Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ard.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6 )
   7
   8 class ARDIE(InfoExtractor):
   9     _VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
  10     _TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>'
  11     _MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)'
  12     _TEST = {
  13         u'url': u'http://www.ardmediathek.de/das-erste/tagesschau-in-100-sek?documentId=14077640',
  14         u'file': u'14077640.mp4',
  15         u'md5': u'6ca8824255460c787376353f9e20bbd8',
  16         u'info_dict': {
  17             u"title": u"11.04.2013 09:23 Uhr - Tagesschau in 100 Sekunden"
  18         },
  19         u'skip': u'Requires rtmpdump'
  20     }
  21
  22     def _real_extract(self, url):
  23         # determine video id from url
  24         m = re.match(self._VALID_URL, url)
  25
  26         numid = re.search(r'documentId=([0-9]+)', url)
  27         if numid:
  28             video_id = numid.group(1)
  29         else:
  30             video_id = m.group('video_id')
  31
  32         # determine title and media streams from webpage
  33         html = self._download_webpage(url, video_id)
  34         title = re.search(self._TITLE, html).group('title')
  35         streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM, html)]
  36         if not streams:
  37             assert '"fsk"' in html
  38             raise ExtractorError(u'This video is only available after 8:00 pm')
  39
  40         # choose default media type and highest quality for now
  41         stream = max([s for s in streams if int(s["media_type"]) == 0],
  42                      key=lambda s: int(s["quality"]))
  43
  44         # there's two possibilities: RTMP stream or HTTP download
  45         info = {'id': video_id, 'title': title, 'ext': 'mp4'}
  46         if stream['rtmp_url']:
  47             self.to_screen(u'RTMP download detected')
  48             assert stream['video_url'].startswith('mp4:')
  49             info["url"] = stream["rtmp_url"]
  50             info["play_path"] = stream['video_url']
  51         else:
  52             assert stream["video_url"].endswith('.mp4')
  53             info["url"] = stream["video_url"]
  54         return [info]