Google Sheet Changes#
Introduce Google sheet connector, supporting read queries.
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-enabledor the session property
size_based_split_weights_enabledto 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-weightor 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
BY_COLUMN_SIZEto improve IO efficiency.
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 NOT NULLqueries to Pinot.
Add support to run
IS NOT NULLqueries on Pinot data from Presto.
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