ray
ray copied to clipboard
[Datasets] Add test coverage for reading from HDFS.
Add test coverage for reading from a pyarrow HadoopFileSystem
. Will need to explore mocking HDFS, or starting and managing a local HDFS cluster via a pytest fixture.
@atomic FYI this PR which is for S3 may serve as an example for HDFS test: https://github.com/ray-project/ray/pull/17158
@jianoaix based on @atomic's bandwidth, we are targeting 11/23 for the completion of this task, let us know if this is reasonable, thanks!