Resource System

Inject external dependencies into tasks and intercept jobs before they enqueue.

Task arguments cross a process boundary, so they must be serializable — but most real dependencies (database pools, HTTP clients, cloud SDKs) are not. The Node SDK keeps those on the worker and passes only plain data through the queue.

import { Queue, useResource } from "@byteveda/flexiq";
import { Pool } from "pg";

const queue = new Queue({ dbPath: "tasks.db" });

queue.resource("db", () => new Pool(), { dispose: (p) => p.end() });

queue.task("sync", async (id: string) => {
  const db = await useResource<Pool>("db");
  await db.query("UPDATE users SET synced = now() WHERE id = $1", [id]);
});

Dependency injectionqueue.resource(), worker/task/pooled scopes, useResource, declarative inject, teardown ConfigurationEvery registration option — scopes, pool tuning, health checks, hot reload Enqueue interceptionThe onEnqueue hook — validate, redact, and rewrite jobs before serialization ProxiesSend a non-serializable resource as a signed reference and rebuild it on the worker ObservabilityLifecycle metrics, cluster-wide health, readiness probes, proxy stats TestingmockResource, teardown assertions, deterministic pool tests

Coming from Python? Node's resource system is intentionally lighter — enqueue interception is a single onEnqueue hook rather than a set of strategies, and resources are registered in code with no TOML loader. Scopes, pooling, health checks, hot reload, and signed proxies all have direct equivalents.