yt_dlp/extractor/noz.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_urllib_parse_unquote,
   7 )
   8 from ..utils import (
   9     int_or_none,
  10     find_xpath_attr,
  11     xpath_text,
  12     update_url_query,
  13 )
  14
  15
  16 class NozIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?noz\.de/video/(?P<id>[0-9]+)/'
  18     _TESTS = [{
  19         'url': 'http://www.noz.de/video/25151/32-Deutschland-gewinnt-Badminton-Lnderspiel-in-Melle',
  20         'info_dict': {
  21             'id': '25151',
  22             'ext': 'mp4',
  23             'duration': 215,
  24             'title': '3:2 - Deutschland gewinnt Badminton-Länderspiel in Melle',
  25             'description': 'Vor rund 370 Zuschauern gewinnt die deutsche Badminton-Nationalmannschaft am Donnerstag ein EM-Vorbereitungsspiel gegen Frankreich in Melle. Video Moritz Frankenberg.',
  26             'thumbnail': r're:^http://.*\.jpg',
  27         },
  28     }]
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32         webpage = self._download_webpage(url, video_id)
  33         description = self._og_search_description(webpage)
  34
  35         edge_url = self._html_search_regex(
  36             r'<script\s+(?:type="text/javascript"\s+)?src="(.*?/videojs_.*?)"',
  37             webpage, 'edge URL')
  38         edge_content = self._download_webpage(edge_url, 'meta configuration')
  39
  40         config_url_encoded = self._search_regex(
  41             r'so\.addVariable\("config_url","[^,]*,(.*?)"',
  42             edge_content, 'config URL'
  43         )
  44         config_url = compat_urllib_parse_unquote(config_url_encoded)
  45
  46         doc = self._download_xml(config_url, 'video configuration')
  47         title = xpath_text(doc, './/title')
  48         thumbnail = xpath_text(doc, './/article/thumbnail/url')
  49         duration = int_or_none(xpath_text(
  50             doc, './/article/movie/file/duration'))
  51         formats = []
  52         for qnode in doc.findall('.//article/movie/file/qualities/qual'):
  53             http_url_ele = find_xpath_attr(
  54                 qnode, './html_urls/video_url', 'format', 'video/mp4')
  55             http_url = http_url_ele.text if http_url_ele is not None else None
  56             if http_url:
  57                 formats.append({
  58                     'url': http_url,
  59                     'format_name': xpath_text(qnode, './name'),
  60                     'format_id': '%s-%s' % ('http', xpath_text(qnode, './id')),
  61                     'height': int_or_none(xpath_text(qnode, './height')),
  62                     'width': int_or_none(xpath_text(qnode, './width')),
  63                     'tbr': int_or_none(xpath_text(qnode, './bitrate'), scale=1000),
  64                 })
  65             else:
  66                 f4m_url = xpath_text(qnode, 'url_hd2')
  67                 if f4m_url:
  68                     formats.extend(self._extract_f4m_formats(
  69                         update_url_query(f4m_url, {'hdcore': '3.4.0'}),
  70                         video_id, f4m_id='hds', fatal=False))
  71                 m3u8_url_ele = find_xpath_attr(
  72                     qnode, './html_urls/video_url',
  73                     'format', 'application/vnd.apple.mpegurl')
  74                 m3u8_url = m3u8_url_ele.text if m3u8_url_ele is not None else None
  75                 if m3u8_url:
  76                     formats.extend(self._extract_m3u8_formats(
  77                         m3u8_url, video_id, 'mp4', 'm3u8_native',
  78                         m3u8_id='hls', fatal=False))
  79         self._sort_formats(formats)
  80
  81         return {
  82             'id': video_id,
  83             'formats': formats,
  84             'title': title,
  85             'duration': duration,
  86             'description': description,
  87             'thumbnail': thumbnail,
  88         }