In this post I want to share my thoughts about two ways to check that Power BI reports are using the right template with Azure DevOps. Like in the below diagram.
I want to share my thoughts about this for various reasons. Including the fact that in my last two posts I covered both of these methods. As listed below.
- Automatically check Power BI reports are using the right template with Azure DevOps.
- Another way to check Power BI reports are using the right template with Azure DevOps.
In the following sections I highlight my thoughts. So that you can decide which you prefer. Along the way I highlight some important points.
Check Power BI reports with PBI Inspector
I covered how to do the checks automatically with PBI Inspector in a previous post. Which is already a popular choice for some well-known Power BI Professionals and they are helping to promoting it.
One key point I must highlight is that there is a PBI Inspector V2 available. Which supports the enhanced metadata file format (PBIR).
Plus, from version 2.3 it now supports the testing of any Fabric items’ JSON metadata, not just Power BI reports. In fact, just before this post was published Nat Van Gulck posted on LinkedIn about this update and introduced Ric, the Fabric Inspector.
This latest update might make PBI/Fabric Inspector a more appealing option to those with a non-Power BI background. Especially those who see themselves working more with Power BI and semantic models in the future.
Check Power BI reports with pytest
In my last post I covered how to do the checks automatically with pytest. Which is a popular Python-based testing framework amongst Data Engineers. I want to highlight a few points about this option below.
First point is that I chose pytest because I had worked with it when I was working with CI/CD in Azure Databricks. Which highlights the fact that this option might be popular for those with a background with other offerings.
Second point is related to the first point. Since pytest is a popular framework, it will be easier to work with Copilot to get answers to various pytest scenarios. Plus, more complex Python scenarios like conditions.
I showed an example of how to create tests with GitHub Copilot in Visual Studio Code in a previous post.
Third point I want to highlight, is that you can replicate the logic and perform similar tests with PowerShell.
However, bear in mind that the majority of the people working in Microsoft Fabric focus on either Power BI or Data Engineering. So, if you want them to create tests it is probably better to provide a solution that is connected to what they are working with.
Final words
I hope that by sharing my thoughts about two ways to check that Power BI reports are using the right template with Azure DevOps I provide some food for thought.
In reality, there are multiple way to test various Microsoft Fabric items at different stages of your CI/CD story. I shared these two since PBI Inspector is good for those with Power BI knowledge and pytest is familiar with a lot of Data Engineers who work with Spark compute.
Of course, if you have any comments or queries about this post feel free to reach out to me
Be First to Comment