mirror of
https://github.com/UberGames/GtkRadiant.git
synced 2024-11-22 11:51:22 +00:00
362 lines
16 KiB
C
362 lines
16 KiB
C
/*
|
|
Copyright (C) 1999-2007 id Software, Inc. and contributors.
|
|
For a list of contributors, see the accompanying CONTRIBUTORS file.
|
|
|
|
This file is part of GtkRadiant.
|
|
|
|
GtkRadiant is free software; you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation; either version 2 of the License, or
|
|
(at your option) any later version.
|
|
|
|
GtkRadiant is distributed in the hope that it will be useful,
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
GNU General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
along with GtkRadiant; if not, write to the Free Software
|
|
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
|
|
*/
|
|
|
|
#ifndef __MATHLIB__
|
|
#define __MATHLIB__
|
|
|
|
// mathlib.h
|
|
#include <math.h>
|
|
#include <float.h>
|
|
|
|
#include "bytebool.h"
|
|
|
|
#ifdef __cplusplus
|
|
extern "C"
|
|
{
|
|
#endif
|
|
|
|
typedef float vec_t;
|
|
typedef vec_t vec3_t[3];
|
|
typedef vec_t vec5_t[5];
|
|
typedef vec_t vec4_t[4];
|
|
|
|
// Smallest positive value for vec_t such that 1.0 + VEC_SMALLEST_EPSILON_AROUND_ONE != 1.0.
|
|
// In the case of 32 bit floats (which is almost certainly the case), it's 0.00000011921.
|
|
// Don't forget that your epsilons should depend on the possible range of values,
|
|
// because for example adding VEC_SMALLEST_EPSILON_AROUND_ONE to 1024.0 will have no effect.
|
|
#define VEC_SMALLEST_EPSILON_AROUND_ONE FLT_EPSILON
|
|
|
|
#define SIDE_FRONT 0
|
|
#define SIDE_ON 2
|
|
#define SIDE_BACK 1
|
|
#define SIDE_CROSS -2
|
|
|
|
// plane types are used to speed some tests
|
|
// 0-2 are axial planes
|
|
#define PLANE_X 0
|
|
#define PLANE_Y 1
|
|
#define PLANE_Z 2
|
|
#define PLANE_NON_AXIAL 3
|
|
|
|
#define Q_PI 3.14159265358979323846f
|
|
|
|
extern vec3_t vec3_origin;
|
|
|
|
#define EQUAL_EPSILON 0.001
|
|
|
|
#ifndef VEC_MAX
|
|
#define VEC_MAX 3.402823466e+38F
|
|
#endif
|
|
|
|
qboolean VectorCompare( vec3_t v1, vec3_t v2 );
|
|
|
|
#define DotProduct( x,y ) ( ( x )[0] * ( y )[0] + ( x )[1] * ( y )[1] + ( x )[2] * ( y )[2] )
|
|
#define VectorSubtract( a,b,c ) ( ( c )[0] = ( a )[0] - ( b )[0],( c )[1] = ( a )[1] - ( b )[1],( c )[2] = ( a )[2] - ( b )[2] )
|
|
#define VectorAdd( a,b,c ) ( ( c )[0] = ( a )[0] + ( b )[0],( c )[1] = ( a )[1] + ( b )[1],( c )[2] = ( a )[2] + ( b )[2] )
|
|
#define VectorIncrement( a,b ) ( ( b )[0] += ( a )[0],( b )[1] += ( a )[1],( b )[2] += ( a )[2] )
|
|
#define VectorCopy( a,b ) ( ( b )[0] = ( a )[0],( b )[1] = ( a )[1],( b )[2] = ( a )[2] )
|
|
#define VectorSet( v, a, b, c ) ( ( v )[0] = ( a ),( v )[1] = ( b ),( v )[2] = ( c ) )
|
|
#define VectorScale( a,b,c ) ( ( c )[0] = ( b ) * ( a )[0],( c )[1] = ( b ) * ( a )[1],( c )[2] = ( b ) * ( a )[2] )
|
|
#define VectorMid( a,b,c ) ( ( c )[0] = ( ( a )[0] + ( b )[0] ) * 0.5f,( c )[1] = ( ( a )[1] + ( b )[1] ) * 0.5f,( c )[2] = ( ( a )[2] + ( b )[2] ) * 0.5f )
|
|
#define VectorNegative( a,b ) ( ( b )[0] = -( a )[0],( b )[1] = -( a )[1],( b )[2] = -( a )[2] )
|
|
#define CrossProduct( a,b,c ) ( ( c )[0] = ( a )[1] * ( b )[2] - ( a )[2] * ( b )[1],( c )[1] = ( a )[2] * ( b )[0] - ( a )[0] * ( b )[2],( c )[2] = ( a )[0] * ( b )[1] - ( a )[1] * ( b )[0] )
|
|
#define VectorClear( x ) ( ( x )[0] = ( x )[1] = ( x )[2] = 0 )
|
|
|
|
#define Q_rint( in ) ( (vec_t)floor( in + 0.5 ) )
|
|
|
|
qboolean VectorIsOnAxis( vec3_t v );
|
|
qboolean VectorIsOnAxialPlane( vec3_t v );
|
|
|
|
vec_t VectorLength( vec3_t v );
|
|
|
|
void VectorMA( const vec3_t va, vec_t scale, const vec3_t vb, vec3_t vc );
|
|
|
|
void _CrossProduct( vec3_t v1, vec3_t v2, vec3_t cross );
|
|
// I need this define in order to test some of the regression tests from time to time.
|
|
// This define affect the precision of VectorNormalize() function only.
|
|
#define MATHLIB_VECTOR_NORMALIZE_PRECISION_FIX 1
|
|
vec_t VectorNormalize( const vec3_t in, vec3_t out );
|
|
vec_t ColorNormalize( const vec3_t in, vec3_t out );
|
|
void VectorInverse( vec3_t v );
|
|
void VectorPolar( vec3_t v, float radius, float theta, float phi );
|
|
|
|
// default snapping, to 1
|
|
void VectorSnap( vec3_t v );
|
|
|
|
// integer snapping
|
|
void VectorISnap( vec3_t point, int snap );
|
|
|
|
// Gef: added snap to float for sub-integer grid sizes
|
|
// TTimo: we still use the int version of VectorSnap when possible
|
|
// to avoid potential rounding issues
|
|
// TTimo: renaming to VectorFSnap for C implementation
|
|
void VectorFSnap( vec3_t point, float snap );
|
|
|
|
// NOTE: added these from Ritual's Q3Radiant
|
|
void ClearBounds( vec3_t mins, vec3_t maxs );
|
|
void AddPointToBounds( vec3_t v, vec3_t mins, vec3_t maxs );
|
|
|
|
void AngleVectors( vec3_t angles, vec3_t forward, vec3_t right, vec3_t up );
|
|
void VectorToAngles( vec3_t vec, vec3_t angles );
|
|
|
|
#define ZERO_EPSILON 1.0E-6
|
|
#define RAD2DEGMULT 57.29577951308232f
|
|
#define DEG2RADMULT 0.01745329251994329f
|
|
#define RAD2DEG( a ) ( ( a ) * RAD2DEGMULT )
|
|
#define DEG2RAD( a ) ( ( a ) * DEG2RADMULT )
|
|
|
|
void VectorRotate( vec3_t vIn, vec3_t vRotation, vec3_t out );
|
|
void VectorRotateOrigin( vec3_t vIn, vec3_t vRotation, vec3_t vOrigin, vec3_t out );
|
|
|
|
// some function merged from tools mathlib code
|
|
|
|
qboolean PlaneFromPoints( vec4_t plane, const vec3_t a, const vec3_t b, const vec3_t c );
|
|
void NormalToLatLong( const vec3_t normal, byte bytes[2] );
|
|
int PlaneTypeForNormal( vec3_t normal );
|
|
void RotatePointAroundVector( vec3_t dst, const vec3_t dir, const vec3_t point, float degrees );
|
|
|
|
// Spog
|
|
// code imported from geomlib
|
|
|
|
/*!
