TM1 Forum

declanr wrote:

Code: Select all

['Value']=>DB('Cube2', !Dim1, !Dim2, ElPar ( 'Product',!Product, 1), 'Value');

Normally I wouldn't use elpar in a feeder due to the possibility that your products have more than 1 parent and I would instead use an attribute. Either way though the concept of feeding to a higher level granularity dimension is fairly straight forward.

Alan Kirk wrote:

declanr wrote:

Code: Select all

['Value']=>DB('Cube2', !Dim1, !Dim2, ElPar ( 'Product',!Product, 1), 'Value');

Normally I wouldn't use elpar in a feeder due to the possibility that your products have more than 1 parent and I would instead use an attribute. Either way though the concept of feeding to a higher level granularity dimension is fairly straight forward.

Am I misreading? I got the impression that Willi was reading the consolidated value from the more detailed cube into the summarised cube. In such a case all that would be needed is the bang operator on the product dimension (assuming that the names were identical).

This theoretically feeds from the detailed consolidation to the summarised N elements. (Though in practice of course it's effectively be a feeder from all of the N level elements from each consolidation in the detail cube to the N elements of the summary cube.)

['Value']=N: DB ( 'Cube Detailed', !Dim1, !Dim2, !Product_MPL, 'value' );

['Value']=>DB('Cube2', !Dim1, !Dim2,!Product, 'Value');

declanr wrote:
My assumption was that a rule exists in the less detailed cube:

Code: Select all

['Value']=N: DB ( 'Cube Detailed', !Dim1, !Dim2, !Product_MPL, 'value' );

In which case the feeder (in more detailed cube):

Code: Select all

['Value']=>DB('Cube2', !Dim1, !Dim2,!Product, 'Value');

Wouldn't work since it just feeds from the n-level detailed elements that don't exist in 'Cube1' (less detailed) hence needing to do something to deduce the consolidation of the N-level element (which is itself the n-level element in Cube2).

Willi wrote:But unfortunately the Problem is that in the detailed cube the Elements in questions has more than one Level of childs. So in my understanding it's not enough to just use ELPAR!?

Willi wrote:Thx, I think 'll give that a try. But Do I have to put this on all N-Elements or only on the Elements I have in the Destination-Dimension?

sDimDetail = 'prod_detail';
sDimSummary = 'prod_summary';
sSubset = 'temp_subset';
sAttribute = 'feeder_product';
AttrInsert ( sDimDetail, '', sAttribute, 's' );

iCount = 1;
iMax = DimSiz ( sDimSummary );
While ( iCount <= iMax );
	sElement = DimNm ( sDimSummary, iCount );
	If ( DType ( sDimSummary, sElement ) @= 'N' );
		sMDX = '{TM1FILTERBYLEVEL( {TM1DRILLDOWNMEMBER( {([' | sDimDetail | '].[' | sElement | '])}, ALL, RECURSIVE )}, 0)}';
		SubsetCreateByMDX ( sSubset, sMDX, sDimDetail );
		nLast = SubsetGetSize ( sDimDetail, sSubset );
		sLast = SubsetGetElementName ( sDimDetail, sSubset, nLast );
		SubsetElementDelete ( sDimDetail, sSubset, nLast );
		SubsetElementInsert ( sDimDetail, sSubset, sLast, nLast );
		iSubCount = 1;
		iSubMax = nLast;
		While ( iSubCount <= iSubMax );
			sSubElement = SubsetGetElementName ( sDimDetail, sSubset, iSubCount );
			AttrPutS ( sElement, sDimDetail, sSubElement, sAttribute );
			iSubCount = iSubCount + 1;
		End;
		SubsetDestroy ( sDimDetail, sSubset );
	EndIf;
	iCount = iCount + 1;
End;

Willi wrote:But my question regarding the N-Elements was not specific enough because I did not mentioned that the elements of the detailed Dimension are C-Elements and the corresponding elements in the other dimensions are N_Elements.

Willi wrote:I currently have the Feeders defined manually like this:

Code: Select all

['Element 1','Value']=>DB('Cube2', !Dim1, !Dim2, 'Element 1', 'Value');
['Element 2','Value']=>DB('Cube2', !Dim1, !Dim2, 'Element 2', 'Value');
...

Where the Elements are C-Elements in Cube 1 (Source of Feeding). And this works. But I don't want to Change the Rule every time a new C-Element is defined. So I thought the Attribute at the C-Elements would be enough!?

['value'] => DB ( 'cube2', !dim1, !dim2, AttrS ( 'product', !product, 'feeder_attribute' ), 'value' );