Description
There are still some missing parts in the file source V2 framework:
1. It doesn't support reporting file scan metrics such as "numOutputRows"/"numFiles"/"fileSize" like `FileSourceScanExec`.
2. It doesn't support partition pruning with subqueries or dynamic partition pruning.
As we are going to code freeze on Jan 31st, I suggest disabling all the V2 file sources in Spark 3.0 by default.
Attachments
Issue Links
- links to