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            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         webpage = self._download_webpage(
 158             'https://accounts.vevo.com/token', None,
 159             note='Retrieving oauth token',
 160             errnote='Unable to retrieve oauth token',
 161             data=json.dumps({
 162                 'client_id': 'SPupX1tvqFEopQ1YS6SS',
 163                 'grant_type': 'urn:vevo:params:oauth:grant-type:anonymous',
 164             }).encode('utf-8'),
 165             headers={
 166                 'Content-Type': 'application/json',
 167             })
 168
 169         if re.search(r'(?i)THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION', webpage):
 170             self.raise_geo_restricted(
 171                 '%s said: This page is currently unavailable in your region' % self.IE_NAME)
 172
 173         auth_info = self._parse_json(webpage, video_id)
 174         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['legacy_token']
 175
 176     def _call_api(self, path, *args, **kwargs):
 177         try:
 178             data = self._download_json(self._api_url_template % path, *args, **kwargs)
 179         except ExtractorError as e:
 180             if isinstance(e.cause, compat_HTTPError):
 181                 errors = self._parse_json(e.cause.read().decode(), None)['errors']
 182                 error_message = ', '.join([error['message'] for error in errors])
 183                 raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
 184             raise
 185         return data
 186
 187     def _real_extract(self, url):
 188         video_id = self._match_id(url)
 189
 190         self._initialize_api(video_id)
 191
 192         video_info = self._call_api(
 193             'video/%s' % video_id, video_id, 'Downloading api video info',
 194             'Failed to download video info')
 195
 196         video_versions = self._call_api(
 197             'video/%s/streams' % video_id, video_id,
 198             'Downloading video versions info',
 199             'Failed to download video versions info',
 200             fatal=False)
 201
 202         # Some videos are only available via webpage (e.g.
 203         # https://github.com/rg3/youtube-dl/issues/9366)
 204         if not video_versions:
 205             webpage = self._download_webpage(url, video_id)
 206             json_data = self._extract_json(webpage, video_id)
 207             if 'streams' in json_data.get('default', {}):
 208                 video_versions = json_data['default']['streams'][video_id][0]
 209             else:
 210                 video_versions = [
 211                     value
 212                     for key, value in json_data['apollo']['data'].items()
 213                     if key.startswith('%s.streams' % video_id)]
 214
 215         uploader = None
 216         artist = None
 217         featured_artist = None
 218         artists = video_info.get('artists')
 219         for curr_artist in artists:
 220             if curr_artist.get('role') == 'Featured':
 221                 featured_artist = curr_artist['name']
 222             else:
 223                 artist = uploader = curr_artist['name']
 224
 225         formats = []
 226         for video_version in video_versions:
 227             version = self._VERSIONS.get(video_version.get('version'), 'generic')
 228             version_url = video_version.get('url')
 229             if not version_url:
 230                 continue
 231
 232             if '.ism' in version_url:
 233                 continue
 234             elif '.mpd' in version_url:
 235                 formats.extend(self._extract_mpd_formats(
 236                     version_url, video_id, mpd_id='dash-%s' % version,
 237                     note='Downloading %s MPD information' % version,
 238                     errnote='Failed to download %s MPD information' % version,
 239                     fatal=False))
 240             elif '.m3u8' in version_url:
 241                 formats.extend(self._extract_m3u8_formats(
 242                     version_url, video_id, 'mp4', 'm3u8_native',
 243                     m3u8_id='hls-%s' % version,
 244                     note='Downloading %s m3u8 information' % version,
 245                     errnote='Failed to download %s m3u8 information' % version,
 246                     fatal=False))
 247             else:
 248                 m = re.search(r'''(?xi)
 249                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 250                     _(?P<vcodec>[a-z0-9]+)
 251                     _(?P<vbr>[0-9]+)
