2004-02-29 13:20:34 +00:00
/*
2008-01-29 09:44:26 +00:00
* $ Id $
2004-02-29 13:20:34 +00:00
*
2008-04-16 22:40:48 +00:00
* * * * * * BEGIN GPL LICENSE BLOCK * * * * *
2004-02-29 13:20:34 +00:00
*
* This program 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
2008-04-16 22:40:48 +00:00
* of the License , or ( at your option ) any later version .
2004-02-29 13:20:34 +00:00
*
* This program 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 this program ; if not , write to the Free Software Foundation ,
* Inc . , 59 Temple Place - Suite 330 , Boston , MA 02111 - 1307 , USA .
*
* The Original Code is Copyright ( C ) 2001 - 2002 by NaN Holding BV .
* All rights reserved .
*
* This is a new part of Blender .
*
2005-05-25 04:52:52 +00:00
* Contributor ( s ) : Joseph Gilbert , Campbell Barton
2004-02-29 13:20:34 +00:00
*
2008-04-16 22:40:48 +00:00
* * * * * * END GPL LICENSE BLOCK * * * * *
2004-02-29 13:20:34 +00:00
*/
2005-07-18 03:50:37 +00:00
# include "Mathutils.h"
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
# include "BLI_arithb.h"
# include "PIL_time.h"
# include "BLI_rand.h"
2005-09-26 15:06:28 +00:00
# include "BKE_utildefines.h"
2006-07-01 15:29:04 +00:00
2004-10-07 19:25:40 +00:00
# include "gen_utils.h"
2005-07-18 03:50:37 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//-------------------------DOC STRINGS ---------------------------
2004-10-07 19:25:40 +00:00
static char M_Mathutils_doc [ ] = " The Blender Mathutils module \n \n " ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
static char M_Mathutils_Vector_doc [ ] = " () - create a new vector object from a list of floats " ;
static char M_Mathutils_Matrix_doc [ ] = " () - create a new matrix object from a list of floats " ;
static char M_Mathutils_Quaternion_doc [ ] = " () - create a quaternion from a list or an axis of rotation and an angle " ;
static char M_Mathutils_Euler_doc [ ] = " () - create and return a new euler object " ;
2004-10-07 19:25:40 +00:00
static char M_Mathutils_Rand_doc [ ] = " () - return a random number " ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
static char M_Mathutils_CrossVecs_doc [ ] = " () - returns a vector perpedicular to the 2 vectors crossed " ;
2004-10-07 19:25:40 +00:00
static char M_Mathutils_CopyVec_doc [ ] = " () - create a copy of vector " ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
static char M_Mathutils_DotVecs_doc [ ] = " () - return the dot product of two vectors " ;
static char M_Mathutils_AngleBetweenVecs_doc [ ] = " () - returns the angle between two vectors in degrees " ;
static char M_Mathutils_MidpointVecs_doc [ ] = " () - return the vector to the midpoint between two vectors " ;
static char M_Mathutils_MatMultVec_doc [ ] = " () - multiplies a matrix by a column vector " ;
static char M_Mathutils_VecMultMat_doc [ ] = " () - multiplies a row vector by a matrix " ;
static char M_Mathutils_ProjectVecs_doc [ ] = " () - returns the projection vector from the projection of vecA onto vecB " ;
static char M_Mathutils_RotationMatrix_doc [ ] = " () - construct a rotation matrix from an angle and axis of rotation " ;
static char M_Mathutils_ScaleMatrix_doc [ ] = " () - construct a scaling matrix from a scaling factor " ;
static char M_Mathutils_OrthoProjectionMatrix_doc [ ] = " () - construct a orthographic projection matrix from a selected plane " ;
static char M_Mathutils_ShearMatrix_doc [ ] = " () - construct a shearing matrix from a plane of shear and a shear factor " ;
2004-10-07 19:25:40 +00:00
static char M_Mathutils_CopyMat_doc [ ] = " () - create a copy of a matrix " ;
2007-06-16 12:24:41 +00:00
static char M_Mathutils_TranslationMatrix_doc [ ] = " (vec) - create a translation matrix from a vector " ;
2004-10-07 19:25:40 +00:00
static char M_Mathutils_CopyQuat_doc [ ] = " () - copy quatB to quatA " ;
static char M_Mathutils_CopyEuler_doc [ ] = " () - copy eulB to eultA " ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
static char M_Mathutils_CrossQuats_doc [ ] = " () - return the mutliplication of two quaternions " ;
static char M_Mathutils_DotQuats_doc [ ] = " () - return the dot product of two quaternions " ;
static char M_Mathutils_Slerp_doc [ ] = " () - returns the interpolation between two quaternions " ;
static char M_Mathutils_DifferenceQuats_doc [ ] = " () - return the angular displacment difference between two quats " ;
static char M_Mathutils_RotateEuler_doc [ ] = " () - rotate euler by an axis and angle " ;
2005-09-26 15:06:28 +00:00
static char M_Mathutils_Intersect_doc [ ] = " (v1, v2, v3, ray, orig, clip=1) - returns the intersection between a ray and a triangle, if possible, returns None otherwise " ;
static char M_Mathutils_TriangleArea_doc [ ] = " (v1, v2, v3) - returns the area size of the 2D or 3D triangle defined " ;
static char M_Mathutils_TriangleNormal_doc [ ] = " (v1, v2, v3) - returns the normal of the 3D triangle defined " ;
static char M_Mathutils_QuadNormal_doc [ ] = " (v1, v2, v3, v4) - returns the normal of the 3D quad defined " ;
static char M_Mathutils_LineIntersect_doc [ ] = " (v1, v2, v3, v4) - returns a tuple with the points on each line respectively closest to the other " ;
2005-11-15 21:45:05 +00:00
static char M_Mathutils_Point_doc [ ] = " Creates a 2d or 3d point object " ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//-----------------------METHOD DEFINITIONS ----------------------
2004-10-07 19:25:40 +00:00
struct PyMethodDef M_Mathutils_methods [ ] = {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
{ " Rand " , ( PyCFunction ) M_Mathutils_Rand , METH_VARARGS , M_Mathutils_Rand_doc } ,
{ " Vector " , ( PyCFunction ) M_Mathutils_Vector , METH_VARARGS , M_Mathutils_Vector_doc } ,
{ " CrossVecs " , ( PyCFunction ) M_Mathutils_CrossVecs , METH_VARARGS , M_Mathutils_CrossVecs_doc } ,
{ " DotVecs " , ( PyCFunction ) M_Mathutils_DotVecs , METH_VARARGS , M_Mathutils_DotVecs_doc } ,
{ " AngleBetweenVecs " , ( PyCFunction ) M_Mathutils_AngleBetweenVecs , METH_VARARGS , M_Mathutils_AngleBetweenVecs_doc } ,
{ " MidpointVecs " , ( PyCFunction ) M_Mathutils_MidpointVecs , METH_VARARGS , M_Mathutils_MidpointVecs_doc } ,
{ " VecMultMat " , ( PyCFunction ) M_Mathutils_VecMultMat , METH_VARARGS , M_Mathutils_VecMultMat_doc } ,
{ " ProjectVecs " , ( PyCFunction ) M_Mathutils_ProjectVecs , METH_VARARGS , M_Mathutils_ProjectVecs_doc } ,
{ " CopyVec " , ( PyCFunction ) M_Mathutils_CopyVec , METH_VARARGS , M_Mathutils_CopyVec_doc } ,
{ " Matrix " , ( PyCFunction ) M_Mathutils_Matrix , METH_VARARGS , M_Mathutils_Matrix_doc } ,
{ " RotationMatrix " , ( PyCFunction ) M_Mathutils_RotationMatrix , METH_VARARGS , M_Mathutils_RotationMatrix_doc } ,
{ " ScaleMatrix " , ( PyCFunction ) M_Mathutils_ScaleMatrix , METH_VARARGS , M_Mathutils_ScaleMatrix_doc } ,
{ " ShearMatrix " , ( PyCFunction ) M_Mathutils_ShearMatrix , METH_VARARGS , M_Mathutils_ShearMatrix_doc } ,
2007-06-16 12:24:41 +00:00
{ " TranslationMatrix " , ( PyCFunction ) M_Mathutils_TranslationMatrix , METH_O , M_Mathutils_TranslationMatrix_doc } ,
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
{ " CopyMat " , ( PyCFunction ) M_Mathutils_CopyMat , METH_VARARGS , M_Mathutils_CopyMat_doc } ,
{ " OrthoProjectionMatrix " , ( PyCFunction ) M_Mathutils_OrthoProjectionMatrix , METH_VARARGS , M_Mathutils_OrthoProjectionMatrix_doc } ,
{ " MatMultVec " , ( PyCFunction ) M_Mathutils_MatMultVec , METH_VARARGS , M_Mathutils_MatMultVec_doc } ,
{ " Quaternion " , ( PyCFunction ) M_Mathutils_Quaternion , METH_VARARGS , M_Mathutils_Quaternion_doc } ,
{ " CopyQuat " , ( PyCFunction ) M_Mathutils_CopyQuat , METH_VARARGS , M_Mathutils_CopyQuat_doc } ,
{ " CrossQuats " , ( PyCFunction ) M_Mathutils_CrossQuats , METH_VARARGS , M_Mathutils_CrossQuats_doc } ,
{ " DotQuats " , ( PyCFunction ) M_Mathutils_DotQuats , METH_VARARGS , M_Mathutils_DotQuats_doc } ,
{ " DifferenceQuats " , ( PyCFunction ) M_Mathutils_DifferenceQuats , METH_VARARGS , M_Mathutils_DifferenceQuats_doc } ,
{ " Slerp " , ( PyCFunction ) M_Mathutils_Slerp , METH_VARARGS , M_Mathutils_Slerp_doc } ,
{ " Euler " , ( PyCFunction ) M_Mathutils_Euler , METH_VARARGS , M_Mathutils_Euler_doc } ,
{ " CopyEuler " , ( PyCFunction ) M_Mathutils_CopyEuler , METH_VARARGS , M_Mathutils_CopyEuler_doc } ,
{ " RotateEuler " , ( PyCFunction ) M_Mathutils_RotateEuler , METH_VARARGS , M_Mathutils_RotateEuler_doc } ,
2005-09-26 15:06:28 +00:00
{ " Intersect " , ( PyCFunction ) M_Mathutils_Intersect , METH_VARARGS , M_Mathutils_Intersect_doc } ,
{ " TriangleArea " , ( PyCFunction ) M_Mathutils_TriangleArea , METH_VARARGS , M_Mathutils_TriangleArea_doc } ,
{ " TriangleNormal " , ( PyCFunction ) M_Mathutils_TriangleNormal , METH_VARARGS , M_Mathutils_TriangleNormal_doc } ,
{ " QuadNormal " , ( PyCFunction ) M_Mathutils_QuadNormal , METH_VARARGS , M_Mathutils_QuadNormal_doc } ,
{ " LineIntersect " , ( PyCFunction ) M_Mathutils_LineIntersect , METH_VARARGS , M_Mathutils_LineIntersect_doc } ,
2005-11-15 21:45:05 +00:00
{ " Point " , ( PyCFunction ) M_Mathutils_Point , METH_VARARGS , M_Mathutils_Point_doc } ,
2004-10-07 19:25:40 +00:00
{ NULL , NULL , 0 , NULL }
} ;
2008-08-05 10:39:13 +00:00
/*----------------------------MODULE INIT-------------------------*/
/* from can be Blender.Mathutils or GameLogic.Mathutils for the BGE */
PyObject * Mathutils_Init ( char * from )
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
{
PyObject * submodule ;
//seed the generator for the rand function
BLI_srand ( ( unsigned int ) ( PIL_check_seconds_timer ( ) *
0x7FFFFFFF ) ) ;
2006-10-03 16:07:48 +00:00
/* needed for getseters */
if ( PyType_Ready ( & vector_Type ) < 0 )
return NULL ;
2007-03-16 11:38:02 +00:00
if ( PyType_Ready ( & matrix_Type ) < 0 )
return NULL ;
if ( PyType_Ready ( & euler_Type ) < 0 )
return NULL ;
if ( PyType_Ready ( & quaternion_Type ) < 0 )
return NULL ;
2006-10-03 16:07:48 +00:00
2008-08-05 10:39:13 +00:00
submodule = Py_InitModule3 ( from , M_Mathutils_methods , M_Mathutils_doc ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return ( submodule ) ;
}
//-----------------------------METHODS----------------------------
//----------------column_vector_multiplication (internal)---------
//COLUMN VECTOR Multiplication (Matrix X Vector)
// [1][2][3] [a]
// [4][5][6] * [b]
// [7][8][9] [c]
//vector/matrix multiplication IS NOT COMMUTATIVE!!!!
