diff options
Diffstat (limited to 'src/mesh/assimp-master/include/assimp/matrix3x3.inl')
-rw-r--r-- | src/mesh/assimp-master/include/assimp/matrix3x3.inl | 359 |
1 files changed, 359 insertions, 0 deletions
diff --git a/src/mesh/assimp-master/include/assimp/matrix3x3.inl b/src/mesh/assimp-master/include/assimp/matrix3x3.inl new file mode 100644 index 0000000..99d9197 --- /dev/null +++ b/src/mesh/assimp-master/include/assimp/matrix3x3.inl @@ -0,0 +1,359 @@ +/* +--------------------------------------------------------------------------- +Open Asset Import Library (assimp) +--------------------------------------------------------------------------- + +Copyright (c) 2006-2022, assimp team + +All rights reserved. + +Redistribution and use of this software in source and binary forms, +with or without modification, are permitted provided that the following +conditions are met: + +* Redistributions of source code must retain the above + copyright notice, this list of conditions and the + following disclaimer. + +* 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. + +* Neither the name of the assimp team, nor the names of its + contributors may be used to endorse or promote products + derived from this software without specific prior + written permission of the assimp team. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"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 THE COPYRIGHT +OWNER 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. +--------------------------------------------------------------------------- +*/ + +/** @file matrix3x3.inl + * @brief Inline implementation of the 3x3 matrix operators + */ +#pragma once +#ifndef AI_MATRIX3X3_INL_INC +#define AI_MATRIX3X3_INL_INC + +#ifdef __GNUC__ +# pragma GCC system_header +#endif + +#ifdef __cplusplus +#include <assimp/matrix3x3.h> +#include <assimp/matrix4x4.h> + +#include <algorithm> +#include <cmath> +#include <limits> + +// ------------------------------------------------------------------------------------------------ +// Construction from a 4x4 matrix. The remaining parts of the matrix are ignored. +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal>::aiMatrix3x3t( const aiMatrix4x4t<TReal>& pMatrix) { + a1 = pMatrix.a1; a2 = pMatrix.a2; a3 = pMatrix.a3; + b1 = pMatrix.b1; b2 = pMatrix.b2; b3 = pMatrix.b3; + c1 = pMatrix.c1; c2 = pMatrix.c2; c3 = pMatrix.c3; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal>& aiMatrix3x3t<TReal>::operator *= (const aiMatrix3x3t<TReal>& m) { + *this = aiMatrix3x3t<TReal>(m.a1 * a1 + m.b1 * a2 + m.c1 * a3, + m.a2 * a1 + m.b2 * a2 + m.c2 * a3, + m.a3 * a1 + m.b3 * a2 + m.c3 * a3, + m.a1 * b1 + m.b1 * b2 + m.c1 * b3, + m.a2 * b1 + m.b2 * b2 + m.c2 * b3, + m.a3 * b1 + m.b3 * b2 + m.c3 * b3, + m.a1 * c1 + m.b1 * c2 + m.c1 * c3, + m.a2 * c1 + m.b2 * c2 + m.c2 * c3, + m.a3 * c1 + m.b3 * c2 + m.c3 * c3); + return *this; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +template <typename TOther> +aiMatrix3x3t<TReal>::operator aiMatrix3x3t<TOther> () const { + return aiMatrix3x3t<TOther>(static_cast<TOther>(a1),static_cast<TOther>(a2),static_cast<TOther>(a3), + static_cast<TOther>(b1),static_cast<TOther>(b2),static_cast<TOther>(b3), + static_cast<TOther>(c1),static_cast<TOther>(c2),static_cast<TOther>(c3)); +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal> aiMatrix3x3t<TReal>::operator* (const aiMatrix3x3t<TReal>& m) const { + aiMatrix3x3t<TReal> temp( *this); + temp *= m; + return temp; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +TReal* aiMatrix3x3t<TReal>::operator[] (unsigned int p_iIndex) { + switch ( p_iIndex ) { + case 0: + return &a1; + case 1: + return &b1; + case 2: + return &c1; + default: + break; + } + return &a1; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +const TReal* aiMatrix3x3t<TReal>::operator[] (unsigned int p_iIndex) const { + switch ( p_iIndex ) { + case 0: + return &a1; + case 1: + return &b1; + case 2: + return &c1; + default: + break; + } + return &a1; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +bool aiMatrix3x3t<TReal>::operator== (const aiMatrix3x3t<TReal>& m) const { + return a1 == m.a1 && a2 == m.a2 && a3 == m.a3 && + b1 == m.b1 && b2 == m.b2 && b3 == m.b3 && + c1 == m.c1 && c2 == m.c2 && c3 == m.c3; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +bool aiMatrix3x3t<TReal>::operator!= (const aiMatrix3x3t<TReal>& m) const { + return !(*this == m); +} + +// --------------------------------------------------------------------------- +template<typename TReal> +AI_FORCE_INLINE +bool aiMatrix3x3t<TReal>::Equal(const aiMatrix3x3t<TReal>& m, TReal epsilon) const { + return + std::abs(a1 - m.a1) <= epsilon && + std::abs(a2 - m.a2) <= epsilon && + std::abs(a3 - m.a3) <= epsilon && + std::abs(b1 - m.b1) <= epsilon && + std::abs(b2 - m.b2) <= epsilon && + std::abs(b3 - m.b3) <= epsilon && + std::abs(c1 - m.c1) <= epsilon && + std::abs(c2 - m.c2) <= epsilon && + std::abs(c3 - m.c3) <= epsilon; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal>& aiMatrix3x3t<TReal>::Transpose() { + // (TReal&) don't remove, GCC complains cause of packed fields + std::swap( (TReal&)a2, (TReal&)b1); + std::swap( (TReal&)a3, (TReal&)c1); + std::swap( (TReal&)b3, (TReal&)c2); + return *this; +} + +// ---------------------------------------------------------------------------------------- +template <typename TReal> +AI_FORCE_INLINE +TReal aiMatrix3x3t<TReal>::Determinant() const { + return a1*b2*c3 - a1*b3*c2 + a2*b3*c1 - a2*b1*c3 + a3*b1*c2 - a3*b2*c1; +} + +// ---------------------------------------------------------------------------------------- +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal>& aiMatrix3x3t<TReal>::Inverse() { + // Compute the reciprocal determinant + TReal det = Determinant(); + if(det == static_cast<TReal>(0.0)) + { + // Matrix not invertible. Setting all elements to nan is not really + // correct in a mathematical sense; but at least qnans are easy to + // spot. XXX we might throw an exception instead, which would + // be even much better to spot :/. + const TReal nan = std::numeric_limits<TReal>::quiet_NaN(); + *this = aiMatrix3x3t<TReal>( nan,nan,nan,nan,nan,nan,nan,nan,nan); + + return *this; + } + + TReal invdet = static_cast<TReal>(1.0) / det; + + aiMatrix3x3t<TReal> res; + res.a1 = invdet * (b2 * c3 - b3 * c2); + res.a2 = -invdet * (a2 * c3 - a3 * c2); + res.a3 = invdet * (a2 * b3 - a3 * b2); + res.b1 = -invdet * (b1 * c3 - b3 * c1); + res.b2 = invdet * (a1 * c3 - a3 * c1); + res.b3 = -invdet * (a1 * b3 - a3 * b1); + res.c1 = invdet * (b1 * c2 - b2 * c1); + res.c2 = -invdet * (a1 * c2 - a2 * c1); + res.c3 = invdet * (a1 * b2 - a2 * b1); + *this = res; + + return *this; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal>& aiMatrix3x3t<TReal>::RotationZ(TReal a, aiMatrix3x3t<TReal>& out) { + out.a1 = out.b2 = std::cos(a); + out.b1 = std::sin(a); + out.a2 = - out.b1; + + out.a3 = out.b3 = out.c1 = out.c2 = 0.f; + out.c3 = 1.f; + + return out; +} + +// ------------------------------------------------------------------------------------------------ +// Returns a rotation matrix for a rotation around an arbitrary axis. +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal>& aiMatrix3x3t<TReal>::Rotation( TReal a, const