[Proposal] Add back minimal HashEachMethods cop

[JacobEvelyn]

Is your feature request related to a problem? Please describe.

I recently noticed that the Performance/HashEachMethods cop has been removed, and reading that commit message and those of other commits it seems it comes down to the fact that it had a lot of false positives for negligible performance gains when trying to correct hash.each { |k, _| ... } to hash.each_key { |k| ... }.

I agree that that portion of the cop wasn't very useful. However, the cop also checked for another case: hash.keys.each { |k| ... } or hash.values.each { |v| ... } instead of the more performant hash.each_key { |k| ... } or hash.each_value { |k| ... }.

I miss that check in my code and would love to bring it back. Even though they're not quite equivalent—hash.keys.each {} will return an array of keys—at our organization we'd almost never use hash.keys.each in ways where we couldn't change it to each_key, and I think it would be fine to use an inline # rubocop:disable Performance/HashEachMethods in the rare exceptions.

Describe the solution you'd like

I'd love to bring back the Performance/HashEachMethods cop, but only the check for keys.each and values.each, and not the more complex code that inferred block argument usage and tended to produce false positives.

Describe alternatives you've considered

If it alleviates confusion we could make a new cop instead, but I actually think that would be less helpful since there are old websites/StackOverflow answers/etc. that refer to the old cop by name and describe it checking for .keys.each and .values.each.

Additional context

It seems like this check is still desired since it's still a part of the Rubocop ruby style guide: https://github.com/rubocop-hq/ruby-style-guide#hash-each

If I'm misunderstanding something about why it was removed, please let me know!

[JacobEvelyn]

@koic is there another way I should be advocating for this change?

[bbatsov]

With a pull request. :-) Generally I'm not opposed to what you're suggesting, although it's debatable if that'd be a performance cop or some style cop. Perhaps in the past we should have just disabled it by default.

[JacobEvelyn]

@bbatsov thanks. Would you recommend this being a performance or style cop?

[jemmaissroff]

I'm working on a PR here and will plan to do it in https://github.com/rubocop-hq/rubocop as a style cop. Would that make sense to you @JacobEvelyn and @bbatsov?

[jemmaissroff]

PR is up here rubocop/rubocop#7677! cc @JacobEvelyn @bbatsov. Full disclosure, this is my first open source PR so please advise me if I did anything incorrectly.

[tejasbubane]

I think this issue can be closed since a new cop has been added to rubocop.