yt_dlp/extractor/vevo.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..networking.exceptions import HTTPError
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     parse_iso8601,
  10     parse_qs,
  11 )
  12
  13
  14 class VevoBaseIE(InfoExtractor):
  15     def _extract_json(self, webpage, video_id):
  16         return self._parse_json(
  17             self._search_regex(
  18                 r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>',
  19                 webpage, 'initial store'),
  20             video_id)
  21
  22
  23 class VevoIE(VevoBaseIE):
  24     """
  25     Accepts urls from vevo.com or in the format 'vevo:{id}'
  26     (currently used by MTVIE and MySpaceIE)
  27     """
  28     _VALID_URL = r'''(?x)
  29         (?:https?://(?:www\.)?vevo\.com/watch/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?|
  30            https?://cache\.vevo\.com/m/html/embed\.html\?video=|
  31            https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
  32            https?://embed\.vevo\.com/.*?[?&]isrc=|
  33            https?://tv\.vevo\.com/watch/artist/(?:[^/]+)/|
  34            vevo:)
  35         (?P<id>[^&?#]+)'''
  36     _EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1']
  37
  38     _TESTS = [{
  39         'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
  40         'md5': '95ee28ee45e70130e3ab02b0f579ae23',
  41         'info_dict': {
  42             'id': 'GB1101300280',
  43             'ext': 'mp4',
  44             'title': 'Hurts - Somebody to Die For',
  45             'timestamp': 1372057200,
  46             'upload_date': '20130624',
  47             'uploader': 'Hurts',
  48             'track': 'Somebody to Die For',
  49             'artist': 'Hurts',
  50             'genre': 'Pop',
  51         },
  52         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
  53     }, {
  54         'note': 'v3 SMIL format',
  55         'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
  56         'md5': 'f6ab09b034f8c22969020b042e5ac7fc',
  57         'info_dict': {
  58             'id': 'USUV71302923',
  59             'ext': 'mp4',
  60             'title': 'Cassadee Pope - I Wish I Could Break Your Heart',
  61             'timestamp': 1392796919,
  62             'upload_date': '20140219',
  63             'uploader': 'Cassadee Pope',
  64             'track': 'I Wish I Could Break Your Heart',
  65             'artist': 'Cassadee Pope',
  66             'genre': 'Country',
  67         },
  68         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
  69     }, {
  70         'note': 'Age-limited video',
  71         'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
  72         'info_dict': {
  73             'id': 'USRV81300282',
  74             'ext': 'mp4',
  75             'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
  76             'age_limit': 18,
  77             'timestamp': 1372888800,
  78             'upload_date': '20130703',
  79             'uploader': 'Justin Timberlake',
  80             'track': 'Tunnel Vision (Explicit)',
  81             'artist': 'Justin Timberlake',
  82             'genre': 'Pop',
  83         },
  84         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
  85     }, {
  86         'note': 'No video_info',
  87         'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000',
  88         'md5': '8b83cc492d72fc9cf74a02acee7dc1b0',
  89         'info_dict': {
  90             'id': 'USUV71503000',
  91             'ext': 'mp4',
  92             'title': 'K Camp ft. T.I. - Till I Die',
  93             'age_limit': 18,
  94             'timestamp': 1449468000,
  95             'upload_date': '20151207',
  96             'uploader': 'K Camp',
  97             'track': 'Till I Die',
  98             'artist': 'K Camp',
  99             'genre': 'Hip-Hop',
 100         },
 101         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
 102     }, {
 103         'note': 'Featured test',
 104         'url': 'https://www.vevo.com/watch/lemaitre/Wait/USUV71402190',
 105         'md5': 'd28675e5e8805035d949dc5cf161071d',
 106         'info_dict': {
 107             'id': 'USUV71402190',
 108             'ext': 'mp4',
 109             'title': 'Lemaitre ft. LoLo - Wait',
 110             'age_limit': 0,
 111             'timestamp': 1413432000,
 112             'upload_date': '20141016',
 113             'uploader': 'Lemaitre',
 114             'track': 'Wait',
 115             'artist': 'Lemaitre',
 116             'genre': 'Electronic',
 117         },
 118         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
 119     }, {
 120         'note': 'Only available via webpage',
 121         'url': 'http://www.vevo.com/watch/GBUV71600656',
 122         'md5': '67e79210613865b66a47c33baa5e37fe',
 123         'info_dict': {
 124             'id': 'GBUV71600656',
 125             'ext': 'mp4',
 126             'title': 'ABC - Viva Love',
 127             'age_limit': 0,
 128             'timestamp': 1461830400,
 129             'upload_date': '20160428',
 130             'uploader': 'ABC',
 131             'track': 'Viva Love',
 132             'artist': 'ABC',
 133             'genre': 'Pop',
 134         },
 135         'expected_warnings': ['Failed to download video versions info'],
 136     }, {
 137         # no genres available
 138         'url': 'http://www.vevo.com/watch/INS171400764',
 139         'only_matching': True,
 140     }, {
 141         # Another case available only via the webpage; using streams/streamsV3 formats
 142         # Geo-restricted to Netherlands/Germany
 143         'url': 'http://www.vevo.com/watch/boostee/pop-corn-clip-officiel/FR1A91600909',
 144         'only_matching': True,
 145     }, {
 146         'url': 'https://embed.vevo.com/?isrc=USH5V1923499&partnerId=4d61b777-8023-4191-9ede-497ed6c24647&partnerAdCode=',
 147         'only_matching': True,
 148     }, {
 149         'url': 'https://tv.vevo.com/watch/artist/janet-jackson/US0450100550',
 150         'only_matching': True,
 151     }]
 152     _VERSIONS = {
 153         0: 'youtube',  # only in AuthenticateVideo videoVersions
 154         1: 'level3',
 155         2: 'akamai',
 156         3: 'level3',
 157         4: 'amazon',
 158     }
 159
 160     def _initialize_api(self, video_id):
 161         webpage = self._download_webpage(
 162             'https://accounts.vevo.com/token', None,
 163             note='Retrieving oauth token',
 164             errnote='Unable to retrieve oauth token',
 165             data=json.dumps({
 166                 'client_id': 'SPupX1tvqFEopQ1YS6SS',
 167                 'grant_type': 'urn:vevo:params:oauth:grant-type:anonymous',
 168             }).encode(),
 169             headers={
 170                 'Content-Type': 'application/json',
 171             })
 172
 173         if re.search(r'(?i)THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION', webpage):
 174             self.raise_geo_restricted(
 175                 f'{self.IE_NAME} said: This page is currently unavailable in your region')
 176
 177         auth_info = self._parse_json(webpage, video_id)
 178         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['legacy_token']
 179
 180     def _call_api(self, path, *args, **kwargs):
 181         try:
 182             data = self._download_json(self._api_url_template % path, *args, **kwargs)
 183         except ExtractorError as e:
 184             if isinstance(e.cause, HTTPError):
 185                 errors = self._parse_json(e.cause.response.read().decode(), None)['errors']
 186                 error_message = ', '.join([error['message'] for error in errors])
 187                 raise ExtractorError(f'{self.IE_NAME} said: {error_message}', expected=True)
 188             raise
 189         return data
 190
 191     def _real_extract(self, url):
 192         video_id = self._match_id(url)
 193
 194         self._initialize_api(video_id)
 195
 196         video_info = self._call_api(
 197             f'video/{video_id}', video_id, 'Downloading api video info',
 198             'Failed to download video info')
 199
 200         video_versions = self._call_api(
 201             f'video/{video_id}/streams', video_id,
 202             'Downloading video versions info',
 203             'Failed to download video versions info',
 204             fatal=False)
 205
 206         # Some videos are only available via webpage (e.g.
