scarlehoff / pyHepGrid

Tool for distributed computing management geared towards HEP applications.
GNU General Public License v3.0
6 stars 4 forks source link

WIP: Auto black #71

Closed marianheil closed 4 years ago

scarlehoff commented 4 years ago

Personally I'd say we want people to run black on their own computers. If someone wants, for some reason (clarity or whatever), to have a line with 101 characters I think they should be able to have it.

marianheil commented 4 years ago

black gives the option to turn the parsing off with # fmt: off. This MR just simplifies committing by taking the formatting part away, in particular since with #69 we enforces pep8. However I don't think this is a must have. Whatever you prefer :wink:

scarlehoff commented 4 years ago

Well, you guys write to this repo way more than me these days, so if it makes your life easier...

The only permission the token needs is workflow, right? How do you plan to trigger the flake8 workflow?

Alternatively, wouldn't it make sense to run the workflow such that you do

Otherwise you will be creating many pushes that basically change the code a bit where there are potential bugs.

marianheil commented 4 years ago

I haven't used github actions before. Let me play around with them on my own mirror so you don't have to do anything for now. Whenever I have a setup that works well, I will come back.

For now we can just keep the current flake8 setup and see how annoying it is.

scarlehoff commented 4 years ago

If you want I can give it a go, I've been moving things from travis to github workflow in the last few months.

marianheil commented 4 years ago

As you want, if you have time and feel like doing it, go ahead. As said its not a must have, just a neat little side project.

scarlehoff commented 4 years ago

just a neat little side project.

I agree, I like it more from the point of view of fiddling with workflows than the actual feature :__ I'll reopen then.

scarlehoff commented 4 years ago

Ok, no, I just changed opinion. Having an autolinter going through all the code everytime will potentially create many "fake changes" which makes reviewing PRs much more difficult, which in turns will make reviews worse.