TheR1sing3un opened a new pull request, #10093: URL: https://github.com/apache/paimon/pull/10093
### Purpose Preserve fields that first appear in later Python source rows passed to `merge().execute()`. Untyped Arrow row inference previously used only the first row's keys, silently omitting later values from updates and storing nulls for inserts. Infer columns from the union of keys across the source rows. Keep source-only fields available to explicit assignments and preserve the existing partial-update rule: target columns absent from the entire source remain unchanged. Input rows are not modified. ### Tests - 29 relevant multimodal table tests passed on both PyArrow 16.0.0 and 19.0.1, covering merge, add, create-table and overwrite paths. - New real-table regressions cover later fields in matched updates and inserts, omitted target columns, mapped keys, and source-only fields. Both regressions fail against the original implementation. - Python 3.6 syntax, flake8, license headers and `git diff --check` passed. -- 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]
