Raphaël G. Git Repositories - youtubedl/blob - 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_urllib_request,
   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 and MySpaceIE)
  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
  25     _TESTS = [{
  26         'url': 'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
  27         "md5": "95ee28ee45e70130e3ab02b0f579ae23",
  28         'info_dict': {
  29             'id': 'GB1101300280',
  30             'ext': 'mp4',
  31             "upload_date": "20130624",
  32             "uploader": "Hurts",
  33             "title": "Somebody to Die For",
  34             "duration": 230.12,
  35             "width": 1920,
  36             "height": 1080,
  37             # timestamp and upload_date are often incorrect; seem to change randomly
  38             'timestamp': int,
  39         }
  40     }, {
  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',
  44         'info_dict': {
  45             'id': 'USUV71302923',
  46             'ext': 'mp4',
  47             'upload_date': '20140219',
  48             'uploader': 'Cassadee Pope',
  49             'title': 'I Wish I Could Break Your Heart',
  50             'duration': 226.101,
  51             'age_limit': 0,
  52             'timestamp': int,
  53         }
  54     }, {
  55         'note': 'Age-limited video',
  56         'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282',
  57         'info_dict': {
  58             'id': 'USRV81300282',
  59             'ext': 'mp4',
  60             'age_limit': 18,
  61             'title': 'Tunnel Vision (Explicit)',
  62             'uploader': 'Justin Timberlake',
  63             'upload_date': 're:2013070[34]',
  64             'timestamp': int,
  65         },
  66         'params': {
  67             'skip_download': 'true',
  68         }
  69     }]
  70     _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com/'
  71
  72     def _real_initialize(self):
  73         req = compat_urllib_request.Request(
  74             'http://www.vevo.com/auth', data=b'')
  75         webpage = self._download_webpage(
  76             req, None,
  77             note='Retrieving oauth token',
  78             errnote='Unable to retrieve oauth token',
  79             fatal=False)
  80         if webpage is False:
  81             self._oauth_token = None
  82         else:
  83             self._oauth_token = self._search_regex(
  84                 r'access_token":\s*"([^"]+)"',
  85                 webpage, 'access token', fatal=False)
  86
  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')
  96
  97         renditions = xml.etree.ElementTree.fromstring(last_version['data'])
  98         formats = []
  99         # Already sorted from worst to best quality
 100         for rend in renditions.findall('rendition'):
 101             attr = rend.attrib
 102             format_note = '%(videoCodec)s@%(videoBitrate)4sk, %(audioCodec)s@%(audioBitrate)3sk' % attr
 103             formats.append({
 104                 'url': attr['url'],
 105                 'format_id': attr['name'],
 106                 'format_note': format_note,
 107                 'height': int(attr['frameheight']),
 108                 'width': int(attr['frameWidth']),
 109             })
 110         return formats
 111
 112     def _formats_from_smil(self, smil_xml):
 113         formats = []
 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')
 116         for el in els:
 117             src = el.attrib['src']
 118             m = re.match(r'''(?xi)
 119                 (?P<ext>[a-z0-9]+):
 120                 (?P<path>
 121                     [/a-z0-9]+     # The directory and main part of the URL
 122                     _(?P<cbr>[0-9]+)k
 123                     _(?P<width>[0-9]+)x(?P<height>[0-9]+)
 124                     _(?P<vcodec>[a-z0-9]+)
 125                     _(?P<vbr>[0-9]+)
 126                     _(?P<acodec>[a-z0-9]+)
 127                     _(?P<abr>[0-9]+)
 128                     \.[a-z0-9]+  # File extension
 129                 )''', src)
 130             if not m:
 131                 continue
 132
 133             format_url = self._SMIL_BASE_URL + m.group('path')
 134             formats.append({
 135                 'url': format_url,
 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')),
 144             })
 145         return formats
 146
 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')
 151             return []
 152
 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(
 156             api_url, video_id,
 157             note='Downloading HLS formats',
 158             errnote='Failed to download HLS format list', fatal=False)
 159         if api_data is None:
 160             return []
 161
 162         m3u8_url = api_data[0]['url']
 163         return self._extract_m3u8_formats(
 164             m3u8_url, video_id, entry_protocol='m3u8_native', ext='mp4',
 165             preference=0)
 166
 167     def _real_extract(self, url):
 168         mobj = re.match(self._VALID_URL, url)
 169         video_id = mobj.group('id')
 170
 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']
 174
 175         if not video_info:
 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')
 179
 180         formats = self._formats_from_json(video_info)
 181
 182         is_explicit = video_info.get('isExplicit')
 183         if is_explicit is True:
 184             age_limit = 18
 185         elif is_explicit is False:
 186             age_limit = 0
 187         else:
 188             age_limit = None
 189
 190         # Download via HLS API
 191         formats.extend(self._download_api_formats(video_id))
 192
 193         self._sort_formats(formats)
 194         timestamp_ms = int(self._search_regex(
 195             r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date'))
 196
 197         return {
 198             'id': video_id,
 199             'title': video_info['title'],
 200             'formats': formats,
 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,
 206         }