Release 0.155
Fix incorrect results when using with
IN
predicates.Fix error messages for failures during commit.
Reduce unnecessary memory usage of map_agg(), and map_union().
Make ,
EXCLUDING
and non-reserved keywords.Remove support for the experimental feature to compute approximate queries based on sampled tables.
Various optimizations to reduce coordinator CPU usage.
Fix schema evolution support in new Parquet reader.
Fix
NoClassDefFoundError
when using Hadoop KMS.Always produce dictionary blocks for DWRF dictionary encoded streams.