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