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_etree_fromstring,
   8     compat_str,
   9     compat_urlparse,
  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     _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com'
 144     _SOURCE_TYPES = {
 145         0: 'youtube',
 146         1: 'brightcove',
 147         2: 'http',
 148         3: 'hls_ios',
 149         4: 'hls',
 150         5: 'smil',  # http
 151         7: 'f4m_cc',
 152         8: 'f4m_ak',
 153         9: 'f4m_l3',
 154         10: 'ism',
 155         13: 'smil',  # rtmp
 156         18: 'dash',
 157     }
 158     _VERSIONS = {
 159         0: 'youtube',  # only in AuthenticateVideo videoVersions
 160         1: 'level3',
 161         2: 'akamai',
 162         3: 'level3',
 163         4: 'amazon',
 164     }
 165
 166     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
 167         formats = []
 168         els = smil.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
 169         for el in els:
 170             src = el.attrib['src']
 171             m = re.match(r'''(?xi)
 172                 (?P<ext>[a-z0-9]+):
 173                 (?P<path>
 174                     [/a-z0-9]+     # The directory and main part of the URL
 175                     _(?P<tbr>[0-9]+)k
 176                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 177                     _(?P<vcodec>[a-z0-9]+)
 178                     _(?P<vbr>[0-9]+)
 179                     _(?P<acodec>[a-z0-9]+)
 180                     _(?P<abr>[0-9]+)
 181                     \.[a-z0-9]+  # File extension
 182                 )''', src)
 183             if not m:
 184                 continue
 185
 186             format_url = self._SMIL_BASE_URL + m.group('path')
 187             formats.append({
 188                 'url': format_url,
 189                 'format_id': 'smil_' + m.group('tbr'),
 190                 'vcodec': m.group('vcodec'),
 191                 'acodec': m.group('acodec'),
 192                 'tbr': int(m.group('tbr')),
 193                 'vbr': int(m.group('vbr')),
 194                 'abr': int(m.group('abr')),
 195                 'ext': m.group('ext'),
 196                 'width': int(m.group('width')),
 197                 'height': int(m.group('height')),
 198             })
 199         return formats
 200
 201     def _initialize_api(self, video_id):
 202         req = sanitized_Request(
 203             'http://www.vevo.com/auth', data=b'')
 204         webpage = self._download_webpage(
 205             req, None,
 206             note='Retrieving oauth token',
 207             errnote='Unable to retrieve oauth token')
 208
 209         if 'THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION' in webpage:
 210             self.raise_geo_restricted(
 211                 '%s said: This page is currently unavailable in your region' % self.IE_NAME)
 212
 213         auth_info = self._parse_json(webpage, video_id)
 214         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token']
 215
 216     def _call_api(self, path, *args, **kwargs):
 217         return self._download_json(self._api_url_template % path, *args, **kwargs)
 218
 219     def _real_extract(self, url):
 220         video_id = self._match_id(url)
 221
 222         json_url = 'http://api.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
 223         response = self._download_json(
 224             json_url, video_id, 'Downloading video info',
 225             'Unable to download info', fatal=False) or {}
 226         video_info = response.get('video') or {}
 227         artist = None
 228         featured_artist = None
 229         uploader = None
 230         view_count = None
 231         formats = []
 232
 233         if not video_info:
 234             try:
 235                 self._initialize_api(video_id)
 236             except ExtractorError:
 237                 ytid = response.get('errorInfo', {}).get('ytid')
 238                 if ytid:
 239                     self.report_warning(
 240                         'Video is geoblocked, trying with the YouTube video %s' % ytid)
 241                     return self.url_result(ytid, 'Youtube', ytid)
 242
 243                 raise
 244
 245             video_info = self._call_api(
 246                 'video/%s' % video_id, video_id, 'Downloading api video info',
 247                 'Failed to download video info')
 248
 249             video_versions = self._call_api(
 250                 'video/%s/streams' % video_id, video_id,
 251                 'Downloading video versions info',
 252                 'Failed to download video versions info',
 253                 fatal=False)
 254
 255             # Some videos are only available via webpage (e.g.
