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 .spiegeltv import SpiegeltvIE
   8 from ..compat import compat_urlparse
   9 from ..utils import (
  10     extract_attributes,
  11     unified_strdate,
  12     get_element_by_attribute,
  13 )
  14
  15
  16 class SpiegelIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed|-iframe)?(?:\.html)?(?:#.*)?$'
  18     _TESTS = [{
  19         'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
  20         'md5': '2c2754212136f35fb4b19767d242f66e',
  21         'info_dict': {
  22             'id': '1259285',
  23             'ext': 'mp4',
  24             'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv',
  25             'description': 'md5:8029d8310232196eb235d27575a8b9f4',
  26             'duration': 49,
  27             'upload_date': '20130311',
  28         },
  29     }, {
  30         'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
  31         'md5': 'f2cdf638d7aa47654e251e1aee360af1',
  32         'info_dict': {
  33             'id': '1309159',
  34             'ext': 'mp4',
  35             'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers',
  36             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
  37             'duration': 983,
  38             'upload_date': '20131115',
  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             'upload_date': '20140904',
  49         }
  50     }, {
  51         'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-iframe.html',
  52         'only_matching': True,
  53     }]
  54
  55     def _real_extract(self, url):
  56         video_id = self._match_id(url)
  57         webpage, handle = self._download_webpage_handle(url, video_id)
  58
  59         # 302 to spiegel.tv, like http://www.spiegel.de/video/der-film-zum-wochenende-die-wahrheit-ueber-maenner-video-99003272.html
  60         if SpiegeltvIE.suitable(handle.geturl()):
  61             return self.url_result(handle.geturl(), 'Spiegeltv')
  62
  63         video_data = extract_attributes(self._search_regex(r'(<div[^>]+id="spVideoElements"[^>]+>)', webpage, 'video element', default=''))
  64
  65         title = video_data.get('data-video-title') or get_element_by_attribute('class', 'module-title', webpage)
  66         description = video_data.get('data-video-teaser') or self._html_search_meta('description', webpage, 'description')
  67
  68         base_url = self._search_regex(
  69             [r'server\s*:\s*(["\'])(?P<url>.+?)\1', r'var\s+server\s*=\s*"(?P<url>[^"]+)\"'],
  70             webpage, 'server URL', group='url')
  71
  72         xml_url = base_url + video_id + '.xml'
  73         idoc = self._download_xml(xml_url, video_id)
  74
  75         formats = []
  76         for n in list(idoc):
  77             if n.tag.startswith('type') and n.tag != 'type6':
  78                 format_id = n.tag.rpartition('type')[2]
  79                 video_url = base_url + n.find('./filename').text
  80                 formats.append({
  81                     'format_id': format_id,
  82                     'url': video_url,
  83                     'width': int(n.find('./width').text),
  84                     'height': int(n.find('./height').text),
  85                     'abr': int(n.find('./audiobitrate').text),
  86                     'vbr': int(n.find('./videobitrate').text),
  87                     'vcodec': n.find('./codec').text,
  88                     'acodec': 'MP4A',
  89                 })
  90         duration = float(idoc[0].findall('./duration')[0].text)
  91
  92         self._check_formats(formats, video_id)
  93         self._sort_formats(formats)
  94
  95         return {
  96             'id': video_id,
  97             'title': title,
  98             'description': description.strip() if description else None,
  99             'duration': duration,
 100             'upload_date': unified_strdate(video_data.get('data-video-date')),
 101             'formats': formats,
 102         }
 103
 104
 105 class SpiegelArticleIE(InfoExtractor):
 106     _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
 107     IE_NAME = 'Spiegel:Article'
 108     IE_DESC = 'Articles on spiegel.de'
 109     _TESTS = [{
 110         'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html',
 111         'info_dict': {
 112             'id': '1516455',
 113             'ext': 'mp4',
 114             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
 115             'description': 're:^Patrick Kämnitz gehört.{100,}',
 116         },
 117     }, {
 118         'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html',
 119         'info_dict': {
 120
 121         },
 122         'playlist_count': 6,
 123     }]
 124
 125     def _real_extract(self, url):
 126         video_id = self._match_id(url)
 127         webpage = self._download_webpage(url, video_id)
 128
 129         # Single video on top of the page
 130         video_link = self._search_regex(
 131             r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage,
 132             'video page URL', default=None)
 133         if video_link:
 134             video_url = compat_urlparse.urljoin(
 135                 self.http_scheme() + '//spiegel.de/', video_link)
 136             return self.url_result(video_url)
 137
 138         # Multiple embedded videos
 139         embeds = re.findall(
 140             r'<div class="vid_holder[0-9]+.*?</div>\s*.*?url\s*=\s*"([^"]+)"',
 141             webpage)
 142         entries = [
 143             self.url_result(compat_urlparse.urljoin(
 144                 self.http_scheme() + '//spiegel.de/', embed_path))
 145             for embed_path in embeds
 146         ]
 147         return self.playlist_result(entries)