Hi all-
I've been creating pivot tables based on AS cubes, then (OLAP tools) converting to cube formulas. I've been asked to start on a new project, where the specific need will be to group data based on a pattern rather than a specific element in the cube. For example, in the past I might have used the following three parameters in a CUBEVALUE formula
[Food].[Fruit].[8133_Apple_Fuji]
[Date].[2014_01]
[measure].[Sales]
Part 1 of my question:
First, in this new project there are situations where I need to find (for example), the total Jan sales forevery fruit that contains the string "berry" (blueberry, strawberry, blackberry, raspberry)
[Food].[Fruit].[1835_Berry_BlueCrop]
Does Excel allow this type of wildcard in a CUBEVALUE formula? If so, what is the appropriate syntax?
Part 2 of my question:
I also need to be able to create categories of exclusion based on wildcards- for example, total Jan sales for every fruit that isn't a berry, or even more specifically, where "berry" isn't at a specific location in the element. For example, let's look at elements with the following naming convention:
[Food].[Fruit].[1835_Berry_BlueCrop] exclude
[Food].[Fruit].[1639_Berry_BlueJersey] exclude
[Food].[Fruit].[1871_Berry_HighBush] exclude
[Food].[Fruit].[1727_Watermelon_BerrySweet] retain
I'd need to exclude only the elements with "berry" starting at character 6, but retain another food which happens to have berry elsewhere in the string. Is there any way to pass location-specific wildcard strings inside a CUBEVALUE formula?
The closest I've found so far is the last post here: http://www.tm1forum.com/viewtopic.php?f=3&t=7406 but I haven't been able to convert that to working CUBESET/CUBEVALUE syntax and I don't know if that is even valid within a cube formula.
Many thanks for any help, or pointers to websites or blogs I haven't found yet