Specify scope in out_of_scope_macro_calls lint

[estebank]

warning: cannot find macro `in_root` in the crate root
  --> $DIR/key-value-expansion-scope.rs:1:10
   |
LL | #![doc = in_root!()]
   |          ^^^^^^^ not found in the crate root
   |
   = warning: this was previously accepted by the compiler but is being phased out; it will become a hard error in a future release!
   = note: for more information, see issue #124535 <https://github.com/rust-lang/rust/issues/124535>
   = help: import `macro_rules` with `use` to make it callable above its definition
   = note: `#[warn(out_of_scope_macro_calls)]` on by default

r? @petrochenkov

[estebank]

Split off from #126810

[petrochenkov]

This PR (and probably some other diagnostics in #126810) mix up two different things

• the set of places in which the name is looked up - "the current scope" (ident), or "module foo" (foo::ident), or "the crate root" (crate::ident) or "extern prelude" (::ident)
• the location from which the name is resolved - "enum E", or "trait Bar", or "the crate root" as well

In the example above

warning: cannot find macro `in_root` in the crate root

in_root is not looked up in the crate root, like it may seem from the message.
It is looked up in the current scope from the crate root module.

warning: cannot find macro `in_root` in the current scope from the crate root module

That's how I'd word it if I extended the message.

I'm not sure what exactly should go after "from" for the clarification to be useful.
The ModuleKind-based location in this PR will only report modules, enums and traits (or nothing in blocks).
Seems arbitrary, but also ok as an approximation.
Or maybe just not do this and keep things as is because the clarification is not useful enough, not sure.

---

Note: renaming "this scope" to "the current scope" would look like a slight improvement to me, but it's a massive mechanical change, probably worth doing last after more meaningful changes settle down.

[JohnCSimon]

@estebank
ping from triage - can you post your status on this PR? This PR has not received an update in a few months.

[petrochenkov]

@estebank Do you plan to return to this?

[estebank]

Rebased and reworded the message to be inline with your comment, @petrochenkov.

---

I'm wondering what we should do for the case that would look like

warning: cannot find macro `in_root` in the current scope from the the current scope

I agree that we should differentiate the two concepts, I'm just also unsure on how to that in a way that can be conveyed to our users in a reasonable way.

Note: renaming "this scope" to "the current scope" would look like a slight improvement to me, but it's a massive mechanical change, probably worth doing last after more meaningful changes settle down.

👍

[petrochenkov]

I'm wondering what we should do for the case that would look like

warning: cannot find macro `in_root` in the current scope from the the current scope

Probably skip the "from" part if the $location is "this scope" (== make the location optional).

[estebank]

Re: the redundant wording/making the location optional, we have no test in the codebase that shows that case, so I'd like to leave the optionality of that output for a separate PR.

[petrochenkov]

@bors r+

[bors]

📌 Commit 835d434 has been approved by petrochenkov

It is now in the queue for this repository.