Release 0.267

Details

Google Sheet Changes

  • Introduce Google sheet connector, supporting read queries.

Hive Changes

  • Add support for weighing splits according to their file size, allowing deeper worker split queues when files are small. This behavior is enabled by default, and can be disabled by setting the configuration property hive.size-based-split-weights-enabled or the session property size_based_split_weights_enabled to false. Additionally, when enabled, the extent to which hive splits are considered “smaller” than standard can be controlled by using the configuration property hive.minimum-assigned-split-weight or session property minimum_assigned_split_weight. The default value is 0.05.

  • Add support for reading LZ4 and ZSTD compressed parquet data.

  • Switch the default value of config hive.orc.writer.stream-layout-type to BY_COLUMN_SIZE to improve IO efficiency.

Pinot Changes

  • Fix query crash when Pinot query returns mismatched schema.

  • Add support to enable Pinot proxy for broker and grpc requests.

  • Add support to pushdown IS NULL and IS NOT NULL queries to Pinot.

  • Add support to run IS NOT NULL queries on Pinot data from Presto.

Credits

Arunachalam Thirupathi, Beinan Wang, Chen, James Petty, James Sun, Pranjal Shankhdhar, Reetika Agrawal, Rongrong Zhong, Sreeni Viswanadha, Varun Gajjala, Xiang Fu, Xinli shang, Zac, Zhenxiao Luo, kriti-sc, tanjialiang, v-jizhang