adiranjan25 commented on PR #19045: URL: https://github.com/apache/iceberg/pull/19045#issuecomment-7023185460
I've completed the initial implementation and regression tests for integer/long truncation overflow. The targeted API tests pass, including a regression test for projection overflow near `Integer.MIN_VALUE`. One design decision remains before I finalize the fix: - **Reject unrepresentable results:** Prevents silent wraparound but may reject valid source values during partition writes. - **Use a defined boundary representation:** Could preserve write support but requires corresponding projection semantics and compatibility handling. Existing tables may also contain wrapped partition values, so reader compatibility needs consideration. Could a maintainer advise which approach is preferred? Once confirmed, I'll finalize the implementation, extend regression coverage, and update this Draft PR for review. Thanks! -- 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]
