Blender Git Loki

Fix T47862: VSE hard cut fails

Regression caused by changing how video length is calculated,
however similar problems could be caused by strip length changing on-disk.

GPU: avoid redundant logic for non-spot lamps

Fix for non spot lamps calculating spot values

Logical error in D1378, also incorrectly used lamp type as a flag.

Include requests' cacert.pem file

This allows us to verify certificates of HTTPS connections, which is
mandatory for logins like on Blender ID.

Reviewers: campbellbarton

Differential Revision: https://developer.blender.org/D1845

Fix T47670: cycles GLSL incorrect normal map node.

Fix T47670: cycles GLSL incorrect layer weight / fresnel.

World space and view space normals were mixed up, we should only convert from
world to view space if a custom normal is connected, otherwise it is already in
view space.

fix for rare crash at loading stored constraints

added a null pointer check so to prevent a crash atleast

another fix attempt for occasional loading crashes

fix, forgot to set userpointer in point constraint

fix, forgot to load participating constraints per meshisland

this caused crashes when trying to simulate with percentage breaking

fix attempt for crash on loading last blend (quit.blend with dynamic fracture)

crash fix for loading files with dynamic fracture enabled

forgot to set runtime listbase pointers of fractureIDs to NULL

fix for occasional crashes with dynamic fracture while freeing the modifier

crash happened with scene objects which had a freed rigidbody world but didnt set the pointer to NULL.

fix for loading splitshards to island blends and clusters properly

was necessary to skip "empty shards" properly and keep cluster index

a bit clean up and attempt to reduce dynamic fracture memory leaks

dont suppress undo save/load with dynamic any more

this seems a bit more stable, but does crash occasionally too, and may even
corrupt the blend. Workaround, open with official blender and FM is gone, save
and blend should be usable again.

fix for different constraint behavior after refracturing / reloading / param change

constraint enabled flags were changed during sim and next sim started with pre-broken
constraints this way... thus different behavior.

store constraints for prefractured mode now too in blend

this automatically creates constraints from scratch as fallback for older files

Fix T47758: rigidbody bug with constraint breaking and disable collisions.

This is a regression in Bullet, reverted the problematic change for now with
a custom patch (backported to Fracture Modifier)

allow multiple dynamic destruction with extra particles or verts

this should allow triggered dynamic destruction with different colliders at different times. Note: WIP and crashes often in relation with undo