@ekwoka/spotify-api

⚡️A tree-shakable, composable, lightweight wrapper for the multiple Spotify APIs🔥

Born from my own difficulties using other wrapper libraries for Spotify, this library seeks to be the best possible API wrapper.

NOTE: This library is still very much a work in progress and will be subject to breaking changes and regular updates until this note is removed.

Why is this good?

• 🏷 Types out of the box!
• 🌴 Treeshakeable and composable!
• ⚡️ Improved performance from the intelligent use of caches and batching!
• 🏗 Consistent Behavior through normalized return types!
• 🦕 Supports Deno and Bun out of the box!

Installation

Simply install with your favourite flavour of Package Manager

npm add @ekwoka/spotify-api
pnpm add @ekwoka/spotify-api
bun install @ekwoka/spotify-api
yarn add @ekwoka/spotify-api

Usage

To get started, you'll need to import the core client from the package, and initialize it.

import { SpotifyApi } from '@ekwoka/spotify-api';

export const client = SpotifyApi('tokenhere'); // client requires an initial token to initialize. Initialize client after recieving token from Spotify.

This will create the core client structure with which you'll manage and run requests to the Spotify Apis.

To update the token during ongoing usage simply, import and use the setToken composable with your client.

import { SpotifyApi, setToken } from '@ekwoka/spotify-api';

const client = SpotifyApi('initial_token'); // original token

// after some event, update the token
client(setToken('my_new_token')); // updated token

As you'll notice, this is not a method on the client object like many other libraries. This is a composable function. The goal is for all interactions with the client and APIs to be composable functions. This will enable very aggressive tree-shaking to keep minimal clients from shipping lots of unused code, as well as enable code-splitting for larger applications. This should be reflected in a much more modest bundle size for the majority of use cases.

Authentication

Includes in this package are some additional helper functions for interacting with Spotify's authentication API. These should only be used on a server, as they require client secrets.

These helpers are:

• makeAuthURL: Generates an Auth URL to direct the user to for logging in with Spotify OAuth
• getTokenFromCode: Accepts a code from the Spotify authentication flow and returns a suite of tokens (access and refresh).
• refreshToken: Accepts a refresh token and returns a new access token.

These currently depend on you setting up and exposing certain environment variables for the functions to access on the process.env object:

• SPOTIFY_CLIENT: Client id from Spotify Developer Dashboard.
• SPOTIFY_SECRET: Client secret.
• REDIRECT: URL to pass to Spotify Auth for handling the user returning to your app.

If these are not defined, the function will throw.

Examples (Pseudo Express code)

import {
  getTokenFromCode,
  refreshToken,
  makeAuthURL,
} from '@ekwoka/spotify-api';

const loginHandler = async (req, res) => {
  const url = makeAuthURL(['user-read-email']);
  res.redirect(302, url);
};

const codeHandler = async (req, res) => {
  try {
    const { code } = JSON.parse(req.body);
    const { access_token, refresh_token } = await getTokenFromCode(code);
    res.cookie('refresh_token', refresh_token);
    res.status(200).json({ access_token });
  } catch (err) {
    res.status(500).json({ error: err.message });
  }
};

const refreshHandler = async (req, res) => {
  try {
    const { refresh_token } = req.cookies;
    const { access_token } = await refreshToken(refresh_token);
    res.status(200).json({ access_token });
  } catch (err) {
    res.status(500).json({ error: err.message });
  }
};

Endpoints

Endpoints are importable both from @ekwoka/spotify-api or @ekwoka/spotify-api/endpoints for convenience.

NOTE: The following documentation uses the same structure as the Official Spotify Docs for organization

Albums

Currently Available methods in the Albums category include:

• getAlbum - Retrieves info about an album by ID
• getAlbums - Retrieves info about multiple albums by ID
• getAlbumTracks - Retrieves info about an albums tracks

While these correspond to 3 different endpoints to Spotify's API, internally these 3 use only the getAlbums endpoints for improved code-reuse.