PyObject * column_vector_multiplication ( MatrixObject * mat , VectorObject * vec )
{
float vecNew [ 4 ] , vecCopy [ 4 ] ;
double dot = 0.0f ;
int x , y , z = 0 ;
if ( mat - > rowSize ! = vec - > size ) {
if ( mat - > rowSize = = 4 & & vec - > size ! = 3 ) {
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
2006-07-26 12:34:56 +00:00
" matrix * vector: matrix row size and vector size must be the same " ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else {
2006-07-26 12:34:56 +00:00
vecCopy [ 3 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
}
}
2004-10-07 19:25:40 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < vec - > size ; x + + ) {
vecCopy [ x ] = vec - > vec [ x ] ;
}
2004-10-07 19:25:40 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < mat - > rowSize ; x + + ) {
for ( y = 0 ; y < mat - > colSize ; y + + ) {
dot + = mat - > matrix [ x ] [ y ] * vecCopy [ y ] ;
}
2005-07-18 03:50:37 +00:00
vecNew [ z + + ] = ( float ) dot ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
dot = 0.0f ;
}
2005-09-27 17:03:28 +00:00
return newVectorObject ( vecNew , vec - > size , Py_NEW ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
}
2005-07-23 13:46:40 +00:00
//This is a helper for point/matrix translation
2006-07-01 15:29:04 +00:00
2005-07-23 13:46:40 +00:00
PyObject * column_point_multiplication ( MatrixObject * mat , PointObject * pt )
{
float ptNew [ 4 ] , ptCopy [ 4 ] ;
double dot = 0.0f ;
int x , y , z = 0 ;
if ( mat - > rowSize ! = pt - > size ) {
if ( mat - > rowSize = = 4 & & pt - > size ! = 3 ) {
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" matrix * point: matrix row size and point size must be the same \n " ) ;
} else {
ptCopy [ 3 ] = 0.0f ;
}
}
for ( x = 0 ; x < pt - > size ; x + + ) {
ptCopy [ x ] = pt - > coord [ x ] ;
}
for ( x = 0 ; x < mat - > rowSize ; x + + ) {
for ( y = 0 ; y < mat - > colSize ; y + + ) {
dot + = mat - > matrix [ x ] [ y ] * ptCopy [ y ] ;
}
ptNew [ z + + ] = ( float ) dot ;
dot = 0.0f ;
}
2005-09-27 17:03:28 +00:00
return newPointObject ( ptNew , pt - > size , Py_NEW ) ;
2005-07-23 13:46:40 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//-----------------row_vector_multiplication (internal)-----------
//ROW VECTOR Multiplication - Vector X Matrix
//[x][y][z] * [1][2][3]
// [4][5][6]
// [7][8][9]
//vector/matrix multiplication IS NOT COMMUTATIVE!!!!
PyObject * row_vector_multiplication ( VectorObject * vec , MatrixObject * mat )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
float vecNew [ 4 ] , vecCopy [ 4 ] ;
double dot = 0.0f ;
2006-10-03 16:07:48 +00:00
int x , y , z = 0 , vec_size = vec - > size ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
2006-10-03 16:07:48 +00:00
if ( mat - > colSize ! = vec_size ) {
if ( mat - > rowSize = = 4 & & vec_size ! = 3 ) {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
2006-07-26 12:34:56 +00:00
" vector * matrix: matrix column size and the vector size must be the same " ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else {
2006-07-26 12:34:56 +00:00
vecCopy [ 3 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
}
}
2006-10-03 16:07:48 +00:00
for ( x = 0 ; x < vec_size ; x + + ) {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
vecCopy [ x ] = vec - > vec [ x ] ;
}
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//muliplication
for ( x = 0 ; x < mat - > colSize ; x + + ) {
for ( y = 0 ; y < mat - > rowSize ; y + + ) {
dot + = mat - > matrix [ y ] [ x ] * vecCopy [ y ] ;
}
2005-07-18 03:50:37 +00:00
vecNew [ z + + ] = ( float ) dot ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
dot = 0.0f ;
}
2006-10-03 16:07:48 +00:00
return newVectorObject ( vecNew , vec_size , Py_NEW ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
}
2005-07-23 13:46:40 +00:00
//This is a helper for the point class
PyObject * row_point_multiplication ( PointObject * pt , MatrixObject * mat )
{
float ptNew [ 4 ] , ptCopy [ 4 ] ;
double dot = 0.0f ;
int x , y , z = 0 , size ;
if ( mat - > colSize ! = pt - > size ) {
if ( mat - > rowSize = = 4 & & pt - > size ! = 3 ) {
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" point * matrix: matrix column size and the point size must be the same \n " ) ;
} else {
ptCopy [ 3 ] = 0.0f ;
}
}
size = pt - > size ;
for ( x = 0 ; x < pt - > size ; x + + ) {
ptCopy [ x ] = pt - > coord [ x ] ;
}
//muliplication
for ( x = 0 ; x < mat - > colSize ; x + + ) {
for ( y = 0 ; y < mat - > rowSize ; y + + ) {
dot + = mat - > matrix [ y ] [ x ] * ptCopy [ y ] ;
}
ptNew [ z + + ] = ( float ) dot ;
dot = 0.0f ;
}
2005-09-27 17:03:28 +00:00
return newPointObject ( ptNew , size , Py_NEW ) ;
2005-07-23 13:46:40 +00:00
}
//-----------------quat_rotation (internal)-----------
//This function multiplies a vector/point * quat or vice versa
//to rotate the point/vector by the quaternion
//arguments should all be 3D
PyObject * quat_rotation ( PyObject * arg1 , PyObject * arg2 )
{
float rot [ 3 ] ;
QuaternionObject * quat = NULL ;
VectorObject * vec = NULL ;
PointObject * pt = NULL ;
if ( QuaternionObject_Check ( arg1 ) ) {
quat = ( QuaternionObject * ) arg1 ;
if ( VectorObject_Check ( arg2 ) ) {
vec = ( VectorObject * ) arg2 ;
rot [ 0 ] = quat - > quat [ 0 ] * quat - > quat [ 0 ] * vec - > vec [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 0 ] * vec - > vec [ 2 ] -
2 * quat - > quat [ 3 ] * quat - > quat [ 0 ] * vec - > vec [ 1 ] + quat - > quat [ 1 ] * quat - > quat [ 1 ] * vec - > vec [ 0 ] +
2 * quat - > quat [ 2 ] * quat - > quat [ 1 ] * vec - > vec [ 1 ] + 2 * quat - > quat [ 3 ] * quat - > quat [ 1 ] * vec - > vec [ 2 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * vec - > vec [ 0 ] - quat - > quat [ 2 ] * quat - > quat [ 2 ] * vec - > vec [ 0 ] ;
rot [ 1 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 2 ] * vec - > vec [ 0 ] + quat - > quat [ 2 ] * quat - > quat [ 2 ] * vec - > vec [ 1 ] +
2 * quat - > quat [ 3 ] * quat - > quat [ 2 ] * vec - > vec [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 3 ] * vec - > vec [ 0 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * vec - > vec [ 1 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * vec - > vec [ 1 ] -
2 * quat - > quat [ 1 ] * quat - > quat [ 0 ] * vec - > vec [ 2 ] - quat - > quat [ 1 ] * quat - > quat [ 1 ] * vec - > vec [ 1 ] ;
rot [ 2 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 3 ] * vec - > vec [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 3 ] * vec - > vec [ 1 ] +
quat - > quat [ 3 ] * quat - > quat [ 3 ] * vec - > vec [ 2 ] - 2 * quat - > quat [ 0 ] * quat - > quat [ 2 ] * vec - > vec [ 0 ] -
quat - > quat [ 2 ] * quat - > quat [ 2 ] * vec - > vec [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 1 ] * vec - > vec [ 1 ] -
quat - > quat [ 1 ] * quat - > quat [ 1 ] * vec - > vec [ 2 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * vec - > vec [ 2 ] ;
2005-09-27 17:03:28 +00:00
return newVectorObject ( rot , 3 , Py_NEW ) ;
2005-07-23 13:46:40 +00:00
} else if ( PointObject_Check ( arg2 ) ) {
pt = ( PointObject * ) arg2 ;
rot [ 0 ] = quat - > quat [ 0 ] * quat - > quat [ 0 ] * pt - > coord [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 0 ] * pt - > coord [ 2 ] -
2 * quat - > quat [ 3 ] * quat - > quat [ 0 ] * pt - > coord [ 1 ] + quat - > quat [ 1 ] * quat - > quat [ 1 ] * pt - > coord [ 0 ] +
2 * quat - > quat [ 2 ] * quat - > quat [ 1 ] * pt - > coord [ 1 ] + 2 * quat - > quat [ 3 ] * quat - > quat [ 1 ] * pt - > coord [ 2 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * pt - > coord [ 0 ] - quat - > quat [ 2 ] * quat - > quat [ 2 ] * pt - > coord [ 0 ] ;
rot [ 1 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 2 ] * pt - > coord [ 0 ] + quat - > quat [ 2 ] * quat - > quat [ 2 ] * pt - > coord [ 1 ] +
2 * quat - > quat [ 3 ] * quat - > quat [ 2 ] * pt - > coord [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 3 ] * pt - > coord [ 0 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * pt - > coord [ 1 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * pt - > coord [ 1 ] -
2 * quat - > quat [ 1 ] * quat - > quat [ 0 ] * pt - > coord [ 2 ] - quat - > quat [ 1 ] * quat - > quat [ 1 ] * pt - > coord [ 1 ] ;
rot [ 2 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 3 ] * pt - > coord [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 3 ] * pt - > coord [ 1 ] +
quat - > quat [ 3 ] * quat - > quat [ 3 ] * pt - > coord [ 2 ] - 2 * quat - > quat [ 0 ] * quat - > quat [ 2 ] * pt - > coord [ 0 ] -
quat - > quat [ 2 ] * quat - > quat [ 2 ] * pt - > coord [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 1 ] * pt - > coord [ 1 ] -
quat - > quat [ 1 ] * quat - > quat [ 1 ] * pt - > coord [ 2 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * pt - > coord [ 2 ] ;
2005-09-27 17:03:28 +00:00
return newPointObject ( rot , 3 , Py_NEW ) ;
2005-07-23 13:46:40 +00:00
}
} else if ( VectorObject_Check ( arg1 ) ) {
vec = ( VectorObject * ) arg1 ;
if ( QuaternionObject_Check ( arg2 ) ) {
quat = ( QuaternionObject * ) arg2 ;
rot [ 0 ] = quat - > quat [ 0 ] * quat - > quat [ 0 ] * vec - > vec [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 0 ] * vec - > vec [ 2 ] -
2 * quat - > quat [ 3 ] * quat - > quat [ 0 ] * vec - > vec [ 1 ] + quat - > quat [ 1 ] * quat - > quat [ 1 ] * vec - > vec [ 0 ] +
2 * quat - > quat [ 2 ] * quat - > quat [ 1 ] * vec - > vec [ 1 ] + 2 * quat - > quat [ 3 ] * quat - > quat [ 1 ] * vec - > vec [ 2 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * vec - > vec [ 0 ] - quat - > quat [ 2 ] * quat - > quat [ 2 ] * vec - > vec [ 0 ] ;
rot [ 1 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 2 ] * vec - > vec [ 0 ] + quat - > quat [ 2 ] * quat - > quat [ 2 ] * vec - > vec [ 1 ] +
