0

I am consuming dimensional data (TM1 cubes) with Report Studio. I want to combine the slice given by two (or more) tuple calls. This is because I want to include two elements from the same dimension that is not logically structured in a dimensional hierarchy. An example would be if I wanted to have combine the actuals for 2012 and 2013 without first making separate data items for both. Then I would have two expressions:

tuple([Actuals], [2012])
tuple([Actuals], [2013])

To illustrate what I want, I would ideally be able to write something similar to these suggestions:

tuple([Actuals], [2012], [2013])
tuple([Actuals], set([2012], [2013]))

However, these suggestions do not work. Hence, I was wondering if there is any other way of combining the slices given by two or more tuple calls, as described earlier in this post. Thanks in advance!

Skovly
  • 224
  • 1
  • 8
  • 22

2 Answers2

2

Combining different elements from 1 dimension means addressing 2 different cells (2012, 2013 in your example) and then doing something (summing up?) to their values. Tuple function always addresses a single cell, making your idea impossible.

But combining values for different tuples is usually something along the lines: tuple([Actuals], [2012]) + tuple([Actuals], [2013])

or

total ([Actuals] within set([2012],[2013]))

ykud
  • 166
  • 3
  • Thank you for your suggestions, ykud! I think I will be able to use them in a workaround to solve my real problem. However, in an effort to find a more elegant solution, I would like to follow-up. From my understanding, the tuple function does not necessarily return a single cell, but rather a slice of the cube (i.e. a set of cells, possibly only one). Using tuple with multiple elements from one dimension is therefore obviously not possible, as there will never be an intersection between these two elements. Hence; is there a different way to combine the slices provided by two tuple calls? – Skovly Nov 21 '13 at 08:01
  • Just for future searches: tuple always returns a single cell, current members in context or default members are picked from all the dimensions you don't specify explicitly. – ykud Nov 23 '13 at 14:06
  • Just for my own understanding, @ykud: What if the default member of some dimension not included in the tuple call is the root node of a hierarchy that covers all elements in that dimension? Would it not return multiple cells then (i.e. all elements in that dimension)? – Skovly Nov 25 '13 at 08:03
  • It will return the default member, single dimension element, that's the way OLAP engines work. You can have a different value calculated on that total member (like avg of children, or even a calculated %) and tuple call will return you that measure value. Unlike relational model, multidimensional operates on the idea that aggregate values exist as cells in the cube, not as combination of detail values. The logic behind aggregate calculation can be anything, not just basic sums and avg's. – ykud Nov 26 '13 at 11:51
0

The answer that I was looking for can be found here:

https://www.ibm.com/developerworks/community/forums/html/topic?id=6ef92a9e-0d0f-494b-984a-cdc3a82ca287&ps=25

Skovly
  • 224
  • 1
  • 8
  • 22