melgenek commented on issue #4462: URL: https://github.com/apache/arrow-datafusion/issues/4462#issuecomment-1374552643
> I also left some feedback on https://github.com/apache/arrow-datafusion/pull/4834#pullrequestreview-1239686856 -- does that make sense? Thank you for the feedback. > For both datafusion and postgres, apply some sort of normalization to floating point values prior to printing (round them to a smaller number of significant figures I think this is a good suggestion. I'll update the Postgres/Datafusion clients to have a normalization step for results. ------ In order to proceed further, I'd like to clarify the way how you'd like to define `.slt` tests. In https://github.com/apache/arrow-datafusion/pull/4834 I implemented the tests similar to the existing `integragration-tests`: there are only queries and results are generated and compared by the test runner. Regular `.slt` are supposed to have expected outputs. And in this comment https://github.com/apache/arrow-datafusion/pull/4834#discussion_r1064024225 you're suggesting having the same approach for Postgres compatibility tests. Having explicit results is a totally viable approach, especially having `--complete` option for running sqllogictests. So do we agree, that the way to go is to force all `.slt` tests to have expected outputs, and then simply verify the same tests with Postgres/Datafusion runnens? -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
