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

   1 # coding: 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_parse_urlencode,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     try_get,
  15 )
  16
  17
  18 class SohuIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
  20
  21     # Sohu videos give different MD5 sums on Travis CI and my machine
  22     _TESTS = [{
  23         'note': 'This video is available only in Mainland China',
  24         'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
  25         'info_dict': {
  26             'id': '382479172',
  27             'ext': 'mp4',
  28             'title': 'MV：Far East Movement《The Illest》',
  29         },
  30         'skip': 'On available in China',
  31     }, {
  32         'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
  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         '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             'info_dict': {
  54                 'id': '78910339_part1',
  55                 'ext': 'mp4',
  56                 'duration': 294,
  57                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  58             }
  59         }, {
  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             'info_dict': {
  68                 'id': '78910339_part3',
  69                 'ext': 'mp4',
  70                 'duration': 150,
  71                 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
  72             }
  73         }]
  74     }, {
  75         'note': 'Video with title containing dash',
  76         'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
  77         'info_dict': {
  78             'id': '78932792',
  79             'ext': 'mp4',
  80             'title': 'youtube-dl testing video',
  81         },
  82         'params': {
  83             'skip_download': True
  84         }
  85     }]
  86
  87     def _real_extract(self, url):
  88
  89         def _fetch_data(vid_id, mytv=False):
  90             if mytv:
  91                 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
  92             else:
  93                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
  94
  95             return self._download_json(
  96                 base_data_url + vid_id, video_id,
  97                 'Downloading JSON data for %s' % vid_id,
  98                 headers=self.geo_verification_headers())
  99
 100         mobj = re.match(self._VALID_URL, url)
 101         video_id = mobj.group('id')
 102         mytv = mobj.group('mytv') is not None
 103
 104         webpage = self._download_webpage(url, video_id)
 105
 106         title = re.sub(r' - 搜狐视频$', '', self._og_search_title(webpage))
 107
 108         vid = self._html_search_regex(
 109             r'var vid ?= ?["\'](\d+)["\']',
 110             webpage, 'video path')
 111         vid_data = _fetch_data(vid, mytv)
 112         if vid_data['play'] != 1:
 113             if vid_data.get('status') == 12:
 114                 raise ExtractorError(
 115                     '%s said: There\'s something wrong in the video.' % self.IE_NAME,
 116                     expected=True)
 117             else:
 118                 self.raise_geo_restricted(
 119                     '%s said: The video is only licensed to users in Mainland China.' % self.IE_NAME)
 120
 121         formats_json = {}
 122         for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
 123             vid_id = vid_data['data'].get('%sVid' % format_id)
 124             if not vid_id:
 125                 continue
 126             vid_id = compat_str(vid_id)
 127             formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
 128
 129         part_count = vid_data['data']['totalBlocks']
 130
 131         playlist = []
 132         for i in range(part_count):
 133             formats = []
 134             for format_id, format_data in formats_json.items():
 135                 allot = format_data['allot']
 136
 137                 data = format_data['data']
 138                 clips_url = data['clipsURL']
 139                 su = data['su']
 140
 141                 video_url = 'newflv.sohu.ccgslb.net'
 142                 cdnId = None
 143                 retries = 0
 144
 145                 while 'newflv.sohu.ccgslb.net' in video_url:
 146                     params = {
 147                         'prot': 9,
 148                         'file': clips_url[i],
 149                         'new': su[i],
 150                         'prod': 'flash',
 151                         'rb': 1,
 152                     }
 153
 154                     if cdnId is not None:
 155                         params['idc'] = cdnId
 156
 157                     download_note = 'Downloading %s video URL part %d of %d' % (
 158                         format_id, i + 1, part_count)
 159
 160                     if retries > 0:
 161                         download_note += ' (retry #%d)' % retries
 162                     part_info = self._parse_json(self._download_webpage(
 163                         'http://%s/?%s' % (allot, compat_urllib_parse_urlencode(params)),
 164                         video_id, download_note), video_id)
 165
 166                     video_url = part_info['url']
 167                     cdnId = part_info.get('nid')
 168
 169                     retries += 1
 170                     if retries > 5:
 171                         raise ExtractorError('Failed to get video URL')
 172
 173                 formats.append({
 174                     'url': video_url,
 175                     'format_id': format_id,
 176                     'filesize': int_or_none(
 177                         try_get(data, lambda x: x['clipsBytes'][i])),
 178                     'width': int_or_none(data.get('width')),
 179                     'height': int_or_none(data.get('height')),
 180                     'fps': int_or_none(data.get('fps')),
 181                 })
 182             self._sort_formats(formats)
 183
 184             playlist.append({
 185                 'id': '%s_part%d' % (video_id, i + 1),
 186                 'title': title,
 187                 'duration': vid_data['data']['clipsDuration'][i],
 188                 'formats': formats,
 189             })
 190
 191         if len(playlist) == 1:
 192             info = playlist[0]
 193             info['id'] = video_id
 194         else:
 195             info = {
 196                 '_type': 'multi_video',
 197                 'entries': playlist,
 198                 'id': video_id,
 199                 'title': title,
 200             }
 201
 202         return info