godot-cpp/include/godot_cpp/core/math.hpp

820 lines
23 KiB
C++
Raw Normal View History

/**************************************************************************/
/* math.hpp */
/**************************************************************************/
/* 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. */
/**************************************************************************/
2021-09-08 18:11:12 +00:00
#ifndef GODOT_MATH_HPP
#define GODOT_MATH_HPP
#include <godot_cpp/core/defs.hpp>
#include <gdextension_interface.h>
#include <cmath>
namespace godot {
#define Math_SQRT12 0.7071067811865475244008443621048490
#define Math_SQRT2 1.4142135623730950488016887242
#define Math_LN2 0.6931471805599453094172321215
#define Math_PI 3.1415926535897932384626433833
#define Math_TAU 6.2831853071795864769252867666
#define Math_E 2.7182818284590452353602874714
#define Math_INF INFINITY
#define Math_NAN NAN
// Make room for our constexpr's below by overriding potential system-specific macros.
#undef ABS
#undef SIGN
#undef MIN
#undef MAX
#undef CLAMP
// Generic ABS function, for math uses please use Math::abs.
template <typename T>
constexpr T ABS(T m_v) {
return m_v < 0 ? -m_v : m_v;
}
template <typename T>
constexpr const T SIGN(const T m_v) {
return m_v == 0 ? 0.0f : (m_v < 0 ? -1.0f : +1.0f);
}
template <typename T, typename T2>
constexpr auto MIN(const T m_a, const T2 m_b) {
return m_a < m_b ? m_a : m_b;
}
template <typename T, typename T2>
constexpr auto MAX(const T m_a, const T2 m_b) {
return m_a > m_b ? m_a : m_b;
}
template <typename T, typename T2, typename T3>
constexpr auto CLAMP(const T m_a, const T2 m_min, const T3 m_max) {
return m_a < m_min ? m_min : (m_a > m_max ? m_max : m_a);
}
// Generic swap template.
#ifndef SWAP
#define SWAP(m_x, m_y) __swap_tmpl((m_x), (m_y))
template <typename T>
inline void __swap_tmpl(T &x, T &y) {
T aux = x;
x = y;
y = aux;
}
#endif // SWAP
/* Functions to handle powers of 2 and shifting. */
// Function to find the next power of 2 to an integer.
static _FORCE_INLINE_ unsigned int next_power_of_2(unsigned int x) {
if (x == 0) {
return 0;
}
--x;
x |= x >> 1;
x |= x >> 2;
x |= x >> 4;
x |= x >> 8;
x |= x >> 16;
return ++x;
}
// Function to find the previous power of 2 to an integer.
static _FORCE_INLINE_ unsigned int previous_power_of_2(unsigned int x) {
x |= x >> 1;
x |= x >> 2;
x |= x >> 4;
x |= x >> 8;
x |= x >> 16;
return x - (x >> 1);
}
// Function to find the closest power of 2 to an integer.
static _FORCE_INLINE_ unsigned int closest_power_of_2(unsigned int x) {
unsigned int nx = next_power_of_2(x);
unsigned int px = previous_power_of_2(x);
return (nx - x) > (x - px) ? px : nx;
}
// Get a shift value from a power of 2.
static inline int get_shift_from_power_of_2(unsigned int p_bits) {
for (unsigned int i = 0; i < 32; i++) {
if (p_bits == (unsigned int)(1 << i)) {
return i;
}
}
return -1;
}
template <typename T>
static _FORCE_INLINE_ T nearest_power_of_2_templated(T x) {
--x;
// The number of operations on x is the base two logarithm
// of the number of bits in the type. Add three to account
// for sizeof(T) being in bytes.
size_t num = get_shift_from_power_of_2(sizeof(T)) + 3;
// If the compiler is smart, it unrolls this loop.
// If it's dumb, this is a bit slow.
for (size_t i = 0; i < num; i++) {
x |= x >> (1 << i);
}
return ++x;
}
// Function to find the nearest (bigger) power of 2 to an integer.
static inline unsigned int nearest_shift(unsigned int p_number) {
for (int i = 30; i >= 0; i--) {
if (p_number & (1 << i)) {
return i + 1;
}
}
return 0;
}
// constexpr function to find the floored log2 of a number
template <typename T>
constexpr T floor_log2(T x) {
return x < 2 ? x : 1 + floor_log2(x >> 1);
}
// Get the number of bits needed to represent the number.
// IE, if you pass in 8, you will get 4.
