+ headers.update(self.geo_verification_headers())
+ self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
+
+ raw_data = self._download_json(req_url, video_id, note=note, headers=headers)
+
+ return raw_data['data']
+
+ video_password = self._downloader.params.get('videopassword')
+
+ # request basic data
+ basic_data_url = 'http://play.youku.com/play/get.json?vid=%s&ct=12' % video_id
+ if video_password:
+ basic_data_url += '&pwd=%s' % video_password
+
+ data = retrieve_data(basic_data_url, 'Downloading JSON metadata')
+
+ error = data.get('error')
+ if error:
+ error_note = error.get('note')
+ if error_note is not None and '因版权原因无法观看此视频' in error_note:
+ raise ExtractorError(
+ 'Youku said: Sorry, this video is available in China only', expected=True)
+ elif error_note and '该视频被设为私密' in error_note:
+ raise ExtractorError(
+ 'Youku said: Sorry, this video is private', expected=True)
+ else:
+ msg = 'Youku server reported error %i' % error.get('code')
+ if error_note is not None:
+ msg += ': ' + error_note
+ raise ExtractorError(msg)
+
+ # get video title
+ title = data['video']['title']
+
+ # generate video_urls_dict
+ video_urls_dict = self.construct_video_urls(data)
+
+ # construct info
+ entries = [{
+ 'id': '%s_part%d' % (video_id, i + 1),
+ 'title': title,
+ 'formats': [],
+ # some formats are not available for all parts, we have to detect
+ # which one has all
+ } for i in range(max(len(v.get('segs')) for v in data['stream']))]
+ for stream in data['stream']:
+ if stream.get('channel_type') == 'tail':
+ continue
+ fm = stream.get('stream_type')
+ video_urls = video_urls_dict[fm]
+ for video_url, seg, entry in zip(video_urls, stream['segs'], entries):
+ entry['formats'].append({
+ 'url': video_url,
+ 'format_id': self.get_format_name(fm),
+ 'ext': self.parse_ext_l(fm),
+ 'filesize': int(seg['size']),
+ 'width': stream.get('width'),
+ 'height': stream.get('height'),
+ })
+
+ return {
+ '_type': 'multi_video',
+ 'id': video_id,
+ 'title': title,
+ 'entries': entries,
+ }
+
+
+class YoukuShowIE(InfoExtractor):
+ _VALID_URL = r'https?://(?:www\.)?youku\.com/show_page/id_(?P<id>[0-9a-z]+)\.html'
+ IE_NAME = 'youku:show'
+
+ _TEST = {
+ 'url': 'http://www.youku.com/show_page/id_zc7c670be07ff11e48b3f.html',
+ 'info_dict': {
+ 'id': 'zc7c670be07ff11e48b3f',
+ 'title': '花千骨 未删减版',
+ 'description': 'md5:578d4f2145ae3f9128d9d4d863312910',
+ },
+ 'playlist_count': 50,
+ }
+
+ _PAGE_SIZE = 40
+
+ def _find_videos_in_page(self, webpage):
+ videos = re.findall(
+ r'<li><a[^>]+href="(?P<url>https?://v\.youku\.com/[^"]+)"[^>]+title="(?P<title>[^"]+)"', webpage)
+ return [
+ self.url_result(video_url, YoukuIE.ie_key(), title)
+ for video_url, title in videos]
+
+ def _real_extract(self, url):
+ show_id = self._match_id(url)
+ webpage = self._download_webpage(url, show_id)
+
+ entries = self._find_videos_in_page(webpage)
+
+ playlist_title = self._html_search_regex(
+ r'<span[^>]+class="name">([^<]+)</span>', webpage, 'playlist title', fatal=False)
+ detail_div = get_element_by_attribute('class', 'detail', webpage) or ''
+ playlist_description = self._html_search_regex(
+ r'<span[^>]+style="display:none"[^>]*>([^<]+)</span>',
+ detail_div, 'playlist description', fatal=False)
+
+ for idx in itertools.count(1):
+ episodes_page = self._download_webpage(
+ 'http://www.youku.com/show_episode/id_%s.html' % show_id,
+ show_id, query={'divid': 'reload_%d' % (idx * self._PAGE_SIZE + 1)},
+ note='Downloading episodes page %d' % idx)
+ new_entries = self._find_videos_in_page(episodes_page)
+ entries.extend(new_entries)
+ if len(new_entries) < self._PAGE_SIZE:
+ break
Rapsys Git / youtubedl / blobdiff