CODEWITHSUNDEEP
c++g++operator-overloading
William
William

Reputation: 11

Operator Overloading in C++

http://pastebin.com/Y7SXnrqr

This is a component of a larger OpenGL project. It defines a three-dimensional vector structure for use in doing some of the physics and graphics in said program. Apparently, my professor can get this to compile problem-free in Visual Studio, but when I try to compile it in any g++-based system it generates errors. These errors are as follows:

/Point3f.h:67: error: no match for 'operator*=' in '((const Vector3f&)((const Vector3f*)v1)) *= scale'
/Point3f.h:77: error: no match for 'operator+=' in '((const Vector3f&)((const Vector3f*)v1)) += v2'
/Point3f.h:82: error: no match for 'operator-=' in '((const Vector3f&)((const Vector3f*)v1)) -= v2'

I understand that lines 40-63 are overloading the *=, +=, and -= operators. What I'm confused about is why Visual Studio lets this compile and G++ wont. Any ideas?

Upvotes: 1

Views: 623

Answers (2)

tpdi
tpdi

Reputation: 35141

Because the first operands of the overloaded functions aren't const. But they shouldn't be (in fact, can't be).

But you don't have the copy ctor that the non-assignment operators use. Try adding the copy ctor.

 Vector3f(const Vector3f& rhs):X(rhs.X), Y(rhs.Y), Z(rhs.Z){}

On edit: or maybe I'm completely off base. :)

And why are the op= operators written as friends? They don't need to be, as they aren't symmetric.

Upvotes: 0

Thomas Edleson
Thomas Edleson

Reputation: 2196

57:  friend Vector3f& operator *= (Vector3f& v1, float scale)
67:  return Vector3f(v1) *= scale;

You can't bind a temporary (in line 67) to a non-const ref (in line 57).

For each operator, write them as:

struct Vector3f {
    // ...
    Vector3f& operator*=(float scale) {
        // ...
        return *this;
    }
    friend Vector3f operator*(Vector3f v, float scale) {
        return v *= scale;
    }
    friend Vector3f operator*(float scale, Vector3f v) {
        return v *= scale;
    }

Upvotes: 2

Related Questions