Hello,


I have been searching newsgroups and the web high and low to find a solution
to my problem. It seems to be a fairly common problem, but I have not found
a proper fix for it.

Below is my MDX, which might as well be off the Foodmart DB:

WITH SET [Date Range] AS '{ [Created].[2002].&[11].[25] :
[Created].[2003].&[9].[9] }'
MEMBER [Created].[Range] AS 'Aggregate([Date Range])'
SELECT { FILTER([Campaign].&[1].Children, [Measures].[Impressions] > 0) }
DIMENSION PROPERTIES [Campaign].[Campaign].Key, [Campaign].[Campaign].Name
ON ROWS,
{ [Measures].[Leads], [Measures].[Impressions], [Measures].[Gross CPL],
[Measures].[eCPL]} ON COLUMNS
FROM iLead_Dashboard
WHERE [Created].[Range]

This works fine. The numbers 2002, 11, 25, 2003, 9 and 9 are populated based
on user input, however. This means, that I cannot be sure that the start
date and end date exist in my "Created" dimension as a members. This, in
turn, would of course cause an error looking something like this "Formula
error - cannot find dimension member ("[Created].[xxxx].&[xx].[xx]") - in a
name-binding function".

How can I work around this (while maintaining functionality) or simply catch
the error and select the nearest date after and before the start date and
end date, respectively?

Any input is appreciated.

Thanks!


/Martin

The best way to solve this is to only present the user with a valid list of
dates from which there is data available in the cube.

Sean

--
--
Sean Boon
SQL Server BI Product Unit

**This posting is provided AS IS, with no warranties, and confers no
rights.**

"main" <martinskeem (AT) hotmail (DOT) com> wrote