paleolimbot opened a new issue, #167:
URL: https://github.com/apache/sedona-db/issues/167

   The framework for writing an aggregate function implementation currently 
doesn't expose the `GroupsAccumulator`:
   
   
https://github.com/apache/sedona-db/blob/a93b322f0f60e5927cb21e33a48afb130ac42c61/rust/sedona-expr/src/aggregate_udf.rs#L162-L181
   
   ...which is an important optimization for low cardinality groups:
   
   
https://github.com/apache/datafusion/blob/5cc0be51f9b60224c0667b89dd0318a591e733d9/datafusion/expr/src/udaf.rs#L581-L605
   
   There are a few potentially interesting things we could exploit in our 
aggregate functions but this is probably the most important one (or perhaps we 
need to separate the dispatch mechanism so that we don't have to repeat 
concepts when certain functions might benefit from a deeper integration with 
DataFusion).


-- 
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