Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vevo.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_str,
   8     compat_urlparse,
   9     compat_HTTPError,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     sanitized_Request,
  15     parse_iso8601,
  16 )
  17
  18
  19 class VevoBaseIE(InfoExtractor):
  20     def _extract_json(self, webpage, video_id):
  21         return self._parse_json(
  22             self._search_regex(
  23                 r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>',
  24                 webpage, 'initial store'),
  25             video_id)
  26
  27
  28 class VevoIE(VevoBaseIE):
  29     '''
  30     Accepts urls from vevo.com or in the format 'vevo:{id}'
  31     (currently used by MTVIE and MySpaceIE)
  32     '''
  33     _VALID_URL = r'''(?x)
  34         (?:https?://(?:www\.)?vevo\.com/watch/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?|
  35            https?://cache\.vevo\.com/m/html/embed\.html\?video=|
  36            https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
  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     _VERSIONS = {
 149         0: 'youtube',  # only in AuthenticateVideo videoVersions
 150         1: 'level3',
 151         2: 'akamai',
 152         3: 'level3',
 153         4: 'amazon',
 154     }
 155
 156     def _initialize_api(self, video_id):
 157         req = sanitized_Request(
 158             'http://www.vevo.com/auth', data=b'')
 159         webpage = self._download_webpage(
 160             req, None,
 161             note='Retrieving oauth token',
 162             errnote='Unable to retrieve oauth token')
 163
 164         if re.search(r'(?i)THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION', webpage):
 165             self.raise_geo_restricted(
 166                 '%s said: This page is currently unavailable in your region' % self.IE_NAME)
 167
 168         auth_info = self._parse_json(webpage, video_id)
 169         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token']
 170
 171     def _call_api(self, path, *args, **kwargs):
 172         try:
 173             data = self._download_json(self._api_url_template % path, *args, **kwargs)
 174         except ExtractorError as e:
 175             if isinstance(e.cause, compat_HTTPError):
 176                 errors = self._parse_json(e.cause.read().decode(), None)['errors']
 177                 error_message = ', '.join([error['message'] for error in errors])
 178                 raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 179             raise
 180         return data
 181
 182     def _real_extract(self, url):
 183         video_id = self._match_id(url)
 184
 185         self._initialize_api(video_id)
 186
 187         video_info = self._call_api(
 188             'video/%s' % video_id, video_id, 'Downloading api video info',
 189             'Failed to download video info')
 190
 191         video_versions = self._call_api(
 192             'video/%s/streams' % video_id, video_id,
 193             'Downloading video versions info',
 194             'Failed to download video versions info',
 195             fatal=False)
 196
 197         # Some videos are only available via webpage (e.g.
 198         # https://github.com/rg3/youtube-dl/issues/9366)
 199         if not video_versions:
 200             webpage = self._download_webpage(url, video_id)
 201             json_data = self._extract_json(webpage, video_id)
 202             if 'streams' in json_data.get('default', {}):
 203                 video_versions = json_data['default']['streams'][video_id][0]
 204             else:
 205                 video_versions = [
 206                     value
 207                     for key, value in json_data['apollo']['data'].items()
 208                     if key.startswith('%s.streams' % video_id)]
 209
 210         uploader = None
 211         artist = None
 212         featured_artist = None
 213         artists = video_info.get('artists')
 214         for curr_artist in artists:
 215             if curr_artist.get('role') == 'Featured':
 216                 featured_artist = curr_artist['name']
 217             else:
 218                 artist = uploader = curr_artist['name']
 219
 220         formats = []
 221         for video_version in video_versions:
 222             version = self._VERSIONS.get(video_version.get('version'), 'generic')
 223             version_url = video_version.get('url')
 224             if not version_url:
 225                 continue
 226
 227             if '.ism' in version_url:
 228                 continue
 229             elif '.mpd' in version_url:
 230                 formats.extend(self._extract_mpd_formats(
 231                     version_url, video_id, mpd_id='dash-%s' % version,
 232                     note='Downloading %s MPD information' % version,
 233                     errnote='Failed to download %s MPD information' % version,
 234                     fatal=False))
 235             elif '.m3u8' in version_url:
 236                 formats.extend(self._extract_m3u8_formats(
 237                     version_url, video_id, 'mp4', 'm3u8_native',
 238                     m3u8_id='hls-%s' % version,
