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 from ..utils import compat_urlparse
   8
   9
  10 class SpiegelIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
  12     _TESTS = [{
  13         'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
  14         'md5': '2c2754212136f35fb4b19767d242f66e',
  15         'info_dict': {
  16             'id': '1259285',
  17             'ext': 'mp4',
  18             'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv',
  19             'description': 'md5:8029d8310232196eb235d27575a8b9f4',
  20             'duration': 49,
  21         },
  22     }, {
  23         'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
  24         'md5': 'f2cdf638d7aa47654e251e1aee360af1',
  25         'info_dict': {
  26             'id': '1309159',
  27             'ext': 'mp4',
  28             'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers',
  29             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
  30             'duration': 983,
  31         },
  32     }]
  33
  34     def _real_extract(self, url):
  35         m = re.match(self._VALID_URL, url)
  36         video_id = m.group('videoID')
  37
  38         webpage = self._download_webpage(url, video_id)
  39
  40         title = self._html_search_regex(
  41             r'<div class="module-title">(.*?)</div>', webpage, 'title')
  42         description = self._html_search_meta('description', webpage, 'description')
  43
  44         base_url = self._search_regex(
  45             r'var\s+server\s*=\s*"([^"]+)\"', webpage, 'server URL')
  46
  47         xml_url = base_url + video_id + '.xml'
  48         idoc = self._download_xml(xml_url, video_id)
  49
  50         formats = [
  51             {
  52                 'format_id': n.tag.rpartition('type')[2],
  53                 'url': base_url + n.find('./filename').text,
  54                 'width': int(n.find('./width').text),
  55                 'height': int(n.find('./height').text),
  56                 'abr': int(n.find('./audiobitrate').text),
  57                 'vbr': int(n.find('./videobitrate').text),
  58                 'vcodec': n.find('./codec').text,
  59                 'acodec': 'MP4A',
  60             }
  61             for n in list(idoc)
  62             # Blacklist type 6, it's extremely LQ and not available on the same server
  63             if n.tag.startswith('type') and n.tag != 'type6'
  64         ]
  65         duration = float(idoc[0].findall('./duration')[0].text)
  66
  67         self._sort_formats(formats)
  68
  69         return {
  70             'id': video_id,
  71             'title': title,
  72             'description': description,
  73             'duration': duration,
  74             'formats': formats,
  75         }
  76
  77
  78 class SpiegelArticleIE(InfoExtractor):
  79     _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
  80     IE_NAME = 'Spiegel:Article'
  81     IE_DESC = 'Articles on spiegel.de'
  82     _TEST = {
  83         'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html',
  84         'info_dict': {
  85             'id': '1516455',
  86             'ext': 'mp4',
  87             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
  88             'description': 're:^Patrick Kämnitz gehört.{100,}',
  89         },
  90     }
  91
  92     def _real_extract(self, url):
  93         m = re.match(self._VALID_URL, url)
  94         video_id = m.group('id')
  95
  96         webpage = self._download_webpage(url, video_id)
  97         video_link = self._search_regex(
  98             r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage,
  99             'video page URL')
 100         video_url = compat_urlparse.urljoin(
 101             self.http_scheme() + '//spiegel.de/', video_link)
 102
 103         return {
 104             '_type': 'url',
 105             'url': video_url,
 106         }