Slicing over a named set

问题

This is a follow up question to this previous question

I have the following working code:

WITH 
SET NewIDs AS
    INTERSECT(NonEmpty([ID].Children, CROSSJOIN([Date].&[T-1], [Category].&[Old]))
        , NonEmpty([ID].Children, CROSSJOIN([Date].&[T], [Category].&[EOD]))
        )
SELECT 
NON EMPTY([Dim2].Children
    ) ON 0
,
NON EMPTY([NewIDs]
    ) ON 1
FROM [MyCube]
WHERE [Measures].[Value]

However I would like to have another dimension on rows, not the ID (let's call it Dim1)

I have tried the below, but it does not work (as I don't think you can slice on named sets)

WITH 
SET NewIDs AS
    INTERSECT(NonEmpty([ID].Children, CROSSJOIN([Date].&[T-1], [Category].&[Old]))
        , NonEmpty([ID].Children, CROSSJOIN([Date].&[T], [Category].&[EOD]))
        )
SELECT 
NON EMPTY([Dim2].Children
    ) ON 0
,
NON EMPTY([Dim1].Children
    ) ON 1
FROM [MyCube]
WHERE [Measures].[Value]
, [NewIDs]

EDIT: Some further clarity. I usually aggregate the Value measure, and look at it in a table of Dim2 on columns and Dim1 on rows (this will include all IDs). However now I want to just have this but sliced only on the [NewIDs] set.

EDIT2: Using the below code almost solves my question, but on rows I have an unwanted column where every value is NewIDs ideally which I would like not to have.

WITH 
SET NewIDs AS
    INTERSECT(NonEmpty([ID].Children, CROSSJOIN([Date].&[T-1], [Category].&[Old]))
        , NonEmpty([ID].Children, CROSSJOIN([Date].&[T], [Category].&[EOD]))
        )
MEMBER [ID].[NewIDs] AS
       Aggregate([NewIDs])
SELECT 
NON EMPTY([Dim2].Children
    ) ON 0
,
NON EMPTY([ID].[NewIDs] * [Dim1].Children
    ) ON 1
FROM [MyCube]
WHERE [Measures].[Value]

回答1:

Do you get an error if you aggregate before adding to the WHERE clause?

WITH 
SET [NewIDs] AS
    INTERSECT(
          NonEmpty(
             [ID].Children, 
             ([Date].&[T-1], [Category].&[Old])
          )
        , NonEmpty(
             [ID].Children, 
             ([Date].&[T], [Category].&[EOD])
          )
        )
MEMBER [ID].[All].NewIDs AS
    AGGREGATE( [NewIDs] )
SELECT 
NON EMPTY([Dim2].Children
    ) ON 0
,
NON EMPTY([Dim1].Children
    ) ON 1
FROM [MyCube]
WHERE 
( [Measures].[Value]
, [ID].[All].NewIDs );

Try chucking the whole expression into the WHERE clause:

SELECT 
     NON EMPTY([Dim2].Children) ON 0
    ,NON EMPTY([Dim1].Children) ON 1
FROM [MyCube]
WHERE 
( [Measures].[Value]
, AGGREGATE( 
  INTERSECT(
          NonEmpty(
             [ID].Children, 
             ([Date].&[T-1], [Category].&[Old])
          )
        , NonEmpty(
             [ID].Children, 
             ([Date].&[T], [Category].&[EOD])
          )
        )
 ) 
);

来源：https://stackoverflow.com/questions/49980765/slicing-over-a-named-set

标签

mdx