SEZ9 commented on PR #12605: URL: https://github.com/apache/seatunnel/pull/12605#issuecomment-6030219343
Thanks @SEPURI-SAI-KRISHNA, the pointers help. **`numberToInt` / `longValue()` (F1)** – Understood that as of `5eb6a8ca9` the `BigDecimal`/`BigInteger` branches go through the `BigInteger` bound comparison and `longValue()` only runs on the Byte/Short/Integer/Long branch. Using 18446744073709551616 as the test value makes sense given `longValue()` yields 0 there. I'll confirm those branches and `testCoalesceRejectsADecimalBeyondLongRange` in the diff at `22090e713` before closing this one. **Javadoc (F4)** – Thanks for tracing the wording across `b3637b7cd`, `0d21566d6` and `5eb6a8ca9`. I'll read the current Javadoc at `22090e713` and close this if the before/after narration is gone. **Error message / code (F3)** – The neutral message template addresses the CAST-specific wording; I'll verify it in the source. On the error code, I accept the consistency argument: keeping `UNSUPPORTED_OPERATION` here and handling a data-oriented code across all sites in a separate change is reasonable, and it is not a blocker for this PR. **Docs (F2)** – You've confirmed the CAST wording is on both pages, but my concern was about what it says rather than whether it is there: the earlier text claimed "casting to an integral type rejects...", which overstates things since BIGINT still wraps for wide numerics, and it read as a changelog entry rather than reference documentation. Could you paste the current CAST note from `sql-functions.md` here, or confirm that it (a) scopes the rejection to the integral types that actually reject and (b) describes current behaviour rather than narrating before/after? Once I've gone through the diff at `22090e713` and the docs wording is confirmed, I expect to be able to approve. <!-- streview-comment:1575 --> -- 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]
