Release 423 (10 Aug 2023)#
General#
- Add support for renaming nested fields in a column via - RENAME COLUMN. (#16757)
- Add support for setting the type of a nested field in a column via - SET DATA TYPE. (#16959)
- Add support for comments on materialized view columns. (#18016) 
- Add support for displaying all Unicode characters in string literals. (#5061) 
- Improve performance of - INSERTand- CREATE TABLE AS ... SELECTqueries. (#18212)
- Improve performance when planning queries involving multiple window functions. (#18491) 
- Improve performance of queries involving - BETWEENclauses. (#18501)
- Improve performance of queries containing redundant - ORDER BYclauses in views or- WITHclauses. This may affect the semantics of queries that incorrectly rely on implementation-specific behavior. The old behavior can be restored via the- skip_redundant_sortsession property or the- optimizer.skip-redundant-sortconfiguration property. (#18159)
- Reduce default values for the - task.partitioned-writer-countand- task.scale-writers.max-writer-countconfiguration properties to reduce the memory requirements of queries that write data. (#18488)
- Remove the deprecated - optimizer.use-mark-distinctconfiguration property, which has been replaced with- optimizer.mark-distinct-strategy. (#18540)
- Fix query planning failure due to dynamic filters in fault tolerant execution mode. (#18383) 
- Fix - EXPLAINfailure when a query contains- WHERE ... IN (NULL). (#18328)
JDBC driver#
- Add support for constrained delegation with Kerberos. (#17853) 
CLI#
- Add support for accepting a single Trino JDBC URL with parameters as an alternative to passing command line arguments. (#12587) 
ClickHouse connector#
- Add support for - CASCADEoption in- DROP SCHEMAstatements. (#18305)
Blackhole connector#
- Add support for the - COMMENT ON VIEWstatement. (#18516)
Delta Lake connector#
- Add - $propertiessystem table which can be queried to inspect Delta Lake table properties. (#17294)
- Add support for reading the - timestamp_ntztype. (#17502)
- Add support for writing the - timestamp with time zonetype on partitioned columns. (#16822)
- Add option to enforce that a filter on a partition key is present for query processing. This can be enabled by setting the - delta.query-partition-filter-requiredconfiguration property or the- query_partition_filter_requiredsession property to- true. (#18345)
- Improve performance of the - $historysystem table. (#18427)
- Improve memory accounting of the Parquet writer. (#18564) 
- Allow metadata changes on Delta Lake tables with identity columns. (#18200) 
- Fix incorrectly creating files smaller than the configured - file_size_thresholdas part of- OPTIMIZE. (#18388)
- Fix query failure when a table has a file with a location ending with whitespace. (#18206) 
Hive connector#
- Add support for changing a column’s type from - varcharto- timestamp. (#18014)
- Improve memory accounting of the Parquet writer. (#18564) 
- Remove the legacy Parquet writer, along with the - parquet.optimized-writer.enabledconfiguration property and the- parquet_optimized_writer_enabledsession property. Replace the- parquet.optimized-writer.validation-percentageconfiguration property with- parquet.writer.validation-percentage. (#18420)
- Disallow coercing Hive - timestamptypes to- varcharfor dates before 1900. (#18004)
- Fix loss of data precision when coercing Hive - timestampvalues. (#18003)
- Fix incorrectly creating files smaller than the configured - file_size_thresholdas part of- OPTIMIZE. (#18388)
- Fix query failure when a table has a file with a location ending with whitespace. (#18206) 
- Fix incorrect results when using S3 Select and a query predicate includes a quote character ( - ") or a decimal column. (#17775)
- Add the - hive.s3select-pushdown.experimental-textfile-pushdown-enabledconfiguration property to enable S3 Select pushdown for- TEXTFILEtables. (#17775)
Hudi connector#
- Fix query failure when a table has a file with a location ending with whitespace. (#18206) 
Iceberg connector#
- Add support for renaming nested fields in a column via - RENAME COLUMN. (#16757)
- Add support for setting the type of a nested field in a column via - SET DATA TYPE. (#16959)
- Add support for comments on materialized view columns. (#18016) 
- Add support for - tinyintand- smallinttypes in the- migrateprocedure. (#17946)
- Add support for reading Parquet files with time stored in millisecond precision. (#18535) 
- Improve performance of - information_schema.columnsqueries for tables managed by Trino with AWS Glue as metastore. (#18315)
- Improve performance of - system.metadata.table_commentswhen querying Iceberg tables backed by AWS Glue as metastore. (#18517)
- Improve performance of - information_schema.columnswhen using the Glue catalog. (#18586)
- Improve memory accounting of the Parquet writer. (#18564) 
- Fix incorrectly creating files smaller than the configured - file_size_thresholdas part of- OPTIMIZE. (#18388)
- Fix query failure when a table has a file with a location ending with whitespace. (#18206) 
- Fix failure when creating a materialized view on a table which has been rolled back. (#18205) 
- Fix query failure when reading ORC files with nullable - timecolumns. (#15606)
- Fix failure to calculate query statistics when referring to - $pathas part of a- WHEREclause. (#18330)
- Fix write conflict detection for - UPDATE,- DELETE, and- MERGEoperations. In rare situations this issue may have resulted in duplicate rows when multiple operations were run at the same time, or at the same time as an- optimizeprocedure. (#18533)
Kafka connector#
- Rename the - ADD_DUMMYvalue for the- kafka.empty-field-strategyconfiguration property and the- empty_field_strategysession property to- MARK(#18485).
Kudu connector#
- Add support for optimized local scheduling of splits. (#18121) 
MariaDB connector#
- Add support for - CASCADEoption in- DROP SCHEMAstatements. (#18305)
MongoDB connector#
- Add support for predicate pushdown on - charand- decimaltype. (#18382)
MySQL connector#
Oracle connector#
SingleStore connector#
- Add support for - CASCADEoption in- DROP SCHEMAstatements. (#18305)
SPI#
- Deprecate the - ConnectorMetadata.getTableHandle(ConnectorSession, SchemaTableName)method signature. Connectors should implement- ConnectorMetadata.getTableHandle(ConnectorSession, SchemaTableName, Optional, Optional)instead. (#18596)
- Remove the deprecated - supportsReportingWrittenBytesmethod from ConnectorMetadata. (#18617)