mirror of
				https://github.com/godotengine/godot.git
				synced 2025-10-31 13:41:03 +00:00 
			
		
		
		
	
		
			
				
	
	
		
			211 lines
		
	
	
	
		
			6.7 KiB
		
	
	
	
		
			C++
		
	
	
	
	
	
			
		
		
	
	
			211 lines
		
	
	
	
		
			6.7 KiB
		
	
	
	
		
			C++
		
	
	
	
	
	
| /**************************************************************************/
 | |
| /*  vector2.cpp                                                           */
 | |
| /**************************************************************************/
 | |
| /*                         This file is part of:                          */
 | |
| /*                             GODOT ENGINE                               */
 | |
| /*                        https://godotengine.org                         */
 | |
| /**************************************************************************/
 | |
| /* Copyright (c) 2014-present Godot Engine contributors (see AUTHORS.md). */
 | |
| /* Copyright (c) 2007-2014 Juan Linietsky, Ariel Manzur.                  */
 | |
| /*                                                                        */
 | |
| /* Permission is hereby granted, free of charge, to any person obtaining  */
 | |
| /* a copy of this software and associated documentation files (the        */
 | |
| /* "Software"), to deal in the Software without restriction, including    */
 | |
| /* without limitation the rights to use, copy, modify, merge, publish,    */
 | |
| /* distribute, sublicense, and/or sell copies of the Software, and to     */
 | |
| /* permit persons to whom the Software is furnished to do so, subject to  */
 | |
| /* the following conditions:                                              */
 | |
| /*                                                                        */
 | |
| /* The above copyright notice and this permission notice shall be         */
 | |
| /* included in all copies or substantial portions of the Software.        */
 | |
| /*                                                                        */
 | |
| /* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,        */
 | |
| /* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF     */
 | |
| /* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. */
 | |
| /* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY   */
 | |
| /* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,   */
 | |
| /* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE      */
 | |
| /* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.                 */
 | |
| /**************************************************************************/
 | |
| 
 | |
| #include "vector2.h"
 | |
| 
 | |
| #include "core/math/vector2i.h"
 | |
| #include "core/string/ustring.h"
 | |
| 
 | |
| real_t Vector2::angle() const {
 | |
| 	return Math::atan2(y, x);
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::from_angle(real_t p_angle) {
 | |
| 	return Vector2(Math::cos(p_angle), Math::sin(p_angle));
 | |
| }
 | |
| 
 | |
| real_t Vector2::length() const {
 | |
| 	return Math::sqrt(x * x + y * y);
 | |
| }
 | |
| 
 | |
| real_t Vector2::length_squared() const {
 | |
| 	return x * x + y * y;
 | |
| }
 | |
| 
 | |
| void Vector2::normalize() {
 | |
| 	real_t l = x * x + y * y;
 | |
| 	if (l != 0) {
 | |
| 		l = Math::sqrt(l);
 | |
| 		x /= l;
 | |
| 		y /= l;
 | |
| 	}
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::normalized() const {
 | |
| 	Vector2 v = *this;
 | |
| 	v.normalize();
 | |
| 	return v;
 | |
| }
 | |
| 
 | |
| bool Vector2::is_normalized() const {
 | |
| 	// use length_squared() instead of length() to avoid sqrt(), makes it more stringent.
 | |
| 	return Math::is_equal_approx(length_squared(), 1, (real_t)UNIT_EPSILON);
 | |
| }
 | |
| 
 | |
| real_t Vector2::distance_to(const Vector2 &p_vector2) const {
 | |
| 	return Math::sqrt((x - p_vector2.x) * (x - p_vector2.x) + (y - p_vector2.y) * (y - p_vector2.y));
 | |
| }
 | |
| 
 | |
| real_t Vector2::distance_squared_to(const Vector2 &p_vector2) const {
 | |
| 	return (x - p_vector2.x) * (x - p_vector2.x) + (y - p_vector2.y) * (y - p_vector2.y);
 | |
| }
 | |
| 
 | |
| real_t Vector2::angle_to(const Vector2 &p_vector2) const {
 | |
| 	return Math::atan2(cross(p_vector2), dot(p_vector2));
 | |
| }
 | |
| 
 | |
| real_t Vector2::angle_to_point(const Vector2 &p_vector2) const {
 | |
