You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently the way we group or stack values is by executing multiple queries and joining them. There is another way though. We could instead pass multiple fields to the groupByFieldsForStatistics and make a single query and then convert the response into the same format that is currently the output of the join operation.
This is how Ops Dahsboard does their split by queries and I feel like it should be the default way that cedar does them. For example, would be much nicer if the radar example had 1 query instead of 9.
In order to do this, we'd likely need to add something like splitBy or splitByFields to the definition so we know how to process the query results. We should look at the Ops Dahsboard API (and maybe even source code 😈) for "inspiration."
The text was updated successfully, but these errors were encountered:
Currently the way we group or stack values is by executing multiple queries and joining them. There is another way though. We could instead pass multiple fields to the
groupByFieldsForStatistics
and make a single query and then convert the response into the same format that is currently the output of thejoin
operation.This is how Ops Dahsboard does their split by queries and I feel like it should be the default way that cedar does them. For example, would be much nicer if the radar example had 1 query instead of 9.
In order to do this, we'd likely need to add something like
splitBy
orsplitByFields
to the definition so we know how to process the query results. We should look at the Ops Dahsboard API (and maybe even source code 😈) for "inspiration."The text was updated successfully, but these errors were encountered: