Hi,
Many of our users want to export the underlying data of our cubes. I have created some reports where they can choose the columns and it builds a rather verbose MDX statement with many nonemptycrossjoins. This has its obvious performance issues when many columns are required in the result set.
I would like to implement DRILLTHROUGH instead, much like Analyzer does in Pentaho EE. So I have the following example query:
However this doesn't work if i just enter the code into the usual 'mdx over mondrianJndi' connector. Should it?
Should I be using the 'olap4j over olap4j' connector instead? If so - I cannot get this to work at all. Does anybody have any good documentation on how to set this up properly (for the same JNDI connection if possible?)
Thanks!
Andrew
Many of our users want to export the underlying data of our cubes. I have created some reports where they can choose the columns and it builds a rather verbose MDX statement with many nonemptycrossjoins. This has its obvious performance issues when many columns are required in the result set.
I would like to implement DRILLTHROUGH instead, much like Analyzer does in Pentaho EE. So I have the following example query:
Code:
DRILLTHROUGH
SELECT
{[Measures].[Activity Count]} ON COLUMNS
,NON EMPTY {[Activity Data.Business Unit].MEMBERS} ON ROWS
FROM [Activity]
RETURN [ActivityType.Activity Type].[activityTypeDesc], [Activity Data.Business Unit].[ownerAgencyName]Should I be using the 'olap4j over olap4j' connector instead? If so - I cannot get this to work at all. Does anybody have any good documentation on how to set this up properly (for the same JNDI connection if possible?)
Thanks!
Andrew