Blender Git Loki

cleanup: use realloc for faces

New 'read' flag to toggle reading of the CacheLibrary as a whole.

This is also used during the bake process to avoid confusion: The read
flag gets disabled for the baking cachelib, so that objects don't try
to read cache data that is supposed to be generated.

cleanup: use 'const' metaelems

cleanup: minor edits to BVH code

cleanup: use realloc & store co/no as float(*)[3]

cleanup: avoid passing by value when we can use const pointers

cleanup: use strict flags

cleanup: style & doxy comments

Delete operator for cache library datablocks.

CacheLibrary datablock has a generic unlink function now, but currently
nothing actually links to cache libs themselves, so it's empty. Still
good to keep this in mind for the future.

Reader/Writer API functions for basic particles.

Also slightly renamed pathcache functions for consistency (use plural
'particles' instead of 'particle').

Removed deprecated and unused alembic API functions.

These are now implemented as virtual functions in the Factor class.

Patch: T43678 BVH tree for Metaball calculation

T43678#291857 by @chrisr (mball_26_02_fix.diff)

Merge branch 'cycles_memory' into gooseberry

Fix blenderplayer

Merge branch 'master' into asset-experiments

Depsgraph: Add relation between parameters and animation lamp components

This way animating lamp parameter will properly cause all related drivers to be
updated as well.

This seems to be actually needed for all non-geometry nodes, but that's for later.

Depsgrpah: Fix for wrong check of object data was already created

Depsgraph: Fix/workaround for a drivers using lamp settings as variables

Not sure what's the designed way dealing with this, so if Joshua can clarify this a
bit that'd be really nice :)

For now it's so called "seems to work in the test file", added a TODO in the code so
we don't forget to doublecheck on this before merge.

Depsgrpah: Remove bone eval check from a while ago

It doesn't really fully working because it's possible only part of armature will be
tagged for update now.

What is not clear currently if it's legit situation when we're updating only parts
of the armature or it could lead to some matrix space mismatch.

Depsgraph: Code cleanup, line wrapping