How to Run a FOX Planning Function only on the data in your planning buffer

[mitko1994]

Hi Everyone,

I created a simple FOX formula PF that divides a KF by 100. It of course runs on a filter. My initial thought was that the PF will run only on the planning buffer, but it looks like it runs on everything in the filter regardless of the planning buffer. This makes sense but causes an unwanted behavior as it continuously divides all records that match the filter by 100. I tried ticking "Process Changed Data" in Analysis but it didn't change the behavior.Is there a way that I only run this PF on the data from the planning buffer(only the cells I change) ? If not, how would you suggest I display 10 but store 0.1 for example?

Thanks,

Dimitar

[mitko1994]

Hi Gregor,

Thank you for your reply! I have a column where the user enters a percentage. They are expected to enter the percent as one to one(ex: enter 1 for 1%), but I would like to store it as 0.01 in the cube afterwards on save. If I scale by 100 then I will store 100 for a user input of 1, right? At least that was my observation the last time I tried it. I think process changed data is exactly what I need but the function doesn't run on the records that are modified by the user only. Instead, it runs on all the records(maybe because of the merged filter?). That's why I have the issue of dividing the same percentage value by 100 over and over again even if the user hasn't modified it. Ideally I would like to have this percentage column with a scaling factor of 100 for display purposes but I would store the

[gregor_dieckmann]

Hi Dimitar,

this look like a work around for something: do you just want to have a kind of scaling factor that the query or Analysis Office does not support?

If yes, you can use an inverse formula: to display a key figure value with a factor of 100 define a formula F = A * 100 and the inverse formula A = F / 100.

Regards,

Gregor

P.S.

'Process changed data' reads the 'changed' data from the buffer, computes the projections of these records to the block characteristics (characteristics not to be changed) and these projections are used as a run time filter used in the function; this run time filter is merged with the original filter. The 'changed' data are the data from the last user 'LUW' (logical unit of work), i.e. the 'delta' records send to the buffer by the last user interaction.

[mitko1994]

Hi Gregor,

Thank you for your reply! I have a column where the user enters a percentage. They are expected to enter the percent as one to one(ex: enter 1 for 1%), but I would like to store it as 0.01 in the cube afterwards on save. If I scale by 100 then I will store 100 for a user input of 1, right? At least that was my observation the last time I tried it. I think process changed data is exactly what I need but the function doesn't run on the records that are modified by the user only. Instead, it runs on all the records(maybe because of the merged filter?). That's why I have the issue of dividing the same percentage value by 100 over and over again even if the user hasn't modified it. Ideally I would like to have this percentage column with a scaling factor of 100 for display purposes but I would store the values as 1/100 of the user's input. This will give me the 1 to 1 appearance and allow me to store the values as 1/100th of the input.