2 * quat - > quat [ 3 ] * quat - > quat [ 2 ] * vec - > vec [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 3 ] * vec - > vec [ 0 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * vec - > vec [ 1 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * vec - > vec [ 1 ] -
2 * quat - > quat [ 1 ] * quat - > quat [ 0 ] * vec - > vec [ 2 ] - quat - > quat [ 1 ] * quat - > quat [ 1 ] * vec - > vec [ 1 ] ;
rot [ 2 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 3 ] * vec - > vec [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 3 ] * vec - > vec [ 1 ] +
quat - > quat [ 3 ] * quat - > quat [ 3 ] * vec - > vec [ 2 ] - 2 * quat - > quat [ 0 ] * quat - > quat [ 2 ] * vec - > vec [ 0 ] -
quat - > quat [ 2 ] * quat - > quat [ 2 ] * vec - > vec [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 1 ] * vec - > vec [ 1 ] -
quat - > quat [ 1 ] * quat - > quat [ 1 ] * vec - > vec [ 2 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * vec - > vec [ 2 ] ;
2005-09-27 17:03:28 +00:00
return newVectorObject ( rot , 3 , Py_NEW ) ;
2005-07-23 13:46:40 +00:00
}
} else if ( PointObject_Check ( arg1 ) ) {
pt = ( PointObject * ) arg1 ;
if ( QuaternionObject_Check ( arg2 ) ) {
quat = ( QuaternionObject * ) arg2 ;
rot [ 0 ] = quat - > quat [ 0 ] * quat - > quat [ 0 ] * pt - > coord [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 0 ] * pt - > coord [ 2 ] -
2 * quat - > quat [ 3 ] * quat - > quat [ 0 ] * pt - > coord [ 1 ] + quat - > quat [ 1 ] * quat - > quat [ 1 ] * pt - > coord [ 0 ] +
2 * quat - > quat [ 2 ] * quat - > quat [ 1 ] * pt - > coord [ 1 ] + 2 * quat - > quat [ 3 ] * quat - > quat [ 1 ] * pt - > coord [ 2 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * pt - > coord [ 0 ] - quat - > quat [ 2 ] * quat - > quat [ 2 ] * pt - > coord [ 0 ] ;
rot [ 1 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 2 ] * pt - > coord [ 0 ] + quat - > quat [ 2 ] * quat - > quat [ 2 ] * pt - > coord [ 1 ] +
2 * quat - > quat [ 3 ] * quat - > quat [ 2 ] * pt - > coord [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 3 ] * pt - > coord [ 0 ] -
quat - > quat [ 3 ] * quat - > quat [ 3 ] * pt - > coord [ 1 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * pt - > coord [ 1 ] -
2 * quat - > quat [ 1 ] * quat - > quat [ 0 ] * pt - > coord [ 2 ] - quat - > quat [ 1 ] * quat - > quat [ 1 ] * pt - > coord [ 1 ] ;
rot [ 2 ] = 2 * quat - > quat [ 1 ] * quat - > quat [ 3 ] * pt - > coord [ 0 ] + 2 * quat - > quat [ 2 ] * quat - > quat [ 3 ] * pt - > coord [ 1 ] +
quat - > quat [ 3 ] * quat - > quat [ 3 ] * pt - > coord [ 2 ] - 2 * quat - > quat [ 0 ] * quat - > quat [ 2 ] * pt - > coord [ 0 ] -
quat - > quat [ 2 ] * quat - > quat [ 2 ] * pt - > coord [ 2 ] + 2 * quat - > quat [ 0 ] * quat - > quat [ 1 ] * pt - > coord [ 1 ] -
quat - > quat [ 1 ] * quat - > quat [ 1 ] * pt - > coord [ 2 ] + quat - > quat [ 0 ] * quat - > quat [ 0 ] * pt - > coord [ 2 ] ;
2005-09-27 17:03:28 +00:00
return newPointObject ( rot , 3 , Py_NEW ) ;
2005-07-23 13:46:40 +00:00
}
}
return ( EXPP_ReturnPyObjError ( PyExc_RuntimeError ,
" quat_rotation(internal): internal problem rotating vector/point \n " ) ) ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.Rand() --------------------
//returns a random number between a high and low value
PyObject * M_Mathutils_Rand ( PyObject * self , PyObject * args )
{
2004-02-29 13:20:34 +00:00
float high , low , range ;
double rand ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//initializers
2004-02-29 13:20:34 +00:00
high = 1.0 ;
low = 0.0 ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " |ff " , & low , & high ) )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Rand(): expected nothing or optional (float, float) \n " ) ) ;
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ( high < low ) | | ( high < 0 & & low > 0 ) )
return ( EXPP_ReturnPyObjError ( PyExc_ValueError ,
" Mathutils.Rand(): high value should be larger than low value \n " ) ) ;
2004-02-29 13:20:34 +00:00
//get the random number 0 - 1
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
rand = BLI_drand ( ) ;
2004-02-29 13:20:34 +00:00
//set it to range
range = high - low ;
rand = rand * range ;
rand = rand + low ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return PyFloat_FromDouble ( rand ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------VECTOR FUNCTIONS---------------------
//----------------------------------Mathutils.Vector() ------------------
2004-02-29 13:20:34 +00:00
// Supports 2D, 3D, and 4D vector objects both int and float values
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
// accepted. Mixed float and int values accepted. Ints are parsed to float
PyObject * M_Mathutils_Vector ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
PyObject * listObject = NULL ;
2005-01-23 17:04:02 +00:00
int size , i ;
float vec [ 4 ] ;
2005-09-27 17:03:28 +00:00
PyObject * v , * f ;
2004-02-29 13:20:34 +00:00
2005-01-23 17:04:02 +00:00
size = PySequence_Length ( args ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( size = = 1 ) {
2005-01-23 17:04:02 +00:00
listObject = PySequence_GetItem ( args , 0 ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( PySequence_Check ( listObject ) ) {
2005-01-23 17:04:02 +00:00
size = PySequence_Length ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else { // Single argument was not a sequence
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Vector(): 2-4 floats or ints expected (optionally in a sequence) \n " ) ;
2005-01-23 17:04:02 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( size = = 0 ) {
//returns a new empty 3d vector
2005-09-27 17:03:28 +00:00
return newVectorObject ( NULL , 3 , Py_NEW ) ;
2005-01-23 17:04:02 +00:00
} else {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
listObject = EXPP_incr_ret ( args ) ;
2004-02-29 13:20:34 +00:00
}
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( size < 2 | | size > 4 ) { // Invalid vector size
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Vector(): 2-4 floats or ints expected (optionally in a sequence) \n " ) ;
2005-01-23 17:04:02 +00:00
}
2005-01-23 20:15:06 +00:00
2005-09-27 17:03:28 +00:00
for ( i = 0 ; i < size ; i + + ) {
2005-01-23 20:15:06 +00:00
v = PySequence_GetItem ( listObject , i ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( v = = NULL ) { // Failed to read sequence
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_RuntimeError ,
" Mathutils.Vector(): 2-4 floats or ints expected (optionally in a sequence) \n " ) ;
2004-02-29 13:20:34 +00:00
}
2005-09-27 17:03:28 +00:00
2005-01-23 20:15:06 +00:00
f = PyNumber_Float ( v ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( f = = NULL ) { // parsed item not a number
2005-01-23 17:04:02 +00:00
Py_DECREF ( v ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Vector(): 2-4 floats or ints expected (optionally in a sequence) \n " ) ;
2005-01-23 17:04:02 +00:00
}
2005-09-27 17:03:28 +00:00
2005-07-18 03:50:37 +00:00
vec [ i ] = ( float ) PyFloat_AS_DOUBLE ( f ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
EXPP_decr2 ( f , v ) ;
2004-02-29 13:20:34 +00:00
}
2005-01-23 17:04:02 +00:00
Py_DECREF ( listObject ) ;
2005-09-27 17:03:28 +00:00
return newVectorObject ( vec , size , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.CrossVecs() ---------------
2004-02-29 13:20:34 +00:00
//finds perpendicular vector - only 3D is supported
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
PyObject * M_Mathutils_CrossVecs ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
PyObject * vecCross = NULL ;
VectorObject * vec1 = NULL , * vec2 = NULL ;
if ( ! PyArg_ParseTuple ( args , " O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2 ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.CrossVecs(): expects (2) 3D vector objects \n " ) ;
if ( vec1 - > size ! = 3 | | vec2 - > size ! = 3 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.CrossVecs(): expects (2) 3D vector objects \n " ) ;
vecCross = newVectorObject ( NULL , 3 , Py_NEW ) ;
Crossf ( ( ( VectorObject * ) vecCross ) - > vec , vec1 - > vec , vec2 - > vec ) ;
2004-02-29 13:20:34 +00:00
return vecCross ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.DotVec() -------------------
//calculates the dot product of two vectors
PyObject * M_Mathutils_DotVecs ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
VectorObject * vec1 = NULL , * vec2 = NULL ;
double dot = 0.0f ;
2004-02-29 13:20:34 +00:00
int x ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2 ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
2007-08-06 14:27:08 +00:00
" Mathutils.DotVecs(): expects (2) vector objects of the same size \n " ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( vec1 - > size ! = vec2 - > size )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
2007-08-06 14:27:08 +00:00
" Mathutils.DotVecs(): expects (2) vector objects of the same size \n " ) ;
2004-09-25 20:30:40 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < vec1 - > size ; x + + ) {
2004-02-29 13:20:34 +00:00
dot + = vec1 - > vec [ x ] * vec2 - > vec [ x ] ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return PyFloat_FromDouble ( dot ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.AngleBetweenVecs() ---------
//calculates the angle between 2 vectors
PyObject * M_Mathutils_AngleBetweenVecs ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
VectorObject * vec1 = NULL , * vec2 = NULL ;
2005-10-07 15:24:32 +00:00
double dot = 0.0f , angleRads , test_v1 = 0.0f , test_v2 = 0.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
int x , size ;
if ( ! PyArg_ParseTuple ( args , " O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2 ) )
2005-10-10 15:28:59 +00:00
goto AttributeError1 ; //not vectors
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( vec1 - > size ! = vec2 - > size )
2005-10-10 15:28:59 +00:00
goto AttributeError1 ; //bad sizes
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//since size is the same....