|
|
\todo
|
|
FIXME test calls such as intersect tests should be named test_
|
|
*/
|
|
|
|
typedef vec_t m3x3_t[9];
|
|
/*!NOTE
|
|
m4x4 looks like this..
|
|
|
|
x y z
|
|
x axis ( 0 1 2)
|
|
y axis ( 4 5 6)
|
|
z axis ( 8 9 10)
|
|
translation (12 13 14)
|
|
scale ( 0 5 10)
|
|
*/
|
|
typedef vec_t m4x4_t[16];
|
|
|
|
#define M4X4_INDEX( m,row,col ) ( m[( col << 2 ) + row] )
|
|
|
|
typedef enum { TRANSLATE, SCALE, ROTATE } transformtype; // legacy, used only in pmesh.cpp
|
|
|
|
typedef enum { eXYZ, eYZX, eZXY, eXZY, eYXZ, eZYX } eulerOrder_t;
|
|
|
|
// constructors
|
|
/*! create m4x4 as identity matrix */
|
|
void m4x4_identity( m4x4_t matrix );
|
|
/*! create m4x4 as a translation matrix, for a translation vec3 */
|
|
void m4x4_translation_for_vec3( m4x4_t matrix, const vec3_t translation );
|
|
/*! create m4x4 as a rotation matrix, for an euler angles (degrees) vec3 */
|
|
void m4x4_rotation_for_vec3( m4x4_t matrix, const vec3_t euler, eulerOrder_t order );
|
|
/*! create m4x4 as a scaling matrix, for a scale vec3 */
|
|
void m4x4_scale_for_vec3( m4x4_t matrix, const vec3_t scale );
|
|
/*! create m4x4 as a rotation matrix, for a quaternion vec4 */
|
|
void m4x4_rotation_for_quat( m4x4_t matrix, const vec4_t rotation );
|
|
/*! create m4x4 as a rotation matrix, for an axis vec3 and an angle (radians) */
|
|
void m4x4_rotation_for_axisangle( m4x4_t matrix, const vec3_t axis, vec_t angle );
|
|
|
|
// a valid m4x4 to be modified is always first argument
|
|
/*! translate m4x4 by a translation vec3 */
|
|
void m4x4_translate_by_vec3( m4x4_t matrix, const vec3_t translation );
|
|
/*! rotate m4x4 by a euler (degrees) vec3 */
|
|
void m4x4_rotate_by_vec3( m4x4_t matrix, const vec3_t euler, eulerOrder_t order );
|
|
/*! scale m4x4 by a scaling vec3 */
|
|
void m4x4_scale_by_vec3( m4x4_t matrix, const vec3_t scale );
|
|
/*! rotate m4x4 by a quaternion vec4 */
|
|
void m4x4_rotate_by_quat( m4x4_t matrix, const vec4_t rotation );
|
|
/*! rotate m4x4 by an axis vec3 and an angle (radians) */
|
|
void m4x4_rotate_by_axisangle( m4x4_t matrix, const vec3_t axis, vec_t angle );
|
|
/*! transform m4x4 by translation/euler/scaling vec3 (transform = translation.euler.scale) */
|
|
void m4x4_transform_by_vec3( m4x4_t matrix, const vec3_t translation, const vec3_t euler, eulerOrder_t order, const vec3_t scale );
|
|
/*! rotate m4x4 around a pivot point by euler(degrees) vec3 */
|
|
void m4x4_pivoted_rotate_by_vec3( m4x4_t matrix, const vec3_t euler, eulerOrder_t order, const vec3_t pivotpoint );
|
|
