1 from __future__ import unicode_literals
6 from .common import InfoExtractor
20 class VevoBaseIE(InfoExtractor):
21 def _extract_json(self, webpage, video_id):
22 return self._parse_json(
24 r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>',
25 webpage, 'initial store'),
29 class VevoIE(VevoBaseIE):
31 Accepts urls from vevo.com or in the format 'vevo:{id}'
32 (currently used by MTVIE and MySpaceIE)
35 (?:https?://(?:www\.)?vevo\.com/watch/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?|
36 https?://cache\.vevo\.com/m/html/embed\.html\?video=|
37 https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
42 'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
43 'md5': '95ee28ee45e70130e3ab02b0f579ae23',
47 'title': 'Hurts - Somebody to Die For',
48 'timestamp': 1372057200,
49 'upload_date': '20130624',
51 'track': 'Somebody to Die For',
55 'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
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',
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',
71 'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
73 'note': 'Age-limited video',
74 'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
78 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
80 'timestamp': 1372888800,
81 'upload_date': '20130703',
82 'uploader': 'Justin Timberlake',
83 'track': 'Tunnel Vision (Explicit)',
84 'artist': 'Justin Timberlake',
87 'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
89 'note': 'No video_info',
90 'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000',
91 'md5': '8b83cc492d72fc9cf74a02acee7dc1b0',
95 'title': 'K Camp ft. T.I. - Till I Die',
97 'timestamp': 1449468000,
98 'upload_date': '20151207',
100 'track': 'Till I Die',
104 'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
106 'note': 'Featured test',
107 'url': 'https://www.vevo.com/watch/lemaitre/Wait/USUV71402190',
108 'md5': 'd28675e5e8805035d949dc5cf161071d',
110 'id': 'USUV71402190',
112 'title': 'Lemaitre ft. LoLo - Wait',
114 'timestamp': 1413432000,
115 'upload_date': '20141016',
116 'uploader': 'Lemaitre',
118 'artist': 'Lemaitre',
119 'genre': 'Electronic',
121 'expected_warnings': ['Unable to download SMIL file', 'Unable to download info'],
123 'note': 'Only available via webpage',
124 'url': 'http://www.vevo.com/watch/GBUV71600656',
125 'md5': '67e79210613865b66a47c33baa5e37fe',
127 'id': 'GBUV71600656',
129 'title': 'ABC - Viva Love',
131 'timestamp': 1461830400,
132 'upload_date': '20160428',
134 'track': 'Viva Love',
138 'expected_warnings': ['Failed to download video versions info'],
140 # no genres available
141 'url': 'http://www.vevo.com/watch/INS171400764',
142 'only_matching': True,
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,
150 0: 'youtube', # only in AuthenticateVideo videoVersions
157 def _initialize_api(self, video_id):
158 post_data = json.dumps({
159 'client_id': 'SPupX1tvqFEopQ1YS6SS',
160 'grant_type': 'urn:vevo:params:oauth:grant-type:anonymous',
163 'Content-Type': 'application/json',
165 req = sanitized_Request(
166 'https://accounts.vevo.com/token', post_data, headers)
167 webpage = self._download_webpage(
169 note='Retrieving oauth token',
170 errnote='Unable to retrieve oauth token')
172 if re.search(r'(?i)THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION', webpage):
173 self.raise_geo_restricted(
174 '%s said: This page is currently unavailable in your region' % self.IE_NAME)
176 auth_info = self._parse_json(webpage, video_id)
177 self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['legacy_token']
179 def _call_api(self, path, *args, **kwargs):
181 data = self._download_json(self._api_url_template % path, *args, **kwargs)
182 except ExtractorError as e:
183 if isinstance(e.cause, compat_HTTPError):
184 errors = self._parse_json(e.cause.read().decode(), None)['errors']
185 error_message = ', '.join([error['message'] for error in errors])
186 raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message), expected=True)
190 def _real_extract(self, url):
191 video_id = self._match_id(url)
193 self._initialize_api(video_id)
195 video_info = self._call_api(
196 'video/%s' % video_id, video_id, 'Downloading api video info',
197 'Failed to download video info')
199 video_versions = self._call_api(
200 'video/%s/streams' % video_id, video_id,
201 'Downloading video versions info',
202 'Failed to download video versions info',
205 # Some videos are only available via webpage (e.g.
