Re: [Boost-users] [mpl::math] exponential metafunction for complex types

From: Chris Weed (chrisweed_at_[hidden])
Date: 2007-02-22 10:50:02

Thanks,
that works great.

I also noticed that there is a problem with the double exponential metafunction.

For example:
#include <boost/mpl/math/double.hpp>
#include <boost/mpl/math/complex_number.hpp>
#include <iostream>

int main()
{
  typedef boost::mpl::math::zero_dispatch<boost::mpl::math::double_tag>
double_zero;
  typedef boost::mpl::math::string_c_to_double<'-',2,'.',0>::type minus_two;
  typedef boost::mpl::times<minus_two,boost::mpl::math::pi_dispatch<boost::mpl::math::double_tag>
>::type minus_two_pi;

  std::cout << boost::mpl::math::exponential<minus_two_pi>::type() << std::endl;
  return 0;
}

outputs:
0.0788433

I increased BOOST_MPL_LIMIT_MATH_EXPONENTIAL_SERIES to 32, but I am
not sure what exactly this should be.

the example program outputs:
0.00186744

Chris

On 2/21/07, Cromwell Enage <sponage_at_[hidden]> wrote:
> --- Cromwell Enage wrote:
> > The problem was in the sine and cosine functions,
> > which by default lose their accuracy when the angle
> > is beyond tha range (-pi/2,pi/2).
>
> Sorry, I meant [-pi,pi].
>
>                               Cromwell D. Enage
>
>
>
>
> ____________________________________________________________________________________
> Don't pick lemons.
> See all the new 2007 cars at Yahoo! Autos.
> http://autos.yahoo.com/new_cars.html
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users
>