size = vec1 - > size ;
2005-10-07 15:24:32 +00:00
for ( x = 0 ; x < size ; x + + ) {
test_v1 + = vec1 - > vec [ x ] * vec1 - > vec [ x ] ;
test_v2 + = vec2 - > vec [ x ] * vec2 - > vec [ x ] ;
}
if ( ! test_v1 | | ! test_v2 ) {
2005-10-10 15:28:59 +00:00
goto AttributeError2 ; //zero-length vector
2004-02-29 13:20:34 +00:00
}
2005-10-10 15:28:59 +00:00
2004-02-29 13:20:34 +00:00
//dot product
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < size ; x + + ) {
2005-10-10 15:28:59 +00:00
dot + = vec1 - > vec [ x ] * vec2 - > vec [ x ] ;
2004-02-29 13:20:34 +00:00
}
2005-10-10 15:28:59 +00:00
dot / = ( sqrt ( test_v1 ) * sqrt ( test_v2 ) ) ;
2005-10-10 19:50:53 +00:00
2008-01-29 09:44:26 +00:00
angleRads = ( double ) saacos ( dot ) ;
2004-02-29 13:20:34 +00:00
2005-10-10 19:50:53 +00:00
return PyFloat_FromDouble ( angleRads * ( 180 / Py_PI ) ) ;
2005-10-10 15:28:59 +00:00
AttributeError1 :
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.AngleBetweenVecs(): expects (2) VECTOR objects of the same size \n " ) ;
AttributeError2 :
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.AngleBetweenVecs(): zero length vectors are not acceptable arguments \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.MidpointVecs() -------------
//calculates the midpoint between 2 vectors
PyObject * M_Mathutils_MidpointVecs ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
VectorObject * vec1 = NULL , * vec2 = NULL ;
float vec [ 4 ] ;
2004-02-29 13:20:34 +00:00
int x ;
2004-10-12 23:58:03 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2 ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.MidpointVecs(): expects (2) vector objects of the same size \n " ) ;
if ( vec1 - > size ! = vec2 - > size )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.MidpointVecs(): expects (2) vector objects of the same size \n " ) ;
for ( x = 0 ; x < vec1 - > size ; x + + ) {
vec [ x ] = 0.5f * ( vec1 - > vec [ x ] + vec2 - > vec [ x ] ) ;
2004-02-29 13:20:34 +00:00
}
2005-09-27 17:03:28 +00:00
return newVectorObject ( vec , vec1 - > size , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.ProjectVecs() -------------
//projects vector 1 onto vector 2
PyObject * M_Mathutils_ProjectVecs ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
VectorObject * vec1 = NULL , * vec2 = NULL ;
float vec [ 4 ] ;
double dot = 0.0f , dot2 = 0.0f ;
int x , size ;
if ( ! PyArg_ParseTuple ( args , " O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2 ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.ProjectVecs(): expects (2) vector objects of the same size \n " ) ;
if ( vec1 - > size ! = vec2 - > size )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.ProjectVecs(): expects (2) vector objects of the same size \n " ) ;
//since they are the same size...
size = vec1 - > size ;
//get dot products
for ( x = 0 ; x < size ; x + + ) {
2004-09-25 20:30:40 +00:00
dot + = vec1 - > vec [ x ] * vec2 - > vec [ x ] ;
dot2 + = vec2 - > vec [ x ] * vec2 - > vec [ x ] ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//projection
2004-09-25 20:30:40 +00:00
dot / = dot2 ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < size ; x + + ) {
vec [ x ] = ( float ) ( dot * vec2 - > vec [ x ] ) ;
2004-02-29 13:20:34 +00:00
}
2005-09-27 17:03:28 +00:00
return newVectorObject ( vec , size , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------MATRIX FUNCTIONS--------------------
//----------------------------------Mathutils.Matrix() -----------------
2004-02-29 13:20:34 +00:00
//mat is a 1D array of floats - row[0][0],row[0][1], row[1][0], etc.
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//create a new matrix type
PyObject * M_Mathutils_Matrix ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
PyObject * listObject = NULL ;
2005-09-27 17:03:28 +00:00
PyObject * argObject , * m , * s , * f ;
MatrixObject * mat ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
int argSize , seqSize = 0 , i , j ;
float matrix [ 16 ] = { 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f ,
0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f } ;
argSize = PySequence_Length ( args ) ;
if ( argSize > 4 ) { //bad arg nums
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Matrix(): expects 0-4 numeric sequences of the same size \n " ) ;
} else if ( argSize = = 0 ) { //return empty 4D matrix
return ( PyObject * ) newMatrixObject ( NULL , 4 , 4 , Py_NEW ) ;
} else if ( argSize = = 1 ) {
//copy constructor for matrix objects
argObject = PySequence_GetItem ( args , 0 ) ;
if ( MatrixObject_Check ( argObject ) ) {
mat = ( MatrixObject * ) argObject ;
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
argSize = mat - > rowSize ; //rows
2005-09-27 17:03:28 +00:00
seqSize = mat - > colSize ; //col
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( i = 0 ; i < ( seqSize * argSize ) ; i + + ) {
matrix [ i ] = mat - > contigPtr [ i ] ;
}
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
Py_DECREF ( argObject ) ;
} else { //2-4 arguments (all seqs? all same size?)
for ( i = 0 ; i < argSize ; i + + ) {
argObject = PySequence_GetItem ( args , i ) ;
if ( PySequence_Check ( argObject ) ) { //seq?
if ( seqSize ) { //0 at first
if ( PySequence_Length ( argObject ) ! = seqSize ) { //seq size not same
2005-09-27 17:03:28 +00:00
Py_DECREF ( argObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Matrix(): expects 0-4 numeric sequences of the same size \n " ) ;
2004-02-29 13:20:34 +00:00
}
2004-09-25 20:30:40 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
seqSize = PySequence_Length ( argObject ) ;
} else { //arg not a sequence
2005-09-27 17:03:28 +00:00
Py_XDECREF ( argObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Matrix(): expects 0-4 numeric sequences of the same size \n " ) ;
2004-02-29 13:20:34 +00:00
}
2005-09-27 17:03:28 +00:00
Py_DECREF ( argObject ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//all is well... let's continue parsing
2005-09-27 17:03:28 +00:00
listObject = args ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( i = 0 ; i < argSize ; i + + ) {
m = PySequence_GetItem ( listObject , i ) ;
if ( m = = NULL ) { // Failed to read sequence
return EXPP_ReturnPyObjError ( PyExc_RuntimeError ,
" Mathutils.Matrix(): failed to parse arguments... \n " ) ;
}
2005-09-27 17:03:28 +00:00
for ( j = 0 ; j < seqSize ; j + + ) {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
s = PySequence_GetItem ( m , j ) ;
if ( s = = NULL ) { // Failed to read sequence
Py_DECREF ( m ) ;
return EXPP_ReturnPyObjError ( PyExc_RuntimeError ,
" Mathutils.Matrix(): failed to parse arguments... \n " ) ;
}
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
f = PyNumber_Float ( s ) ;
if ( f = = NULL ) { // parsed item is not a number
EXPP_decr2 ( m , s ) ;
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Matrix(): expects 0-4 numeric sequences of the same size \n " ) ;
}
2005-09-27 17:03:28 +00:00
2005-07-18 03:50:37 +00:00
matrix [ ( seqSize * i ) + j ] = ( float ) PyFloat_AS_DOUBLE ( f ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
EXPP_decr2 ( f , s ) ;
}
Py_DECREF ( m ) ;
2005-05-22 17:40:00 +00:00
}
}
2005-09-27 17:03:28 +00:00
return newMatrixObject ( matrix , argSize , seqSize , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.RotationMatrix() ----------
2004-02-29 13:20:34 +00:00
//mat is a 1D array of floats - row[0][0],row[0][1], row[1][0], etc.
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//creates a rotation matrix
PyObject * M_Mathutils_RotationMatrix ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
2005-05-22 17:40:00 +00:00
VectorObject * vec = NULL ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
char * axis = NULL ;
2004-02-29 13:20:34 +00:00
int matSize ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
float angle = 0.0f , norm = 0.0f , cosAngle = 0.0f , sinAngle = 0.0f ;
float mat [ 16 ] = { 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f ,
0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f } ;
if ( ! PyArg_ParseTuple
( args , " fi|sO! " , & angle , & matSize , & axis , & vector_Type , & vec ) ) {
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.RotationMatrix(): expected float int and optional string and vector \n " ) ;
2004-02-29 13:20:34 +00:00
}
2006-08-08 14:34:38 +00:00
/* Clamp to -360:360 */
while ( angle < - 360.0f )
angle + = 360.0 ;
while ( angle > 360.0f )
angle - = 360.0 ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize ! = 2 & & matSize ! = 3 & & matSize ! = 4 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.RotationMatrix(): can only return a 2x2 3x3 or 4x4 matrix \n " ) ;
if ( matSize = = 2 & & ( axis ! = NULL | | vec ! = NULL ) )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.RotationMatrix(): cannot create a 2x2 rotation matrix around arbitrary axis \n " ) ;
if ( ( matSize = = 3 | | matSize = = 4 ) & & axis = = NULL )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.RotationMatrix(): please choose an axis of rotation for 3d and 4d matrices \n " ) ;
if ( axis ) {
if ( ( ( strcmp ( axis , " r " ) = = 0 ) | |
( strcmp ( axis , " R " ) = = 0 ) ) & & vec = = NULL )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.RotationMatrix(): please define the arbitrary axis of rotation \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( vec ) {
if ( vec - > size ! = 3 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.RotationMatrix(): the arbitrary axis must be a 3D vector \n " ) ;
2004-02-29 13:20:34 +00:00
}
//convert to radians
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
angle = angle * ( float ) ( Py_PI / 180 ) ;
if ( axis = = NULL & & matSize = = 2 ) {
2004-02-29 13:20:34 +00:00
//2D rotation matrix
2005-07-14 15:59:42 +00:00
mat [ 0 ] = ( float ) cos ( angle ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
mat [ 1 ] = ( float ) sin ( angle ) ;
mat [ 2 ] = - ( ( float ) sin ( angle ) ) ;
mat [ 3 ] = ( float ) cos ( angle ) ;
} else if ( ( strcmp ( axis , " x " ) = = 0 ) | | ( strcmp ( axis , " X " ) = = 0 ) ) {
2004-02-29 13:20:34 +00:00
//rotation around X
2004-09-25 20:30:40 +00:00
mat [ 0 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
mat [ 4 ] = ( float ) cos ( angle ) ;
mat [ 5 ] = ( float ) sin ( angle ) ;
mat [ 7 ] = - ( ( float ) sin ( angle ) ) ;
mat [ 8 ] = ( float ) cos ( angle ) ;
} else if ( ( strcmp ( axis , " y " ) = = 0 ) | | ( strcmp ( axis , " Y " ) = = 0 ) ) {
2004-02-29 13:20:34 +00:00
//rotation around Y
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
mat [ 0 ] = ( float ) cos ( angle ) ;
mat [ 2 ] = - ( ( float ) sin ( angle ) ) ;
2004-09-25 20:30:40 +00:00
mat [ 4 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
mat [ 6 ] = ( float ) sin ( angle ) ;
mat [ 8 ] = ( float ) cos ( angle ) ;
} else if ( ( strcmp ( axis , " z " ) = = 0 ) | | ( strcmp ( axis , " Z " ) = = 0 ) ) {
2004-02-29 13:20:34 +00:00
//rotation around Z
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
mat [ 0 ] = ( float ) cos ( angle ) ;
mat [ 1 ] = ( float ) sin ( angle ) ;
mat [ 3 ] = - ( ( float ) sin ( angle ) ) ;
mat [ 4 ] = ( float ) cos ( angle ) ;
2004-09-25 20:30:40 +00:00
mat [ 8 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( strcmp ( axis , " r " ) = = 0 ) | | ( strcmp ( axis , " R " ) = = 0 ) ) {
2004-02-29 13:20:34 +00:00
//arbitrary rotation
//normalize arbitrary axis
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
norm = ( float ) sqrt ( vec - > vec [ 0 ] * vec - > vec [ 0 ] +
2004-09-25 20:30:40 +00:00
vec - > vec [ 1 ] * vec - > vec [ 1 ] +
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
vec - > vec [ 2 ] * vec - > vec [ 2 ] ) ;
2004-09-25 20:30:40 +00:00
vec - > vec [ 0 ] / = norm ;
vec - > vec [ 1 ] / = norm ;
vec - > vec [ 2 ] / = norm ;
2008-06-22 18:12:46 +00:00
if ( isnan ( vec - > vec [ 0 ] ) | | isnan ( vec - > vec [ 1 ] ) | | isnan ( vec - > vec [ 2 ] ) ) {
/* zero length vector, return an identity matrix, could also return an error */
mat [ 0 ] = mat [ 4 ] = mat [ 8 ] = 1.0f ;
} else {
/* create matrix */
cosAngle = ( float ) cos ( angle ) ;
sinAngle = ( float ) sin ( angle ) ;
mat [ 0 ] = ( ( vec - > vec [ 0 ] * vec - > vec [ 0 ] ) * ( 1 - cosAngle ) ) +
cosAngle ;
mat [ 1 ] = ( ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) * ( 1 - cosAngle ) ) +
( vec - > vec [ 2 ] * sinAngle ) ;
mat [ 2 ] = ( ( vec - > vec [ 0 ] * vec - > vec [ 2 ] ) * ( 1 - cosAngle ) ) -
( vec - > vec [ 1 ] * sinAngle ) ;
mat [ 3 ] = ( ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) * ( 1 - cosAngle ) ) -
( vec - > vec [ 2 ] * sinAngle ) ;
mat [ 4 ] = ( ( vec - > vec [ 1 ] * vec - > vec [ 1 ] ) * ( 1 - cosAngle ) ) +
cosAngle ;
mat [ 5 ] = ( ( vec - > vec [ 1 ] * vec - > vec [ 2 ] ) * ( 1 - cosAngle ) ) +
( vec - > vec [ 0 ] * sinAngle ) ;
mat [ 6 ] = ( ( vec - > vec [ 0 ] * vec - > vec [ 2 ] ) * ( 1 - cosAngle ) ) +
( vec - > vec [ 1 ] * sinAngle ) ;
mat [ 7 ] = ( ( vec - > vec [ 1 ] * vec - > vec [ 2 ] ) * ( 1 - cosAngle ) ) -
( vec - > vec [ 0 ] * sinAngle ) ;
mat [ 8 ] = ( ( vec - > vec [ 2 ] * vec - > vec [ 2 ] ) * ( 1 - cosAngle ) ) +
cosAngle ;
}
2004-09-25 20:30:40 +00:00
} else {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.RotationMatrix(): unrecognizable axis of rotation type - expected x,y,z or r \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize = = 4 ) {
2004-02-29 13:20:34 +00:00
//resize matrix
2004-09-25 20:30:40 +00:00
mat [ 10 ] = mat [ 8 ] ;
mat [ 9 ] = mat [ 7 ] ;
mat [ 8 ] = mat [ 6 ] ;
mat [ 7 ] = 0.0f ;
mat [ 6 ] = mat [ 5 ] ;
mat [ 5 ] = mat [ 4 ] ;
mat [ 4 ] = mat [ 3 ] ;
2004-02-29 13:20:34 +00:00
mat [ 3 ] = 0.0f ;
}
//pass to matrix creation
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return newMatrixObject ( mat , matSize , matSize , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.TranslationMatrix() -------
//creates a translation matrix
2007-06-16 12:24:41 +00:00
PyObject * M_Mathutils_TranslationMatrix ( PyObject * self , VectorObject * vec )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
float mat [ 16 ] = { 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f ,
0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f } ;
2007-06-16 12:24:41 +00:00
if ( ! VectorObject_Check ( vec ) ) {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.TranslationMatrix(): expected vector \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( vec - > size ! = 3 & & vec - > size ! = 4 ) {
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.TranslationMatrix(): vector must be 3D or 4D \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//create a identity matrix and add translation
Mat4One ( ( float ( * ) [ 4 ] ) mat ) ;
2004-02-29 13:20:34 +00:00
mat [ 12 ] = vec - > vec [ 0 ] ;
mat [ 13 ] = vec - > vec [ 1 ] ;
mat [ 14 ] = vec - > vec [ 2 ] ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return newMatrixObject ( mat , 4 , 4 , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.ScaleMatrix() -------------
2004-02-29 13:20:34 +00:00
//mat is a 1D array of floats - row[0][0],row[0][1], row[1][0], etc.