/*! scale m4x4 around a pivot point by scaling vec3 */
|
|
void m4x4_pivoted_scale_by_vec3( m4x4_t matrix, const vec3_t scale, const vec3_t pivotpoint );
|
|
/*! transform m4x4 around a pivot point by translation/euler/scaling vec3 */
|
|
void m4x4_pivoted_transform_by_vec3( m4x4_t matrix, const vec3_t translation, const vec3_t euler, eulerOrder_t order, const vec3_t scale, const vec3_t pivotpoint );
|
|
/*! rotate m4x4 around a pivot point by quaternion vec4 */
|
|
void m4x4_pivoted_rotate_by_quat( m4x4_t matrix, const vec4_t rotation, const vec3_t pivotpoint );
|
|
/*! rotate m4x4 around a pivot point by axis vec3 and angle (radians) */
|
|
void m4x4_pivoted_rotate_by_axisangle( m4x4_t matrix, const vec3_t axis, vec_t angle, const vec3_t pivotpoint );
|
|
/*! post-multiply m4x4 by another m4x4 */
|
|
void m4x4_multiply_by_m4x4( m4x4_t matrix, const m4x4_t other );
|
|
/*! pre-multiply m4x4 by another m4x4 */
|
|
void m4x4_premultiply_by_m4x4( m4x4_t matrix, const m4x4_t other );
|
|
|
|
/*! multiply a point (x,y,z,1) by matrix */
|
|
void m4x4_transform_point( const m4x4_t matrix, vec3_t point );
|
|
/*! multiply a normal (x,y,z,0) by matrix */
|
|
void m4x4_transform_normal( const m4x4_t matrix, vec3_t normal );
|
|
/*! multiply a vec4 (x,y,z,w) by matrix */
|
|
void m4x4_transform_vec4( const m4x4_t matrix, vec4_t vector );
|
|
|
|
/*! multiply a point (x,y,z,1) by matrix */
|
|
void m4x4_transform_point( const m4x4_t matrix, vec3_t point );
|
|
/*! multiply a normal (x,y,z,0) by matrix */
|
|
void m4x4_transform_normal( const m4x4_t matrix, vec3_t normal );
|
|
|
|
/*! transpose a m4x4 */
|
|
void m4x4_transpose( m4x4_t matrix );
|
|
/*! invert an orthogonal 4x3 subset of a 4x4 matrix */
|
|
void m4x4_orthogonal_invert( m4x4_t matrix );
|
|
/*! invert any m4x4 using Kramer's rule.. return 1 if matrix is singular, else return 0 */
|
|
int m4x4_invert( m4x4_t matrix );
|
|
|
|
/*!
|
|
\todo object/ray intersection functions should maybe return a point rather than a distance?
|
|
*/
|
|
|
|
/*!
|
|
aabb_t - "axis-aligned" bounding box...
|
|
origin: centre of bounding box...
|
|
extents: +/- extents of box from origin...
|
|
radius: cached length of extents vector...
|
|
*/
|
|
typedef struct aabb_s
|
|
{
|
|
vec3_t origin;
|
|
vec3_t extents;
|
|
vec_t radius;
|
|
} aabb_t;
|
|
|
|
/*!
|
|
bbox_t - oriented bounding box...
|
|
aabb: axis-aligned bounding box...
|
|
axes: orientation axes...
|
|
*/
|
|
typedef struct bbox_s
|
|
{
|
|
aabb_t aabb;
|
|
vec3_t axes[3];
|
|
} bbox_t;
|
|
|
|
/*!
