Youtubei
is made to replace my other library scrape-yt. Instead of scrapping data from Youtube page, youtubei
fetches data by sending a request directly to https://www.youtube.com/youtubei/v1
, which should be faster and provide more reliable result.
Requires Node >= 16
npm i youtubei
or use the dev build directly from GitHub:
npm i git://github.com/suspiciouslookingowl/youtubei.git#dist
const { Client, MusicClient } = require("youtubei");
// or for TS / ES6
import { Client, MusicClient } from "youtubei";
const youtube = new Client();
const music = new MusicClient();
const run = async () => {
const videos = await youtube.search("Never gonna give you up", {
type: "video", // video | playlist | channel | all
});
console.log(videos.items.length); // 20
const nextVideos = await videos.next(); // load next page
console.log(nextVideos.length); // 18-20, inconsistent next videos count from youtube
console.log(videos.items.length); // 38 - 40
// you can also pass the video URL
const video = await youtube.getVideo("dQw4w9WgXcQ");
const channelVideos = await video.channel.videos.next();
const channelPlaylists = await video.channel.playlists.next();
// you can also pass the playlist URL
const playlist = await youtube.getPlaylist("UUHnyfMqiRRG1u-2MsSQLbXA");
console.log(playlist.videos.items.length); // first 100 videos;
let newVideos = await playlist.videos.next(); // load next 100 videos
console.log(playlist.videos.items.length); // 200 videos;
await playlist.videos.next(0); // load the rest videos in the playlist
// search using music client
const shelves = await music.search("Never gonna give you up");
console.log(shelves);
};
run();