Release 391 (22 Jul 2022)#
General#
- Improve performance of repeated aggregations with - CASEexpressions. (#12548)
- Improve query latency when there is high concurrency. (#13213) 
- Improve planning performance for join queries when tables have statistics. (#13047) 
- Fail queries that get stuck in long-running regular expression functions. (#12392) 
- Fix potential query failure when the - UUIDtype is used. (#13265)
- Set the default value of the - optimizer.force-single-node-outputconfiguration property to false. (#13217)
BigQuery connector#
Delta Lake connector#
- Improve optimized Parquet writer performance. (#13203, #13208) 
- Store query ID when creating a new schema. (#13242) 
- Fix incorrect - schema already existserror caused by a client timeout when creating a new schema. (#13242)
- Fix incorrect query results when reading a table with an outdated cached representation of its active data files. (#13181) 
Druid connector#
- Fix potential query failure when using the - querytable function with metadata caching and the underlying table schema is changed via Trino. (#12526)
Hive connector#
- Add support for AWS Athena partition projection. (#11305) 
- Improve optimized Parquet writer performance. (#13203, #13208) 
- Fix potential failure when creating empty ORC bucket files while using ZSTD compression. (#9775) 
- Fix query failure or potentially incorrect statistics when running concurrent - CREATE TABLE ASqueries with the- IF NOT EXISTSclause for the same non-existent table. (#12895)
- Fix incorrect results when using the Glue metastore with queries that contain - IS NULLand additional filters. (#13122)
Iceberg connector#
MariaDB connector#
- Fix potential query failure when using the - querytable function with metadata caching and the underlying table schema is changed via Trino. (#12526)
MySQL connector#
- Fix potential query failure when using the - querytable function with metadata caching and the underlying table schema is changed via Trino. (#12526)
Oracle connector#
- Fix potential query failure when using the - querytable function with metadata caching and the underlying table schema is changed via Trino. (#12526)
PostgreSQL connector#
- Fix potential query failure when using the - querytable function with metadata caching and the underlying table schema is changed via Trino. (#12526)
Redshift connector#
- Fix potential query failure when using the - querytable function with metadata caching and the underlying table schema is changed via Trino. (#12526)
SQL Server connector#
- Fix potential query failure when using the - querytable function with metadata caching and the underlying table schema is changed via Trino. (#12526)
SPI#
- Removed deprecated methods and classes related to the grouped execution feature. (#13125)