[Bug fortran/107266] Reject kind=4 characters for BIND(C) – it invalid and generates wrong code

["sgk at troutmask dot apl.washington.edu" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107266

--- Comment #6 from Steve Kargl <sgk at troutmask dot apl.washington.edu> ---
On Fri, Oct 14, 2022 at 09:36:59PM +0000, burnus at gcc dot gnu.org wrote:
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107266
> 
> --- Comment #5 from Tobias Burnus <burnus at gcc dot gnu.org> ---
> I do note that we already have bind(C) + kind=4 examples:
> https://gcc.gnu.org/git/?p=gcc.git;a=blob;f=gcc/testsuite/gfortran.dg/PR100906.f90
>   integer, parameter :: c_ucs4_char = 4
> and more.

F2023 does not have an interoperable type of c_ucs4_char.
F2023 has SELECTED_CHAR_KIND("ISO_10646"), which can be
used as in the example code (see 22-007r1.pdf, 16.9.180):

14 SUBROUTINE create_date_string(string)
15    INTRINSIC date_and_time,selected_char_kind
16    INTEGER,PARAMETER :: ucs4 = selected_char_kind("ISO_10646")
17    CHARACTER(1,UCS4),PARAMETER :: nen=CHAR(INT(Z'5e74'),UCS4), & !year
18      gatsu=CHAR(INT(Z'6708'),UCS4), & !month
19      nichi=CHAR(INT(Z'65e5'),UCS4) !day
20    CHARACTER(len= *, kind= ucs4) string
21    INTEGER values(8)
22    CALL date_and_time(values=values)
23    WRITE(string,1) values(1),nen,values(2),gatsu,values(3),nichi
24    1 FORMAT(I0,A,I0,A,I0,A)
25 END SUBROUTINE


Neither UCS4, ISO_10646, ISO 10646, nor 10646 appear in 
F2023 Section 18.

On J3 Fortran github, there are no proposals/requests for
interoperability with ISO 10646.

> => Keep permitting it but fix it

I'll update my patch to use "gfc_notify_std (GFC_STD_GNU, ...)"
as this is clearly an extension to the Fortran standard.