 239                     note='Downloading %s m3u8 information' % version,
 240                     errnote='Failed to download %s m3u8 information' % version,
 241                     fatal=False))
 242             else:
 243                 m = re.search(r'''(?xi)
 244                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 245                     _(?P<vcodec>[a-z0-9]+)
 246                     _(?P<vbr>[0-9]+)
 247                     _(?P<acodec>[a-z0-9]+)
 248                     _(?P<abr>[0-9]+)
 249                     \.(?P<ext>[a-z0-9]+)''', version_url)
 250                 if not m:
 251                     continue
 252
 253                 formats.append({
 254                     'url': version_url,
 255                     'format_id': 'http-%s-%s' % (version, video_version['quality']),
 256                     'vcodec': m.group('vcodec'),
 257                     'acodec': m.group('acodec'),
 258                     'vbr': int(m.group('vbr')),
 259                     'abr': int(m.group('abr')),
 260                     'ext': m.group('ext'),
 261                     'width': int(m.group('width')),
 262                     'height': int(m.group('height')),
 263                 })
 264         self._sort_formats(formats)
 265
 266         track = video_info['title']
 267         if featured_artist:
 268             artist = '%s ft. %s' % (artist, featured_artist)
 269         title = '%s - %s' % (artist, track) if artist else track
 270
 271         genres = video_info.get('genres')
 272         genre = (
 273             genres[0] if genres and isinstance(genres, list) and
 274             isinstance(genres[0], compat_str) else None)
 275
 276         is_explicit = video_info.get('isExplicit')
 277         if is_explicit is True:
 278             age_limit = 18
 279         elif is_explicit is False:
 280             age_limit = 0
 281         else:
 282             age_limit = None
 283
 284         return {
 285             'id': video_id,
 286             'title': title,
 287             'formats': formats,
 288             'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
 289             'timestamp': parse_iso8601(video_info.get('releaseDate')),
 290             'uploader': uploader,
 291             'duration': int_or_none(video_info.get('duration')),
 292             'view_count': int_or_none(video_info.get('views', {}).get('total')),
 293             'age_limit': age_limit,
 294             'track': track,
 295             'artist': uploader,
 296             'genre': genre,
 297         }
 298
 299
 300 class VevoPlaylistIE(VevoBaseIE):
 301     _VALID_URL = r'https?://(?:www\.)?vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
 302
 303     _TESTS = [{
 304         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 305         'info_dict': {
 306             'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 307             'title': 'Best-Of: Birdman',
 308         },
 309         'playlist_count': 10,
 310     }, {
 311         'url': 'http://www.vevo.com/watch/genre/rock',
 312         'info_dict': {
 313             'id': 'rock',
 314             'title': 'Rock',
 315         },
 316         'playlist_count': 20,
 317     }, {
 318         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
 319         'md5': '32dcdfddddf9ec6917fc88ca26d36282',
 320         'info_dict': {
 321             'id': 'USCMV1100073',
 322             'ext': 'mp4',
 323             'title': 'Birdman - Y.U. MAD',
 324             'timestamp': 1323417600,
 325             'upload_date': '20111209',
 326             'uploader': 'Birdman',
 327             'track': 'Y.U. MAD',
 328             'artist': 'Birdman',
 329             'genre': 'Rap/Hip-Hop',
 330         },
 331         'expected_warnings': ['Unable to download SMIL file'],
 332     }, {
 333         'url': 'http://www.vevo.com/watch/genre/rock?index=0',
 334         'only_matching': True,
 335     }]
 336
 337     def _real_extract(self, url):
 338         mobj = re.match(self._VALID_URL, url)
 339         playlist_id = mobj.group('id')
 340         playlist_kind = mobj.group('kind')
 341
 342         webpage = self._download_webpage(url, playlist_id)
 343
 344         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 345         index = qs.get('index', [None])[0]
 346
 347         if index:
 348             video_id = self._search_regex(
 349                 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
 350                 webpage, 'video id', default=None, group='id')
 351             if video_id:
 352                 return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
 353
 354         playlists = self._extract_json(webpage, playlist_id)['default']['%ss' % playlist_kind]
 355
 356         playlist = (list(playlists.values())[0]
 357                     if playlist_kind == 'playlist' else playlists[playlist_id])
 358
 359         entries = [
 360             self.url_result('vevo:%s' % src, VevoIE.ie_key())
 361             for src in playlist['isrcs']]
 362
 363         return self.playlist_result(
 364             entries, playlist.get('playlistId') or playlist_id,
 365             playlist.get('name'), playlist.get('description'))