Re: [boost] type_traits: more state-of-the-union

From: David Abrahams (david.abrahams_at_[hidden])
Date: 2002-04-16 06:58:44

----- Original Message -----
From: "John Maddock" <john_maddock_at_[hidden]>
>
> There are so many *?*! workarounds that's the problem :-(
> You are right though, we may need to rework the headers sometime.

I think we have primitive definitions for enough of the traits that you
might be surprised at how nicely it decomposes.

> > 8. I think we need some new config macros, but I'd like someone who
has
> > done this before to add them if possible
> >
> >     BOOST_ELIPSIS_ARGS_DO_NOT_INVOKE_COPY_CONSTRUCTOR - for
compilers
> > that can resolve overloads to functions accepting elipsis arguments
> > without invoking the copy constructor of the actual argument. This
is
> > nonconforming behavior AFAICT but most compilers seem to allow it,
and
> > it's what allows is_convertible<T,U> to work when T is noncopyable.
>
> I think we can fix that by not passing object through elipses in
> is_convertible - see gcc implementation.

If that works, you could've saved me about 2 days' work by telling my
earlier ;-(
Oh, well... I leave it in your hands now.

-Dave