discord-request

A Discord HTTP client that handles global and resource rate limits automatically.

Looking for an API Wrapper? Check out @discord-interactions/api instead. This package is low level, and works best in libraries as opposed to applications.

Usage

This package works best in conjuction with discord-api-types. To begin, install both:

npm install discord-api discord-api-types

Then you can combine the two (with or without typing the results):

import Client from "discord-request";
import {
    Routes,
    RESTGetAPIApplicationGuildCommandsResult,
} from "discord-api-types/v10";

const instance = new Client();

const guildCommands = instance.get(
    Routes.applicationGuildCommands(applicationId, guildId)
) as Promise<RESTGetAPIApplicationGuildCommandsResult>;

Options

When creating a new client, you can configure it by passing in optional arguments:

const instance = new Client({
    retries: 0,
    timeout: 1000,
    globalRequestsPerSecond: 100,
    onRateLimit: (data) => console.log({ data }),
});

Every parameter listed below is optional.

• api: The API URL to use. Defaults to https://discord.com/api.
• version: The API version to use. Defaults to 10.
• cdn: The CDN URL to use. Defaults to https://cdn.discordapp.com.
• headers: An object of additional headers to send with each request.
• userAgent: The user agent to use. Defaults to Discord Request v0.
• retries: The number of times to retry a request if it fails. Defaults to 3.
• timeout: The number of milliseconds to wait before timing out a request. Defaults to 15000 (15 seconds).
• globalRequestsPerSecond: The number of global requests per second to allow. Defaults to 50.
• shutdownSignal: An AbortSignal to use when you need to cancel all unfinished requests to shut down the application.
• queueSweepInterval: The number of milliseconds to wait between sweeping the queue. Defaults to 0 (no sweeping). If you use discord-request in a persistent environment, you should set this value.
• bucketSweepInterval: The number of milliseconds to wait between sweeping the buckets. Defaults to 0 (no sweeping). If you use discord-request in a persistent environment, you should set this value.
• onBucketSweep: See callbacks below.
• onQueueSweep: See callbacks below.
• onRateLimit: See callbacks below.
• onRequest: See callbacks below.

Callbacks

onBucketSweep?: (swept: Map<string, Bucket>) => void;

Runs when a bucket sweep finishes. Returns a Map of removed buckets.

onQueueSweep?: (swept: Map<string, Queue>) => void;

Runs when a queue sweep finishes. Returns a Map of removed queues.

onRateLimit?: (data: RateLimitData) => void;

Runs when a rate limit is encountered. Returns information about the rate limit.

onRequest?: (parameters: Route, resource: string, init: RequestInit, options: RequestOptions, retries: number) => void;

Runs when a request is sent to the Discord API. Returns information used to send the request.

Credits

This code is derived from @IanMitchell's Interaction Kit: discord-request & discord-error and distributed under the Apache 2 license. It was changed to merge these packages together.