Deployment Pipeline compare using Fabric REST API

[SanderTK]

Hi all,

I am working with significant number of workspaces that each operate using deployment pipelines. 

The goal is to build a notebook that runs every evening to check for changes in the pipelines between development and production to make sure that people don't forget to deploy neccesary changes to production (we have found that this is an element of human error at the moment).

So far I have build a python function that can detect if items are paired or not and thereby see if there are items in dev that have yet to be deployed to prod but what I am missing is a check to see if the paired items between the stages are actually the same or not. I don't need to know what the differences are between items (I feel like I would need git integration for that) but I was wondering if I could get the basic compare function from the UI from the API endpoint as well.

[v-karpurapud]

Hi @SanderTK 

Thank you for reaching out to the Microsoft Fabric community forum.

I would also like to thank you for your active participation  @lbendlin and  @deborshi_nag for sharing solutions within the community forum. 

I hope the information provided helps resolve your issue. If you have any further questions or need additional assistance, please feel free to contact us. We are always here to help.

Best regards,
Community Support Team.

[lbendlin]

 to make sure that people don't forget to deploy neccesary changes to production 

How do you define what a "necessary" change is? Shouldn't you let the developer make that decision?

[SanderTK]

What I mean with this is that fixes often get implemented in DEV without then being deployed to PROD. People simply forget to deploy their own fixes, features, etc.

I know it sounds a bit ridiculous but it is the current situation.

[deborshi_nag]

Hello @SanderTK 

There isn’t a public “Compare” API that gives you the same green or orange “identical” or “different” result as you see in the Deployment Pipelines screen. 

What you can do for now is use the Pipelines REST APIs to get a list of items in each stage and spot the differences yourself, based on things like IDs, names, paths, and when they were last deployed. If you want to dig deeper, especially with Reports and Dataflows, you can use Git-integrated PBIR for a better check on how similar they are.

GET /v1.0/myorg/pipelines/{pipelineId}/stages/{stageOrder}/artifacts will show you the supported items (datasets, reports, dataflows, dashboards, and so on) with details like:

• artifactId, artifactName (or artifactDisplayName)
• sourceArtifactId, targetArtifactId (if paired)
• lastDeploymentTime

For Power BI Reports, if you’re using Git integration with PBIR (Power BI Project), the report is defined in JSON. You can create a standard JSON hash, ignoring fields like timestamps that change every time.

For Dataflows, you can pull their JSON definitions using the Dataflow APIs and create a hash as well.