Blender Git Commits

Move priority of widget handler back for node editor. Still not very
clean but should allow gooseberrians to work a little better

Initialize dropboxes and widgets together before opening the homefile,
makes sure areas get their proper handlers registered

Merge branch 'master' into wiggly-widgets

Conflicts:
source/blender/blenloader/intern/versioning_270.c
source/blender/editors/object/object_intern.h
source/blender/editors/object/object_ops.c
source/blender/editors/space_sequencer/sequencer_draw.c
source/blender/editors/space_view3d/view3d_draw.c
source/blender/makesdna/DNA_customdata_types.h
source/blender/makesrna/intern/rna_object.c

Merge branch 'master' into gooseberry

Conflicts:
source/blender/blenkernel/BKE_image.h
source/blender/blenkernel/intern/sequencer.c
source/blender/imbuf/intern/anim_movie.c

Merge remote-tracking branch 'origin/master' into multiview

(aka the typo fix commit in master, just to prevent having to solve this
conflict if someone else had fixed this first ;)

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

Merge commit 'origin/master~' into multiview

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

Fix crash when appending/linking (double free...).

Merge branch 'master' into asset-experiments

Conflicts:
source/blender/blenloader/intern/versioning_270.c

Fix for unfreed memory after bad merge (re: sequencer proxies)

Pedantic style edit...

Merge branch 'master' into temp_custom_loop_normals

Fix own stupid mistake in removing of 'use_current_clnors' option...

Sequencer: Implement priority queue based on frame number instead of
memory size (Thanks to sergey for suggestion)

Possibly dangerous feature - don't scale images if they come from a
proxy.

More name cleanup (spaceset -> spacearr).

Fix clnors no more importing correctly.

Found out why I needed CD_MASK_NORMAL in CD_MASK_MESH: BKE_mesh_validate() removes
any CD layer that does not match that mask! And it is mandatory for importers to
be able to call mesh.validate() without deleting those 'temp' CD layers.

So instead of 'poluting' CD_MASK_MESH here, added an option to BKE_mesh_validate
to not remove those temp CD layers...

Minor fix comments.

New space enum option for the particle instance modifier, to chose which
space the particle data should be interpreted in.

By default the space will now be `Local`, meaning that copies of the
mesh are made with the offset of the respective particle in the particle
object's local space, rather than using the world space offset of the
particle //inside the modified object space//. This behavior is much
more intuitive and consistent with true duplicators, such as face duplis.
Old files will still have `World` setting, so existing blend files are
not changed.

Don't reference NULL in RNA docs

rename vars