Calculation Group Format String Expression: Local Model & remote PowerBI Dataset, hidden measures

[alexrobe]

Calculation groups (CG) in a remotely connected Dataset and local CGs seem to work just fine - despite some limitations listed here.

One downside though, that is really preventing so much functionality of this otherwise great combining feature:
There seems currently to be the lack of the ability to combine 2 datasets with CG Format String Expressions:

The Combined Dataset (local Model + remote DirectQuery PowerBI-Dataset) works beautifully.

As soon as i create a CG with a local Format String Expression with Tabular Editor, I cannot even save the model back.
It looks like, TE/TOM are creating hidden measures (starting with "_") again - would be no big deal but it also creates hidden measures for the remote Dataset in the local model again. This already has of course the hidden measures because of its own CG with Format String Expressions.

Trying to save the Model gives of course the error message "Measure "_xxxx" is already existing, try to rename..."

• Anyone knows how to overcome this behaviour, that these hidden measures are not created for the remotely connected datset too?
• Do we have to wait for MS to address this or can this be overcome with a Tabular Editor tweak?

Thanks a lot,

Alex

[Anonymous]

@alexrobe 
I doubt this limitation can be lifted, it is not something can be fix like bug.

Best 

Paul

[Anonymous · Accepted solution]

@alexrobe 
I guess you used this method to turn dataset to direct query, scrolldown and there is a limitation for remote source.

Using DirectQuery for datasets and Azure Analysis Services (preview) - Power BI | Microsoft Docs

Best Regards
Paul Zheng _ Community Support Team

[alexrobe]

Thank you very much for your reply! 
Do you know, if you colleagues in the PowerBI dev team will allow in the future, to have format strings also allowed in combined models.
btw: Using 2 calculation groups (1 in remote model via DQ and 1 additional in local model) works otherwise perfectly fine - other than your linked information might suggest 🙂

[alexrobe]

Thank you very much, Paul!
Do you have any information, if this restriction will be lifted anytime soon?

🙏