class BloombergIE(InfoExtractor):
- _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?)\.html'
+ _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<id>.+?)\.html'
_TEST = {
'url': 'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
- 'md5': '7bf08858ff7c203c870e8a6190e221e5',
+ # The md5 checksum changes
'info_dict': {
'id': 'qurhIVlJSB6hzkVi229d8g',
'ext': 'flv',
}
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- name = mobj.group('name')
+ name = self._match_id(url)
webpage = self._download_webpage(url, name)
+
f4m_url = self._search_regex(
r'<source src="(https?://[^"]+\.f4m.*?)"', webpage,
'f4m url')
return {
'id': name.split('-')[-1],
'title': title,
- 'url': f4m_url,
- 'ext': 'flv',
+ 'formats': self._extract_f4m_formats(f4m_url, name),
'description': self._og_search_description(webpage),
'thumbnail': self._og_search_thumbnail(webpage),
}