Introduce black linting #6586
Introduce black linting #6586
Conversation
webbyfox
commented
Apr 13, 2019
•
webbyfox
commented
- Added black configuration -> f2eacd3
- Run black to fix formatting -> f418b4e
|
@tomchristie, Seems like they always released alpha/beta versions till the date as they admit it not stable yet but the number of projects using it. |
|
So, I really like this, but I am finding it hard to correctly review without having the "introduce to the linting" split out from the "apply black". It might be a good idea for us to start with a failing test case that only adds black into the linting, but doesn't actually apply it. That way I can review it more thoroughly. |
|
Just throwing in my 2 cents, but I am -1 on black. In general, I'm really biased against black, but I also think it would be somewhat disruptive to introduce it this late into the project's history. Thoughts:
If we do want to go with black though, per the third point, we'd probably want someone from within the org (e.g., @tomchristie) to apply black so we can trust that there aren't any erroneous/additional changes, unintentional or otherwise (e.g., bugs in black). We may also want to consider using |
|
I think the points re. git history, and impact on large mature codebases are valid. |
|
OK, @tomchristie, I will create separate PR for applying black configuration soon. |
|
|
||
| ret = renderer.render(self.data, accepted_media_type, context) | ||
| if isinstance(ret, six.text_type): | ||
| assert charset, ( | ||
| 'renderer returned unicode, and did not specify ' | ||
| 'a charset value.' | ||
| "renderer returned unicode, and did not specify " "a charset value." |
There was a problem hiding this comment.
These two strings can be one now. This is a known upstream bug psf/black#26.
| self.assertEqual( | ||
| 1, | ||
| len(s.errors["address"]), | ||
| "Unexpected number of validation errors: " "{0}".format(s.errors), |
|
Closing per #6586 (comment) and the creation of #6630. |
