1 from __future__ import unicode_literals
5 from .common import InfoExtractor
6 from ..compat import compat_etree_fromstring
14 class VevoIE(InfoExtractor):
16 Accepts urls from vevo.com or in the format 'vevo:{id}'
17 (currently used by MTVIE and MySpaceIE)
20 (?:https?://www\.vevo\.com/watch/(?:[^/]+/(?:[^/]+/)?)?|
21 https?://cache\.vevo\.com/m/html/embed\.html\?video=|
22 https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
27 'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
28 "md5": "95ee28ee45e70130e3ab02b0f579ae23",
32 "upload_date": "20130624",
34 "title": "Somebody to Die For",
38 # timestamp and upload_date are often incorrect; seem to change randomly
42 'note': 'v3 SMIL format',
43 'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
44 'md5': 'f6ab09b034f8c22969020b042e5ac7fc',
48 'upload_date': '20140219',
49 'uploader': 'Cassadee Pope',
50 'title': 'I Wish I Could Break Your Heart',
56 'note': 'Age-limited video',
57 'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
62 'title': 'Tunnel Vision (Explicit)',
63 'uploader': 'Justin Timberlake',
64 'upload_date': 're:2013070[34]',
68 'skip_download': 'true',
71 _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/'
73 def _real_initialize(self):
74 req = sanitized_Request(
75 'http://www.vevo.com/auth', data=b'')
76 webpage = self._download_webpage(
78 note='Retrieving oauth token',
79 errnote='Unable to retrieve oauth token',
82 self._oauth_token = None
84 if 'THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION' in webpage:
85 raise ExtractorError('%s said: This page is currently unavailable in your region.' % self.IE_NAME, expected=True)
87 self._oauth_token = self._search_regex(
88 r'access_token":\s*"([^"]+)"',
89 webpage, 'access token', fatal=False)
91 def _formats_from_json(self, video_info):
92 last_version = {'version': -1}
93 for version in video_info['videoVersions']:
94 # These are the HTTP downloads, other types are for different manifests
95 if version['sourceType'] == 2:
96 if version['version'] > last_version['version']:
97 last_version = version
98 if last_version['version'] == -1:
99 raise ExtractorError('Unable to extract last version of the video')
101 renditions = compat_etree_fromstring(last_version['data'])
103 # Already sorted from worst to best quality
104 for rend in renditions.findall('rendition'):
106 format_note = '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr
109 'format_id': attr['name'],
110 'format_note': format_note,
111 'height': int(attr['frameheight']),
112 'width': int(attr['frameWidth']),
116 def _formats_from_smil(self, smil_xml):
118 smil_doc = compat_etree_fromstring(smil_xml.encode('utf-8'))
119 els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
121 src = el.attrib['src']
122 m = re.match(r'''(?xi)
125 [/a-z0-9]+ # The directory and main part of the URL
127 _(?P<width>[0-9]+)x(?P<height>[0-9]+)
128 _(?P<vcodec>[a-z0-9]+)
130 _(?P<acodec>[a-z0-9]+)
132 \.[a-z0-9]+ # File extension
137 format_url = self._SMIL_BASE_URL + m.group('path')
140 'format_id': 'SMIL_' + m.group('cbr'),
141 'vcodec': m.group('vcodec'),
142 'acodec': m.group('acodec'),
143 'vbr': int(m.group('vbr')),
144 'abr': int(m.group('abr')),
145 'ext': m.group('ext'),
146 'width': int(m.group('width')),
147 'height': int(m.group('height')),
151 def _download_api_formats(self, video_id):
152 if not self._oauth_token:
153 self._downloader.report_warning(
154 'No oauth token available, skipping API HLS download')
157 api_url = 'https://apiv2.vevo.com/video/%s/streams/hls?token=%s' % (
158 video_id, self._oauth_token)
159 api_data = self._download_json(
161 note='Downloading HLS formats',
162 errnote='Failed to download HLS format list', fatal=False)
166 m3u8_url = api_data[0]['url']
167 return self._extract_m3u8_formats(
168 m3u8_url, video_id, entry_protocol='m3u8_native', ext='mp4',
171 def _real_extract(self, url):
172 mobj = re.match(self._VALID_URL, url)
173 video_id = mobj.group('id')
175 json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
176 response = self._download_json(json_url, video_id)
177 video_info = response['video']
180 if 'statusMessage' in response:
181 raise ExtractorError('%s said: %s' % (self.IE_NAME, response['statusMessage']), expected=True)
182 raise ExtractorError('Unable to extract videos')
184 formats = self._formats_from_json(video_info)
186 is_explicit = video_info.get('isExplicit')
187 if is_explicit is True:
189 elif is_explicit is False:
194 # Download via HLS API
195 formats.extend(self._download_api_formats(video_id))
198 smil_blocks = sorted((
199 f for f in video_info['videoVersions']
200 if f['sourceType'] == 13),
201 key=lambda f: f['version'])
202 smil_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (
203 self._SMIL_BASE_URL, video_id, video_id.lower())
205 smil_url_m = self._search_regex(
206 r'url="([^"]+)"', smil_blocks[-1]['data'], 'SMIL URL',
208 if smil_url_m is not None:
209 smil_url = smil_url_m
211 smil_xml = self._download_webpage(
212 smil_url, video_id, 'Downloading SMIL info', fatal=False)
214 formats.extend(self._formats_from_smil(smil_xml))
216 self._sort_formats(formats)
217 timestamp_ms = int_or_none(self._search_regex(
219 video_info['launchDate'], 'launch date', fatal=False))
223 'title': video_info['title'],
225 'thumbnail': video_info['imageUrl'],
226 'timestamp': timestamp_ms // 1000,
227 'uploader': video_info['mainArtists'][0]['artistName'],
228 'duration': video_info['duration'],
229 'age_limit': age_limit,