| 	return (p_vector2 - *this).angle();
 | |
| }
 | |
| 
 | |
| real_t Vector2::dot(const Vector2 &p_other) const {
 | |
| 	return x * p_other.x + y * p_other.y;
 | |
| }
 | |
| 
 | |
| real_t Vector2::cross(const Vector2 &p_other) const {
 | |
| 	return x * p_other.y - y * p_other.x;
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::sign() const {
 | |
| 	return Vector2(SIGN(x), SIGN(y));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::floor() const {
 | |
| 	return Vector2(Math::floor(x), Math::floor(y));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::ceil() const {
 | |
| 	return Vector2(Math::ceil(x), Math::ceil(y));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::round() const {
 | |
| 	return Vector2(Math::round(x), Math::round(y));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::rotated(real_t p_by) const {
 | |
| 	real_t sine = Math::sin(p_by);
 | |
| 	real_t cosi = Math::cos(p_by);
 | |
| 	return Vector2(
 | |
| 			x * cosi - y * sine,
 | |
| 			x * sine + y * cosi);
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::posmod(real_t p_mod) const {
 | |
| 	return Vector2(Math::fposmod(x, p_mod), Math::fposmod(y, p_mod));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::posmodv(const Vector2 &p_modv) const {
 | |
| 	return Vector2(Math::fposmod(x, p_modv.x), Math::fposmod(y, p_modv.y));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::project(const Vector2 &p_to) const {
 | |
| 	return p_to * (dot(p_to) / p_to.length_squared());
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::clamp(const Vector2 &p_min, const Vector2 &p_max) const {
 | |
| 	return Vector2(
 | |
| 			CLAMP(x, p_min.x, p_max.x),
 | |
| 			CLAMP(y, p_min.y, p_max.y));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::clampf(real_t p_min, real_t p_max) const {
 | |
| 	return Vector2(
 | |
| 			CLAMP(x, p_min, p_max),
 | |
| 			CLAMP(y, p_min, p_max));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::snapped(const Vector2 &p_step) const {
 | |
| 	return Vector2(
 | |
| 			Math::snapped(x, p_step.x),
 | |
| 			Math::snapped(y, p_step.y));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::snappedf(real_t p_step) const {
 | |
| 	return Vector2(
 | |
| 			Math::snapped(x, p_step),
 | |
| 			Math::snapped(y, p_step));
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::limit_length(real_t p_len) const {
 | |
| 	const real_t l = length();
 | |
| 	Vector2 v = *this;
 | |
| 	if (l > 0 && p_len < l) {
 | |
| 		v /= l;
 | |
| 		v *= p_len;
 | |
| 	}
 | |
| 
 | |
| 	return v;
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::move_toward(const Vector2 &p_to, real_t p_delta) const {
 | |
| 	Vector2 v = *this;
 | |
| 	Vector2 vd = p_to - v;
 | |
| 	real_t len = vd.length();
 | |
| 	return len <= p_delta || len < (real_t)CMP_EPSILON ? p_to : v + vd / len * p_delta;
 | |
| }
 | |
| 
 | |
| // slide returns the component of the vector along the given plane, specified by its normal vector.
 | |
| Vector2 Vector2::slide(const Vector2 &p_normal) const {
 | |
| #ifdef MATH_CHECKS
 | |
| 	ERR_FAIL_COND_V_MSG(!p_normal.is_normalized(), Vector2(), "The normal Vector2 " + p_normal.operator String() + "must be normalized.");
 | |
| #endif
 | |
| 	return *this - p_normal * dot(p_normal);
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::bounce(const Vector2 &p_normal) const {
 | |
| 	return -reflect(p_normal);
 | |
| }
 | |
| 
 | |
| Vector2 Vector2::reflect(const Vector2 &p_normal) const {
 | |
| #ifdef MATH_CHECKS
 | |
| 	ERR_FAIL_COND_V_MSG(!p_normal.is_normalized(), Vector2(), "The normal Vector2 " + p_normal.operator String() + "must be normalized.");
 | |
| #endif
 | |
| 	return 2.0f * p_normal * dot(p_normal) - *this;
 | |
| }
 | |
| 
 | |
| bool Vector2::is_equal_approx(const Vector2 &p_v) const {
 | |
| 	return Math::is_equal_approx(x, p_v.x) && Math::is_equal_approx(y, p_v.y);
 | |
| }
 | |
| 
 | |
| bool Vector2::is_zero_approx() const {
 | |
| 	return Math::is_zero_approx(x) && Math::is_zero_approx(y);
 | |
| }
 | |
| 
 | |
| bool Vector2::is_finite() const {
 | |
| 	return Math::is_finite(x) && Math::is_finite(y);
 | |
| }
 | |
| 
 | |
| Vector2::operator String() const {
 | |
| 	return "(" + String::num_real(x, false) + ", " + String::num_real(y, false) + ")";
 | |
| }
 | |
| 
 | |
| Vector2::operator Vector2i() const {
 | |
| 	return Vector2i(x, y);
 | |
| }
 | 
