QlikSense Distinct Filter

[#1]

New guy here....

I'm sure this is simple and I'm searching the forums, but I think I'm probably searching on the wrong terms or trying to tackle this problem the wrong way.

I have a simple table that I want to filter out duplicates in certain dimensions but not others (in the same way I can filter out null values in some dimensions but leave them others).  So in my table below, I want to filter out the duplicate values in the WRKORDNBR dimension.  Ultimately, I want to create a measure that will total the AMTBILLED dimension where WRKORDNBR is unique.  So on the example below, I only care about the first row where WRKORDNBR was 962217 and would like to hide the second row (with the later SUMMIT DATE).

Thanks!

[sinanozdemir]

Hi Drew,

If you look at the dimensions such as SUMMIT DATE, there are different values for the entry 962217 so it is not a real duplication and in order to remove the repitition, just don't bring the SUMMIT DATE dimension. Again, it is not a duplication, there are unique records in SUMMIT DATE dimension.

Hope this helps

[#3]

Sinan - I realize that the rows are not duplications and if I removed the SUMMIT DATE dimension my problem would be solved.  Unfortunately I need to show the SUMMIT DATE, but only care about the first SUMMIT DATE for a WRKORDNBR.

[jonathandienst]

Then I suggest that you remove the SUMMIT DATE dimension and add an expression such as Min(SUMMIT DATE) to display the first summit date.

[#5]

I get an "Invalid dimension" when I try to do this.

[jonathandienst]

I think you are trying to enter the expression where a dimension is expected. I don't have a Sense model handy to check the entry name, but it will be something like Expressions or Measures

[#7]

This is where I'm adding it.  I've added other expressions that are working fine.

[sinanozdemir]

Hi Drew,

You need to use  "Measure" option, not "Dimension". Please see the below screenshot:

When you hit "Add column", choose "Measure" and insert your Min([SUMMIT DATE]) expression:

If you choose "Dimension", it will throw an error of "Invalid dimension":

And finally, your final table will look like the below:

Let me know if this is what you are looking for.

Thanks