andygrove commented on PR #6457: URL: https://github.com/apache/datafusion-comet/pull/6457#issuecomment-5950932010
@parthchandra Thanks for the review. Addressed the actionable points: - The signed-zero SQL checks were rewritten in `eb5c7af4f` to use ordered windows (`first_value`/`last_value` by `id`) instead of depending on aggregate scan order. - `greatest`/`least` now validate Spark's minimum of two arguments in both `return_type` and invocation, with zero- and one-argument regression coverage. - `min_max_udaf` now routes only `Float32`/`Float64` through `SparkMinMax`; an internal `Float16` plan stays on DataFusion's min/max path rather than reaching an unsupported Spark accumulator. - No change to the grouped `get_unchecked_mut` path, as noted. The latest hardening is in `d460aff81`. Validation passed with `cargo test -p datafusion-comet-spark-expr greatest_least`, `cargo check -p datafusion-comet`, `make core`, formatting/diff checks, and the focused `CometSqlFileTestSuite min_max_floating_point` fixture in both strict and non-strict modes (2 tests). The earlier unrelated Scalafix dependency-resolution failure was left untouched because the fix is already on main. -- 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]
