[][src]Derive Macro big_brain_derive::Action

#[derive(Action)]
{
    // Attributes available to this derive:
    #[action]
}

Actions in big-brain are defined through this derive macro. Once defined, they can be freely used in a .ron file. They define actual behaviors that an actor will perform when the Thinker engine picks it as the active action, based on Considerations.

Definition Example

This example is not tested
use specs::{Component, Entity, System, WriteStorage};
use big_brain::{Action, ActionState};

// These are your game's components.
use crate::components;

// This will be used to create `Action` components. They MUST implement the
// `specs::Component` trait.
#[derive(Debug, Clone, Component, Action)]
pub struct Eat {
    // All actions **must** have a public `actor` field. This will be populated
    // with the actual actor performing the Action. The `Entity` associated with
    // the `Action` itself is distinct from the actor.
    pub actor: Entity,

    // `default` fields will be populated using default::Default() when the
    // Action is instantiated. These cannot be used as params.
    #[action(default)]
    pub foo: f32,

    // `param` fields will be populated using the value passed in through the
    // `.ron` file.
    #[action(param)]
    pub reduce_by: f32,
}

// Once an Action component is defined, we define a System that can act on it.
pub struct EatSystem;

impl<'a> System<'a> for EatSystem {
    type SystemData = (
        WriteStorage<'a, components::Hunger>,
        // This is the actual Eat component.
        WriteStorage<'a, Eat>,
        // An ActionState component is attached to every Action Entity.
        // It contains the current running status of the Action, and will be
        // updated as needed by the actor's Thinker.
        WriteStorage<'a, ActionState>,
    );
    fn run(&mut self, (mut hungers, mut eat_actions, mut states): Self::SystemData) {
        // You can join the Eat and ActionState together. They're on the same component.
        for (state, eat_action) in (&mut states, &mut eat_actions).join() {
            // Any components attached to the actor must be fetched separately.
            if let Some(hunger) = hungers.get_mut(eat_action.actor.clone()) {
                match state {
                    // At the very least, every Action should handle the
                    // `Requested` state.
                    ActionState::Requested => {
                        hunger.hunger -= eat_action.reduce_by;
                        // Success tells the Thinker that this action succeeded!
                        *state = ActionState::Success;
                    }
                    // Make sure to handle Cancelled for long-running Actions.
                    // The Thinker will not continue until the state is either
                    // Success or Failure.
                    ActionState::Cancelled => {
                        *state = ActionState::Failure;
                    }
                    _ => {}
                }
            }
        }
    }
}

Usage Example

This example is not tested
(
    picker: {"FirstToScore": ()},
    // Actions are defined using the `then` param to Choices
    choices: [(
        consider: [{"Hunger": ()}],
        // We can use the param defined in our derive definition here.
        // The `foo` field will be defaulted and cannot be defined here.
        then: {"Eat": (reduce_by: 80.0)},
    )]
)