Skip to content

Latest commit

 

History

History
196 lines (140 loc) · 6.83 KB

README.md

File metadata and controls

196 lines (140 loc) · 6.83 KB
Raw

YoutubeExplodeDart

This is a port of the YoutubeExplode library from C#, most of the functions, doc comments, readme information, is taken from YoutubeExplode repository.

Pub Version GitHub Workflow Status License Effective Dart

YoutubeExplode is a library that provides an interface to query metadata of YouTube videos, playlists and channels, as well as to resolve and download video streams and closed caption tracks. Behind a layer of abstraction, the library parses raw page content and uses reverse-engineered AJAX requests to retrieve information. As it doesn't use the official API, there's also no need for an API key and there are no usage quotas.

Features from YoutubeExplode

  • Retrieve metadata on videos, playlists, channels, streams, and closed captions
  • Execute search queries and get resulting videos.
  • Get or download video streams.
  • Get closed captions.
  • Get video comments.
  • All model extend Equatable to easily perform equality checks

Differences from YoutubeExplode

  • The entry point is YoutubeExplode, not YoutubeClient.
  • Download closed captions as srt is not supported yet.
  • Search queries can be fetched from the search page as well (thus fetch Videos, Channels and Playlists).

Usage

Install

Add the dependency to the pubspec.yaml (Check for the latest version)

youtube_explode_dart: ^1.3.0

Import the library

import 'package:youtube_explode_dart/youtube_explode_dart.dart';

Getting metadata of a video

The following example shows how you can extract various metadata from a YouTube video:

// You can provider either a video ID or URL as String or an instance of `VideoId`.
var video = yt.videos.get('https://youtube.com/watch?v=Dpp1sIL1m5Q'); // Returns a Video instance.

var title = video.title; // "Scamazon Prime"
var author = video.author; // "Jim Browning"
var duration = video.duration; // Instance of Duration - 0:19:48.00000

Downloading a video stream

Every YouTube video has a number of streams available. These streams may have different containers, video quality, bitrate, etc.

On top of that, depending on the content of the stream, the streams are further divided into 3 categories:

  • Muxed streams -- contain both video and audio
  • Audio-only streams -- contain only audio -- Video-only streams -- contain only video

You can request the stream manifest to get available streams for a particular video:

var yt = YoutubeExplode();

var manifest = yt.videos.streamsClient.getManifest('Dpp1sIL1m5Q');

Once you get the manifest, you can filter through the streams and choose the one you're interested in downloading:

// Get highest quality muxed stream
var streamInfo = streamManifest.muxed.withHigestVideoQuality();

// ...or highest bitrate audio-only stream
var streamInfo = streamManifest.audioOnly.withHigestBitrate()

// ...or highest quality MP4 video-only stream
var streamInfo.videoOnly.where((e) => e.container == Container)

Finally, you can get the actual Stream object represented by the metadata:

if (streamInfo != null) {
  // Get the actual stream
  var stream = yt.video.streamClient.get(streamInfo);
  
  // Open a file for writing.
  var file = File(filePath);
  var fileStream = file.openWrite();

  // Pipe all the content of the stream into the file.
  await stream.pipe(fileStream);

  // Close the file.
  await fileStream.flush();
  await fileStream.close();
}

While it may be tempting to just always use muxed streams, it's important to note that they are limited in quality. Muxed streams don't go beyond 720p30.

If you want to download the video in maximum quality, you need to download the audio-only and video-only streams separately and then mux them together on your own. There are tools like FFmpeg that let you do that.

Working with playlists

Among other things, YoutubeExplode also supports playlists:

var yt = YoutubeExplode();

// Get playlist metadata.
var playlist = await yt.playlists.get('xxxxx');

var title = playlist.title;
var author = playlist.author;

  await for (var video in yt.playlists.getVideos(playlist.id)) {
    var videoTitle = video.title;
    var videoAuthor = video.author;
  }

var playlistVideos = await yt.playlists.getVideos(playlist.id);

// Get first 20 playlist videos.
var somePlaylistVideos = await yt.playlists.getVideos(playlist.id).take(20);

Extracting closed captions

Similarly, to streams, you can extract closed captions by getting the manifest and choosing the track you're interested in:

  var yt = YoutubeExplode();

  var trackManifest = await yt.videos.closedCaptions.getManifest('_QdPW8JrYzQ')

  var trackInfo = manifest.getByLanguage('en'); // Get english caption.
  
  if (trackInfo != null)
  {
     // Get the actual closed caption track.
     var track = await youtube.videos.closedCaptions.get(trackInfo);
      
    // Get the caption displayed at 1:01
    var caption = track.getByTime(Duration(seconds: 61));
    var text = caption?.text; // "And the game was afoot."
  }

Getting comments

You can easily get the video comments of a given video, the comment pages are fetched automatically, and every page has 20 videos (that means that there is no point to take 25 comments to save some resources, but you could just take 40)

// Here we are getting the first 20 comments of the video.
// The stream wont close until all the video comments are fetched.
var comments = await yt.videos.commentsClient.getComments(video).take(20).toList()

Currently getting comment replies is not supported.

Cleanup

You need to close YoutubeExplode's http client, when done otherwise this could halt the dart process.

yt.close();

Examples:

More examples available on GitHub.

Check the api documentation for additional information.

Credits