The OneStream Community is temporarily frozen until June 29th due to the ongoing maintenance. Please read the blog post here to learn more.
Forum Discussion
Solved
Change derived data to stored
Hi, I am loading periodic data and in some periods there are derived amounts (due to no periodic data for some periods). The YTD values appear as derived data. However, the parent entity isn't roll...
From my experience, you have 2 options: either rectify the problem in Staging or the Cube.
Staging: this will depend on the source data. If that imports the Cube dimension members (ie all the transformation rules are * to *), then you could potentially write a derivative rule that creates a record with amount = 0 for all the intersections that are in the previous period but not the current. If the dimension members need transforming, then it is more tricky as you need to reference transformed data and create a new target data record.
Cube: have a calculation that checks the derived status (api.Data.GetDataCell("POV").CellStatus.IsDerivedData) and sets the value to be 0 if necessary. You can restrict the check to Origin Import.
Staging I think is probably the way to go if the source data does not need transforming. If that is not the case I think Cube is probably safer and certainly easier.
From my experience, you have 2 options: either rectify the problem in Staging or the Cube.
Staging: this will depend on the source data. If that imports the Cube dimension members (ie all the transformation rules are * to *), then you could potentially write a derivative rule that creates a record with amount = 0 for all the intersections that are in the previous period but not the current. If the dimension members need transforming, then it is more tricky as you need to reference transformed data and create a new target data record.
Cube: have a calculation that checks the derived status (api.Data.GetDataCell("POV").CellStatus.IsDerivedData) and sets the value to be 0 if necessary. You can restrict the check to Origin Import.
Staging I think is probably the way to go if the source data does not need transforming. If that is not the case I think Cube is probably safer and certainly easier.