mbutrovich opened a new pull request, #7698: URL: https://github.com/apache/datafusion-comet/pull/7698
## Which issue does this PR close? Closes #7695. ## Rationale for this change See #7695. ## What changes are included in this PR? - Try-mode `+`, `-`, and `*` on `Int8` through `Int64` use a new kernel. One pass computes the wrapping values and whether any row overflowed. A second pass, which builds the validity of the rows that didn't overflow, runs only when one did. - A scalar operand is no longer expanded to an array. Two scalar operands return a scalar, as the ANSI path already does. - Values under null slots are left as computed instead of zeroed. The Arrow format doesn't require a null slot to hold a particular value ([`Columnar.rst`](https://github.com/apache/arrow/blob/373d64c02cb23663b63c5853099fca43a46c39fc/docs/source/format/Columnar.rst?plain=1#L354-L356)). - Try-mode division keeps the existing kernel. - `benches/checked_arithmetic.rs` gets try-mode cases for `Int32` and `Int64` `add` and `mul`. The inputs have no overflow, one overflow per batch, random overflows in about 1% of rows, and overflows only under nulls. There are also a scalar case and a chained `try_add(try_add(a, b), c)`. The ANSI cases are the control, since this PR doesn't touch them. ## How are these changes tested? The tests are in `native/spark-expr/src/math_funcs/checked_arithmetic.rs`. `test_try_integer_arithmetic_matches_i128` compares every row with exact `i128` arithmetic for each type and op. It sweeps pairs of edge values in the array, scalar, and null-scalar layouts. It also moves one overflowing row through sliced batches of 300 rows, with the row valid and with it null, and with overflowing values under the other nulls. It also passes against `main`, so it checks that the new kernel changes no result. ## Benchmarks TODO # AI usage disclosure I wrote this PR with help from Claude but understand and support the core approach, implementation, and test coverage. -- 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]
