would convert from quat to matrix and back if the bone had any constraint,
but did not normalize the quat first as done in other places, giving a
sudden jump when starting transform on some bones with constraints. Two
changes:
* Normalize quaternion first.
* Only do this conversion on bones with limit rotation constraints,
instead of all bones with any constraint.
* new icon for split view (at least temporary)
* icon buttons in header rather than popup menus for better efficiency (can easily be changed in python UI script again)
Bone would get a '.' added even when there was no extension.
(center limit would still be useful)
- name flipping function used sizeof() incorrectly.
- ED_lorem should be extern.
fixing
[#21014] SEQUENCER: Can no longer apply colour balance filter to colour effect strip.
Also: reordered some UI elements so that
geometry, time and color modifications are grouped together.
- collada export would run MEM_freeN on an un-initialized pointer in some cases.
- makesrna was missing a call to close a file.
- text cursor update function was missing a NULL check for st->text.
- possible (unlikely) un-initialized return value for bge python lamp.type, set error instead.
- possible (unlikely) missing NULL terminator with strncpy for ffmpeg.
(http://freestyle.sourceforge.net/doc/html/index.html) has been
incorporated into the Blender/Freestyle Python API implementation
in the form of Sphinx-based embedded docstrings. Some C++-specific
descriptions of classes and functions were revised so that they are
suitable for Python programmers. Missing docstrings were filled,
and sparse descriptions were extended. By means of the new
documentation system for Blender, an up-to-date Freestyle Python
API reference will be part of the Blender 2.5 documentation.
- BGE Shader.setSampler(name, index): index range check was wrong.
- Compositor check for an invalid channel was incorrect.
- getting the center of selected verts used an uninitalized z axis.
- do_init_render_material() used && rather then & when testing for MA_TRANSP.
- weight paint activate flipped bone used && rather then & for flag checking.
The patch exposes mouse and keyboard read-only properties in the GameLogic module
Also renames bge.keys to bge.events (* Note: name of bge submodules (logic, render, ...) may change before 2.5 final release [right Campbell?]).
"""
This patch adds two new types to the BGE:
SCA_PythonKeyboard
SCA_PythonMouse
These two types allow users to make use of the keyboard and mouse without the need for a keyboard or mouse sensor.
SCA_PythonKeyboard has an events property that acts just like SCA_KeyboardSensor.events.
SCA_PythonMouse also has an events property to check for mouse events. Further more it supports getting and setting normalized cursor position (from 0.0 to 1.0) with SCA_PythonMouse.position. The cursor can be shown/hidden using SCA_PythonMouse.visible.
"""
Its use is similar with current mouse and keyboard controllers. With the exception of mouse position being normalized and writable as well (replacing Rasterizer.setMousePosition).
Code Sample:
######
from bge import logic, events
mouse = logic.mouse
keyboard = logic.keyboard
for key,status in keyboard.events:
if status == logic.KX_INPUT_JUST_ACTIVATED:
if key == events.WKEY:
print(mouse.position)
# move_forward()
mouse.visible = True # turn cursor visible
mouse.position = 0.5,0.5 # centralize mouse - use tuple
######
* Important Note: mouse.position still will not work properly for Letterbox mode.
In order to fix letterboxing I may need to move the set x,y mouse function to inside the canvas code (to avoid duplicated code between mouse sensor and bge.logic.mouse). I'll leave this for another commit though.
Thanks Mitchell for the work on that.
I heard that the actual solution would be to remove #include <pthread.h> from BLI_threads.h
But in the mean time is not fair to CMake/MSVC to be the only system not building ;)
own fault when adding mathutils callbacks, generic destructor didnt free the matrix accessor array, made the array apart of the matrix struct since its not worth malloc'ing to save at most 16bytes.
Force mesh out of face select mode and into edge mode when doing loop cut -
it left an invalid edge selection in face mode, especially with edge slider afterwards.
the same coordinates as points.
Also send additional notification at the end of conversion operator,
which is needed for correct interface updating (active object could
be changed in operator but interface wouldn't be updated).
In this day and age most people have 3 button mice, so this option can now be the
exception not the rule - causes enough issues with conflicts. This commit won't
disturb existing saved configs.
* Division by zero fix for TNT SVD code.
* Sound fix, in case ffmpeg decode fails, don't use the samples.
* Fix for incorrect bounds of transformed objects in new raytracing code.
* Gave memory arena's a name used for allocations for easier memory
usage debugging.
* Dupligroup no_draw option was using layers but not restrict view/render
setting. (not a bugfix exactly but would do display list context switching
while drawing for no reason).
* Fix objects instanced on hair particles not giving consistent results
when the object is transformed.
* New math functions: madd_v4_v4fl, len_squared_v3v3, interp_v4_v4v4v4,
mul_v4_m4v4, SH and form factor functions, box_minmax_bounds_m4.
* mul_m4_m4m4 and mul_m3_m3m3 now accept the same pointers for multiple
arguments.
* endjob callback for WM jobs system.
* Geometry node uv/color layer now has search list/autocomplete.
* Various small buildsystem tweaks, not strictly needed yet in trunk.