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//creates a scaling matrix
PyObject * M_Mathutils_ScaleMatrix ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
VectorObject * vec = NULL ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
float norm = 0.0f , factor ;
int matSize , x ;
float mat [ 16 ] = { 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f ,
0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f } ;
if ( ! PyArg_ParseTuple
( args , " fi|O! " , & factor , & matSize , & vector_Type , & vec ) ) {
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.ScaleMatrix(): expected float int and optional vector \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize ! = 2 & & matSize ! = 3 & & matSize ! = 4 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.ScaleMatrix(): can only return a 2x2 3x3 or 4x4 matrix \n " ) ;
if ( vec ) {
if ( vec - > size > 2 & & matSize = = 2 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.ScaleMatrix(): please use 2D vectors when scaling in 2D \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( vec = = NULL ) { //scaling along axis
if ( matSize = = 2 ) {
2004-02-29 13:20:34 +00:00
mat [ 0 ] = factor ;
mat [ 3 ] = factor ;
2004-09-25 20:30:40 +00:00
} else {
2004-02-29 13:20:34 +00:00
mat [ 0 ] = factor ;
mat [ 4 ] = factor ;
mat [ 8 ] = factor ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else { //scaling in arbitrary direction
2004-02-29 13:20:34 +00:00
//normalize arbitrary axis
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < vec - > size ; x + + ) {
2004-02-29 13:20:34 +00:00
norm + = vec - > vec [ x ] * vec - > vec [ x ] ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
norm = ( float ) sqrt ( norm ) ;
for ( x = 0 ; x < vec - > size ; x + + ) {
2004-02-29 13:20:34 +00:00
vec - > vec [ x ] / = norm ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize = = 2 ) {
mat [ 0 ] = 1 + ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 0 ] ) ) ;
mat [ 1 ] = ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ) ;
mat [ 2 ] = ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ) ;
mat [ 3 ] = 1 + ( ( factor - 1 ) * ( vec - > vec [ 1 ] * vec - > vec [ 1 ] ) ) ;
2004-09-25 20:30:40 +00:00
} else {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
mat [ 0 ] = 1 + ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 0 ] ) ) ;
mat [ 1 ] = ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ) ;
mat [ 2 ] = ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 2 ] ) ) ;
mat [ 3 ] = ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ) ;
mat [ 4 ] = 1 + ( ( factor - 1 ) * ( vec - > vec [ 1 ] * vec - > vec [ 1 ] ) ) ;
mat [ 5 ] = ( ( factor - 1 ) * ( vec - > vec [ 1 ] * vec - > vec [ 2 ] ) ) ;
mat [ 6 ] = ( ( factor - 1 ) * ( vec - > vec [ 0 ] * vec - > vec [ 2 ] ) ) ;
mat [ 7 ] = ( ( factor - 1 ) * ( vec - > vec [ 1 ] * vec - > vec [ 2 ] ) ) ;
mat [ 8 ] = 1 + ( ( factor - 1 ) * ( vec - > vec [ 2 ] * vec - > vec [ 2 ] ) ) ;
2004-02-29 13:20:34 +00:00
}
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize = = 4 ) {
2004-02-29 13:20:34 +00:00
//resize matrix
2004-09-25 20:30:40 +00:00
mat [ 10 ] = mat [ 8 ] ;
mat [ 9 ] = mat [ 7 ] ;
mat [ 8 ] = mat [ 6 ] ;
mat [ 7 ] = 0.0f ;
mat [ 6 ] = mat [ 5 ] ;
mat [ 5 ] = mat [ 4 ] ;
mat [ 4 ] = mat [ 3 ] ;
mat [ 3 ] = 0.0f ;
2004-02-29 13:20:34 +00:00
}
//pass to matrix creation
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return newMatrixObject ( mat , matSize , matSize , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.OrthoProjectionMatrix() ---
2004-02-29 13:20:34 +00:00
//mat is a 1D array of floats - row[0][0],row[0][1], row[1][0], etc.
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//creates an ortho projection matrix
PyObject * M_Mathutils_OrthoProjectionMatrix ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
2005-05-22 17:40:00 +00:00
VectorObject * vec = NULL ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
char * plane ;
int matSize , x ;
2004-02-29 13:20:34 +00:00
float norm = 0.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
float mat [ 16 ] = { 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f ,
0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f } ;
if ( ! PyArg_ParseTuple
( args , " si|O! " , & plane , & matSize , & vector_Type , & vec ) ) {
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.OrthoProjectionMatrix(): expected string and int and optional vector \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize ! = 2 & & matSize ! = 3 & & matSize ! = 4 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.OrthoProjectionMatrix(): can only return a 2x2 3x3 or 4x4 matrix \n " ) ;
if ( vec ) {
if ( vec - > size > 2 & & matSize = = 2 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.OrthoProjectionMatrix(): please use 2D vectors when scaling in 2D \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( vec = = NULL ) { //ortho projection onto cardinal plane
if ( ( ( strcmp ( plane , " x " ) = = 0 )
| | ( strcmp ( plane , " X " ) = = 0 ) ) & & matSize = = 2 ) {
2004-02-29 13:20:34 +00:00
mat [ 0 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " y " ) = = 0 )
| | ( strcmp ( plane , " Y " ) = = 0 ) )
& & matSize = = 2 ) {
2004-02-29 13:20:34 +00:00
mat [ 3 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " xy " ) = = 0 )
| | ( strcmp ( plane , " XY " ) = = 0 ) )
& & matSize > 2 ) {
2004-02-29 13:20:34 +00:00
mat [ 0 ] = 1.0f ;
mat [ 4 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " xz " ) = = 0 )
| | ( strcmp ( plane , " XZ " ) = = 0 ) )
& & matSize > 2 ) {
2004-02-29 13:20:34 +00:00
mat [ 0 ] = 1.0f ;
mat [ 8 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " yz " ) = = 0 )
| | ( strcmp ( plane , " YZ " ) = = 0 ) )
& & matSize > 2 ) {
2004-02-29 13:20:34 +00:00
mat [ 4 ] = 1.0f ;
mat [ 8 ] = 1.0f ;
2004-09-25 20:30:40 +00:00
} else {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.OrthoProjectionMatrix(): unknown plane - expected: x, y, xy, xz, yz \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else { //arbitrary plane
2004-02-29 13:20:34 +00:00
//normalize arbitrary axis
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < vec - > size ; x + + ) {
2004-02-29 13:20:34 +00:00
norm + = vec - > vec [ x ] * vec - > vec [ x ] ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
norm = ( float ) sqrt ( norm ) ;
for ( x = 0 ; x < vec - > size ; x + + ) {
2004-02-29 13:20:34 +00:00
vec - > vec [ x ] / = norm ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ( ( strcmp ( plane , " r " ) = = 0 )
| | ( strcmp ( plane , " R " ) = = 0 ) ) & & matSize = = 2 ) {
mat [ 0 ] = 1 - ( vec - > vec [ 0 ] * vec - > vec [ 0 ] ) ;
mat [ 1 ] = - ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ;
mat [ 2 ] = - ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ;
mat [ 3 ] = 1 - ( vec - > vec [ 1 ] * vec - > vec [ 1 ] ) ;
} else if ( ( ( strcmp ( plane , " r " ) = = 0 )
| | ( strcmp ( plane , " R " ) = = 0 ) )
& & matSize > 2 ) {
mat [ 0 ] = 1 - ( vec - > vec [ 0 ] * vec - > vec [ 0 ] ) ;
mat [ 1 ] = - ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ;
mat [ 2 ] = - ( vec - > vec [ 0 ] * vec - > vec [ 2 ] ) ;
mat [ 3 ] = - ( vec - > vec [ 0 ] * vec - > vec [ 1 ] ) ;
mat [ 4 ] = 1 - ( vec - > vec [ 1 ] * vec - > vec [ 1 ] ) ;
mat [ 5 ] = - ( vec - > vec [ 1 ] * vec - > vec [ 2 ] ) ;
mat [ 6 ] = - ( vec - > vec [ 0 ] * vec - > vec [ 2 ] ) ;
mat [ 7 ] = - ( vec - > vec [ 1 ] * vec - > vec [ 2 ] ) ;
mat [ 8 ] = 1 - ( vec - > vec [ 2 ] * vec - > vec [ 2 ] ) ;
2004-09-25 20:30:40 +00:00
} else {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.OrthoProjectionMatrix(): unknown plane - expected: 'r' expected for axis designation \n " ) ;
2004-02-29 13:20:34 +00:00
}
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize = = 4 ) {
2004-02-29 13:20:34 +00:00
//resize matrix
2004-09-25 20:30:40 +00:00
mat [ 10 ] = mat [ 8 ] ;
mat [ 9 ] = mat [ 7 ] ;
mat [ 8 ] = mat [ 6 ] ;
mat [ 7 ] = 0.0f ;
mat [ 6 ] = mat [ 5 ] ;
mat [ 5 ] = mat [ 4 ] ;
mat [ 4 ] = mat [ 3 ] ;
2004-02-29 13:20:34 +00:00
mat [ 3 ] = 0.0f ;
}
//pass to matrix creation
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return newMatrixObject ( mat , matSize , matSize , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.ShearMatrix() -------------
//creates a shear matrix
PyObject * M_Mathutils_ShearMatrix ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
int matSize ;
char * plane ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
float factor ;
float mat [ 16 ] = { 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f ,
0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f } ;
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " sfi " , & plane , & factor , & matSize ) ) {
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.ShearMatrix(): expected string float and int \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize ! = 2 & & matSize ! = 3 & & matSize ! = 4 )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.ShearMatrix(): can only return a 2x2 3x3 or 4x4 matrix \n " ) ;
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ( ( strcmp ( plane , " x " ) = = 0 ) | | ( strcmp ( plane , " X " ) = = 0 ) )
& & matSize = = 2 ) {
2004-09-25 20:30:40 +00:00
mat [ 0 ] = 1.0f ;
mat [ 2 ] = factor ;
mat [ 3 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " y " ) = = 0 )
| | ( strcmp ( plane , " Y " ) = = 0 ) ) & & matSize = = 2 ) {
2004-09-25 20:30:40 +00:00
mat [ 0 ] = 1.0f ;
mat [ 1 ] = factor ;
mat [ 3 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " xy " ) = = 0 )
| | ( strcmp ( plane , " XY " ) = = 0 ) ) & & matSize > 2 ) {
2004-09-25 20:30:40 +00:00
mat [ 0 ] = 1.0f ;
