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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class SpiegelIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
  11     _TESTS = [{
  12         'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
  13         'md5': '2c2754212136f35fb4b19767d242f66e',
  14         'info_dict': {
  15             'id': '1259285',
  16             'ext': 'mp4',
  17             'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv',
  18             'description': 'md5:8029d8310232196eb235d27575a8b9f4',
  19             'duration': 49,
  20         },
  21     }, {
  22         'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
  23         'md5': 'f2cdf638d7aa47654e251e1aee360af1',
  24         'info_dict': {
  25             'id': '1309159',
  26             'ext': 'mp4',
  27             'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers',
  28             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
  29             'duration': 983,
  30         },
  31     }, {
  32         'url': 'http://www.spiegel.de/video/johann-westhauser-videobotschaft-des-hoehlenforschers-video-1502367.html',
  33         'md5': '54f58ba0e752e3c07bc2a26222dd0acf',
  34         'info_dict': {
  35             'id': '1502367',
  36             'ext': 'mp4',
  37             'title': 'Videobotschaft: Höhlenforscher Westhauser dankt seinen Rettern',
  38             'description': 'md5:c6f1ec11413ebd1088b6813943e5fc91',
  39             'duration': 42,
  40         },
  41     }]
  42
  43     def _real_extract(self, url):
  44         m = re.match(self._VALID_URL, url)
  45         video_id = m.group('videoID')
  46
  47         webpage = self._download_webpage(url, video_id)
  48
  49         title = self._html_search_regex(
  50             r'<div class="module-title">(.*?)</div>', webpage, 'title')
  51         description = self._html_search_meta('description', webpage, 'description')
  52
  53         base_url = self._search_regex(
  54             r'var\s+server\s*=\s*"([^"]+)\"', webpage, 'server URL')
  55
  56         xml_url = base_url + video_id + '.xml'
  57         idoc = self._download_xml(xml_url, video_id)
  58
  59         formats = [
  60             {
  61                 'format_id': n.tag.rpartition('type')[2],
  62                 'url': base_url + n.find('./filename').text,
  63                 'width': int(n.find('./width').text),
  64                 'height': int(n.find('./height').text),
  65                 'abr': int(n.find('./audiobitrate').text),
  66                 'vbr': int(n.find('./videobitrate').text),
  67                 'vcodec': n.find('./codec').text,
  68                 'acodec': 'MP4A',
  69             }
  70             for n in list(idoc)
  71             # Blacklist type 6, it's extremely LQ and not available on the same server
  72             if n.tag.startswith('type') and n.tag != 'type6'
  73         ]
  74         duration = float(idoc[0].findall('./duration')[0].text)
  75
  76         self._sort_formats(formats)
  77
  78         return {
  79             'id': video_id,
  80             'title': title,
  81             'description': description,
  82             'duration': duration,
  83             'formats': formats,
  84         }