[Bug middle-end/62029] Requesting new warning: missing const-qualifier

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62029

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |diagnostic
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2017-07-22
                 CC|                            |msebor at gcc dot gnu.org
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=80711
     Ever confirmed|0                           |1

--- Comment #1 from Martin Sebor <msebor at gcc dot gnu.org> ---
Confirmed.  This enhancement would be especially useful in C++.  See also
pr80711 for a related  request.  It would make sense to consider both of these
together.

In C, one caveat/question specific to this request (pointers) is what to do
about C functions like strchr:

  void f (char *p)
  {
    char *q = strchr (p, '\0');
    *q = 'x';
  }

or strtol:

  void g (char *p)
  {
    char *q;
    strtol (p, &q, 0);
    *q = '\0';
  }

To avoid false positives on this code the warning would have to handle these
cases specially.  That's not difficult to do for functions like strchr that GCC
has intrinsic knowledge of but not for others (like strtol).