kumarUjjawal commented on code in PR #24565:
URL: https://github.com/apache/datafusion/pull/24565#discussion_r3873915411
##########
datafusion/optimizer/src/analyzer/type_coercion.rs:
##########
@@ -1083,6 +1083,8 @@ fn extract_window_frame_target_type(col_type: &DataType)
-> Result<DataType> {
Ok(DataType::Interval(IntervalUnit::MonthDayNano))
} else if let DataType::Dictionary(_, value_type) = col_type {
extract_window_frame_target_type(value_type)
+ } else if let DataType::RunEndEncoded(_, value_type) = col_type {
Review Comment:
> perhaps do no.3 then raise an upstream issue for no.1
>
> thoughts @kumarUjjawal?
I was thinking we can put the unwrap back here, and make
`supports_offset_arithmetic` false for a run-encoded order key? That keeps the
plan time rejection for `1 PRECEDING` and `1 FOLLOWING`. An offset bound is the
one case that needs `add_checked` on a wrapped value. The user then gets a plan
error, not the internal "file a bug report" error.
--
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]
