2 from __future__ import unicode_literals
6 from .common import InfoExtractor
11 from ..utils import url_sanitize_consecutive_slashes
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》',
27 'cn_verification_proxy': 'proxy.uku.im:8888'
30 'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
31 'md5': '699060e75cf58858dd47fb9c03c42cfb',
35 'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
38 'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
39 'md5': '9bf34be48f2f4dadcb226c74127e203c',
43 'title': '【爱范品】第31期:MWC见不到的奇葩手机',
47 def _real_extract(self, url):
49 def _fetch_data(vid_id, mytv=False):
51 base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
53 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
55 req = compat_urllib_request.Request(base_data_url + vid_id)
57 cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
58 if cn_verification_proxy:
59 req.add_header('Ytdl-request-proxy', cn_verification_proxy)
61 return self._download_json(req, video_id,
62 'Downloading JSON data for %s' % vid_id)
64 mobj = re.match(self._VALID_URL, url)
65 video_id = mobj.group('id')
66 mytv = mobj.group('mytv') is not None
68 webpage = self._download_webpage(url, video_id)
69 raw_title = self._html_search_regex(
70 r'(?s)<title>(.+?)</title>',
71 webpage, 'video title')
72 title = raw_title.partition('-')[0].strip()
74 vid = self._html_search_regex(
75 r'var vid ?= ?["\'](\d+)["\']',
76 webpage, 'video path')
77 vid_data = _fetch_data(vid, mytv)
80 for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
81 vid_id = vid_data['data'].get('%sVid' % format_id)
84 vid_id = compat_str(vid_id)
85 formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
87 part_count = vid_data['data']['totalBlocks']
90 for i in range(part_count):
92 for format_id, format_data in formats_json.items():
93 allot = format_data['allot']
94 prot = format_data['prot']
96 data = format_data['data']
97 clips_url = data['clipsURL']
100 part_str = self._download_webpage(
101 'http://%s/?prot=%s&file=%s&new=%s' %
102 (allot, prot, clips_url[i], su[i]),
104 'Downloading %s video URL part %d of %d'
105 % (format_id, i + 1, part_count))
107 part_info = part_str.split('|')
109 video_url = url_sanitize_consecutive_slashes(
110 '%s%s?key=%s' % (part_info[0], su[i], part_info[3]))
114 'format_id': format_id,
115 'filesize': data['clipsBytes'][i],
116 'width': data['width'],
117 'height': data['height'],
120 self._sort_formats(formats)
123 'id': '%s_part%d' % (video_id, i + 1),
125 'duration': vid_data['data']['clipsDuration'][i],
129 if len(playlist) == 1:
131 info['id'] = video_id