Qlik Community

QlikView App Dev

Discussion Board for collaboration related to QlikView App Development.

Announcements
Now Live: Qlik Sense SaaS Simplified Authoring – Analytics Creation for Everyone: READ DETAILS
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Multi Use Dashboards

I have been playing around with QV for quite a while now and my company is now looking to make some dashboards live so i need to really think about the best practices for building the data model. One dashboard I am building has a couple of data sources from two different databases that are currently not connected which will mean the finished dashboard will likely have one set of tabs for one source and another set of tabs for the second source. So the question is, how do i script the data load such that when a front end designer adds, for example, a straight table and brings in dimensions from both schemas they don't generate a cartesian join and end up crashing their PC?

1 Reply
Not applicable
Author

My recommendation would be to provide your front-end designer a naming convention that denotes when to use which fields.  For example, you could have the fields DB1CustomerID, DB1Amount, DB2CustomerID, and DB2Amount.  Now your designer will know that putting DB1CustomerID with DB2Amount would not be valid.

What you will need to address (and maybe you are fine with this) is that your user will not be able to select 'Customer' in a DB1 tab and then have that same filter apply if they move to a DB2 tab, which could cause confusion among some users.  It would still be selected and visible in the Current Selections, but it wouldn't filter the data at all.  One way to get around this may be to set a Clear All action when the user switches tabs, so that it will reset the filters and force the user to re-select the Customer, this time in the correct DB field.  Does that make sense?