sam-1112 commented on PR #7198: URL: https://github.com/apache/datafusion-comet/pull/7198#issuecomment-6555425728
Thanks for the detailed feedback, @comphead. I've addressed the independent changes: - Restored the actual and expected column counts in the mismatch error, with regression coverage for both too few and too many columns. - Included the offending UUID value in the error message, limited to 128 Unicode characters with a truncation marker and escaped control characters. - Removed `child_visibility` and shared `repeated_visibility`. Conversion now uses `Option<&NullBuffer>`; structs skip the mask when everything is visible and reuse an existing ancestor bitmap when possible. The null-parent and sliced list/map cases remain covered. - Updated the eligibility table to list the UUID partition/sort-source restrictions. The #7100 integration is still pending. I'll coordinate with @yunchipang to preserve top-level `unknown` pruning when aligning batches, and replace the UUID + unknown fallback case with a native-write parity test on the integrated changes. -- 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]
