numerix 1.0.0

data/ext/numerix/vector3.c

@@ -0,0 +1,805 @@
+
+#include "vector3.h"
+
+void Init_vector3(VALUE outer) {
+    rb_define_alloc_func(rb_cVector3, rb_vector3_alloc);
+
+    // Instance Methods
+    rb_define_method(rb_cVector3, "initialize", rb_vector3_initialize, -1);
+    rb_define_method(rb_cVector3, "length", rb_vector3_length, 0);
+    rb_define_method(rb_cVector3, "length_squared", rb_vector3_length_squared, 0);
+    rb_define_method(rb_cVector3, "one?", rb_vector3_one_p, 0);
+    rb_define_method(rb_cVector3, "zero?", rb_vector3_zero_p, 0);
+    rb_define_method(rb_cVector3, "min_value", rb_vector3_min_value, 0);
+    rb_define_method(rb_cVector3, "max_value", rb_vector3_max_value, 0);
+    rb_define_method(rb_cVector3, "distance", rb_vector3_distance, 1);
+    rb_define_method(rb_cVector3, "distance_squared", rb_vector3_distance_squared, 1);
+    rb_define_method(rb_cVector3, "normalize", rb_vector3_normalize, 0);
+    rb_define_method(rb_cVector3, "normalize!", rb_vector3_normalize_bang, 0);
+    rb_define_method(rb_cVector3, "lerp", rb_vector3_lerp, 2);
+    rb_define_method(rb_cVector3, "lerp!", rb_vector3_lerp_bang, 2);
+    rb_define_method(rb_cVector3, "transform", rb_vector3_transform, 1);
+    rb_define_method(rb_cVector3, "transform!", rb_vector3_transform_bang, 1);
+    rb_define_method(rb_cVector3, "abs", rb_vector3_abs, 0);
+    rb_define_method(rb_cVector3, "sqrt", rb_vector3_sqrt, 0);
+    rb_define_method(rb_cVector3, "dot", rb_vector3_dot, 1);
+    rb_define_method(rb_cVector3, "clamp", rb_vector3_clamp, 2);
+    rb_define_method(rb_cVector3, "clamp!", rb_vector3_clamp_bang, 2);
+    rb_define_method(rb_cVector3, "cross", rb_vector3_cross, 1);
+    rb_define_method(rb_cVector3, "cross!", rb_vector3_cross_bang, 1);
+    rb_define_method(rb_cVector3, "reflect", rb_vector3_reflect, 1);
+    rb_define_method(rb_cVector3, "reflect!", rb_vector3_reflect_bang, 1);
+    rb_define_method(rb_cVector3, "angle", rb_vector3_angle, 1);
+    rb_define_method(rb_cVector3, "transform_normal", rb_vector3_transform_normal, 1);
+    rb_define_method(rb_cVector3, "transform_normal!", rb_vector3_transform_normal_bang, 1);
+    rb_define_method(rb_cVector3, "map", rb_vector3_map, 0);
+    rb_define_method(rb_cVector3, "map!", rb_vector3_map_bang, 0);
+
+    // Conversion
+    rb_define_method(rb_cVector3, "to_s", rb_vector3_to_s, 0);
+    rb_define_method(rb_cVector3, "to_a", rb_vector3_to_a, 0);
+    rb_define_method(rb_cVector3, "to_h", rb_vector3_to_h, 0);
+    rb_define_method(rb_cVector3, "to_quaternion", rb_vector3_to_quaternion, 0);
+    rb_define_method(rb_cVector3, "to_plane", rb_vector3_to_plane, 0);
+    rb_define_method(rb_cVector3, "to_vec2", rb_vector3_to_vec2, 0);
+    rb_define_method(rb_cVector3, "to_vec4", rb_vector3_to_vec4, 0);
+
+    // Operators
+    rb_define_method(rb_cVector3, "+", rb_vector3_add, 1);
+    rb_define_method(rb_cVector3, "-", rb_vector3_subtract, 1);
+    rb_define_method(rb_cVector3, "*", rb_vector3_multiply, 1);
+    rb_define_method(rb_cVector3, "/", rb_vector3_divide, 1);
+    rb_define_method(rb_cVector3, "==", rb_vector3_equal, 1);
