yt_dlp/extractor/vevo.py

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