regression: default topmost parent selection broken on TEST

[Rdataflow]

Describe the bug
a regression occurs
on TEST by default no longer topmost parent is selected...

To Reproduce
Steps to reproduce the behavior:

1. Go to https://test.visualize.admin.ch/create/new?cube=https://environment.ld.admin.ch/foen/nfi/C-616/cube/1&dataSource=Int
2. See BS/BL selected as default region
   

Expected behavior

1. See in https://int.visualize.admin.ch/create/new?cube=https://environment.ld.admin.ch/foen/nfi/C-616/cube/1&dataSource=Int
2. default selected region is "Switzerland"

observed env: TEST

cc @bprusinowski

[bprusinowski]

Hi @Rdataflow, thanks for noticing this! 🙇‍♂️ It should be fixed by #1018. By the way, it seems that the cube mentioned in the issue is broken? Both on TEST and INT there is a no dimensions error.

I used NFI: Topics by ownership to test the behavior and the changes in #1018 make it correctly select Switzerland.

However, as it appears on INT right now, the Switzerland that is selected is a part of the hierarchy but does not appear in the dimension values, thus resulting in an empty selection (which can be fixed by manually selecting Switzerland from a filter dropdown).

Is this an intended behavior that there are values in hierarchy that are not used in the dimension values (I guess yes)? Is then the behavior visible for the cube above a bug? If so, should we also filter out the values from the hierarchy that are not in the dimension values so this scenario does not happen?

I think we should do it this way, but just wanted to make sure if I don't miss something important.