The documentation of GGL (Generic Geometry Library) is still scattered and hard to follow. As usual the first test I do on a library is to see how well it can be adapted to existing code. GGL seems to be able to adapt c_arrays and tuples as nD points. I was wondering if boost::array and fusion::vector can be adapted as well. The manual says it is but the header file boost/geometry/geometries/adapted/boost_array_as_ring.hpp doesn't seem to have that purpose. In the following example, p4 doesn't work as a point #include<boost/geometry/geometry.hpp> #include<boost/geometry/geometries/cartesian3d.hpp> #include<boost/geometry/geometries/adapted/boost_array_as_ring.hpp> #include <boost/geometry/geometries/adapted/c_array_cartesian.hpp> #include <boost/geometry/geometries/adapted/tuple_cartesian.hpp> #include<iostream> int main(){ using namespace boost::geometry; point<double, 3, cs::cartesian> p1(1,2,3); double p2[3] = {4,5,6}; boost::tuple<double, double, double> p3(7,8,9); boost::array<double, 3> p4={{10,11,12}}; std::clog << distance(p1,p2) << std::endl; std::clog << distance(p2,p3) << std::endl; std::clog << distance(p3,p4) << std::endl; return 0; } In other words, why there is not boost/geometry/geometries/adapted/boost_array_cartesian.hpp ? And it indeed this adaptation of boost::array is not included in the library, what steps should I follow to make a boost::array adapted to GGL? Thank you, Alfredo