+    rb_define_method(rb_cVector3, "-@", rb_vector3_negate, 0);
+    rb_define_method(rb_cVector3, "**", rb_vector3_pow, 1);
+
+    // Alias
+    rb_define_alias(rb_cVector3, "magnitude", "length");
+    rb_define_alias(rb_cVector3, "elements", "to_a");
+    rb_define_alias(rb_cVector3, "collect", "map");
+    rb_define_alias(rb_cVector3, "collect!", "map!");
+
+    // Singleton Methods
+    rb_define_singleton_method(rb_cVector3, "zero", rb_vector3_alloc, 0);
+    rb_define_singleton_method(rb_cVector3, "one", rb_vector3_one, 0);
+    rb_define_singleton_method(rb_cVector3, "unit_x", rb_vector3_unit_x, 0);
+    rb_define_singleton_method(rb_cVector3, "unit_y", rb_vector3_unit_y, 0);
+    rb_define_singleton_method(rb_cVector3, "unit_z", rb_vector3_unit_z, 0);
+    rb_define_singleton_method(rb_cVector3, "create_norm", rb_vector3_create_norm, 3);
+    rb_define_singleton_method(rb_cVector3, "clamp", rb_vector3_clamp_s, 3);
+    rb_define_singleton_method(rb_cVector3, "lerp", rb_vector3_lerp_s, 3);
+    rb_define_singleton_method(rb_cVector3, "min", rb_vector3_min_s, 2);
+    rb_define_singleton_method(rb_cVector3, "max", rb_vector3_max_s, 2);
+}
+
+static VALUE rb_vector3_alloc(VALUE klass) {
+    void *v = ALLOC(Vector3);
+    memset(v, 0, sizeof(Vector3));
+    return NUMERIX_WRAP(klass, v);
+}
+
+VALUE rb_vector3_initialize(int argc, VALUE *argv, VALUE self) {
+    VECTOR3();
+    switch (argc) {
+        case 0:
+            break;
+        case 1: {
+            float value = NUM2FLT(argv[0]);
+            v->x = value;
+            v->y = value;
+            v->z = value;
+            break;
+        }
+        case 2: {
+            Vector2 *v2;
+            Data_Get_Struct(argv[0], Vector2, v2);
+            v->x = v2->x;
+            v->y = v2->y;
+            v->z = NUM2FLT(argv[1]);
+            break;
+        }
+        case 3: {
+            v->x = NUM2FLT(argv[0]);
+            v->y = NUM2FLT(argv[1]);
+            v->z = NUM2FLT(argv[2]);
+            break;
+        }
+        default:
+            rb_raise(rb_eArgError, "wrong number of arguments (given %d, expected 0, 1, 2, 3)", argc);
+            break;
+    }
+    return Qnil;
+}
+
+VALUE rb_vector3_one(VALUE klass) {
+    Vector3 *v = ALLOC(Vector3);
+    v->x = 1.0f;
+    v->y = 1.0f;
+    v->z = 1.0f;
+    return NUMERIX_WRAP(klass, v);
+}
+
+VALUE rb_vector3_unit_x(VALUE klass) {
+    Vector3 *v = ALLOC(Vector3);
+    v->x = 1.0f;
+    v->y = 0.0f;
+    v->z = 0.0f;
+    return NUMERIX_WRAP(klass, v);
+}
+
+VALUE rb_vector3_unit_y(VALUE klass) {
+    Vector3 *v = ALLOC(Vector3);
+    v->x = 0.0f;
+    v->y = 1.0f;
+    v->z = 0.0f;
+    return NUMERIX_WRAP(klass, v);
+}
+
+VALUE rb_vector3_unit_z(VALUE klass) {
+    Vector3 *v = ALLOC(Vector3);
+    v->x = 0.0f;
+    v->y = 0.0f;
+    v->z = 1.0f;
+    return NUMERIX_WRAP(klass, v);
+}
+
+VALUE rb_vector3_length(VALUE self) {
+    VECTOR3();
+    return DBL2NUM(sqrtf(v->x * v->x + v->y * v->y + v->z * v->z));
+}
+
+VALUE rb_vector3_length_squared(VALUE self) {
+    VECTOR3();
+    return DBL2NUM(v->x * v->x + v->y * v->y + v->z * v->z);
+}
+
