andygrove opened a new pull request, #3494:
URL: https://github.com/apache/datafusion-comet/pull/3494
## Summary
- Fix incorrect timestamp timezone handling in the schema adapter on the
df52 branch
- INT96 Parquet timestamps coerced to `Timestamp(us, None)` by DataFusion
were being routed through Spark's `Cast` expression when the logical schema
expected `Timestamp(us, Some("UTC"))`. Spark's Cast treats `None`-timezone as
TimestampNTZ (local time) and applies a timezone conversion, shifting values by
the session timezone offset (e.g., -5h45m for Asia/Kathmandu)
- Route `Timestamp -> Timestamp` mismatches through `CometCastColumnExpr`
which delegates to `spark_parquet_convert`, handling this as a metadata-only
timezone relabel
## Test plan
- [ ] Existing test "SortMergeJoin with unsupported key type should fall
back to Spark" in `CometJoinSuite` should now pass
- [ ] Existing Rust tests (`parquet_roundtrip_int_as_string`,
`parquet_roundtrip_unsigned_int`) continue to pass
- [ ] CI passes
🤖 Generated with [Claude Code](https://claude.com/claude-code)
--
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]
