Enable the Variable Library as a centralized location for storing all environment‑specific variables, allowing us to adjust them for promotion scenarios (e.g., from dev to prod) without relying on deployment pipelines.
It would be ideal if semantic model parameters or data sources could reference a Variable Library, similar to how deployment pipelines handle variables today.
When using CI/CD Option 1 — Git‑based deployments — it isn’t possible to hardcode environment‑specific values or data sources, since these differ across environments.
Example:
If I build a report in a dev workspace/branch that uses a dev data source, and then promote it to a prod workspace/branch via PR, I have no way to automatically switch the data source to the production version. I would need to submit an additional PR just to update the prod branch, resulting in the same report with different hardcoded sources.
2 Comments
- hedeencharles
Advocate I
This would be extremely helpful to have added! When doing git-based deployments, there is no easy way to update semantic model sources. This is only possible using Fabric Deployment Pipelines w/ deployment rules. git based deployments - https://learn.microsoft.com/en-us/fabric/cicd/manage-deployment#option-1---git--based-deployments - sjrrkb123
Helper III
ugh lame that this does not exist...would totally remove my need for fabric deployments from azure dev ops.
Recent ideas
Adjust column width in SQL Analytics Endpoint user interface
I can't see the column names and cell values in some of the columns in my tables in SQL Analytics Endpoint because the column width is too narrow. Also, the tooltip also doesn't show the entire...frithjof_v1 day agoCommunity ChampionCompleted4.2KViews75likes8Comments