youtube_dl/extractor/spiegel.py

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