Raphaël G. Git Repositories - youtubedl/blob - 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     }
  21
  22     def _real_extract(self, url):
  23
  24         def _fetch_data(vid_id, mytv=False):
  25             if mytv:
  26                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  27             else:
  28                 base_data_url = u'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  29             data_url = base_data_url + str(vid_id)
  30             data_json = self._download_webpage(
  31                 data_url, video_id,
  32                 note=u'Downloading JSON data for ' + str(vid_id))
  33             return json.loads(data_json)
  34
  35         mobj = re.match(self._VALID_URL, url)
  36         video_id = mobj.group('id')
  37         mytv = mobj.group('mytv') is not None
  38
  39         webpage = self._download_webpage(url, video_id)
  40         raw_title = self._html_search_regex(r'(?s)<title>(.+?)</title>',
  41                                             webpage, u'video title')
  42         title = raw_title.partition('-')[0].strip()
  43
  44         vid = self._html_search_regex(r'var vid ?= ?["\'](\d+)["\']', webpage,
  45                                       u'video path')
  46         data = _fetch_data(vid, mytv)
  47
  48         QUALITIES = ('ori', 'super', 'high', 'nor')
  49         vid_ids = [data['data'][q + 'Vid']
  50                    for q in QUALITIES
  51                    if data['data'][q + 'Vid'] != 0]
  52         if not vid_ids:
  53             raise ExtractorError(u'No formats available for this video')
  54
  55         # For now, we just pick the highest available quality
  56         vid_id = vid_ids[-1]
  57
  58         format_data = data if vid == vid_id else _fetch_data(vid_id, mytv)
  59         part_count = format_data['data']['totalBlocks']
  60         allot = format_data['allot']
  61         prot = format_data['prot']
  62         clipsURL = format_data['data']['clipsURL']
  63         su = format_data['data']['su']
  64
  65         playlist = []
  66         for i in range(part_count):
  67             part_url = ('http://%s/?prot=%s&file=%s&new=%s' %
  68                         (allot, prot, clipsURL[i], su[i]))
  69             part_str = self._download_webpage(
  70                 part_url, video_id,
  71                 note=u'Downloading part %d of %d' % (i+1, part_count))
  72
  73             part_info = part_str.split('|')
  74             video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
  75
  76             video_info = {
  77                 'id': '%s_part%02d' % (video_id, i + 1),
  78                 'title': title,
  79                 'url': video_url,
  80                 'ext': 'mp4',
  81             }
  82             playlist.append(video_info)
  83
  84         if len(playlist) == 1:
  85             info = playlist[0]
  86             info['id'] = video_id
  87         else:
  88             info = {
  89                 '_type': 'playlist',
  90                 'entries': playlist,
  91                 'id': video_id,
  92             }
  93
  94         return info