Struct JobQueue 

pub struct JobQueue { /* private fields */ }

The job queue engine.

Thread-safe: all internal state is behind mutexes, and a Condvar wakes blocked workers when new jobs arrive.

Implementations

impl JobQueue

pub fn new(max_history: usize) -> Self

pub fn attach_store(&self, store: Arc<JobStore>)

Attach a persistent store. After this, every enqueue, state change, and terminal event is mirrored to the store. Call once at startup.

pub fn register(&self, job_name: &str, handler: JobHandler)

Register a handler for a job type.

pub fn enqueue(&self, name: &str, payload: Value) -> String

Enqueue a new job with default options. Returns the job ID.

pub fn enqueue_with_options( &self, name: &str, payload: Value, priority: Priority, delay_secs: u64, max_retries: u32, queue: &str, ) -> String

Enqueue with full options. Returns the job id, or an empty string if the persistent store rejected the write. Prefer [try_enqueue_with_options] in new code so persist failures don’t look like success to the caller.

pub fn try_enqueue_with_options( &self, name: &str, payload: Value, priority: Priority, delay_secs: u64, max_retries: u32, queue: &str, ) -> Result<String, String>

Result-returning variant of [enqueue_with_options]. Use this from any path where a silent failure would propagate as an apparent success (e.g. the TS scheduler hook returning id: "" to the user).

pub fn dequeue(&self, timeout: Duration) -> Option<Job>

Dequeue the highest-priority pending job whose delay_secs has elapsed. Blocks up to timeout if nothing is ready.

pub fn dequeue_from(&self, queue: &str, timeout: Duration) -> Option<Job>

Dequeue from a specific queue. Blocks up to timeout if nothing in the queue is ready (delay-respecting).

pub fn complete(&self, job_id: &str)

Mark a job as completed.

pub fn fail(&self, job_id: &str, error: &str)

Mark a job as failed. Retries if under max_retries.

pub fn process_one(&self) -> bool

Process the next available job using registered handlers. Returns true if a job was processed.

pub fn get_job(&self, id: &str) -> Option<Job>

Get job by ID (searches pending, running, history, dead letters).

pub fn stats(&self) -> QueueStats

Get queue statistics.

pub fn pending_count(&self) -> usize

Get pending job count.

pub fn running_count(&self) -> usize

Get running job count.

pub fn dead_letters(&self) -> Vec<Job>

Get dead letter queue contents.

pub fn retry_dead(&self, job_id: &str) -> bool

Retry a dead letter by moving it back to pending.

pub fn recent_history(&self, limit: usize) -> Vec<Job>

Get recent job history.

pub fn list_jobs( &self, status: Option<&str>, queue: Option<&str>, limit: usize, ) -> Vec<Job>

List pending jobs with optional status/queue filters.

pub fn restore_from(&self, store: &JobStore) -> usize

Restore pending/running/retrying jobs from a persistent store.

Jobs that were Running at the time of the crash are reset to Pending so they will be re-processed. Returns the number of jobs restored.

Call this once at startup, before workers begin processing.

Trait Implementations

impl JobOps for JobQueue

fn enqueue( &self, name: &str, payload: Value, priority: &str, delay_secs: u64, max_retries: u32, queue: &str, ) -> String

fn stats(&self) -> Value

fn dead_letters(&self) -> Value

fn retry_dead(&self, id: &str) -> bool

fn list_jobs( &self, status: Option<&str>, queue: Option<&str>, limit: usize, ) -> Value

fn get_job(&self, id: &str) -> Option<Value>