improve serialization based on feedback: make it easier to serialize a single object, or single shape, or single bvh.
Modified Bullet/Demos/ConcaveDemo to use btSerializer to serialize the btOptimizedBvh (or optionally the trimesh+bvh)
You can also serialize trimesh without BVH, use:
serializer->setSerializationFlags(BT_SERIALIZE_NO_BVH);// or BT_SERIALIZE_NO_TRIANGLEINFOMAP
Add flags to disable bvh and triangleinfomap serialization
btSerializer::setSerializationFlags(flags)
flag can have value BT_SERIALIZE_NO_BVH, BT_SERIALIZE_NO_TRIANGLEINFOMAP
move btTriangleInfoMap.h into its own header, into the BulletCollision/CollisionShapes folder
add btTriangleInfoMap pointer to btBvhTriangleMeshShape, so we don't need to use the 'userpointer' anymore
preparation for serialization of btTriangleInfoMap.
See also http://bulletphysics.org/Bullet/phpBB3/viewtopic.php?f=9&t=4603&start=30
So it is different from the in-place method (btQuantizedBvh::serializeInPlace/deserializeInPlace).
It is also possible to serialize/deserialize just the bvh, using the btSerializer (needs some code snippet/helper)
See also http://bulletphysics.org/Bullet/phpBB3/viewtopic.php?f=9&t=4770
updated serialization structures and header files (for new 'm_name' field, replacing user pointer)
fix issue with btGeneric6DofConstraint::setParam in headerfile
(no motor support or advanced settings yet)
added btStaticPlaneShape serialization
Added toggle in cmake for BenchmarksDemo to enable/disable graphics rendering
See also request here: http://code.google.com/p/bullet/issues/detail?id=325
Also added a warning. This warning also applies to all methods that change the collision shape:
"changing a collision shape while the body is in the world is not recommended,
it is best to remove the body from the world, then make the change, and re-add it
alternatively flush the contact points, see documentation for 'cleanProxyFromPairs'"
Minor fixes (#ifndef SPU_COLLISION_OBJECT_WRAPPER_H block etc)
Only update 'setNumTasks' when necessary
Revert vectormath.h include (can't resolve conflicts with PS3 SDK, need additional #include path)
Make btTypedConstraint and btPersistentManifold both derive from btTypedObject to make SPU-side generic constraint solver easier.
Note: all build systems need to be updated: remove SpuVoronoiSimplexSolver.cpp, SpuGjkPairDetector.cpp, SpuEpaPenetrationDepthSolver.cpp, SpuGjkEpa2.cpp
+ Allow user to enable useConvexConservativeDistanceUtil . Use dynamicsWorld->getDispatchInfo().m_useConvexConservativeDistanceUtil = true;
(see Demos/Benchmarks/Benchmark4 (convex objects falling down)
+ Fix for plane drawing (just wire-frame)
+ Gimpact: use collision margin of 0.07 for demo (because BULLET_TRIANGLE_COLLISION is used)
+ replace dot,cross,distance,angle,triple in btVector3 by btDot, btCross,btDistance,btAngle,btDistance to avoid naming conflicts
+ Some fixes in GJK penetration depth normal direction (broken in a previous commit)
+ fix in calculateDiffAxisAngleQuaternion to make ConvexConservativeDistanceUtil work properly
+ allow debug drawing to debug btContinuousConvexCollision
+ add comment/warning that btTriangleMesh::findOrAddVertex is an internal method, users should use addTriangle instead
If user wants more accurate inertia tensor for multi-sphere shape, use btCompoundShape instead
(amd btCompoundShape::calculatePrincipalAxisTransform)
Thanks rcharlton for bringing this up.
Erwin
1) add AabbCaching versions of btPolyhedralConvexShape and btMultiSphereShape (this speeds up btMultiSphereShape 'getAabb', and reduces size of btBoxShape)
2) btCylinderShape doesn't derive from btBoxShape anymore
+ Minor fixes in drawing for btMultiSphereShape, btBoxShape.
+ Don't re-generate btDebugFont every frame
+ Disabled velocity prediction for btDbvtBroadphase. Previous default can be restored using btDbvtBroadphase->setVelocityPrediction(1./2.);
+ applied patch to mix double/single precision meshes independent from double/single precision Bullet build
Thanks to Ole for the patch, http://code.google.com/p/bullet/issues/detail?id=213
+ re-enable warming starting in constraint solver, it was disabled by accident
+ fix btConvexHullShape constructor, so accept vertices with non-16-byte striding
Thanks Shawn Baird for report and fix: http://code.google.com/p/bullet/issues/detail?id=204
