EmilyMatt commented on PR #22945:
URL: https://github.com/apache/datafusion/pull/22945#issuecomment-4707198506

   > It would be great to construct a end-to-end reproducer (ideally at SQL 
level) for the following reasons.
   > 
   >     * I think this implementation makes sense to me, but it inevatiably 
introduced some complexity inside operator, if we can know the end-to-end goal 
from the specific workload, we can try to think about is there any easier 
alternative.
   > 
   >     * The UTs are very low-level, if we decide to refactor the 
implementation someday, this coverage is very likely to get lost, an e2e test 
would be more robust to the rapidly moving codebase.
   
   I believe I've added an e2e test, but let me know if it is not what you 
intended. can confirm it fails without this implementation


-- 
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]

Reply via email to