David Abrahams <dave@boost-consulting.com> wrote in message news:<uvfskfk54.fsf@boost-consulting.com>...
> stlsoft@hotmail.com (Matthew Wilson) writes:
>
> > I've argued in the past to Walter that the expression x == y should be rewritten by the compiler as
> >
> >   (x !== null && y !== null)
> >     ? x.eq(y)
> >     : false
>
> So null != null?

He, he.

I've never tried it. Trying now ...


... Bizarre.

The following code

  class X
  {};

  int main(char[][] args)
  {
    if(null != null)
    {
      printf("null Different\n");
    }
    else
    {
      printf("null Same\n");
    }

    X x1;
    X x2;

    if(x1 != x1)
    {
      printf("X Different\n");
    }
    else
    {
      printf("X Same\n");
    }

    return 0;
  }

prints


  "null Same"
  "Error: Access Violation"

which is a surprise to me, as I'd expected an access violation for both. This is inconsistent, to say the least. (I shall post this on D ng now, methinks).

This is a constant beef between me Walter. It's a curious battle, since I have right on my side, and he has absolute authority on his. (Kind of like being married <g>)

Matthew Wilson

STLSoft moderator and C++ monomaniac

mailto:matthew@stlsoft.org
http://www.stlsoft.org
news://news.digitalmars.com/c++.stlsoft

"You can tell a Yorkshireman, but you can't tell him much!" -- Uncle Michael

----------------------------------------------------------------------------
---