Release 0.245#


There is a bug causing failure at startup if function namespace manager is enabled and Thrift is not configured (#15501).


There is a bug causing failure when reading ORC files with ARRAY/MAP/ROW of VARCHAR columns using the selective stream readers for some corner cases (#15549)


  • New enum_key UDF to get the key for an enum value.

  • Configurable query result compression.


General Changes#

  • Fix wrong results issue for queries with multiple lambda expressions that differ only in array/map/row constants that have the same length (#15424).

  • Fix serialization bug causing maps with string-valued enum keys to be displayed as illegible strings in Presto CLI.

  • Fix parsing logic for prepared statements to be consistent with regular statements.

  • Improve memory usage for fragment result caching.

  • Add support for disabling query result compression via client and server-side configuration properties. Clients can disable compressed responses using the --disable-compression flag or disableCompression driver property. Compression can be disabled server-wide by using the configuration property: query-results.compression-enabled=false

  • Add support in affinity scheduler to enable cache for bucketed table scan that has remote source.

  • Add enum_key to get the key corresponding to an enum value: ENUM_KEY(EnumType) -> VARCHAR.

Hive Connector Changes#

  • Improve parquet metadata reader by preloading data and reducing the number of reads (#15421).

SPI Changes#

  • Add getSchema to ConnectorSession.


Andrii Rosa, Beinan Wang, Bhavani Hari, Daniel Ohayon, Dong Shi, Ge Gao, James Petty, James Sun, Ke Wang, Leiqing Cai, Masha Basmanova, Mayank Garg, Nikhil Collooru, Rebecca Schlussel, Rongrong Zhong, Saksham Sachdev, Shixuan Fan, Timothy Meehan, Vic Zhang, Wenlei Xie, Ying Su, Zhenxiao Luo