In my MDX report based on cube, the input help for date must be a calendar, therefore Date/Time type of parameter is essential. The field in time dimension by which I filter data is Integer. Example value: 20130827.
My dataset query looks like this:
SELECT NON EMPTY { [Measures].[Hours In Track] } ON COLUMNS, NON EMPTY {
([Dim Date].[Date ID].[Date ID].ALLMEMBERS * [Dim Division].[Hierarchy].[Division ID].ALLMEMBERS ) }
DIMENSION PROPERTIES MEMBER_CAPTION, MEMBER_UNIQUE_NAME ON ROWS FROM
( SELECT ( STRTOSET(@DimDivisionHierarchy, CONSTRAINED) ) ON COLUMNS FROM
( SELECT ( STRTOMEMBER(@FromDimDateDateID, CONSTRAINED) : STRTOMEMBER(@ToDimDateDateID, CONSTRAINED) ) ON COLUMNS FROM [BicepsArveCube]))
CELL PROPERTIES VALUE, BACK_COLOR, FORE_COLOR, FORMATTED_VALUE, FORMAT_STRING, FONT_NAME, FONT_SIZE, FONT_FLAGS
I used a text field, which showed me, that date/time type parameter value looks as follows: 2013-08-05 00:00:00, while my DateID is Integer type, so I need to do conversion.
When in dataset parameters tab I specify the parameter value as expression:
="[Dim Date].[Date ID].&["
& Replace(Replace("2013-08-05 00:00:00", "-", ""), " 00:00:00", "")
& "]"
I get the data as expected, everything works fine. But when I changed the hardcoded date/time value to parameter value (type date/time ):
="[Dim Date].[Date ID].&["
& Replace(Replace(Parameters!FromDimDateDateID.Value, "-", ""), " 00:00:00", "")
& "]"
I get a constraint violation error. I don't know why, because as written above, the Parameters!FromDimDateDateID.Value looks exactly the same as the hardcoded value I used.
I used text field to check what expression I get after conversion in both cases (hardcoded date and the same date chosen from calendar and passed as parameter value) and it looks exactly the same:
[Dim Date].[Date ID].&[20130805]