HeartSaVioR edited a comment on pull request #31427: URL: https://github.com/apache/spark/pull/31427#issuecomment-776580168
I tend to agree that limiting the namespace only on session catalog isn't good and I really would like to see bright solution which addresses this, but the new solution shouldn't conflict with the existing V1 space (table, view, temp view) which has no concept of "custom" catalog. Btw, looks like we are still not clear about the "role" of session catalog. Is it for extending V1 space, or make V2 catalog simply used by default? I see both claims from previous discussion and existing implementation out there. If it's limited to V1, limiting the namespace is natural as there's no catalog. If not, we'd better elaborate the problem(s) we have by limiting the namespace, and discuss what would be the ideal/realistic approach for Spark to deal with. (I can roughly imagine that that's from an use case that table in other catalog should be accessible in 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: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
