yt_dlp/extractor/genius.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     js_to_json,
   5     smuggle_url,
   6     str_or_none,
   7     traverse_obj,
   8     unescapeHTML,
   9 )
  10
  11
  12 class GeniusIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?genius\.com/videos/(?P<id>[^?/#]+)'
  14     _TESTS = [{
  15         'url': 'https://genius.com/videos/Vince-staples-breaks-down-the-meaning-of-when-sparks-fly',
  16         'md5': '64c2ad98cfafcfda23bfa0ad0c512f4c',
  17         'info_dict': {
  18             'id': '6313303597112',
  19             'ext': 'mp4',
  20             'title': 'Vince Staples Breaks Down The Meaning Of “When Sparks Fly”',
  21             'description': 'md5:bc15e00342c537c0039d414423ae5752',
  22             'tags': 'count:1',
  23             'uploader_id': '4863540648001',
  24             'duration': 388.416,
  25             'upload_date': '20221005',
  26             'timestamp': 1664982341,
  27             'thumbnail': r're:^https?://.*\.jpg$',
  28         },
  29     }, {
  30         'url': 'https://genius.com/videos/Breaking-down-drakes-certified-lover-boy-kanye-beef-way-2-sexy-cudi',
  31         'md5': 'b8ed87a5efd1473bd027c20a969d4060',
  32         'info_dict': {
  33             'id': '6271792014001',
  34             'ext': 'mp4',
  35             'title': 'md5:c6355f7fa8a70bc86492a3963919fc15',
  36             'description': 'md5:1774638c31548b31b037c09e9b821393',
  37             'tags': 'count:3',
  38             'uploader_id': '4863540648001',
  39             'duration': 2685.099,
  40             'upload_date': '20210909',
  41             'timestamp': 1631209167,
  42             'thumbnail': r're:^https?://.*\.jpg$',
  43         },
  44     }]
  45
  46     def _real_extract(self, url):
  47         display_id = self._match_id(url)
  48         webpage = self._download_webpage(url, display_id)
  49
  50         metadata = self._search_json(
  51             r'<meta content="', webpage, 'metadata', display_id, transform_source=unescapeHTML)
  52         video_id = traverse_obj(
  53             metadata, ('video', 'provider_id'),
  54             ('dfp_kv', lambda _, x: x['name'] == 'brightcove_video_id', 'values', 0), get_all=False)
  55         if not video_id:
  56             raise ExtractorError('Brightcove video id not found in webpage')
  57
  58         config = self._search_json(r'var\s*APP_CONFIG\s*=', webpage, 'config', video_id, default={})
  59         account_id = config.get('brightcove_account_id', '4863540648001')
  60         player_id = traverse_obj(
  61             config, 'brightcove_standard_web_player_id', 'brightcove_standard_no_autoplay_web_player_id',
  62             'brightcove_modal_web_player_id', 'brightcove_song_story_web_player_id', default='S1ZcmcOC1x')
  63
  64         return self.url_result(
  65             smuggle_url(
  66                 f'https://players.brightcove.net/{account_id}/{player_id}_default/index.html?videoId={video_id}',
  67                 {'referrer': url}), 'BrightcoveNew', video_id)
  68
  69
  70 class GeniusLyricsIE(InfoExtractor):
  71     _VALID_URL = r'https?://(?:www\.)?genius\.com/(?P<id>[^?/#]+)-lyrics[?/#]?'
  72     _TESTS = [{
  73         'url': 'https://genius.com/Lil-baby-heyy-lyrics',
  74         'playlist_mincount': 2,
  75         'info_dict': {
  76             'id': '8454545',
  77             'title': 'Heyy',
  78             'description': 'Heyy by Lil Baby',
  79         },
  80     }, {
  81         'url': 'https://genius.com/Outkast-two-dope-boyz-in-a-cadillac-lyrics',
  82         'playlist_mincount': 1,
  83         'info_dict': {
  84             'id': '36239',
  85             'title': 'Two Dope Boyz (In a Cadillac)',
  86             'description': 'Two Dope Boyz (In a Cadillac) by OutKast',
  87         },
  88     }, {
  89         'url': 'https://genius.com/Playboi-carti-rip-lyrics',
  90         'playlist_mincount': 1,
  91         'info_dict': {
  92             'id': '3710582',
  93             'title': 'R.I.P.',
  94             'description': 'R.I.P. by Playboi Carti',
  95         },
  96     }]
  97
  98     def _real_extract(self, url):
  99         display_id = self._match_id(url)
 100         webpage = self._download_webpage(url, display_id)
 101
 102         json_string = self._search_json(
 103             r'window\.__PRELOADED_STATE__\s*=\s*JSON\.parse\(', webpage, 'json string',
 104             display_id, transform_source=js_to_json, contains_pattern=r'\'{(?s:.+)}\'')
 105         song_info = self._parse_json(json_string, display_id)
 106         song_id = str_or_none(traverse_obj(song_info, ('songPage', 'song')))
 107         if not song_id:
 108             raise ExtractorError('Song id not found in webpage')
 109
 110         title = traverse_obj(
 111             song_info, ('songPage', 'trackingData', lambda _, x: x['key'] == 'Title', 'value'),
 112             get_all=False, default='untitled')
 113         artist = traverse_obj(
 114             song_info, ('songPage', 'trackingData', lambda _, x: x['key'] == 'Primary Artist', 'value'),
 115             get_all=False, default='unknown artist')
 116         media = traverse_obj(
 117             song_info, ('entities', 'songs', song_id, 'media'), expected_type=list, default=[])
 118
 119         entries = []
 120         for m in media:
 121             if m.get('type') in ('video', 'audio') and m.get('url'):
 122                 if m.get('provider') == 'spotify':
 123                     self.to_screen(f'{song_id}: Skipping Spotify audio embed')
 124                 else:
 125                     entries.append(self.url_result(m['url']))
 126
 127         return self.playlist_result(entries, song_id, title, f'{title} by {artist}')