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