youtube_dl/extractor/spiegel.py

   1 import re
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5
   6
   7 class SpiegelIE(InfoExtractor):
   8     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
   9     _TESTS = [{
  10         u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
  11         u'file': u'1259285.mp4',
  12         u'md5': u'2c2754212136f35fb4b19767d242f66e',
  13         u'info_dict': {
  14             u"title": u"Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv"
  15         }
  16     },
  17     {
  18         u'url': u'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
  19         u'file': u'1309159.mp4',
  20         u'md5': u'f2cdf638d7aa47654e251e1aee360af1',
  21         u'info_dict': {
  22             u'title': u'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers'
  23         }
  24     }]
  25
  26     def _real_extract(self, url):
  27         m = re.match(self._VALID_URL, url)
  28         video_id = m.group('videoID')
  29
  30         webpage = self._download_webpage(url, video_id)
  31
  32         video_title = self._html_search_regex(
  33             r'<div class="module-title">(.*?)</div>', webpage, u'title')
  34
  35         xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml'
  36         xml_code = self._download_webpage(
  37             xml_url, video_id,
  38             note=u'Downloading XML', errnote=u'Failed to download XML')
  39
  40         idoc = xml.etree.ElementTree.fromstring(xml_code)
  41
  42         formats = [
  43             {
  44                 'format_id': n.tag.rpartition('type')[2],
  45                 'url': u'http://video2.spiegel.de/flash/' + n.find('./filename').text,
  46                 'width': int(n.find('./width').text),
  47                 'height': int(n.find('./height').text),
  48                 'abr': int(n.find('./audiobitrate').text),
  49                 'vbr': int(n.find('./videobitrate').text),
  50                 'vcodec': n.find('./codec').text,
  51                 'acodec': 'MP4A',
  52             }
  53             for n in list(idoc)
  54             # Blacklist type 6, it's extremely LQ and not available on the same server
  55             if n.tag.startswith('type') and n.tag != 'type6'
  56         ]
  57         formats.sort(key=lambda f: f['vbr'])
  58         duration = float(idoc[0].findall('./duration')[0].text)
  59
  60         info = {
  61             'id': video_id,
  62             'title': video_title,
  63             'duration': duration,
  64             'formats': formats,
  65         }
  66         return info