[Bf-committers] Re: [Bf-blender-cvs] CVS commit: blender SConstruct blender/bin/.blender .Blanguages blender/extern/bFTGL/src Makefile blender/extern/bullet/Bullet/NarrowPhaseCollision CollisionMargin.h blender/intern/SoundSystem/openal SND_OpenALDevice.cpp blender/intern/bmfont/intern BMF_BitmapFont.cpp blender/intern/boolop/intern ...

Toni Alatalo antont at kyperjokki.fi
Mon Jan 30 13:27:44 CET 2006


On Saturday 28 January 2006 19:33, Chris Want wrote:
> hos (Chris Want) 2006/01/28 19:33:28 CET
>   Modified files:
>     blender/source/blender/python/api2_2x Armature.c Bone.h NLA.c
>   The code from the orange branch has now been copied into HEAD.
>   - bone layers

as mentioned in some other of these messages, the version of Armature I'd 
committed (khughes actually wrote the draft for the wrapper) with .layers 
property (it a manipulable list of active layers, e.g. [1,4,7] or [2,8]) is 
now in there - up to the py team to accept / reject and for me to document if 
it is accepted then (it is currently undocumented)

also, parts of the work I and Roland Hess did in December for NLA action / 
path baking were in orange cvs and now in bf head. there is a function 
insert_matrixkey, that replicates insert_key() but tries to use the current 
object matrix as it is displayed on screen. and i have some calls for that in 
the Object class, seems that four were added in the experimentation 
alltogether:

static PyObject *Object_insertIpoKey( BPy_Object * self, PyObject * args );
static PyObject *Object_insertPoseKey( BPy_Object * self, PyObject * args );
static PyObject *Object_insertCurrentPoseKey( BPy_Object * self, PyObject * 
args );
static PyObject *Object_insertMatrixKey( BPy_Object * self, PyObject * args );
static PyObject *Object_bake_to_action( BPy_Object * self, PyObject * args );

i have some working py scripts that use those differently,

but also Roland / harkyman has a patch on the c side with much later versions 
of the NLA baking code, that we have tested a little here at orange (we have 
not actually tested the path baking yet). as reported by Roland himself 
earlier here, the code should definitely be reviewed and perhaps rewritten if 
it was to make it to BF .. bake_all_to_action(); is the function in 
editnla.c , Roland could perhaps fill in here with some status info of the 
code and a link to the patch? (the code in cvs now is 3 weeks older, i guess 
from mid December, than his later that he worked on in early January at 
least). if nothing else we/i must at least clean up those before 2.42 
release, but if it is possible to get the new functionality in it would be 
kinda nice too.. ;)

(but that is not to distract Ton from render coding)

>   -Ton-

~Toni


More information about the Bf-committers mailing list