Release 449 (31 May 2024)

General

Add OpenLineage event listener. ({issue}21265)
Fix rare query failure or incorrect results for array types when the data is dictionary encoded. ({issue}21911)
Fix JMX metrics not exporting for resource groups. ({issue}21343)

Improve performance when listing schemas while the bigquery.case-insensitive-name-matching configuration property is enabled. ({issue}22033)

Add support for pushing down execution of the count(distinct), corr, covar_samp, and covar_pop functions to the underlying database. ({issue}7100)
Improve performance when pushing down equality predicates on textual types. ({issue}7100)

Add support for the $partitions system table. ({issue}18590)
Add support for reading from and writing to tables with VACUUM Protocol Check. ({issue}21398)
Add support for configuring the request retry policy on the native S3 filesystem with the s3.retry-mode and s3.max-error-retries configuration properties. ({issue}21900)
Automatically use timestamp(6) in struct types as a type during table creation when timestamp is specified. ({issue}21511)
Improve performance of writing data files. ({issue}22089)
Fix query failure when the hive.metastore.glue.catalogid configuration property is set. ({issue}22048)

Add support for specifying a catalog name in the Thrift metastore with the hive.metastore.thrift.catalog-name configuration property. ({issue}10287)
Add support for configuring the request retry policy on the native S3 filesystem with the s3.retry-mode and s3.max-error-retries configuration properties. ({issue}21900)
Improve performance of writing to Parquet files. ({issue}22089)
Allow usage of filesystem caching on the Trino coordinator when node-scheduler.include-coordinator is enabled. ({issue}21987)
Fix failure when listing Hive tables with unsupported syntax. ({issue}21981)
Fix query failure when the hive.metastore.glue.catalogid configuration property is set. ({issue}22048)
Fix failure when running the flush_metadata_cache table procedure with the Glue v2 metastore. ({issue}22075)

Add support for configuring the request retry policy on the native S3 filesystem with the s3.retry-mode and s3.max-error-retries configuration properties. ({issue}21900)

Add support for views when using the Iceberg REST catalog. ({issue}19818)
Add support for configuring the request retry policy on the native S3 filesystem with the s3.retry-mode and s3.max-error-retries configuration properties. ({issue}21900)
Automatically use varchar in struct types as a type during table creation when char is specified. ({issue}21511)
Automatically use microsecond precision for temporal types in struct types during table creation. ({issue}21511)
Improve performance and memory usage when equality delete files are used. ({issue}18396)
Improve performance of writing to Parquet files. ({issue}22089)
Fix failure when writing to tables with Iceberg VARBINARY values. ({issue}22072)

{{breaking}} Remove support for non-gRPC clients and the pinot.grpc.enabled and pinot.estimated-size-in-bytes-for-non-numeric-column configuration properties. ({issue}22213)