Details
-
Type:
Bug
-
Status:
Closed
-
Priority:
Minor
-
Resolution: Fixed
-
Affects Version/s: 3.1 GA
-
Fix Version/s: Future Release
-
Component/s: None
-
Labels:None
Description
The query
select
{[Measures].[Store Sales],[Gender].[M]} on columns,
{[Store].[All Stores].[USA].children} on rows
from Sales
gives values for ([Measures].[Store Sales], [Gender].[All
Gender]) in the 2nd column, but it should be the default
measure, ([Measures].[Unit Sales], [Gender].[M]).
The problem is that the axis is heterogeneous --
[Measures].[Store Sales] and [Gender].[M] are from
different hierarchies -- and mondrian optimizes query
execution by assuming that they are homogeneous.
For now, we should throw an error that we do not
support heterogeneous axes. (MSOLAP does not.)
If we ever allow heterogeneous axes, we need to be
more selective how we apply the optimization.
select
{[Measures].[Store Sales],[Gender].[M]} on columns,
{[Store].[All Stores].[USA].children} on rows
from Sales
gives values for ([Measures].[Store Sales], [Gender].[All
Gender]) in the 2nd column, but it should be the default
measure, ([Measures].[Unit Sales], [Gender].[M]).
The problem is that the axis is heterogeneous --
[Measures].[Store Sales] and [Gender].[M] are from
different hierarchies -- and mondrian optimizes query
execution by assuming that they are homogeneous.
For now, we should throw an error that we do not
support heterogeneous axes. (MSOLAP does not.)
If we ever allow heterogeneous axes, we need to be
more selective how we apply the optimization.