aboutsummaryrefslogtreecommitdiffstats
path: root/config-model/src/main/java/com/yahoo/schema/processing
Commit message (Collapse)AuthorAgeFilesLines
* Merge pull request #28836 from vespa-engine/bratseth/copyright-2Jon Bratseth2023-10-1073-72/+73
|\ | | | | Update copyright
| * Update copyrightJon Bratseth2023-10-0973-72/+73
| |
* | Support inheriting multiple document summariesJon Bratseth2023-10-071-1/+1
|/
* Remove stacktrace printingJon Bratseth2023-09-291-1/+0
|
* Return the expected outputJon Bratseth2023-09-273-17/+19
| | | | | | | | | | | 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 OptionalIntJon Bratseth2023-09-221-4/+2
|
* Correctly classify references to missing fieldsJon Bratseth2023-08-141-5/+5
|
* Always use stricter validation for complex fields with struct field attributes.Geir Storli2023-08-081-2/+2
|
* Remove unused, stale stuff related to index-toHarald Musum2023-07-081-1/+1
| | | | Mostly in schemas, support remove years ago, no funcation changes
* Only log warning when the stricter validation triggers.Geir Storli2023-07-041-2/+2
| | | | | The stricter validation will eventually be made default after all cloud applications are upgraded.
* Classify missing node tag for cloud prod as illegal argumentJon Bratseth2023-05-161-4/+2
|
* Add vmsfields config for nearest neighbor search on supported tensor fields.Geir Storli2023-04-141-0/+9
|
* Replace reflection by visitorJon Bratseth2023-03-311-1/+1
|
* Add raw attributes to config model.Tor Egge2023-03-172-5/+6
|
* wire global-phase through config-model exactly like second-phaseArne Juul2023-02-032-0/+4
|
* Update comment for DynamicSummaryTransformUtils.getSource.Tor Egge2023-01-271-2/+4
|
* Use original source for dynamic summary transform.Tor Egge2023-01-271-3/+15
|
* Allow hnsw index for tensors with 1 mapped and 1 indexed dimension.Geir Storli2023-01-131-2/+11
|
* Revert "Revert "Let list handling catch up with Java 17""Henning Baldersheim2022-12-021-9/+6
|
* Revert "Let list handling catch up with Java 17"Bjørn Christian Seime2022-12-021-6/+9
|
* Let list handling catch up with Java 17Henning Baldersheim2022-12-021-9/+6
|
* Use well defined order where we output text and generate config. Makes ↵Henning Baldersheim2022-12-011-1/+1
| | | | config stable and simple tests predictable.
* Optimize importsHarald Musum2022-10-241-3/+1
|
* Allow paged setting for tensor attributes without fast-rank setting.Tor Egge2022-10-101-4/+4
|
* Merge pull request #24285 from ↵Jon Marius Venstad2022-10-031-1/+1
|\ | | | | | | | | vespa-engine/bratseth/optimize-primitives-only-take-2 Bratseth/optimize primitives only take 2
| * Only check types when necessaryJon Bratseth2022-10-031-1/+1
| |
* | Update terminologyJon Bratseth2022-10-031-3/+3
|/
* Do not log disk access warning when using an imported complex field in summary.Geir Storli2022-09-191-1/+4
|
* Allow dynamic transforms on array<string> summary fields.Geir Storli2022-09-135-22/+82
|
* Remove summarymap config that is no longer used.Geir Storli2022-09-081-1/+0
|
* Warn oonce per summaryJon Bratseth2022-09-051-6/+12
|
* Add 'paged' support for reference attribute fields.Geir Storli2022-08-241-1/+0
| | | | | | | | The memory usage of a reference attribute can be calculated as follows: num_child_docs * 8 + num_parent_docs * 24 + num_child_docs_where_parent_reference_is_set * 4. When using 'paged' the btrees used in the reverse mapping from parent lid to child lids are still memory allocated. This is the num_child_docs_where_parent_reference_is_set part of the equation above.
* No functional changesJon Bratseth2022-07-251-2/+1
|
* Remove textextractor summary transform.Geir Storli2022-07-141-1/+0
| | | | This has been dead for many years.
* Merge pull request #23465 from ↵v8.17.19Geir Storli2022-07-123-0/+34
|\ | | | | | | | | vespa-engine/geirst/summary-transform-for-documentid-fields Add new summary transform for all 'documentid' summary fields.
| * Add new summary transform for all 'documentid' summary fields.Geir Storli2022-07-123-0/+34
| | | | | | | | | | This prepares for no longer inserting 'documentid' fields in the docsum blob in the backend. Instead such fields will be produced using a docsum writer (transform) that uses the document instance to get the id.
* | Don't say we expect declaration in query profilesJon Bratseth2022-07-111-3/+3
|/
* Style fixes. Add member function comment.Tor Egge2022-06-291-5/+5
|
* Use CopyDFW for struct field summaries with explicit source.Tor Egge2022-06-291-0/+14
|
* No functional changesJon Bratseth2022-06-151-0/+3
|
* update commentArne H Juul2022-06-081-1/+1
|
* GC deprecation warnings.Henning Baldersheim2022-06-081-1/+2
|
* move internal details from public APIArne H Juul2022-06-083-6/+16
|
* no more tensors in attribute prefetchArne H Juul2022-06-083-5/+10
| | | | * less hardcoded constants
* Rename the 'searchdefinition' package to 'schema'Jon Bratseth2022-05-1971-0/+5586