Bullet.jl

Author goretkin
Popularity
10 Stars
Updated Last
8 Months Ago
Started In
May 2019

Bullet.jl

Code Generation

Raw

src/generator/bullet_parse_headers.jl makes Julia wrappers to call the Bullet C API (b3 functions).

e.g.

sm = Bullet.Raw.b3ConnectPhysicsDirect()

Warning, these functions are thin wrappers around ccall calls. If you pass a [1, 1, 1] where the Bullet API expects a Ptr{Cdouble}, Bullet will see [5.0e-324, 5.0e-324, 5.0e-324] (binary representation of integers interpreted as floats. Also, I don't remember what happens if you pass in an array with different byte size).

Safe

src/generator/make_julian_api.jl makes Julia wrappers that are safer and more Julian (using e.g. ColorTypes)

e.g.

sm = Bullet.Raw.b3ConnectPhysicsDirect()
command = Bullet.Raw.b3InitPhysicsParamCommand(sm)
Bullet.Safe.PhysicsParamSetGravity(command, [0, 0, -9.8])
#= equivalent to Bullet.Raw.b3PhysicsParamSetGravity(commandHandle, 0, 0, -9.8) =#

C or C++

This package illustrates two different methods of interfacing with Bullet, because "Bullet" has multiple interfaces.

  • There is the core physics code written entirely in C++.
  • There is the command processor, which runs in its own POSIX process, that handles a special-made protocol (which could be over shared memory, or the network), and calls the core code above.
  • and then there is C code for sending/receiving command messages.

The popular library pybullet uses the last layer. The pybullet project is C code that uses the Python C API to expose a Python interface to the C code for sending/receive command messages.

There are aspects of the C++ code that are not exposed via the command processor (after all, it's supposed to be a sort of abstraction over at least some parts of the specific engine implementation), and this package tries to demonstrate how to use Cxx.jl for the purpose of accessing the physics engine directly.

Note that pybullet, taken to mean what I described above, does not do any physics calculation, and goretkin/Bullet.jl roughly rewrites at least part of what resembles PyBullet. Except that it's a bit better, because our ecosystem has some packages like ColorTypes.jl, Rotations.jl, GeometryTypes.jl that aim to be used in interfaces. Bullet.jl wraps both a Bullet C API function, e.g. setOrientation(Float x, Float y, Float z, Float w), directly, and also exposes an interface with richer data types, e.g. setOrientation(r::Rotation), allowing you to easily use whatever parameterization of rotation (quaternion, rpy, ...) you'd like in the interface (the rotation quantity will be converted to a unit quaternion).