Git Function Branch Workflow

Question

I contribute to the github project and find myself in a bit of a pickle.

There is a new function branch in the original repo, and I will actively send download requests. I would prefer that you do not have to constantly send requests for traction, and then wait for the merge, and then remove the fork branch and create a new fork from the updated code base. My question is the best way to do this.

Example for clarification:
main = original repo
mine = my forked repo
main created a function branch.
I forked this branch of functions to mine.
I made changes to mine and sent a draft request.
the main combined my request for traction

My question is: should I remove the mine and redeploy the function branch to avoid duplicating the retrieval request of my previous tensile request or is there a better way to accomplish this.

UPDATE For a real live example, I'm working on a codeigniter system, and there is a big change in the authentication system. In the original repo, a new branch of signs was created for this. My concern is that there is still no problem, it is a very fluid / rapidly changing industry. Therefore, to expand my question, this is not a bunch of corrections, but rather a large number of corrections.

Answer 1

Good question! First, I would speak with (the project developer / person who will accept requests for traction) to find out what his preferences are. Believe me, you want to make life easier for the support person and make sure that you are doing the “Right Thing”. Creating a traction request workflow that is convenient for him will be long.

OK, that said the nature of the changes (and pull requests) that you make? Are they “named functions” or are they “small” bug fixes?

Also, is there a significant overlap in the lines you touch (so that merge conflicts between pull requests can occur) or are they mostly orthogonal?

If they are "named functions", with little overlap / merge conflicts, I would bind one new named branch for each function. If they overlap and / or are tiny bug fixes, I would probably go with “constant pull requests from the same forked branch” if it's OK with the maintainer.

You may know this, but I put it in posterity; git branches are cheap, cheap, cheap. If in doubt, create a new branch.