206 # https://github.com/rg3/youtube-dl/issues/9366)
207 if not video_versions:
208 webpage = self._download_webpage(url, video_id)
209 json_data = self._extract_json(webpage, video_id)
210 if 'streams' in json_data.get('default', {}):
211 video_versions = json_data['default']['streams'][video_id][0]
215 for key, value in json_data['apollo']['data'].items()
216 if key.startswith('%s.streams' % video_id)]
220 featured_artist = None
221 artists = video_info.get('artists')
222 for curr_artist in artists:
223 if curr_artist.get('role') == 'Featured':
224 featured_artist = curr_artist['name']
226 artist = uploader = curr_artist['name']
229 for video_version in video_versions:
230 version = self._VERSIONS.get(video_version.get('version'), 'generic')
231 version_url = video_version.get('url')
235 if '.ism' in version_url:
237 elif '.mpd' in version_url:
238 formats.extend(self._extract_mpd_formats(
239 version_url, video_id, mpd_id='dash-%s' % version,
240 note='Downloading %s MPD information' % version,
241 errnote='Failed to download %s MPD information' % version,
243 elif '.m3u8' in version_url:
244 formats.extend(self._extract_m3u8_formats(
245 version_url, video_id, 'mp4', 'm3u8_native',
246 m3u8_id='hls-%s' % version,
247 note='Downloading %s m3u8 information' % version,
248 errnote='Failed to download %s m3u8 information' % version,
251 m = re.search(r'''(?xi)
252 _(?P<width>[0-9]+)x(?P<height>[0-9]+)
253 _(?P<vcodec>[a-z0-9]+)
255 _(?P<acodec>[a-z0-9]+)
257 \.(?P<ext>[a-z0-9]+)''', version_url)
263 'format_id': 'http-%s-%s' % (version, video_version['quality']),
264 'vcodec': m.group('vcodec'),
265 'acodec': m.group('acodec'),
266 'vbr': int(m.group('vbr')),
267 'abr': int(m.group('abr')),
268 'ext': m.group('ext'),
269 'width': int(m.group('width')),
270 'height': int(m.group('height')),
272 self._sort_formats(formats)
274 track = video_info['title']
276 artist = '%s ft. %s' % (artist, featured_artist)
277 title = '%s - %s' % (artist, track) if artist else track
279 genres = video_info.get('genres')
281 genres[0] if genres and isinstance(genres, list) and
282 isinstance(genres[0], compat_str) else None)
284 is_explicit = video_info.get('isExplicit')
285 if is_explicit is True:
287 elif is_explicit is False:
296 'thumbnail': video_info.get('imageUrl') or video_info.get('thumbnailUrl'),
297 'timestamp': parse_iso8601(video_info.get('releaseDate')),
298 'uploader': uploader,
299 'duration': int_or_none(video_info.get('duration')),
300 'view_count': int_or_none(video_info.get('views', {}).get('total')),
301 'age_limit': age_limit,
308 class VevoPlaylistIE(VevoBaseIE):
309 _VALID_URL = r'https?://(?:www\.)?vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
312 'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
314 'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
315 'title': 'Best-Of: Birdman',
317 'playlist_count': 10,
319 'url': 'http://www.vevo.com/watch/genre/rock',
324 'playlist_count': 20,
326 'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
327 'md5': '32dcdfddddf9ec6917fc88ca26d36282',
329 'id': 'USCMV1100073',
331 'title': 'Birdman - Y.U. MAD',
332 'timestamp': 1323417600,
333 'upload_date': '20111209',
334 'uploader': 'Birdman',
337 'genre': 'Rap/Hip-Hop',
339 'expected_warnings': ['Unable to download SMIL file'],
341 'url': 'http://www.vevo.com/watch/genre/rock?index=0',
342 'only_matching': True,
345 def _real_extract(self, url):
346 mobj = re.match(self._VALID_URL, url)
347 playlist_id = mobj.group('id')
348 playlist_kind = mobj.group('kind')
350 webpage = self._download_webpage(url, playlist_id)
352 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
353 index = qs.get('index', [None])[0]
356 video_id = self._search_regex(
357 r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
358 webpage, 'video id', default=None, group='id')
360 return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
362 playlists = self._extract_json(webpage, playlist_id)['default']['%ss' % playlist_kind]
364 playlist = (list(playlists.values())[0]
365 if playlist_kind == 'playlist' else playlists[playlist_id])
368 self.url_result('vevo:%s' % src, VevoIE.ie_key())
369 for src in playlist['isrcs']]
371 return self.playlist_result(
372 entries, playlist.get('playlistId') or playlist_id,
373 playlist.get('name'), playlist.get('description'))