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, item):
  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)['default'][item]
  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     _VERSIONS = {
 144         0: 'youtube',  # only in AuthenticateVideo videoVersions
 145         1: 'level3',
 146         2: 'akamai',
 147         3: 'level3',
 148         4: 'amazon',
 149     }
 150
 151     def _initialize_api(self, video_id):
 152         req = sanitized_Request(
 153             'http://www.vevo.com/auth', data=b'')
 154         webpage = self._download_webpage(
 155             req, None,
 156             note='Retrieving oauth token',
 157             errnote='Unable to retrieve oauth token')
 158
 159         if re.search(r'(?i)THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION', webpage):
 160             self.raise_geo_restricted(
 161                 '%s said: This page is currently unavailable in your region' % self.IE_NAME)
 162
 163         auth_info = self._parse_json(webpage, video_id)
 164         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token']
 165
 166     def _call_api(self, path, *args, **kwargs):
 167         try:
 168             data = self._download_json(self._api_url_template % path, *args, **kwargs)
 169         except ExtractorError as e:
 170             if isinstance(e.cause, compat_HTTPError):
 171                 errors = self._parse_json(e.cause.read().decode(), None)['errors']
 172                 error_message = ', '.join([error['message'] for error in errors])
 173                 raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 174             raise
 175         return data
 176
 177     def _real_extract(self, url):
 178         video_id = self._match_id(url)
 179
 180         self._initialize_api(video_id)
 181
 182         video_info = self._call_api(
 183             'video/%s' % video_id, video_id, 'Downloading api video info',
 184             'Failed to download video info')
 185
 186         video_versions = self._call_api(
 187             'video/%s/streams' % video_id, video_id,
 188             'Downloading video versions info',
 189             'Failed to download video versions info',
 190             fatal=False)
 191
 192         # Some videos are only available via webpage (e.g.
 193         # https://github.com/rg3/youtube-dl/issues/9366)
 194         if not video_versions:
 195             webpage = self._download_webpage(url, video_id)
 196             video_versions = self._extract_json(webpage, video_id, 'streams')[video_id][0]
 197
 198         uploader = None
 199         artist = None
 200         featured_artist = None
 201         artists = video_info.get('artists')
 202         for curr_artist in artists:
 203             if curr_artist.get('role') == 'Featured':
 204                 featured_artist = curr_artist['name']
 205             else:
 206                 artist = uploader = curr_artist['name']
 207
 208         formats = []
 209         for video_version in video_versions:
 210             version = self._VERSIONS.get(video_version['version'])
 211             version_url = video_version.get('url')
 212             if not version_url:
 213                 continue
 214
 215             if '.ism' in version_url:
 216                 continue
 217             elif '.mpd' in version_url:
 218                 formats.extend(self._extract_mpd_formats(
 219                     version_url, video_id, mpd_id='dash-%s' % version,
 220                     note='Downloading %s MPD information' % version,
 221                     errnote='Failed to download %s MPD information' % version,
 222                     fatal=False))
 223             elif '.m3u8' in version_url:
 224                 formats.extend(self._extract_m3u8_formats(
 225                     version_url, video_id, 'mp4', 'm3u8_native',
 226                     m3u8_id='hls-%s' % version,
 227                     note='Downloading %s m3u8 information' % version,
 228                     errnote='Failed to download %s m3u8 information' % version,
 229                     fatal=False))
 230             else:
 231                 m = re.search(r'''(?xi)
 232                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 233                     _(?P<vcodec>[a-z0-9]+)
 234                     _(?P<vbr>[0-9]+)
 235                     _(?P<acodec>[a-z0-9]+)
 236                     _(?P<abr>[0-9]+)
 237                     \.(?P<ext>[a-z0-9]+)''', version_url)
 238                 if not m:
 239                     continue
 240
