My scenario:
I have a published dataset with a bunch of tables and columns. Now two years after publishing it I want to change some names, but doing so breaks references in measures used outside the model.
So all lean reports connected to this model will be subject to errors in those cases where they:
use measures contained in the source dataset. 2. create own report measures referencing tables/columns in the source dataset.
Is there any (good) way to deal with this kind of scenario?
Putting some time and effort into naming is of course a given, but sometimes there are instances where name changes in sources cannot be predicted.
@Kristoffer.Dahlquist Assuming your Source is SQL, for these kind of situations a View is the best options, it acts as a layer between the tables and Power BI. Power BI will always have the same name, and you can manage the name changes in that view.
If you don’t have a database then use Dataflows/Datamart as a view
Thanks for your reply, however my scenario is more towards name changes in the dataset published in the service.
Hypothetical situtation: a measure resided in the model might be [Sum of sales] which sums the column Sales[Sales]. Some lean report then connects to the model and uses this measure in a visualization. But then for some reason I want to change the column name in the dataset to Sales[Current Sales].
After model refresh, the lean report using this measure will break as the column reference is no longer the same. Same would happen if the lean report creates its own measure referencing the old name Sales[Sales].
Keep the dataset interface the same, and adjust the dataset measure(s) accordingly. If you are using measures from the thin reports, I’m afraid you’re out-of-luck, and each thin report will need to be adjusted manually.
Greg
Just following up if the response above help you solve your inquiry?
We’ve noticed that no response was received from you on the post above. In case there won’t be any activity on it in the next few days, we’ll be tagging this post as Solved.
Due to inactivity, a response on this post has been tagged as “Solution”.
If you have a follow question or concern related to this topic, please remove the Solution tag first by clicking the three dots beside Reply and then untick the check box.
Thanks for your assistance. The issue is sadly non-solvable it sounds like.
This of course is a huge drawback when dealing with enterprise-wide datasets with a bunch of users creating their own reports from a dataset. It makes having nearly perfect naming conventions necessary from the start since you might not be able to edit the names later on without a lot of problems.