nsivabalan commented on issue #16941: URL: https://github.com/apache/hudi/issues/16941#issuecomment-4800695681
This appears to be fixed on current master. Commit [`3e45b527`](https://github.com/apache/hudi/commit/3e45b5279aab8412c7c866d23fddab529aad5c6a) (#13993, Sep 2025) introduced `HoodieRecord.getOrderingValueAsJava(...)` and changed `HoodieAppendHandle` to use it when constructing the `DeleteRecord` written to log files (see `HoodieAppendHandle.java:668`). `HoodieSparkRecord` overrides `getOrderingValueAsJava` (HoodieSparkRecord.java:373) to extract column values through `getColumnValueAsJava`, returning engine-agnostic Java types instead of Spark-internal types like `org.apache.spark.sql.types.Decimal`. A follow-up [`adbea027`](https://github.com/apache/hudi/commit/adbea027727abed49612a6d51743d8bc58c7c217) (#13998) added `OrderingValueEngineTypeConverter` to handle timestamp-millis / decimal type conversion on the read side as well. Could you verify on current master and close if resolved? -- 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]
