Blender Git Loki

Move fx_flag into the GPUFXSettings struct (simplifies things a bit)

Cleanup.

Merge branch 'master' into asset-experiments

Add previews to scenes, and enhance batch preview generation.

Batch preview generation can now also generate 'internal' previews (materials, textures etc.),
as well as scene ones. Also, some basic Cycles handling is there now.
You can now chose which kind of data to generate previews for, and whether the files are
trusted or not (for py autorun and drivers).
And rendering happens in background again, issue was actually a bug in creator.c, fixed in master.
Main TODOs reamining here: OpenGL rendering, and find a way to prevent '.blend1' when
saving over (either as a new option of save op, or simply by removing those files in script?).

Added a batch cleanup of previews too.

Also, added path of files as their tooltip 'image' button in main filebrowser view,
quite useful with the 'recursive' option...

Cleanup: replaced the convoluted BKE_ptcache_id_clear function with
specific variants for clearing and truncating the cache frame range.

CLEAR_BEFORE is not used anywhere (it doesn't make much sense anyway).
CLEAR_FRAME is just a special version for deleting a single frame and
only used internally prior to replacing a cache frame.

Various changes:

- use term 'settings' rather then 'options', (follows rest of rna)
- use const where possible
- minor changes to arg order

Merge branch 'master' into alembic_pointcache

Conflicts:
source/blender/blenkernel/intern/cloth.c

Removed unused function.

Merge branch 'master' into viewport_master_merge

3D Viewport only uses DOF when rendered from camera now.

Read/Write implementation for cloth/hair data in alembic files.

This data is very simple, just deform positions, velocities and goal
positions. Future cloth might become more complex, but for now a plain
Abc::Point schema is sufficient.

Store a custom context pointer "cache_user" for point cache operators,
so that all point caches can use the export feature.

This was enabled only for particles. There is a lot of weird unexplained
code with special cases all over the place - unified point cache should
be possible in the end, but currently it's still a mess.

Reuse focal length and sensor from camera

More naming tidying up

Don't allocate FX, naming of eGPUFXFlags

Cleanup naming, use sane defaults

Improve Ptex filter border updates

- Texture paint looks better now

- Still need to fix corners

Depsgraph: Initial support of dupligroups

This is an attempt to bring the dupligroups support at least on the same
level as they used to work in the old dependency graph. This means we don't
bother with local storage, time offset etc for now and building the group
graph in the same scene graph (as opposed t ideal sub-graph),

The reason for that is that it'll allow to have working system sooner,
and even such a simplified approach seems to have issues with updates.

Separation of the main point cache API from the Alembic implementation.

Even though we probably will use only the Alembic backend in the
foreseeable future, it is still good design to have a bit of abstraction
in this place and not directly call Alembic functions.

Depsgraph: Change to how id tagging happens

Basically the idea is:

- Builder re-sets the LIB_DOIT flag, so it's guaranteed no left over
flags happened (it's not guaranteed LIB_DOIT is reset everywhere).

- Functions checks this flag directly, without using wrapping the check
with functions which kinda hides actual logic (that wouldn't be a
problem if this flag was only used by depsgraph).

- Removed flag check/set wrapping functions. Found them being rather
obscure and not really necessary from the beginning.