Struct Worker 

pub struct Worker { /* private fields */ }

Polls a queue and executes registered handlers with bounded concurrency.

Implementations

impl Worker

pub fn new(queue: &Queue) -> Self

Creates a worker with 16 concurrent slots and five-minute leases.

pub fn poll_interval(self, milliseconds: u64) -> Self

Sets the idle polling interval in milliseconds.

pub fn max_concurrency(self, limit: usize) -> Self

Sets the hard upper bound for simultaneously executing handlers.

pub fn job_timeout(self, timeout: Duration) -> Self

Sets the maximum duration of an individual handler execution.

A handler still running at the deadline is aborted and its job is failed as timed out. A handler that cannot be interrupted (for example one that blocks its thread) and then returns is recorded from its own result instead.

pub fn stalled_after(self, age: Duration) -> Self

Sets the age after which a processing lease counts as stalled.

The worker returns stalled leases to pending when it starts and then every min(age, 60 s) (at least every second). Recovery is queue-wide: it covers every processing lease in the shared SQLite table or Redis namespace, including leases held by other workers. age must exceed this worker’s Self::job_timeout, and every worker sharing a queue must use an age longer than the longest job_timeout among them, or a worker with a short age requeues another worker’s running job and it runs twice.

pub fn register<F, Fut>( &mut self, name: impl Into<String>, handler: F, ) -> &mut Self

where F: Fn(Value) -> Fut + Send + Sync + 'static, Fut: Future<Output = Result<(), Box<dyn Error + Send + Sync>>> + Send + 'static,

Registers a handler for a job name.

pub fn run(&self) -> Result<WorkerHandle, QueueError>

Starts the polling loop and returns an observable lifecycle handle.

Errors

Returns a typed error for zero concurrency, a zero polling interval, or invocation outside an active Tokio runtime. No task is spawned on error.