Re: [boost] Re: enable_if and BOOST_NO_SFINAE

From: Douglas Paul Gregor (gregod_at_[hidden])
Date: 2003-12-03 17:12:28

On Wed, 3 Dec 2003, David Abrahams wrote:
> Jaakko Jarvi <jajarvi_at_[hidden]> writes:
>
> >
> >> > #   if __GNUC__ == 2 && __GNUC_MINOR__ <= 95
> >>
> >> The regression tests seem to disagree that 2.95.3 can handle enable_if,
> >> and in Function I'm using:
> >>
> >> #if BOOST_WORKAROUND(__GNUC__, < 3)
> >
> >> > #   if __GNUC__ == 2 && __GNUC_MINOR__ <= 95
> >
> > This should flag 2.95.3 as BOOST_NO_SFINAE ???
>
> I think these tests are equivalent.

My mistake. I read the <= as < and saw all of the regression failures
and... anyway, the tests are equivalent.

> >> > #   if (_MSC_VER <= 1300)
> >>
> >> That's BOOST_MSVC, I hope?
> >
> > I don't know the etymology of these macros, but _MSC_VER is used
> > throughout /boost/boost/config/compiler/visualc.hpp.
> > The same file defines BOOST_MSVC as _MSC_VER.
>
> No!  _MSC_VER comes from the compiler (and from lots of other
> compilers which want to be able to compile the MFC headers).  Use
> BOOST_MSVC.

I think Jaakko's right since he's using _MSC_VER from inside visualc.hpp,
where the config system has already made sure that this is a _real_
Visual C++, not an imposter. BOOST_MSVC needs to be used outside of this
file, of course, and that's what I was reacting to.

        Doug