+VALUE rb_vector3_add(VALUE self, VALUE other) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    result->x = v1->x + v2->x;
+    result->y = v1->y + v2->y;
+    result->z = v1->z + v2->z;
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_subtract(VALUE self, VALUE other) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    result->x = v1->x - v2->x;
+    result->y = v1->y - v2->y;
+    result->z = v1->z - v2->z;
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_multiply(VALUE self, VALUE other) {
+    Vector3 *v1, *result;
+    Data_Get_Struct(self, Vector3, v1);
+    result = ALLOC(Vector3);
+
+    if (NUMERIX_TYPE_P(other, rb_cVector3)) {
+        Vector3 *v2;
+        Data_Get_Struct(other, Vector3, v2);
+        result->x = v1->x * v2->x;
+        result->y = v1->x * v2->y;
+        result->z = v1->z * v2->z;
+    } else {
+        float scalar = NUM2FLT(other);
+        result->x = v1->x * scalar;
+        result->y = v1->y * scalar;
+        result->z = v1->z * scalar;
+    }
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_divide(VALUE self, VALUE other) {
+    Vector3 *v1, *result;
+    Data_Get_Struct(self, Vector3, v1);
+    result = ALLOC(Vector3);
+
+    if (NUMERIX_TYPE_P(other, rb_cVector3)) {
+        Vector3 *v2;
+        Data_Get_Struct(other, Vector3, v2);
+        result->x = v1->x / v2->x;
+        result->y = v1->y / v2->y;
+        result->z = v1->z / v2->z;
+    } else {
+        float scalar = 1.0f / NUM2FLT(other);
+        result->x = v1->x * scalar;
+        result->y = v1->y * scalar;
+        result->z = v1->z * scalar;
+    }
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_equal(VALUE self, VALUE other) {
+    if (CLASS_OF(other) != CLASS_OF(self))
+        return Qfalse;
+    Vector3 *v1, *v2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+    return FLT_EQUAL(v1->x, v2->x) && FLT_EQUAL(v1->y, v2->y) && FLT_EQUAL(v1->z, v2->z) ? Qtrue : Qfalse;
+}
+
+VALUE rb_vector3_pow(VALUE self, VALUE exponent) {
+    VECTOR3();
+    Vector3 *result = ALLOC(Vector3);
+    float e = fabsf(NUM2FLT(exponent));
+
+    result->x = powf(fabsf(v->x), e);
+    result->y = powf(fabsf(v->y), e);
+    result->z = powf(fabsf(v->z), e);
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_negate(VALUE self) {
+    Vector3 *v, *result;
+    Data_Get_Struct(self, Vector3, v);
+    result = ALLOC(Vector3);
+
+    result->x = -v->x;
+    result->y = -v->y;
+    result->z = -v->z;
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_map(VALUE self) {
+    VECTOR3();
+    Vector3 *result = ALLOC(Vector3);
+
+    result->x = NUM2FLT(rb_yield(DBL2NUM(v->x)));
+    result->y = NUM2FLT(rb_yield(DBL2NUM(v->y)));
+    result->z = NUM2FLT(rb_yield(DBL2NUM(v->z)));
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_map_bang(VALUE self) {
+    VECTOR3();
+
+    v->x = NUM2FLT(rb_yield(DBL2NUM(v->x)));
+    v->y = NUM2FLT(rb_yield(DBL2NUM(v->y)));
+    v->z = NUM2FLT(rb_yield(DBL2NUM(v->z)));
+
+    return self;
+}
+
+VALUE rb_vector3_to_s(VALUE self) {
+    VECTOR3();
+    return rb_sprintf("<%f, %f, %f>", v->x, v->y, v->z);
+}
+
+VALUE rb_vector3_to_a(VALUE self) {
+    VECTOR3();
+    VALUE ary = rb_ary_new_capa(3);
