SEZ9 commented on issue #12940: URL: https://github.com/apache/seatunnel/issues/12940#issuecomment-6487260090
Thanks for opening #13026 now that #12900 has merged. Since this issue was investigation-first, please make sure the PR description explains how the allocation numbers were measured (tool, fixture shapes, before/after runs) so the reported gain can be reproduced during review. On the approach: whichever reuse strategy the PR takes, please avoid anything keyed only by field name, since two ROWs with the same leaf name under different parents must keep distinct schemas, and keep the existing byte-compatibility and nested Kafka/Pulsar round-trip regressions as the minimum boundary. Coverage for null and present ROWs at every depth, including ROWs inside ARRAY and MAP, under both the default `avro_row_nullable = false` layout and the nullable union layout would also help confirm that serialized bytes and record names are unchanged. I'll take a look at the PR. <!-- streview-comment:2121 --> -- 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]
