Multiple unused formatting argument notes should use multispan

[brson]

After #37613, the output for format! errors looks like:

error: multiple unused formatting arguments
  --> $DIR/format-foreign.rs:12:5
   |
12 |     println!("%.*3$s %s!/n", "Hello,", "World", 4);
   |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   |
note: argument never used
  --> $DIR/format-foreign.rs:12:30
   |
12 |     println!("%.*3$s %s!/n", "Hello,", "World", 4);
   |                              ^^^^^^^^
note: argument never used
  --> $DIR/format-foreign.rs:12:40
   |
12 |     println!("%.*3$s %s!/n", "Hello,", "World", 4);
   |                                        ^^^^^^^
note: argument never used
  --> $DIR/format-foreign.rs:12:49
   |
12 |     println!("%.*3$s %s!/n", "Hello,", "World", 4);
   |                                                 ^
   = help: `%.*3$s` should be written as `{:.2$}`
   = help: `%s` should be written as `{}`
   = note: printf formatting not supported; see the documentation for `std::fmt`
= note: this error originates in a macro outside of the current crate

that's a lot of lines dedicated to underlining macro acruments. They should all be output in one multispan.

cc @DanielKeep @estebank @jonathandturner @GuillaumeGomez

[GuillaumeGomez]

Hum indeed. If no one has taken it in 12 hours, I'll give it a look.

[DanielKeep]

@GuillaumeGomez I'm poking at it right now. I think I need to fix ExtCtxt's methods to accept <S: Into<MultiSpan>> instead of just Span, and then I can use MultiSpan directly.

After all, I created the problem, I should probably fix it. :P

[GuillaumeGomez]

Perfect. Good luck! :)

[cramertj]

I believe this should be closed.

[GuillaumeGomez]

Hold on, it seems the PR is still open.

[perryprog]

Seeing as there isn’t anyone working on this anymore, I’ll be happy to take over.

This is my first issue, (woot), so I’m not exactly sure where to start. I assume I should just fork the repo, change format.rs to use span_lable, and fix the tests?

Pinging @Nickforall and @DanielKeep in case you want to work on this still.

[Nickforall]

@perryprog I am not working on this anymore.

[Mark-Simulacrum]

@perryprog Those steps may be correct; @GuillaumeGomez or someone else may be able to verify for you. You can also look at #41256 and #37729 which appear to be previous PRs which attempted to do this. Let us know if you'd like any help!

[perryprog]

@Mark-Simulacrum Awesome! I'm super excited. Anyone specific I should r??

[estebank]

@perryprog bors will select someone automatically if you don't specify anyone, but I think any of me, @GuillaumeGomez, @Mark-Simulacrum, @jonathandturner or @nikomatsakis would be available for review and or guidance.

[perryprog]

For some reason I seem to be getting test errors unrelated to my changes. (ran ./x.py test --stage 1)

[Mark-Simulacrum]

Those tests shouldn't actually be running in --stage 1... cc @alexcrichton. For these changes, I would suggest ./x.py test --stage 1 src/test/{compile-fail,ui}, I think that should be enough.

[perryprog]

Gotcha, it’s possible I copy pasted from the wrong command (on mobile can’t check right now), but I’ll use that command for testing and see if it works.

I’m kinda busy for the rest of the week but I’ll keep working on it in my free time. I’ll comment here until I get a PR ready with any updates I have.