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 ..compat import compat_urlparse
   8 from .spiegeltv import SpiegeltvIE
   9
  10
  11 class SpiegelIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed)?(?:\.html)?(?:#.*)?$'
  13     _TESTS = [{
  14         'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
  15         'md5': '2c2754212136f35fb4b19767d242f66e',
  16         'info_dict': {
  17             'id': '1259285',
  18             'ext': 'mp4',
  19             'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv',
  20             'description': 'md5:8029d8310232196eb235d27575a8b9f4',
  21             'duration': 49,
  22         },
  23     }, {
  24         'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
  25         'md5': 'f2cdf638d7aa47654e251e1aee360af1',
  26         'info_dict': {
  27             'id': '1309159',
  28             'ext': 'mp4',
  29             'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers',
  30             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
  31             'duration': 983,
  32         },
  33     }, {
  34         'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-embed.html',
  35         'md5': 'd8eeca6bfc8f1cd6f490eb1f44695d51',
  36         'info_dict': {
  37             'id': '1519126',
  38             'ext': 'mp4',
  39             'description': 'SPIEGEL ONLINE-Nutzer durften den deutschen Astronauten Alexander Gerst über sein Leben auf der ISS-Station befragen. Hier kommen seine Antworten auf die besten sechs Fragen.',
  40             'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"',
  41         }
  42     }]
  43
  44     def _real_extract(self, url):
  45         video_id = self._match_id(url)
  46         webpage, handle = self._download_webpage_handle(url, video_id)
  47
  48         # 302 to spiegel.tv, like http://www.spiegel.de/video/der-film-zum-wochenende-die-wahrheit-ueber-maenner-video-99003272.html
  49         if SpiegeltvIE.suitable(handle.geturl()):
  50             return self.url_result(handle.geturl(), 'Spiegeltv')
  51
  52         title = re.sub(r'\s+', ' ', self._html_search_regex(
  53             r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>',
  54             webpage, 'title'))
  55         description = self._html_search_meta('description', webpage, 'description')
  56
  57         base_url = self._search_regex(
  58             r'var\s+server\s*=\s*"([^"]+)\"', webpage, 'server URL')
  59
  60         xml_url = base_url + video_id + '.xml'
  61         idoc = self._download_xml(xml_url, video_id)
  62
  63         formats = [
  64             {
  65                 'format_id': n.tag.rpartition('type')[2],
  66                 'url': base_url + n.find('./filename').text,
  67                 'width': int(n.find('./width').text),
  68                 'height': int(n.find('./height').text),
  69                 'abr': int(n.find('./audiobitrate').text),
  70                 'vbr': int(n.find('./videobitrate').text),
  71                 'vcodec': n.find('./codec').text,
  72                 'acodec': 'MP4A',
  73             }
  74             for n in list(idoc)
  75             # Blacklist type 6, it's extremely LQ and not available on the same server
  76             if n.tag.startswith('type') and n.tag != 'type6'
  77         ]
  78         duration = float(idoc[0].findall('./duration')[0].text)
  79
  80         self._sort_formats(formats)
  81
  82         return {
  83             'id': video_id,
  84             'title': title,
  85             'description': description,
  86             'duration': duration,
  87             'formats': formats,
  88         }
  89
  90
  91 class SpiegelArticleIE(InfoExtractor):
  92     _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
  93     IE_NAME = 'Spiegel:Article'
  94     IE_DESC = 'Articles on spiegel.de'
  95     _TESTS = [{
  96         'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html',
  97         'info_dict': {
  98             'id': '1516455',
  99             'ext': 'mp4',
 100             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
 101             'description': 're:^Patrick Kämnitz gehört.{100,}',
 102         },
 103     }, {
 104         'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html',
 105         'info_dict': {
 106
 107         },
 108         'playlist_count': 6,
 109     }]
 110
 111     def _real_extract(self, url):
 112         video_id = self._match_id(url)
 113         webpage = self._download_webpage(url, video_id)
 114
 115         # Single video on top of the page
 116         video_link = self._search_regex(
 117             r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage,
 118             'video page URL', default=None)
 119         if video_link:
 120             video_url = compat_urlparse.urljoin(
 121                 self.http_scheme() + '//spiegel.de/', video_link)
 122             return self.url_result(video_url)
 123
 124         # Multiple embedded videos
 125         embeds = re.findall(
 126             r'<div class="vid_holder[0-9]+.*?</div>\s*.*?url\s*=\s*"([^"]+)"',
 127             webpage)
 128         entries = [
 129             self.url_result(compat_urlparse.urljoin(
 130                 self.http_scheme() + '//spiegel.de/', embed_path))
 131             for embed_path in embeds
 132         ]
 133         return self.playlist_result(entries)