BlakeOrth commented on PR #17050: URL: https://github.com/apache/datafusion/pull/17050#issuecomment-3189831540
> Thank you for your patience No worries at all, it's pretty obvious to me you have quite a few plates spinning. I wanted to ensure you had time to review, but simultaneously didn't want to be too bothersome with requests/@ mentions etc. Hopefully I struck an ok balance. > I am sorry but I still don't understand this PR's logic -- probably mostly because I don't understand the current behavior of how ListingTable is supposed to automatically infer partitioning columns > > I worry if we merge this PR as is we may break some other existing behavior Sure, that makes sense. Just for the sake of clarity here, the `ListingTable` itself does partition inference without any issues. The problems seen here stem from how the `ListingTableFactory` configures the options passed to a `ListingTable`. I guess the one caveat to that is fix for filtering of non-hive compliant directories, which was exposed by an existing test once partition inference was introduced. > Is this something you might be willing to try? Sure, I can absolutely give this a shot. Should I add those tests to this PR or a new PR that just introduces new .slt tests be better? -- 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. To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
