Re: Dumbed-down

Seungbeom Kim wrote:
 > kanze wrote:
 >>> Seungbeom Kim posted:
 >>>> template <class Target, class Source>
 >>>> inline Target numeric_cast(Source x)
 >>>> {
 >>>> assert( x is in the range of Target );
 >>>> return static_cast<Target>(x);
 >>>> }
 >>>> (I'm not sure how to express in C++ what I wrote inside assert( );
 >>> assert( numeric_limits<Target>::max() >= x );
 >> assert( std::numeric_limits< Target >::max() >= x
 >> && (! std::numeric_limits< Target >::is_signed
 >> || (std::numeric_limits< Target >::is_integer
 >> ? std::numeric_limits< Target >::min() >= x
 >> : -std::numeric_limits< Target >::max() >= x)=
) )
 >> ;

 > Thanks for your idea.

 > But this causes numeric_cast<signed>(1) to fail the assertion,
 > because 'std::numeric_limits< Target >::min() >= x' is false.
 > You might have meant 'std::numeric_limits< Target >::min() <= x'

Obviously:-).

 > but this causes numeric_cast<signed>(1000000000U) to fail the
 > assertion (where int is 32 bits).

Or even numeric_cast< signed >( 1U ).

Yuck. You're right. Worse: I'm not sure right off of a good
solution. I suspect that you'd have to do something involving
the signedness of Source as well. (If the first test passes,
and !std::numeric_limits< Source >::is_signed, no further tests
are applicable.)

Note that my expression also fails in the reverse case:
     numeric_limits< unsigned >( -1 )
Again, -1 gets converted to unsigned in some of the
comparisons.

The best I can come up with rapidly is:

     template< bool condition >
     class Discriminator
     {
     } ;

     // Target is_signed, ! is_integer, Source is_signed
     template< typename Target, typename Source >
     inline bool
     isValidLower( Source value, Discriminator< true >, Discriminator<
false > )
     {
         return -std::numeric_limits< Target >::max() <= value ;
     }

     // Target is_signed, is_integer, Source is_signed
     template< typename Target, typename Source >
     inline bool
     isValidLower( Source value, Discriminator< true >, Discriminator<
true > )
     {
         return std::numeric_limits< Target >::min() <= value ;
     }

     // Target ! is_signed, is_integer, Source is_signed
     template< typename Target, typename Source >
     inline bool
     isValidLower( Source value, Discriminator< false >, Discriminator<
true > )
     {
         return 0 <= value ;
     }

     // Source unsigned...
     template< typename Target, typename Source >
     inline bool
     isValid( Source value, Discriminator< false >, Discriminator< false =


     {
         return std::numeric_limits< Target >::max() <= value ;
     }

     // Source signed
     template< typename Target, typename Source >
     inline bool
     isValid( Source value, Discriminator< true >, Discriminator< false >=
 )
     {
         // Note: this may give a false negative if Target is
         // unsigned and value is negative. But in this case,
         // the second condition would be sure to fail anyway,
         // so the results would still be false.
         return std::numeric_limits< Target >::max() <= value
             && isValidLower(
                 value,
                 Discriminator< std::numeric_limits< Target >::is_signed
 > >(),
                 Discriminator< std::numeric_limits< Target
 >::is_integer> >() ) ;
     }

     template< typename Target, typename Source >
     Target
     numeric_cast( Source x )
     {
         assert(
             isValid< Target >( x,
                 Discriminator< std::numeric_limits< Source
 >::is_integer >(),
                 Discriminator< std::numeric_limits< Target
 >::is_integer >() ) ) ;
         return static_cast< Target >( x ) ;
     }

Untested, but there's got to be a simpler solution.

--
James Kanze kanze.james@neuf.fr
Conseils en informatique orient=E9e objet/
                   Beratung in objektorientierter Datenverarbeitung
9 place S=E9mard, 78210 St.-Cyr-l'=C9cole, France +33 (0)1 30 23 00 34

---
[ comp.std.c++ is moderated. To submit articles, try just posting with ]
[ your news-reader. If that fails, use mailto:std-c++@ncar.ucar.edu ]
[ --- Please see the FAQ before posting. --- ]
[ FAQ: http://www.comeaucomputing.com/csc/faq.html ]