Reputation: 13
SubSelect MDX Query as filtered list of main query
Hi all I want to write MDX query like to SQL:
select a, b, sum(x)
from table1
where b = "True" and a in (select distinct c from table2 where c is not null and d="True")
group by a,b
I try something like this:
`Hi all I want to write MDX query like to SQL:
select a, b, sum(x)
from table1
where b = "True" and a in (select distinct c from table2 where c is not null and d="True")
group by a,b
I try something like this:
SELECT
NON EMPTY { [Measures].[X] } ON COLUMNS,
NON EMPTY { [A].[Name].[Name]
*[B].[Name].[Name].&[True]
} ON ROWS
FROM
(
SELECT
{ ([A].[Name].[Name] ) } ON 0
FROM
( SELECT (
{EXCEPT([C].[Name].ALLMEMBERS, [C].[Name].[ALL].UNKNOWNMEMBER) }) ON COLUMNS
FROM
( SELECT (
{ [D].[Name].&[True] } ) ON COLUMNS
FROM [CUBE]))
)
But it returns me the sum of x from subquery.
How it should look like? '
Upvotes: 1
Views: 501
Reputation: 1484
Does X's measure group have relationship with D dimension? If it's true, the following code must just work:
Select
[Measures].[X] on 0,
Non Empty [A].[Name].[Name].Members * [B].[Name].&[True] on 1
From [CUBE]
Where ([D].[Name].&[True])
If you have many-to-many relationship, you need an extra measure (say Y):
Select
[Measures].[X] on 0,
Non Empty NonEmpty([A].[Name].[Name].Members,[Measures].[Y]) * [B].[Name].&[True] on 1
From [CUBE]
Where ([D].[Name].&[True])
Upvotes: 1