All, I think I've discovered a bug in IP but I want to verify that this isn't a case of "a feature, not a bug".
1. I create a r/t cube for planning, a simple DSO (to create a characteristic relationship), and then I create the aforementioned characteristic relationship.
For simplicity sake, my cube involves 0MATERIAL, 0VERSION, 0SALESORG, 0SALES_OFF, 0SALES_GRP, 0CURRENCY, and 0AMOUNT. My characteristic relationship only uses 0SALESORG, 0SALES_OFF, and 0SALES_GRP and is set to "Excl. #". The "Excl #" makes it display better.
2. Because I want to plan on some NULL (#) values for 0SALES_OFF and 0SALES_GRP, I include them in my DSO. For example, I have a line w/ 1100 for 0SALESORG, 0003 for 0SALES_OFF, and no value for 0SALES_GRP. This allows me plan data at that level on my cube even though my char relationship is set to "Excl #".
3. I then create a simple planning function to copy data, only changing the 0VERSION. However, when I try to execute the planning function IP throws the following error:
"To-value 1 for sub-process 1 produces following errors
Cannot find reference data for characteristic combination"
If I change the char relationships not to "Excl #" then I can run the Copy Planning Function without a problem.
It seems to me if I can plan data manually then I should also be able to plan it with a function. So my question is .... is this a bug? Should I be able to do this? If not, what part shouldn't I be able to do?
I'd open an OSS note but experience tells me SAP will dismiss it as a "consulting issue".
Ok, just to inform all. I opened up this issue as an OSS message and received an interesting response from SAP.
"I would like you to refer to note 994853 and pdf document attached with it. This note describes the usage of 'Excl #' in aggregation level, This flag disables the automatic valid combination for the relation. It is used in relation to 'top down distribution' and disaggregation of key figure.
The disaggregation supports only manual input and an aid in input queries. This is not applicable for planning functions. For the same reason when you try to use this flag with copy function it raises a message. Please refer to document for further reference. I hope this will help."
In the pdf attached to this note section 2.4 is supposed to provide a planning scenario using this setting. But section 2.4 of this document just says
"Coming soon ... (sorry)."
So I inquired about 2 things.
First, I asked about the incomplete document.
Second, I said that from this response it sounds like in top-down planning scenarios one can either manually plan or use planning functions but cannot reasonably do both because in order to manually plan without generating a cartesian product you must use the "Excl. #" option. I asked if this was correct.
My question about the completeness of the document was not acknowledged. To my second question I was told that basically this is correct and that "Top down functionality is getting improved. Please discuss the issue in SDN forum for further help".