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

   1 # encoding: utf-8
   2
   3 import re
   4 import json
   5 import time
   6 import logging
   7 import urllib2
   8
   9 from .common import InfoExtractor
  10 from ..utils import compat_urllib_request, clean_html
  11
  12
  13 class SohuIE(InfoExtractor):
  14     _VALID_URL = r'https?://tv\.sohu\.com/\d+?/n(?P<id>\d+)\.shtml.*?'
  15
  16     _TEST = {
  17         u'url': u'http://tv.sohu.com/20130724/n382479172.shtml#super',
  18         u'file': u'382479172.flv',
  19         u'md5': u'cc84eed6b6fbf0f2f9a8d3cb9da1939b',
  20         u'info_dict': {
  21             u'title': u'The Illest - Far East Movement Riff Raff',
  22         },
  23     }
  24
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group('id')
  29         webpage = self._download_webpage(url, video_id)
  30         pattern = r'<title>(.+?)</title>'
  31         compiled = re.compile(pattern, re.DOTALL)
  32         title = self._search_regex(compiled, webpage, u'video title')
  33         title = clean_html(title).split('-')[0].strip()
  34         pattern = re.compile(r'var vid="(\d+)"')
  35         result = re.search(pattern, webpage)
  36         if not result:
  37             logging.info('[Sohu] could not get vid')
  38             return None
  39         vid = result.group(1)
  40         logging.info('vid: %s' % vid)
  41         base_url_1 = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  42         url_1 = base_url_1 + vid
  43         logging.info('json url: %s' % url_1)
  44         webpage = self._download_webpage(url_1, vid)
  45         json_1 = json.loads(webpage)
  46         # get the highest definition video vid and json infomation.
  47         vids = []
  48         qualities = ('oriVid', 'superVid', 'highVid', 'norVid')
  49         for vid_name in qualities:
  50             vids.append(json_1['data'][vid_name])
  51         clearest_vid = 0
  52         for i, v in enumerate(vids):
  53             if v != 0:
  54                 clearest_vid = v
  55                 logging.info('quality definition: %s' % qualities[i][:-3])
  56                 break
  57         if not clearest_vid:
  58             logging.warning('could not find valid clearest_vid')
  59             return None
  60         if vid != clearest_vid:
  61             url_1 = '%s%d' % (base_url_1, clearest_vid)
  62             logging.info('highest definition json url: %s' % url_1)
  63             json_1 = json.loads(urllib2.urlopen(url_1).read())
  64         allot = json_1['allot']
  65         prot = json_1['prot']
  66         clipsURL = json_1['data']['clipsURL']
  67         su = json_1['data']['su']
  68         num_of_parts = json_1['data']['totalBlocks']
  69         logging.info('Total parts: %d' % num_of_parts)
  70         base_url_3 = 'http://allot/?prot=prot&file=clipsURL[i]&new=su[i]'
  71         files_info = []
  72         for i in range(num_of_parts):
  73             middle_url = 'http://%s/?prot=%s&file=%s&new=%s' % (allot, prot, clipsURL[i], su[i])
  74             logging.info('middle url part %d: %s' % (i, middle_url))
  75             middle_info = urllib2.urlopen(middle_url).read().split('|')
  76             middle_part_1 = middle_info[0]
  77             download_url = '%s%s?key=%s' % (middle_info[0], su[i], middle_info[3])
  78
  79             info = {
  80                 'id': '%s_part%02d' % (video_id, i + 1),
  81                 'title': title,
  82                 'url': download_url,
  83                 'ext': 'mp4',
  84             }
  85             files_info.append(info)
  86             time.sleep(1)
  87         if num_of_parts == 1:
  88             info =  files_info[0]
  89             info['id'] = video_id
  90             return info
  91         return files_info