youtube_dl/extractor/vevo.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import xml.etree.ElementTree
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_HTTPError,
   9     ExtractorError,
  10 )
  11
  12
  13 class VevoIE(InfoExtractor):
  14     """
  15     Accepts urls from vevo.com or in the format 'vevo:{id}'
  16     (currently used by MTVIE)
  17     """
  18     _VALID_URL = r'''(?x)
  19         (?:https?://www\.vevo\.com/watch/(?:[^/]+/[^/]+/)?|
  20            https?://cache\.vevo\.com/m/html/embed\.html\?video=|
  21            https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
  22            vevo:)
  23         (?P<id>[^&?#]+)'''
  24     _TESTS = [{
  25         'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
  26         "md5": "06bea460acb744eab74a9d7dcb4bfd61",
  27         'info_dict': {
  28             'id': 'GB1101300280',
  29             'ext': 'mp4',
  30             "upload_date": "20130624",
  31             "uploader": "Hurts",
  32             "title": "Somebody to Die For",
  33             "duration": 230.12,
  34             "width": 1920,
  35             "height": 1080,
  36             'timestamp': 1372057200,
  37         }
  38     }, {
  39         'note': 'v3 SMIL format',
  40         'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923',
  41         'md5': '893ec0e0d4426a1d96c01de8f2bdff58',
  42         'info_dict': {
  43             'id': 'USUV71302923',
  44             'ext': 'mp4',
  45             'upload_date': '20140219',
  46             'uploader': 'Cassadee Pope',
  47             'title': 'I Wish I Could Break Your Heart',
  48             'duration': 226.101,
  49             'age_limit': 0,
  50             'timestamp': 1392796919,
  51         }
  52     }, {
  53         'note': 'Age-limited video',
  54         'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
  55         'info_dict': {
  56             'id': 'USRV81300282',
  57             'ext': 'mp4',
  58             'age_limit': 18,
  59             'title': 'Tunnel Vision (Explicit)',
  60             'uploader': 'Justin Timberlake',
  61             'upload_date': '20130704',
  62             'timestamp': 1372906800,
  63         },
  64         'params': {
  65             'skip_download': 'true',
  66         }
  67     }]
  68     _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/'
  69
  70     def _formats_from_json(self, video_info):
  71         last_version = {'version': -1}
  72         for version in video_info['videoVersions']:
  73             # These are the HTTP downloads, other types are for different manifests
  74             if version['sourceType'] == 2:
  75                 if version['version'] > last_version['version']:
  76                     last_version = version
  77         if last_version['version'] == -1:
  78             raise ExtractorError('Unable to extract last version of the video')
  79
  80         renditions = xml.etree.ElementTree.fromstring(last_version['data'])
  81         formats = []
  82         # Already sorted from worst to best quality
  83         for rend in renditions.findall('rendition'):
  84             attr = rend.attrib
  85             format_note = '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr
  86             formats.append({
  87                 'url': attr['url'],
  88                 'format_id': attr['name'],
  89                 'format_note': format_note,
  90                 'height': int(attr['frameheight']),
  91                 'width': int(attr['frameWidth']),
  92             })
  93         return formats
  94
  95     def _formats_from_smil(self, smil_xml):
  96         formats = []
  97         smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8'))
  98         els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
  99         for el in els:
 100             src = el.attrib['src']
 101             m = re.match(r'''(?xi)
 102                 (?P<ext>[a-z0-9]+):
 103                 (?P<path>
 104                     [/a-z0-9]+     # The directory and main part of the URL
 105                     _(?P<cbr>[0-9]+)k
 106                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 107                     _(?P<vcodec>[a-z0-9]+)
 108                     _(?P<vbr>[0-9]+)
 109                     _(?P<acodec>[a-z0-9]+)
 110                     _(?P<abr>[0-9]+)
 111                     \.[a-z0-9]+  # File extension
 112                 )''', src)
 113             if not m:
 114                 continue
 115
 116             format_url = self._SMIL_BASE_URL + m.group('path')
 117             formats.append({
 118                 'url': format_url,
 119                 'format_id': 'SMIL_' + m.group('cbr'),
 120                 'vcodec': m.group('vcodec'),
 121                 'acodec': m.group('acodec'),
 122                 'vbr': int(m.group('vbr')),
 123                 'abr': int(m.group('abr')),
 124                 'ext': m.group('ext'),
 125                 'width': int(m.group('width')),
 126                 'height': int(m.group('height')),
 127             })
 128         return formats
 129
 130     def _real_extract(self, url):
 131         mobj = re.match(self._VALID_URL, url)
 132         video_id = mobj.group('id')
 133
 134         json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
 135         video_info = self._download_json(json_url, video_id)['video']
 136
 137         formats = self._formats_from_json(video_info)
 138
 139         is_explicit = video_info.get('isExplicit')
 140         if is_explicit is True:
 141             age_limit = 18
 142         elif is_explicit is False:
 143             age_limit = 0
 144         else:
 145             age_limit = None
 146
 147         # Download SMIL
 148         smil_blocks = sorted((
 149             f for f in video_info['videoVersions']
 150             if f['sourceType'] == 13),
 151             key=lambda f: f['version'])
 152
 153         smil_url = '%s/Video/V2/VFILE/%s/%sr.smil' % (
 154             self._SMIL_BASE_URL, video_id, video_id.lower())
 155         if smil_blocks:
 156             smil_url_m = self._search_regex(
 157                 r'url="([^"]+)"', smil_blocks[-1]['data'], 'SMIL URL',
 158                 fatal=False)
 159             if smil_url_m is not None:
 160                 smil_url = smil_url_m
 161
 162         try:
 163             smil_xml = self._download_webpage(smil_url, video_id,
 164                                               'Downloading SMIL info')
 165             formats.extend(self._formats_from_smil(smil_xml))
 166         except ExtractorError as ee:
 167             if not isinstance(ee.cause, compat_HTTPError):
 168                 raise
 169             self._downloader.report_warning(
 170                 'Cannot download SMIL information, falling back to JSON ..')
 171
 172         timestamp_ms = int(self._search_regex(
 173             r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date'))
 174
 175         return {
 176             'id': video_id,
 177             'title': video_info['title'],
 178             'formats': formats,
 179             'thumbnail': video_info['imageUrl'],
 180             'timestamp': timestamp_ms // 1000,
 181             'uploader': video_info['mainArtists'][0]['artistName'],
 182             'duration': video_info['duration'],
 183             'age_limit': age_limit,
 184         }