Zouxxyy opened a new pull request, #9322: URL: https://github.com/apache/paimon/pull/9322
### Purpose Fix Spark key-dynamic writes for Paimon tables containing timestamps whose precision cannot be represented in Spark schemas. The bootstrap path serializes rows with the table's original `RowType`, while the global-index output was decoded with a `RowType` reconstructed from the Spark schema. Spark loses the Paimon timestamp precision, so a `TIMESTAMP(3)` binary row was decoded as `TIMESTAMP(6)` and failed. Decode global-index data rows with `fileStoreTable.rowType()` and append the Spark row-kind and bucket columns separately. ### Tests - Added a cross-partition dynamic-bucket regression test using `TIMESTAMP(3)` and deletion vectors. - Spark 3 `DynamicBucketTableTest`: 8 tests passed. - Spark 4 shared sources compiled successfully. -- 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]
