topic Measures based on the same Dimension but different characteristics in App Development

Measures based on the same Dimension but different characteristics

hendriks_steffe — Fri, 28 Sep 2018 07:58:32 GMT

Hi Qlik Community!

I have a problem which I can't find the solution right now and I have twisted my head around this for some days.

The data I get are in the following pattern:

IITEM NO | CATALOGUE | NUMBER OF SALES

1000 A 1000

1000 B 2000

1000 C 3000

What I need to do is create a table with following function:

ITEM NO | NUMBER OF SALES CATALOGUE SELECTION 1 | NUMBER OF SALES CATALOGUE SELECTION 2

1000 1000 3000

LIST WITH CATALOGUE SELECTION 1:

A (selected in the above example)

LIST WITH CATALOGUE SELECTION 2:

C (select in the above example)

The obvious problem is I can't do this via 2 normal filters, because one selection cancels the other out.

I tried different approaches via set analysis but nothing works, because it goes against the functionality of filters.

Anyone got any ideas how to solve this without the use of extensions (we are not allowed to use them cause of maybe discontinuity in the future)?

My only other approach would be to create a new data model with some kind of matrix where every catalogue has every other catalogue corresponding to them in the data, but that would create a lot of data.

Best regards

Re: Measures based on the same Dimension but different characteristics

timpoismans — Fri, 28 Sep 2018 08:28:05 GMT

I understand why you can't use Extensions, you never know when they might stop working and if they will be updated or not.

But this is a typical example of alternate states which, as far as I know, can't be solved without an extension.

Re: Measures based on the same Dimension but different characteristics

hendriks_steffe — Fri, 28 Sep 2018 09:12:02 GMT

Thanks for the clarification, I feared it would be that way, so I have to solve this via SQL...what a mess.

Re: Measures based on the same Dimension but different characteristics

timpoismans — Fri, 28 Sep 2018 09:22:25 GMT

The only (quick and dirty) solution I can think of right now would be the following:

Test:

Load * Inline [

ITEM NO,CATALOGUE1,NUMBER OF SALES

1000,A,1000

1000,B,2000

1000,C,3000

];

Test2:

Load * Inline [

ITEM NO,CATALOGUE2,NUMBER OF SALES2

1000,A,1000

1000,B,2000

1000,C,3000

];

But this would effectively double your data.

It would allow you to have different selections in the "same" dimension.