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 (
   8     compat_urlparse,
   9     compat_HTTPError,
  10 )
  11 from ..utils import (
  12     HEADRequest,
  13     ExtractorError,
  14 )
  15 from .spiegeltv import SpiegeltvIE
  16
  17
  18 class SpiegelIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed)?(?:\.html)?(?:#.*)?$'
  20     _TESTS = [{
  21         'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
  22         'md5': '2c2754212136f35fb4b19767d242f66e',
  23         'info_dict': {
  24             'id': '1259285',
  25             'ext': 'mp4',
  26             'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv',
  27             'description': 'md5:8029d8310232196eb235d27575a8b9f4',
  28             'duration': 49,
  29         },
  30     }, {
  31         'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
  32         'md5': 'f2cdf638d7aa47654e251e1aee360af1',
  33         'info_dict': {
  34             'id': '1309159',
  35             'ext': 'mp4',
  36             'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers',
  37             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
  38             'duration': 983,
  39         },
  40     }, {
  41         'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-embed.html',
  42         'md5': 'd8eeca6bfc8f1cd6f490eb1f44695d51',
  43         'info_dict': {
  44             'id': '1519126',
  45             'ext': 'mp4',
  46             '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.',
  47             'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"',
  48         }
  49     }]
  50
  51     def _real_extract(self, url):
  52         video_id = self._match_id(url)
  53         webpage, handle = self._download_webpage_handle(url, video_id)
  54
  55         # 302 to spiegel.tv, like http://www.spiegel.de/video/der-film-zum-wochenende-die-wahrheit-ueber-maenner-video-99003272.html
  56         if SpiegeltvIE.suitable(handle.geturl()):
  57             return self.url_result(handle.geturl(), 'Spiegeltv')
  58
  59         title = re.sub(r'\s+', ' ', self._html_search_regex(
  60             r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>',
  61             webpage, 'title'))
  62         description = self._html_search_meta('description', webpage, 'description')
  63
  64         base_url = self._search_regex(
  65             r'var\s+server\s*=\s*"([^"]+)\"', webpage, 'server URL')
  66
  67         xml_url = base_url + video_id + '.xml'
  68         idoc = self._download_xml(xml_url, video_id)
  69
  70         formats = []
  71         for n in list(idoc):
  72             if n.tag.startswith('type') and n.tag != 'type6':
  73                 format_id = n.tag.rpartition('type')[2]
  74                 video_url = base_url + n.find('./filename').text
  75                 # Test video URLs beforehand as some of them are invalid
  76                 try:
  77                     self._request_webpage(
  78                         HEADRequest(video_url), video_id,
  79                         'Checking %s video URL' % format_id)
  80                 except ExtractorError as e:
  81                     if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
  82                         self.report_warning(
  83                             '%s video URL is invalid, skipping' % format_id, video_id)
  84                         continue
  85                 formats.append({
  86                     'format_id': format_id,
  87                     'url': video_url,
  88                     'width': int(n.find('./width').text),
  89                     'height': int(n.find('./height').text),
  90                     'abr': int(n.find('./audiobitrate').text),
  91                     'vbr': int(n.find('./videobitrate').text),
  92                     'vcodec': n.find('./codec').text,
  93                     'acodec': 'MP4A',
  94                 })
  95         duration = float(idoc[0].findall('./duration')[0].text)
  96
  97         self._sort_formats(formats)
  98
  99         return {
 100             'id': video_id,
 101             'title': title,
 102             'description': description,
 103             'duration': duration,
 104             'formats': formats,
 105         }
 106
 107
 108 class SpiegelArticleIE(InfoExtractor):
 109     _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
 110     IE_NAME = 'Spiegel:Article'
 111     IE_DESC = 'Articles on spiegel.de'
 112     _TESTS = [{
 113         'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html',
 114         'info_dict': {
 115             'id': '1516455',
 116             'ext': 'mp4',
 117             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
 118             'description': 're:^Patrick Kämnitz gehört.{100,}',
 119         },
 120     }, {
 121         'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html',
 122         'info_dict': {
 123
 124         },
 125         'playlist_count': 6,
 126     }]
 127
 128     def _real_extract(self, url):
 129         video_id = self._match_id(url)
 130         webpage = self._download_webpage(url, video_id)
 131
 132         # Single video on top of the page
 133         video_link = self._search_regex(
 134             r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage,
 135             'video page URL', default=None)
 136         if video_link:
 137             video_url = compat_urlparse.urljoin(
 138                 self.http_scheme() + '//spiegel.de/', video_link)
 139             return self.url_result(video_url)
 140
 141         # Multiple embedded videos
 142         embeds = re.findall(
 143             r'<div class="vid_holder[0-9]+.*?</div>\s*.*?url\s*=\s*"([^"]+)"',
 144             webpage)
 145         entries = [
 146             self.url_result(compat_urlparse.urljoin(
 147                 self.http_scheme() + '//spiegel.de/', embed_path))
 148             for embed_path in embeds
 149         ]
 150         return self.playlist_result(entries)