To make it work, I had to:
- update Operators to include all of its methods in the Python API
- redefine the Freestyle modules as local classes, allowing predicates to support the __call__ method (I found out that Python does not allow modules to be callable). The API is defined in the new freestyle_init.py file.
- remove all references to former Python Freestyle API (located in source/blender/freestyle/python/, under the Freestyle.py file) and replace it with the new API file.
- normalize all constants related to the Nature class
- redefine the logical_operators.py file so that AndUP1D, OrUP1D and NotUP1D would be subclasses of a UnaryPredicate1D subclass having a operator() implementation. Using UnaryPredicate1D as a superclass would somehow prevent calls to the operator(). For the time being, I chose ContourUP1D (even though it does not really matter which subclass it is, since the __call__ method is redefined in all of these classes). I will implement this classes in C++ to get rid of the problem altogether.
- turn off a few classes that somehow are not well recognized (Noise, Curve...). I will reenable them later once I understand what's going on.
To make it work, I had to:
- update Operators to include all of its methods in the Python API
- redefine the Freestyle modules as local classes, allowing predicates to support the __call__ method (I found out that Python does not allow modules to be callable). The API is defined in the new freestyle_init.py file.
- remove all references to former Python Freestyle API (located in source/blender/freestyle/python/, under the Freestyle.py file) and replace it with the new API file.
- normalize all constants related to the Nature class
- redefine the logical_operators.py file so that AndUP1D, OrUP1D and NotUP1D would be subclasses of a UnaryPredicate1D subclass having a operator() implementation. Using UnaryPredicate1D as a superclass would somehow prevent calls to the operator(). For the time being, I chose ContourUP1D (even though it does not really matter which subclass it is, since the __call__ method is redefined in all of these classes). I will implement this classes in C++ to get rid of the problem altogether.
- turn off a few classes that somehow are not well recognized (Noise, Curve...). I will reenable them later once I understand what's going on.
Quite a few modifications were made to finish the API:
- Freestyle's SConscript was modified to catch all files within the intern/python directory, allowing integration of future shaders implemented in C++.
- the Operators class was ported, with a special care of making its methods static (using the METH_STATIC flag in the tp_methods method definitions)
- all of the type-checking functions [ BPy_[class name]_Check(obj) ] were changed to allow subclasses to be seen as that type too: instead on looking at the ob_type value, the PyObject_IsInstance function is used.
- all of the iterators can now retrieve the object pointed to by the operator, using the getObject() method. A directedViewEdge pair is returned as a list of the two elements in the pair.
- all of the style modules were copied to a style_modules_blender/ folder and were modified to use Freestyle as a Blender's submodule. IntegrationType and MediumType was also integrated (for example, changing MEAN to IntegrationType.MEAN).
Testing now begins. If everything works correctly, I'll move on to lib3ds removal right away.
There is just one more class remaining to port (and probably the most important): Operators. After that, I'll be able to test whether Freestyle functions well without SWIG.
- Takes less than half the time to format a full document
- Where possible only the required lines are re-parsed when text is changed (was the whole file, for every key press!)
- Memory is allocated in one place only (there were all sorts of problems here)
- Should be easily extensible for other scripting languages
- Lots of comments to make it very easy to follow / change
- def and class are now properly coloured. They had a theme colour but the checks didn't work.
Bone-locking should only be set if the bone (and not its parent/children) are selected. Transposition error from the patch (I had to manually apply as there were some oddities about the patch file).
The pasting code for text-buttons (when hovering over the button), was incorrectly using the drawstr (i.e. the string that is displayed in the text-button, and includes the 'name' of the button) as the old value. This meant that the renaming code could not correctly detect matches to rename.
This bug-report revealed a critical error made in the menus for Constraints Owner/Target spaces. It seems that the wrong values have been used for 'pose space' and 'local with parent' space. This has now been rectified, so that old rigs won't break (just that the labels will be a bit different).
Submitted by: Lorenzo Pierfederici (lento)
This patch adds the ability to lock transformation on bones in edit mode, to protect them from accidental editing.
Bones can be locked from the editing buttons, the transform property panel, the specials popup menu or the python api.
This now works even when there is a proxy-protected bone selected with a few differences.
* The Transform Locks, IK-Flags, and Custom Bone Shape settings are unavailable, as they would be reset by the proxy protection anyway
Also, the Copy Constraints options have been altered so that they do NOT free all existing constraints on the target bones before copying.
* Constraints are now appended to the end of the constraint stacks, with those added to bones belonging on proxy protected layers getting tagged as 'local' constraints.
* This is more useful for riggers than the previous behaviour (which was quite nasty too).
Render/postrender events were missing from bg rendering (and also from rendering called inside scripts). Found this because of bug #17389, the code to prevent race conditions with pynodes is currently inside BPY_do_all_scripts (that runs scriptlinks) and so was not being called in bg mode or rendering via scripts.
http://projects.blender.org/tracker/?func=detail&atid=125&aid=17389&group_id=9
This commit attempts to fix some of the problems with the accuracy of strokes recorded.
I've disabled the post-draw smoothing of strokes as it deviated from the source stroke too much. This was introduced to try to eliminate the effects of fine 'jitter' (which is especially noticable when using optical mice + fast computer).
Now, I've introduced two thresholds which determine the 'minimum' distances that the mouse needs to have moved from the previous mouse-coordinates recorded, to be recorded. These are currently hardcoded, but may be exposed in the UI if there is significant need to do so.
- The first is for the distance that the mouse must have moved on both axes to be considered.
- The second is for the 'pythagorean' distance that needs to have been travelled by the mouse.
* For constraint mesh-targets, the check for whether the target mesh was in EditMode didn't check to see if the target mesh was the mesh in EditMode.
Thanks jaguarandi for catching that!
* Bumped up size of buffer for grease-pencil drawing after seeing a program written in 1989 that had a 100 times larger buffer
