somandal commented on issue #10628: URL: https://github.com/apache/pinot/issues/10628#issuecomment-1513536612
> however in Pinot's use case this should actually be sufficient to use INNER JOIN b/c the left userUUID we can guarantee to be non-nullable (otherwise we can throw?) thus we don't need to handle this situation. was wondering if that's better. > > @somandal was adding the BOOL_AND and BOOL_OR resolves the issue? it is "one" solution but i don't think it is ideal b/c it makes several extra operators that does basically nothing :-/ BOOL_AND/BOOL_OR should solve the issue (have to actually try it out) but yes this may not be the ideal solution for this problem. Are you recommending a rule rewrite for this scenario? In general though, if a plan tries to run MIN/MAX on a Boolean column though, shouldn't the aggregation functions handle this scenario correctly? I still think adding a fix to handle bools correctly is a useful fix to add. Today won't queries that run MIN()/MAX() on a boolean column fail? > but i don't think it is ideal b/c it makes several extra operators that does basically nothing :-/ I didn't understand this part, why do you say it makes extra operators that does basically nothing? The nodes and operators will still be the exact same as specified in the plan, we just decide which function to use based on the type. -- 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]
