osipovartem opened a new issue, #25380: URL: https://github.com/apache/datafusion/issues/25380
### Bug description `UpperFunc` and `LowerFunc` rely on the default `ScalarUDFImpl::return_field_from_args`, which marks their result fields nullable even when the input field is non-nullable. Both functions are null-preserving: a non-null input produces a non-null output, while a nullable input remains nullable. This makes planned schemas less precise and affects clients that consume Arrow/SQL result metadata. ### Expected behavior `UPPER` and `LOWER` result fields should preserve the input field nullability and string data type. ### Performance constraint The fix should be schema-only through `return_field_from_args`; it should not add runtime dispatch, per-row conversion, or allocation to the vectorized kernels. -- 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]
