Update codebase to Python >= 3.6 #2612
Conversation
|
Note that in #2301 I already went over all the occurrences of |
There was a problem hiding this comment.
To give a clearer comment...
Now I will point out the particular places where subjectively I think the code "after" is harder to understand than the code "before".
Sorry that I'm not giving the reasoning.
For an easy way forward, maybe you could just undo those parts. The rest is all fine.
There was a problem hiding this comment.
Further discussion is also possible.
Certainly there's a lot of merit to setting a clear guideline once and for all.
Some examples of guidelines:
- Always use f-strings, not
.format
(what seems like you're trying). - Use
.formatif one of the items isn't a plain variable or member access
(seems like the current situation, I don't mind it). - If one of the items isn't a plain variable or member access, first assign it to a variable and then always use f-strings anyway
(I could be open to this, but I think it can also be messy).
|
The last commit - I don't think it's good to clump it together with this pull request. Please drop it. Also not sure about "Rewrite Nav config type validation error message" - ideally would be separate |
|
Thanks a lot! |
|
I'm stupid and merged in the wrong order @mondeja , I apologize for the conflict here... |
|
No problem, it's fixed now 👍🏼 |
.formatcalls with f-strings.'r'mode argument foropenfunction as it is uneeded, but maybe maintainers want to keep it explicit?__future__import andcoding: utf-8comment from a file.