+    rb_ary_store(ary, 0, DBL2NUM(v->x));
+    rb_ary_store(ary, 1, DBL2NUM(v->y));
+    rb_ary_store(ary, 2, DBL2NUM(v->z));
+    return ary;
+}
+
+VALUE rb_vector3_to_h(VALUE self) {
+    VECTOR3();
+    VALUE hash = rb_hash_new();
+    rb_hash_aset(hash, ID2SYM(rb_intern("x")), DBL2NUM(v->x));
+    rb_hash_aset(hash, ID2SYM(rb_intern("y")), DBL2NUM(v->y));
+    rb_hash_aset(hash, ID2SYM(rb_intern("z")), DBL2NUM(v->z));
+    return hash;
+}
+
+VALUE rb_vector3_to_vec2(VALUE self) {
+    VECTOR3();
+    Vector2 *vec = ALLOC(Vector2);
+    vec->x = v->x;
+    vec->y = v->y;
+    return NUMERIX_WRAP(rb_cVector2, vec);
+}
+
+VALUE rb_vector3_to_vec4(VALUE self) {
+    VECTOR3();
+    Vector4 *vec = ALLOC(Vector4);
+    vec->x = v->x;
+    vec->y = v->y;
+    vec->z = v->z;
+    vec->w = 0.0f;
+    return NUMERIX_WRAP(rb_cVector4, vec);
+}
+
+VALUE rb_vector3_to_quaternion(VALUE self) {
+    VECTOR3();
+    Quaternion *q = ALLOC(Quaternion);
+    memcpy(q, v, sizeof(Vector3));
+    q->w = 0.0f;
+    return NUMERIX_WRAP(rb_cQuaternion, q);
+}
+
+VALUE rb_vector3_to_plane(VALUE self) {
+    VECTOR3();
+    Plane *p = ALLOC(Plane);
+    memcpy(p, v, sizeof(Vector3));
+    p->distance = 0.0f;
+    return NUMERIX_WRAP(rb_cPlane, p);
+}
+
+VALUE rb_vector3_min_value(VALUE self) {
+    VECTOR3();
+    float m = NUMERIX_MIN(v->x, NUMERIX_MIN(v->y, v->z));
+    return DBL2NUM(m);
+}
+
+VALUE rb_vector3_max_value(VALUE self) {
+    VECTOR3();
+    float m = NUMERIX_MAX(v->x, NUMERIX_MAX(v->y, v->z));
+    return DBL2NUM(m);
+}
+
+VALUE rb_vector3_distance(VALUE self, VALUE other) {
+    Vector3 *v1, *v2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+
+    float dx = v1->x - v2->x;
+    float dy = v1->y - v2->y;
+    float dz = v1->z - v2->z;
+
+    return DBL2NUM(sqrtf(dx * dx + dy * dy + dz * dz));
+}
+
+VALUE rb_vector3_distance_squared(VALUE self, VALUE other) {
+    Vector3 *v1, *v2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+
+    float dx = v1->x - v2->x;
+    float dy = v1->y - v2->y;
+    float dz = v1->z - v2->z;
+
+    return DBL2NUM(dx * dx + dy * dy + dz * dz);
+}
+
+VALUE rb_vector3_normalize(VALUE self) {
+    VECTOR3();
+    Vector3 *result = ALLOC(Vector3);
+
+    float inv = 1.0f / sqrtf(v->x * v->x + v->y * v->y + v->z * v->z);
+    result->x = v->x * inv;
+    result->y = v->y * inv;
+    result->z = v->z * inv;
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_lerp(VALUE self, VALUE other, VALUE amount) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    float w = NUMERIX_CLAMP(NUM2FLT(amount), 0.0f, 1.0f);
+
+    result->x = v1->x + (v2->x - v1->x) * w;
+    result->y = v1->y + (v2->y - v1->y) * w;
+    result->z = v1->z + (v2->z - v1->z) * w;
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_transform(VALUE self, VALUE other) {
+    Vector3 *v, *result;
+    Data_Get_Struct(self, Vector3, v);
+    result = ALLOC(Vector3);
+
+    if (NUMERIX_TYPE_P(other, rb_cMatrix4x4)) {
+        Matrix4x4 *m;
+        Data_Get_Struct(other, Matrix4x4, m);
+
+        result->x = v->x * m->m11 + v->y * m->m21 + v->z * m->m31 + m->m41;
