Core & Components

The core code is designed to remain untouched and unexpanded. Its members are first-class citizens of the asset and must have no knowledge of the behaviour components. These are the three core members and their responsibilities:

1. Prop

   • Owns the RigidBody3D target body and holds a reference to the active Interactor, acting as the central hub between physics, input and behaviour.

   • Dispatches interaction lifecycle events (start, end) and camera movement requests as signals, decoupling core logic from behaviour implementation.

2. Interactor

   • Casts a ray each physics frame, either from the screen center (CAMERA_MODE) or from the cursor position (CURSOR_MODE), to detect and bind Prop nodes within reach.

   • Tracks and exposes input state (pressed, just_pressed, released) and mouse delta for consumption by behaviour components via get_action_state() and get_mouse_delta().

3. PropBehaviour

   • Defines the base interface for all behaviour components, exposing a reference to the owning Prop and its active Camera3D through the setup() method.

   • Declares the execute() method as the single entry point called each frame by Prop.update_interaction(), to be overridden by each concrete behaviour.

The components code is considered “second-class”, inheriting from PropBehaviour and meant to be modified, expanded and/or replaced. Behaviour components can (and should) know about the internals of core members, but not the other way around.

Currently there are three implementations of behaviour components, though more may be added in the future:

1. GrabBehaviour

2. SpinBehaviour

3. ThrowBehaviour