viirya commented on code in PR #8410:
URL: https://github.com/apache/arrow-datafusion/pull/8410#discussion_r1413217944
##########
datafusion/expr/src/window_frame.rs:
##########
@@ -148,18 +148,20 @@ impl WindowFrame {
pub fn regularize(mut frame: WindowFrame, order_bys: usize) ->
Result<WindowFrame> {
if frame.units == WindowFrameUnits::Range && order_bys != 1 {
// Normally, RANGE frames require an ORDER BY clause with exactly one
- // column. However, an ORDER BY clause may be absent in two edge cases.
+ // column. However, an ORDER BY clause may be absent in two edge cases:
+ // 1. start bound is UNBOUNDED or CURRENT ROW
+ // 2. end bound is CURRENT ROW or UNBOUNDED.
+ // In these cases, we regularize the RANGE frame to be equivalent to a
ROWS
+ // frame with the UNBOUNDED bounds.
Review Comment:
Note that actually Spark doesn't allow empty ORDER BY clause for RANGE frame
in any cases. I updated this according to the comment.
--
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]