mat [ 4 ] = 1.0f ;
mat [ 6 ] = factor ;
mat [ 7 ] = factor ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " xz " ) = = 0 )
| | ( strcmp ( plane , " XZ " ) = = 0 ) ) & & matSize > 2 ) {
2004-09-25 20:30:40 +00:00
mat [ 0 ] = 1.0f ;
mat [ 3 ] = factor ;
mat [ 4 ] = 1.0f ;
mat [ 5 ] = factor ;
mat [ 8 ] = 1.0f ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( ( ( strcmp ( plane , " yz " ) = = 0 )
| | ( strcmp ( plane , " YZ " ) = = 0 ) ) & & matSize > 2 ) {
2004-09-25 20:30:40 +00:00
mat [ 0 ] = 1.0f ;
mat [ 1 ] = factor ;
mat [ 2 ] = factor ;
mat [ 4 ] = 1.0f ;
2004-02-29 13:20:34 +00:00
mat [ 8 ] = 1.0f ;
2004-09-25 20:30:40 +00:00
} else {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.ShearMatrix(): expected: x, y, xy, xz, yz or wrong matrix size for shearing plane \n " ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( matSize = = 4 ) {
2004-02-29 13:20:34 +00:00
//resize matrix
2004-09-25 20:30:40 +00:00
mat [ 10 ] = mat [ 8 ] ;
mat [ 9 ] = mat [ 7 ] ;
mat [ 8 ] = mat [ 6 ] ;
mat [ 7 ] = 0.0f ;
mat [ 6 ] = mat [ 5 ] ;
mat [ 5 ] = mat [ 4 ] ;
mat [ 4 ] = mat [ 3 ] ;
2004-02-29 13:20:34 +00:00
mat [ 3 ] = 0.0f ;
}
//pass to matrix creation
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return newMatrixObject ( mat , matSize , matSize , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------QUATERNION FUNCTIONS-----------------
//----------------------------------Mathutils.Quaternion() --------------
PyObject * M_Mathutils_Quaternion ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
PyObject * listObject = NULL , * n , * q , * f ;
int size , i ;
float quat [ 4 ] ;
double norm = 0.0f , angle = 0.0f ;
size = PySequence_Length ( args ) ;
if ( size = = 1 | | size = = 2 ) { //seq?
listObject = PySequence_GetItem ( args , 0 ) ;
if ( PySequence_Check ( listObject ) ) {
size = PySequence_Length ( listObject ) ;
if ( ( size = = 4 & & PySequence_Length ( args ) ! = 1 ) | |
( size = = 3 & & PySequence_Length ( args ) ! = 2 ) | | ( size > 4 | | size < 3 ) ) {
// invalid args/size
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
if ( size = = 3 ) { //get angle in axis/angle
n = PyNumber_Float ( PySequence_GetItem ( args , 1 ) ) ;
if ( n = = NULL ) { // parsed item not a number or getItem fail
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
angle = PyFloat_AS_DOUBLE ( n ) ;
Py_DECREF ( n ) ;
}
} else {
listObject = PySequence_GetItem ( args , 1 ) ;
2007-05-28 21:53:49 +00:00
if ( size > 1 & & PySequence_Check ( listObject ) ) {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
size = PySequence_Length ( listObject ) ;
if ( size ! = 3 ) {
// invalid args/size
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
n = PyNumber_Float ( PySequence_GetItem ( args , 0 ) ) ;
if ( n = = NULL ) { // parsed item not a number or getItem fail
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
angle = PyFloat_AS_DOUBLE ( n ) ;
Py_DECREF ( n ) ;
} else { // argument was not a sequence
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else if ( size = = 0 ) { //returns a new empty quat
2005-09-27 17:03:28 +00:00
return newQuaternionObject ( NULL , Py_NEW ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else {
listObject = EXPP_incr_ret ( args ) ;
2004-02-29 13:20:34 +00:00
}
2005-05-22 17:40:00 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( size = = 3 ) { // invalid quat size
if ( PySequence_Length ( args ) ! = 2 ) {
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
} else {
if ( size ! = 4 ) {
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
}
2004-02-29 13:20:34 +00:00
}
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( i = 0 ; i < size ; i + + ) { //parse
q = PySequence_GetItem ( listObject , i ) ;
if ( q = = NULL ) { // Failed to read sequence
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_RuntimeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
f = PyNumber_Float ( q ) ;
if ( f = = NULL ) { // parsed item not a number
2005-09-27 17:03:28 +00:00
EXPP_decr2 ( q , listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Quaternion(): 4d numeric sequence expected or 3d vector and number \n " ) ;
}
2005-09-27 17:03:28 +00:00
2005-07-18 03:50:37 +00:00
quat [ i ] = ( float ) PyFloat_AS_DOUBLE ( f ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
EXPP_decr2 ( f , q ) ;
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( size = = 3 ) { //calculate the quat based on axis/angle
norm = sqrt ( quat [ 0 ] * quat [ 0 ] + quat [ 1 ] * quat [ 1 ] + quat [ 2 ] * quat [ 2 ] ) ;
2005-07-18 03:50:37 +00:00
quat [ 0 ] / = ( float ) norm ;
quat [ 1 ] / = ( float ) norm ;
quat [ 2 ] / = ( float ) norm ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
angle = angle * ( Py_PI / 180 ) ;
quat [ 3 ] = ( float ) ( sin ( angle / 2.0f ) ) * quat [ 2 ] ;
quat [ 2 ] = ( float ) ( sin ( angle / 2.0f ) ) * quat [ 1 ] ;
quat [ 1 ] = ( float ) ( sin ( angle / 2.0f ) ) * quat [ 0 ] ;
quat [ 0 ] = ( float ) ( cos ( angle / 2.0f ) ) ;
}
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
Py_DECREF ( listObject ) ;
2005-09-27 17:03:28 +00:00
return newQuaternionObject ( quat , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.CrossQuats() ----------------
//quaternion multiplication - associate not commutative
PyObject * M_Mathutils_CrossQuats ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
QuaternionObject * quatU = NULL , * quatV = NULL ;
float quat [ 4 ] ;
2005-05-22 17:40:00 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " O!O! " , & quaternion_Type , & quatU ,
& quaternion_Type , & quatV ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError , " Mathutils.CrossQuats() : expected Quaternion types " ) ;
QuatMul ( quat , quatU - > quat , quatV - > quat ) ;
2005-05-22 17:40:00 +00:00
2005-09-27 17:03:28 +00:00
return newQuaternionObject ( quat , Py_NEW ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
}
//----------------------------------Mathutils.DotQuats() ----------------
//returns the dot product of 2 quaternions
PyObject * M_Mathutils_DotQuats ( PyObject * self , PyObject * args )
2005-05-22 17:40:00 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
QuaternionObject * quatU = NULL , * quatV = NULL ;
double dot = 0.0f ;
2004-02-29 13:20:34 +00:00
int x ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " O!O! " , & quaternion_Type , & quatU ,
& quaternion_Type , & quatV ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError , " Mathutils.DotQuats() : expected Quaternion types " ) ;
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < 4 ; x + + ) {
2004-02-29 13:20:34 +00:00
dot + = quatU - > quat [ x ] * quatV - > quat [ x ] ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return PyFloat_FromDouble ( dot ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.DifferenceQuats() ---------
//returns the difference between 2 quaternions
PyObject * M_Mathutils_DifferenceQuats ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
QuaternionObject * quatU = NULL , * quatV = NULL ;
float quat [ 4 ] , tempQuat [ 4 ] ;
double dot = 0.0f ;
2004-02-29 13:20:34 +00:00
int x ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " O!O! " , & quaternion_Type ,
& quatU , & quaternion_Type , & quatV ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError , " Mathutils.DifferenceQuats() : expected Quaternion types " ) ;
2004-02-29 13:20:34 +00:00
tempQuat [ 0 ] = quatU - > quat [ 0 ] ;
tempQuat [ 1 ] = - quatU - > quat [ 1 ] ;
tempQuat [ 2 ] = - quatU - > quat [ 2 ] ;
tempQuat [ 3 ] = - quatU - > quat [ 3 ] ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
dot = sqrt ( tempQuat [ 0 ] * tempQuat [ 0 ] + tempQuat [ 1 ] * tempQuat [ 1 ] +
tempQuat [ 2 ] * tempQuat [ 2 ] + tempQuat [ 3 ] * tempQuat [ 3 ] ) ;
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
for ( x = 0 ; x < 4 ; x + + ) {
2005-07-18 03:50:37 +00:00
tempQuat [ x ] / = ( float ) ( dot * dot ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
QuatMul ( quat , tempQuat , quatV - > quat ) ;
2005-09-27 17:03:28 +00:00
return newQuaternionObject ( quat , Py_NEW ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.Slerp() ------------------
//attemps to interpolate 2 quaternions and return the result
PyObject * M_Mathutils_Slerp ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
QuaternionObject * quatU = NULL , * quatV = NULL ;
float quat [ 4 ] , quat_u [ 4 ] , quat_v [ 4 ] , param ;
double x , y , dot , sinT , angle , IsinT ;
int z ;
if ( ! PyArg_ParseTuple ( args , " O!O!f " , & quaternion_Type ,
& quatU , & quaternion_Type , & quatV , & param ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Slerp(): expected Quaternion types and float " ) ;
if ( param > 1.0f | | param < 0.0f )
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Slerp(): interpolation factor must be between 0.0 and 1.0 " ) ;
//copy quats
for ( z = 0 ; z < 4 ; z + + ) {
quat_u [ z ] = quatU - > quat [ z ] ;
quat_v [ z ] = quatV - > quat [ z ] ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//dot product
dot = quat_u [ 0 ] * quat_v [ 0 ] + quat_u [ 1 ] * quat_v [ 1 ] +
quat_u [ 2 ] * quat_v [ 2 ] + quat_u [ 3 ] * quat_v [ 3 ] ;
//if negative negate a quat (shortest arc)
if ( dot < 0.0f ) {
quat_v [ 0 ] = - quat_v [ 0 ] ;
quat_v [ 1 ] = - quat_v [ 1 ] ;
quat_v [ 2 ] = - quat_v [ 2 ] ;
quat_v [ 3 ] = - quat_v [ 3 ] ;
dot = - dot ;
}
if ( dot > .99999f ) { //very close
2004-02-29 13:20:34 +00:00
x = 1.0f - param ;
y = param ;
2004-09-25 20:30:40 +00:00
} else {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//calculate sin of angle
sinT = sqrt ( 1.0f - ( dot * dot ) ) ;
//calculate angle
angle = atan2 ( sinT , dot ) ;
//caluculate inverse of sin(theta)
