User bio
404 bio not found
Member since Nov 17, 2020

Hi Alexander, thanks for the help! 

Yes, crossjoining too many dimensions like that is generally a terrible practice if it is not a small dataset.  This is more a hack attempt to allow a mostly textual matrix (member names) to appear on the dash along with charts that retrieve from the cube.  I was hoping to use the cube to populate this matrix/pivot so that the filters that control the charts also control the matrix.  But it appears the best way it to use an SQL based KPI to retrieve from the underlying cube class, with the drawback that custom filtering has to be coded in the %OnGetSQL of the KPI with the additional challenge to mix MDX filters with an SQL query.

For the dashboard, in my case, the HIDDEN = ON appears to not do anything (column remains) and HIDDEN = OFF strangely only shows that column, it makes all the other columns disappear.  Maybe this is a bug with the 2020.1 version.  But this helps me understand what the data properties are supposed to do.  Since they didn't work right I wasn't too sure when I first checked them out.  I didn't realize you can format columns from here.  This helped so marking as answer.  Thanks!

HEAD returns a set, even if that set only has one member in it.  You will not be able to use the result of HEAD as an argument to a function that takes a member argument.  Some MDX implementations use some sort of iterator function that allows selecting the nth member of a set (e.g. Microsoft uses an "item()" function).  It sounds like there may not be something like that in IRIS MDX.

Certifications & Credly badges:
Lee has no Certifications & Credly badges yet.
Global Masters badges:
Lee has no Global Masters badges yet.
Lee has no followers yet.
Lee has not followed anybody yet.