 256             # https://github.com/rg3/youtube-dl/issues/9366)
 257             if not video_versions:
 258                 webpage = self._download_webpage(url, video_id)
 259                 video_versions = self._extract_json(webpage, video_id, 'streams')[video_id][0]
 260
 261             timestamp = parse_iso8601(video_info.get('releaseDate'))
 262             artists = video_info.get('artists')
 263             for curr_artist in artists:
 264                 if curr_artist.get('role') == 'Featured':
 265                     featured_artist = curr_artist['name']
 266                 else:
 267                     artist = uploader = curr_artist['name']
 268             view_count = int_or_none(video_info.get('views', {}).get('total'))
 269
 270             for video_version in video_versions:
 271                 version = self._VERSIONS.get(video_version['version'])
 272                 version_url = video_version.get('url')
 273                 if not version_url:
 274                     continue
 275
 276                 if '.ism' in version_url:
 277                     continue
 278                 elif '.mpd' in version_url:
 279                     formats.extend(self._extract_mpd_formats(
 280                         version_url, video_id, mpd_id='dash-%s' % version,
 281                         note='Downloading %s MPD information' % version,
 282                         errnote='Failed to download %s MPD information' % version,
 283                         fatal=False))
 284                 elif '.m3u8' in version_url:
 285                     formats.extend(self._extract_m3u8_formats(
 286                         version_url, video_id, 'mp4', 'm3u8_native',
 287                         m3u8_id='hls-%s' % version,
 288                         note='Downloading %s m3u8 information' % version,
 289                         errnote='Failed to download %s m3u8 information' % version,
 290                         fatal=False))
 291                 else:
 292                     m = re.search(r'''(?xi)
 293                         _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 294                         _(?P<vcodec>[a-z0-9]+)
 295                         _(?P<vbr>[0-9]+)
 296                         _(?P<acodec>[a-z0-9]+)
 297                         _(?P<abr>[0-9]+)
 298                         \.(?P<ext>[a-z0-9]+)''', version_url)
 299                     if not m:
 300                         continue
 301
 302                     formats.append({
 303                         'url': version_url,
 304                         'format_id': 'http-%s-%s' % (version, video_version['quality']),
 305                         'vcodec': m.group('vcodec'),
 306                         'acodec': m.group('acodec'),
 307                         'vbr': int(m.group('vbr')),
 308                         'abr': int(m.group('abr')),
 309                         'ext': m.group('ext'),
 310                         'width': int(m.group('width')),
 311                         'height': int(m.group('height')),
 312                     })
 313         else:
 314             timestamp = int_or_none(self._search_regex(
 315                 r'/Date\((\d+)\)/',
 316                 video_info['releaseDate'], 'release date', fatal=False),
 317                 scale=1000)
 318             artists = video_info.get('mainArtists')
 319             if artists:
 320                 artist = uploader = artists[0]['artistName']
 321
 322             featured_artists = video_info.get('featuredArtists')
 323             if featured_artists:
 324                 featured_artist = featured_artists[0]['artistName']
 325
 326             smil_parsed = False
 327             for video_version in video_info['videoVersions']:
 328                 version = self._VERSIONS.get(video_version['version'])
 329                 if version == 'youtube':
 330                     continue
 331                 else:
 332                     source_type = self._SOURCE_TYPES.get(video_version['sourceType'])
 333                     renditions = compat_etree_fromstring(video_version['data'])
 334                     if source_type == 'http':
 335                         for rend in renditions.findall('rendition'):
 336                             attr = rend.attrib
 337                             formats.append({
 338                                 'url': attr['url'],
 339                                 'format_id': 'http-%s-%s' % (version, attr['name']),
 340                                 'height': int_or_none(attr.get('frameheight')),
 341                                 'width': int_or_none(attr.get('frameWidth')),
 342                                 'tbr': int_or_none(attr.get('totalBitrate')),
 343                                 'vbr': int_or_none(attr.get('videoBitrate')),
 344                                 'abr': int_or_none(attr.get('audioBitrate')),
