Updated id properties interface as per
discussed in python meeting. Basically,
id properties are now entirely accessed
through the dict-like interface if IDGroupType.
Also, tp_getsetters are used throughout the code
now.
Using the dict interface allowed for a major cleanup
of the wrapping code. The biggest change is that ID
properties are no longer wrapped in a structure with
.type .name and .data members; instead when you get
properties from the group it returns the direct value.
Ints, strings and floats return simple python types,
while arrays and groups return special wrappers though.
This means to detect the type of an ID property, you
have to use type(). For string and int types this is
easy; for group and array types (which of course have
their own wrappers) you use type() with Blender.IDGroupType
or Blender.IDArrayType.
Update of epydocs plus a temporary gui script will be
forthcoming; the gui script will be removed before release
as of course by then we'll have a built-in gui for id
properties.
off_export - minor changes
uv_archimap - cleanup. slightly slower but less duplicate code
uv_from_adjacent - nothing
BPY_menus - renamed py slot UvCalculation to UVCalculation to be consistant
added python api stuff to deal with Color and UV/Image layers.
me.activeUvLayer - int
me.activeColorLayer - int
me.totUvLayers - int
me.totColorLayers - int
me.addUvLayer()
me.addColorLayer()
me.removeUvLayer()
me.removeColorLayer()
Variable names may need changing.
These can be created and deleted in the Mesh panel in the same place as
before. There is always one active UV and vertex color layer, that is
edited and displayed.
Important things to do:
- Render engine, material support
- Multires and NMesh now lose non active layers
Also CustomData changes to support muliple layers of the same type, and
changes to layer allocation, updated documentation is here:
http://mediawiki.blender.org/index.php/BlenderDev/BlenderArchitecture/CustomData
added restrictDraw/Select/Render to objects in python
updated group and scene docs for last commit
made 3ds import use new scn.objects rather then Object.New() - (removed import as instance for now)
fixes off import error from this report http://blenderartists.org/forum/showthread.php?t=84182
----------
Bugfix #5373: creating a curve or text object using Object.New() without
linking any data to the object would later cause a segfault when ob->data
was later dereferenced. This problem will be fixed (hopefully soon) in the
API when new objects are created with data and linked to scenes all in one
step, but for now check for curves that ob->data is defined before using,
otherwise print an error message to the console and skip the object.
This commit upgrades the modifier stack to only calculate the data which is
needed, either by modifiers further down the stack or by other functions at
the end of the stack (e.g. drawing functions).
This speeds up modifier stack recalculation, especially where vertex
groups and UV coordinates are concerned. For example, a mesh with an Armature
modifier followed by a Subsurf modifier would previously have required the
Subsurf modifier to interpolate all the vertex groups in the mesh, slowing
down modifier calculations considerably. With this update, vertex group data
is not propagated beyond the Armature modifier, so calculations are faster.
Note that this depends on the order of modifiers in the stack. If the Armature
and Subsurf modifiers were swapped in the above example, the Subsurf modifier
would have to interpolate vertex groups, as they are needed by the Armature
modifier.
BPy conventions were not being followed for PyObject_New;
now they are. I still doubt the usefullness of doing this,
however its important to maintain module standards.
Which reminds me, anyone know where I can get that source code
prettier that was used on the api files? It was kindof
hard following the coding style of bpy with it being so
different from normal blender, and also what with having
to write the IDProperty code from scratch. :)
Another bug from the tracker, reported by Mike Stramba.
A duplicated Py_XDECREF in the wrong place made assigning
arrays from Vector objects not work.
Also, fixed nasty bug in C API of idproperties (the function
to look up a property from a group was broken). Fixed a memory
leak too.
In addition, made "del group['property']" delete properties
from group; previously this would just crash (or at least it
should have). Added a small addition to the example in the
epydocs for IDGroup.
Fixed the bug of "print type(property)" crashed blender;
just needed to add PyType_Ready init code for all ID property
python types; I made a function IDProp_Init_Types and put
code to call it in types_initAll().
