andygrove commented on PR #7130: URL: https://github.com/apache/datafusion-comet/pull/7130#issuecomment-6949531398
Thanks for following up on #7028 so quickly @dwsmith1983. The flipped checks and the new disjoint-struct test look right to me. That test is a nice proof that `A` now gets its own field id in `CometScanRule`. Could you add a `TIMESTAMP` child next to `x` in "nested struct field renamed by case only"? You checked timestamps locally, but that was the exact shape @rich7420 hit. A test would catch a future iceberg-rust bump that regresses it. The Spark 3.5 and 4.0 branches of the `a`/`A` tests are the ones that matter most here, and PR CI only runs 4.1. I've added `run-all-spark-profiles` so they run before merge. One note for whoever backports this. `branch-1.2` is still on the iceberg-rust pin from before #7028, so this fix would need #7028 to go first there. On its own it brings back the NULL nested reads. -- 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]
