sunchao commented on PR #5376: URL: https://github.com/apache/datafusion-comet/pull/5376#issuecomment-5337982345
Thanks @comphead! Moved the behavioral cases into the existing `arrays_overlap.sql` in 247e0b173. The SQL suite now compares against the running Spark version and covers runtime-generated NaNs, signed zeros, nulls, reversed operands, and both the short-array and hash-lookup paths for `FLOAT` and `DOUBLE`. I kept a small Scala raw-bit test because SQL equality cannot distinguish NaN representations. It verifies that runtime negation produces noncanonical NaNs after Parquet has canonicalized the stored values. Both focused integration tests pass locally on Spark 3.5, 4.0, and 4.2, and all 23 targeted Rust tests pass. -- 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]
