Geir Storli2024-04-152-22/+31
* Adjust cost model params for IN/weightedSet/dotProduct after benchmarking.Geir Storli2024-04-121-2/+4
* Merge pull request #30835 from vespa-engine/geirst/more-than-64k-child-iterat...Henning Baldersheim2024-04-052-15/+35
|\
| * Support more than 64k child iterators for IN and weightedSet.Geir Storli2024-04-052-15/+35
* | force strictness where allowed and cheaperHåvard Pettersen2024-04-043-5/+5
|/
* remove array_cost since all strict ORs are now heapsHåvard Pettersen2024-04-021-1/+1
* use thread_local for blueprint optionsHåvard Pettersen2024-03-254-6/+6
* tag blueprints with strictnessHåvard Pettersen2024-03-2015-56/+78
* stop using non-heap strict ORHåvard Pettersen2024-03-131-5/+13
* Integrate flow tuning for attribute, memory and disk index search.Geir Storli2024-03-052-8/+34
* Add noexcept and use emplaceHenning Baldersheim2024-02-291-6/+2
* Use auto, braced initialiser and noexceptHenning Baldersheim2024-02-191-5/+4
* Merge pull request #30296 from vespa-engine/toregge/add-search-predicate-attr...Geir Storli2024-02-165-22/+110
|\
| * Add search::PredicateAttributeSaver.Tor Egge2024-02-165-22/+110
* | Check buffer size when loading predicate attribute.Tor Egge2024-02-161-0/+6
|/
* Add search::predicate::PredicateIndexSaver.Tor Egge2024-02-151-2/+9
* Add search::predicate::SimpleIndexSaver.Tor Egge2024-02-141-19/+1
* baseline flow stats for complex leafsHåvard Pettersen2024-02-134-9/+80
* Add extra logging when saving predicate attribute.Tor Egge2024-02-121-1/+19
* Add noexcept, remove virtual and use = defaultHenning Baldersheim2024-02-122-13/+7
* Use smaller buffer for reading file headers. And GC some unused code.Henning Baldersheim2024-02-092-2/+2
* make default flow stats more explicitHåvard Pettersen2024-02-064-0/+23
* Align flow stats calculation for leaf blueprints.Geir Storli2024-02-051-3/+15
* Tag hit estimates from attribute search contexts as unknown when applicable.Geir Storli2024-02-0214-46/+47
* Include algorithm to get declarations of std::adjacent_find and std::sortTor Egge2024-01-291-0/+1
* Support IDocidWithWeightPostingStore for more attribute data types.Geir Storli2024-01-244-54/+33
* Remove optimization for single term searching in wset string/int attribute.Geir Storli2024-01-231-74/+2
* Revert "Support IDocidWithWeightPostingStore for more attribute data types."Henning Baldersheim2024-01-224-34/+49
* Support IDocidWithWeightPostingStore for more attribute data types.Geir Storli2024-01-224-49/+34
* Support hash filter iterator in DirectMultiTermBlueprint.Geir Storli2024-01-195-9/+63
* Merge pull request #29964 from vespa-engine/vekterli/use-string-view-for-non-...Tor Egge2024-01-183-6/+17
|\
| * Use `string_view` for standalone DFA fuzzy match functionTor Brede Vekterli2024-01-183-6/+17
* | Stop using MultiTermOrFilterSearch for InTerm and WeightedSetTerm.Geir Storli2024-01-184-26/+17
* | Merge pull request #29660 from vespa-engine/balder/sliced-parallell-orv8.288.15Henning Baldersheim2024-01-161-2/+6
|\ \
| * | Merge branch 'master' into balder/sliced-parallell-orHenning Baldersheim2024-01-1514-120/+226
| |\|
| * | Slice the vectors and use 1 thread per slice when computing the OR.Henning Baldersheim2023-12-141-2/+6
* | | Rename MultiTermFilter to MultiTermHashFilter.Geir Storli2024-01-163-13/+13
| |/ |/|
* | Reapply Use optimized implementation for IN query operator.Geir Storli2024-01-121-22/+32
* | Cache docids from iterators to reduce cache misses.Geir Storli2024-01-111-3/+12
* | Revert "Use optimized implementation for IN query operator."Henning Baldersheim2024-01-061-32/+22
* | Use optimized implementation for IN query operator.Geir Storli2024-01-051-22/+32
* | Support the IN query operator in DirectMultiTermBlueprint.Geir Storli2024-01-054-16/+47