Blender Git Loki

Fix #30231: Drag and Droping Parenting from Outliner giving unnecessary Loop Error

In fact, error was much deeper and DND parenting in outliner used to parent all selected
objects to drop target, not just currently dragging object.

This was caused by code sharing between Parent Set operator and this DND operator which
was iterating all selected objects in scene and was setting parent to it.

Solved issue by separating actual parenting code which makes specified object be parented
to specified parent and moved iterating to Parent Set exec callback.

Now both of discovered issues (unneeded loop error and parenting all selected objects)
are solved, but more extensive testing of this changes are welcome.

Attempt to fix scons compilation of blenderplayer.
Added bmesh to player libraries and changed the priority of blenlib so as to avoid duplicate definitions (looks like there are actually functions defined twice that cause conflicts if one changes the order of linking...figures).

Only tested this on mingw-windows, I hope it works elsewhere too.

Stitch tool fully functional again. Many thanks to Howard Trickey for proposing the loop winding criterion for normal disambiguation of edges. Unfortunately some extra memory has to be allocated for this to work correctly. If the tool had been initially written for bmesh I would have used the already present adjacency information to make it work, avoiding some extra allocations. Maybe a project for another day though, when I am more proficient with bmesh internals.

change UvElement to directly use BMLoop * instead of tfindex. This saves quite some lookups on uv sculpting and stitching. Based on BMESH todo by Campbell, thanks for the idea!

uv stitch static island highlight ported to bmesh system.

Unfortunately since we are missing a way to tesselate uvs, the result will look good only on concave faces since we are essentially using a triangle fan to draw. When a tesselation of uv polygons is possible the tool should also change to use that instead. This will also influence the preview drawing code, coming next.

Text fix:
Added #blender to list of recommended irc channels.

Remove debug print code.
It was harmless because it ifdef-ed, but it's likelly not needed anymore.

Remove unused code left from old multires interpolation stuff

- Removed validate() callback for MDisps layer. It wouldn't actually work
correct from CustomData layer and all needed data might be validated from
BMesh interpolation level. Also this callback was never actually used in
BMesh, so can't see why we'll want to have it in structures.
- Removed layrInterp_mdisps callback. Interpolation now happens from
another level (bmesh_interp) and this callback isn't needed anymore.
- Removed all function from multires.c which were used by old interpolation
stuff and seems to be useless for other usages.
- multires_topology_changed is still marked as a TODO, Probably it's not
needed anymore, buy better to keep for now until it'll be 100% clear this
function isn't needed and all needed re-allocations happens in bmesh_interp.
Otherwise, it'll be needed to be ported to new system.

- Remove unused code in bmesh interpolation module and switch it to using mathutils
functions instead of own macros for vectors.
- Double precision isn't needed here at all. It only makes it impossible to use
mathutils, adds extra float<->double conversions without any benefits.
(current multires interpolation is already an approximate method and it can't
be improved using double precision)

Improved displacement flipping algorithm to work properly in cases when target grid isn't orthogonal.

Subdivision of face now works pretty smooth, only producing minor unwanted displacement on the along edges.

Flip displacement direction after interpolation in bmesh_loop_interp_mdisps

This is needed because displacement might been interpolated from a grid
with different orientation and in this case X and Y components of displacement
are need to be flipped in needed order.

Order of flipping is determining by projecting source grids axis orientation
on target grid axis. This probably will give some unwanted artifacts when
interpolating non-planar face but currently can't think about better way
to determine how to flip displacement.

This commit makes operators like Subdivide works much more unpredictable
for sculpting data, but this stuff should be rethinked much more global
because current approach is not acceptable.

Fix part of #30272: cycles preview render not updating when changing particle
emission from with instanced objects.

Fix [#30281] Left/Right vertex weights flipped.

Mirror modifier was flipping vgroups on org vertices, not mirrored ones!

Fix #30270, #30265: cycles not displaying textured objects, and not rendering
text/curve objects after the bmesh merge. Also removed a debug print.

Somehow typo was made just before commit in previous patch.

Camera tracing: search area was always centered to marker's position when tracking

Another fix for non-VBO flat-shading in sculpt mode, this time for non-multires meshes.

As with multires, this change calculates face normals rather than
using vertex normals when the node is flat-shaded.

Flat-shading with VBO on non-multires meshes is still wrong, but
fixing that would require larger changes to our vertex buffers.

Remove call to GPU_update_mesh_buffers from GPU_build_mesh_buffers.

Building the mesh buffers already gets the PBVH_UpdateDrawBuffers flag
set, so this was double-updating the vertex buffer.

fix [#30264] B-Mesh: Crash when unwrapping ngon mesh with Smart UV project

Merged changes in the trunk up to revision 44288.

Conflicts resolved:
source/blender/blenkernel/intern/cdderivedmesh.c