ryukobayashi commented on PR #6794: URL: https://github.com/apache/hive/pull/6794#issuecomment-6187329761
@deniskuzZ I addressed your review comments in commit 4b01004b4b. The main changes are: - Replaced the unbounded `count(*) OVER (PARTITION BY file_path)` with a per-file `GROUP BY file_path` count joined to the representative row, avoiding unnecessary buffering and spill. - Included MERGE INSERT rows in the affected-row count, including insert-only MERGE and MERGE statements where no target rows match. - Separated the physical `RECORDS_OUT_*` counters from the logical `TOTAL_TABLE_ROWS_WRITTEN` affected-row count. - Added fallback behavior for mixed writers and protected writer closing/affected-row accounting from double processing. - Deduplicated replaced Iceberg data files by file path. - Made the CoW marker CTE name collision-safe and updated DELETE, UPDATE, and MERGE rewrites to use the generated CTE name. - Updated the Iceberg CRUD tests and qtest expected outputs. -- 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]
