cloud-fan commented on issue #24768: [SPARK-27919][SQL] Add v2 session catalog URL: https://github.com/apache/spark/pull/24768#issuecomment-509899359 I think I understand the behavior proposed here now. It's better to put the final rule in the PR description, instead of only the updates. IIUC, the final rule is: for a table name `a.b` and `a` is not a registered catalog 1. if the default catalog is set, resolve the table name with this user-specified default catalog 2. otherwise, go with the hive catalog. It has 2 cases: 2.1 if the table provider is v1, go with the old code path (old command plans) 2.2 if the provider is v2, go with v2 session catalog. I agree with it and I'm happy to merge it except one concern: I think it's a bad UX to ask end-users to config both default catalog and session catalog. It's really confusing to expose these 2 configs together to end-users, and I don't expect to see custom implements of session catalog.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
