Add generic OpenLineage context configuration for propagating parent-run info#4682@kacpermudaAdds one JSON context payload for parent-run info, replacing 6 separate config keys in dbt and Spark.
Dbt: Capture incremental-strategy config and run-wide full_refresh#4716@himakolavennuCaptures incremental model settings, such as strategy and partitioning, plus a run-wide full_refresh flag.
dbt: Emit exposures as a dataset facet on dbt model datasets#4705@himakolavennuAdds exposures, such as dashboards and notebooks, as a dataset facet, so lineage now includes table-to-exposure links.
Flink: Allow disabling checkpoint tracking to reduce excessive RUNNING events#4744@wangxiaojingAdds a setting to turn off checkpoint tracking, cutting unnecessary RUNNING events and REST calls.
GreatExpectations: Implement FileSizeExpectationsParser for file asset validation#4698@maitraymukeshkumarmodi-aimlImplements file size checks, so file size expectation results now produce a fileSize facet instead of being ignored.
Spec: Extend ParentRunFacet with parent and root facets#4680@kacpermudaAdds an optional facets field to ParentRunFacet, so producers can forward parent and root facets to child events.
Client/Python: Add new Datadog DCs#4711@HeroCCAdds the new UK1 and US2 FedRAMP Datadog sites to the transport's accepted site list.
Client/Java: Fix deserialization crash for facet fields literally named additionalProperties#4743@zerafachrisFixes a crash when a facet field is named additionalProperties, which confused the JSON deserialiser.
Client/Python: Don't recurse deep_merge_dicts into a non-dict value#4731@chuenchen309Fixes a crash when merging a scalar config value with a dict value from another source.
Client/Python: Fix async HTTP transport shutdown to wait for events sharing an ID#4728@mattfaltynFixes shutdown so it waits for every event, even when several events share the same ID.
Client/Python: Fix several async transport issues in the Datadog transport#4694@mobuchowskiFixes high CPU use and a memory leak in the Datadog transport's async worker.
Client/Python: Make with_additional_properties() facets picklable#4729@zerafachrisFixes a pickling error for facets built with with_additional_properties().
Client/Python: Match tag keys case-insensitively in both directions#4722@chuenchen309Fixes tag matching so a user tag overrides an integration tag, regardless of letter case.
Client/Python: Restore HTTP debuglevel after emit failure#4696@hcthakur2004Restores the HTTP debug level after a failed emit, so it does not stay switched on for later requests.
Common/dbt: Consume --openlineage-dbt-job-name in its --opt=value form#4730@chuenchen309Fixes job name handling so --openlineage-dbt-job-name=value no longer breaks the dbt run.
Common/dbt: Let the ~/.dbt profiles directory fallback run#4725@chuenchen309Fixes the profiles directory lookup so it now falls back to ~/.dbt/ as intended.
Common/dbt: Resolve dbt test seed inputs by alias, not logical name#4732@chuenchen309Fixes seed input resolution so tests use the seed's alias instead of its logical name.
Common/dbt: Separate the default spark port from the host#4724@chuenchen309Fixes a missing colon in the default Spark port, which broke the dataset namespace.
Common/dbt: Use the dataset namespace for the SQL query externalQuery facet#4723@chuenchen309Fixes the SQL query externalQuery facet to use the dataset namespace, not a placeholder value.
Sql: Include WHERE subquery inputs in table lineage#4752@mattfaltynIncludes tables referenced only in a WHERE subquery, such as WHERE EXISTS, in table lineage.