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
The time_agg operator has several definition issues that would be worth discussing. A first list here:
periodIndFrom parameter: Looks like it is not necessary. The period in from is taken from the input time or time_period. On top, examples 2 to 4 are written without the second parameter,
The operator should apply only to time_period types. With dates, we would need to do an implicit casting, which is not possible. For time, they don't have a period, so they should be casted to time_period, and it is not possible.
We should separate the cases for aggregation and the cases for datasets without aggregation. Execution inside an aggregate operator is actually at component level, not dataset level. In fact, the manual says that at dataset level, the execution applies to a measure, while when applied for aggregation, it applies to the identifier. I think this point should be clarified in the manual.
Proposed Solution
Review the definition of the operator
The text was updated successfully, but these errors were encountered:
Issue Description
The time_agg operator has several definition issues that would be worth discussing. A first list here:
Proposed Solution
Review the definition of the operator
The text was updated successfully, but these errors were encountered: