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     compat_urllib_parse,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14 )
  15
  16
  17 class SohuIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  19
  20     _TESTS = [{
  21         'note': 'This video is available only in Mainland China',
  22         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  23         'md5': '29175c8cadd8b5cc4055001e85d6b372',
  24         'info_dict': {
  25             'id': '382479172',
  26             'ext': 'mp4',
  27             'title': 'MV：Far East Movement《The Illest》',
  28         },
  29         'skip': 'On available in China',
  30     }, {
  31         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
  32         'md5': '699060e75cf58858dd47fb9c03c42cfb',
  33         'info_dict': {
  34             'id': '409385080',
  35             'ext': 'mp4',
  36             'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
  37         }
  38     }, {
  39         'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
  40         'md5': '9bf34be48f2f4dadcb226c74127e203c',
  41         'info_dict': {
  42             'id': '78693464',
  43             'ext': 'mp4',
  44             'title': '【爱范品】第31期：MWC见不到的奇葩手机',
  45         }
  46     }, {
  47         'note': 'Multipart video',
  48         'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml',
  49         'info_dict': {
  50             'id': '78910339',
  51             'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  52         },
  53         'playlist': [{
  54             'md5': 'bdbfb8f39924725e6589c146bc1883ad',
  55             'info_dict': {
  56                 'id': '78910339_part1',
  57                 'ext': 'mp4',
  58                 'duration': 294,
  59                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  60             }
  61         }, {
  62             'md5': '3e1f46aaeb95354fd10e7fca9fc1804e',
  63             'info_dict': {
  64                 'id': '78910339_part2',
  65                 'ext': 'mp4',
  66                 'duration': 300,
  67                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  68             }
  69         }, {
  70             'md5': '8407e634175fdac706766481b9443450',
  71             'info_dict': {
  72                 'id': '78910339_part3',
  73                 'ext': 'mp4',
  74                 'duration': 150,
  75                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  76             }
  77         }]
  78     }, {
  79         'note': 'Video with title containing dash',
  80         'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
  81         'info_dict': {
  82             'id': '78932792',
  83             'ext': 'mp4',
  84             'title': 'youtube-dl testing video',
  85         },
  86         'params': {
  87             'skip_download': True
  88         }
  89     }]
  90
  91     def _real_extract(self, url):
  92
  93         def _fetch_data(vid_id, mytv=False):
  94             if mytv:
  95                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  96             else:
  97                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  98
  99             req = compat_urllib_request.Request(base_data_url + vid_id)
 100
 101             cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
 102             if cn_verification_proxy:
 103                 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 104
 105             return self._download_json(
 106                 req, video_id,
 107                 'Downloading JSON data for %s' % vid_id)
 108
 109         mobj = re.match(self._VALID_URL, url)
 110         video_id = mobj.group('id')
 111         mytv = mobj.group('mytv') is not None
 112
 113         webpage = self._download_webpage(url, video_id)
 114
 115         title = re.sub(r' - 搜狐视频$', '', self._og_search_title(webpage))
 116
 117         vid = self._html_search_regex(
 118             r'var vid ?= ?["\'](\d+)["\']',
 119             webpage, 'video path')
 120         vid_data = _fetch_data(vid, mytv)
 121         if vid_data['play'] != 1:
 122             if vid_data.get('status') == 12:
 123                 raise ExtractorError(
 124                     'Sohu said: There\'s something wrong in the video.',
 125                     expected=True)
 126             else:
 127                 raise ExtractorError(
 128                     'Sohu said: The video is only licensed to users in Mainland China.',
 129                     expected=True)
 130
 131         formats_json = {}
 132         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
 133             vid_id = vid_data['data'].get('%sVid' % format_id)
 134             if not vid_id:
 135                 continue
 136             vid_id = compat_str(vid_id)
 137             formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
 138
 139         part_count = vid_data['data']['totalBlocks']
 140
 141         playlist = []
 142         for i in range(part_count):
 143             formats = []
 144             for format_id, format_data in formats_json.items():
 145                 allot = format_data['allot']
 146
 147                 data = format_data['data']
 148                 clips_url = data['clipsURL']
 149                 su = data['su']
 150
 151                 video_url = 'newflv.sohu.ccgslb.net'
 152                 cdnId = None
 153                 retries = 0
 154
 155                 while 'newflv.sohu.ccgslb.net' in video_url:
 156                     params = {
 157                         'prot': 9,
 158                         'file': clips_url[i],
 159                         'new': su[i],
 160                         'prod': 'flash',
 161                     }
 162
 163                     if cdnId is not None:
 164                         params['idc'] = cdnId
 165
 166                     download_note = 'Downloading %s video URL part %d of %d' % (
 167                         format_id, i + 1, part_count)
 168
 169                     if retries > 0:
 170                         download_note += ' (retry #%d)' % retries
 171                     part_info = self._parse_json(self._download_webpage(
 172                         'http://%s/?%s' % (allot, compat_urllib_parse.urlencode(params)),
 173                         video_id, download_note), video_id)
 174
 175                     video_url = part_info['url']
 176                     cdnId = part_info.get('nid')
 177
 178                     retries += 1
 179                     if retries > 5:
 180                         raise ExtractorError('Failed to get video URL')
 181
 182                 formats.append({
 183                     'url': video_url,
 184                     'format_id': format_id,
 185                     'filesize': data['clipsBytes'][i],
 186                     'width': data['width'],
 187                     'height': data['height'],
 188                     'fps': data['fps'],
 189                 })
 190             self._sort_formats(formats)
 191
 192             playlist.append({
 193                 'id': '%s_part%d' % (video_id, i + 1),
 194                 'title': title,
 195                 'duration': vid_data['data']['clipsDuration'][i],
 196                 'formats': formats,
 197             })
 198
 199         if len(playlist) == 1:
 200             info = playlist[0]
 201             info['id'] = video_id
 202         else:
 203             info = {
 204                 '_type': 'multi_video',
 205                 'entries': playlist,
 206                 'id': video_id,
 207                 'title': title,
 208             }
 209
 210         return info