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