// If you want to know how many bits are needed to store 8 values however, pass in (8 - 1).
template <typename T>
constexpr T get_num_bits(T x) {
return floor_log2(x);
}
// Swap 16, 32 and 64 bits value for endianness.
#if defined(__GNUC__)
#define BSWAP16(x) __builtin_bswap16(x)
#define BSWAP32(x) __builtin_bswap32(x)
#define BSWAP64(x) __builtin_bswap64(x)
#else
static inline uint16_t BSWAP16(uint16_t x) {
return (x >> 8) | (x << 8);
}
static inline uint32_t BSWAP32(uint32_t x) {
return ((x << 24) | ((x << 8) & 0x00FF0000) | ((x >> 8) & 0x0000FF00) | (x >> 24));
}
static inline uint64_t BSWAP64(uint64_t x) {
x = (x & 0x00000000FFFFFFFF) << 32 | (x & 0xFFFFFFFF00000000) >> 32;
x = (x & 0x0000FFFF0000FFFF) << 16 | (x & 0xFFFF0000FFFF0000) >> 16;
x = (x & 0x00FF00FF00FF00FF) << 8 | (x & 0xFF00FF00FF00FF00) >> 8;
return x;
}
#endif
namespace Math {
// This epsilon should match the one used by Godot for consistency.
// Using `f` when `real_t` is float.
#define CMP_EPSILON 0.00001f
#define CMP_EPSILON2 (CMP_EPSILON * CMP_EPSILON)
// This epsilon is for values related to a unit size (scalar or vector len).
#ifdef PRECISE_MATH_CHECKS
#define UNIT_EPSILON 0.00001
#else
// Tolerate some more floating point error normally.
#define UNIT_EPSILON 0.001
#endif
// Functions reproduced as in Godot's source code `math_funcs.h`.
// Some are overloads to automatically support changing real_t into either double or float in the way Godot does.
inline double fmod(double p_x, double p_y) {
return ::fmod(p_x, p_y);
}
inline float fmod(float p_x, float p_y) {
return ::fmodf(p_x, p_y);
}
inline double fposmod(double p_x, double p_y) {
double value = Math::fmod(p_x, p_y);
if ((value < 0 && p_y > 0) || (value > 0 && p_y < 0)) {
value += p_y;
}
value += 0.0;
return value;
}
inline float fposmod(float p_x, float p_y) {
float value = Math::fmod(p_x, p_y);
if ((value < 0 && p_y > 0) || (value > 0 && p_y < 0)) {
value += p_y;
}
value += 0.0f;
return value;
}
inline float fposmodp(float p_x, float p_y) {
float value = Math::fmod(p_x, p_y);
if (value < 0) {
value += p_y;
}
value += 0.0f;
return value;
}
inline double fposmodp(double p_x, double p_y) {
double value = Math::fmod(p_x, p_y);
if (value < 0) {
value += p_y;
}
value += 0.0;
return value;
}
inline int64_t posmod(int64_t p_x, int64_t p_y) {
int64_t value = p_x % p_y;
if ((value < 0 && p_y > 0) || (value > 0 && p_y < 0)) {
value += p_y;
}
return value;
}
inline double floor(double p_x) {
return ::floor(p_x);
}
inline float floor(float p_x) {
return ::floorf(p_x);
}
inline double ceil(double p_x) {
return ::ceil(p_x);
}
inline float ceil(float p_x) {
return ::ceilf(p_x);
}
inline double exp(double p_x) {
return ::exp(p_x);
}
inline float exp(float p_x) {
return ::expf(p_x);
}
inline double sin(double p_x) {
return ::sin(p_x);
}
inline float sin(float p_x) {
return ::sinf(p_x);
}
inline double cos(double p_x) {
return ::cos(p_x);
}
inline float cos(float p_x) {
return ::cosf(p_x);
}
inline double tan(double p_x) {
return ::tan(p_x);
}
inline float tan(float p_x) {
return ::tanf(p_x);
}
inline double sinh(double p_x) {
return ::sinh(p_x);
}
inline float sinh(float p_x) {
return ::sinhf(p_x);
}
inline float sinc(float p_x) {
return p_x == 0 ? 1 : ::sin(p_x) / p_x;
}
inline double sinc(double p_x) {
return p_x == 0 ? 1 : ::sin(p_x) / p_x;
}
inline float sincn(float p_x) {
return (float)sinc(Math_PI * p_x);
}
