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
When grouping by an entity, we add all the columns of the entity in the grouping key.
For entity which are mapped as JSON,
Should we add all the properties? If yes, how would we add the ordinal property which is non-persisted.
Given the JSON entity is going to be single SQL function call, we can just add the function call.
First option aligns with regular entity but would cause a lot of JSON scalar function calls. Second option will only have one call but the grouping key will contain owned entities in aggregate too. Further, grouping by a large JSON blob may not be something which user should do.
The text was updated successfully, but these errors were encountered:
Variation for #17653
When grouping by an entity, we add all the columns of the entity in the grouping key.
For entity which are mapped as JSON,
First option aligns with regular entity but would cause a lot of JSON scalar function calls. Second option will only have one call but the grouping key will contain owned entities in aggregate too. Further, grouping by a large JSON blob may not be something which user should do.
The text was updated successfully, but these errors were encountered: