yt_dlp/extractor/noz.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     find_xpath_attr,
   5     xpath_text,
   6     update_url_query,
   7 )
   8 from ..compat import compat_urllib_parse_unquote
   9
  10
  11 class NozIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?noz\.de/video/(?P<id>[0-9]+)/'
  13     _TESTS = [{
  14         'url': 'http://www.noz.de/video/25151/32-Deutschland-gewinnt-Badminton-Lnderspiel-in-Melle',
  15         'info_dict': {
  16             'id': '25151',
  17             'ext': 'mp4',
  18             'duration': 215,
  19             'title': '3:2 - Deutschland gewinnt Badminton-Länderspiel in Melle',
  20             'description': 'Vor rund 370 Zuschauern gewinnt die deutsche Badminton-Nationalmannschaft am Donnerstag ein EM-Vorbereitungsspiel gegen Frankreich in Melle. Video Moritz Frankenberg.',
  21             'thumbnail': r're:^http://.*\.jpg',
  22         },
  23     }]
  24
  25     def _real_extract(self, url):
  26         video_id = self._match_id(url)
  27         webpage = self._download_webpage(url, video_id)
  28         description = self._og_search_description(webpage)
  29
  30         edge_url = self._html_search_regex(
  31             r'<script\s+(?:type="text/javascript"\s+)?src="(.*?/videojs_.*?)"',
  32             webpage, 'edge URL')
  33         edge_content = self._download_webpage(edge_url, 'meta configuration')
  34
  35         config_url_encoded = self._search_regex(
  36             r'so\.addVariable\("config_url","[^,]*,(.*?)"',
  37             edge_content, 'config URL'
  38         )
  39         config_url = compat_urllib_parse_unquote(config_url_encoded)
  40
  41         doc = self._download_xml(config_url, 'video configuration')
  42         title = xpath_text(doc, './/title')
  43         thumbnail = xpath_text(doc, './/article/thumbnail/url')
  44         duration = int_or_none(xpath_text(
  45             doc, './/article/movie/file/duration'))
  46         formats = []
  47         for qnode in doc.findall('.//article/movie/file/qualities/qual'):
  48             http_url_ele = find_xpath_attr(
  49                 qnode, './html_urls/video_url', 'format', 'video/mp4')
  50             http_url = http_url_ele.text if http_url_ele is not None else None
  51             if http_url:
  52                 formats.append({
  53                     'url': http_url,
  54                     'format_name': xpath_text(qnode, './name'),
  55                     'format_id': '%s-%s' % ('http', xpath_text(qnode, './id')),
  56                     'height': int_or_none(xpath_text(qnode, './height')),
  57                     'width': int_or_none(xpath_text(qnode, './width')),
  58                     'tbr': int_or_none(xpath_text(qnode, './bitrate'), scale=1000),
  59                 })
  60             else:
  61                 f4m_url = xpath_text(qnode, 'url_hd2')
  62                 if f4m_url:
  63                     formats.extend(self._extract_f4m_formats(
  64                         update_url_query(f4m_url, {'hdcore': '3.4.0'}),
  65                         video_id, f4m_id='hds', fatal=False))
  66                 m3u8_url_ele = find_xpath_attr(
  67                     qnode, './html_urls/video_url',
  68                     'format', 'application/vnd.apple.mpegurl')
  69                 m3u8_url = m3u8_url_ele.text if m3u8_url_ele is not None else None
  70                 if m3u8_url:
  71                     formats.extend(self._extract_m3u8_formats(
  72                         m3u8_url, video_id, 'mp4', 'm3u8_native',
  73                         m3u8_id='hls', fatal=False))
  74
  75         return {
  76             'id': video_id,
  77             'formats': formats,
  78             'title': title,
  79             'duration': duration,
  80             'description': description,
  81             'thumbnail': thumbnail,
  82         }