inline double sincn(double p_x) {
return sinc(Math_PI * p_x);
}
inline double cosh(double p_x) {
return ::cosh(p_x);
}
inline float cosh(float p_x) {
return ::coshf(p_x);
}
inline double tanh(double p_x) {
return ::tanh(p_x);
}
inline float tanh(float p_x) {
return ::tanhf(p_x);
}
inline double asin(double p_x) {
return ::asin(p_x);
}
inline float asin(float p_x) {
return ::asinf(p_x);
}
inline double acos(double p_x) {
return ::acos(p_x);
}
inline float acos(float p_x) {
return ::acosf(p_x);
}
inline double atan(double p_x) {
return ::atan(p_x);
}
inline float atan(float p_x) {
return ::atanf(p_x);
}
inline double atan2(double p_y, double p_x) {
return ::atan2(p_y, p_x);
}
inline float atan2(float p_y, float p_x) {
return ::atan2f(p_y, p_x);
}
inline double sqrt(double p_x) {
return ::sqrt(p_x);
}
inline float sqrt(float p_x) {
return ::sqrtf(p_x);
}
inline double pow(double p_x, double p_y) {
return ::pow(p_x, p_y);
}
inline float pow(float p_x, float p_y) {
return ::powf(p_x, p_y);
}
inline double log(double p_x) {
return ::log(p_x);
}
inline float log(float p_x) {
return ::logf(p_x);
}
inline float lerp(float minv, float maxv, float t) {
return minv + t * (maxv - minv);
}
inline double lerp(double minv, double maxv, double t) {
return minv + t * (maxv - minv);
}
inline double lerp_angle(double p_from, double p_to, double p_weight) {
double difference = fmod(p_to - p_from, Math_TAU);
double distance = fmod(2.0 * difference, Math_TAU) - difference;
return p_from + distance * p_weight;
}
inline float lerp_angle(float p_from, float p_to, float p_weight) {
float difference = fmod(p_to - p_from, (float)Math_TAU);
float distance = fmod(2.0f * difference, (float)Math_TAU) - difference;
return p_from + distance * p_weight;
}
inline double cubic_interpolate(double p_from, double p_to, double p_pre, double p_post, double p_weight) {
return 0.5 *
((p_from * 2.0) +
(-p_pre + p_to) * p_weight +
(2.0 * p_pre - 5.0 * p_from + 4.0 * p_to - p_post) * (p_weight * p_weight) +
(-p_pre + 3.0 * p_from - 3.0 * p_to + p_post) * (p_weight * p_weight * p_weight));
}
inline float cubic_interpolate(float p_from, float p_to, float p_pre, float p_post, float p_weight) {
return 0.5f *
((p_from * 2.0f) +
(-p_pre + p_to) * p_weight +
(2.0f * p_pre - 5.0f * p_from + 4.0f * p_to - p_post) * (p_weight * p_weight) +
(-p_pre + 3.0f * p_from - 3.0f * p_to + p_post) * (p_weight * p_weight * p_weight));
}
inline double cubic_interpolate_angle(double p_from, double p_to, double p_pre, double p_post, double p_weight) {
double from_rot = fmod(p_from, Math_TAU);
double pre_diff = fmod(p_pre - from_rot, Math_TAU);
double pre_rot = from_rot + fmod(2.0 * pre_diff, Math_TAU) - pre_diff;
double to_diff = fmod(p_to - from_rot, Math_TAU);
double to_rot = from_rot + fmod(2.0 * to_diff, Math_TAU) - to_diff;
double post_diff = fmod(p_post - to_rot, Math_TAU);
double post_rot = to_rot + fmod(2.0 * post_diff, Math_TAU) - post_diff;
return cubic_interpolate(from_rot, to_rot, pre_rot, post_rot, p_weight);
}
inline float cubic_interpolate_angle(float p_from, float p_to, float p_pre, float p_post, float p_weight) {
float from_rot = fmod(p_from, (float)Math_TAU);
float pre_diff = fmod(p_pre - from_rot, (float)Math_TAU);
float pre_rot = from_rot + fmod(2.0f * pre_diff, (float)Math_TAU) - pre_diff;
float to_diff = fmod(p_to - from_rot, (float)Math_TAU);
float to_rot = from_rot + fmod(2.0f * to_diff, (float)Math_TAU) - to_diff;
