2 from __future__ import unicode_literals
11 from .common import InfoExtractor
12 from ..compat import (
13 compat_urllib_parse_urlencode,
18 get_element_by_attribute,
22 class YoukuIE(InfoExtractor):
27 http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
29 (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
34 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
36 'id': 'XMTc1ODE5Njcy_part1',
37 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
41 'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
42 'only_matching': True,
44 'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html',
46 'id': 'XODgxNjg1Mzk2',
50 'skip': 'Available in China only',
52 'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
54 'id': 'XMTI1OTczNDM5Mg',
59 'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
60 'note': 'Video protected with password',
62 'id': 'XNjA1NzA2Njgw',
63 'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
67 'videopassword': '100600',
70 # /play/get.json contains streams with "channel_type":"tail"
71 'url': 'http://v.youku.com/v_show/id_XOTUxMzg4NDMy.html',
73 'id': 'XOTUxMzg4NDMy',
74 'title': '我的世界☆明月庄主☆车震猎杀☆杀人艺术Minecraft',
79 def construct_video_urls(self, data):
85 t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
86 ls[i], ls[t] = ls[t], ls[i]
89 for i in range(len(s2)):
92 ls[x], ls[y] = ls[y], ls[x]
93 s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
97 b'becaf9be', base64.b64decode(data['security']['encrypt_string'].encode('ascii'))
98 ).decode('ascii').split('_')
101 oip = data['security']['ip']
104 for stream in data['stream']:
105 if stream.get('channel_type') == 'tail':
107 format = stream.get('stream_type')
108 fileid = stream['stream_fileid']
109 fileid_dict[format] = fileid
111 def get_fileid(format, n):
112 number = hex(int(str(n), 10))[2:].upper()
114 number = '0' + number
115 streamfileids = fileid_dict[format]
116 fileid = streamfileids[0:8] + number + streamfileids[10:]
120 def generate_ep(format, n):
121 fileid = get_fileid(format, n)
124 ('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
126 ep = base64.b64encode(ep_t).decode('ascii')
129 # generate video_urls
131 for stream in data['stream']:
132 if stream.get('channel_type') == 'tail':
134 format = stream.get('stream_type')
136 for dt in stream['segs']:
137 n = str(stream['segs'].index(dt))
140 'hd': self.get_hd(format),
147 'ep': generate_ep(format, n)
150 'http://k.youku.com/player/getFlvPath/' + \
153 '/st/' + self.parse_ext_l(format) + \
154 '/fileid/' + get_fileid(format, n) + '?' + \
155 compat_urllib_parse_urlencode(param)
156 video_urls.append(video_url)
157 video_urls_dict[format] = video_urls
159 return video_urls_dict
163 return '%d%s' % (int(time.time()), ''.join([
164 random.choice(string.ascii_letters) for i in range(3)]))
166 def get_hd(self, fm):
179 return hd_id_dict[fm]
181 def parse_ext_l(self, fm):
196 def get_format_name(self, fm):
211 def _real_extract(self, url):
212 video_id = self._match_id(url)
214 self._set_cookie('youku.com', '__ysuid', self.get_ysuid())
216 def retrieve_data(req_url, note):
220 headers.update(self.geo_verification_headers())
221 self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
223 raw_data = self._download_json(req_url, video_id, note=note, headers=headers)
225 return raw_data['data']
227 video_password = self._downloader.params.get('videopassword')
230 basic_data_url = 'http://play.youku.com/play/get.json?vid=%s&ct=12' % video_id
232 basic_data_url += '&pwd=%s' % video_password
234 data = retrieve_data(basic_data_url, 'Downloading JSON metadata')
236 error = data.get('error')
238 error_note = error.get('note')
239 if error_note is not None and '因版权原因无法观看此视频' in error_note:
240 raise ExtractorError(
241 'Youku said: Sorry, this video is available in China only', expected=True)
242 elif error_note and '该视频被设为私密' in error_note:
243 raise ExtractorError(
244 'Youku said: Sorry, this video is private', expected=True)
246 msg = 'Youku server reported error %i' % error.get('code')
247 if error_note is not None:
248 msg += ': ' + error_note
249 raise ExtractorError(msg)
252 title = data['video']['title']
254 # generate video_urls_dict
255 video_urls_dict = self.construct_video_urls(data)
259 'id': '%s_part%d' % (video_id, i + 1),
262 # some formats are not available for all parts, we have to detect
264 } for i in range(max(len(v.get('segs')) for v in data['stream']))]
265 for stream in data['stream']:
266 if stream.get('channel_type') == 'tail':
268 fm = stream.get('stream_type')
269 video_urls = video_urls_dict[fm]
270 for video_url, seg, entry in zip(video_urls, stream['segs'], entries):
271 entry['formats'].append({
273 'format_id': self.get_format_name(fm),
274 'ext': self.parse_ext_l(fm),
275 'filesize': int(seg['size']),
276 'width': stream.get('width'),
277 'height': stream.get('height'),
281 '_type': 'multi_video',
288 class YoukuShowIE(InfoExtractor):
289 _VALID_URL = r'https?://(?:www\.)?youku\.com/show_page/id_(?P<id>[0-9a-z]+)\.html'
290 IE_NAME = 'youku:show'
293 'url': 'http://www.youku.com/show_page/id_zc7c670be07ff11e48b3f.html',
295 'id': 'zc7c670be07ff11e48b3f',
297 'description': 'md5:578d4f2145ae3f9128d9d4d863312910',
299 'playlist_count': 50,
304 def _find_videos_in_page(self, webpage):
306 r'<li><a[^>]+href="(?P<url>https?://v\.youku\.com/[^"]+)"[^>]+title="(?P<title>[^"]+)"', webpage)
308 self.url_result(video_url, YoukuIE.ie_key(), title)
309 for video_url, title in videos]
311 def _real_extract(self, url):
312 show_id = self._match_id(url)
313 webpage = self._download_webpage(url, show_id)
315 entries = self._find_videos_in_page(webpage)
317 playlist_title = self._html_search_regex(
318 r'<span[^>]+class="name">([^<]+)</span>', webpage, 'playlist title', fatal=False)
319 detail_div = get_element_by_attribute('class', 'detail', webpage) or ''
320 playlist_description = self._html_search_regex(
321 r'<span[^>]+style="display:none"[^>]*>([^<]+)</span>',
322 detail_div, 'playlist description', fatal=False)
324 for idx in itertools.count(1):
325 episodes_page = self._download_webpage(
326 'http://www.youku.com/show_episode/id_%s.html' % show_id,
327 show_id, query={'divid': 'reload_%d' % (idx * self._PAGE_SIZE + 1)},
328 note='Downloading episodes page %d' % idx)
329 new_entries = self._find_videos_in_page(episodes_page)
330 entries.extend(new_entries)
331 if len(new_entries) < self._PAGE_SIZE:
334 return self.playlist_result(entries, show_id, playlist_title, playlist_description)