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|-iframe)?(?:\.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         'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-iframe.html',
  44         'only_matching': True,
  45     }]
  46
  47     def _real_extract(self, url):
  48         video_id = self._match_id(url)
  49         webpage, handle = self._download_webpage_handle(url, video_id)
  50
  51         # 302 to spiegel.tv, like http://www.spiegel.de/video/der-film-zum-wochenende-die-wahrheit-ueber-maenner-video-99003272.html
  52         if SpiegeltvIE.suitable(handle.geturl()):
  53             return self.url_result(handle.geturl(), 'Spiegeltv')
  54
  55         title = re.sub(r'\s+', ' ', self._html_search_regex(
  56             r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>',
  57             webpage, 'title'))
  58         description = self._html_search_meta('description', webpage, 'description')
  59
  60         base_url = self._search_regex(
  61             r'var\s+server\s*=\s*"([^"]+)\"', webpage, 'server URL')
  62
  63         xml_url = base_url + video_id + '.xml'
  64         idoc = self._download_xml(xml_url, video_id)
  65
  66         formats = []
  67         for n in list(idoc):
  68             if n.tag.startswith('type') and n.tag != 'type6':
  69                 format_id = n.tag.rpartition('type')[2]
  70                 video_url = base_url + n.find('./filename').text
  71                 formats.append({
  72                     'format_id': format_id,
  73                     'url': video_url,
  74                     'width': int(n.find('./width').text),
  75                     'height': int(n.find('./height').text),
  76                     'abr': int(n.find('./audiobitrate').text),
  77                     'vbr': int(n.find('./videobitrate').text),
  78                     'vcodec': n.find('./codec').text,
  79                     'acodec': 'MP4A',
  80                 })
  81         duration = float(idoc[0].findall('./duration')[0].text)
  82
  83         self._check_formats(formats, video_id)
  84         self._sort_formats(formats)
  85
  86         return {
  87             'id': video_id,
  88             'title': title,
  89             'description': description,
  90             'duration': duration,
  91             'formats': formats,
  92         }
  93
  94
  95 class SpiegelArticleIE(InfoExtractor):
  96     _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
  97     IE_NAME = 'Spiegel:Article'
  98     IE_DESC = 'Articles on spiegel.de'
  99     _TESTS = [{
 100         'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html',
 101         'info_dict': {
 102             'id': '1516455',
 103             'ext': 'mp4',
 104             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
 105             'description': 're:^Patrick Kämnitz gehört.{100,}',
 106         },
 107     }, {
 108         'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html',
 109         'info_dict': {
 110
 111         },
 112         'playlist_count': 6,
 113     }]
 114
 115     def _real_extract(self, url):
 116         video_id = self._match_id(url)
 117         webpage = self._download_webpage(url, video_id)
 118
 119         # Single video on top of the page
 120         video_link = self._search_regex(
 121             r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage,
 122             'video page URL', default=None)
 123         if video_link:
 124             video_url = compat_urlparse.urljoin(
 125                 self.http_scheme() + '//spiegel.de/', video_link)
 126             return self.url_result(video_url)
 127
 128         # Multiple embedded videos
 129         embeds = re.findall(
 130             r'<div class="vid_holder[0-9]+.*?</div>\s*.*?url\s*=\s*"([^"]+)"',
 131             webpage)
 132         entries = [
 133             self.url_result(compat_urlparse.urljoin(
 134                 self.http_scheme() + '//spiegel.de/', embed_path))
 135             for embed_path in embeds
 136         ]
 137         return self.playlist_result(entries)