tyrobro commented on PR #301: URL: https://github.com/apache/datafusion-spark/pull/301#issuecomment-6943874902
Thanks for informing me, have made the following changes to fix the issues: 1. Moved the string cast above the is_volatile() check in simplify(), so a volatile expr1 paired with a STRING argument now gets the cast args in its Original fallback, and invoke_with_args only ever sees Float32/Float64. Added an explicit type-validation loop in invoke_with_args over all arguments (not just args[0]) that returns a clear exec_err! instead of letting a type mismatch surface as an Arrow zip error. Added the nanvl(random(), s) test case with a string column. 2. Fixed the -0.0 sign test to use string formatting (CAST(... AS VARCHAR) LIKE '-%') instead of 1/x, so it matches Spark’s actual divide-by-zero behavior rather than relying on DataFusion-specific division semantics. 3. Fixed the missing trailing newline. Noted on the '1d'/'1f' cast-compatibility gap. -- 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]
