_ Git - youtube-dl/blob - youtube_dl/extractor/vevo.py

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