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 )
  10
  11
  12 class NozIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?noz\.de/video/(?P<id>[0-9]+)/'
  14     _TESTS = [{
  15         'url': 'http://www.noz.de/video/25151/32-Deutschland-gewinnt-Badminton-Lnderspiel-in-Melle',
  16         'info_dict': {
  17             'id': '25151',
  18             'ext': 'mp4',
  19             'duration': 215,
  20             'title': '3:2 - Deutschland gewinnt Badminton-Länderspiel in Melle',
  21             'description': 'Vor rund 370 Zuschauern gewinnt die deutsche Badminton-Nationalmannschaft am Donnerstag ein EM-Vorbereitungsspiel gegen Frankreich in Melle. Video Moritz Frankenberg.',
  22             'thumbnail': 're:^http://.*\.jpg',
  23         },
  24     }]
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28         webpage = self._download_webpage(url, video_id)
  29         description = self._og_search_description(webpage)
  30
  31         edge_url = self._html_search_regex(
  32             r'<script\s+(?:type="text/javascript"\s+)?src="(.*?/videojs_.*?)"',
  33             webpage, 'edge URL')
  34         edge_content = self._download_webpage(edge_url, 'meta configuration')
  35
  36         config_url_encoded = self._search_regex(
  37             r'so\.addVariable\("config_url","[^,]*,(.*?)"',
  38             edge_content, 'config URL'
  39         )
  40         config_url = compat_urllib_parse_unquote(config_url_encoded)
  41
  42         doc = self._download_xml(config_url, 'video configuration')
  43         title = xpath_text(doc, './/title')
  44         thumbnail = xpath_text(doc, './/article/thumbnail/url')
  45         duration = int_or_none(xpath_text(
  46             doc, './/article/movie/file/duration'))
  47         formats = []
  48         for qnode in doc.findall('.//article/movie/file/qualities/qual'):
  49             video_node = qnode.find('./html_urls/video_url[@format="video/mp4"]')
  50             if video_node is None:
  51                 continue  # auto
  52             formats.append({
  53                 'url': video_node.text,
  54                 'format_name': xpath_text(qnode, './name'),
  55                 'format_id': 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         self._sort_formats(formats)
  61
  62         return {
  63             'id': video_id,
  64             'formats': formats,
  65             'title': title,
  66             'duration': duration,
  67             'description': description,
  68             'thumbnail': thumbnail,
  69         }