pepijnve commented on PR #17813: URL: https://github.com/apache/datafusion/pull/17813#issuecomment-3505112907
> I am really concerned about this PR adding lots of code and complexity for relatively little gain (fixing a schema mismatch) That's fair. I have no stake in this one TBH, it just solves the problem at hand. FWIW nobody has come up with a better solution in the meantime. I do think the goal in this PR is not so much to fix a schema mismatch, but to improve the accuracy of `nullable` for case expressions. At the moment it reports nullable in many situations where it's provably not. Fixing the schema mismatch is a nice side effect. > Mostly i am concerned it that it is complicated enough that we won't be able to catch additional issues / bugs and that it may introduce very subtle and hard to track down bugs of its own Also a fair comment and I share your concern which is why I at least started adding unit tests. I'm going to assume the complexity concern is wrt the logical implementation and in particular the code in `predicate_eval.rs` since the changes to other files are pretty straightforward. Is there a way we could mitigate these concerns? How is the `cp_solver` code for instance tested for correctness? -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
