Remove pointless calls to *iter() and iter_mut()

[Veedrac]

Pull request for #26188.

[rust-highfive]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @alexcrichton (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 The way Github handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed.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 the contribution instructions for more information.

[steveklabnik]

🎊

[steveklabnik]

This kind of thing is extra interesting, because this looks like a change in semantics, but since err_map is a reference, it's not.

[Veedrac]

IMHO, this isn't particularly worrisome because the default - move if it's a value or take references if it's a reference - is always either the right thing or a compiler error. YMMV.

[steveklabnik]

Yeah, I don't think it's a problem at all, just an interesting side-effect of looking at the diff.

[alexcrichton]

Thanks for the PR @Veedrac! I personally wouldn't so aggressively remove calls to iterators, as sometimes it's more clear (for example where the replacement uses &*). Could you revert the replacements to &* to keep using iterators?

In general I personally prefer for x in foo.iter() over for x in &foo as well, but it's definitely a matter of personal taste :). I don't think we're in a position to eliminate calls to iter(), though, so we don't necessarily need to aggressively try to prune them.

[Veedrac]

There are a few cases here

1. .zip(pats.iter()) → .zip(&*pats)
2. .zip(pats.iter()) → .zip(&pats)
3. .zip(pats[a..b].iter()) → .zip(&pats[a..b]) (special-case of the prior)
4. .zip(pats.iter_mut()) → .zip(&mut pats)
5. for pat in pats.iter() → for pat in &*pats
6. for pat in pats.iter() → for pat in &pats
7. for pat in pats[a..b].iter() → for pat in &pats[a..b] (special-case of the prior)
8. for pat in pats.iter_mut() → for pat in &mut pats

Can I get a definitive statement of which of these should be removed? It's not much hassle, but I'd nevertheless like to do it only once. :)

[alexcrichton]

I would go for .iter() over &*, but if there's only one & then it's fine to leave as-is.

[steveklabnik]

Same.

[Veedrac]

Reborrows actually seemed to be more common than calling .iter(), so this new commit changes more code than before.

FWIW, I've included reborrows of the form &**foo, which this changes to foo.iter().

[Veedrac]

I just stumbled on #21830, which changed several for x in y.iter() to for x in &*y. Not sure what the best course of action is.

[alexcrichton]

This all looks great to me, thanks @Veedrac! It's sometimes tough to give a close review to all changed code in large patches like #21830, so I think these cases just fell through the cracks.

@bors: r+ d7f5fa4

[bors]

⌛ Testing commit d7f5fa4 with merge b5b3a99...

[alexcrichton]

This passed all builds except one mac build which failed b/c the bot disappeared, so merging manually.