There are dozens of your type out there that I don’t understand. The “I don’t want no commit hooks running on my machine. I and only I control what happens on my local machine” purists. There’s at least one in every job I work.
If the thing is going to block in ci anyway, why are you opting for a push and pray approach? Why arbitrarily increase your feedback loop time and add waiting time for each loop in ci? Chances are the time to get feedback is at least a couple orders of magnitude faster locally, you’re paying not only the startup time to register the runner and bring it online, get dependencies installed etc but also the manual time to context switch to the CI window etc. Just do all your linting and auto formatting and whatever on commit. It’s all work you’re going to have to do anyway, why introduce some extra less efficient step on yourself to slow yourself down?
If the thing is going to block in ci anyway, why are you opting for a push and pray approach? Why arbitrarily increase your feedback loop time and add waiting time for each loop in ci? Chances are the time to get feedback is at least a couple orders of magnitude faster locally, you’re paying not only the startup time to register the runner and bring it online, get dependencies installed etc but also the manual time to context switch to the CI window etc. Just do all your linting and auto formatting and whatever on commit. It’s all work you’re going to have to do anyway, why introduce some extra less efficient step on yourself to slow yourself down?