aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorremitamine <remitamine@gmail.com>2016-04-29 19:21:17 +0100
committerremitamine <remitamine@gmail.com>2016-04-29 19:21:17 +0100
commit65a3bfb379c9d5e53cac874af097d2071ee4ac4d (patch)
tree194bfcac5ebf476c864ec91811e7e037e84262b5
parentcef3f3011f9d3a67de3ff064a5185a1a4bcf40e7 (diff)
[dfb] extract m3u8 formats
-rw-r--r--youtube_dl/extractor/dfb.py33
1 files changed, 20 insertions, 13 deletions
diff --git a/youtube_dl/extractor/dfb.py b/youtube_dl/extractor/dfb.py
index cdfeccacb..a4d0448c2 100644
--- a/youtube_dl/extractor/dfb.py
+++ b/youtube_dl/extractor/dfb.py
@@ -12,39 +12,46 @@ class DFBIE(InfoExtractor):
_TEST = {
'url': 'http://tv.dfb.de/video/u-19-em-stimmen-zum-spiel-gegen-russland/11633/',
- # The md5 is different each time
+ 'md5': 'ac0f98a52a330f700b4b3034ad240649',
'info_dict': {
'id': '11633',
'display_id': 'u-19-em-stimmen-zum-spiel-gegen-russland',
- 'ext': 'flv',
+ 'ext': 'mp4',
'title': 'U 19-EM: Stimmen zum Spiel gegen Russland',
'upload_date': '20150714',
},
}
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- video_id = mobj.group('id')
- display_id = mobj.group('display_id')
+ display_id, video_id = re.match(self._VALID_URL, url).groups()
- webpage = self._download_webpage(url, display_id)
player_info = self._download_xml(
'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id,
display_id)
video_info = player_info.find('video')
-
- f4m_info = self._download_xml(
- self._proto_relative_url(video_info.find('url').text.strip()), display_id)
- token_el = f4m_info.find('token')
- manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0'
- formats = self._extract_f4m_formats(manifest_url, display_id)
+ stream_access_url = self._proto_relative_url(video_info.find('url').text.strip())
+
+ formats = []
+ # see http://tv.dfb.de/player/js/ajax.js for the method to extract m3u8 formats
+ for sa_url in (stream_access_url, stream_access_url + '&area=&format=iphone'):
+ stream_access_info = self._download_xml(sa_url, display_id)
+ token_el = stream_access_info.find('token')
+ manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth']
+ if '.f4m' in manifest_url:
+ formats.extend(self._extract_f4m_formats(
+ manifest_url + '&hdcore=3.2.0',
+ display_id, f4m_id='hds', fatal=False))
+ else:
+ formats.extend(self._extract_m3u8_formats(
+ manifest_url, display_id, 'mp4',
+ 'm3u8_native', m3u8_id='hls', fatal=False))
self._sort_formats(formats)
return {
'id': video_id,
'display_id': display_id,
'title': video_info.find('title').text,
- 'thumbnail': self._og_search_thumbnail(webpage),
+ 'thumbnail': 'http://tv.dfb.de/images/%s_640x360.jpg' % video_id,
'upload_date': unified_strdate(video_info.find('time_date').text),
'formats': formats,
}