|
|
ray_t - origin point and direction unit-vector
|
|
*/
|
|
typedef struct ray_s
|
|
{
|
|
vec3_t origin;
|
|
vec3_t direction;
|
|
} ray_t;
|
|
|
|
|
|
/*! Generate AABB from min/max. */
|
|
void aabb_construct_for_vec3( aabb_t *aabb, const vec3_t min, const vec3_t max );
|
|
/*! Update bounding-sphere radius. */
|
|
void aabb_update_radius( aabb_t *aabb );
|
|
/*! Initialise AABB to negative size. */
|
|
void aabb_clear( aabb_t *aabb );
|
|
|
|
/*! Extend AABB to include point. */
|
|
void aabb_extend_by_point( aabb_t *aabb, const vec3_t point );
|
|
/*! Extend AABB to include aabb_src. */
|
|
void aabb_extend_by_aabb( aabb_t *aabb, const aabb_t *aabb_src );
|
|
/*! Extend AABB by +/- extension vector. */
|
|
void aabb_extend_by_vec3( aabb_t *aabb, vec3_t extension );
|
|
|
|
/*! Return 2 if point is inside, else 1 if point is on surface, else 0. */
|
|
int aabb_intersect_point( const aabb_t *aabb, const vec3_t point );
|
|
/*! Return 2 if aabb_src intersects, else 1 if aabb_src touches exactly, else 0. */
|
|
int aabb_intersect_aabb( const aabb_t *aabb, const aabb_t *aabb_src );
|
|
/*! Return 2 if aabb is behind plane, else 1 if aabb intersects plane, else 0. */
|
|
int aabb_intersect_plane( const aabb_t *aabb, const float *plane );
|
|
/*! Return 1 if aabb intersects ray, else 0... dist = closest intersection. */
|
|
int aabb_intersect_ray( const aabb_t *aabb, const ray_t *ray, vec_t *dist );
|
|
/*! Return 1 if aabb intersects ray, else 0. Faster, but does not provide point of intersection */
|
|
int aabb_test_ray( const aabb_t* aabb, const ray_t* ray );
|
|
|
|
/*! Generate AABB from oriented bounding box. */
|
|
void aabb_for_bbox( aabb_t *aabb, const bbox_t *bbox );
|
|
/*! Generate AABB from 2-dimensions of min/max, specified by axis. */
|
|
void aabb_for_area( aabb_t *aabb, vec3_t area_tl, vec3_t area_br, int axis );
|
|
/*! Generate AABB to contain src * transform. NOTE: transform must be orthogonal */
|
|
void aabb_for_transformed_aabb( aabb_t* dst, const aabb_t* src, const m4x4_t transform );
|
|
|
|
|
|
/*! Generate oriented bounding box from AABB and transformation matrix. */
|
|
/*!\todo Remove need to specify euler/scale. */
|
|
void bbox_for_oriented_aabb( bbox_t *bbox, const aabb_t *aabb,
|
|
const m4x4_t matrix, const vec3_t euler, const vec3_t scale );
|
|
/*! Return 2 is bbox is behind plane, else return 1 if bbox intersects plane, else return 0. */
|
|
int bbox_intersect_plane( const bbox_t *bbox, const vec_t* plane );
|
|
|
|
|
|
/*! Generate a ray from an origin point and a direction unit-vector */
|
|
void ray_construct_for_vec3( ray_t *ray, const vec3_t origin, const vec3_t direction );
|
|
|
|
/*! Transform a ray */
|
|
void ray_transform( ray_t *ray, const m4x4_t matrix );
|
|
|
|
/*! return true if point intersects cone formed by ray, divergence and epsilon */
|
|
vec_t ray_intersect_point( const ray_t *ray, const vec3_t point, vec_t epsilon, vec_t divergence );
|
|
/*! return true if triangle intersects ray... dist = dist from intersection point to ray-origin */
|
|
vec_t ray_intersect_triangle( const ray_t *ray, qboolean bCullBack, const vec3_t vert0, const vec3_t vert1, const vec3_t vert2 );
|
|
|
|
|
|
////////////////////////////////////////////////////////////////////////////////
|
|
// Below is double-precision math stuff. This was initially needed by the new
|
|
// "base winding" code in q3map2 brush processing in order to fix the famous
|
|
// "disappearing triangles" issue. These definitions can be used wherever extra
|
|
// precision is needed.
