Make last structs indexes definitions use newtype_index macro

[spastorino]

This PR makes the last two index structs not using newtype_index macro to use it and also fixes this #45763 issue.

[rust-highfive]

r? @eddyb

(rust_highfive has picked a reviewer for you, use r? to override)

[spastorino]

/cc @nikomatsakis @Nashenas88

[spastorino]

r? @nikomatsakis

[durka]

How come this one isn't in the macro call?

[spastorino]

Because we need graph::DepNodeIndex::INVALID and the macro generates graph::INVALID

[durka]

As a side note, this could be using $pub:vis instead of $($pub:tt)*.

[spastorino]

I could add that as a separate commit.

[spastorino]

My attempt to make this change https://gist.github.com/spastorino/715079ea4839effe67fd9f0c3bee1c70

And I'm getting a lot of these errors ...

error: no rules expected the token `]`
  --> src/librustc/dep_graph/graph.rs:48:1
   |
48 | newtype_index!(DepNodeIndex);
   | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   |
   = note: this error originates in a macro outside of the current crate

Ran with trace macros ...

note: trace_macro
  --> src/librustc/dep_graph/graph.rs:48:1
   |
48 | newtype_index!(DepNodeIndex);
   | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   |
   = note: expanding `newtype_index! { DepNodeIndex }`
   = note: to `newtype_index ! (
           @ type [ DepNodeIndex ] @ max [ :: std :: u32 :: MAX ] @ debug_format [ "{}" ]
           ) ;`
   = note: expanding `newtype_index! { @ type [ DepNodeIndex ] @ max [ :: std :: u32 :: MAX ] @ debug_format [ "{}" ] }`
   = note: to `newtype_index ! (
           @ vis [  ] @ type [ DepNodeIndex ] @ max [ ::std::u32::MAX ] @ debug_format [
           "{}" ] ) ;`
   = note: expanding `newtype_index! { @ vis [  ] @ type [ DepNodeIndex ] @ max [ ::std::u32::MAX ] @ debug_format [
           "{}" ] }`

error: no rules expected the token `]`
  --> src/librustc/dep_graph/serialized.rs:17:1
   |
17 | newtype_index!(SerializedDepNodeIndex);
   | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   |
   = note: this error originates in a macro outside of the current crate

[durka]

This could be #41949. Probably not worth it to fix in this PR :)

[spastorino]

@nikomatsakis I'd merge this if you are ok and then this vis thing could be done in a separate PR.

[Nashenas88]

Nice work!

[nikomatsakis]

@bors r+

[bors]

📌 Commit d19faea has been approved by nikomatsakis

[bors]

⌛ Testing commit d19faea with merge 44990e5...

[bors]

☀️ Test successful - status-appveyor, status-travis
Approved by: nikomatsakis
Pushing 44990e5 to master...