marko-radoicic-db commented on PR #58777: URL: https://github.com/apache/spark/pull/58777#issuecomment-5834099437
> Do we need `assert_column_resolution_error` here? In a Connect parity suite, an inherited test normally indicates identical Classic and Connect behavior, including the error condition/message. That is also the convention we have been using when analyzing parity tests. Overriding this helper lets the same test accept different errors, hiding an observable behavior difference. Could we inline the expected condition instead? If the divergence is intentional, the relevant Connect tests should be explicitly overridden and documented. Good point - here's a followup that does just that: https://github.com/apache/spark/pull/59040 -- 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]