Cachekey: 'albums.[id]' Batching Limit: 20

getAlbum

Gets details of an Album by ID.

const album = client(getAlbum('6tLZvqqoWszgPagzzNNQQF'));
const albumInMarket = client(getAlbum('6tLZvqqoWszgPagzzNNQQF', 'KR'));

getAlbums

Gets Details about multiple Albums at once.

const albums = client(
  getAlbums(['6tLZvqqoWszgPagzzNNQQF', '6XBIkDFhDgc3PQOUEcO2fd'])
);
const albumsInMarket = client(
  getAlbums(['6tLZvqqoWszgPagzzNNQQF', '6XBIkDFhDgc3PQOUEcO2fd'], 'KR')
);

getAlbumTracks

Gets Track Info about Album

const album = client(getAlbumTracks('6tLZvqqoWszgPagzzNNQQF'));
const albumInMarket = client(getAlbumTracks('6tLZvqqoWszgPagzzNNQQF', 'KR'));

Users

Currently Available methods in the Users category include:

• getCurrentUser - Retrieves data related to the actively logged in user (as defined by access token)
• getTopItems - Retrieves info covering the Users Top Tracks and Artists
• getUserProfile - Retrieves public info regarding another User by ID

getCurrentUser

Gets details of the currently logged in user (identified by token)

const user = client(getCurrentUser());

console.log(user); // should log user

cache key: 'user'

getTopItems

Gets the users top tracks or artists

const topTracks = await client(
  getTopItems('tracks', { limit: 50, time_range: 'long_term' })
);
const topArtists = await client(
  getTopItems('artists', { offset: 20, limit: 10 })
);

Options:

• limit: The number of items to return. Default: 20. Maximum: 50.
• offset: The index of the first item to return. Default: 0.
• time_range: Over what time frame the data is retrieved. Options: short_term, medium_term, long_term. Default: medium_term.

getUserProfile

Get's another user's profile by ID

const thekwoka = await client(getUserProfile('thekwoka'));

Batching and Caching

One of the unique features of this API wrapper is the use of batched requests and intelligent caching. These features serve the purpose of reducing the number of requests make to the Spotify API and generally improving the responsiveness of your application.

All of these takes place under the hood, and the implementation of your code should be relatively agnostic to the fact this is happening.

Batching

Many endpoints that exist with the Spotify API accept the request for multiple items at a single time. For example, even though a singular album endpoint exists, there is also an albums endpoint that returns the same information, but simply using an array of album IDs instead of a singular one.

So, the basic idea is that, if the application is making calls for many individual items ( like album), and those items can be batched (into albums), this wrapper will automatically do so and distribute the results.

Example:

const ids = ['6tLZvqqoWszgPagzzNNQQF', '6XBIkDFhDgc3PQOUEcO2fd'];
ids.forEach(async (id) => {
  const album = await client(getAlbum(id));
  console.log(album.name);
});

While this isa bit contrived, in a component based framework, you might have these actual calls happening in places far away from each other.

In the background, these two ids will be bunched together (if they come in close enough to eachother) sent as a single request to Spotify, and resolved from the return. This also includes combining multiple requests for a single album into only the one reference requested from Spotify.

Splitting Large Lists

This system also comes with another nice feature. These multiple item endpoints have limits on the total number of items in a single request. For the above example of albums it's 20.

Once again, in the background, as these requests come in, the total number of albums may be greater than 20. Or you may even do a direct request for more than 20 albums.

In the background, the request will be bunched into groups of 20, sent as requests, and, once again, distributed back to whence they came.

Caching

As this data changes infrequently, responses will be cached and reused when the same information is requested again. This works with the above batching, as well. So if you make a bulk request for 10 albums, 3 of which you've already searched for before, those 3 will be returned from cache and the other 7 will be fetched anew, all without any adjustments to how your code behaves.

Special Utilities

Cache Busting

As noted, a major benefit of this API wrapper is the intelligent use of caches. However, caches may not always be accurate, or may introduce other issues in certain contexts. As such, there is a special utility for cache busting.

import { resetCache } from '@ekwoka/spotify-api';

// clears entire cache
client(resetCache());

// clears specific cached value
client(resetCache('user')); // should clear user cache only

Where caches are utilized, the documentation for those endpoints will include information about the cache key(s) used.