Raphaël G. Git Repositories - youtubedl/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 .common import compat_str
   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         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  15         'md5': 'bde8d9a6ffd82c63a1eefaef4eeefec7',
  16         'info_dict': {
  17             'id': '382479172',
  18             'ext': 'mp4',
  19             'title': 'MV：Far East Movement《The Illest》',
  20         },
  21         'skip': 'Only available from China',
  22     }
  23
  24     def _real_extract(self, url):
  25
  26         def _fetch_data(vid_id, mytv=False):
  27             if mytv:
  28                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  29             else:
  30                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  31
  32             return self._download_json(
  33                 base_data_url + vid_id, video_id,
  34                 'Downloading JSON data for %s' % vid_id)
  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(
  42             r'(?s)<title>(.+?)</title>',
  43             webpage, 'video title')
  44         title = raw_title.partition('-')[0].strip()
  45
  46         vid = self._html_search_regex(
  47             r'var vid ?= ?["\'](\d+)["\']',
  48             webpage, 'video path')
  49         vid_data = _fetch_data(vid, mytv)
  50
  51         formats_json = {}
  52         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
  53             vid_id = vid_data['data'].get('%sVid' % format_id)
  54             if not vid_id:
  55                 continue
  56             vid_id = compat_str(vid_id)
  57             formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
  58
  59         part_count = vid_data['data']['totalBlocks']
  60
  61         playlist = []
  62         for i in range(part_count):
  63             formats = []
  64             for format_id, format_data in formats_json.items():
  65                 allot = format_data['allot']
  66                 prot = format_data['prot']
  67
  68                 data = format_data['data']
  69                 clips_url = data['clipsURL']
  70                 su = data['su']
  71
  72                 part_str = self._download_webpage(
  73                     'http://%s/?prot=%s&file=%s&new=%s' %
  74                     (allot, prot, clips_url[i], su[i]),
  75                     video_id,
  76                     'Downloading %s video URL part %d of %d'
  77                     % (format_id, i + 1, part_count))
  78
  79                 part_info = part_str.split('|')
  80                 video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
  81
  82                 formats.append({
  83                     'url': video_url,
  84                     'format_id': format_id,
  85                     'filesize': data['clipsBytes'][i],
  86                     'width': data['width'],
  87                     'height': data['height'],
  88                     'fps': data['fps'],
  89                 })
  90             self._sort_formats(formats)
  91
  92             playlist.append({
  93                 'id': '%s_part%d' % (video_id, i + 1),
  94                 'title': title,
  95                 'duration': vid_data['data']['clipsDuration'][i],
  96                 'formats': formats,
  97             })
  98
  99         if len(playlist) == 1:
 100             info = playlist[0]
 101             info['id'] = video_id
 102         else:
 103             info = {
 104                 '_type': 'playlist',
 105                 'entries': playlist,
 106                 'id': video_id,
 107             }
 108
 109         return info