+        result->y = v->x * m->m12 + v->y * m->m22 + v->z * m->m32 + m->m42;
+        result->z = v->x * m->m13 + v->y * m->m23 + v->z * m->m33 + m->m43;
+    } else {
+        Quaternion *q;
+        Data_Get_Struct(other, Quaternion, q);
+
+        float x2 = q->x + q->x;
+        float y2 = q->y + q->y;
+        float z2 = q->z + q->z;
+
+        float wx2 = q->w * x2;
+        float wy2 = q->w * y2;
+        float wz2 = q->w * z2;
+        float xx2 = q->x * x2;
+        float xy2 = q->x * y2;
+        float xz2 = q->x * z2;
+        float yy2 = q->y * y2;
+        float yz2 = q->y * z2;
+        float zz2 = q->z * z2;
+
+        result->x = v->x * (1.0f - yy2 - zz2) + v->y * (xy2 - wz2) + v->z * (xz2 + wy2);
+        result->y = v->x * (xy2 + wz2) + v->y * (1.0f - xx2 - zz2) + v->z * (yz2 - wx2);
+        result->z = v->x * (xz2 - wy2) + v->y * (yz2 + wx2) + v->z * (1.0f - xx2 - yy2);
+    }
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_abs(VALUE self) {
+    VECTOR3();
+    Vector3 *result = ALLOC(Vector3);
+
+    result->x = fabsf(v->x);
+    result->y = fabsf(v->y);
+    result->z = fabsf(v->z);
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_sqrt(VALUE self) {
+    VECTOR3();
+    Vector3 *result = ALLOC(Vector3);
+
+    result->x = sqrtf(v->x);
+    result->y = sqrtf(v->y);
+    result->z = sqrtf(v->z);
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_dot(VALUE self, VALUE other) {
+    Vector3 *v1, *v2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+
+    return DBL2NUM(v1->x * v2->x + v1->y * v2->y + v1->z * v2->z);
+}
+
+VALUE rb_vector3_clamp(VALUE self, VALUE min, VALUE max) {
+    return rb_vector3_clamp_s(CLASS_OF(self), self, min, max);
+}
+
+VALUE rb_vector3_normalize_bang(VALUE self) {
+    VECTOR3();
+    float inv = 1.0f / sqrtf(v->x * v->x + v->y * v->y + v->z * v->z);
+    v->x = v->x * inv;
+    v->y = v->y * inv;
+    v->z = v->z * inv;
+
+    return self;
+}
+
+VALUE rb_vector3_lerp_bang(VALUE self, VALUE other, VALUE amount) {
+    Vector3 *v1, *v2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+    float w = NUMERIX_CLAMP(NUM2FLT(amount), 0.0f, 1.0f);
+
+    v1->x = v1->x + (v2->x - v1->x) * w;
+    v1->y = v1->y + (v2->y - v1->y) * w;
+    v1->z = v1->z + (v2->z - v1->z) * w;
+
+    return self;
+}
+
+VALUE rb_vector3_transform_bang(VALUE self, VALUE other) {
+    VECTOR3();
+
+    if (NUMERIX_TYPE_P(other, rb_cMatrix4x4)) {
+        Matrix4x4 *m;
+        Data_Get_Struct(other, Matrix4x4, m);
+
+        v->x = v->x * m->m11 + v->y * m->m21 + v->z * m->m31 + m->m41;
+        v->y = v->x * m->m12 + v->y * m->m22 + v->z * m->m32 + m->m42;
+        v->z = v->x * m->m13 + v->y * m->m23 + v->z * m->m33 + m->m43;
+    } else {
+        Quaternion *q;
+        Data_Get_Struct(other, Quaternion, q);
+
+        float x2 = q->x + q->x;
+        float y2 = q->y + q->y;
+        float z2 = q->z + q->z;
+
+        float wx2 = q->w * x2;
+        float wy2 = q->w * y2;
+        float wz2 = q->w * z2;
+        float xx2 = q->x * x2;
+        float xy2 = q->x * y2;
+        float xz2 = q->x * z2;
+        float yy2 = q->y * y2;
+        float yz2 = q->y * z2;
+        float zz2 = q->z * z2;
