youtube_dl/extractor/sohu.py

   1 # encoding: utf-8
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import ExtractorError
   8
   9
  10 class SohuIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  12
  13     _TEST = {
  14         u'url': u'http://tv.sohu.com/20130724/n382479172.shtml#super',
  15         u'file': u'382479172.mp4',
  16         u'md5': u'bde8d9a6ffd82c63a1eefaef4eeefec7',
  17         u'info_dict': {
  18             u'title': u'MV：Far East Movement《The Illest》',
  19         },
  20         u'skip': u'Only available from China',
  21     }
  22
  23     def _real_extract(self, url):
  24
  25         def _fetch_data(vid_id, mytv=False):
  26             if mytv:
  27                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  28             else:
  29                 base_data_url = u'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  30             data_url = base_data_url + str(vid_id)
  31             data_json = self._download_webpage(
  32                 data_url, video_id,
  33                 note=u'Downloading JSON data for ' + str(vid_id))
  34             return json.loads(data_json)
  35
  36         mobj = re.match(self._VALID_URL, url)
  37         video_id = mobj.group('id')
  38         mytv = mobj.group('mytv') is not None
  39
  40         webpage = self._download_webpage(url, video_id)
  41         raw_title = self._html_search_regex(r'(?s)<title>(.+?)</title>',
  42                                             webpage, u'video title')
  43         title = raw_title.partition('-')[0].strip()
  44
  45         vid = self._html_search_regex(r'var vid ?= ?["\'](\d+)["\']', webpage,
  46                                       u'video path')
  47         data = _fetch_data(vid, mytv)
  48
  49         QUALITIES = ('ori', 'super', 'high', 'nor')
  50         vid_ids = [data['data'][q + 'Vid']
  51                    for q in QUALITIES
  52                    if data['data'][q + 'Vid'] != 0]
  53         if not vid_ids:
  54             raise ExtractorError(u'No formats available for this video')
  55
  56         # For now, we just pick the highest available quality
  57         vid_id = vid_ids[-1]
  58
  59         format_data = data if vid == vid_id else _fetch_data(vid_id, mytv)
  60         part_count = format_data['data']['totalBlocks']
  61         allot = format_data['allot']
  62         prot = format_data['prot']
  63         clipsURL = format_data['data']['clipsURL']
  64         su = format_data['data']['su']
  65
  66         playlist = []
  67         for i in range(part_count):
  68             part_url = ('http://%s/?prot=%s&file=%s&new=%s' %
  69                         (allot, prot, clipsURL[i], su[i]))
  70             part_str = self._download_webpage(
  71                 part_url, video_id,
  72                 note=u'Downloading part %d of %d' % (i+1, part_count))
  73
  74             part_info = part_str.split('|')
  75             video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
  76
  77             video_info = {
  78                 'id': '%s_part%02d' % (video_id, i + 1),
  79                 'title': title,
  80                 'url': video_url,
  81                 'ext': 'mp4',
  82             }
  83             playlist.append(video_info)
  84
  85         if len(playlist) == 1:
  86             info = playlist[0]
  87             info['id'] = video_id
  88         else:
  89             info = {
  90                 '_type': 'playlist',
  91                 'entries': playlist,
  92                 'id': video_id,
  93             }
  94
  95         return info