normal_vector< K > Class Template Reference

#include <qsc_approximation_fcts.hpp>

List of all members.

Public Member Functions

• normal_vector ()
• normal_vector (const point< K > &a, const point< K > &b, const point< K > &c)
• point< K > middle ()
• point< K > eval_unit (const K &t)

---

Detailed Description

template<class K>
class mmx::shape::normal_vector< K >

Definition at line 954 of file qsc_approximation_fcts.hpp.

---

Constructor & Destructor Documentation

normal_vector

(

)

[inline]

Definition at line 957 of file qsc_approximation_fcts.hpp.

{} ;

normal_vector	(	const point< K > &	a,
		const point< K > &	b,
		const point< K > &	c
	)			[inline]

Definition at line 968 of file qsc_approximation_fcts.hpp.

References KVPOL, and mmx::shape::orth_vect().

{
  data = KVPOL(1,1); //defines a degree one polynomial
  data[0]=orth_vect(2*(b-a));
  data[1]=orth_vect(2*(a-2*b+c));

  std::cout<< Bezier_curve<K>(a,b,c).normal(0)<<" toto " <<data[0]<<"\n";
  std::cout<< Bezier_curve<K>(a,b,c).normal(1)<<" toto " <<data[0]+data[1];

}

---

Member Function Documentation

point< K > eval_unit

(

const K &

t

)

[inline]

Definition at line 980 of file qsc_approximation_fcts.hpp.

References mmx::shape::dot(), mmx::shape::myeval(), mmx::sqrt(), and VECT.

Referenced by normal_vector< K >::middle(), and quad_supp_func< K >::quad_supp_func().

{
  VECT v;
  v= myeval(data,t);
  return v / std::sqrt(dot(v,v));
}

point< K > middle

(

)

[inline]

Definition at line 988 of file qsc_approximation_fcts.hpp.

References mmx::shape::dot(), normal_vector< K >::eval_unit(), mmx::shape::orth_vect(), mmx::sqrt(), and VECT.

Referenced by quad_supp_func< K >::quad_supp_func().

            : est ce meilleur que de prendre tout simplement t=1/2 ? */
{
  VECT n0=eval_unit(0);
  VECT n1=eval_unit(1);
  VECT n2;
  if ( n0 == -n1 ) 
    n2=orth_vect(n0); // Achtung ! check later if there are some numerical problems (with numerical computations, we should never have n0=-n1, but n0+n1 could be very small !
  else 
    {
      n2=n0+n1;
      n2=n2/std::sqrt(dot(n2,n2));
    }
  VECT n=eval_unit(0.5);
  K p1=dot(n,n2);
  K p2=dot(n0,n2);
  if ( p1 < p2 ) 
    return -n2;
  else 
    return n2;
}

---

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

• include/shape/qsc_approximation_fcts.hpp