david-mollitor-db commented on PR #58972:
URL: https://github.com/apache/spark/pull/58972#issuecomment-5798354970
Thanks, @dongjoon-hyun -- both addressed:
1. **Description / commit message.** You're right, corrected both.
`CheckOverflow` never mixed in the `NullIntolerant` trait (checked
`a84ca5e46e6^`: there `UnscaledValue` and `MakeDecimal` had the trait,
`CheckOverflow` did not), so it was never declared null-intolerant -- a
pre-existing gap, independent of SPARK-50241.
2. **Tests.** Added both of your examples:
- `InferFiltersFromConstraintsSuite`: `IsNotNull('d)` is inferred from
`CheckOverflow('d, ...) > 0`.
- a new `NullPropagationSuite`: `CheckOverflow(Literal(null, ...), ...)`
folds to a null literal under a NullPropagation-only batch, so the fold is
attributable to the new null-intolerance rather than plain constant folding.
Force-pushed. Thanks again for the careful review!
--
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]