float post_diff = fmod(p_post - to_rot, (float)Math_TAU);
float post_rot = to_rot + fmod(2.0f * post_diff, (float)Math_TAU) - post_diff;
return cubic_interpolate(from_rot, to_rot, pre_rot, post_rot, p_weight);
}
inline double cubic_interpolate_in_time(double p_from, double p_to, double p_pre, double p_post, double p_weight,
double p_to_t, double p_pre_t, double p_post_t) {
/* Barry-Goldman method */
double t = Math::lerp(0.0, p_to_t, p_weight);
double a1 = Math::lerp(p_pre, p_from, p_pre_t == 0 ? 0.0 : (t - p_pre_t) / -p_pre_t);
double a2 = Math::lerp(p_from, p_to, p_to_t == 0 ? 0.5 : t / p_to_t);
double a3 = Math::lerp(p_to, p_post, p_post_t - p_to_t == 0 ? 1.0 : (t - p_to_t) / (p_post_t - p_to_t));
double b1 = Math::lerp(a1, a2, p_to_t - p_pre_t == 0 ? 0.0 : (t - p_pre_t) / (p_to_t - p_pre_t));
double b2 = Math::lerp(a2, a3, p_post_t == 0 ? 1.0 : t / p_post_t);
return Math::lerp(b1, b2, p_to_t == 0 ? 0.5 : t / p_to_t);
}
inline float cubic_interpolate_in_time(float p_from, float p_to, float p_pre, float p_post, float p_weight,
float p_to_t, float p_pre_t, float p_post_t) {
/* Barry-Goldman method */
float t = Math::lerp(0.0f, p_to_t, p_weight);
float a1 = Math::lerp(p_pre, p_from, p_pre_t == 0 ? 0.0f : (t - p_pre_t) / -p_pre_t);
float a2 = Math::lerp(p_from, p_to, p_to_t == 0 ? 0.5f : t / p_to_t);
float a3 = Math::lerp(p_to, p_post, p_post_t - p_to_t == 0 ? 1.0f : (t - p_to_t) / (p_post_t - p_to_t));
float b1 = Math::lerp(a1, a2, p_to_t - p_pre_t == 0 ? 0.0f : (t - p_pre_t) / (p_to_t - p_pre_t));
float b2 = Math::lerp(a2, a3, p_post_t == 0 ? 1.0f : t / p_post_t);
return Math::lerp(b1, b2, p_to_t == 0 ? 0.5f : t / p_to_t);
}
inline double cubic_interpolate_angle_in_time(double p_from, double p_to, double p_pre, double p_post, double p_weight,
double p_to_t, double p_pre_t, double p_post_t) {
double from_rot = fmod(p_from, Math_TAU);
double pre_diff = fmod(p_pre - from_rot, Math_TAU);
double pre_rot = from_rot + fmod(2.0 * pre_diff, Math_TAU) - pre_diff;
double to_diff = fmod(p_to - from_rot, Math_TAU);
double to_rot = from_rot + fmod(2.0 * to_diff, Math_TAU) - to_diff;
double post_diff = fmod(p_post - to_rot, Math_TAU);
double post_rot = to_rot + fmod(2.0 * post_diff, Math_TAU) - post_diff;
return cubic_interpolate_in_time(from_rot, to_rot, pre_rot, post_rot, p_weight, p_to_t, p_pre_t, p_post_t);
}
inline float cubic_interpolate_angle_in_time(float p_from, float p_to, float p_pre, float p_post, float p_weight,
float p_to_t, float p_pre_t, float p_post_t) {
float from_rot = fmod(p_from, (float)Math_TAU);
float pre_diff = fmod(p_pre - from_rot, (float)Math_TAU);
float pre_rot = from_rot + fmod(2.0f * pre_diff, (float)Math_TAU) - pre_diff;
float to_diff = fmod(p_to - from_rot, (float)Math_TAU);
float to_rot = from_rot + fmod(2.0f * to_diff, (float)Math_TAU) - to_diff;
float post_diff = fmod(p_post - to_rot, (float)Math_TAU);
float post_rot = to_rot + fmod(2.0f * post_diff, (float)Math_TAU) - post_diff;
return cubic_interpolate_in_time(from_rot, to_rot, pre_rot, post_rot, p_weight, p_to_t, p_pre_t, p_post_t);
}
inline double bezier_interpolate(double p_start, double p_control_1, double p_control_2, double p_end, double p_t) {
/* Formula from Wikipedia article on Bezier curves. */
double omt = (1.0 - p_t);
double omt2 = omt * omt;
double omt3 = omt2 * omt;
double t2 = p_t * p_t;
double t3 = t2 * p_t;
