mod_plugins

Crates.iomod_plugins
lib.rsmod_plugins
version
sourcesrc
created_at2024-03-23 07:39:22.311412
updated_at2024-12-03 20:47:03.339273
descriptionReduce plugin boilerplate in the Bevy game engine.
homepage
repositoryhttps://github.com/DaylightNebula/mod_plugins_bevy
max_upload_size
id1183408
Cargo.toml error:TOML parse error at line 17, column 1 | 17 | autolib = false | ^^^^^^^ unknown field `autolib`, expected one of `name`, `version`, `edition`, `authors`, `description`, `readme`, `license`, `repository`, `homepage`, `documentation`, `build`, `resolver`, `links`, `default-run`, `default_dash_run`, `rust-version`, `rust_dash_version`, `rust_version`, `license-file`, `license_dash_file`, `license_file`, `licenseFile`, `license_capital_file`, `forced-target`, `forced_dash_target`, `autobins`, `autotests`, `autoexamples`, `autobenches`, `publish`, `metadata`, `keywords`, `categories`, `exclude`, `include`
size0
(DaylightNebula)

documentation

README

Module Plugins for Bevy

This is an experiment/project in removing more boilerplate from the Bevy game engine plugins system. While the plugins system does not have much boiler plate, this is an experiment in reducing it even more using an attribute macro. This is achieved by applying the plugin attribute macro to a mod module in your code, the mod block is effectively removed and a plugin in generated. Here is an example of this:

// Create a plugin named `TestPlugin` (snake case names of the mod is turned into cammel case).
#[plugin]
mod test_plugin {

    // This system will be run in the `Startup` schedule.
    #[startup]
    fn setup() { some system ... }
}

Marker Attributes

Systems, functions, structs and enums can be marked by attributes (like #[startup] above) that apply various plugin related functionality.

Startup and Update Systems

The Startup and Update schedules are crucial to Bevy plugins. So to add systems to those schedules simply mark them with #[startup] or #[update].

// Create a plugin named `TestPlugin`.
#[plugin]
mod test_plugin {

    // This system will be run in the `Startup` schedule.
    #[startup]
    fn setup() { some system ... }

    // This system will be run in the `Update` schedule.
    #[update]
    fn update() { some system ... }
}

Run on Enter/Exit State Systems

Being able to run systems on enter and exit from states in central to Bevy. Usually this can be done with the OnEnter(<some state>) and OnExit(<some state>) schedules. This can be done with mod plugins by applying the following to your system #[enter(<some state>)] or #[exit(<some state>)] just like you could with #[startup] or #[update] above.

Run on Event Systems

When using Bevy, it is common to run something when an event is fired. This marker attribute can be used to run a system when an event is fired. This attribute is #[event(<some event type>)]. These systems will automatically read the Res<Current<the same event type>> resource that is temporarily added to the world to run these systems so that these systems have access to the event that was fired.

// Create a plugin named `TestPlugin`.
#[plugin]
mod test_plugin {
    // This system is run every time a `KeyboardInput` is fired.
    #[event(KeyboardInput)]
    fn keyboard_input() {
        println!("Input {current:?}");
    }
}

Resource Initialization Factories and Systems

Bevy plugins are also responsible for initializing resources, while you can initialize resources by their default implementation which we will discuss later, the plugin macro gives two two marker attributes that can be used to initialize those resources.

The first is #[resource_factory] which allows you to mark a function that takes no inputs and return a created resource. This function is ran when the plugin is built, and adds the returned resource to the app. Here's an example:

#[plugin]
mod test_plugin {
    // When the plugin is built, this creates and adds `ResourceB` to the app.
    #[resource_factory]
    fn create_b() -> ResourceB { ResourceB(2) }
}

The second is #[resource_system] which allows you to mark a system that returns a resource. That system is run on startup and the returned resource is added to the world. Here's an example:

#[plugin]
mod test_plugin {
    // On `Startup`, this creates and adds `ResourceC` to the app.
    #[resource_system]
    fn create_resource(
        mut commands: Commands,
        a: Res<ResourceA>,
        b: Res<ResourceB>
    ) -> ResourceC {
        // camera
        commands.spawn(Camera3dBundle {
            transform: Transform::from_xyz(-2.5, 4.5, 9.0).looking_at(Vec3::ZERO, Vec3::Y),
            ..default()
        });

        ResourceC(a.0 + b.0)
    }
}

Build Functions

Sometimes, however, it may be necessary for you to access the app when the plugin is built like you would with a normal Bevy plugin. You can do this by marking a function that returns nothing and takes in a mutable reference to App marked with #[build]. Here's an example of how you can do this:

#[plugin]
mod test_plugin {
    // This is run when the plugin is built.
    #[build]
    fn on_build(app: &mut App) {
        ... do whatever with the app
    }
}

Auto-Init Events

Plugins need to be able to add their events to the App. You can do this by adding the #[init_event] marker attribute to the event you created in a #[plugin] mod and the event will be added to he App by the plugin when it is built.

#[plugin]
mod test_plugin {
    #[init_event]
    #[derive(Event)]
    pub struct SomeEvent(pub i32);
}

Auto-Init Resources

Resources have the option of being initialized by their Default implementation that they may have. To do this, just mark the resource that is created in a mod marked with #[plugin] with #[init_resource]. The resource will then be added to the App by their Default implementation when the plugin is built.

#[plugin]
mod test_plugin {
    #[init_resource]
    #[derive(Resource)]
    pub struct SomeResource {
        name: String,
        num: i32
    }
}

Auto-Init States

Plugins are also responsible for adding States to the App when they are built. You can do this by two ways, either by the States Default implementation or by specifying a starting State. You can do this by marking the State created in a #[plugin] mod with #[init_state] if the State has a Default implementation. Otherwise, you will need to specify the starting State by marking the State with #[init_state(State::Kind)].

#[plugin]
mod test_plugin {
    #[init_state]
    #[derive(States, Clone, Debug, Default, Hash, PartialEq, Eq, PartialOrd, Ord)]
    pub enum ExampleState {
        #[default]
        StateA,
        StateB
    }

    ----------- OR -----------

    #[init_state(ExampleState::StateA)]
    #[derive(States, Clone, Debug, Hash, PartialEq, Eq, PartialOrd, Ord)]
    pub enum ExampleState {
        StateA,
        StateB
    }
}

Register Types

For reflection, you will need to register types with the App. Usually you would do this by calling the register_type function on the App when the plugin is built. You can do this in #[plugin] mod by marking a struct to register with the #[register] attribute marker.

#[plugin]
mod test_plugin {
    #[register]
    pub struct SomeType(pub u32);
}

Executables

Something useful could be adding systems to data like structs or enums. This would be useful for generic types where different systems may have to be run depending on the actual type. For example, a server sends an "action" to the client, and the client runs the system to apply that "action". Here's how you could make a struct an "executable" struct:

pub struct ExecutableData {
    name: String,
    data: i32
}

#[executable(ExecutableData)]
fn execute_data() { ... some system }

To execute the above system, I could take an instance of ExecutableData and call its execute function that has been implemented for ExecutableData which takes a mutable reference to World. This runs the system for the given world.

fn use_executable(world: &mut World, data: ExecutableData) {
    data.execute(world);
}
Commit count: 48

cargo fmt