I am using dimensional reporting and my requirement is
I have a customer dimension , in which the customers are the lowest level.
There are n customers . I will group certain customers and give it a name CustomerGroup1 and I will group certain other customers and give it a name Customer Group2
Now my value prompt will have these two data items (i.e) CustomerGroup1 and CustomerGroup2 and when user selects any of these my report should filter based on the selected group.
How should I achieve this ?
Thanks,
Nithya
Render variable?
Sounds to me like these groups should be a level in your dimension. Then you use standard filter expressions.
If they are not levels, then how are you grouping them? If it is based on a calculated data item, you can still use standard filters.
If these are custom groupings in your report, then as BigChris said, you may need some render variables and perhaps multiple data containers (i.e. crosstabs).
I am using a calculated data item to group the members in lower level .
How do I use standard filters in here.
Say for example I have grouped few customers like
member(Total(currentMeasure within set set([c1],[c2],[c3]),'Group1','Group1')
Now I need to have this Group1 , similarly Group2 in a prompt.
How would I add these Group1 and Group2 in prompt and how would it filter my data accordingly?
I have used two static values in my prompt and added a slicer expression to filter based on the above groups.. is this the write approach?
Thanks,
Nithya
Hello nithya1224,
Create your sets as separate data items. So, the expression for the data item [CustomerGroup1] would be set([c1],[c2],[c3]) and similarly the expression for the data item [CustomerGroup2] could be set([c4],[c5],[c6]).
In your prompt associated with parameter p_CustomerGroup for example, create two static choices with USE value set as [CustomerGroup1] and [CustomerGroup2] to match the data item names in your query.
Finally, drag a slicer into your query with the following expression:
#prompt('p_Group','token','[CustomerGroup1]')#
And that should filter the data.
-Cognos810