Yang Zhang
Yang Zhang
@PHILO-HE thank you!
Velox use libhdfs3 to connect HDFS, find complete supported config here. https://github.com/apache/hawq/blob/master/src/backend/utils/misc/etc/hdfs-client.xml
We can provide options to choose building static or dynamic artifacts, and enable static by default. Just not force ban dynamic packaging directly, thank you.
I'd like support hex and unhex. update: hex and unhex has already supported in Gluten.
I'd like take percentile agg function.
@raulcd Hi, I am trying to use Velox + Gluten develop Spark. After this commit, folly looks not installed in centos8 by default, but still installed in ubuntu [code](https://github.com/facebookincubator/velox/blob/d4225fa88d5438d38594e48c2d30493bff420e1e/scripts/setup-ubuntu.sh#L96), so...
@jinchengchenghh would you support array type in ArrayColumnarRow?
We found same issue.
> I see test `Gluten - Filter applied on merged Parquet schema with new column should work` is removed from 3.3 / 3.4 / 3.5 UTs but is preserved in...
> If `Gluten - Filter applied on merged Parquet schema with new column should work` is removed, can we directly use vanilla Spark's `Filter applied on merged Parquet schema with...