Merge 2 MDX queries | 易学教程

问题

I would like to show outcome of these 2 queries in one table.

SELECT
{[Measures].[Udzial Wieznia w Aktywnosci Count]} ON COLUMNS,
{[Aktywnosc].[Id Wiezienia].Members} ON ROWS
FROM [Wiezienia HD2]
WHERE [Udzial Wieznia w Aktywnosci].[Id Ucieczki].[2] : [Udzial Wieznia w Aktywnosci].[Id Ucieczki].[101]

and

SELECT
{[Measures].[Udzial Wieznia w Aktywnosci Count]} ON COLUMNS,
{[Aktywnosc].[Id Wiezienia].Members} ON ROWS
FROM [Wiezienia HD2]

First one shows how many escapes from jail there were in each jail, and second one show how many people are in each jail.

回答1:

This is the suggestion of Aaron West ported over to AdvWrks:

SELECT 
  {
    [Measures].[Internet Sales Amount]
   ,Aggregate
    (
        [Date].[Calendar].[Calendar Year].&[2006]
      : [![enter image description here][1]][1]
        [Date].[Calendar].[Calendar Year].&[2007]
     ,[Measures].[Internet Sales Amount]
    )
  } ON 0
 ,[Product].[Category].[Category] ON 1
FROM [Adventure Works];

As pointed out by Greg Galloway it errors:

Query (2, 3) The function expects a tuple set expression for the 2 argument. A string or numeric expression was used.

If we move the calculation to a WITH statement:

WITH 
  MEMBER [Measures].[SalesIn2006/2007] AS 
    Aggregate
    (
        [Date].[Calendar].[Calendar Year].&[2006]
      : 
        [Date].[Calendar].[Calendar Year].&[2007]
     ,[Measures].[Internet Sales Amount]
    ) 
SELECT 
  {
    [Measures].[Internet Sales Amount]
   ,[Measures].[SalesIn2006/2007]
  } ON 0
 ,[Product].[Category].[Category] ON 1
FROM [Adventure Works];

We now get a result:

So against your cube it'll probably look a little like this:

WITH 
  MEMBER [Measures].[Count freeAtLast] AS 
    Aggregate
    (
        [Udzial Wieznia w Aktywnosci].[Id Ucieczki].[2]
      : 
        [Udzial Wieznia w Aktywnosci].[Id Ucieczki].[101]
     ,[Measures].[Udzial Wieznia w Aktywnosci Count]
    ) 
SELECT 
  {
    [Measures].[Udzial Wieznia w Aktywnosci Count]
   ,[Measures].[Count freeAtLast]
  } ON 0
 ,[Aktywnosc].[Id Wiezienia].MEMBERS ON 1
FROM [Wiezienia HD2];

回答2:

WITH MEMBER [Measures].x AS 
    AGGREGATE( [Udzial Wieznia w Aktywnosci].[Id Ucieczki].[2] 
             : [Udzial Wieznia w Aktywnosci].[Id Ucieczki].[101]
            , [Measures].[Udzial Wieznia w Aktywnosci Count] )

SELECT { [Measures].[Udzial Wieznia w Aktywnosci Count]
, Measures.x
    } ON COLUMNS
, [Aktywnosc].[Id Wiezienia].Members ON ROWS
FROM [Wiezienia HD2]

If AGGREGATE doesn't work, try SUM. By the way, be careful that the OrderBy is set to KeyColumn (which should be a numeric datatype) on that attribute that is 2 to 101, or your range won't work correctly.

Edit: Perhaps my incorrect first answer put the Aggregate inline because I was thinking of the fact that a tuple set expression can act a little like a SUM, if you have a single hierarchy member to tuple with. Eg, if you created a dimension attribute for the range 2 to 101, you could use an expression like this:

( [Udzial Wieznia w Aktywnosci].[new attribute name].[2_to_101]
  , [Measures].[Udzial Wieznia w Aktywnosci Count] )

But then you would need to ensure that the dimensionality of the tuples is the same in all members on the axis, eg:

SELECT { ( [Udzial Wieznia w Aktywnosci].[new attribute name].[2_to_101]
  , [Measures].[Udzial Wieznia w Aktywnosci Count] )
    ( [Udzial Wieznia w Aktywnosci].[new attribute name].[All]
      , [Measures].[Udzial Wieznia w Aktywnosci Count] )
    } ON COLUMNS
, [Aktywnosc].[Id Wiezienia].Members ON ROWS
FROM [Wiezienia HD2]

来源：https://stackoverflow.com/questions/34232013/merge-2-mdx-queries

标签

merge

mdx