 252                     _(?P<acodec>[a-z0-9]+)
 253                     _(?P<abr>[0-9]+)
 254                     \.(?P<ext>[a-z0-9]+)''', version_url)
 255                 if not m:
 256                     continue
 257
 258                 formats.append({
 259                     'url': version_url,
 260                     'format_id': 'http-%s-%s' % (version, video_version['quality']),
 261                     'vcodec': m.group('vcodec'),
 262                     'acodec': m.group('acodec'),
 263                     'vbr': int(m.group('vbr')),
 264                     'abr': int(m.group('abr')),
 265                     'ext': m.group('ext'),
 266                     'width': int(m.group('width')),
 267                     'height': int(m.group('height')),
 268                 })
 269         self._sort_formats(formats)
 270
 271         track = video_info['title']
 272         if featured_artist:
 273             artist = '%s ft. %s' % (artist, featured_artist)
 274         title = '%s - %s' % (artist, track) if artist else track
 275
 276         genres = video_info.get('genres')
 277         genre = (
 278             genres[0] if genres and isinstance(genres, list) and
 279             isinstance(genres[0], compat_str) else None)
 280
 281         is_explicit = video_info.get('isExplicit')
 282         if is_explicit is True:
 283             age_limit = 18
 284         elif is_explicit is False:
 285             age_limit = 0
 286         else:
 287             age_limit = None
 288
 289         return {
 290             'id': video_id,
 291             'title': title,
 292             'formats': formats,
 293             'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
 294             'timestamp': parse_iso8601(video_info.get('releaseDate')),
 295             'uploader': uploader,
 296             'duration': int_or_none(video_info.get('duration')),
 297             'view_count': int_or_none(video_info.get('views', {}).get('total')),
 298             'age_limit': age_limit,
 299             'track': track,
 300             'artist': uploader,
 301             'genre': genre,
 302         }
 303
 304
 305 class VevoPlaylistIE(VevoBaseIE):
 306     _VALID_URL = r'https?://(?:www\.)?vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
 307
 308     _TESTS = [{
 309         'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 310         'info_dict': {
 311             'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
 312             'title': 'Best-Of: Birdman',
 313         },
 314         'playlist_count': 10,
 315     }, {
 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/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
 324         'md5': '32dcdfddddf9ec6917fc88ca26d36282',
 325         'info_dict': {
 326             'id': 'USCMV1100073',
 327             'ext': 'mp4',
 328             'title': 'Birdman - Y.U. MAD',
 329             'timestamp': 1323417600,
 330             'upload_date': '20111209',
 331             'uploader': 'Birdman',
 332             'track': 'Y.U. MAD',
 333             'artist': 'Birdman',
 334             'genre': 'Rap/Hip-Hop',
 335         },
 336         'expected_warnings': ['Unable to download SMIL file'],
 337     }, {
 338         'url': 'http://www.vevo.com/watch/genre/rock?index=0',
 339         'only_matching': True,
 340     }]
 341
 342     def _real_extract(self, url):
 343         mobj = re.match(self._VALID_URL, url)
 344         playlist_id = mobj.group('id')
 345         playlist_kind = mobj.group('kind')
 346
 347         webpage = self._download_webpage(url, playlist_id)
 348
 349         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 350         index = qs.get('index', [None])[0]
 351
 352         if index:
 353             video_id = self._search_regex(
 354                 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
 355                 webpage, 'video id', default=None, group='id')
 356             if video_id:
 357                 return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
 358
 359         playlists = self._extract_json(webpage, playlist_id)['default']['%ss' % playlist_kind]
 360
 361         playlist = (list(playlists.values())[0]
 362                     if playlist_kind == 'playlist' else playlists[playlist_id])
 363
 364         entries = [
 365             self.url_result('vevo:%s' % src, VevoIE.ie_key())
 366             for src in playlist['isrcs']]
 367
 368         return self.playlist_result(
 369             entries, playlist.get('playlistId') or playlist_id,
 370             playlist.get('name'), playlist.get('description'))