3 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)
19 (?:https?://www\.vevo\.com/watch/(?:[^/]+/[^/]+/)?|
20 https?://cache\.vevo\.com/m/html/embed\.html\?video=|
24 u'url': u'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
25 u'file': u'GB1101300280.mp4',
26 u"md5": u"06bea460acb744eab74a9d7dcb4bfd61",
28 u"upload_date": u"20130624",
29 u"uploader": u"Hurts",
30 u"title": u"Somebody to Die For",
36 _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/'
38 def _formats_from_json(self, video_info):
39 last_version = {'version': -1}
40 for version in video_info['videoVersions']:
41 # These are the HTTP downloads, other types are for different manifests
42 if version['sourceType'] == 2:
43 if version['version'] > last_version['version']:
44 last_version = version
45 if last_version['version'] == -1:
46 raise ExtractorError(u'Unable to extract last version of the video')
48 renditions = xml.etree.ElementTree.fromstring(last_version['data'])
50 # Already sorted from worst to best quality
51 for rend in renditions.findall('rendition'):
53 format_note = '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr
56 'format_id': attr['name'],
57 'format_note': format_note,
58 'height': int(attr['frameheight']),
59 'width': int(attr['frameWidth']),
63 def _formats_from_smil(self, smil_xml):
65 smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8'))
66 els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
68 src = el.attrib['src']
69 m = re.match(r'''(?xi)
72 [/a-z0-9]+ # The directory and main part of the URL
74 _(?P<width>[0-9]+)x(?P<height>[0-9]+)
75 _(?P<vcodec>[a-z0-9]+)
77 _(?P<acodec>[a-z0-9]+)
79 \.[a-z0-9]+ # File extension
84 format_url = self._SMIL_BASE_URL + m.group('path')
87 'format_id': u'SMIL_' + m.group('cbr'),
88 'vcodec': m.group('vcodec'),
89 'acodec': m.group('acodec'),
90 'vbr': int(m.group('vbr')),
91 'abr': int(m.group('abr')),
92 'ext': m.group('ext'),
93 'width': int(m.group('width')),
94 'height': int(m.group('height')),
98 def _real_extract(self, url):
99 mobj = re.match(self._VALID_URL, url)
100 video_id = mobj.group('id')
102 json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
103 info_json = self._download_webpage(json_url, video_id, u'Downloading json info')
104 video_info = json.loads(info_json)['video']
106 formats = self._formats_from_json(video_info)
108 smil_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (
109 self._SMIL_BASE_URL, video_id, video_id.lower())
110 smil_xml = self._download_webpage(smil_url, video_id,
111 u'Downloading SMIL info')
112 formats.extend(self._formats_from_smil(smil_xml))
113 except ExtractorError as ee:
114 if not isinstance(ee.cause, compat_HTTPError):
116 self._downloader.report_warning(
117 u'Cannot download SMIL information, falling back to JSON ..')
119 timestamp_ms = int(self._search_regex(
120 r'/Date\((\d+)\)/', video_info['launchDate'], u'launch date'))
121 upload_date = datetime.datetime.fromtimestamp(timestamp_ms // 1000)
124 'title': video_info['title'],
126 'thumbnail': video_info['imageUrl'],
127 'upload_date': upload_date.strftime('%Y%m%d'),
128 'uploader': video_info['mainArtists'][0]['artistName'],
129 'duration': video_info['duration'],