IsinT = 1.0f / sinT ;
x = sin ( ( 1.0f - param ) * angle ) * IsinT ;
y = sin ( param * angle ) * IsinT ;
2005-05-22 17:40:00 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//interpolate
2005-07-18 03:50:37 +00:00
quat [ 0 ] = ( float ) ( quat_u [ 0 ] * x + quat_v [ 0 ] * y ) ;
quat [ 1 ] = ( float ) ( quat_u [ 1 ] * x + quat_v [ 1 ] * y ) ;
quat [ 2 ] = ( float ) ( quat_u [ 2 ] * x + quat_v [ 2 ] * y ) ;
quat [ 3 ] = ( float ) ( quat_u [ 3 ] * x + quat_v [ 3 ] * y ) ;
2005-05-22 17:40:00 +00:00
2005-09-27 17:03:28 +00:00
return newQuaternionObject ( quat , Py_NEW ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
}
//----------------------------------EULER FUNCTIONS----------------------
//----------------------------------Mathutils.Euler() -------------------
//makes a new euler for you to play with
PyObject * M_Mathutils_Euler ( PyObject * self , PyObject * args )
2004-02-29 13:20:34 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
PyObject * listObject = NULL ;
int size , i ;
float eul [ 3 ] ;
2005-09-27 17:03:28 +00:00
PyObject * e , * f ;
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
size = PySequence_Length ( args ) ;
if ( size = = 1 ) {
listObject = PySequence_GetItem ( args , 0 ) ;
if ( PySequence_Check ( listObject ) ) {
size = PySequence_Length ( listObject ) ;
} else { // Single argument was not a sequence
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Euler(): 3d numeric sequence expected \n " ) ;
}
} else if ( size = = 0 ) {
//returns a new empty 3d euler
2005-09-27 17:03:28 +00:00
return newEulerObject ( NULL , Py_NEW ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
} else {
listObject = EXPP_incr_ret ( args ) ;
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
}
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( size ! = 3 ) { // Invalid euler size
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Euler(): 3d numeric sequence expected \n " ) ;
}
2005-05-22 17:40:00 +00:00
2005-09-27 17:03:28 +00:00
for ( i = 0 ; i < size ; i + + ) {
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
e = PySequence_GetItem ( listObject , i ) ;
if ( e = = NULL ) { // Failed to read sequence
2005-09-27 17:03:28 +00:00
Py_DECREF ( listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_RuntimeError ,
" Mathutils.Euler(): 3d numeric sequence expected \n " ) ;
}
2005-09-27 17:03:28 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
f = PyNumber_Float ( e ) ;
if ( f = = NULL ) { // parsed item not a number
2005-09-27 17:03:28 +00:00
EXPP_decr2 ( e , listObject ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Euler(): 3d numeric sequence expected \n " ) ;
}
2005-09-27 17:03:28 +00:00
2005-07-18 03:50:37 +00:00
eul [ i ] = ( float ) PyFloat_AS_DOUBLE ( f ) ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
EXPP_decr2 ( f , e ) ;
}
Py_DECREF ( listObject ) ;
2005-09-27 17:03:28 +00:00
return newEulerObject ( eul , Py_NEW ) ;
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
}
2005-11-15 21:45:05 +00:00
//----------------------------------POINT FUNCTIONS---------------------
//----------------------------------Mathutils.Point() ------------------
PyObject * M_Mathutils_Point ( PyObject * self , PyObject * args )
{
PyObject * listObject = NULL ;
int size , i ;
float point [ 3 ] ;
PyObject * v , * f ;
size = PySequence_Length ( args ) ;
if ( size = = 1 ) {
listObject = PySequence_GetItem ( args , 0 ) ;
if ( PySequence_Check ( listObject ) ) {
size = PySequence_Length ( listObject ) ;
} else { // Single argument was not a sequence
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Point(): 2-3 floats or ints expected (optionally in a sequence) \n " ) ;
}
} else if ( size = = 0 ) {
//returns a new empty 3d point
return newPointObject ( NULL , 3 , Py_NEW ) ;
} else {
listObject = EXPP_incr_ret ( args ) ;
}
if ( size < 2 | | size > 3 ) { // Invalid vector size
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_AttributeError ,
" Mathutils.Point(): 2-3 floats or ints expected (optionally in a sequence) \n " ) ;
}
for ( i = 0 ; i < size ; i + + ) {
v = PySequence_GetItem ( listObject , i ) ;
if ( v = = NULL ) { // Failed to read sequence
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_RuntimeError ,
" Mathutils.Point(): 2-3 floats or ints expected (optionally in a sequence) \n " ) ;
}
f = PyNumber_Float ( v ) ;
if ( f = = NULL ) { // parsed item not a number
Py_DECREF ( v ) ;
Py_XDECREF ( listObject ) ;
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.Point(): 2-3 floats or ints expected (optionally in a sequence) \n " ) ;
}
point [ i ] = ( float ) PyFloat_AS_DOUBLE ( f ) ;
EXPP_decr2 ( f , v ) ;
}
Py_DECREF ( listObject ) ;
return newPointObject ( point , size , Py_NEW ) ;
}
2005-09-26 15:06:28 +00:00
//---------------------------------INTERSECTION FUNCTIONS--------------------
//----------------------------------Mathutils.Intersect() -------------------
PyObject * M_Mathutils_Intersect ( PyObject * self , PyObject * args )
{
2005-09-27 17:03:28 +00:00
VectorObject * ray , * ray_off , * vec1 , * vec2 , * vec3 ;
2005-09-26 15:06:28 +00:00
float dir [ 3 ] , orig [ 3 ] , v1 [ 3 ] , v2 [ 3 ] , v3 [ 3 ] , e1 [ 3 ] , e2 [ 3 ] , pvec [ 3 ] , tvec [ 3 ] , qvec [ 3 ] ;
float det , inv_det , u , v , t ;
int clip = 1 ;
if ( ! PyArg_ParseTuple
( args , " O!O!O!O!O!|i " , & vector_Type , & vec1 , & vector_Type , & vec2
, & vector_Type , & vec3 , & vector_Type , & ray , & vector_Type , & ray_off , & clip ) )
return ( EXPP_ReturnPyObjError
( PyExc_TypeError , " expected 5 vector types \n " ) ) ;
if ( vec1 - > size ! = 3 | | vec2 - > size ! = 3 | | vec3 - > size ! = 3 | |
ray - > size ! = 3 | | ray_off - > size ! = 3 )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" only 3D vectors for all parameters \n " ) ) ;
VECCOPY ( v1 , vec1 - > vec ) ;
VECCOPY ( v2 , vec2 - > vec ) ;
VECCOPY ( v3 , vec3 - > vec ) ;
VECCOPY ( dir , ray - > vec ) ;
2007-04-04 13:18:41 +00:00
Normalize ( dir ) ;
2005-09-26 15:06:28 +00:00
VECCOPY ( orig , ray_off - > vec ) ;
/* find vectors for two edges sharing v1 */
VecSubf ( e1 , v2 , v1 ) ;
VecSubf ( e2 , v3 , v1 ) ;
/* begin calculating determinant - also used to calculated U parameter */
Crossf ( pvec , dir , e2 ) ;
/* if determinant is near zero, ray lies in plane of triangle */
det = Inpf ( e1 , pvec ) ;
if ( det > - 0.000001 & & det < 0.000001 ) {
return EXPP_incr_ret ( Py_None ) ;
}
inv_det = 1.0f / det ;
/* calculate distance from v1 to ray origin */
VecSubf ( tvec , orig , v1 ) ;
/* calculate U parameter and test bounds */
u = Inpf ( tvec , pvec ) * inv_det ;
if ( clip & & ( u < 0.0f | | u > 1.0f ) ) {
return EXPP_incr_ret ( Py_None ) ;
}
/* prepare to test the V parameter */
Crossf ( qvec , tvec , e1 ) ;
/* calculate V parameter and test bounds */
v = Inpf ( dir , qvec ) * inv_det ;
if ( clip & & ( v < 0.0f | | u + v > 1.0f ) ) {
return EXPP_incr_ret ( Py_None ) ;
}
/* calculate t, ray intersects triangle */
t = Inpf ( e2 , qvec ) * inv_det ;
VecMulf ( dir , t ) ;
VecAddf ( pvec , orig , dir ) ;
return newVectorObject ( pvec , 3 , Py_NEW ) ;
}
//----------------------------------Mathutils.LineIntersect() -------------------
/* Line-Line intersection using algorithm from mathworld.wolfram.com */
PyObject * M_Mathutils_LineIntersect ( PyObject * self , PyObject * args )
{
PyObject * tuple ;
2005-09-27 17:03:28 +00:00
VectorObject * vec1 , * vec2 , * vec3 , * vec4 ;
2005-09-26 15:06:28 +00:00
float v1 [ 3 ] , v2 [ 3 ] , v3 [ 3 ] , v4 [ 3 ] , i1 [ 3 ] , i2 [ 3 ] ;
if ( ! PyArg_ParseTuple
( args , " O!O!O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2
, & vector_Type , & vec3 , & vector_Type , & vec4 ) )
return ( EXPP_ReturnPyObjError
( PyExc_TypeError , " expected 4 vector types \n " ) ) ;
if ( vec1 - > size ! = vec2 - > size | | vec1 - > size ! = vec3 - > size | | vec1 - > size ! = vec2 - > size )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" vectors must be of the same size \n " ) ) ;
if ( vec1 - > size = = 3 | | vec1 - > size = = 2 ) {
2008-06-24 05:02:01 +00:00
int result ;
2005-09-26 15:06:28 +00:00
if ( vec1 - > size = = 3 ) {
VECCOPY ( v1 , vec1 - > vec ) ;
VECCOPY ( v2 , vec2 - > vec ) ;
VECCOPY ( v3 , vec3 - > vec ) ;
VECCOPY ( v4 , vec4 - > vec ) ;
}
else {
v1 [ 0 ] = vec1 - > vec [ 0 ] ;
v1 [ 1 ] = vec1 - > vec [ 1 ] ;
v1 [ 2 ] = 0.0f ;
v2 [ 0 ] = vec2 - > vec [ 0 ] ;
v2 [ 1 ] = vec2 - > vec [ 1 ] ;
v2 [ 2 ] = 0.0f ;
v3 [ 0 ] = vec3 - > vec [ 0 ] ;
v3 [ 1 ] = vec3 - > vec [ 1 ] ;
v3 [ 2 ] = 0.0f ;
v4 [ 0 ] = vec4 - > vec [ 0 ] ;
v4 [ 1 ] = vec4 - > vec [ 1 ] ;
v4 [ 2 ] = 0.0f ;
}
2008-06-24 05:02:01 +00:00
result = LineIntersectLine ( v1 , v2 , v3 , v4 , i1 , i2 ) ;
2005-09-26 15:06:28 +00:00
2008-06-24 05:02:01 +00:00
if ( result = = 0 ) {
2005-09-26 15:06:28 +00:00
/* colinear */
return EXPP_incr_ret ( Py_None ) ;
}
else {
2008-06-24 05:02:01 +00:00
tuple = PyTuple_New ( 2 ) ;
PyTuple_SetItem ( tuple , 0 , newVectorObject ( i1 , vec1 - > size , Py_NEW ) ) ;
PyTuple_SetItem ( tuple , 1 , newVectorObject ( i2 , vec1 - > size , Py_NEW ) ) ;
return tuple ;
2005-09-26 15:06:28 +00:00
}
}
else {
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" 2D/3D vectors only \n " ) ) ;
}
}
2006-07-01 15:29:04 +00:00
2005-09-26 15:06:28 +00:00
//---------------------------------NORMALS FUNCTIONS--------------------
//----------------------------------Mathutils.QuadNormal() -------------------
PyObject * M_Mathutils_QuadNormal ( PyObject * self , PyObject * args )
{
VectorObject * vec1 ;
VectorObject * vec2 ;
VectorObject * vec3 ;
VectorObject * vec4 ;
float v1 [ 3 ] , v2 [ 3 ] , v3 [ 3 ] , v4 [ 3 ] , e1 [ 3 ] , e2 [ 3 ] , n1 [ 3 ] , n2 [ 3 ] ;
if ( ! PyArg_ParseTuple