 207         # https://github.com/ytdl-org/youtube-dl/issues/9366)
 208         if not video_versions:
 209             webpage = self._download_webpage(url, video_id)
 210             json_data = self._extract_json(webpage, video_id)
 211             if 'streams' in json_data.get('default', {}):
 212                 video_versions = json_data['default']['streams'][video_id][0]
 213             else:
 214                 video_versions = [
 215                     value
 216                     for key, value in json_data['apollo']['data'].items()
 217                     if key.startswith(f'{video_id}.streams')]
 218
 219         uploader = None
 220         artist = None
 221         featured_artist = None
 222         artists = video_info.get('artists')
 223         for curr_artist in artists:
 224             if curr_artist.get('role') == 'Featured':
 225                 featured_artist = curr_artist['name']
 226             else:
 227                 artist = uploader = curr_artist['name']
 228
 229         formats = []
 230         for video_version in video_versions:
 231             version = self._VERSIONS.get(video_version.get('version'), 'generic')
 232             version_url = video_version.get('url')
 233             if not version_url:
 234                 continue
 235
 236             if '.ism' in version_url:
 237                 continue
 238             elif '.mpd' in version_url:
 239                 formats.extend(self._extract_mpd_formats(
 240                     version_url, video_id, mpd_id=f'dash-{version}',
 241                     note=f'Downloading {version} MPD information',
 242                     errnote=f'Failed to download {version} MPD information',
 243                     fatal=False))
 244             elif '.m3u8' in version_url:
 245                 formats.extend(self._extract_m3u8_formats(
 246                     version_url, video_id, 'mp4', 'm3u8_native',
 247                     m3u8_id=f'hls-{version}',
 248                     note=f'Downloading {version} m3u8 information',
 249                     errnote=f'Failed to download {version} m3u8 information',
 250                     fatal=False))
 251             else:
 252                 m = re.search(r'''(?xi)
 253                     _(?P<quality>[a-z0-9]+)
 254                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 255                     _(?P<vcodec>[a-z0-9]+)
 256                     _(?P<vbr>[0-9]+)
 257                     _(?P<acodec>[a-z0-9]+)
 258                     _(?P<abr>[0-9]+)
 259                     \.(?P<ext>[a-z0-9]+)''', version_url)
 260                 if not m:
 261                     continue
 262
 263                 formats.append({
 264                     'url': version_url,
 265                     'format_id': f'http-{version}-{video_version.get("quality") or m.group("quality")}',
 266                     'vcodec': m.group('vcodec'),
 267                     'acodec': m.group('acodec'),
 268                     'vbr': int(m.group('vbr')),
 269                     'abr': int(m.group('abr')),
 270                     'ext': m.group('ext'),
 271                     'width': int(m.group('width')),
 272                     'height': int(m.group('height')),
 273                 })
 274
 275         track = video_info['title']
 276         if featured_artist:
 277             artist = f'{artist} ft. {featured_artist}'
 278         title = f'{artist} - {track}' if artist else track
 279
 280         genres = video_info.get('genres')
 281         genre = (
 282             genres[0] if genres and isinstance(genres, list)
 283             and isinstance(genres[0], str) else None)
 284
 285         is_explicit = video_info.get('isExplicit')
 286         if is_explicit is True:
 287             age_limit = 18
 288         elif is_explicit is False:
 289             age_limit = 0
 290         else:
 291             age_limit = None
 292
 293         return {
 294             'id': video_id,
 295             'title': title,
 296             'formats': formats,
 297             'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
 298             'timestamp': parse_iso8601(video_info.get('releaseDate')),
 299             'uploader': uploader,
 300             'duration': int_or_none(video_info.get('duration')),
 301             'view_count': int_or_none(video_info.get('views', {}).get('total')),
 302             'age_limit': age_limit,
 303             'track': track,
 304             'artist': uploader,
 305             'genre': genre,
 306         }
 307
 308
 309 class VevoPlaylistIE(VevoBaseIE):
 310     _VALID_URL = r'https?://(?:www\.)?vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
 311
 312     _TESTS = [{
 313         'url': 'http://www.vevo.com/watch/genre/rock',
 314         'info_dict': {
 315             'id': 'rock',
 316             'title': 'Rock',
 317         },
 318         'playlist_count': 20,
 319     }, {
 320         'url': 'http://www.vevo.com/watch/genre/rock?index=0',
 321         'only_matching': True,
 322     }]
 323
 324     def _real_extract(self, url):
 325         mobj = self._match_valid_url(url)
 326         playlist_id = mobj.group('id')
 327         playlist_kind = mobj.group('kind')
 328
 329         webpage = self._download_webpage(url, playlist_id)
 330
 331         qs = parse_qs(url)
 332         index = qs.get('index', [None])[0]
 333
 334         if index:
 335             video_id = self._search_regex(
 336                 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
 337                 webpage, 'video id', default=None, group='id')
 338             if video_id:
 339                 return self.url_result(f'vevo:{video_id}', VevoIE.ie_key())
 340
 341         playlists = self._extract_json(webpage, playlist_id)['default'][f'{playlist_kind}s']
 342
 343         playlist = (next(iter(playlists.values()))
 344                     if playlist_kind == 'playlist' else playlists[playlist_id])
 345
 346         entries = [
 347             self.url_result(f'vevo:{src}', VevoIE.ie_key())
 348             for src in playlist['isrcs']]
 349
 350         return self.playlist_result(
 351             entries, playlist.get('playlistId') or playlist_id,
 352             playlist.get('name'), playlist.get('description'))