[GitHub] mistercrunch commented on issue #5570: Make aggregation not mandatory

[GitBox <gi...@apache.org>]

mistercrunch commented on issue #5570: Make aggregation not mandatory
URL: https://github.com/apache/incubator-superset/issues/5570#issuecomment-410773587
 
 
   This is more complex than it may seem. The assumption that a metric is an aggregate function runs deep in Superset. The clear and easy workaround is to use a dummy aggregation function. min, max, sum should all provide the same answer in cases where dimension combinations are unique in the source data. The downside is a bit more work on the database side. Note that in some cases that can hurt quite a bit. For example for geospatial data visualizations where hundreds of thousands of data points can be surfaced, grouping by lat&long is expensive and often unnecessary.
   
   The committers have had lengthy conversations about this, and the key may be something around the `MetricsControl` allowing users to specify not to aggregate. It may have to be coupled with validation/assertions on resulting dataframes making sure that dimension combinations are not duplicated in some cases, as that may cause problems with some of the visualizations.
   
   I don't see this work getting prioritized anytime soon though as the work would be complex and there are clear easy workarounds.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: notifications-unsubscribe@superset.apache.org
For additional commands, e-mail: notifications-help@superset.apache.org