Synced defs.hpp with godot's typedefs.h

(cherry picked from commit d634a5f8c4)
This commit is contained in:
Tom
2025-04-03 00:45:56 +01:00
committed by David Snopek
parent bdeb1987f6
commit b139e2fb76
10 changed files with 362 additions and 211 deletions

View File

@@ -31,6 +31,7 @@
#pragma once
#include <godot_cpp/core/defs.hpp>
#include <godot_cpp/core/math_defs.hpp>
#include <gdextension_interface.h>
@@ -38,185 +39,8 @@
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.
@@ -834,3 +658,5 @@ inline float snap_scalar_separation(float p_offset, float p_step, float p_target
} // namespace Math
} // namespace godot
#include "math.compat.inc"