Gauss-Lobatto Adaptive Integrator

A C++ port of the adaptive quadrature algorithm by Gander and Gautschi. The same algorithm is implemented in Numerical Recipes, but their license is very restrictive. I translated the code from the original paper so it can be used freely, as outlined in the MIT License.

Usage

A sample example file,example.cpp, is provided. Simply include the header AdaptiveIntegrator.hpp in your code:

    #include "AdaptiveIntegrator.hpp"

Then create an integrator object specialized for the type of functional you wish to integrate. In the simplest case this is just a normal function with signature double(const double, void*), so the declaration looks like

    AdaptiveIntegrator<double(const double, void*)> test;

Finally to integrate your function, func, invoke the integrate method with integral bounds a and b, desired tolerance tol and users' parameters param:

    test.integrate(func, a, b, tol, param);

The users' parameters structure is defined in Integrator.hpp file as:

typedef struct
{
  vector<double>* value;
} *Parameters;

In main.cpp, user defined parameters are defined as:

   double coef = 2.0;
   
   Parameters param;
   param = (Parameters) malloc(sizeof *param);
   param->value = new vector<double>();
 
   param->value->push_back(coef);
   
   // <some code lines...>
   
   free(param);