+ items_json = self._search_regex(
+ r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
+ default=None)
+ if items_json is None:
+ CONTENT_ID_REGEXES = [
+ r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
+ r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"'
+ ]
+ long_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')
+ video_id = long_id
+ else:
+ items = json.loads(items_json)
+ info = items['mediaItems']['query']['results']['mediaObj'][0]
+ # The 'meta' field is not always in the video webpage, we request it
+ # from another page
+ long_id = info['id']
+ return self._get_info(long_id, video_id, webpage)
+
+ def _get_info(self, long_id, video_id, webpage):
+ query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
+ ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"'
+ ' AND protocol="http"' % long_id)
+ data = compat_urllib_parse.urlencode({
+ 'q': query,
+ 'env': 'prod',
+ 'format': 'json',
+ })
+ query_result = self._download_json(
+ 'http://video.query.yahoo.com/v1/public/yql?' + data,
+ video_id, 'Downloading video info')
+ info = query_result['query']['results']['mediaObj'][0]
+ meta = info['meta']
+
+ formats = []
+ for s in info['streams']:
+ format_info = {
+ 'width': int_or_none(s.get('width')),
+ 'height': int_or_none(s.get('height')),
+ 'tbr': int_or_none(s.get('bitrate')),
+ }
+
+ host = s['host']
+ path = s['path']
+ if host.startswith('rtmp'):
+ format_info.update({
+ 'url': host,
+ 'play_path': path,
+ 'ext': 'flv',
+ })
+ else:
+ format_url = compat_urlparse.urljoin(host, path)
+ format_info['url'] = format_url
+ formats.append(format_info)
+
+ self._sort_formats(formats)
+
+ return {
+ 'id': video_id,
+ 'title': meta['title'],
+ 'formats': formats,
+ 'description': clean_html(meta['description']),
+ 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
+ }
+
+
+class YahooNewsIE(YahooIE):
+ IE_NAME = 'yahoo:news'
+ _VALID_URL = r'http://news\.yahoo\.com/video/.*?-(?P<id>\d*?)\.html'
+
+ _TESTS = [{
+ 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
+ 'md5': '67010fdf3a08d290e060a4dd96baa07b',
+ 'info_dict': {
+ 'id': '104538833',
+ 'ext': 'mp4',
+ 'title': 'China Moses Is Crazy About the Blues',
+ 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
+ },
+ }]
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ video_id = mobj.group('id')
+ webpage = self._download_webpage(url, video_id)
+ long_id = self._search_regex(r'contentId: \'(.+?)\',', webpage, 'long id')
+ return self._get_info(long_id, video_id, webpage)