+
+        v->x = v->x * (1.0f - yy2 - zz2) + v->y * (xy2 - wz2) + v->z * (xz2 + wy2);
+        v->y = v->x * (xy2 + wz2) + v->y * (1.0f - xx2 - zz2) + v->z * (yz2 - wx2);
+        v->z = v->x * (xz2 - wy2) + v->y * (yz2 + wx2) + v->z * (1.0f - xx2 - yy2);
+    }
+
+    return self;
+}
+
+VALUE rb_vector3_clamp_bang(VALUE self, VALUE min, VALUE max) {
+    struct RData *rdata = RDATA(self);
+    VALUE result = rb_vector3_clamp_s(rdata->basic.klass, self, min, max);
+    Vector3 *src;
+    Data_Get_Struct(result, Vector3, src);
+    memcpy(rdata->data, src, sizeof(Vector3));
+    return self;
+}
+
+VALUE rb_vector3_one_p(VALUE self) {
+    VECTOR3();
+    return v->x == 1.0f && v->y == 1.0f && v->z == 1.0f ? Qtrue : Qfalse;
+}
+
+VALUE rb_vector3_zero_p(VALUE self) {
+    VECTOR3();
+    return v->x == 0.0f && v->y == 0.0f && v->z == 0.0f ? Qtrue : Qfalse;
+}
+
+VALUE rb_vector3_cross(VALUE self, VALUE other) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    result->x = v1->y * v2->z - v1->z * v2->y;
+    result->y = v1->z * v2->x - v1->x * v2->z;
+    result->z = v1->x * v2->y - v1->y * v2->x;
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_cross_bang(VALUE self, VALUE other) {
+    Vector3 *v1, *v2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+
+    v1->x = v1->y * v2->z - v1->z * v2->y;
+    v1->y = v1->z * v2->x - v1->x * v2->z;
+    v1->z = v1->x * v2->y - v1->y * v2->x;
+
+    return self;
+}
+
+VALUE rb_vector3_reflect(VALUE self, VALUE other) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    float dot = v1->x * v2->x + v1->y * v2->y + v1->z * v2->z;
+    result->x = v1->x = (v2->x * dot * 2.0f);
+    result->y = v1->y = (v2->y * dot * 2.0f);
+    result->z = v1->z = (v2->z * dot * 2.0f);
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_reflect_bang(VALUE self, VALUE other) {
+    Vector3 *v1, *v2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+
+    float dot = v1->x * v2->x + v1->y * v2->y + v1->z * v2->z;
+    v1->x = v1->x = (v2->x * dot * 2.0f);
+    v1->y = v1->y = (v2->y * dot * 2.0f);
+    v1->z = v1->z = (v2->z * dot * 2.0f);
+
+    return self;
+}
+
+VALUE rb_vector3_angle(VALUE self, VALUE other) {
+    Vector3 *v1, *v2, *n1, *n2;
+    Data_Get_Struct(self, Vector3, v1);
+    Data_Get_Struct(other, Vector3, v2);
+
+    n1 = ruby_xmalloc(sizeof(Vector3));
+    n2 = ruby_xmalloc(sizeof(Vector3));
+
+    // Create normalized copies of the given vectors
+    float inv = 1.0f / sqrtf(v1->x * v1->x + v1->y * v1->y + v1->z * v1->z);
+    n1->x = v1->x * inv;
+    n1->y = v1->y * inv;
+    n1->z = v1->z * inv;
+
+    inv = 1.0f / sqrtf(v2->x * v2->x + v2->y * v2->y + v2->z * v2->z);
+    n2->x = v2->x * inv;
+    n2->y = v2->y * inv;
+    n2->z = v2->z * inv;
+
+    float ratio = n1->x * n2->x + n1->y * n2->y + n1->z * n2->z;
+
+    // The "straight forward" method of acos(u.v) has large precision
+    // issues when the dot product is near +/-1.  This is due to the
+    // steep slope of the acos function as we approach +/- 1.  Slight
+    // precision errors in the dot product calculation cause large
+    // variation in the output value.
