rangareddy commented on issue #17247: URL: https://github.com/apache/hudi/issues/17247#issuecomment-5366114759
This issue was reviewed as part of the JIRA-migrated backlog triage (HUDI-7306). **Findings: needs a reproduction against a current build before it can be dispositioned.** The report is concrete: CustomKeyGenerator writes `2023-03-04 14:44:42.046000` for an input of `2023-03-04 14:44:42.046661`, i.e. it truncates the sub-millisecond digits. Reproducing it looks tractable - the description includes a complete Faker/pandas script that produces the wrong value directly. I have not called this from the source. The surrounding code has been rewritten substantially since this was filed - table versions 8 and 9, the `BufferedRecordMerger` family replacing payload-based merging, the file-group-reader migration, the indexer abstraction in #18348 / #18372, and the `hudi-common` package moves in #19193 / #19195. Inferring either "still broken" or "silently fixed" through that much churn would be a guess. There is also no commit anywhere in the tree referencing HUDI-7306, so if it was fixed it was incidental rather than deliberate - another reason to verify rather than assume. The output in the description is precise and diagnostic: `.046000` from `.046661` is not rounding, it is millisecond truncation zero-padded back to six digits. That points at a millisecond-granularity intermediate (a `long` of epoch millis, or a `SimpleDateFormat`/`DateTimeFormatter` pattern with `SSS`) in the partition-value formatting path. Triage together with #17224 (HUDI-4430), which also isolates a `CustomKeyGenerator`-specific timestamp fault, and with the wider timestamp set #17222 and #17302. Keeping this open. -- 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]
