limitie

Installing

$ npm install limitie

Requests per Interval

Lets say you have a site that only lets you make 10 requests per second across all of its API endpoints. You could create a limiter for that service and use it in all of the methods for that service.

import createLimitie from 'limitie';

// Create a rate limiter for this service, limiting all requests through this limiter to
// 10 per second.
// Any requests in excess of that amount will be queued up and updated when available.
const limitie = createLimitie({ requests: 10, interval: 1000 });

async function listItems() {
  await limitie.request();
  // List items from this service
}

async function getDetailedItem(itemId: string) {
  await limitie.request();
  // Get detailed info for this item
}

Tokens per Interval

Tokens are a lower level abstraction than requests. This could be used for services that have a rate limit in place of something like 50000 tokens per minute. In that case, you would build a rate limiter like this:

import createLimitie from 'limitie';

// Create a rate limiter for this service, limiting all requests through this limiter to
// 50000 tokens per minute.
const limitie = createLimitie({
  tokens: {
    regen: 50000,
  },
  interval: 1000,
});

async function listItems() {
  // Here we use the lower level reserve API, where we can specify a specific number of tokens for this request.
  await limitie.reserve(10000).promise();
  // List items from this service
}

async function getDetailedItem(itemId: string) {
  // Here we use the lower level reserve API, where we can specify a specific number of tokens for this request.
  await limitie.reserve(500).promise();
  // Get detailed info for this item
}

Full API

Detailed token config

// Create a rate limiter for this service, limiting all requests through this limiter to
// 50000 tokens per minute.
// This service starts us off at 0 tokens, though, so we have to wait for tokens to regen
// before we can make requests, so we pass tokens.initial of 0.
// This service also allows us to pool up tokens to a maximum of 99999, so we can pass
// tokens.max of 99999
const limitie = createLimitie({
  tokens: {
    max: 99999,
    regen: 50000,
    initial: 0,
  },
  interval: 1000,
});

Getting time estimates

Let's say I've reserved to use 50 tokens, but I have to wait until they're available. This means that the limiter has handled all of the reserved cases before mine, and that sufficient tokens have pooled to accommodate the request.

const reservation = limitie.reserve(50);
// I can see the time left until this is handled by doing
const timeLeft = limitie.getTimeUntilReady(reservation.id);

// And it is awaited like this
await reservation.promise;

Contributing

We welcome contributions! Please see our contributing guidelines for more information.

License

MIT