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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import compat_str
   8 from ..compat import compat_urllib_request
   9
  10
  11 class SohuIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  13
  14     _TESTS = [{
  15         'note': 'This video is available only in Mainland China',
  16         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  17         'md5': '29175c8cadd8b5cc4055001e85d6b372',
  18         'info_dict': {
  19             'id': '382479172',
  20             'ext': 'mp4',
  21             'title': 'MV：Far East Movement《The Illest》',
  22         },
  23         'params': {
  24             'cn_verification_proxy': 'proxy.uku.im:8888'
  25         }
  26     }, {
  27         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
  28         'md5': '699060e75cf58858dd47fb9c03c42cfb',
  29         'info_dict': {
  30             'id': '409385080',
  31             'ext': 'mp4',
  32             'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
  33         }
  34     }, {
  35         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
  36         'md5': '9bf34be48f2f4dadcb226c74127e203c',
  37         'info_dict': {
  38             'id': '78693464',
  39             'ext': 'mp4',
  40             'title': '【爱范品】第31期：MWC见不到的奇葩手机',
  41         }
  42     }]
  43
  44     def _real_extract(self, url):
  45
  46         def _fetch_data(vid_id, mytv=False):
  47             if mytv:
  48                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  49             else:
  50                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  51
  52             req = compat_urllib_request.Request(base_data_url + vid_id)
  53
  54             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
  55             if cn_verification_proxy:
  56                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
  57
  58             return self._download_json(req, video_id,
  59                                        'Downloading JSON data for %s' % vid_id)
  60
  61         mobj = re.match(self._VALID_URL, url)
  62         video_id = mobj.group('id')
  63         mytv = mobj.group('mytv') is not None
  64
  65         webpage = self._download_webpage(url, video_id)
  66         raw_title = self._html_search_regex(
  67             r'(?s)<title>(.+?)</title>',
  68             webpage, 'video title')
  69         title = raw_title.partition('-')[0].strip()
  70
  71         vid = self._html_search_regex(
  72             r'var vid ?= ?["\'](\d+)["\']',
  73             webpage, 'video path')
  74         vid_data = _fetch_data(vid, mytv)
  75
  76         formats_json = {}
  77         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
  78             vid_id = vid_data['data'].get('%sVid' % format_id)
  79             if not vid_id:
  80                 continue
  81             vid_id = compat_str(vid_id)
  82             formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
  83
  84         part_count = vid_data['data']['totalBlocks']
  85
  86         playlist = []
  87         for i in range(part_count):
  88             formats = []
  89             for format_id, format_data in formats_json.items():
  90                 allot = format_data['allot']
  91                 prot = format_data['prot']
  92
  93                 data = format_data['data']
  94                 clips_url = data['clipsURL']
  95                 su = data['su']
  96
  97                 part_str = self._download_webpage(
  98                     'http://%s/?prot=%s&file=%s&new=%s' %
  99                     (allot, prot, clips_url[i], su[i]),
 100                     video_id,
 101                     'Downloading %s video URL part %d of %d'
 102                     % (format_id, i + 1, part_count))
 103
 104                 part_info = part_str.split('|')
 105
 106                 # Sanitize URL to prevent download failure
 107                 if part_info[0][-1] == '/' and su[i][0] == '/':
 108                     su[i] = su[i][1:]
 109
 110                 video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
 111
 112                 formats.append({
 113                     'url': video_url,
 114                     'format_id': format_id,
 115                     'filesize': data['clipsBytes'][i],
 116                     'width': data['width'],
 117                     'height': data['height'],
 118                     'fps': data['fps'],
 119                 })
 120             self._sort_formats(formats)
 121
 122             playlist.append({
 123                 'id': '%s_part%d' % (video_id, i + 1),
 124                 'title': title,
 125                 'duration': vid_data['data']['clipsDuration'][i],
 126                 'formats': formats,
 127             })
 128
 129         if len(playlist) == 1:
 130             info = playlist[0]
 131             info['id'] = video_id
 132         else:
 133             info = {
 134                 '_type': 'playlist',
 135                 'entries': playlist,
 136                 'id': video_id,
 137             }
 138
 139         return info