Re: [Boost-users] challange to MPL gurus

Subject: Re: [Boost-users] challange to MPL gurus
From: Larry Evans (cppljevans_at_[hidden])
Date: 2011-12-12 17:30:29

On 12/12/11 13:41, Robert Ramey wrote:
> I am trying to implement concept checking for a couple of MPL concepts as
> part of a large project.  I've having a heck of a time with some aspects of 
> this.
> 
> The example below illustrates one of the problems I'm having.
> The intent is to test the concept.  for next<I>, the concept

Does ForwardIterator<I> attempt to test that all elements in
the iterator, I, satisfy some concept?  If so, then I don't
see any test for that within the is_end<false_> branch.  Are you
planning on adding it once you get this immediate problem solved?

Maybe the if_recur template shown here:

http://svn.boost.org/svn/boost/sandbox/variadic_templates/libs/mpl/sandbox/if_recur.cpp

would be some help.

> check is invoked recurrsively until the end of iteration is
> encountered.  (I'm assuming it's a list in this test).  But the
> is_end<false_> branch is always invoked leading to failure
> on both MSVC 9.0 and gcc 4.5.3 compilers

I couldn't get this code to compile with gcc4.7, even after
making the modifications shown in the attached.  Instead,
I get the error message:

ramey_iter_concept_check.cpp:34:10: error: invalid explicit
specialization before '>' token
ramey_iter_concept_check.cpp:34:10: error: enclosing class templates are
not explicitly specialized
ramey_iter_concept_check.cpp:35:28: error: template parameters not used
in partial specialization:
ramey_iter_concept_check.cpp:35:28: error:         'I'

The reason the specialization was moved outside of the containing
class was because gcc said you couldn't specialize a nested template
in the class declaration; however, even when the is_end specialization
was moved outside the class declaration, the above error message occurred.

I'm surprised gcc 4.5.3 compiles your code because, IIRC, the problem
of nested template specializations has been around for some time.

> 
> I'll be quite impressed with anyone who can figure out
> how to make this work.
> 
> Robert Ramey
> 
> 
> template<typename I>
> struct ForwardIterator {
>     template<typename B>
>     struct is_end {};
>     template<>
>     struct is_end<false_> {
>         typedef typename deref<I>::type t1;
>         typedef typename next<I>::type t2;
>         BOOST_CONCEPT_ASSERT(( ForwardIterator<t2> ));
>     };
>     typedef typename boost::mpl::print<
>         I
>     >::type t1;
>     typedef typename is_end<
>         typename boost::is_same<
>             boost::mpl::l_iter<boost::mpl::l_end>,
>             I
>         >
>         > type;
>         BOOST_MPL_ASSERT((
>                 is_convertible<
>                     typename I::category,
>                     forward_iterator_tag
>                 >
>         ));
> };
> 
> template<typename S>
> struct ForwardSequence {
>     typedef typename end<S>::type t2;
>     BOOST_CONCEPT_ASSERT(( ForwardIterator<t2> ));
> };
> 
> #include <boost/mpl/list.hpp>
> void test_list_sequence(){
>     BOOST_CONCEPT_ASSERT(( boost::mpl::ForwardSequence<
>             boost::mpl::list<int, char>
>         >
>     ));
> }