Repository Creation Process Should Take Email Addresses Into Account

Problem to solve

I have multiple emails associated with my account and want to create individual repos with not all the same email addresses for the initial commit, but have an option to choose which one to use if there are more than one.

Intended users

At least every person creating a repository inside Gitlab, who also has multiple email addresses associated with their account, will probably use this. This might most likely be Developers, Project Managers or Team Leads.

Further details

Currently, when I create a repository inside Gitlab, then I can not choose the email address for the initial commit. As a result, when I clone the repository and then commit further, there is one commit with the wrong email address. I know how to rectify that, but if I do, then I get merge conflicts later down the road due to "unrelated branches". This collides with the code review process - otherwise I could just push -f, but updating the email address for said commit results in all approvals to be anulled. Being able to choose an email address from the list of registered email addresses would solve this problem.

Please note that I am currently using a personal account on gitlab.com, but are registered as a developer for some projects of a paying customer (I develop for them). I also have an email from them, which I want to use on repos created for them.

Proposal

You would change the repository creation process to conditionally include a field for the initial email address if the user has registered more than one email address to their account. Then you would configure the newly created repo to use that email address for the first commit.

Later on you can get fancy and have some elaborate management of emails and repo patterns to make the process smoother for people with many emails or many repositories.

Permissions and Security

Nobody needs to have more permissions than they have already.

Documentation

Testing

What does success look like, and how can we measure that?

Links / references