Blender Git Loki

Added writers for derived mesh result and hair paths.

Use explicit name strings for all readers/writers in the cache instead
of constructing names internally.

This helps prevent name collisions and guarantees a consistent naming
scheme for putting multiple items in the same cache.

New reader/writer for DerivedMesh.

Used as a base for existing PointCacheModifier classes, but can be used
on its own.

General support for export to/import from Alembic files based on Cache
Libraries.

Deleted unimplemented parts of the previous pointcache system, to make
refactoring simpler.

Since the new approach is not tied to the old set of point cache data
types there is really no need to carry their empty husks around. Once
the new caching system is defined in detail we can add back whatever
makes sense.

Disabled specialized Point Cache modifier handling in the modifier
stack, this will be replaced and is not correct anyway.

Refactoring of archive handling in readers/writers.

Now instead of each reader/writer creating its own archive, the archive
is created by the caller in advance and passed as a constructor
argument. This means that multiple items can be stored together in the
same archive.

Removed unused Scene pointer from the base Reader/Writer classes.

Operator skeleton for a cache bake operator, using the job system.

Fix for undefined return value.

Removed unused line.

Unlink and cleanup functions when objects or the cached group get
unlinked or changed.

Make sure cache item pointers are lib-linked correctly.

Name string construction for cache items, to use as a unique identifier
in the cache library.

Draw some UI buttons for all the current cache item types.

Updated hash table for lookup of enabled cache items and added basic
UI support.

Improved collections in RNA for looping over potential caching items
in the object group.

Define a collection of "object_caches" in the CacheLibrary RNA for
iterating over all objects eligible for caching in the group.

This allows for nested dupligroup instances as well. All objects that
are instantiated at least once by any group can be included in the
respective cache, but get represented only once because the cached data
is the same anyway.

Link main CacheLibrary pointers in readfile.

Simplified cache item path description, based on plain ID blocks with
subtype/index.

This omits possible instancing and recursion by dupli groups, but since
the data of instances is always the same in Blender at this point, there
is little need to take them into account for caching.

At some point in the future it may become desirable to store full dupli
hierarchies in the cache, but it doesn't make sense to try to design a
fully-fledged path descriptor for this hypothetical case now.