Browse Source

added ARD InfoExtractor (german state television)

master
Michael Walter 11 years ago
parent
commit
df2dedeefb
1 changed files with 37 additions and 0 deletions
  1. 37
      youtube_dl/InfoExtractors.py

37
youtube_dl/InfoExtractors.py

@ -4356,6 +4356,42 @@ class LiveLeakIE(InfoExtractor):
return [info]
class ARDIE(InfoExtractor):
IE_NAME = 'ard'
_VALID_URL = r'^(?:http?://)?mediathek\.daserste\.de/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
_TITLE = r'<h1 class="boxTopHeadline">(?P<title>.*)</h1>'
_MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)'
def _real_extract(self, url):
# determine video id from url
m = re.match(self._VALID_URL, url)
video_id = m.group('video_id')
# determine title and media streams from webpage
html = self._download_webpage(url, video_id)
title = re.search(self._TITLE, html).group('title')
streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
if not streams:
assert '"fsk"' in html
self._downloader.report_error(u'this video is only available after 8:00 pm')
return
# choose default media type and highest quality for now
stream = max([s for s in streams if int(s["media_type"]) == 0], key=lambda s: int(s["quality"]))
#stream = streams[-1]
# there's two possibilities: RTMP stream or HTTP download
info = {'id': video_id, 'title': title, 'ext': 'mp4'}
if stream['rtmp_url']:
self._downloader.to_screen(u'[%s] RTMP download detected' % self.IE_NAME)
assert stream['video_url'].startswith('mp4:')
info["url"] = stream["rtmp_url"]
info["play_path"] = stream['video_url']
else:
assert stream["video_url"].endswith('.mp4')
info["url"] = stream["video_url"]
return [info]
def gen_extractors():
""" Return a list of an instance of every supported extractor.
@ -4409,5 +4445,6 @@ def gen_extractors():
MySpassIE(),
SpiegelIE(),
LiveLeakIE(),
ARDIE(),
GenericIE()
]
Loading…
Cancel
Save