2 from __future__ import unicode_literals
6 from .common import InfoExtractor
9 compat_urllib_parse_urlencode,
11 from ..utils import ExtractorError
14 class SohuIE(InfoExtractor):
15 _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
17 # Sohu videos give different MD5 sums on Travis CI and my machine
19 'note': 'This video is available only in Mainland China',
20 'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
24 'title': 'MV:Far East Movement《The Illest》',
26 'skip': 'On available in China',
28 'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
32 'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
35 'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
39 'title': '【爱范品】第31期:MWC见不到的奇葩手机',
42 'note': 'Multipart video',
43 'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml',
46 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
50 'id': '78910339_part1',
53 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
57 'id': '78910339_part2',
60 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
64 'id': '78910339_part3',
67 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
71 'note': 'Video with title containing dash',
72 'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
76 'title': 'youtube-dl testing video',
83 def _real_extract(self, url):
85 def _fetch_data(vid_id, mytv=False):
87 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
89 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
91 return self._download_json(
92 base_data_url + vid_id, video_id,
93 'Downloading JSON data for %s' % vid_id,
94 headers=self.geo_verification_headers())
96 mobj = re.match(self._VALID_URL, url)
97 video_id = mobj.group('id')
98 mytv = mobj.group('mytv') is not None
100 webpage = self._download_webpage(url, video_id)
102 title = re.sub(r' - 搜狐视频$', '', self._og_search_title(webpage))
104 vid = self._html_search_regex(
105 r'var vid ?= ?["\'](\d+)["\']',
106 webpage, 'video path')
107 vid_data = _fetch_data(vid, mytv)
108 if vid_data['play'] != 1:
109 if vid_data.get('status') == 12:
110 raise ExtractorError(
111 'Sohu said: There\'s something wrong in the video.',
114 raise ExtractorError(
115 'Sohu said: The video is only licensed to users in Mainland China.',
119 for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
120 vid_id = vid_data['data'].get('%sVid' % format_id)
123 vid_id = compat_str(vid_id)
124 formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
126 part_count = vid_data['data']['totalBlocks']
129 for i in range(part_count):
131 for format_id, format_data in formats_json.items():
132 allot = format_data['allot']
134 data = format_data['data']
135 clips_url = data['clipsURL']
138 video_url = 'newflv.sohu.ccgslb.net'
142 while 'newflv.sohu.ccgslb.net' in video_url:
145 'file': clips_url[i],
151 if cdnId is not None:
152 params['idc'] = cdnId
154 download_note = 'Downloading %s video URL part %d of %d' % (
155 format_id, i + 1, part_count)
158 download_note += ' (retry #%d)' % retries
159 part_info = self._parse_json(self._download_webpage(
160 'http://%s/?%s' % (allot, compat_urllib_parse_urlencode(params)),
161 video_id, download_note), video_id)
163 video_url = part_info['url']
164 cdnId = part_info.get('nid')
168 raise ExtractorError('Failed to get video URL')
172 'format_id': format_id,
173 'filesize': data['clipsBytes'][i],
174 'width': data['width'],
175 'height': data['height'],
178 self._sort_formats(formats)
181 'id': '%s_part%d' % (video_id, i + 1),
183 'duration': vid_data['data']['clipsDuration'][i],
187 if len(playlist) == 1:
189 info['id'] = video_id
192 '_type': 'multi_video',