+    //
+    //        |                   |
+    //         \__                |
+    //            ---___          |
+    //                  ---___    |
+    //                        ---_|_
+    //                            | ---___
+    //                            |       ---___
+    //                            |             ---__
+    //                            |                  \
+    //                            |                   |
+    //       -|-------------------+-------------------|-
+    //       -1                   0                   1
+    //
+    //                         acos(x)
+    //
+    // To avoid this we use an alternative method which finds the
+    // angle bisector by (u-v)/2:
+    //
+    //                            _>
+    //                       u  _-  \ (u-v)/2
+    //                        _-  __-v
+    //                      _=__--
+    //                    .=----------->
+    //                            v
+    //
+    // Because u and v and unit vectors, (u-v)/2 forms a right angle
+    // with the angle bisector.  The hypotenuse is 1, therefore
+    // 2*asin(|u-v|/2) gives us the angle between u and v.
+    //
+    // The largest possible value of |u-v| occurs with perpendicular
+    // vectors and is sqrt(2)/2 which is well away from extreme slope
+    // at +/-1.
+
+    float theta;
+    float length, x, y, z;
+    if (ratio < 0.0f) {
+        // Compute length of difference
+        x = -n1->x - n2->x;
+        y = -n1->y - n2->y;
+        z = -n1->z - n2->z;
+        length = sqrtf(x * x + y * y + z * z);
+
+        theta = NUMERIX_PI - 2.0f * asinf(length / 2.0f);
+    } else {
+        // Compute length of difference
+        x = n1->x - n2->x;
+        y = n1->y - n2->y;
+        z = n1->z - n2->z;
+        length = sqrtf(x * x + y * y + z * z);
+
+        theta = 2.0f * asinf(length / 2.0f);
+    }
+
+    ruby_xfree(n1);
+    ruby_xfree(n2);
+
+    // Convert radians to degrees and return
+    const float rad = 180.0f / NUMERIX_PI;
+    return DBL2NUM(theta * rad);
+}
+
+VALUE rb_vector3_transform_normal(VALUE self, VALUE matrix) {
+    VECTOR3();
+    Vector3 *result = ALLOC(Vector3);
+    Matrix4x4 *m4;
+    Data_Get_Struct(matrix, Matrix4x4, m4);
+
+    result->x = v->x * m4->m11 + v->y * m4->m21 + v->z * m4->m31;
+    result->y = v->x * m4->m12 + v->y * m4->m22 + v->z * m4->m32;
+    result->z = v->x * m4->m13 + v->y * m4->m23 + v->z * m4->m33;
+
+    return NUMERIX_WRAP(CLASS_OF(self), result);
+}
+
+VALUE rb_vector3_transform_normal_bang(VALUE self, VALUE matrix) {
+    VECTOR3();
+    Matrix4x4 *m4;
+    Data_Get_Struct(matrix, Matrix4x4, m4);
+
+    v->x = v->x * m4->m11 + v->y * m4->m21 + v->z * m4->m31;
+    v->y = v->x * m4->m12 + v->y * m4->m22 + v->z * m4->m32;
+    v->z = v->x * m4->m13 + v->y * m4->m23 + v->z * m4->m33;
+
+    return self;
+}
+
+static inline VALUE rb_vector3_clamp_s(VALUE klass, VALUE vector, VALUE minimum, VALUE maximum) {
+    Vector3 *v, *result;
+    Data_Get_Struct(vector, Vector3, v);
+    result = ALLOC(Vector3);
+    float x = v->x, y = v->y, z = v->z;
+
+    // This compare order is very important!!!
