Question

I’m using a data structure not written by myself that returns a realVec. This is the declaration of realVec (in typedef.h):

typedef TNT::Vector<PCReal> realVec;

For the definition of TNT Vector please see: http://calicam.sourceforge.net/doxygen/tnt__vector_8h_source.html

Definition of PCReal:

typedef double PCReal;

I need to convert this realVec into the following vec3:

struct vec3 {

GLfloat  x;
GLfloat  y;
GLfloat  z;

//
//  --- Constructors and Destructors ---
//

vec3( GLfloat s = GLfloat(0.0) ) :
x(s), y(s), z(s) {}

vec3( GLfloat _x, GLfloat _y, GLfloat _z ) :
x(_x), y(_y), z(_z) {}

vec3( const vec3& v ) { x = v.x;  y = v.y;  z = v.z; }

vec3( const vec2& v, const float f ) { x = v.x;  y = v.y;  z = f; }
...

I’m very new to C++ so my confusion probably lies in using TNT::Vector’s iterator and converting values returned by it. I’m thinking something like the below, so tell me if it makes sense. It seems to compile (no ‘make’ errors):

realVec normal = this->meshPoints.getNormalForVertex(i);
PCReal* iter = normal.begin();

vec3(*iter++, *iter++, *iter++);

I need this because I’m doing gl programming, and it’s convenient for my shaders to take vec3’s as input.

Answer 1

What you have might work, but you could improve it with some safety checks, and you don’t really have to use iterators at all. It appears realVec provides operator[] like most other vector classes.

realVec normal = this->meshPoints.getNormalForVertex(i);
if (normal.size() >= 3)
{
   vec3 x(normal[0], normal[1], normal[2]);
}