Skip to main content
Microsoft Idea

Power BI

Needs Votes

6 versions of "Report1" with the same name in the same workspace - Duplicated names for Reports and Dataset

Vote (2) Share
Pär Adeen's profile image

Pär Adeen on 11 Mar 2021 14:03:06

There's a hard to understand logic for when you will have duplicated names of reports and Datasets in the Power BI Service, but I will listed what I have discovered

1) Uploading a report with a specific name (e.g. Report1 ) doing live connections to two different datasets will generate two reports with the same name but with Different GUIs.
E.G. Report1 and Report1
Report1 connects to Dataset1 and Report1 connects to Dataset2, but both have the same name.

2) And now it gets interesting. Using the Onedrive sync feature (https://docs.microsoft.com/en-us/power-bi/collaborate-share/service-connect-to-files-in-app-workspace-onedrive-for-business), you also get duplicated reports. So lets say you already have Report1 that does live connection to Dataset1 manualy publised to the workspace and then sync the same report (Report1 that does live connection to Dataset1) from onedrive (or the other way around), then you will have two different Report1. Both will do live connections to Dataset1

And remember... We already had manage to create two Report1 in step (1), and therefor we now have tree versions of Report1
Report1 (manaully published) --> Dataset1
Report1 (manaully published) --> Dataset2
Report1 (published using the onedrive sync) --> Dataset1
So now we have three different Report1.

And yes, we can make this even worse. We can of cause also upload a new Dataset1 using the onedrive sync, and this way we now have two different versions of Dataset1, but with the same name.
And lets say we also manually publish a report that does live connections towards that new Dataset1, then we have four version of Report1.

And then don't forget the posibility to use the ondrive sync to upload the same report, then we have 5 versions of Report1.

And why not also imaging that we have a Report1 with a built in dataset, then we get 6 versions of Report1.

And why is this important...

Besides it's very confusing to have 2 (or even 6 or more) reports with the same name, it's sometimes important to have and keep the same GUID for the report and sometimes also a dataset.

Imaging we would like to set up the ondrive sync for various reasons. The onedrive sync goes quite fast, but you don't get any indications on when the file also has been published to the workspace. You could either just wait untill the file has been published or you might think you could publish manually. If you publish manually, users will not get you changes, since they are looking at a different report.

You can also imaging we would like have different versions of datasets active at the same time. Lets say central BI department provides
Dataset1_Generation1 and Dataset1_Generation2.
Or they Provides different release-rings of the datasets:
Dataset1_Latest and Dataset1_Stable.

When users of the datasets changes (in a controlled way) what datasets their reports does a live connection to, they also changes the GUID of their report. The customers for the report might have choosen to embedd the report (and this operation uses the GUID), of they have just saved a bookmark to the report.
We also have the sam problem when we drill through to other reports. Etc, etc...

I would say that the easiest way to solve this is to always re-use the same GUID, no matther how the report is published, and no matter towards what dataset the reports does live conections (of if the report embeds the dataset).