aiVector3t<TReal>& axis, aiMatrix3x3t<TReal>& out) { + TReal c = std::cos( a), s = std::sin( a), t = 1 - c; + TReal x = axis.x, y = axis.y, z = axis.z; + + // Many thanks to MathWorld and Wikipedia + out.a1 = t*x*x + c; out.a2 = t*x*y - s*z; out.a3 = t*x*z + s*y; + out.b1 = t*x*y + s*z; out.b2 = t*y*y + c; out.b3 = t*y*z - s*x; + out.c1 = t*x*z - s*y; out.c2 = t*y*z + s*x; out.c3 = t*z*z + c; + + return out; +} + +// ------------------------------------------------------------------------------------------------ +template <typename TReal> +AI_FORCE_INLINE +aiMatrix3x3t<TReal>& aiMatrix3x3t<TReal>::Translation( const aiVector2t<TReal>& v, aiMatrix3x3t<TReal>& out) { + out = aiMatrix3x3t<TReal>(); + out.a3 = v.x; + out.b3 = v.y; + return out; +} + +// ---------------------------------------------------------------------------------------- +/** A function for creating a rotation matrix that rotates a vector called + * "from" into another vector called "to". + * Input : from[3], to[3] which both must be *normalized* non-zero vectors + * Output: mtx[3][3] -- a 3x3 matrix in colum-major form + * Authors: Tomas Möller, John Hughes + * "Efficiently Building a Matrix to Rotate One Vector to Another" + * Journal of Graphics Tools, 4(4):1-4, 1999 + */ +// ---------------------------------------------------------------------------------------- +template <typename TReal> +AI_FORCE_INLINE aiMatrix3x3t<TReal>& aiMatrix3x3t<TReal>::FromToMatrix(const aiVector3t<TReal>& from, + const aiVector3t<TReal>& to, aiMatrix3x3t<TReal>& mtx) { + const TReal e = from * to; + const TReal f = (e < 0)? -e:e; + + if (f > static_cast<TReal>(1.0) - static_cast<TReal>(0.00001)) /* "from" and "to"-vector almost parallel */ + { + aiVector3D u,v; /* temporary storage vectors */ + aiVector3D x; /* vector most nearly orthogonal to "from" */ + + x.x = (from.x > 0.0)? from.x : -from.x; + x.y = (from.y > 0.0)? from.y : -from.y; + x.z = (from.z > 0.0)? from.z : -from.z; + + if (x.x < x.y) + { + if (x.x < x.z) + { + x.x = static_cast<TReal>(1.0); + x.y = x.z = static_cast<TReal>(0.0); + } + else + { + x.z = static_cast<TReal>(1.0); + x.x = x.y = static_cast<TReal>(0.0); + } + } + else + { + if (x.y < x.z) + { + x.y = static_cast<TReal>(1.0); + x.x = x.z = static_cast<TReal>(0.0); + } + else + { + x.z = static_cast<TReal>(1.0); + x.x = x.y = static_cast<TReal>(0.0); + } + } + + u.x = x.x - from.x; u.y = x.y - from.y; u.z = x.z - from.z; + v.x = x.x - to.x; v.y = x.y - to.y; v.z = x.z - to.z; + + const TReal c1_ = static_cast<TReal>(2.0) / (u * u); + const TReal c2_ = static_cast<TReal>(2.0) / (v * v); + const TReal c3_ = c1_ * c2_ * (u * v); + + for (unsigned int i = 0; i < 3; i++) + { + for (unsigned int j = 0; j < 3; j++) + { + mtx[i][j] = - c1_ * u[i] * u[j] - c2_ * v[i] * v[j] + + c3_ * v[i] * u[j]; + } + mtx[i][i] += static_cast<TReal>(1.0); + } + } + else /* the most common case, unless "from"="to", or "from"=-"to" */ + { + const aiVector3D v = from ^ to; + /* ... use this hand optimized version (9 mults less) */ + const TReal h = static_cast<TReal>(1.0)/(static_cast<TReal>(1.0) + e); /* optimization by Gottfried Chen */ + const TReal hvx = h * v.x; + const TReal hvz = h * v.z; + const TReal hvxy = hvx * v.y; + const TReal hvxz = hvx * v.z; + const TReal hvyz = hvz * v.y; + mtx[0][0] = e + hvx * v.x; + mtx[0][1] = hvxy - v.z; + mtx[0][2] = hvxz + v.y; + + mtx[1][0] = hvxy + v.z; + mtx[1][1] = e + h * v.y * v.y; + mtx[1][2] = hvyz - v.x; + + mtx[2][0] = hvxz - v.y; + mtx[2][1] = hvyz + v.x; + mtx[2][2] = e + hvz * v.z; + } + return mtx; +} + +#endif // __cplusplus +#endif // AI_MATRIX3X3_INL_INC |