Add the ability to disable specific upgrades

[vmarkovtsev]

E.g. our style guide forces printf-like syntax (we switch to C-like languages often and cut the corners) and we have to revert the corresponding upgrades as in src-d/ml#322 (review)

So would be nice to enable/disable features besides the master switch by Python version.

[asottile]

I'd suggest letting go of the past. printf-style formatting has been replaced (twice now! first with .format() and next with f-strings).

pyupgrade is an opinionated tool and will not be enabling opt-out of features except by language version.

[asottile]

If you're not convinced, I'd suggest reading through https://pyformat.info -- and perhaps this list of oddities

[vmarkovtsev]

@asottile No need to convince me since I mentioned the reason which has nothing to do with Python at all. Let me troll you some more: we also use double quotes hehe.

No worries. I will fork the project for our org and will maintain it under a different package name.
Thanks for pyupgrade btw! ❤️

[hugovk]

That was my PR, and the .format() change is the most common criticism in pyupgrade PRs to other projects too.

One is that .format() is slower than % formatting (eg. 367 nsec v. 243 nsec), even if that probably is premature optimisation. At least I find .format() clearer, and clearer code is usually better than ever so slightly faster code.

Another is that % formatting isn't deprecated so there's no point changing it, or that there's nothing wrong with % formatting so it's not worth the code churn.

Some just prefer the old style, especially for those switching between languages (like @vmarkovtsev).

@vmarkovtsev Please let me know when the fork is ready, it'd be useful for those cases when needing to revert and redo. Thanks!

[asottile]

Without conversion there's no direct conversion to f-strings which are faster.

If you're worried about performance differences between % formatting and .format() you're using the wrong language, python is slow.

[asottile]

you know, now that I see how simple the change would be to opt out of percent formatting, make a PR -- I'll accept it 😭

sorry about the runaround

[asottile]

This is available in 1.8.0.

[vmarkovtsev]

LOL

I knew your nerves would not bare this 😆 That's a pity - I have just passed through the CI checks and prepared the fork for pypi.

[nonchris]

Hi. It's been some time.
But I'd love to hop in and ask for the same disable feature for the "redundant open" correction.
Yeah, I know. It's redundant, we could let it go.
But as the second sentence of the zen of python says:

Explicit is better than implicit.
Just make it possible for us to state explicitly that this is just a read-operation. It makes the code better readable.

Thanks for considering my request!

[asottile]

no means no