return p_start * omt3 + p_control_1 * omt2 * p_t * 3.0 + p_control_2 * omt * t2 * 3.0 + p_end * t3;
}
inline float bezier_interpolate(float p_start, float p_control_1, float p_control_2, float p_end, float p_t) {
/* Formula from Wikipedia article on Bezier curves. */
float omt = (1.0f - p_t);
float omt2 = omt * omt;
float omt3 = omt2 * omt;
float t2 = p_t * p_t;
float t3 = t2 * p_t;
return p_start * omt3 + p_control_1 * omt2 * p_t * 3.0f + p_control_2 * omt * t2 * 3.0f + p_end * t3;
}
template <typename T>
inline T clamp(T x, T minv, T maxv) {
if (x < minv) {
return minv;
}
if (x > maxv) {
return maxv;
}
return x;
}
template <typename T>
inline T min(T a, T b) {
return a < b ? a : b;
}
template <typename T>
inline T max(T a, T b) {
return a > b ? a : b;
}
template <typename T>
inline T sign(T x) {
return static_cast<T>(SIGN(x));
}
template <typename T>
inline T abs(T x) {
return std::abs(x);
}
inline double deg_to_rad(double p_y) {
return p_y * Math_PI / 180.0;
}
inline float deg_to_rad(float p_y) {
return p_y * static_cast<float>(Math_PI) / 180.f;
}
inline double rad_to_deg(double p_y) {
return p_y * 180.0 / Math_PI;
}
inline float rad_to_deg(float p_y) {
return p_y * 180.f / static_cast<float>(Math_PI);
}
inline double inverse_lerp(double p_from, double p_to, double p_value) {
return (p_value - p_from) / (p_to - p_from);
}
inline float inverse_lerp(float p_from, float p_to, float p_value) {
return (p_value - p_from) / (p_to - p_from);
}
inline double remap(double p_value, double p_istart, double p_istop, double p_ostart, double p_ostop) {
return Math::lerp(p_ostart, p_ostop, Math::inverse_lerp(p_istart, p_istop, p_value));
}
inline float remap(float p_value, float p_istart, float p_istop, float p_ostart, float p_ostop) {
return Math::lerp(p_ostart, p_ostop, Math::inverse_lerp(p_istart, p_istop, p_value));
}
inline bool is_nan(float p_val) {
return std::isnan(p_val);
}
inline bool is_nan(double p_val) {
return std::isnan(p_val);
}
inline bool is_inf(float p_val) {
return std::isinf(p_val);
}
inline bool is_inf(double p_val) {
return std::isinf(p_val);
}
2024-04-10 19:18:46 +00:00
inline bool is_finite(float p_val) {
return std::isfinite(p_val);
}
inline bool is_finite(double p_val) {
return std::isfinite(p_val);
}
inline bool is_equal_approx(float a, float b) {
// Check for exact equality first, required to handle "infinity" values.
if (a == b) {
return true;
}
// Then check for approximate equality.
float tolerance = (float)CMP_EPSILON * abs(a);
if (tolerance < (float)CMP_EPSILON) {
tolerance = (float)CMP_EPSILON;
}
return abs(a - b) < tolerance;
}
inline bool is_equal_approx(float a, float b, float tolerance) {
// Check for exact equality first, required to handle "infinity" values.
if (a == b) {
return true;
}
// Then check for approximate equality.
return abs(a - b) < tolerance;
}
inline bool is_zero_approx(float s) {
return abs(s) < (float)CMP_EPSILON;
}
inline bool is_equal_approx(double a, double b) {
// Check for exact equality first, required to handle "infinity" values.
if (a == b) {
return true;
}
// Then check for approximate equality.
double tolerance = CMP_EPSILON * abs(a);
if (tolerance < CMP_EPSILON) {
tolerance = CMP_EPSILON;
}
return abs(a - b) < tolerance;
}
inline bool is_equal_approx(double a, double b, double tolerance) {
// Check for exact equality first, required to handle "infinity" values.
if (a == b) {
return true;
}
// Then check for approximate equality.
