Fix #[derive] for empty structs with braces

[petrochenkov]

Fixes #29548

[rust-highfive]

r? @aturon

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

[alexcrichton]

Could this be extracted at the same time that ident is extracted above?

[alexcrichton]

Looks good to me, thanks @petrochenkov! Just one small nit but otherwise r=me

[petrochenkov]

Updated.

[alexcrichton]

@bors: r+ a3f3c4a

[bors]

⌛ Testing commit a3f3c4a with merge 3263316...

[bors]

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-debug-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-linux-cross-opt, auto-linux-musl-64-opt, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-mac-ios-opt, auto-win-gnu-32-nopt-t, auto-win-gnu-32-opt, auto-win-gnu-64-nopt-t, auto-win-gnu-64-opt, auto-win-msvc-32-opt, auto-win-msvc-64-opt

[durka]

Should we back out this complexity now that braced_empty_structs is stable?

[petrochenkov]

Yeah, I think derive can be cleaned up a bit now. Tuple structs/variants still need to generate Pat::TupleStruct/ExprCall, but everything else can generate braced forms.

[Manishearth]

Please [breaking-batch] future changes to the derive machinery.

[durka]

@Manishearth what kind of changes? Changes to the public API (?) of libsyntax_ext? Or any changes at all to the expansion of #[derive(...)]?

[Manishearth]

The API. We're trying to avoid libsyntax breakages that go on and break large swaths of the ecosystem. The derive machinery and the AST are both things that do this.