@arkstack/jobs

Jobs module for Arkstack, providing dispatchable job classes and a dispatcher on top of @arkstack/queue.

@arkstack/jobs is the authoring layer; @arkstack/queue is the transport. This package gives you the Job base class, the dispatch() helper, and a registry so workers can reconstruct your job classes from a stored payload.

Writing a job

ark make:job SendWelcomeEmail

// src/app/jobs/SendWelcomeEmail.ts
import { Job } from '@arkstack/jobs'

export class SendWelcomeEmail extends Job {
  constructor(public userId: number) {
    super()
  }

  async handle() {
    // ... send the email
  }
}

Dispatching

import { dispatch } from '@arkstack/jobs'
import { SendWelcomeEmail } from '@app/jobs/SendWelcomeEmail'

// static helper
await SendWelcomeEmail.dispatch(user.id)
await SendWelcomeEmail.dispatch(user.id).onQueue('mail').withDelay(60)
await SendWelcomeEmail.dispatchSync(user.id) // run inline now

// functional helper
await dispatch(new SendWelcomeEmail(user.id))
await dispatch(new SendWelcomeEmail(user.id), { queue: 'mail', delay: 60 })

With the default sync queue connection the job runs immediately. Configure a database or redis connection (see @arkstack/queue) and run a worker to process jobs in the background:

ark queue:work

Setup

Importing @arkstack/jobs anywhere wires the queue (de)serialization automatically. For an explicit bootstrap hook (next to @arkstack/database/setup):

import '@arkstack/jobs/setup'

How reconstruction works

Each Job registers itself with the JobRegistry when constructed. When a worker pops a payload, the registry rebuilds the instance (bypassing the constructor) and assigns the serialized data back onto it. For dedicated worker processes, make sure your job modules are imported — or call JobRegistry.register(MyJob) — so the names are known before processing.

Override serialize() on a job for custom payloads.