MML: a Minimalistic Maths Library

The MML was born as a collection of various maths tools that I needed at some point in some of my projects. It is now a rather unified library that focuses on fundamental 2D shapes.

It is a "headers only" library, as it relies on templates / macros to decide what types to use for value storage and for intermediate computations, and what relational operations to use on them.

Shapes

The MML expose a variadic Shape structure, implemented with boost variant. Each shape can be one of those fundamental shapes:

• a Point / Vector;
• a Line segment;
• a Rect (which is really an AABB);
• a Circle;
• a generic Polygon (a basic list of points).

#include <nauths/mml/mml.hh>

typedef mml::shapes<int>::Point  Point;
typedef mml::shapes<int>::Rect   Rect;
typedef mml::shapes<int>::Circle Circle;
typedef mml::shapes<int>::Shape  Shape;

void test()
{
  Shape circle = Circle(Point(10, 10), 10);
  Shape box = Rect(Point(0, 0), Point(20, 20));
  Rect circle_bbox = circle.bounding_rect();

  assert(circle_bbox == box);

  std::cout << "circle area: " << circle.area() << std::endl
            << "box area:    " << box.area()    << std::endl;
}

Algorithms

Some high-level algorithms that work on shapes include inclusion test and overlapping tests. One can even use a given optional macro to define relational operators on shapes based on the inclusion notion.

#define MML_OPERATORS // enables < and > relations
#define MML_MACROS    // enables _contains_ and _is_in_ alias

#include <nauths/mml/mml.hh>

typedef mml::shapes<int>::Shape Shape;

void test(Shape const& s1, Shape const& s2)
{
  assert(mml::contains(s1, s2) == (s1 > s2));
  assert(mml::contains(s1, s2) == (s1 _contains_ s2));
  assert(mml::is_in(s1, s2) == (s1 < s2));
  assert(mml::is_in(s1, s2) == (s1 _is_in_ s2));
}

Tools

The MML implements several other tools, such as a basic Perlin Noise generator and a regular / semi-regular tiling generator. For more information, please see the full documentation in the Wiki.