When I run jobs, I can point to a file on Github or Azure Devops, and specify the branch I want the job to read from.
However when I create a DLT pipeline, I can point only to files on Databricks, and I cannot specify a branch. Pointing to a shared repo on a git provider would make sense... what is the recommended way to run DLT pipelines without this ability?
It's not supported yet. The solution would be to have a checkouts of repository on the specific branch and point DLT pipeline to it. This demo repository shows how to achieve this using separate checkouts for development, staging & production.