Blender Git Loki

RNA function for defining point cache API.

Set the WITH_ALEMBIC define in subfolders that actually use it, follows the pattern in other libs.

Some test code for defining a particles schema, based on I/OPointsSchema of alembic.

Add WITH_ALEMBIC as a cmake option and set a WITH_ALEMBIC define to use in C code.

Library linking for Alembic.

Very basic sub-folder structure for new pointcache implementation.

Code cleanup: As Sergey stated on bf-committers one should use id_us_min() if the user counter has to be decremented and not decrement the counter by hand.

Correct fix for T37617, (own recent regression)

Fix T37617: "Add plane" was adding a 2*2 grid

Was a regression from rBaa3c06b41ca9, hope this time all things are OK again (note the X/Y subdivision values still are different than before (-1 for same result), but imho they make more sense this way).

Code Cleanup: rename vars for detecting change to be more consistent

rename change/is_change/is_changed/modified -> changed
also use bools over int/short/char and once accidental float.

fix: marking edges as sharp in modifier did not work any more

partial fix for fracturing meshes where origin is not in bounds center

some hacks to keep baked files with clipping working...

hopefully fixed atleast the fracture with text converted to mesh/remeshed and objects whose origin is not at bounds center

changed (i think corrected) the centroid calculation for fractured objects with boolean

fixing crash at refreshing fractured text, cause was improper handling of the necessary temp object for boolean function (yuck)

attempt to make text and curves work with rigidbody modifiers too, still failing (flying apart... crashes...)

fix for autorefresh and depgraph callback, strangely did only work for the testblend and not for others...

using direct function for dissolve in object conversion too

using dissolve function now directly to circumvent (hopefully) the va_arg problems under windows