+    // We must follow HLSL behavior in the case user specified min value is bigger than max value.
+    if (NUMERIX_TYPE_P(minimum, rb_cVector3) && NUMERIX_TYPE_P(maximum, rb_cVector3)) {
+        Vector3 *min, *max;
+        Data_Get_Struct(minimum, Vector3, min);
+        Data_Get_Struct(maximum, Vector3, max);
+
+        x = NUMERIX_MIN(x, max->x);
+        x = NUMERIX_MAX(x, min->x);
+
+        y = NUMERIX_MIN(y, max->y);
+        y = NUMERIX_MAX(y, min->y);
+
+        z = NUMERIX_MIN(z, max->z);
+        z = NUMERIX_MAX(z, min->z);
+    } else {
+        float minf = NUM2FLT(minimum);
+        float maxf = NUM2FLT(maximum);
+
+        x = NUMERIX_MIN(x, maxf);
+        x = NUMERIX_MAX(x, minf);
+
+        y = NUMERIX_MIN(y, maxf);
+        y = NUMERIX_MAX(y, minf);
+
+        z = NUMERIX_MIN(z, maxf);
+        z = NUMERIX_MAX(z, minf);
+    }
+
+    result->x = x;
+    result->y = y;
+    result->z = z;
+    return NUMERIX_WRAP(klass, result);
+}
+
+static inline VALUE rb_vector3_lerp_s(VALUE klass, VALUE vec1, VALUE vec2, VALUE amount) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(vec1, Vector3, v1);
+    Data_Get_Struct(vec2, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    float w = NUMERIX_CLAMP(NUM2FLT(amount), 0.0f, 1.0f);
+
+    result->x = v1->x + (v2->x - v1->x) * w;
+    result->y = v1->y + (v2->y - v1->y) * w;
+    result->z = v1->z + (v2->z - v1->z) * w;
+
+    return NUMERIX_WRAP(klass, result);
+}
+
+static inline VALUE rb_vector3_min_s(VALUE klass, VALUE vec1, VALUE vec2) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(vec1, Vector3, v1);
+    Data_Get_Struct(vec2, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    result->x = NUMERIX_MIN(v1->x, v2->x);
+    result->y = NUMERIX_MIN(v1->y, v2->y);
+    result->z = NUMERIX_MIN(v1->z, v2->z);
+
+    return NUMERIX_WRAP(klass, result);
+}
+
+static inline VALUE rb_vector3_max_s(VALUE klass, VALUE vec1, VALUE vec2) {
+    Vector3 *v1, *v2, *result;
+    Data_Get_Struct(vec1, Vector3, v1);
+    Data_Get_Struct(vec2, Vector3, v2);
+    result = ALLOC(Vector3);
+
+    result->x = NUMERIX_MAX(v1->x, v2->x);
+    result->y = NUMERIX_MAX(v1->y, v2->y);
+    result->z = NUMERIX_MAX(v1->z, v2->z);
+
+    return NUMERIX_WRAP(klass, result);
+}
+
+VALUE rb_vector3_create_norm(VALUE klass, VALUE x, VALUE y, VALUE z) {
+    Vector3 *v = ALLOC(Vector3);
+    float vx = NUM2FLT(x);
+    float vy = NUM2FLT(y);
+    float vz = NUM2FLT(z);
+
+    float inv = 1.0f / sqrtf(vx * vx + vy * vy + vz * vz);
+    v->x = vx * inv;
+    v->y = vy * inv;
+    v->z = vz * inv;
+
+    return NUMERIX_WRAP(klass, v);
+}