1.35.0 - 2025-07-11

Added

• Spark: Include spark_applicationDetails facet to all events #3848 @dolfinus
  Add spark_applicationDetails facet to all OpenLineage events emitted by the Spark integration
• Spark: Support additional data in Catalog facet #3850 @ddebowczyk92
  Adds support additional data in Catalog in Spark integration
• Spark: disable connector by Spark config parameter #3880 @pawel-big-lebowski
  Add spark.openlineage.disabled entry to disable OpenLineage integration through Spark config parameters
• Spark: Fine-grained timeout config #3779 @pawel-big-lebowski
  Add extra timeout options to emit incomplete OpenLineage events in case of timeout when building facets. See buildDatasetsTimePercentage and facetsBuildingTimePercentage in docs for more details
• Python: Asynchronous HTTP transport implementation #3812 @mobuchowski
  Adds high-performance asynchronous HTTP transport with event ordering guarantees, configurable concurrency, and comprehensive error handling. Features START-before-completion event ordering, bounded queues, and real-time statistics
• dbt: Add DbtRun facet to dbt run events #3764 @dolfinus
  Adds DbtRun facet for tracking dbt run information
• Python: Add continue_on_success and sorting transport in CompositeTransport #3829 @kacpermuda
  Adds configuration options for CompositeTransport to control behavior and ordering
• Hive: Add jobType facet #3789 @dolfinus
  Adds jobType facet to Hive integration
• Hive: Add dialect=hive to SqlJobFacet #3863 @dolfinus
  Adds dialect field to SqlJobFacet for Hive integration
• Java: Add Debezium job naming convention #3834 @mvitale
  Adds standardized job naming convention for Debezium integration
• Spec: SqlJobFacet now contains dialect #3819 @mobuchowski
  Adds dialect field to SqlJobFacet specification
• Spec: Formalize job naming #3826 @ddebowczyk92
  Formalizes job naming conventions in the specification
• Spec: Formalize dataset naming #3775 @ddebowczyk92
  Formalizes dataset naming conventions in the specification
• Website: Add Collibra as consumer in ecosystem page #3732 @marccampa
  Adds Collibra to the ecosystem page as an OpenLineage consumer

Changed

• Spark: Use Hive as default Iceberg catalog implementation #3858 @dolfinus
  Updates Spark integration to use Hive as the default catalog implementation for Iceberg tables
• Spark: Replace weak hash-map with a map with weak keys and entries #3856 @ddebowczyk92
  Improves memory management in Spark integration by replacing weak hash-map implementation
• Spark: Support latest databricks runtime #3811 @pawel-big-lebowski
  Updates Spark integration to support the latest Databricks runtime
• Python: Remove transport.wait_for_completion() #3881 @dolfinus
  Removes wait_for_completion() method from Python transport interface
• Python: Reuse session in sync HttpTransport #3843 @dolfinus
  Improves performance by reusing HTTP sessions in synchronous transport
• Python: Implement Transport.close() for Datazone and Kinesis #3857 @dolfinus
  Adds proper cleanup methods for Datazone and Kinesis transports
• Python: Implement TransformTransport.close #3855 @dolfinus
  Adds cleanup method for TransformTransport
• Python: Implement KafkaTransport.wait_for_completion() and .close() #3838 @dolfinus
  Adds proper cleanup and completion methods for Kafka transport
• Java: Make CompositeTransport.close() more reliable #3841 @dolfinus
  Improves reliability of CompositeTransport cleanup process
• Java: Cover OpenLineageClient.close() with tests #3839 @dolfinus
  Adds test coverage for OpenLineageClient cleanup methods
• Java: Name threads used in Java client #3817 @mobuchowski
  Adds meaningful names to threads used in Java client for better debugging
• Flink: Close OpenLineageClient in onJobExecuted #3854 @dolfinus
  Ensures proper cleanup of OpenLineageClient in Flink 1.x integration
• Flink: Fixed a bug incorrectly loading configuration in Event Emitter #3799 @fetta
  Fixes configuration loading issue in Flink Event Emitter
• dbt: Make invocation_id field optional #3796 @dolfinus
  Makes invocation_id field optional in dbt integration
• dbt: More resiliency for missing dbt nodes #3836 @mobuchowski
  Improves error handling for missing dbt nodes
• Hive: Add docker-compose example for local testing #3800 @dolfinus
  Adds docker-compose setup for local Hive integration testing
• Airflow: Send pending events after Airflow DAG is finished #3849 @dolfinus
  Ensures all pending events are sent after DAG completion in Airflow integration

Fixed

• dbt: Fix log path, more precise file reading #3793 @mobuchowski
  Improves log file handling in dbt integration
• dbt: Fix deprecated configs #3859 @kacpermuda
  Replaces deprecated dbt configurations with current alternatives
• Spark: Fix missing .db suffix in database/namespace location name for BigQueryMetastoreCatalog #3874 @ddebowczyk92
  Fixes database naming issue in BigQuery Metastore catalog implementation
• Spark: Fix missing output's dataset catalog facet when running CTAS queries on Iceberg tables #3835 @ddebowczyk92
  Fixes missing catalog facet in output datasets for CTAS queries on Iceberg tables
• Spark: Delta merge with column #3871 @pawel-big-lebowski
  Fixes Delta merge operation handling with column-level lineage
• Spark: Support delta 3.3.2 #3861 @pawel-big-lebowski
  Adds support for Delta Lake version 3.3.2
• Spark: Call version utils method when dataset can be identified #3832 @pawel-big-lebowski
  Fixes version handling when dataset can be properly identified
• Java: Raise error if no events were emitted by composite transport #3853 @kacpermuda
  Adds error handling when CompositeTransport fails to emit any events
• Java: Fix IntelliJ reload #3825 @mobuchowski
  Fixes IntelliJ project reload issues
• Java: Fix spotless in hive integration #3806 @mobuchowski
  Fixes code formatting issues in Hive integration
• Java: Fix missing field in SQL facet test #3830 @mobuchowski
  Fixes missing field in SQL facet test case
• Rust: Fix new rust formatting warnings #3814 @mobuchowski
  Fixes formatting warnings in Rust code