msaule commented on issue #21246:
URL: https://github.com/apache/datafusion/issues/21246#issuecomment-5171633679

   
   > Well I think the question is "is the programatically created plan valid, 
or does it violate some (potentially unstated) invariant"?
   > 
   > If there is some undocumented / unenforced invariant, we could potentially 
look into enforcing that / erroring rather than allowing
   
   Thanks, that makes sense. `ProjectionExec::try_new` currently accepts 
duplicate aliases and physical columns carry an index, so the plan is 
executable today. but since SQL rejects this shape earlier, I agree that does 
not necessarily mean it is supported.
   
   so i'd love to hear what you would prefer, that I make 
`ProjectionExec::try_new` reject duplicate output names with a test, or should 
filter pushdown preserve the column index for directly constructed physical 
plans?


-- 
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