Gabriel39 opened a new pull request, #69002:
URL: https://github.com/apache/doris/pull/69002
### What problem does this PR solve?
The OceanBase JDBC all-types golden adds an extra eight hours to non-null
TIMESTAMP values even though the fixture and Doris result rendering use
Asia/Shanghai. Regenerate the expected output from the existing fixture and add
a comment explaining the timezone expectation.
### Release note
None
### Check List (For Author)
- Test:
- [x] Reproduce the timestamp mismatch with the original golden.
- [x] Regenerate the golden with the regression runner and pass ordinary
comparison mode.
- [x] Confirm only the three non-null TIMESTAMP cells change.
- Behavior changed: No production behavior changes.
- Does this need documentation: No.
Validation used the repository OceanBase 4.2.5-test image, unchanged
initialization SQL, and OceanBase JDBC driver 2.4.8. The source session/global
timezone was +08:00; switching the source session to UTC changed timestamp
display but preserved the epoch. The original golden failed by eight hours;
generated output passed ordinary comparison (1 suite, 0 failures, 0 skipped).
Only three non-null TIMESTAMP cells changed; all other columns and NULLs stayed
identical.
Local validation reused existing Doris binaries (FE `5024d4e894`, BE
`a48a5221001`), rather than a fresh build of this master checkout. The
unrelated binlog service was not used in this JDBC validation.
--
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]
