yt_dlp/extractor/noz.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     find_xpath_attr,
   6     int_or_none,
   7     update_url_query,
   8     xpath_text,
   9 )
  10
  11
  12 class NozIE(InfoExtractor):
  13     _WORKING = False
  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': r'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 = 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_ele = find_xpath_attr(
  51                 qnode, './html_urls/video_url', 'format', 'video/mp4')
  52             http_url = http_url_ele.text if http_url_ele is not None else None
  53             if http_url:
  54                 formats.append({
  55                     'url': http_url,
  56                     'format_name': xpath_text(qnode, './name'),
  57                     'format_id': '{}-{}'.format('http', xpath_text(qnode, './id')),
  58                     'height': int_or_none(xpath_text(qnode, './height')),
  59                     'width': int_or_none(xpath_text(qnode, './width')),
  60                     'tbr': int_or_none(xpath_text(qnode, './bitrate'), scale=1000),
  61                 })
  62             else:
  63                 f4m_url = xpath_text(qnode, 'url_hd2')
  64                 if f4m_url:
  65                     formats.extend(self._extract_f4m_formats(
  66                         update_url_query(f4m_url, {'hdcore': '3.4.0'}),
  67                         video_id, f4m_id='hds', fatal=False))
  68                 m3u8_url_ele = find_xpath_attr(
  69                     qnode, './html_urls/video_url',
  70                     'format', 'application/vnd.apple.mpegurl')
  71                 m3u8_url = m3u8_url_ele.text if m3u8_url_ele is not None else None
  72                 if m3u8_url:
  73                     formats.extend(self._extract_m3u8_formats(
  74                         m3u8_url, video_id, 'mp4', 'm3u8_native',
  75                         m3u8_id='hls', fatal=False))
  76
  77         return {
  78             'id': video_id,
  79             'formats': formats,
  80             'title': title,
  81             'duration': duration,
  82             'description': description,
  83             'thumbnail': thumbnail,
  84         }