|
|
////////////////////////////////////////////////////////////////////////////////
|
|
|
|
typedef double vec_accu_t;
|
|
typedef vec_accu_t vec3_accu_t[3];
|
|
|
|
// Smallest positive value for vec_accu_t such that 1.0 + VEC_ACCU_SMALLEST_EPSILON_AROUND_ONE != 1.0.
|
|
// In the case of 64 bit doubles (which is almost certainly the case), it's 0.00000000000000022204.
|
|
// Don't forget that your epsilons should depend on the possible range of values,
|
|
// because for example adding VEC_ACCU_SMALLEST_EPSILON_AROUND_ONE to 1024.0 will have no effect.
|
|
#define VEC_ACCU_SMALLEST_EPSILON_AROUND_ONE DBL_EPSILON
|
|
|
|
vec_accu_t VectorLengthAccu( const vec3_accu_t v );
|
|
|
|
// I have a feeling it may be safer to break these #define functions out into actual functions
|
|
// in order to avoid accidental loss of precision. For example, say you call
|
|
// VectorScaleAccu(vec3_t, vec_t, vec3_accu_t). The scale would take place in 32 bit land
|
|
// and the result would be cast to 64 bit, which would cause total loss of precision when
|
|
// scaling by a large factor.
|
|
//#define DotProductAccu(x, y) ((x)[0] * (y)[0] + (x)[1] * (y)[1] + (x)[2] * (y)[2])
|
|
//#define VectorSubtractAccu(a, b, c) ((c)[0] = (a)[0] - (b)[0], (c)[1] = (a)[1] - (b)[1], (c)[2] = (a)[2] - (b)[2])
|
|
//#define VectorAddAccu(a, b, c) ((c)[0] = (a)[0] + (b)[0], (c)[1] = (a)[1] + (b)[1], (c)[2] = (a)[2] + (b)[2])
|
|
//#define VectorCopyAccu(a, b) ((b)[0] = (a)[0], (b)[1] = (a)[1], (b)[2] = (a)[2])
|
|
//#define VectorScaleAccu(a, b, c) ((c)[0] = (b) * (a)[0], (c)[1] = (b) * (a)[1], (c)[2] = (b) * (a)[2])
|
|
//#define CrossProductAccu(a, b, c) ((c)[0] = (a)[1] * (b)[2] - (a)[2] * (b)[1], (c)[1] = (a)[2] * (b)[0] - (a)[0] * (b)[2], (c)[2] = (a)[0] * (b)[1] - (a)[1] * (b)[0])
|
|
//#define Q_rintAccu(in) ((vec_accu_t) floor(in + 0.5))
|
|
|
|
vec_accu_t DotProductAccu( const vec3_accu_t a, const vec3_accu_t b );
|
|
void VectorSubtractAccu( const vec3_accu_t a, const vec3_accu_t b, vec3_accu_t out );
|
|
void VectorAddAccu( const vec3_accu_t a, const vec3_accu_t b, vec3_accu_t out );
|
|
void VectorCopyAccu( const vec3_accu_t in, vec3_accu_t out );
|
|
void VectorScaleAccu( const vec3_accu_t in, vec_accu_t scaleFactor, vec3_accu_t out );
|
|
void CrossProductAccu( const vec3_accu_t a, const vec3_accu_t b, vec3_accu_t out );
|
|
vec_accu_t Q_rintAccu( vec_accu_t val );
|
|
|
|
void VectorCopyAccuToRegular( const vec3_accu_t in, vec3_t out );
|
|
void VectorCopyRegularToAccu( const vec3_t in, vec3_accu_t out );
|
|
vec_accu_t VectorNormalizeAccu( const vec3_accu_t in, vec3_accu_t out );
|
|
|
|
#ifdef __cplusplus
|
|
}
|
|
#endif
|
|
|
|
#endif /* __MATHLIB__ */
|