@warlock.js/logger

Structured, multi-channel logging for Node.js — five severity levels, a clean three-argument API, and full TypeScript support. Non-blocking by default; synchronous flush on demand.

Install

npm install @warlock.js/logger
# or
yarn add @warlock.js/logger

30-second tour

import { log, ConsoleLog, FileLog } from "@warlock.js/logger";

log.setChannels([
  new ConsoleLog(),
  new FileLog({ chunk: "daily" }),
]);

await log.info("users", "register", "New user created");
await log.error("payments", "charge", new Error("Card declined"));

The logger starts with no channels — nothing is printed or written until you register at least one.

The module · action · message pattern

Every call carries three pieces of context. Modules and actions become searchable keys in file and JSON channels:

await log.info("auth", "login", "User signed in");
await log.warn("api", "rateLimitApproaching", "80% of quota used");
await log.success("payments", "captured", "Payment of $49.99 captured");

Pass an object instead of positional args when you need context metadata:

await log.error({
  module: "orders",
  action: "checkout",
  message: "Card declined",
  context: { orderId: "ord_9f2a", amount: 4999 },
});

Levels

debug · info · warn · error · success — each has a shorthand method on the log singleton (and on every Logger instance).

Built-in channels

All three share the BasicLogConfigurations options (levels, filter, dateFormat); FileLog / JSONFileLog add storage, chunking, rotation, and grouping options on top.

log and Logger

The package exports one default singleton and one class:

• log — a pre-instantiated Logger. Day-to-day logging and configuration both go through it: log.info(...), log.configure(...), log.flushSync(), log.addChannel(...).
• Logger — the class. Use it when you need an isolated logger (libraries, sandboxes, parallel test suites).

log is a Logger instance, not a function — every level shortcut and configuration method is reachable on it. The bare-callable log(data) form was removed; use log.log(data) for the data-object form, or log.info(...) / log.error(...) / etc. for the positional form.

Graceful shutdown

FileLog and JSONFileLog buffer entries. Tell the logger to drain them on exit:

log.configure({
  channels: [new ConsoleLog(), new FileLog({ chunk: "daily" })],
  autoFlushOn: ["SIGINT", "SIGTERM", "beforeExit"],
});

Signal events flush then re-raise (so Node exits with its normal signal semantics); beforeExit flushes in place. For full control, call log.flushSync() manually inside your own handler instead.

Capturing unhandled errors

import { captureAnyUnhandledRejection } from "@warlock.js/logger";

captureAnyUnhandledRejection();

Registers process-level handlers for unhandledRejection and uncaughtException, forwarding both to log.error("app", ...). Call once at startup, after channels are registered.

Custom channels

Extend LogChannel and implement log(data):

import { LogChannel, type LoggingData } from "@warlock.js/logger";

export class SlackLog extends LogChannel<{ webhookUrl: string }> {
  public name = "slack";

  public async log(data: LoggingData) {
    if (!this.shouldBeLogged(data)) return;
    await fetch(this.config("webhookUrl"), {
      method: "POST",
      headers: { "Content-Type": "application/json" },
      body: JSON.stringify({
        text: `[${data.type}] [${data.module}][${data.action}]: ${data.message}`,
      }),
    });
  }
}

Set terminal = true on the class if your channel writes to a TTY — otherwise ANSI codes are auto-stripped from string messages.

Full documentation

The complete guide lives in the project docs:

• Getting Started
• Configuration
• Channels (ConsoleLog, FileLog, JSONFileLog)
• Lifecycle & Flushing
• Capturing Unhandled Errors
• Custom Channels
• Recipes
• API Reference
• Types

Tests

This package uses Vitest. From the repo root:

npx vitest run --root @warlock.js/logger

License

MIT