Modular implementation of two or more components

[H4ppy-04]

Reading through the breakout example I've noticed that single functions have been chosen to handle the functionality of all components¹.

• Collision for multiple sprites are all handled in the check_for_collisions function.
• The apply_velocity function manages the speed of multiple components.

This begs the question: What if components have individual functionality such as differing speed and/or acceleration? Movement and any other attributes that apply to specific components would have to be handled on an individual level.

Given components with individual properties:

#[derive(Component)]
struct Projectile {
    position: Vec2,
    velocity: Vec2
}

#[derive(Component)]
struct Enemy {
    health: i32
}

How might one go about using impl to implement individual functionality for Projectile and Enemy? Thank you in advance.

Footnotes

1. This has been proven to be the most efficient approach in this context. ↩

[Vrixyz]

You have different options:

• You can add more data, so you can fine tune each entity.

/// The current acceleration, so the speed can augment progressively
#[derive(Component)]
struct Acceleration {
    current: f32,
}
/// a configuration for each entity, like the acceleration rate.
#[derive(Component)]
struct AccelerationChange {
    change_per_frame: f32,
}
/// Another configuration to know when the acceleration is too much, or not get an unsupported speed.
#[derive(Component)]
struct MaxSpeed(f32);

• Those additional data could also be stored in a Resource, might be interesting if you have multiple similarly configured entities.
• If you need more specific logic, it might be practical to "specialize" each system, by adding marker components:

#[derive(Component)]
struct MarkerFastProjectile;
#[derive(Component)]
struct MarkerSinProjectile;

fn system_move_fast(mut q_projectiles: Query<&mut Transform, With<MarkerFastProjectile>>) {
  for t in q_projectiles.iter() {
    // We could modify Speed, to defer actual movement to another system
    t.translation += Vec3::X * 5f32;
  }
}

fn system_move_sin(time: Res<Time>, mut q_projectiles: Query<&mut Transform, With<MarkerSinProjectile>>) {
  for t in q_projectiles.iter() {
    t.translation += Vec3::X * 1f32 + Vec3::Y * time.elapsed_seconds().sin();
  }
}