Blender Git Loki

disallow adding tessfaces to a mesh with polygons (only allowed case is creating a new mesh with tessfaces and later converting to polygons, which wont work if polygons exist)

setting the active texture layer from python would get the UV layer out of sync. (entering editmode would show the wrong UV layer)

style cleanup: editors/mesh - mostly whitespace around operators

fix [#30657] New UV layers created with Mesh.uv_textures.new reset previous ones.

adding UV's from python was resetting the active UV layer but not setting the new layer to active, resetting existing UV's.

rename lattice influence to strength from r45144 (other deform modifiers call it strength too)

fix [#30672] Mesh menu duplicity

own fault - added this before desolve menu existed.

style cleanup: mostly whitespace around operators

Code style cleaup for motion-tracking modules.
Should be no functional changes.

style cleanup: dont use 'else for' for extended blocks of code (quite confusing)

style cleanup: add braces around checks - 'if ELEM() {...}', confuses some parsers that done expand macros.

Patch: [#30652] Influence slider for Lattice Modifier

* This patch adds a influence slider for the lattice modifier, which affects the strength of the deformation.

Patch by Patrick Boelens (senshi), thanks a lot!

fix #30668 and the rest of #30638.

Port angle stretch code for uvs to bmesh.

Also added comment to clarify previous commit.

Comment out the projection to ngon normal plane when calculating ngon area. This is hardly worth the cost since we use an approximation anyway.

Fix #30638 and part of #30646.

Problem was that area calculation of polygons was done relative to the xy plane, and with a very obscure (to me at least) algorithm. That meant that vertical ngons would get 0 area.

Commented initial code in case this is a strange optimization case that someone wants to use and used a cleaner algorithm: first project vertices to the ngon plane, defined by the normal of the ngon and the center (mean) of the ngon vertices. This will only be exact for convex and mostly planar ngons, still it is much better than the previous code.

Also fixed memory leak when stretch display was on.

Fix bug #30673, "Crash: Bridge a pair of edges."

Fix edge case for clamp_index() with any negative 'x' that is a
multiple of 'len', was returning 'len' which is invalid index.

Maybe the expression can be simplified back to a one-liner?

Cycles / CUDA:
* Make Cycles aware of Computing Capability 3.0, used by the new Kepler Cards.
You'll need the CUDA 4.2 Toolkit to compile it.

style cleanup: issues missed last commit

bmesh inset: another small improvement - use the shared edge vector rather then the cross product between 2 faces if the faces infact share an edge - works best for non planer faces.

also added utility function - BM_loop_other_vert_loop

inset tool now works better when insetting around corners - the 2 faces normals are now used to calculate the inset edge vector if the faces are different and not planer.

style cleanup: conform to style guide - mostly operator whitespace changes