[Bug c++/57170] Diagnostic for a negative case when switching over unsigned comes from -Wsign-conversion instead of -Wpedantic or -Wnarrowing

["egallager at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57170

Eric Gallager <egallager at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |accepts-invalid
             Status|RESOLVED                    |NEW
   Last reconfirmed|                            |2017-08-24
         Resolution|INVALID                     |---
            Summary|No diagnostic for a         |Diagnostic for a negative
                   |negative case when          |case when switching over
                   |switching over unsigned     |unsigned comes from
                   |                            |-Wsign-conversion instead
                   |                            |of -Wpedantic or
                   |                            |-Wnarrowing
     Ever confirmed|0                           |1

--- Comment #3 from Eric Gallager <egallager at gcc dot gnu.org> ---
(In reply to TC from comment #2)
> As comment #0 alluded to - perhaps a bit opaquely - this code is ill-formed
> in C++11 and later. [stmt.switch] requires the case expression to be "a
> converted constant expression ([expr.const]) of the promoted type of the
> switch condition", and a narrowing conversion cannot be used in a converted
> constant expression. The standard requires a diagnostic, which GCC fails to
> provide even with -pedantic.
> 
> (Meanwhile, using unsigned char or unsigned short there is actually
> well-formed if they get promoted to int as they usually do.)
> 
> So, unless you are saying that GCC's full conformance mode requires
> -Wsign-conversion, I don't see how this bug is invalid.

OK, reopening and retitling to make it clearer.