Re: [I] Filegroup name seems incorrect for log file created with NBCC [hudi]

["nsivabalan (via GitHub)" <gi...@apache.org>]

nsivabalan commented on issue #16671:
URL: https://github.com/apache/hudi/issues/16671#issuecomment-4795630728

   This appears to be fixed on master by [HUDI-8866 / #12627](https://github.com/apache/hudi/pull/12627) (commit `a93c6eec6788` — "Unified the file naming rules in NBCC mode").
   
   The root cause was that NBCC bulk-insert and upsert paths went through `BucketIdentifier.newBucketFileIdPrefix(bucketId, isNonBlockingConcurrencyControl)`, which used `replaceFirst(".{8}", bucketIdStr)` and produced an inconsistent file-id shape depending on which write path emitted it — matching the malformed `00000000-0000-0000-0000-0_…log.1_…` you observed.
   
   The fix introduces a dedicated `BucketIdentifier.newBucketFileIdForNBCC(...)` that produces a deterministic `{bucket_id}-0000-0000-0000-000000000000-0` format, and routes all NBCC bulk-insert / stream-write paths through it — `SparkBucketIndexPartitioner` / `BaseSparkBucketIndexBucketInfoGetter` (Spark) and `BucketStreamWriteFunction` / `BucketBulkInsertWriterHelper` (Flink). The PR also added `testBulkInsertAndInsertConcurrentCase1..5` in `TestSparkNonBlockingConcurrencyControl`, which cover the bulk-insert-after-upsert scenario from this report.
   
   If you can confirm with a current master build, this can probably be closed.


-- 
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: commits-unsubscribe@hudi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org