Lap< arg, Sys_eqs, CENTRAL_SYM > Class Template Reference

Laplacian second order approximation CENTRAL Scheme (with central derivative in the single) More...

Detailed Description

template<typename arg, typename Sys_eqs>
class Lap< arg, Sys_eqs, CENTRAL_SYM >

Laplacian second order approximation CENTRAL Scheme (with central derivative in the single)

         1.0
          *
          | -4.0
  1.0 *---+---* 1.0
          |
          *
         1.0

* 

Definition at line 133 of file Laplacian.hpp.

#include <Laplacian.hpp>

Static Public Member Functions
static void	value (const typename stub_or_real< Sys_eqs, Sys_eqs::dims, typename Sys_eqs::stype, typename Sys_eqs::b_grid::decomposition::extended_type >::type &g_map, grid_dist_key_dx< Sys_eqs::dims > &kmap, const grid_sm< Sys_eqs::dims, void > &gs, typename Sys_eqs::stype(&spacing)[Sys_eqs::dims], std::unordered_map< long int, typename Sys_eqs::stype > &cols, typename Sys_eqs::stype coeff)
	Calculate which colums of the Matrix are non zero. More...

Member Function Documentation

template<typename arg , typename Sys_eqs >

static void Lap< arg, Sys_eqs, CENTRAL_SYM >::value ( const typename stub_or_real< Sys_eqs, Sys_eqs::dims, typename Sys_eqs::stype, typename Sys_eqs::b_grid::decomposition::extended_type >::type &  g_map, grid_dist_key_dx< Sys_eqs::dims > &  kmap, const grid_sm< Sys_eqs::dims, void > &  gs, typename Sys_eqs::stype(&)  spacing[Sys_eqs::dims], std::unordered_map< long int, typename Sys_eqs::stype > &  cols, typename Sys_eqs::stype  coeff  )

inlinestatic

Calculate which colums of the Matrix are non zero.

stub_or_real it is just for change the argument type when testing, in normal conditions it is a distributed map

Parameters

g_map	map grid
kmap	position in the grid
spacing	of the grid
gs	Grid info
cols	non-zero colums calculated by the function
coeff	coefficent (constant in front of the derivative)

Example

    // grid size
    size_t sz[2]={16,16};

    // grid_sm
    grid_sm<2,void> ginfo(sz);

    // spacing
    float spacing[2] = {0.5,0.3};

    // grid_dist_testing
    grid_dist_testing<2> g_map(sz);

    // Create several keys
    grid_dist_key_dx<2> key11(0,grid_key_dx<2>(1,1));
    grid_dist_key_dx<2> key22(0,grid_key_dx<2>(2,2));
    grid_dist_key_dx<2> key1414(0,grid_key_dx<2>(14,14));

    // filled colums
    std::unordered_map<long int,float> cols;

    Lap<Field<V,sys_pp>,sys_pp>::value(g_map,key11,ginfo,spacing,cols,1);

    BOOST_REQUIRE_EQUAL(cols.size(),5ul);

    BOOST_REQUIRE_EQUAL(cols[1],1/spacing[1]/spacing[1]);
    BOOST_REQUIRE_EQUAL(cols[17-1],1/spacing[0]/spacing[0]);
    BOOST_REQUIRE_EQUAL(cols[17+1],1/spacing[0]/spacing[0]);
    BOOST_REQUIRE_EQUAL(cols[17+16],1/spacing[1]/spacing[1]);

    BOOST_REQUIRE_EQUAL(cols[17],-2/spacing[0]/spacing[0] - 2/spacing[1]/spacing[1]);

Definition at line 155 of file Laplacian.hpp.

---

The documentation for this class was generated from the following file:

• openfpm_numerics/src/FiniteDifference/Laplacian.hpp