/* * Copyright (C) 2006, 2007, 2008 Apple Inc. All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * * THIS SOFTWARE IS PROVIDED BY APPLE COMPUTER, INC. ``AS IS'' AND ANY * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE COMPUTER, INC. OR * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #ifndef WTF_MathExtras_h #define WTF_MathExtras_h #include #include #include #include #if OS(SOLARIS) #include #endif #if OS(OPENBSD) #include #include #endif #if (OS(QNX) && defined(_CPPLIB_VER)) || COMPILER(INTEL) // FIXME: Look into a way to have cmath import its functions into both the standard and global // namespace. For now, we include math.h since the QNX cmath header only imports its functions // into the standard namespace. #include // These macros from math.h conflict with the real functions in the std namespace. #undef signbit #undef isnan #undef isinf #undef isfinite #endif #ifndef M_PI const double piDouble = 3.14159265358979323846; const float piFloat = 3.14159265358979323846f; #else const double piDouble = M_PI; const float piFloat = static_cast(M_PI); #endif #ifndef M_PI_4 const double piOverFourDouble = 0.785398163397448309616; const float piOverFourFloat = 0.785398163397448309616f; #else const double piOverFourDouble = M_PI_4; const float piOverFourFloat = static_cast(M_PI_4); #endif #if OS(DARWIN) // Work around a bug in the Mac OS X libc where ceil(-0.1) return +0. inline double wtf_ceil(double x) { return copysign(ceil(x), x); } #define ceil(x) wtf_ceil(x) #endif #if OS(SOLARIS) && __cplusplus < 201103L namespace std { #ifndef isfinite inline bool isfinite(double x) { return finite(x) && !isnand(x); } #endif #ifndef signbit inline bool signbit(double x) { return copysign(1.0, x) < 0; } #endif #ifndef isinf inline bool isinf(double x) { return !finite(x) && !isnand(x); } #endif } // namespace std #endif #if OS(OPENBSD) && __cplusplus < 201103L namespace std { #ifndef isfinite inline bool isfinite(double x) { return finite(x); } #endif #ifndef signbit inline bool signbit(double x) { struct ieee_double *p = (struct ieee_double *)&x; return p->dbl_sign; } #endif } // namespace std #endif #if (COMPILER(MSVC) && _MSC_VER < 1800) || COMPILER(RVCT) // We must not do 'num + 0.5' or 'num - 0.5' because they can cause precision loss. static double round(double num) { double integer = ceil(num); if (num > 0) return integer - num > 0.5 ? integer - 1.0 : integer; return integer - num >= 0.5 ? integer - 1.0 : integer; } static float roundf(float num) { float integer = ceilf(num); if (num > 0) return integer - num > 0.5f ? integer - 1.0f : integer; return integer - num >= 0.5f ? integer - 1.0f : integer; } inline long long llround(double num) { return static_cast(round(num)); } inline long long llroundf(float num) { return static_cast(roundf(num)); } inline long lround(double num) { return static_cast(round(num)); } inline long lroundf(float num) { return static_cast(roundf(num)); } inline double trunc(double num) { return num > 0 ? floor(num) : ceil(num); } #endif #if COMPILER(MSVC) namespace std { inline bool isinf(double num) { return !_finite(num) && !_isnan(num); } inline bool isnan(double num) { return !!_isnan(num); } inline bool isfinite(double x) { return _finite(x); } #if _MSC_VER < 1800 inline bool signbit(double num) { return _copysign(1.0, num) < 0; } #endif } // namespace std inline double nextafter(double x, double y) { return _nextafter(x, y); } inline float nextafterf(float x, float y) { return x > y ? x - FLT_EPSILON : x + FLT_EPSILON; } inline double copysign(double x, double y) { return _copysign(x, y); } // Work around a bug in Win, where atan2(+-infinity, +-infinity) yields NaN instead of specific values. inline double wtf_atan2(double x, double y) { double posInf = std::numeric_limits::infinity(); double negInf = -std::numeric_limits::infinity(); double nan = std::numeric_limits::quiet_NaN(); double result = nan; if (x == posInf && y == posInf) result = piOverFourDouble; else if (x == posInf && y == negInf) result = 3 * piOverFourDouble; else if (x == negInf && y == posInf) result = -piOverFourDouble; else if (x == negInf && y == negInf) result = -3 * piOverFourDouble; else result = ::atan2(x, y); return result; } // Work around a bug in the Microsoft CRT, where fmod(x, +-infinity) yields NaN instead of x. extern "C" inline double wtf_fmod(double x, double y) { return (!std::isinf(x) && std::isinf(y)) ? x : fmod(x, y); } // Work around a bug in the Microsoft CRT, where pow(NaN, 0) yields NaN instead of 1. extern "C" inline double wtf_pow(double x, double y) { return y == 0 ? 1 : pow(x, y); } #define atan2(x, y) wtf_atan2(x, y) #define fmod(x, y) wtf_fmod(x, y) #define pow(x, y) wtf_pow(x, y) #endif // COMPILER(MSVC) inline double deg2rad(double d) { return d * piDouble / 180.0; } inline double rad2deg(double r) { return r * 180.0 / piDouble; } inline double deg2grad(double d) { return d * 400.0 / 360.0; } inline double grad2deg(double g) { return g * 360.0 / 400.0; } inline double turn2deg(double t) { return t * 360.0; } inline double deg2turn(double d) { return d / 360.0; } inline double rad2grad(double r) { return r * 200.0 / piDouble; } inline double grad2rad(double g) { return g * piDouble / 200.0; } inline float deg2rad(float d) { return d * piFloat / 180.0f; } inline float rad2deg(float r) { return r * 180.0f / piFloat; } inline float deg2grad(float d) { return d * 400.0f / 360.0f; } inline float grad2deg(float g) { return g * 360.0f / 400.0f; } inline float turn2deg(float t) { return t * 360.0f; } inline float deg2turn(float d) { return d / 360.0f; } inline float rad2grad(float r) { return r * 200.0f / piFloat; } inline float grad2rad(float g) { return g * piFloat / 200.0f; } #if COMPILER(MINGW64) && (!defined(__MINGW64_VERSION_RC) || __MINGW64_VERSION_RC < 1) inline double wtf_pow(double x, double y) { // MinGW-w64 has a custom implementation for pow. // This handles certain special cases that are different. if ((x == 0.0 || std::isinf(x)) && std::isfinite(y)) { double f; if (modf(y, &f) != 0.0) return ((x == 0.0) ^ (y > 0.0)) ? std::numeric_limits::infinity() : 0.0; } if (x == 2.0) { int yInt = static_cast(y); if (y == yInt) return ldexp(1.0, yInt); } return pow(x, y); } #define pow(x, y) wtf_pow(x, y) #endif // COMPILER(MINGW64) && (!defined(__MINGW64_VERSION_RC) || __MINGW64_VERSION_RC < 1) #endif // #ifndef WTF_MathExtras_h