youtube_dl/extractor/mdr.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6 )
   7
   8 class MDRIE(InfoExtractor):
   9     _VALID_URL = r'^(?P<domain>(?:https?://)?(?:www\.)?mdr\.de)/mediathek/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)_.*'
  10     _TITLE = r'<h2>(?P<title1>[^<]+)<span>(?P<title2>[^<]+)</span></h2>'
  11
  12     _MEDIA_XML = r'(?P<xmlurl>/mediathek/(.+)/(video|audio)([0-9]+)-avCustom.xml)'
  13     _MEDIA_STREAM_VIDEO = r'<asset>.*<frameWidth>(?P<frameWidth>[0-9]+)</frameWidth>.*<flashMediaServerApplicationURL>(?P<flashMediaServerApplicationURL>[^<]+)</flashMediaServerApplicationURL><flashMediaServerURL>(?P<flashMediaServerURL>[^<]+)</flashMediaServerURL>.*<progressiveDownloadUrl>(?P<progressiveDownloadUrl>[^<]+)</progressiveDownloadUrl></asset>'
  14     _MEDIA_STREAM_AUDIO = r'<asset>.*<mediaType>(?P<mediaType>[A-Z0-9]+)</mediaType><bitrateAudio>(?P<bitrateAudio>[0-9]+)</bitrateAudio>.*<flashMediaServerApplicationURL>(?P<flashMediaServerApplicationURL>[^<]+)</flashMediaServerApplicationURL><flashMediaServerURL>(?P<flashMediaServerURL>[^<]+)</flashMediaServerURL>.*<progressiveDownloadUrl>(?P<progressiveDownloadUrl>[^<]+)</progressiveDownloadUrl></asset>'
  15     _TESTS = [{
  16         u'url': u'http://www.mdr.de/mediathek/themen/nachrichten/video165624_zc-c5c7de76_zs-3795826d.html',
  17         u'file': u'165624.mp4',
  18         u'md5': u'95165945756198b8fa2dea10f0b04614',
  19         u'info_dict': {
  20             u"title": u"MDR aktuell Eins30 09.12.2013, 22:48 Uhr"
  21         },
  22         #u'skip': u'Requires rtmpdump'    # rtmp is optional
  23     },
  24     {
  25         u'url': u' http://www.mdr.de/mediathek/radio/mdr1-radio-sachsen/audio718370_zc-67b21197_zs-1b9b2483.html',
  26         u'file': u'718370.mp4',
  27         u'md5': u'4a5b1fbb5519fb0d929c384b6ff7cb8b',
  28         u'info_dict': {
  29             u"title": u"MDR 1 RADIO SACHSEN 10.12.2013, 05:00 Uhr"
  30         },
  31         #u'skip': u'Requires rtmpdump'    # rtmp is optional
  32     }]
  33
  34     def _real_extract(self, url):
  35
  36         # determine video id from url
  37         m = re.match(self._VALID_URL, url)
  38         video_id = m.group('video_id')
  39         domain = m.group('domain')
  40         mediatype = m.group('type')
  41
  42         # determine title and media streams from webpage
  43         html = self._download_webpage(url, video_id)
  44         t = re.search(self._TITLE, html)
  45         if not t:
  46             raise ExtractorError(u'no title found')
  47         title = t.group('title1') + t.group('title2')
  48         m = re.search(self._MEDIA_XML, html)
  49         if not m:
  50             raise ExtractorError(u'no xml found')
  51         xmlurl = m.group('xmlurl')
  52         xml = self._download_webpage(domain+xmlurl, video_id, 'download XML').replace('\n','').replace('\r','').replace('<asset>','\n<asset>').replace('</asset>','</asset>\n')
  53         if(mediatype == "video"):
  54             streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM_VIDEO, xml)]
  55             if not streams:
  56                 raise ExtractorError(u'no media found')
  57             # choose default media type and highest quality for now
  58             stream = max([s for s in streams if s["progressiveDownloadUrl"].startswith("http://") ],
  59                          key=lambda s: int(s["frameWidth"]))
  60         else:
  61             streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM_AUDIO, xml)]
  62             if not streams:
  63                 raise ExtractorError(u'no media found')
  64             # choose default media type (MP4) and highest quality for now
  65             stream = max([s for s in streams if s["progressiveDownloadUrl"].startswith("http://") and s["mediaType"] == "MP4" ],
  66                          key=lambda s: int(s["bitrateAudio"]))
  67
  68         # there's two possibilities: RTMP stream or HTTP download
  69         info = {'id': video_id, 'title': title, 'ext': 'mp4'}
  70         if not stream["progressiveDownloadUrl"]:
  71             self.to_screen(u'RTMP download detected')
  72             assert stream['flashMediaServerURL'].startswith('mp4:')
  73             info["url"] = stream["flashMediaServerApplicationURL"]
  74             info["play_path"] = stream['flashMediaServerURL']
  75         else:
  76             assert stream["progressiveDownloadUrl"].endswith('.mp4')
  77             info["url"] = stream["progressiveDownloadUrl"]
  78         return [info]