Also added GPL/BL headers to idproperty files.
Misspelled __members__ __member__ in IDProperty's
getattr function; this made "print dir(prop)" not
work. Eventually I need to replace the whole thing
with getsetters.
Still need to find out why "print type(a_property)" crashes
blender. :S
IDProperties was doing a name check where it shouldn't;
thise made executing "prop['value'] = something" not work
if the property 'value' already existed. Fixed it by
making the code replace existing properties, like it
should.
This update changes Object.properties to Object.game_properties
(as discussed) so .properties can be used for ID Properties.
This should be fine as, after all, .properties was undocumented anyway :)
Crash using Shape Widget Wizard script, was an error in customdata copy.
The script still throws a python error though, but that seems unrelated
to this crash.
==========
Bugfix #4951: This might be a serious change for python script writers.
-- ob.getMatrix('localspace') did not return the correct matrix when the object
had a parent; this has been corrected.
-- ob.setMatrix(m) did not work predictably (more correctly, in an easy-to-
predict manner) when an object had a parent. The method has been changed
so that if thee is a parent, it ASSUMES the matrix is "localspace",
relative to the parent. The documentation now states this.
-- ob.mat and ob.matrix are now read-only attributes, while ob.matrixLocal
(which calls Object_setMatrix() ) is now read-write.
Ton is not thrilled (is that a fair summary, Ton? ;-) with this method since
the ob->obmat is calculated from the loc/rot/size attributes of an object.
I'll let him speak on this, but I believe his desire is for this method to be
deprecated in the future and replaced with something else.
The epydocs are now updated to have idproperties;
all the modules that have bindings for ID properties
now has docs for them. E.g Materials have a .properties
members, Image, Texture, Scene, Object, NMEsh, and Mesh.
I realized that .properties was already taken in
Objects, so I renamed it to .idproperties. There was
also a nasty little problem with an example inside
Object.getType; the entire example was being pasted inside
the return field. I fixed it by just moving the return
definition to after the example, like it should be.
All data layers, including MVert/MEdge/MFace, are now managed as custom
data layers. The pointers like Mesh.mvert, Mesh.dvert or Mesh.mcol are
still used of course, but allocating, copying or freeing these arrays
should be done through the CustomData API.
Work in progress documentation on this is here:
http://mediawiki.blender.org/index.php/BlenderDev/BlenderArchitecture/CustomData
Replaced TFace by MTFace:
This is the same struct, except that it does not contain color, that now
always stays separated in MCol. This was not a good design decision to
begin with, and it is needed for adding multiple color layers later. Note
that this does mean older Blender versions will not be able to read UV
coordinates from the next release, due to an SDNA limitation.
Removed DispListMesh:
This now fully replaced by DerivedMesh. To provide access to arrays of
vertices, edges and faces, like DispListMesh does. The semantics of the
DerivedMesh.getVertArray() and similar functions were changed to return
a pointer to an array if one exists, or otherwise allocate a temporary
one. On releasing the DerivedMesh, this temporary array will be removed
automatically.
Removed ssDM and meshDM DerivedMesh backends:
The ssDM backend was for DispListMesh, so that became obsolete automatically.
The meshDM backend was replaced by the custom data backend, that now figures
out which layers need to be modified, and only duplicates those.
This changes code in many places, and overall removes 2514 lines of code.
So, there's a good chance this might break some stuff, although I've been
testing it for a few days now. The good news is, adding multiple color and
uv layers should now become easy.
ID Properties binding have now been added for textures. Also,
the beginnings of supporting "del IDProperty Object" (which
basically removes the property from it's parent group then frees
it) in python were done; really the only thing now is to figure
out exactly *how* you overload the del operator. :S
Using acos(dot(u, v)) to find the angle between two vectors is quite
inaccurate, and there's a better way to do it, as explained here:
http://www.plunk.org/~hatch/rightway.php
Also changed the use of atan for computing roll to atan2 in some places,
the latter avoids accuracy and division by zero issues.