 241                 formats.append({
 242                     'url': version_url,
 243                     'format_id': 'http-%s-%s' % (version, video_version['quality']),
 244                     'vcodec': m.group('vcodec'),
 245                     'acodec': m.group('acodec'),
 246                     'vbr': int(m.group('vbr')),
 247                     'abr': int(m.group('abr')),
 248                     'ext': m.group('ext'),
 249                     'width': int(m.group('width')),
 250                     'height': int(m.group('height')),
 251                 })
 252         self._sort_formats(formats)
 253
 254         track = video_info['title']
 255         if featured_artist:
 256             artist = '%s ft. %s' % (artist, featured_artist)
 257         title = '%s - %s' % (artist, track) if artist else track
 258
 259         genres = video_info.get('genres')
 260         genre = (
 261             genres[0] if genres and isinstance(genres, list) and
 262             isinstance(genres[0], compat_str) else None)
 263
 264         is_explicit = video_info.get('isExplicit')
 265         if is_explicit is True:
 266             age_limit = 18
 267         elif is_explicit is False:
 268             age_limit = 0
 269         else:
 270             age_limit = None
 271
 272         return {
 273             'id': video_id,
 274             'title': title,
 275             'formats': formats,
 276             'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
 277             'timestamp': parse_iso8601(video_info.get('releaseDate')),
 278             'uploader': uploader,
 279             'duration': int_or_none(video_info.get('duration')),
 280             'view_count': int_or_none(video_info.get('views', {}).get('total')),
 281             'age_limit': age_limit,
 282             'track': track,
 283             'artist': uploader,
 284             'genre': genre,
 285         }
 286
 287
 288 class VevoPlaylistIE(VevoBaseIE):
 289     _VALID_URL = r'https?://(?:www\.)?vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
 290
 291     _TESTS = [{
 292         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 293         'info_dict': {
 294             'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 295             'title': 'Best-Of: Birdman',
 296         },
 297         'playlist_count': 10,
 298     }, {
 299         'url': 'http://www.vevo.com/watch/genre/rock',
 300         'info_dict': {
 301             'id': 'rock',
 302             'title': 'Rock',
 303         },
 304         'playlist_count': 20,
 305     }, {
 306         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
 307         'md5': '32dcdfddddf9ec6917fc88ca26d36282',
 308         'info_dict': {
 309             'id': 'USCMV1100073',
 310             'ext': 'mp4',
 311             'title': 'Birdman - Y.U. MAD',
 312             'timestamp': 1323417600,
 313             'upload_date': '20111209',
 314             'uploader': 'Birdman',
 315             'track': 'Y.U. MAD',
 316             'artist': 'Birdman',
 317             'genre': 'Rap/Hip-Hop',
 318         },
 319         'expected_warnings': ['Unable to download SMIL file'],
 320     }, {
 321         'url': 'http://www.vevo.com/watch/genre/rock?index=0',
 322         'only_matching': True,
 323     }]
 324
 325     def _real_extract(self, url):
 326         mobj = re.match(self._VALID_URL, url)
 327         playlist_id = mobj.group('id')
 328         playlist_kind = mobj.group('kind')
 329
 330         webpage = self._download_webpage(url, playlist_id)
 331
 332         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 333         index = qs.get('index', [None])[0]
 334
 335         if index:
 336             video_id = self._search_regex(
 337                 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
 338                 webpage, 'video id', default=None, group='id')
 339             if video_id:
 340                 return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
 341
 342         playlists = self._extract_json(webpage, playlist_id, '%ss' % playlist_kind)
 343
 344         playlist = (list(playlists.values())[0]
 345                     if playlist_kind == 'playlist' else playlists[playlist_id])
 346
 347         entries = [
 348             self.url_result('vevo:%s' % src, VevoIE.ie_key())
 349             for src in playlist['isrcs']]
 350
 351         return self.playlist_result(
 352             entries, playlist.get('playlistId') or playlist_id,
 353             playlist.get('name'), playlist.get('description'))