_ 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 ..compat import (
   8     compat_str,
   9     compat_urllib_request
  10 )
  11
  12
  13 class SohuIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  15
  16     _TESTS = [{
  17         'note': 'This video is available only in Mainland China',
  18         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  19         'md5': '29175c8cadd8b5cc4055001e85d6b372',
  20         'info_dict': {
  21             'id': '382479172',
  22             'ext': 'mp4',
  23             'title': 'MV：Far East Movement《The Illest》',
  24         },
  25         'params': {
  26             'cn_verification_proxy': 'proxy.uku.im:8888'
  27         }
  28     }, {
  29         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
  30         'md5': '699060e75cf58858dd47fb9c03c42cfb',
  31         'info_dict': {
  32             'id': '409385080',
  33             'ext': 'mp4',
  34             'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
  35         }
  36     }, {
  37         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
  38         'md5': '9bf34be48f2f4dadcb226c74127e203c',
  39         'info_dict': {
  40             'id': '78693464',
  41             'ext': 'mp4',
  42             'title': '【爱范品】第31期：MWC见不到的奇葩手机',
  43         }
  44     }]
  45
  46     def _real_extract(self, url):
  47
  48         def _fetch_data(vid_id, mytv=False):
  49             if mytv:
  50                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  51             else:
  52                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  53
  54             req = compat_urllib_request.Request(base_data_url + vid_id)
  55
  56             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
  57             if cn_verification_proxy:
  58                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
  59
  60             return self._download_json(req, video_id,
  61                                        'Downloading JSON data for %s' % vid_id)
  62
  63         mobj = re.match(self._VALID_URL, url)
  64         video_id = mobj.group('id')
  65         mytv = mobj.group('mytv') is not None
  66
  67         webpage = self._download_webpage(url, video_id)
  68         raw_title = self._html_search_regex(
  69             r'(?s)<title>(.+?)</title>',
  70             webpage, 'video title')
  71         title = raw_title.partition('-')[0].strip()
  72
  73         vid = self._html_search_regex(
  74             r'var vid ?= ?["\'](\d+)["\']',
  75             webpage, 'video path')
  76         vid_data = _fetch_data(vid, mytv)
  77
  78         formats_json = {}
  79         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
  80             vid_id = vid_data['data'].get('%sVid' % format_id)
  81             if not vid_id:
  82                 continue
  83             vid_id = compat_str(vid_id)
  84             formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
  85
  86         part_count = vid_data['data']['totalBlocks']
  87
  88         playlist = []
  89         for i in range(part_count):
  90             formats = []
  91             for format_id, format_data in formats_json.items():
  92                 allot = format_data['allot']
  93                 prot = format_data['prot']
  94
  95                 data = format_data['data']
  96                 clips_url = data['clipsURL']
  97                 su = data['su']
  98
  99                 part_str = self._download_webpage(
 100                     'http://%s/?prot=%s&file=%s&new=%s' %
 101                     (allot, prot, clips_url[i], su[i]),
 102                     video_id,
 103                     'Downloading %s video URL part %d of %d'
 104                     % (format_id, i + 1, part_count))
 105
 106                 part_info = part_str.split('|')
 107
 108                 # Sanitize URL to prevent download failure
 109                 if part_info[0][-1] == '/' and su[i][0] == '/':
 110                     su[i] = su[i][1:]
 111
 112                 video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
 113
 114                 formats.append({
 115                     'url': video_url,
 116                     'format_id': format_id,
 117                     'filesize': data['clipsBytes'][i],
 118                     'width': data['width'],
 119                     'height': data['height'],
 120                     'fps': data['fps'],
 121                 })
 122             self._sort_formats(formats)
 123
 124             playlist.append({
 125                 'id': '%s_part%d' % (video_id, i + 1),
 126                 'title': title,
 127                 'duration': vid_data['data']['clipsDuration'][i],
 128                 'formats': formats,
 129             })
 130
 131         if len(playlist) == 1:
 132             info = playlist[0]
 133             info['id'] = video_id
 134         else:
 135             info = {
 136                 '_type': 'playlist',
 137                 'entries': playlist,
 138                 'id': video_id,
 139             }
 140
 141         return info