return abs(a - b) < tolerance;
}
inline bool is_zero_approx(double s) {
return abs(s) < CMP_EPSILON;
}
inline float absf(float g) {
union {
float f;
uint32_t i;
} u;
u.f = g;
u.i &= 2147483647u;
return u.f;
}
inline double absd(double g) {
union {
double d;
uint64_t i;
} u;
u.d = g;
u.i &= (uint64_t)9223372036854775807ull;
return u.d;
}
inline double smoothstep(double p_from, double p_to, double p_weight) {
if (is_equal_approx(static_cast<real_t>(p_from), static_cast<real_t>(p_to))) {
return p_from;
}
double x = clamp((p_weight - p_from) / (p_to - p_from), 0.0, 1.0);
return x * x * (3.0 - 2.0 * x);
}
inline float smoothstep(float p_from, float p_to, float p_weight) {
if (is_equal_approx(p_from, p_to)) {
return p_from;
}
float x = clamp((p_weight - p_from) / (p_to - p_from), 0.0f, 1.0f);
return x * x * (3.0f - 2.0f * x);
}
inline double move_toward(double p_from, double p_to, double p_delta) {
return std::abs(p_to - p_from) <= p_delta ? p_to : p_from + sign(p_to - p_from) * p_delta;
}
inline float move_toward(float p_from, float p_to, float p_delta) {
return std::abs(p_to - p_from) <= p_delta ? p_to : p_from + sign(p_to - p_from) * p_delta;
}
inline double linear2db(double p_linear) {
return log(p_linear) * 8.6858896380650365530225783783321;
}
inline float linear2db(float p_linear) {
return log(p_linear) * 8.6858896380650365530225783783321f;
}
inline double db2linear(double p_db) {
return exp(p_db * 0.11512925464970228420089957273422);
}
inline float db2linear(float p_db) {
return exp(p_db * 0.11512925464970228420089957273422f);
}
inline double round(double p_val) {
return (p_val >= 0) ? floor(p_val + 0.5) : -floor(-p_val + 0.5);
}
inline float round(float p_val) {
return (p_val >= 0) ? floor(p_val + 0.5f) : -floor(-p_val + 0.5f);
}
inline int64_t wrapi(int64_t value, int64_t min, int64_t max) {
int64_t range = max - min;
return range == 0 ? min : min + ((((value - min) % range) + range) % range);
}
inline float wrapf(real_t value, real_t min, real_t max) {
const real_t range = max - min;
return is_zero_approx(range) ? min : value - (range * floor((value - min) / range));
}
inline float fract(float value) {
return value - floor(value);
}
inline double fract(double value) {
return value - floor(value);
}
inline float pingpong(float value, float length) {
return (length != 0.0f) ? abs(fract((value - length) / (length * 2.0f)) * length * 2.0f - length) : 0.0f;
}
inline double pingpong(double value, double length) {
return (length != 0.0) ? abs(fract((value - length) / (length * 2.0)) * length * 2.0 - length) : 0.0;
}
// This function should be as fast as possible and rounding mode should not matter.
inline int fast_ftoi(float a) {
static int b;
#if (defined(_WIN32_WINNT) && _WIN32_WINNT >= 0x0603) || WINAPI_FAMILY == WINAPI_FAMILY_PHONE_APP // windows 8 phone?
b = (int)((a > 0.0) ? (a + 0.5) : (a - 0.5));
#elif defined(_MSC_VER) && _MSC_VER < 1800
__asm fld a __asm fistp b
/*#elif defined( __GNUC__ ) && ( defined( __i386__ ) || defined( __x86_64__ ) )
// use AT&T inline assembly style, document that
// we use memory as output (=m) and input (m)
__asm__ __volatile__ (
"flds %1 \n\t"
"fistpl %0 \n\t"
: "=m" (b)
: "m" (a));*/
#else
b = lrintf(a); // assuming everything but msvc 2012 or earlier has lrint
#endif
return b;
}
inline double snapped(double p_value, double p_step) {
if (p_step != 0) {
p_value = Math::floor(p_value / p_step + 0.5) * p_step;
}
return p_value;
}
inline float snap_scalar(float p_offset, float p_step, float p_target) {
return p_step != 0 ? Math::snapped(p_target - p_offset, p_step) + p_offset : p_target;
}
inline float snap_scalar_separation(float p_offset, float p_step, float p_target, float p_separation) {
if (p_step != 0) {
float a = Math::snapped(p_target - p_offset, p_step + p_separation) + p_offset;
float b = a;
if (p_target >= 0) {
b -= p_separation;
} else {
b += p_step;
}
return (Math::abs(p_target - a) < Math::abs(p_target - b)) ? a : b;
}
return p_target;
}
} // namespace Math
} // namespace godot
#endif // GODOT_MATH_HPP