Hi all
I need to build an active form looping two dimensions. The initial idea was to use 2 TM1RPTROW functions, but the problem is I need to filter dimension2 element based on the current element of dimension1 and AFAIK that's impossible.
Another solution would be to build a flag cube, but I'm wondering if I could build just a combined dimension to use it in TM1RPTROW. I tried that but Perspectives stops working when I use a dimension in TM1RPTROW which does not belong to the cube used in TM1RPTVIEW. For testing I even tried to copy 1st dimension (assuring it contains the same elements) and use it in TM1RPTROW, but it seems TM1RPTROW expects always to use a dimension from a cube used in TM1RPTVIEW.
1). Is it true?
2). Is there any other way to solve this without building a flag cube?
Thanks in advance!
TM1RPTROW driving dimension
-
- MVP
- Posts: 3698
- Joined: Fri Mar 13, 2009 11:14 am
- OLAP Product: TableManager1
- Version: PA 2.0.x
- Excel Version: Office 365
- Location: Switzerland
Re: TM1RPTROW driving dimension
First it sin't really clear from your explanation exactly what you want to do. Here might be a case of a (good) picture tells a lot of words.
Does TM1RptRow have to use a dimension from the cube references in TM1RptView? Yes. Of course yes! This is logical. And the dimension must be a dimension NOT referenced within the TM1RptView with a TM1RptTitle.
"I need to filter dimension2 element based on the current element of dimension1 and AFAIK that's impossible." Well not necessarily. You can reference the current member of another dimension in the cube as long as that dimension is on title/page/filter/context that is the member must be a constant value not variable. But you can't reference CurrentMember on another dimension if that dimension element is on rows or columns and therefore variable. Possibly you could use the MDX GENERATE function to achieve the "looping on another dimension" and combined with zero suppression it might give you what you want (which I think is akin to an asymetric selection) but GENERATE normally doesn't perform very well.
"Another solution would be to build a flag cube..." Yes possibly, or probably. Either build a numeric flag cube and use the flag cube for the active form row definition (just hide the column) then pull the rest of the values normally with an additional VIEW formula + DBRW. Or else your idea of a combined dimension in a cube and using that for the rowset then hide the TM1RptRow column and lookup the dimension combination from the hidden column. Tip: if you want to drive asymmetric row selection then a string based flag cube is the way to go as then the elements shown for consolidations can be independent of what is entered in leaf level for the descendants.
Does TM1RptRow have to use a dimension from the cube references in TM1RptView? Yes. Of course yes! This is logical. And the dimension must be a dimension NOT referenced within the TM1RptView with a TM1RptTitle.
"I need to filter dimension2 element based on the current element of dimension1 and AFAIK that's impossible." Well not necessarily. You can reference the current member of another dimension in the cube as long as that dimension is on title/page/filter/context that is the member must be a constant value not variable. But you can't reference CurrentMember on another dimension if that dimension element is on rows or columns and therefore variable. Possibly you could use the MDX GENERATE function to achieve the "looping on another dimension" and combined with zero suppression it might give you what you want (which I think is akin to an asymetric selection) but GENERATE normally doesn't perform very well.
"Another solution would be to build a flag cube..." Yes possibly, or probably. Either build a numeric flag cube and use the flag cube for the active form row definition (just hide the column) then pull the rest of the values normally with an additional VIEW formula + DBRW. Or else your idea of a combined dimension in a cube and using that for the rowset then hide the TM1RptRow column and lookup the dimension combination from the hidden column. Tip: if you want to drive asymmetric row selection then a string based flag cube is the way to go as then the elements shown for consolidations can be independent of what is entered in leaf level for the descendants.
Please place all requests for help in a public thread. I will not answer PMs requesting assistance.
- vovanenok
- Posts: 89
- Joined: Mon Jun 23, 2014 4:54 pm
- OLAP Product: TM1
- Version: 2.0.9
- Excel Version: Office 365
- Location: Toronto, Canada
- Contact:
Re: TM1RPTROW driving dimension
Thanks lotsaram
By dimension 1 and dimension 2 I meant dimensions in two TM1RPTROW formulas and as those are both row dimensions that's impossible to reference CurrentMember because those two TM1RPTROW functions cannot interact with each other.
By dimension 1 and dimension 2 I meant dimensions in two TM1RPTROW formulas and as those are both row dimensions that's impossible to reference CurrentMember because those two TM1RPTROW functions cannot interact with each other.
I meant using a different dimension but containing the same elements. The idea was to build a driving dimension and use its leaf elements which are exactly the same elements as in one of my report cube dimension, but also contain some hierarchies / attributes telling which elements should be selected for other cube dimensions. That would be kind a way to combine 2 dimensions in 1 and use it in TM1RPTROW function. Anyway it seems in TM1RPTROW function you can only use a dimension from TM1RptView cubelotsaram wrote:Does TM1RptRow have to use a dimension from the cube references in TM1RptView? Yes. Of course yes! This is logical. And the dimension must be a dimension NOT referenced within the TM1RptView with a TM1RptTitle.
-
- Regular Participant
- Posts: 152
- Joined: Sat May 25, 2013 10:32 am
- OLAP Product: TM1
- Version: 9.5.2; 10.2.2
- Excel Version: 2007
Re: TM1RPTROW driving dimension
Hi Dude,
It's about dimension 1 and dimension 2 having similar elements, dimension 2 should be dependent on dimension 1 is what I understood from details. Hope I got it correctly.
I would have tried with MDX statement which can be part of TM1RPTROW.
Regards,
Deepak Jain
It's about dimension 1 and dimension 2 having similar elements, dimension 2 should be dependent on dimension 1 is what I understood from details. Hope I got it correctly.
I would have tried with MDX statement which can be part of TM1RPTROW.
Regards,
Deepak Jain