 345                                 'vcodec': attr.get('videoCodec'),
 346                                 'acodec': attr.get('audioCodec'),
 347                             })
 348                     elif source_type == 'hls':
 349                         formats.extend(self._extract_m3u8_formats(
 350                             renditions.find('rendition').attrib['url'], video_id,
 351                             'mp4', 'm3u8_native', m3u8_id='hls-%s' % version,
 352                             note='Downloading %s m3u8 information' % version,
 353                             errnote='Failed to download %s m3u8 information' % version,
 354                             fatal=False))
 355                     elif source_type == 'smil' and version == 'level3' and not smil_parsed:
 356                         formats.extend(self._extract_smil_formats(
 357                             renditions.find('rendition').attrib['url'], video_id, False))
 358                         smil_parsed = True
 359         self._sort_formats(formats)
 360
 361         track = video_info['title']
 362         if featured_artist:
 363             artist = '%s ft. %s' % (artist, featured_artist)
 364         title = '%s - %s' % (artist, track) if artist else track
 365
 366         genres = video_info.get('genres')
 367         genre = (
 368             genres[0] if genres and isinstance(genres, list) and
 369             isinstance(genres[0], compat_str) else None)
 370
 371         is_explicit = video_info.get('isExplicit')
 372         if is_explicit is True:
 373             age_limit = 18
 374         elif is_explicit is False:
 375             age_limit = 0
 376         else:
 377             age_limit = None
 378
 379         duration = video_info.get('duration')
 380
 381         return {
 382             'id': video_id,
 383             'title': title,
 384             'formats': formats,
 385             'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
 386             'timestamp': timestamp,
 387             'uploader': uploader,
 388             'duration': duration,
 389             'view_count': view_count,
 390             'age_limit': age_limit,
 391             'track': track,
 392             'artist': uploader,
 393             'genre': genre,
 394         }
 395
 396
 397 class VevoPlaylistIE(VevoBaseIE):
 398     _VALID_URL = r'https?://(?:www\.)?vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
 399
 400     _TESTS = [{
 401         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 402         'info_dict': {
 403             'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 404             'title': 'Best-Of: Birdman',
 405         },
 406         'playlist_count': 10,
 407     }, {
 408         'url': 'http://www.vevo.com/watch/genre/rock',
 409         'info_dict': {
 410             'id': 'rock',
 411             'title': 'Rock',
 412         },
 413         'playlist_count': 20,
 414     }, {
 415         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
 416         'md5': '32dcdfddddf9ec6917fc88ca26d36282',
 417         'info_dict': {
 418             'id': 'USCMV1100073',
 419             'ext': 'mp4',
 420             'title': 'Birdman - Y.U. MAD',
 421             'timestamp': 1323417600,
 422             'upload_date': '20111209',
 423             'uploader': 'Birdman',
 424             'track': 'Y.U. MAD',
 425             'artist': 'Birdman',
 426             'genre': 'Rap/Hip-Hop',
 427         },
 428         'expected_warnings': ['Unable to download SMIL file'],
 429     }, {
 430         'url': 'http://www.vevo.com/watch/genre/rock?index=0',
 431         'only_matching': True,
 432     }]
 433
 434     def _real_extract(self, url):
 435         mobj = re.match(self._VALID_URL, url)
 436         playlist_id = mobj.group('id')
 437         playlist_kind = mobj.group('kind')
 438
 439         webpage = self._download_webpage(url, playlist_id)
 440
 441         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 442         index = qs.get('index', [None])[0]
 443
 444         if index:
 445             video_id = self._search_regex(
 446                 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
 447                 webpage, 'video id', default=None, group='id')
 448             if video_id:
 449                 return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
 450
 451         playlists = self._extract_json(webpage, playlist_id, '%ss' % playlist_kind)
 452
 453         playlist = (list(playlists.values())[0]
 454                     if playlist_kind == 'playlist' else playlists[playlist_id])
 455
 456         entries = [
 457             self.url_result('vevo:%s' % src, VevoIE.ie_key())
 458             for src in playlist['isrcs']]
 459
 460         return self.playlist_result(
 461             entries, playlist.get('playlistId') or playlist_id,
 462             playlist.get('name'), playlist.get('description'))