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 ..compat import (
   8     compat_str,
   9     compat_urllib_request
  10 )
  11 from ..utils import ExtractorError
  12
  13
  14 class SohuIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  16
  17     _TESTS = [{
  18         'note': 'This video is available only in Mainland China',
  19         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  20         'md5': '29175c8cadd8b5cc4055001e85d6b372',
  21         'info_dict': {
  22             'id': '382479172',
  23             'ext': 'mp4',
  24             'title': 'MV：Far East Movement《The Illest》',
  25         },
  26         'params': {
  27             'cn_verification_proxy': 'proxy.uku.im:8888'
  28         }
  29     }, {
  30         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
  31         'md5': 'ac9a5d322b4bf9ae184d53e4711e4f1a',
  32         'info_dict': {
  33             'id': '409385080',
  34             'ext': 'mp4',
  35             'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
  36         }
  37     }, {
  38         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
  39         'md5': '49308ff6dafde5ece51137d04aec311e',
  40         'info_dict': {
  41             'id': '78693464',
  42             'ext': 'mp4',
  43             'title': '【爱范品】第31期：MWC见不到的奇葩手机',
  44         }
  45     }, {
  46         'note': 'Multipart video',
  47         'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml',
  48         'info_dict': {
  49             'id': '78910339',
  50             'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  51         },
  52         'playlist': [{
  53             'md5': '492923eac023ba2f13ff69617c32754a',
  54             'info_dict': {
  55                 'id': '78910339_part1',
  56                 'ext': 'mp4',
  57                 'duration': 294,
  58                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  59             }
  60         }, {
  61             'md5': 'de604848c0e8e9c4a4dde7e1347c0637',
  62             'info_dict': {
  63                 'id': '78910339_part2',
  64                 'ext': 'mp4',
  65                 'duration': 300,
  66                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  67             }
  68         }, {
  69             'md5': '93584716ee0657c0b205b8aa3d27aa13',
  70             'info_dict': {
  71                 'id': '78910339_part3',
  72                 'ext': 'mp4',
  73                 'duration': 150,
  74                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  75             }
  76         }]
  77     }, {
  78         'note': 'Video with title containing dash',
  79         'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
  80         'info_dict': {
  81             'id': '78932792',
  82             'ext': 'mp4',
  83             'title': 'youtube-dl testing video',
  84         },
  85         'params': {
  86             'skip_download': True
  87         }
  88     }]
  89
  90     def _real_extract(self, url):
  91
  92         def _fetch_data(vid_id, mytv=False):
  93             if mytv:
  94                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  95             else:
  96                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  97
  98             req = compat_urllib_request.Request(base_data_url + vid_id)
  99
 100             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 101             if cn_verification_proxy:
 102                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 103
 104             return self._download_json(
 105                 req, video_id,
 106                 'Downloading JSON data for %s' % vid_id)
 107
 108         mobj = re.match(self._VALID_URL, url)
 109         video_id = mobj.group('id')
 110         mytv = mobj.group('mytv') is not None
 111
 112         webpage = self._download_webpage(url, video_id)
 113
 114         title = re.sub(r' - 搜狐视频$', '', self._og_search_title(webpage))
 115
 116         vid = self._html_search_regex(
 117             r'var vid ?= ?["\'](\d+)["\']',
 118             webpage, 'video path')
 119         vid_data = _fetch_data(vid, mytv)
 120         if vid_data['play'] != 1:
 121             if vid_data.get('status') == 12:
 122                 raise ExtractorError(
 123                     'Sohu said: There\'s something wrong in the video.',
 124                     expected=True)
 125             else:
 126                 raise ExtractorError(
 127                     'Sohu said: The video is only licensed to users in Mainland China.',
 128                     expected=True)
 129
 130         formats_json = {}
 131         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
 132             vid_id = vid_data['data'].get('%sVid' % format_id)
 133             if not vid_id:
 134                 continue
 135             vid_id = compat_str(vid_id)
 136             formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
 137
 138         part_count = vid_data['data']['totalBlocks']
 139
 140         playlist = []
 141         for i in range(part_count):
 142             formats = []
 143             for format_id, format_data in formats_json.items():
 144                 data = format_data['data']
 145
 146                 # URLs starts with http://newflv.sohu.ccgslb.net/ is not usable
 147                 # so retry until got a working URL
 148                 video_url = 'newflv.sohu.ccgslb.net'
 149                 retries = 0
 150                 while 'newflv.sohu.ccgslb.net' in video_url and retries < 5:
 151                     download_note = 'Download information from CDN gateway for format ' + format_id
 152                     if retries > 0:
 153                         download_note += ' (retry #%d)' % retries
 154                     retries += 1
 155                     cdn_info = self._download_json(
 156                         'http://data.vod.itc.cn/cdnList?new=' + data['su'][i],
 157                         video_id, download_note)
 158                     video_url = cdn_info['url']
 159
 160                 formats.append({
 161                     'url': video_url,
 162                     'format_id': format_id,
 163                     'filesize': data['clipsBytes'][i],
 164                     'width': data['width'],
 165                     'height': data['height'],
 166                     'fps': data['fps'],
 167                 })
 168             self._sort_formats(formats)
 169
 170             playlist.append({
 171                 'id': '%s_part%d' % (video_id, i + 1),
 172                 'title': title,
 173                 'duration': vid_data['data']['clipsDuration'][i],
 174                 'formats': formats,
 175             })
 176
 177         if len(playlist) == 1:
 178             info = playlist[0]
 179             info['id'] = video_id
 180         else:
 181             info = {
 182                 '_type': 'multi_video',
 183                 'entries': playlist,
 184                 'id': video_id,
 185                 'title': title,
 186             }
 187
 188         return info