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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_urlparse,
  10     compat_HTTPError,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     int_or_none,
  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            https?://embed\.vevo\.com/.*?[?&]isrc=|
  38            vevo:)
  39         (?P<id>[^&?#]+)'''
  40
  41     _TESTS = [{
  42         'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
  43         'md5': '95ee28ee45e70130e3ab02b0f579ae23',
  44         'info_dict': {
  45             'id': 'GB1101300280',
  46             'ext': 'mp4',
  47             'title': 'Hurts - Somebody to Die For',
  48             'timestamp': 1372057200,
  49             'upload_date': '20130624',
  50             'uploader': 'Hurts',
  51             'track': 'Somebody to Die For',
  52             'artist': 'Hurts',
  53             'genre': 'Pop',
  54         },
  55         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
  56     }, {
  57         'note': 'v3 SMIL format',
  58         'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
  59         'md5': 'f6ab09b034f8c22969020b042e5ac7fc',
  60         'info_dict': {
  61             'id': 'USUV71302923',
  62             'ext': 'mp4',
  63             'title': 'Cassadee Pope - I Wish I Could Break Your Heart',
  64             'timestamp': 1392796919,
  65             'upload_date': '20140219',
  66             'uploader': 'Cassadee Pope',
  67             'track': 'I Wish I Could Break Your Heart',
  68             'artist': 'Cassadee Pope',
  69             'genre': 'Country',
  70         },
  71         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
  72     }, {
  73         'note': 'Age-limited video',
  74         'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
  75         'info_dict': {
  76             'id': 'USRV81300282',
  77             'ext': 'mp4',
  78             'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
  79             'age_limit': 18,
  80             'timestamp': 1372888800,
  81             'upload_date': '20130703',
  82             'uploader': 'Justin Timberlake',
  83             'track': 'Tunnel Vision (Explicit)',
  84             'artist': 'Justin Timberlake',
  85             'genre': 'Pop',
  86         },
  87         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
  88     }, {
  89         'note': 'No video_info',
  90         'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000',
  91         'md5': '8b83cc492d72fc9cf74a02acee7dc1b0',
  92         'info_dict': {
  93             'id': 'USUV71503000',
  94             'ext': 'mp4',
  95             'title': 'K Camp ft. T.I. - Till I Die',
  96             'age_limit': 18,
  97             'timestamp': 1449468000,
  98             'upload_date': '20151207',
  99             'uploader': 'K Camp',
 100             'track': 'Till I Die',
 101             'artist': 'K Camp',
 102             'genre': 'Hip-Hop',
 103         },
 104         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
 105     }, {
 106         'note': 'Featured test',
 107         'url': 'https://www.vevo.com/watch/lemaitre/Wait/USUV71402190',
 108         'md5': 'd28675e5e8805035d949dc5cf161071d',
 109         'info_dict': {
 110             'id': 'USUV71402190',
 111             'ext': 'mp4',
 112             'title': 'Lemaitre ft. LoLo - Wait',
 113             'age_limit': 0,
 114             'timestamp': 1413432000,
 115             'upload_date': '20141016',
 116             'uploader': 'Lemaitre',
 117             'track': 'Wait',
 118             'artist': 'Lemaitre',
 119             'genre': 'Electronic',
 120         },
 121         'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
 122     }, {
 123         'note': 'Only available via webpage',
 124         'url': 'http://www.vevo.com/watch/GBUV71600656',
 125         'md5': '67e79210613865b66a47c33baa5e37fe',
 126         'info_dict': {
 127             'id': 'GBUV71600656',
 128             'ext': 'mp4',
 129             'title': 'ABC - Viva Love',
 130             'age_limit': 0,
 131             'timestamp': 1461830400,
 132             'upload_date': '20160428',
 133             'uploader': 'ABC',
 134             'track': 'Viva Love',
 135             'artist': 'ABC',
 136             'genre': 'Pop',
 137         },
 138         'expected_warnings': ['Failed to download video versions info'],
 139     }, {
 140         # no genres available
 141         'url': 'http://www.vevo.com/watch/INS171400764',
 142         'only_matching': True,
 143     }, {
 144         # Another case available only via the webpage; using streams/streamsV3 formats
 145         # Geo-restricted to Netherlands/Germany
 146         'url': 'http://www.vevo.com/watch/boostee/pop-corn-clip-officiel/FR1A91600909',
 147         'only_matching': True,
 148     }, {
 149         'url': 'https://embed.vevo.com/?isrc=USH5V1923499&partnerId=4d61b777-8023-4191-9ede-497ed6c24647&partnerAdCode=',
 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('utf-8'),
