Blender Git Commits

Bugfix #2803

Transparent shadow could accidentally intersect first with a face being
farther away, due to the fact larger faces can fill up multiple octree
nodes. Had to use the same exception handling as for raytracing mirror or
glass.

Changed type for new textimeout and texcollectrate into "int".

Please note that the "long" should be avoided in dna structs... it is a
reserved type for use in our code for pointer storage. It is 32 or 64 bits,
depending CPU architecture (pointer size).

The "long" type currently is in the makesdna.c code, and interpreted as 32
bits only, which is actually wrong yes... I should remove it or code it
correct.

BTW; this gave a warning in space.c, another case where MSVC forgives you?
:)

Armature;

- IK constraint now uses Ipo from action again
- Bug in last commit; the object action was ignored... did NLA always

Editmesh:

- when there are edges in mesh, it now only copies selection in edges
when you have selectmode edge

New! Pose-constraint Ipos now are integrated in NLA. ALso cleaned up the
weird disabled code that was hanging out there for ages.

Also cleaned up NLA blending itself, it was copying far too much data
around. Should be three times faster or so... need good test!

And restored Action Baking.

- add missing includes for lazy programmers who don't read error msgs

- fix what appeared to be obvious bug, "if (fclose) fclose(file)" ->
"if (file) fclose(file)", Text3d.c owner should check.

Armatures; the bones menu used for "Make Parent" (CTRL+P) used wrong
indices still.
Cleaned the code, moved to editobject.c

Armature drawing fixes;

- Dashed line to parent bone was drawn to wrong bone in Rest Position
- Dashed line had selection color of parent, not of own bone.

- Fix for action constraints; for Bone targets (target is input bone
actually) only the relative rotation is used.

- Added scale=1.0 initializer in saving files, this fixes a little bit
better upward compatibility

- Still there are cases where bones flip 180 degrees when you read it with
older Blenders... not sure what it is caused by

Commented out one more line to make debug version of engine compile, needed
because API for armatures changed.

Two little bugs showing up in workshop;

- update_for_new_frame() (on frame advance, alt+a) crashed when no 3d view
was opened ever (NULL pointer)
- using proportional mode on Curve editing wrote beyond end of a malloced
block. Just a minus one needed here...

Two small fixes;

- ALT+R clear rotation on PoseMode didn't work when an Action was assigned
- 'Delete object' didn't set object pointers to NULL for Armature/Pose
constraints (old bug)

Well, it's better if the spelling is corrected everywhere the variable is used... (why must I make that kind of error?!)

After much debate on IRC, it was decided to spell "cancelled" properly...

Pressing ESC on armature posemode now restores position correctly again.

Reverted accidental change in compiling Python for OSX. Now it is default
again to use the Framework library.

Image garbage collection settings init was in the wrong do_version branch (236 instead of 237).

3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html


Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.

Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)


Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.


Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.


Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.

Make WITH_FREETYPE2 and INTERNATIONAL set to true by default when
building under linux (set to false in user-def.mk if you don't
want them).

cursors.c for blender's special cursors was half done.

GetBlenderCursor was prototyped but never defined. I filled it in, might be useful at one point (and for some uncommited evil Py code of mine).