]>
jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/vevo.py
1 from __future__
import unicode_literals
4 import xml
.etree
.ElementTree
6 from .common
import InfoExtractor
13 class VevoIE(InfoExtractor
):
15 Accepts urls from vevo.com or in the format 'vevo:{id}'
16 (currently used by MTVIE and MySpaceIE)
19 (?:https?://www\.vevo\.com/watch/(?:[^/]+/(?:[^/]+/)?)?|
20 https?://cache\.vevo\.com/m/html/embed\.html\?video=|
21 https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
26 'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
27 "md5": "95ee28ee45e70130e3ab02b0f579ae23",
31 "upload_date": "20130624",
33 "title": "Somebody to Die For",
37 # timestamp and upload_date are often incorrect; seem to change randomly
41 'note': 'v3 SMIL format',
42 'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
43 'md5': 'f6ab09b034f8c22969020b042e5ac7fc',
47 'upload_date': '20140219',
48 'uploader': 'Cassadee Pope',
49 'title': 'I Wish I Could Break Your Heart',
55 'note': 'Age-limited video',
56 'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
61 'title': 'Tunnel Vision (Explicit)',
62 'uploader': 'Justin Timberlake',
63 'upload_date': 're:2013070[34]',
67 'skip_download': 'true',
70 _SMIL_BASE_URL
= 'http://smil.lvl3.vevo.com/'
72 def _real_initialize(self
):
73 req
= compat_urllib_request
.Request(
74 'http://www.vevo.com/auth', data
=b
'')
75 webpage
= self
._download
_webpage
(
77 note
='Retrieving oauth token',
78 errnote
='Unable to retrieve oauth token',
81 self
._oauth
_token
= None
83 self
._oauth
_token
= self
._search
_regex
(
84 r
'access_token":\s*"([^"]+)"',
85 webpage
, 'access token', fatal
=False)
87 def _formats_from_json(self
, video_info
):
88 last_version
= {'version': -1}
89 for version
in video_info
['videoVersions']:
90 # These are the HTTP downloads, other types are for different manifests
91 if version
['sourceType'] == 2:
92 if version
['version'] > last_version
['version']:
93 last_version
= version
94 if last_version
['version'] == -1:
95 raise ExtractorError('Unable to extract last version of the video')
97 renditions
= xml
.etree
.ElementTree
.fromstring(last_version
['data'])
99 # Already sorted from worst to best quality
100 for rend
in renditions
.findall('rendition'):
102 format_note
= '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr
105 'format_id': attr
['name'],
106 'format_note': format_note
,
107 'height': int(attr
['frameheight']),
108 'width': int(attr
['frameWidth']),
112 def _formats_from_smil(self
, smil_xml
):
114 smil_doc
= xml
.etree
.ElementTree
.fromstring(smil_xml
.encode('utf-8'))
115 els
= smil_doc
.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
117 src
= el
.attrib
['src']
118 m
= re
.match(r
'''(?xi)
121 [/a-z0-9]+ # The directory and main part of the URL
123 _(?P<width>[0-9]+)x(?P<height>[0-9]+)
124 _(?P<vcodec>[a-z0-9]+)
126 _(?P<acodec>[a-z0-9]+)
128 \.[a-z0-9]+ # File extension
133 format_url
= self
._SMIL
_BASE
_URL
+ m
.group('path')
136 'format_id': 'SMIL_' + m
.group('cbr'),
137 'vcodec': m
.group('vcodec'),
138 'acodec': m
.group('acodec'),
139 'vbr': int(m
.group('vbr')),
140 'abr': int(m
.group('abr')),
141 'ext': m
.group('ext'),
142 'width': int(m
.group('width')),
143 'height': int(m
.group('height')),
147 def _download_api_formats(self
, video_id
):
148 if not self
._oauth
_token
:
149 self
._downloader
.report_warning(
150 'No oauth token available, skipping API HLS download')
153 api_url
= 'https://apiv2.vevo.com/video/%s/streams/hls?token=%s' % (
154 video_id
, self
._oauth
_token
)
155 api_data
= self
._download
_json
(
157 note
='Downloading HLS formats',
158 errnote
='Failed to download HLS format list', fatal
=False)
162 m3u8_url
= api_data
[0]['url']
163 return self
._extract
_m
3u8_formats
(
164 m3u8_url
, video_id
, entry_protocol
='m3u8_native', ext
='mp4',
167 def _real_extract(self
, url
):
168 mobj
= re
.match(self
._VALID
_URL
, url
)
169 video_id
= mobj
.group('id')
171 json_url
= 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
172 response
= self
._download
_json
(json_url
, video_id
)
173 video_info
= response
['video']
176 if 'statusMessage' in response
:
177 raise ExtractorError('%s said: %s' % (self
.IE_NAME
, response
['statusMessage']), expected
=True)
178 raise ExtractorError('Unable to extract videos')
180 formats
= self
._formats
_from
_json
(video_info
)
182 is_explicit
= video_info
.get('isExplicit')
183 if is_explicit
is True:
185 elif is_explicit
is False:
190 # Download via HLS API
191 formats
.extend(self
._download
_api
_formats
(video_id
))
193 self
._sort
_formats
(formats
)
194 timestamp_ms
= int(self
._search
_regex
(
195 r
'/Date\((\d+)\)/', video_info
['launchDate'], 'launch date'))
199 'title': video_info
['title'],
201 'thumbnail': video_info
['imageUrl'],
202 'timestamp': timestamp_ms
// 1000,
203 'uploader': video_info
['mainArtists'][0]['artistName'],
204 'duration': video_info
['duration'],
205 'age_limit': age_limit
,