Any advantages/disadvantages to put this into the clang format run?

Do you mean combining the workflows?
That would be possible, but I don't think we could restrict this workflow to just shell files in that case (meaning that it will always run, something which we are trying to reduce...? :)).

Spinning up a runner also costs. Maybe even more?

Does anyone manually run the clang-format check? Would they be expecting it to start worrying about scripts?

Conversely, would anyone want to only run the script check without the full clang-format check?

Given how quick running "make clang_format" is, I'd expect spinning up the builder to be the most expensive part of the process in terms of elapse time, if nothing else. On the whole, I'm tending towards just bundling it all together. I'd keep separate "make" options, though.

Yeah. Probably putting it together is cleaner and sustainable?

@hoffie I think we should put it into the clang format file and run it only if there are changes to cpp, mm, sh, h, ... files.

@hoffie I think we should put it into the clang format file and run it only if there are changes to cpp, mm, sh, h, ... files.

Done. It's a bit unnatural to do this in a single job, but splitting into multiple jobs (even as part of the same workflow) would probably require fresh runners so it wouldn't save us anything.