 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                 '%s said: This page is currently unavailable in your region' % self.IE_NAME)
 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, compat_HTTPError):
 185                 errors = self._parse_json(e.cause.read().decode(), None)['errors']
 186                 error_message = ', '.join([error['message'] for error in errors])
 187                 raise ExtractorError('%s said: %s' % (self.IE_NAME, 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             'video/%s' % video_id, video_id, 'Downloading api video info',
 198             'Failed to download video info')
 199
 200         video_versions = self._call_api(
 201             'video/%s/streams' % video_id, 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('%s.streams' % video_id)]
 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='dash-%s' % version,
 241                     note='Downloading %s MPD information' % version,
 242                     errnote='Failed to download %s MPD information' % version,
 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='hls-%s' % version,
 248                     note='Downloading %s m3u8 information' % version,
 249                     errnote='Failed to download %s m3u8 information' % version,
 250                     fatal=False))
 251             else:
 252                 m = re.search(r'''(?xi)
 253                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 254                     _(?P<vcodec>[a-z0-9]+)
 255                     _(?P<vbr>[0-9]+)
 256                     _(?P<acodec>[a-z0-9]+)
 257                     _(?P<abr>[0-9]+)
 258                     \.(?P<ext>[a-z0-9]+)''', version_url)
 259                 if not m:
 260                     continue
 261
 262                 formats.append({
 263                     'url': version_url,
 264                     'format_id': 'http-%s-%s' % (version, video_version['quality']),
 265                     'vcodec': m.group('vcodec'),
 266                     'acodec': m.group('acodec'),
 267                     'vbr': int(m.group('vbr')),
 268                     'abr': int(m.group('abr')),
 269                     'ext': m.group('ext'),
 270                     'width': int(m.group('width')),
 271                     'height': int(m.group('height')),
 272                 })
 273         self._sort_formats(formats)
 274
 275         track = video_info['title']
 276         if featured_artist:
 277             artist = '%s ft. %s' % (artist, featured_artist)
 278         title = '%s - %s' % (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], compat_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/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 314         'info_dict': {
 315             'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 316             'title': 'Best-Of: Birdman',
 317         },
 318         'playlist_count': 10,
 319     }, {
 320         'url': 'http://www.vevo.com/watch/genre/rock',
 321         'info_dict': {
 322             'id': 'rock',
 323             'title': 'Rock',
 324         },
 325         'playlist_count': 20,
 326     }, {
 327         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
 328         'md5': '32dcdfddddf9ec6917fc88ca26d36282',
 329         'info_dict': {
 330             'id': 'USCMV1100073',
 331             'ext': 'mp4',
 332             'title': 'Birdman - Y.U. MAD',
 333             'timestamp': 1323417600,
 334             'upload_date': '20111209',
 335             'uploader': 'Birdman',
 336             'track': 'Y.U. MAD',
 337             'artist': 'Birdman',
 338             'genre': 'Rap/Hip-Hop',
 339         },
 340         'expected_warnings': ['Unable to download SMIL file'],
 341     }, {
 342         'url': 'http://www.vevo.com/watch/genre/rock?index=0',
 343         'only_matching': True,
 344     }]
 345
 346     def _real_extract(self, url):
 347         mobj = re.match(self._VALID_URL, url)
 348         playlist_id = mobj.group('id')
 349         playlist_kind = mobj.group('kind')
 350
 351         webpage = self._download_webpage(url, playlist_id)
 352
 353         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 354         index = qs.get('index', [None])[0]
 355
 356         if index:
 357             video_id = self._search_regex(
 358                 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
 359                 webpage, 'video id', default=None, group='id')
 360             if video_id:
 361                 return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
 362
 363         playlists = self._extract_json(webpage, playlist_id)['default']['%ss' % playlist_kind]
 364
 365         playlist = (list(playlists.values())[0]
 366                     if playlist_kind == 'playlist' else playlists[playlist_id])
 367
 368         entries = [
 369             self.url_result('vevo:%s' % src, VevoIE.ie_key())
 370             for src in playlist['isrcs']]
 371
 372         return self.playlist_result(
 373             entries, playlist.get('playlistId') or playlist_id,
 374             playlist.get('name'), playlist.get('description'))