[Bug c++/110911] New: Confusing diagnostic on misplaced static_assert

[Bug c++/110911] New: Confusing diagnostic on misplaced static_assert

[pacoarjonilla at yahoo dot es]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110911

            Bug ID: 110911
           Summary: Confusing diagnostic on misplaced static_assert
           Product: gcc
           Version: og13 (devel/omp/gcc-13)
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pacoarjonilla at yahoo dot es
  Target Milestone: ---

This code (note static_assert instead of static_cast)


#include <complex>
void f(float m)
{
    std::complex<double> c = std::complex<double>(static_assert<double>(m),0);
}


generates the diagnostic

<source>: In function 'void f(float)':
<source>:4:50: error: expected primary-expression before '(' token
    4 |     std::complex<double> c =
std::complex<double>(static_assert<double>(m),0);
      |                                                  ^
<source>:4:51: error: expected primary-expression before 'static_assert'
    4 |     std::complex<double> c =
std::complex<double>(static_assert<double>(m),0);
      |                      


But the problem is not before the '(' token, nor before static_assert.
A suggestion such as "static_assert not valid here. Did you mean
'static_cast'?" would be nice.

[Bug c++/110911] Confusing diagnostic on misplaced static_assert

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110911

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2023-08-05
             Status|UNCONFIRMED                 |NEW

--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> ---
I think this would need a very specific custom diagnostic just for this case.
But I think I've made the same typo myself and I agree it would be nice, if
it's possible to do.

[Bug c++/110911] Confusing diagnostic on misplaced static_assert

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110911

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Although Dave Malcolm has recommended that we don't use "did you mean...?"
phrasing, but something like "static_assert not valid here, static_cast would
be valid".