implement fmt traits for collections

[skullzzz]

reopen #20932
closes #19670

Still need to fix up tests and move them to libcollectionstest, but figured I should get comments on this implementation.

[rust-highfive]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @huonw (or someone else) soon.

If any changes to this PR are deemed necessary, please add them as extra commits. This ensures that the reviewer can see what has changed since they last reviewed the code. The way Github handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed. Large or tricky changes may require several passes of review and changes.

Please see CONTRIBUTING.md for more information.

[Gankra]

r? @gankro

[Gankra]

Missing a space after colon

[Gankra]

(this is pretty much everywhere in this PR)

[Gankra]

This is a pretty nice approach to the problem 👍

[ftxqxd]

I feel like this would require an RFC—there are alternative approaches to this (e.g., adding formatting syntax for explicitly applying a formatter to the items in an iterable). Also, the precise details aren’t easy to settle on; e.g., should [1, 2, 3] format with Display as 1, 2, 3 or 1, 2, and 3? Or should sequences/maps not have Display formatters at all?

[Gankra]

@P1start There have been two RFCs, both of which provide guidance on collection formatting:

https://github.com/rust-lang/rfcs/blob/master/text/0565-show-string-guidelines.md
https://github.com/rust-lang/rfcs/blob/master/text/0640-debug-improvements.md

[Gankra]

Actually the first RFC explicitly states that Display should:

Not be implemented for generic containers like Vec or even Result<T, E>, where there is no useful, general way to tailor the output for user consumption.

[Gankra]

CC @aturon

[Gankra]

Oh! Just saw you added commits (Github doesn't notify anyone when that happens). Are you still working on stuff or are you ready for another pass?

[skullzzz]

Oops, I assumed new commits would send a notification.
I think I've covered all previously pointed out issues, so yeah, ready for another pass.

[Gankra]

r? @alexcrichton

It has come to my attention that current standards are ambiguous: Is Octal et al. Displayish, or Debugish? Needs decision.

[bors]

☔ The latest upstream changes (presumably #25989) made this pull request unmergeable. Please resolve the merge conflicts.

[alexcrichton]

Thanks again for the PR @skullzzz! For now, though, there is enough consensus to merge this. There are remaining concerns about how ambitious collections should be in their formatting (especially when it comes to the various flags formatters have). It's also possible to add these formattings as adaptors if necessary, or just writing a helper method.

If you're interested in pursuing this, though, I'd recommend one of two courses of action:

• I'd primarily recommend opening up a discuss post to gauge the interest in a feature like this along with how ambitious it should be. For example should this attempt to provide anything beyond the bare-bones formatting? If so, how would that be done?
• You can also open up an RFC to get broader discussion about this.

One of the unfortunate parts about stability today is that any trait implementation is insta-stable, so there's a bit higher of a bar for entry for trait impls than new structures (as we can't introduce this as stable functionality).