Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/sohu.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import ExtractorError
   9
  10
  11 class SohuIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  13
  14     _TEST = {
  15         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  16         'md5': 'bde8d9a6ffd82c63a1eefaef4eeefec7',
  17         'info_dict': {
  18             'id': '382479172',
  19             'ext': 'mp4',
  20             'title': 'MV：Far East Movement《The Illest》',
  21         },
  22         'skip': 'Only available from China',
  23     }
  24
  25     def _real_extract(self, url):
  26
  27         def _fetch_data(vid_id, mytv=False):
  28             if mytv:
  29                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  30             else:
  31                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  32             data_url = base_data_url + str(vid_id)
  33             data_json = self._download_webpage(
  34                 data_url, video_id,
  35                 note='Downloading JSON data for ' + str(vid_id))
  36             return json.loads(data_json)
  37
  38         mobj = re.match(self._VALID_URL, url)
  39         video_id = mobj.group('id')
  40         mytv = mobj.group('mytv') is not None
  41
  42         webpage = self._download_webpage(url, video_id)
  43         raw_title = self._html_search_regex(r'(?s)<title>(.+?)</title>',
  44                                             webpage, 'video title')
  45         title = raw_title.partition('-')[0].strip()
  46
  47         vid = self._html_search_regex(r'var vid ?= ?["\'](\d+)["\']', webpage,
  48                                       'video path')
  49         data = _fetch_data(vid, mytv)
  50
  51         QUALITIES = ('ori', 'super', 'high', 'nor')
  52         vid_ids = [data['data'][q + 'Vid']
  53                    for q in QUALITIES
  54                    if data['data'][q + 'Vid'] != 0]
  55         if not vid_ids:
  56             raise ExtractorError('No formats available for this video')
  57
  58         # For now, we just pick the highest available quality
  59         vid_id = vid_ids[-1]
  60
  61         format_data = data if vid == vid_id else _fetch_data(vid_id, mytv)
  62         part_count = format_data['data']['totalBlocks']
  63         allot = format_data['allot']
  64         prot = format_data['prot']
  65         clipsURL = format_data['data']['clipsURL']
  66         su = format_data['data']['su']
  67
  68         playlist = []
  69         for i in range(part_count):
  70             part_url = ('http://%s/?prot=%s&file=%s&new=%s' %
  71                         (allot, prot, clipsURL[i], su[i]))
  72             part_str = self._download_webpage(
  73                 part_url, video_id,
  74                 note='Downloading part %d of %d' % (i + 1, part_count))
  75
  76             part_info = part_str.split('|')
  77             video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
  78
  79             video_info = {
  80                 'id': '%s_part%02d' % (video_id, i + 1),
  81                 'title': title,
  82                 'url': video_url,
  83                 'ext': 'mp4',
  84             }
  85             playlist.append(video_info)
  86
  87         if len(playlist) == 1:
  88             info = playlist[0]
  89             info['id'] = video_id
  90         else:
  91             info = {
  92                 '_type': 'playlist',
  93                 'entries': playlist,
  94                 'id': video_id,
  95             }
  96
  97         return info