Hi everyone,
Since @idan accepted the oauthlib community migration, as a team, we should list what do we need to progress as a true community. I suggest to start with a small list, and please anyone, feel free to participate by adding any suggestions :-)
Define/Improve release process:
github's releases
, 2.0.5 in __init__.py
and 2.0.6 on pypi
...something wrong for sureTRAVIS_TAG
(see example at https://github.com/thomsonreuters/bottle-oauthlib/blob/master/setup.py instead of our current hard-coded value: https://github.com/oauthlib/oauthlib/blob/master/oauthlib/__init__.py ). Also, I've seen @ib-lundgren is the actual publisher of the pypi package, he looks inactive since years, but dunno if it's an issue. Future, roadmap:
Also, a quick round table could be great. I start the introductions, I'm currently working on a OAuth2.0 RequestValidator implementation with bottle, and I've never worked with OAuth1.0, nor Django nor Pyramid nor Flask. However, I'm trying to have a good knowledge around the RFC involved here (oauth2, introspect, revocation, jwt...). I have not started yet the OpenID integration, but it will come soon.
All great ideas @JonathanHuot!
And what do people think about:
@JonathanHuot I'm trying clear things a little bit here https://github.com/oauthlib/oauthlib/issues/512
Anything I can do just ask.
chiming in :) while working on a PR I noticed there is no coding style around which makes things sometimes hard to follow. I'd like to work on that if you guys are ok. Maybe starting with a issue containing a proposal and after that being accepted, starting to :nail_care: the codebase?
Hi @MattBlack85, it's a good idea, any work in this direction is welcomed ! :-)
_re: coding style_. I've found in projects I've worked on that by using autopep8 and yapf, I can basically let tooling clean up the coding style so I don't have to worry about it (except in cases where the cleaned up version is far less useful than having it un-cleaned-up, usually to do with line-lengths that would be clearer and only exceed the length boundary by a character or two). I use elpy-mode in Emacs to make that easy, but I suspect it could easily be done in command-line, and CI, as well.
Having an .editorconfig
in the root of the repo has also been useful.
Just thinking we probably want to have a run merging the currently open PRs before passing the code through a formatter. I'm all for pep8/flake8/yapf.
:+1:
👍
I'm a big fan of just keeping things PEP8.
Most helpful comment
Just thinking we probably want to have a run merging the currently open PRs before passing the code through a formatter. I'm all for pep8/flake8/yapf.