Struct wadm::workers::EventWorker

pub struct EventWorker<StateStore, C: Clone, P: Clone> { /* private fields */ }

Implementations

impl<StateStore, C, P> EventWorker<StateStore, C, P>

where StateStore: Store + Send + Sync + Clone + 'static, C: ClaimsSource + InventorySource + LinkSource + Clone + Send + Sync + 'static, P: Publisher + Clone + Send + Sync + 'static,

pub fn new( store: StateStore, ctl_client: C, command_publisher: CommandPublisher<P>, status_publisher: StatusPublisher<P>, manager: ScalerManager<StateStore, P, C> ) -> EventWorker<StateStore, C, P>

Creates a new event worker configured to use the given store and control interface client for fetching state

Trait Implementations

impl<StateStore, C, P> Worker for EventWorker<StateStore, C, P>

where StateStore: Store + Send + Sync + Clone + 'static, C: ClaimsSource + InventorySource + LinkSource + Clone + Send + Sync + 'static, P: Publisher + Clone + Send + Sync + 'static,

type Message = Event

The actual message type to expect, such as Event

fn do_work<'life0, 'async_trait>( &'life0 self, message: ScopedMessage<Self::Message> ) -> Pin<Box<dyn Future<Output = WorkResult<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Process the given work to completion. Almost all errors returned are things that could be retried. But if for some reason a fatal error occurs, return WorkError::Fatal to indicate that work should stop. Any worker MUST handle acking the message (or passing it to another worker). By default, when a ScopedMessage is dropped, it will nack it