[ #5458 ] Some transforms are not accurate while pressing shift key
Shrink/Fatten & Push/Pull: Shift slowed the transformation but didn't add precision. (integer division. fixed by applying the same solution used in Grab)
Warp: Shift was ignored. Fixed by changing warp to use the same input method has To Sphere. That is, the 3D view acts like a big horizontal slider: Left -> No effect, Right -> Full effect. This is somewhat of a "big" change from how warp used to work but this is more predictable and is not limited by the original position of the mouse pointer. (of course, this works with Shift correctly too).
Render: Ambient Occlusion was being triggered now with the Material flag
(Shader panel) "Shadow". Nice idea, but that was not so in 2.42a and
before, breaking how previously renders looked.
Now the "Shadow" button in Shaders panel only triggers shadow again.
(Note: the Ambient slider in material triggers AO)
Since 2.40 (and a few pre-releases around then), armature joining has not
worked correctly. Constraints and other attributes of bones in posemode
(IK DOF limits, transform locks, custom shapes, etc.) were not preserved
on the armature(s) that were joined onto the last selected armature. This
was a serious production problem, as it meant that you could not easily
add pre-made rig segments and merge them with the rest of your rigs without
having to redo all the constraints. After a few attempts, I've finally
managed to fix this.
All constraints and parenting relationships get name corrections for the post-
merge armatures. Action channels in actions don't really get any corrections
yet unless the action is being used by an Action Constraint.
Python-API people: beware, I may have broken something in this commit.
Two sensible methods left in: armatureobject.insertPoseKey and .insertCurrentPoseKey,
but still left unofficial i.e. not published in the api doc yet, because of additional
error checking is still needed and more testing too. But at least is now in better shape to
work on to be published in a later release.
- Bugfix #5714
New option "Save changed images" didn't respect relative file names
- Bugfix #5621
New camera (shiftx, shifty) was missing in IpoWindow as animatable option
UV rendering error: the third value of a UV texture coordinate should be
set to zero always, this enables 3d textures to work uniform with UV.
Also fixed: UV pass output for non_OSA is now equal to OSA, meaning only
used UV face values are in such passes.
Turned out somehow I managed to miss adding the proper
type refs in Blender.Types for IDGroupType and IDArrayType,
which made the script not work. So, I've got it all fixed now.
Or at least hopefully :)
After extensive testing, I've concluded that VisualRot keyframing
didn't work at all for bones. On the other hand, VisualRot for Objects
worked fine. Now I've come up with a fix, which should make this
actually work for IK/FK switching (I hope).
The fix I'm committing is slightly hacky as it tempolarily overwrites the
bone's rotation/quat values with the 'visual' ones, inserts keyframes,
then resets the values it wrote on before.
This hack was needed as for some strange reason, the values being
added as keyframes were not the ones that had been derived. Note
that on some bones, there will be a little bit of strange rotation after
frame changes while their constraints are left on.
Vertex color node worked only if VCol Paint/Light was enabled. Fixed
that, and removed the vertex color node making it part of the geometry
node instead.
Also, preview.blend had black vertex colors for the sphere, so set them
to white like the other primitives.
Loss of UV coordinates reading from 2.42 .blend files, if there were
both vertex colors and tfaces (these contained colors also). It should
have been impossible to create them both in 2.42, but even if invalid
we should read them correct.
Rewrote the core-function responsible for ipo-cleaning. Now, it is in a less
wacko form. What is still not done is the conversion of a bunch of points
describing an arc to an arc defined by the handles of the keyframes on
either side of the arc. That will have to wait for the next development cycle.
This rewrite fixes two big bugs with the code:
* All but first curve got hidden aften cleaning
* Cleaning a curve with only two verts resulted in only one vert, even though
the values were not the same.
Material Node Shader crash: new "layered UV" option required a better
version patch... now patches all 2.42.x files, and not only the 2.42.2 and
older :). Yes, subversioning doesn't always make things easier.
A few tweaks to this little tool I added yesterday. These tweaks are so
that bugs like #5625 and #5636 are less likely to occur due to extreme
scaling.
Now, for action strips added using this:
* Auto-calculation of action range is disabled for that strip
* Strip Start and Action Start are set to the current frame value, when the
strip was added
* Strip End and Action End are set to either the end frame value or 100
frames after the current frame value, whichever is higher.
----------
Added CurNurb.recalc() method. I thought I had commited this before as part
of a bugfix but obviously not. It allows control point handles be recalculated
after changing a curve's bezTriples.
