On Wed, Nov 20, 2024 at 11:08 PM mauro russo <ing.russomauro@gmail.com> wrote:

>> I see this lack in both the text of draft n4849 and the current draft text.
>> e.g.
>> auto (*p)() -> auto; // ill-formed (as all major compilers act)

> For this example we have to check whether both `auto`s are legal.

> The first one is legal under p3. The second one is not legal: it appears in the type-specifier-seq of the trailing return type, but the declarator in this case doesn't declare a function. So, the entire declaration is ill-formed.

Dear Brian, thanks for the answer.

Please, note that my example "auto (*p)() -> auto;" was exactly discussed as the illegal case due to the second 'auto'.
My discussion was exactly focused on the fact that the overall text of [decl.spec.auto.general] seems not excluding this case.

I'm not seeing what the problem is. The existing text disallows the second `auto`.

[dcl.spec.auto.general]/4 contains the "declares a function" restriction. Consequently there is no case in [dcl.spec.auto.general] that allows the second `auto` and it is therefore ill-formed by [dcl.spec.auto.general]/8.

It seems you haven't really read my 3 points, just focused on the example.

I think if this example is not meant to actually illustrate the defect you perceive in the wording then you should provide another example that does.

Please, be patient and check again details of my discussion.

That detail is not even present in [dcl.fct]-p(1.3) that just forces to have return type auto in decl-specifiers in case there is trailing-return-type,
and I really believe that detail shuld be precise in [decl.spec.auto.general].

Thanks,
Mauro.

Brian Bi