Release 451 (27 Jun 2024)#
General#
Add support for configuring a proxy for the S3 native filesystem with the
s3.http-proxy.username
,s3.http-proxy.password
,s3.http-proxy.non-proxy-hosts
, ands3.http-proxy.preemptive-basic-auth
configuration properties. (#22207)Improve performance of reading JSON array data. (#22379)
Improve performance of certain queries involving the
row_number()
,rank()
, ordense_rank()
window functions with partitioning and filters. (#22509)Fix error when reading empty files with the native S3 file system. (#22469)
Fix rare error where query execution could hang when fault-tolerant execution is enabled. (#22472)
Fix incorrect results for CASE expressions of the form
CASE WHEN ... THEN true ELSE false END
. (#22530)
Delta Lake connector#
Improve performance of reading from Parquet files with large schemas. (#22451)
Hive connector#
Improve performance of reading from Parquet files with large schemas. (#22451)
Hudi connector#
Improve performance of reading from Parquet files with large schemas. (#22451)
Iceberg connector#
Add support for incremental refresh for basic materialized views. (#20959)
Add support for adding and dropping fields inside an array. (#22232)
Add support for specifying a resource prefix in the Iceberg REST catalog. (#22441)
Add support for partitioning on nested
ROW
fields. (#15712)Add support for writing Parquet Bloom filters. (#21570)
Add support for uppercase characters in the
partitioning
table property. (#12668)Improve performance of reading from Parquet files with large schemas. (#22451)
Kudu connector#
SPI#
Add a
Connector.getInitialMemoryRequirement()
API for pre-allocating memory during catalog initialization. (#22197)