parthchandra commented on PR #4256:
URL:
https://github.com/apache/datafusion-comet/pull/4256#issuecomment-4435954353
> Thanks @parthchandra. This LGTM. Could you add this test?
>
> ```
> -- MinSparkVersion: 4.1
> -- Config: spark.sql.timeType.enabled=true
>
> statement
> CREATE TABLE test_to_time_col_fmt(s STRING, f STRING) USING parquet
>
> statement
> INSERT INTO test_to_time_col_fmt VALUES
> ('14.30.00', 'HH.mm.ss'),
> ('1230', 'HHmm')
>
> -- A non-foldable format column should fall back to Spark because Comet
does
> -- not implement the format-pattern variant of to_time. Spark's
> -- ToTime.replacement uses Invoke(Literal(ToTimeParser(None)), "parse",
> -- Seq(str, format)) when the format is non-foldable, so parser.fmt is
None
> -- and the current shim guard (parser.fmt.isEmpty) is satisfied. The shim
> -- should additionally require args.size == 1 to let this case fall
through.
> query expect_fallback(invoke is not supported)
> SELECT to_time(s, f) FROM test_to_time_col_fmt
> ```
added the test and the fix needed for it. Good catch!!
--
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]
