From: Christoph Ludwig (cludwig_at_[hidden])
Date: 2004-06-25 04:48:37
On Thu, Jun 24, 2004 at 05:51:15PM +0200, Daniel Frey wrote:
> Robert Ramey wrote:
> >BOOST_STATIC_ASSERT(false)
>
> This *should* work. IMHO, if it doesn't, we should fix it, not work
> around it like Michiel suggested.
BOOST_STATIC_ASSERT(false) expands to a typedef that does not contain
any dependent expression. I didn't check the detailed rules in my copy
of the template book by Vandevoorde and Josuttis, but I think a
compiler should check this typedef even if the function template is
not instantiated.
> >However, its doesn't work
> >with compilers that are too smart for their own good. In order to make
> >this
> >work I need something like:
>
> Which compiler and which version of boost and the Compiler are you
> using? Can you provide a minimal but complete example? Also, you might
> want to have a look at checked_delete where we had similar problems in
> the past IIRC.
Try the following program:
#include <boost/static_assert.hpp>
template <typename T>
void f(T t) {
#ifdef DEPENDENT_ASSERT_CONDITION
BOOST_STATIC_ASSERT(sizeof(T) && false);
#else
BOOST_STATIC_ASSERT(false);
#endif
}
#ifdef INSTANTIATE_F
void g(int i) {
f(i);
}
#endif
g++ 3.3.1 accepts it unless you define INSTANTIATE_F.
g++ 3.4.0 rejects it unless you define DEPENDENT_ASSERT_CONDITION (and
don't define INSTANTIATE_F, of course).
I don't know if my solution (sizeof(T) && false) has any
(dis-)advantages compared to Michiel's solution ((T*) 0).
Regards
Christoph
PS: Ramey, I patched your serialization library (based on
serialization18.zip) such that g++ 3.4 compiles the
examples. Besides above problem and the many errors due to the
unqualified use of dependent names I had to work around a bug in
libstdc++v3 (PR#16154). But I didn't have the time yet for
extensive tests.
If you are interested then I can send you my patches.
-- http://www.informatik.tu-darmstadt.de/TI/Mitarbeiter/cludwig.html LiDIA: http://www.informatik.tu-darmstadt.de/TI/LiDIA/Welcome.html