apf::Vector< N > Class Template Reference

template-generic vector of N doubles More...

Inherits Array< double, N >.

Public Member Functions
	Vector ()
	mandatory
	Vector (double const *v)
	construct from array
Vector< N >	operator+ (Vector< N > const &b) const
	add two vectors
Vector< N > &	operator+= (Vector< N > const &b)
	add a vector to this vector
Vector< N >	operator- (Vector< N > const &b) const
	subtract two vectors
Vector< N >	operator* (double s) const
	multiply a vector times a scalar More...
Vector< N >	operator/ (double s) const
	divide a vector by a scalar More...
double	operator* (Vector< N > const &b) const
	vector dot product More...
double	getLength () const
	get the vector magnitude
Vector< N >	normalize () const
	divide the vector by its magnitude
void	zero ()
	zero the vector

Detailed Description

template<std::size_t N>
class apf::Vector< N >

template-generic vector of N doubles

this class implements a linear algebra vector whose size is known at compile time. It is intended to be used for small, dense vectors whose operations benefit from the optimization offered by templates and inline methods because they are so small that the overhead of function calls and conditionals is high.

For vectors sized at runtime, see apf::DynamicVector. For sparse structures or parallel vectors, look outside of APF.

apf::Vector objects should be used in a functional programming style, i.e. treated as immutable values for the most part, then we expect the compiler to optimize the assignments into faster code.

Definition at line 26 of file apfVector.h.

Member Function Documentation

template<std::size_t N>

Vector<N> apf::Vector< N >::operator* ( double  s ) const

inline

multiply a vector times a scalar

currently there is no scalar-times-vector operator, so do be sure to put scalar on the right hand side.

Definition at line 83 of file apfVector.h.

    {
      Vector<N> c;
      for (std::size_t i=0; i < N; ++i)
        c[i] = (*this)[i] * s;
      return c;
    }

template<std::size_t N>

double apf::Vector< N >::operator* ( Vector< N > const &  b ) const

inline

vector dot product

We chose the default vector-vector multiplication operator to be the dot product. So far this seems to have been a good choice

Definition at line 103 of file apfVector.h.

    {
      double r=0;
      for (std::size_t i=0; i < N; ++i)
        r += (*this)[i] * b[i];
      return r;
    }

template<std::size_t N>

Vector<N> apf::Vector< N >::operator/ ( double  s ) const

inline

divide a vector by a scalar

equivalent to scaling by 1/s

Definition at line 92 of file apfVector.h.

    {
      Vector<N> c;
      for (std::size_t i=0; i < N; ++i)
        c[i] = (*this)[i] / s;
      return c;
    }

---

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

• apfVector.h