( args , " O!O!O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2
, & vector_Type , & vec3 , & vector_Type , & vec4 ) )
return ( EXPP_ReturnPyObjError
( PyExc_TypeError , " expected 4 vector types \n " ) ) ;
if ( vec1 - > size ! = vec2 - > size | | vec1 - > size ! = vec3 - > size | | vec1 - > size ! = vec4 - > size )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" vectors must be of the same size \n " ) ) ;
if ( vec1 - > size ! = 3 )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" only 3D vectors \n " ) ) ;
VECCOPY ( v1 , vec1 - > vec ) ;
VECCOPY ( v2 , vec2 - > vec ) ;
VECCOPY ( v3 , vec3 - > vec ) ;
VECCOPY ( v4 , vec4 - > vec ) ;
/* find vectors for two edges sharing v2 */
VecSubf ( e1 , v1 , v2 ) ;
VecSubf ( e2 , v3 , v2 ) ;
Crossf ( n1 , e2 , e1 ) ;
2007-04-04 13:18:41 +00:00
Normalize ( n1 ) ;
2005-09-26 15:06:28 +00:00
/* find vectors for two edges sharing v4 */
VecSubf ( e1 , v3 , v4 ) ;
VecSubf ( e2 , v1 , v4 ) ;
Crossf ( n2 , e2 , e1 ) ;
2007-04-04 13:18:41 +00:00
Normalize ( n2 ) ;
2005-09-26 15:06:28 +00:00
/* adding and averaging the normals of both triangles */
VecAddf ( n1 , n2 , n1 ) ;
2007-04-04 13:18:41 +00:00
Normalize ( n1 ) ;
2005-09-26 15:06:28 +00:00
return newVectorObject ( n1 , 3 , Py_NEW ) ;
}
//----------------------------Mathutils.TriangleNormal() -------------------
PyObject * M_Mathutils_TriangleNormal ( PyObject * self , PyObject * args )
{
2005-09-27 17:03:28 +00:00
VectorObject * vec1 , * vec2 , * vec3 ;
2005-09-26 15:06:28 +00:00
float v1 [ 3 ] , v2 [ 3 ] , v3 [ 3 ] , e1 [ 3 ] , e2 [ 3 ] , n [ 3 ] ;
if ( ! PyArg_ParseTuple
( args , " O!O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2
, & vector_Type , & vec3 ) )
return ( EXPP_ReturnPyObjError
( PyExc_TypeError , " expected 3 vector types \n " ) ) ;
if ( vec1 - > size ! = vec2 - > size | | vec1 - > size ! = vec3 - > size )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" vectors must be of the same size \n " ) ) ;
if ( vec1 - > size ! = 3 )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" only 3D vectors \n " ) ) ;
VECCOPY ( v1 , vec1 - > vec ) ;
VECCOPY ( v2 , vec2 - > vec ) ;
VECCOPY ( v3 , vec3 - > vec ) ;
/* find vectors for two edges sharing v2 */
VecSubf ( e1 , v1 , v2 ) ;
VecSubf ( e2 , v3 , v2 ) ;
Crossf ( n , e2 , e1 ) ;
2007-04-04 13:18:41 +00:00
Normalize ( n ) ;
2005-09-26 15:06:28 +00:00
return newVectorObject ( n , 3 , Py_NEW ) ;
}
//--------------------------------- AREA FUNCTIONS--------------------
//----------------------------------Mathutils.TriangleArea() -------------------
PyObject * M_Mathutils_TriangleArea ( PyObject * self , PyObject * args )
{
2005-09-27 17:03:28 +00:00
VectorObject * vec1 , * vec2 , * vec3 ;
2005-09-26 15:06:28 +00:00
float v1 [ 3 ] , v2 [ 3 ] , v3 [ 3 ] ;
if ( ! PyArg_ParseTuple
( args , " O!O!O! " , & vector_Type , & vec1 , & vector_Type , & vec2
, & vector_Type , & vec3 ) )
return ( EXPP_ReturnPyObjError
( PyExc_TypeError , " expected 3 vector types \n " ) ) ;
if ( vec1 - > size ! = vec2 - > size | | vec1 - > size ! = vec3 - > size )
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" vectors must be of the same size \n " ) ) ;
if ( vec1 - > size = = 3 ) {
VECCOPY ( v1 , vec1 - > vec ) ;
VECCOPY ( v2 , vec2 - > vec ) ;
VECCOPY ( v3 , vec3 - > vec ) ;
return PyFloat_FromDouble ( AreaT3Dfl ( v1 , v2 , v3 ) ) ;
}
else if ( vec1 - > size = = 2 ) {
v1 [ 0 ] = vec1 - > vec [ 0 ] ;
v1 [ 1 ] = vec1 - > vec [ 1 ] ;
v2 [ 0 ] = vec2 - > vec [ 0 ] ;
v2 [ 1 ] = vec2 - > vec [ 1 ] ;
v3 [ 0 ] = vec3 - > vec [ 0 ] ;
v3 [ 1 ] = vec3 - > vec [ 1 ] ;
return PyFloat_FromDouble ( AreaF2Dfl ( v1 , v2 , v3 ) ) ;
}
else {
return ( EXPP_ReturnPyObjError ( PyExc_TypeError ,
" only 2D,3D vectors are supported \n " ) ) ;
}
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//#############################DEPRECATED################################
//#######################################################################
//----------------------------------Mathutils.CopyMat() -----------------
//copies a matrix into a new matrix
PyObject * M_Mathutils_CopyMat ( PyObject * self , PyObject * args )
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
PyObject * matrix = NULL ;
2007-03-31 15:31:37 +00:00
static char warning = 1 ;
if ( warning ) {
printf ( " Mathutils.CopyMat(): deprecated :use Mathutils.Matrix() to copy matrices \n " ) ;
- - warning ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
matrix = M_Mathutils_Matrix ( self , args ) ;
if ( matrix = = NULL )
return NULL ; //error string already set if we get here
else
return matrix ;
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.CopyVec() -----------------
//makes a new vector that is a copy of the input
PyObject * M_Mathutils_CopyVec ( PyObject * self , PyObject * args )
{
PyObject * vec = NULL ;
2007-03-31 15:31:37 +00:00
static char warning = 1 ;
if ( warning ) {
printf ( " Mathutils.CopyVec(): Deprecated: use Mathutils.Vector() to copy vectors \n " ) ;
- - warning ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
vec = M_Mathutils_Vector ( self , args ) ;
if ( vec = = NULL )
return NULL ; //error string already set if we get here
else
return vec ;
}
//----------------------------------Mathutils.CopyQuat() --------------
//Copies a quaternion to a new quat
PyObject * M_Mathutils_CopyQuat ( PyObject * self , PyObject * args )
{
PyObject * quat = NULL ;
2007-03-31 15:31:37 +00:00
static char warning = 1 ;
if ( warning ) {
printf ( " Mathutils.CopyQuat(): Deprecated: use Mathutils.Quaternion() to copy vectors \n " ) ;
- - warning ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
quat = M_Mathutils_Quaternion ( self , args ) ;
if ( quat = = NULL )
return NULL ; //error string already set if we get here
else
return quat ;
}
//----------------------------------Mathutils.CopyEuler() ---------------
//copies a euler to a new euler
PyObject * M_Mathutils_CopyEuler ( PyObject * self , PyObject * args )
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
PyObject * eul = NULL ;
2007-03-31 15:31:37 +00:00
static char warning = 1 ;
if ( warning ) {
printf ( " Mathutils.CopyEuler(): deprecated:use Mathutils.Euler() to copy vectors \n " ) ;
- - warning ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
eul = M_Mathutils_Euler ( self , args ) ;
if ( eul = = NULL )
return NULL ; //error string already set if we get here
else
return eul ;
}
//----------------------------------Mathutils.RotateEuler() ------------
//rotates a euler a certain amount and returns the result
//should return a unique euler rotation (i.e. no 720 degree pitches :)
PyObject * M_Mathutils_RotateEuler ( PyObject * self , PyObject * args )
{
EulerObject * Eul = NULL ;
2004-02-29 13:20:34 +00:00
float angle ;
char * axis ;
2007-03-31 15:31:37 +00:00
static char warning = 1 ;
if ( warning ) {
printf ( " Mathutils.RotateEuler(): Deprecated:use Euler.rotate() to rotate a euler \n " ) ;
- - warning ;
}
2004-02-29 13:20:34 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
if ( ! PyArg_ParseTuple ( args , " O!fs " , & euler_Type , & Eul , & angle , & axis ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.RotateEuler(): expected euler type & float & string " ) ;
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-05-20 19:28:04 +00:00
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
Euler_Rotate ( Eul , Py_BuildValue ( " fs " , angle , axis ) ) ;
2007-05-26 04:39:31 +00:00
Py_RETURN_NONE ;
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
}
//----------------------------------Mathutils.MatMultVec() --------------
//COLUMN VECTOR Multiplication (Matrix X Vector)
PyObject * M_Mathutils_MatMultVec ( PyObject * self , PyObject * args )
{
MatrixObject * mat = NULL ;
VectorObject * vec = NULL ;
2007-03-31 15:31:37 +00:00
static char warning = 1 ;
if ( warning ) {
printf ( " Mathutils.MatMultVec(): Deprecated: use matrix * vec to perform column vector multiplication \n " ) ;
- - warning ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//get pyObjects
if ( ! PyArg_ParseTuple ( args , " O!O! " , & matrix_Type , & mat , & vector_Type , & vec ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.MatMultVec(): MatMultVec() expects a matrix and a vector object - in that order \n " ) ;
2005-09-27 17:03:28 +00:00
return column_vector_multiplication ( mat , vec ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//----------------------------------Mathutils.VecMultMat() ---------------
//ROW VECTOR Multiplication - Vector X Matrix
PyObject * M_Mathutils_VecMultMat ( PyObject * self , PyObject * args )
2005-05-22 17:40:00 +00:00
{
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
MatrixObject * mat = NULL ;
VectorObject * vec = NULL ;
2007-03-31 15:31:37 +00:00
static char warning = 1 ;
if ( warning ) {
printf ( " Mathutils.VecMultMat(): Deprecated: use vec * matrix to perform row vector multiplication \n " ) ;
- - warning ;
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//get pyObjects
if ( ! PyArg_ParseTuple ( args , " O!O! " , & vector_Type , & vec , & matrix_Type , & mat ) )
return EXPP_ReturnPyObjError ( PyExc_TypeError ,
" Mathutils.VecMultMat(): VecMultMat() expects a vector and matrix object - in that order \n " ) ;
2005-09-27 17:03:28 +00:00
return row_vector_multiplication ( vec , mat ) ;
2004-02-29 13:20:34 +00:00
}
Mathutils update
- also included is some fixes for preprocessor inclues and some clean up of the previous commit
-rewrite and bugfixes
----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
2005-07-14 03:34:56 +00:00
//#######################################################################
//#############################DEPRECATED################################