cpp_double_fp documentation error

[jrade]

The page
https://www.boost.org/doc/libs/latest/libs/multiprecision/doc/html/boost_multiprecision/tut/floats/cpp_double_fp_backend.html
states that one should include the header
boost/multiprecision/cpp_double_fp_backend.hpp
This is wrong. There is no such header. The correct header is
boost/multiprecision/cpp_double_fp.hpp

[ckormanyos]

Thank you @jrade I will fix this. We just published 1.90 and that release is closed. So the correction won't actually make it into Boost until spring for Boost 1.91.

[ckormanyos]

See also #748

[ckormanyos]

Test locally with the following.

#include <boost/math/constants/constants.hpp>
#include <boost/multiprecision/cpp_double_fp.hpp>

#include <iomanip>
#include <iostream>
#include <limits>
#include <sstream>

auto main() -> int;

auto main() -> int
{
  using double_fp_type = boost::multiprecision::cpp_double_double;

  const double_fp_type sqr { sqrt(boost::math::constants::pi<double_fp_type>()) };

  std::stringstream strm { };

  strm << std::setprecision(std::numeric_limits<double_fp_type>::digits10) << std::fixed << sqr;

  std::cout << strm.str() << std::endl;
}

[ckormanyos]

Fixed by #748