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.*?'
18 'note': 'This video is available only in Mainland China',
19 'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
20 'md5': '29175c8cadd8b5cc4055001e85d6b372',
24 'title': 'MV:Far East Movement《The Illest》',
26 'skip': 'On available in China',
28 'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
29 'md5': '699060e75cf58858dd47fb9c03c42cfb',
33 'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
36 'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
37 'md5': '9bf34be48f2f4dadcb226c74127e203c',
41 'title': '【爱范品】第31期:MWC见不到的奇葩手机',
44 'note': 'Multipart video',
45 'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml',
48 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
51 'md5': 'bdbfb8f39924725e6589c146bc1883ad',
53 'id': '78910339_part1',
56 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
59 'md5': '3e1f46aaeb95354fd10e7fca9fc1804e',
61 'id': '78910339_part2',
64 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
67 'md5': '8407e634175fdac706766481b9443450',
69 'id': '78910339_part3',
72 'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
76 'note': 'Video with title containing dash',
77 'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
81 'title': 'youtube-dl testing video',
88 def _real_extract(self, url):
90 def _fetch_data(vid_id, mytv=False):
92 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
94 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
96 return self._download_json(
97 base_data_url + vid_id, video_id,
98 'Downloading JSON data for %s' % vid_id,
99 headers=self.geo_verification_headers())
101 mobj = re.match(self._VALID_URL, url)
102 video_id = mobj.group('id')
103 mytv = mobj.group('mytv') is not None
105 webpage = self._download_webpage(url, video_id)
107 title = re.sub(r' - 搜狐视频$', '', self._og_search_title(webpage))
109 vid = self._html_search_regex(
110 r'var vid ?= ?["\'](\d+)["\']',
111 webpage, 'video path')
112 vid_data = _fetch_data(vid, mytv)
113 if vid_data['play'] != 1:
114 if vid_data.get('status') == 12:
115 raise ExtractorError(
116 'Sohu said: There\'s something wrong in the video.',
119 raise ExtractorError(
120 'Sohu said: The video is only licensed to users in Mainland China.',
124 for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
125 vid_id = vid_data['data'].get('%sVid' % format_id)
128 vid_id = compat_str(vid_id)
129 formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
131 part_count = vid_data['data']['totalBlocks']
134 for i in range(part_count):
136 for format_id, format_data in formats_json.items():
137 allot = format_data['allot']
139 data = format_data['data']
140 clips_url = data['clipsURL']
143 video_url = 'newflv.sohu.ccgslb.net'
147 while 'newflv.sohu.ccgslb.net' in video_url:
150 'file': clips_url[i],
156 if cdnId is not None:
157 params['idc'] = cdnId
159 download_note = 'Downloading %s video URL part %d of %d' % (
160 format_id, i + 1, part_count)
163 download_note += ' (retry #%d)' % retries
164 part_info = self._parse_json(self._download_webpage(
165 'http://%s/?%s' % (allot, compat_urllib_parse_urlencode(params)),
166 video_id, download_note), video_id)
168 video_url = part_info['url']
169 cdnId = part_info.get('nid')
173 raise ExtractorError('Failed to get video URL')
177 'format_id': format_id,
178 'filesize': data['clipsBytes'][i],
179 'width': data['width'],
180 'height': data['height'],
183 self._sort_formats(formats)
186 'id': '%s_part%d' % (video_id, i + 1),
188 'duration': vid_data['data']['clipsDuration'][i],
192 if len(playlist) == 1:
194 info['id'] = video_id
197 '_type': 'multi_video',