RE: [boost] Submission: typelist

From: Hamish Mackenzie (boost_at_[hidden])
Date: 2001-11-27 09:56:35

On Tue, 2001-11-27 at 14:11, Emily Winch wrote:
> So long as we put in an assert that head_value and Tail::head_value are the
> same type. Otherwise people could make really counterintuitive types by
> putting

I think you mean head_type.

In fact if it should realy be value_type if the list is not going to be
heterogenous (which was what I was thinking).

template< class Type, Type value, class Tail >
class static_const_value_list
{
public:
  BOOST_STATIC_ASSERT(
      is_same< Type, typename Tail::value_type >::value )

  static const Type head_value = Value;
  typedef Type value_type;
  typedef Tail tail_type;
  class nil
  {
    typedef Type value_type;
  };
};

 
> Unless we _wanted_ a heterogenous compile time value list. Is there a use
> for that?

Yes, how about a compile time sprintf function.

Of course we would need a good compile time string system to be able to
use it easily.

If ctstring( "abc" ) expanded to recusive< char_list, 'a', 'b', 'c' >
that would do it.

Then you would define

template< class FormatString, class Args >
class compile_time_sprintf
{
public:
  typedef ???? result;
};

Result would be another char_list with the formated numbers in it. 

Hamish

> 
> Emily
> 
> Info: http://www.boost.org  Unsubscribe: <mailto:boost-unsubscribe_at_[hidden]> 
> 
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ 
> 
>