Re-deployment of Data factory resource makes publish process messy

Question

I have added role assignments with my data factory template and therefore I sometimes need to re-deploy data factory resource.

Repo configuration(1) includes property called lastCommitId that is not part of my template.

LastCommitId is updated by data factory itself and I do not use that property on my template. As soon as I re-deploy my template lastCommitId property is removed:

then running ADF instance thinks that it must re-create "all" objects when publishing from the collaboration branch:

I tried well-known workarounds in my template:

• lastCommitId: json('null')
• lastCommitId: null

but both workarounds end up deleting the lastCommitId property instead of ignoring the value change.

This makes publish process really messy with larger projects after Data factory resource re-deployment.

Is there any good workarounds for this?

I can think of at least two workarounds:

• Capture lastCommitId before deployment and then set it back
• or capture lastCommitId before deployment and pass it as a template parameter

Any better workarounds without adding complexity to the template/pipeline?

EDIT:
As a workaround, I read the value of lastCommitId in pre-commit scripts and inject the value as template parameter value into the deployment script. That is working for me but I would like not to add extra steps to maintain the idempotency of the deployment templates.

(1) https://learn.microsoft.com/en-us/azure/templates/microsoft.datafactory/factories?tabs=bicep#factoryrepoconfiguration