Re: [boost] Re: Copy Constructible Concept

From: Douglas Gregor (gregod_at_[hidden])
Date: 2002-11-15 16:13:42

On Friday 15 November 2002 03:45 pm, Douglas Gregor wrote:
> On Friday 15 November 2002 03:36 pm, Gennaro Prota wrote:
> > of inventing a separate concept (say "Addressable") for them. The problem
> > I see in your resolution is that AFAIK addressof() is not guaranteed to
> > work by the standard. Yes, it has an extremely high probability to do the
> > right thing, but not a guarantee.
>
> You'll have to back that up with some standardese. AFAICT, 5.2.10/10 lets
> addressof() work:
>   "That is, a reference cast reinterpret_cast<T&>(x) has the same effect as
> the conversion *reinterpret_cast<T*>(&x) with the builtin & and *
> operators."
>
> (And that reinterpret casting T* -> U* -> T* preserves the original value).
>
> 	Doug

Just to spell it out in agonizing detail, so we can look for any holes in my 
logic:

  template <typename T> T* addressof(T& v)
  {
    return reinterpret_cast<T*>(
         &const_cast<char&>(reinterpret_cast<const volatile char &>(v)));
  }

By 5.2.10/10, "reinterpret_cast<const volatile char &>(v)" is equivalent to 
"*reinterpret_cast<const volatile char *>(&v)" with the built-in operators & 
and *.

Substitute this into the original expression:

  reinterpret_cast<T*>(
         &const_cast<char&>(&*reinterpret_cast<const volatile char *>(&v)));

The const/volatile qualifiers don't matter at all in reinterpret_cast or 
const_cast and the object is guaranteed to be the same object through the 
const_cast (5.2.10/7 and 5.2.11/3, respectively). Now we just have the case 
where we're reinterpret_casting T*->char*->T*, and that's covered by 
5.2.10/7.

        Doug