Do not input private or sensitive data. View Qlik Privacy & Cookie Policy.
Skip to main content

Announcements
Posting on Qlik Community is temporarily paused for some users. Thank you for patience and we'll update you soon!
cancel
Showing results for 
Search instead for 
Did you mean: 
sirpod90
Contributor III
Contributor III

Qlik (Sense) is not using dimensions for calculation

Hello everybody,

I am running at the moment in really weird problem and can not come up with any conclusions.

My Data looks like that:

Container-NameAction-DateStep-Name
Container1201.01.2016StepName-1
Container1202.01.2016StepName-2
Container1203.01.2016StepName-3
Container1204.01.2016StepName-1-V
Container4301.01.2016StepName-A
Container4301.01.2016StepName-B

I want to have table with the max step name the container past in time

So my result should look like that:

Container-NameAction-DateStep-Name
Container1204.01.2016StepName-1-V
Container4301.01.2016StepName-B


I set up a table with Container-Name as dimension and AGGR(max(Action-Date), Container-Name) as second dimension. If I now add the fact MaxString(Step-Name) the result looks for max string in all containers and not with the maximum date.


So my result looks like that:

Container-NameAction-DateStep-Name
Container1204.01.2016StepName-3
Container4301.01.2016StepName-B


It seams like the fact column would ignore the two dimensions before hand?

Why?? Do I have a wrong understanding of dimensions and facts?

When I tried to solve the problem with set analysis and only the container as dimension:

maxstring({$<Action-Date= {'&(=max(Action-Date))'}>}StepName)

My result where right but just for one Container, so Qlik took the max-Date from all dates, ignoring the dimension again.

Container-NameStep-Name
Container12
StepName-1-V
Container43-

I also tried to combine the set analysis with AGGR but the result is still wrong.

Is there something I really miss or does Qlik have a problem?

I am thank full for any help.

Regards!

10 Replies
sirpod90
Contributor III
Contributor III
Author

It just means, that the ID is devided by 10 to the power 10

ID / 10000000000

So that the first parameter (Date/Time) weight more than the ID.