mustafasrepo opened a new issue, #6035:
URL: https://github.com/apache/arrow-datafusion/issues/6035

   ### Is your feature request related to a problem or challenge?
   
   Current `BoundedWindowAggExec` implementation assumes that its input is 
ordered both by `PARTITION BY` and `ORDER BY` columns. This enables us to 
calculate partition separation points easily and fast. However, this assumption 
results in some queries to break pipeline. Even if there is no theoretical 
reason to break pipeline.
   
   For instance, the query below
   ```
   SELECT SUM(inc_col) OVER(PARTITION BY unsorted_col ORDER BY low_card_col1 
ASC ROWS BETWEEN 2 PRECEDING AND 1 FOLLOWING) as sum1
   FROM annotated_data2
   ```
   can run theoretically when its input is ordered by `low_card_col1`. However, 
current executor requires its input to be ordered by `unsorted_col`, 
`low_card_col1`. To be able to calculate partition boundaries easily. 
   
   
   
   ### Describe the solution you'd like
   
   By changing partition calculation algorithm, we can accomplish so.
   
   ### Describe alternatives you've considered
   
   _No response_
   
   ### Additional context
   
   _No response_


-- 
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]

Reply via email to