aboutsummaryrefslogtreecommitdiffstats
path: root/config-model/src/main/java/com/yahoo/schema
Commit message (Collapse)AuthorAgeFilesLines
* Dynamic summary fields don't require tokenizer.Tor Egge2023-11-021-16/+2
| | | | | | Tokenization is enabled by the presence of an index expression in the indexing script for the field, and the annotated strings containing the result of tokenization are written to the document field.
* Use "copy" summary transform when field is only defined as summary field.Tor Egge2023-11-011-0/+31
|
* Revert "validate for array/wset attributes"Harald Musum2023-11-011-22/+2
|
* validate for array/wset attributesArne Juul2023-11-011-2/+22
|
* Merge pull request #29172 from ↵Geir Storli2023-10-316-8/+8
|\ | | | | | | | | vespa-engine/toregge/consolidate-on-document-summary Use "document-summary" instead of "summary class" or "document summary".
| * Use "document-summary" instead of "summary class" or "document summary".Tor Egge2023-10-316-8/+8
| |
* | Merge pull request #29167 from vespa-engine/toregge/fixup-copyrightsTor Egge2023-10-311-1/+1
|\ \ | |/ |/| Update copyrights.
| * Update copyrights.Tor Egge2023-10-311-1/+1
| |
* | Add attribute-tokens summary transform to config model.Tor Egge2023-10-315-2/+27
|/
* handle functions with argumentsArne Juul2023-10-301-11/+24
| | | | | * should work as match-features and summary-features now * and they will become hidden/implicit match-feature if used from global-phase
* Validate data type for tokens summary transform.Tor Egge2023-10-272-0/+51
|
* Merge pull request #29096 from vespa-engine/arnej/more-reserved-function-namesLester Solbakken2023-10-271-2/+23
|\ | | | | add some more built-in function names
| * add more reserved functions namesArne Juul2023-10-261-0/+8
| |
| * add some more built-in function namesArne Juul2023-10-251-2/+15
| |
* | Merge pull request #29120 from ↵Tor Egge2023-10-264-71/+83
|\ \ | | | | | | | | | | | | vespa-engine/toregge/adjust-multiple-summary-transforms-in-shared-class Adjust multiple summary transforms in shared class.
| * | Adjust multiple summary transforms in shared class.Tor Egge2023-10-264-71/+83
| | |
* | | keep the implicit (hidden) match features in a separate set in RankProfileArne Juul2023-10-262-12/+9
|/ /
* | avoid losing inherited match-features when adding implicit onesArne Juul2023-10-261-2/+4
| |
* | Don't add summary fields with tokens transform to default summary.Tor Egge2023-10-251-0/+1
|/
* Extend class comment for SummaryFieldTypeResolver.Tor Egge2023-10-241-0/+4
|
* Prepare for emitting warning if summary field type is specified.Tor Egge2023-10-246-37/+140
| | | | Improve resolving of summary field type from source field type.
* Derive summary field type from source.Tor Egge2023-10-191-0/+6
|
* Rename linguistics-tokens to tokens.Tor Egge2023-10-195-8/+8
|
* Non-functional changes onlyJon Bratseth2023-10-181-2/+1
|
* Merge pull request #28919 from vespa-engine/arnej/global-phase-config-7Arne H Juul2023-10-175-18/+289
|\ | | | | allow configuring normalizers
| * allow configuring normalizersArne Juul2023-10-135-18/+289
| |
* | Add linguistics tokens document field writer.Tor Egge2023-10-165-2/+11
|/
* Merge pull request #28836 from vespa-engine/bratseth/copyright-2Jon Bratseth2023-10-10202-195/+202
|\ | | | | Update copyright
| * Update copyrightJon Bratseth2023-10-09202-195/+202
| |
* | Support inheriting multiple document summariesJon Bratseth2023-10-073-8/+5
|/
* Merge pull request #28699 from vespa-engine/bratseth/smarter-inputJon Bratseth2023-09-294-22/+22
|\ | | | | Bratseth/smarter input
| * Remove stacktrace printingJon Bratseth2023-09-291-1/+0
| |
| * Return the expected outputJon Bratseth2023-09-274-22/+23
| | | | | | | | | | | | | | | | | | | | | | In if-else expressions, return the output of the executed branch rather than the input. The current behavior was undocumented and quite unexpected, so I suggest we treat that as a bug. Also return the last executed expression in a script as its output (rather than nothing. In addition, improve some error messages.
* | use equals to compare OptionalInt instancesArne Juul2023-09-281-1/+1
|/
* Merge pull request #28623 from ↵Arne H Juul2023-09-253-13/+22
|\ | | | | | | | | vespa-engine/bratseth/dynamic-summary-ngram-fields Bratseth/dynamic summary ngram fields
| * Use OptionalIntJon Bratseth2023-09-223-13/+15
| |
| * Output gram for summary fields where the source is gramJon Bratseth2023-09-221-1/+8
| |
* | declare types for more rank featuresArne Juul2023-09-211-4/+18
|/
* Merge pull request #28566 from ↵Arne H Juul2023-09-212-0/+9
|\ | | | | | | | | vespa-engine/arnej/avoid-recomputing-match-features if something is already a match-feature, avoid computing it again in …
| * if something is already a match-feature, avoid computing it again in ↵Arne Juul2023-09-182-0/+9
| | | | | | | | global-phase
* | get rid of input which is already gotten from queryArne Juul2023-09-181-3/+1
|/
* Add utility function for custom token idsLester Solbakken2023-09-111-18/+43
|
* Shallow clone is sufficient for onnxmodelLester Solbakken2023-08-301-11/+7
|
* Add cluster specific settings for model evaluationLester Solbakken2023-08-293-2/+30
|
* get line number and contextArne Juul2023-08-241-1/+1
|
* Control the auto-adjustment of targetHits in ANN using post-filtering.Geir Storli2023-08-154-0/+23
| | | | | | | | | | | | | | When searching the HNSW index in a post-filtering case, targetHits is auto-adjusted in an effort to still expose targetHits hits to first-phase ranking after post-filtering. The following formula is now used to ensure an upper bound of adjustedTargetHits, avoiding that the search in the HNSW index takes too long. adjustedTargetHits = min(targetHits / estimatedHitRatio, targetHits * targetHitsMaxAdjustmentFactor). The target-hits-max-adjustment-factor can be set in a rank profile and overriden per query. The value is in the range [1.0,inf], with the default being 20.0. When setting this to 1.0, auto-adjustment of targetHits is effectively disabled.
* Correctly classify references to missing fieldsJon Bratseth2023-08-143-12/+9
|
* Always use stricter validation for complex fields with struct field attributes.Geir Storli2023-08-084-21/+16
|
* Don't scale down if we are likely to scale back upJon Bratseth2023-07-201-4/+4
| | | | | | | | | When we decide to scale number of nodes down we'll see a lower ideal load because we need to be able to handle one node going down. This may lead us to be closer to ideal (at current peak load) than otherwise anticipated, such that we are quite likely to soon scale back up. This checks for that and avoids scaling down dimensions where this is the case.
* Remove unused, stale stuff related to index-toHarald Musum2023-07-081-1/+1
| | | | Mostly in schemas, support remove years ago, no funcation changes