_ Git - youtube-dl/blob - youtube_dl/extractor/noz.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_urllib_parse_unquote
   6 from ..utils import (
   7     int_or_none,
   8     xpath_text,
   9     update_url_query,
  10 )
  11
  12
  13 class NozIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?noz\.de/video/(?P<id>[0-9]+)/'
  15     _TESTS = [{
  16         'url': 'http://www.noz.de/video/25151/32-Deutschland-gewinnt-Badminton-Lnderspiel-in-Melle',
  17         'info_dict': {
  18             'id': '25151',
  19             'ext': 'mp4',
  20             'duration': 215,
  21             'title': '3:2 - Deutschland gewinnt Badminton-Länderspiel in Melle',
  22             'description': 'Vor rund 370 Zuschauern gewinnt die deutsche Badminton-Nationalmannschaft am Donnerstag ein EM-Vorbereitungsspiel gegen Frankreich in Melle. Video Moritz Frankenberg.',
  23             'thumbnail': 're:^http://.*\.jpg',
  24         },
  25     }]
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29         webpage = self._download_webpage(url, video_id)
  30         description = self._og_search_description(webpage)
  31
  32         edge_url = self._html_search_regex(
  33             r'<script\s+(?:type="text/javascript"\s+)?src="(.*?/videojs_.*?)"',
  34             webpage, 'edge URL')
  35         edge_content = self._download_webpage(edge_url, 'meta configuration')
  36
  37         config_url_encoded = self._search_regex(
  38             r'so\.addVariable\("config_url","[^,]*,(.*?)"',
  39             edge_content, 'config URL'
  40         )
  41         config_url = compat_urllib_parse_unquote(config_url_encoded)
  42
  43         doc = self._download_xml(config_url, 'video configuration')
  44         title = xpath_text(doc, './/title')
  45         thumbnail = xpath_text(doc, './/article/thumbnail/url')
  46         duration = int_or_none(xpath_text(
  47             doc, './/article/movie/file/duration'))
  48         formats = []
  49         for qnode in doc.findall('.//article/movie/file/qualities/qual'):
  50             http_url = xpath_text(
  51                 qnode, './html_urls/video_url[@format="video/mp4"]')
  52             if http_url:
  53                 formats.append({
  54                     'url': http_url,
  55                     'format_name': xpath_text(qnode, './name'),
  56                     'format_id': '%s-%s' % ('http', xpath_text(qnode, './id')),
  57                     'height': int_or_none(xpath_text(qnode, './height')),
  58                     'width': int_or_none(xpath_text(qnode, './width')),
  59                     'tbr': int_or_none(xpath_text(qnode, './bitrate'), scale=1000),
  60                 })
  61             else:
  62                 f4m_url = xpath_text(qnode, 'url_hd2')
  63                 if f4m_url:
  64                     formats.extend(self._extract_f4m_formats(
  65                         update_url_query(f4m_url, {'hdcore': '3.4.0'}),
  66                         video_id, f4m_id='hds', fatal=False))
  67                 m3u8_url = xpath_text(
  68                     qnode, './html_urls/video_url[@format="application/vnd.apple.mpegurl"]')
  69                 if m3u8_url:
  70                     formats.extend(self._extract_m3u8_formats(
  71                         m3u8_url, video_id, 'mp4', 'm3u8_native',
  72                         m3u8_id='hls', fatal=False))
  73         self._sort_formats(formats)
  74
  75         return {
  76             'id': video_id,
  77             'formats': formats,
  78             'title': title,
  79             'duration': duration,
  80             'description': description,
  81             'thumbnail': thumbnail,
  82         }