jojochuang commented on PR #11780: URL: https://github.com/apache/ozone/pull/11780#issuecomment-6509684202
Could you extend the regression tests with these scenarios, asserting `INVALID_REQUEST`? - Repeated source: `a -> b`, `a -> c` (directly covers response-map entry replacement). - Repeated destination: `a -> c`, `b -> c`. - Nonadjacent overlap: `a -> b`, `c -> d`, `b -> e` (ensures names from all earlier pairs remain tracked). Please also cover and clarify the compatibility restriction for a batch that previously allowed a valid partial result: suppose `a` exists and `b` and `c` do not, with `b -> c` followed by `a -> b`. The first pair fails because its source is missing, while the second can successfully rename `a` to `b`. The new name-only preExecute check instead rejects the whole batch because `b` is shared, even though the first pair would make no metadata changes. I understand that rejecting overlaps is intentional, but this is broader than rejecting combinations that actually corrupt persistence. Please add a test documenting that behavior and mention the change from partial success to whole-request rejection in the PR description. Generated-by: Codex (GPT-6) -- 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]
