Blender Git Commits

added support for inner vertex group... WARNING: this is still UNSTABLE

forgot to change one BLI_temporary_dir() call to BLI_temp_dir_base()

change: rebuilding constraints should work now, BUT you have to put a modifier on all rigidbody objects now, otherwise the regular active rigidbody object will be reset when continuing an interrupted simulation, need to investigate why...

fix for re-enabling the constraints, takes now all objects into account in the first sim step

fix for constraint update bug, all constraints must be re-enabled after a fracturing run if they are not recreated

removed compound shape from RNA (did not work with modifier)

crash fix with other pointsources, forgot to initialize counts to 0...

fix for losing texture images after load (pink object) but this still adds the image to the inner faces unintendedly, only should affect viewport and viewport render, glsl and rendering looks ok

convert to objects fix: now should take textures into account too

windows compile fix: replace BLI_temporary_dir() by BLI_temp_dir_base()

fix for: fix normals, can use modifiers before fracture now too to get correct result

revert last change again, was just a test

gitmodules test

merge related compile fixes

Merge branch 'master' into fracture_modifier_oldsim

Conflicts:
extern/CMakeLists.txt
source/blender/blenkernel/intern/rigidbody.c
source/blender/editors/interface/interface_templates.c
source/blender/editors/object/object_edit.c
source/creator/CMakeLists.txt

forgot to initialize totvert variable prior to building a kdtree

fix attempt for halving case, passive groups didnt work properly there

added inner material support for boolean and bisect fill

added a fix normals option to let cracks in smooth objects nearly disappear, also rotates the vertex normals with the verts... WIP... because i need somehow to get the derivedmesh of the previous modifier in stack if there is one

added vertweight display for fracture modifier, and attempt to update data in case vgroups are changed, but this doesnt work, seems to require a refracture to interpolate data again