Re: Follow up on static bug (is it a bug?)

[Martin Sebor <sebor at roguewave dot com>]

David Durham wrote:
> 
> To repeat...
> 
> I believe this is a bug in gcc 3.0

No, it isn't. 9.2, p9:

    Except when used to form a pointer to member (5.3.1), when used
    in the body of a nonstatic member function of its class or of
    a class derived from its class (9.3.1), or when used in
    a mem-initializer for a constructor for its class or for a class
    derived from its class (12.6.2), a nonstatic data or function member
    of a class shall only be referred to with the class member access
    syntax (5.2.5).

Regards
Martin

> Forgive me if it's already known about...
> 
> The following code used to be valid ( <=2.95.3 ):
> 
> struct foo
> {
>         char field[25];
> 
>         static int getSize()
>         {
>                 return(sizeof(field));
                         ^^^^^^^^^^^^^

This will work since the argument of sizeof isn't evaluated:

                  return sizeof ((foo*)0)->field;


>         }
> };
> 
> 
> Here it reports that in foo::getSize, foo::field cannot be referenced
> since it is non-static and foo::getSize is static, however the used of
> sizeof/typeof/etc should make it valid.  Is there something about ANSI
> C99 and the fact that arrays can have non-constant sizes or something?
> 
> ...
> 
> Also, since then I've discovered that the following is also invalid
> 
> sizeof(foo::field)
> 
> ... or ...
> 
> (&foo::field)   <- which should give the offset of the field
> 
> This is invalid in any ole non-member function
> 
> Thanks,
>     -- Davy