This is an interesting bug since it is likely the cause of many other suspicious python crashes in blender.
sys.last_traceback would store references to PyObjects at the point of the crash.
it would only free these when sys.last_traceback was set again or on exit.
This caused many crashes in the BGE while testing since python would end up freeing invalid game objects -
When running scripts with errors, Blender would crash every 2-5 runs - in my test just now it crashed after 4 trys.
It could also segfault blender, when (for eg) you run a script that has objects referenced. then load a new file and run another script that raises an error.
In this case all the invalid Blender-Object's user counts would be decremented, even though none of the pointers were still valid.
"Warning: binarysearch_bezt_index encountered invalid array" errors were being displayed in the console. Was caused by 3d-view show-keyframe for infostring stuff, when an IPO being checked had no keyframes.
- Adding constraint using button in panel still didn't update Armature Editing buttons properly.
- Minor code tidying of earlier bugfix for armatures
- 'For Transform' option for Limit constraints is now only taken into account for constraints that are enabled.
Second attempt at fixing this bug. Previous fix caused segfault when all bones in a chain are selected. Now it should segments which are selected (i.e. get swapped) will get unparented from segments that aren't (i.e. aren't swapped, so are still in old orientation)
After some discussion with Campbell, changed the way cstruct sizeof is fetched.
Moved DataSize() to Blender.Types.CSizeof(Blendertype). Supported types return sizeof(data struct), otherwise -1.
To quickly check what types are supported:
import Blender.Types as bt
x = dir(bt)
for t in x:
if t[0] != '_':
s = 'bt.CSizeof(bt.' + t + ')'
print t,"=", eval(s)
[#17867] Adds option to SCONS to generate Python API documentation
Added patch from Brandano with some small improvements (BF_DOCDIR, clean) by yours truly.
To use make sure you have epydoc installed. Enable with WITH_BF_BPYDOC=1.
* fix two typos in RenderLayer API (renderosiy -> renderosity in two places. Will break .py's saved with render_save_layers.py, just fix passRadiosiy and passRadiosiyXOR)
* add some docs on RenderLayer API
* fix some copy/paste leftover in render_save_layers.py
source/blender/blenlib/intern/fileops.c - zero length strings would check for a slash before the strings first char.
source/gameengine/GameLogic/SCA_JoystickSensor.cpp - m_istrig_prev was not initialized
source/blender/src/editmesh.c - active face pointer was not set to NULL in free_editMesh()
- Forgot to make SCA_ISensor::UnregisterToManager() virtual to intercept active-inactive transition on collision sensor to clear colliders reference.
- Don't record collision on inactive sensor.
This situation occurs when an object with an inactive collision sensor collides with an object with an active collision sensor: the collision handler triggers both sensors.
The result of this bug was pending references that eventually cause temporary memory leak (until the sensor is reactivated).
- Reset hit object pointer at end of frame of touch sensor to avoid returning invalid pointer to getHitObject().
- Clear all references in KX_TouchSensor::m_colliders when the sensor is disabled to avoid loose references.
- Test GetSGNode() systematically for all KX_GameObject functions that can be called from python in case a python controller keeps a reference in GameLogic (bad practice anyway).
Particle system crashed in convertblender, missing NULL check.
This fixes crash, render survives, but I doubt it was meant so...
Will leave it Janne to evaluate later.
regression files, it showed small dark outline errors on envmap.
This commit rewinds bugfix #8437, which actually had to be fixed on
another location in code, which was done a few weeks ago. :)
Index OB pass didn't support FSA for Ztransp.
Also made buttons to set black/white for non-RGBA images hide in Image Window,
the Curves color code only supports 4 channels atm.
When Action Editor was maximised, Grease Pencil data disappeared. This was due to the screen-swapping that went on. Now, it uses the old-screen that was stored in the maximised Action Editor's screen (should be safe...)
* Onion-skinning with GStep > 0 was not showing enough of a noticable difference between ghosts. Improved method of calculating this.
* Clicking in a Grease-Pencil datablock channel in the Action Editor would crash
