Blender Git Commits

more work on getting bmesh dirty index flags reliable.

now there very close, some hard to redo cases to check on.

minor changes
- comment BM_SetIndex as setting dirty.
- have alloc_flag_layer validate the index data and clear the dirty flag since it loops on the data anyway.

fix for minor memory leak for BMO_VInitOpf() in an error case, also use the BLI version of strdup().

basic bevel modifier working again in bmesh, closely matches the edge split modifier since it calls out to the bevel operator.
- edge angles are ok
- width is interpreted as percent

added BMESH_TODO comments with details about what needs working on.

change edge split modifier to loop over all edges and compare the face angles, rather then looping over all faces and looking at every faces-edges-fance which would compare faces twice.

use lower angle when checking to draw an edge, adding uv sphere looked strange

Superficial changes for bmesh bevel to sync with bmesh branch, also noticed length comparison can be done without sqrt for baking.

when comparing lengths no need to sqrt

2 of the 3 places that check 2 edges share a vertex were incorrect (silly mistake)

so make into a function: BM_Edge_Share_Vert()

svn merge ^/trunk/blender -r42194:42197

remove redundant check

quiet some warnings and logical errors.
- curve map insert point had a nested loop which used the same value to index different arrays.
- dynamic paint used ternary operator where both outcomes were the same.

svn merge ^/trunk/blender -r42182:42194

BLI_mempool_findelem() only worked when no elements were freed, use the iterator for now.

Merging r42178 through r42192 form trunk into soc-2011-tomato

Do not use search area of unselected tracks when searching for nearest
track on mouse selection operator.

Buildbot configuration tweaks
- Force libsamplerate linking statically
- Remove global compiler flags for sse2 and sse.
Tests of release builds didn't show big difference in time of
tracking using SAD algorithm or time of solving,
so this shouldn't introduce big slowdown.

Camera tracking: refactor of tracks synchronization stuff

Currently this stuff is used for 2d tracking job only, but the same
thing would necessary for camera solver job, so moved it into more
general utility stuff.

enable multires baking code, this works between different multires levels but currently not baking to level 0, added TODO in code.

accessing rna 'default_flag' attribute of a non enum-flag could crash. eg.

bpy.types.Modifier.bl_rna.properties["type"].default_flag

now check the default/default_flag match the enum property they are used with.