_ Git - youtube-dl/blob - youtube_dl/extractor/tagesschau.py

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class TagesschauIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/video/video(?P<id>-?[0-9]+)\.html'
  11
  12     _TESTS = [{
  13         'url': 'http://www.tagesschau.de/multimedia/video/video1399128.html',
  14         'md5': 'bcdeac2194fb296d599ce7929dfa4009',
  15         'info_dict': {
  16             'id': '1399128',
  17             'ext': 'mp4',
  18             'title': 'Harald Range, Generalbundesanwalt, zu den Ermittlungen',
  19             'description': 'md5:69da3c61275b426426d711bde96463ab',
  20             'thumbnail': 're:^http:.*\.jpg$',
  21         },
  22     }]
  23
  24     _FORMATS = {
  25         's': {'width': 256, 'height': 144, 'quality': 1},
  26         'm': {'width': 512, 'height': 288, 'quality': 2},
  27         'l': {'width': 960, 'height': 544, 'quality': 3},
  28     }
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         video_id = mobj.group('id')
  33
  34         if video_id.startswith('-'):
  35             display_id = video_id.strip('-')
  36         else:
  37             display_id = video_id
  38
  39         webpage = self._download_webpage(url, display_id)
  40
  41         playerpage = self._download_webpage(
  42             'http://www.tagesschau.de/multimedia/video/video%s~player_autoplay-true.html' % video_id,
  43             display_id, 'Downloading player page')
  44
  45         medias = re.findall(
  46             r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"',
  47             playerpage)
  48
  49         formats = []
  50         for url, ext, res in medias:
  51             f = {
  52                 'format_id': res + '_' + ext,
  53                 'url': url,
  54                 'ext': ext,
  55             }
  56             f.update(self._FORMATS.get(res, {}))
  57             formats.append(f)
  58
  59         self._sort_formats(formats)
  60
  61         thumbnail = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1]
  62
  63         return {
  64             'id': display_id,
  65             'title': self._og_search_title(webpage).strip(),
  66             'thumbnail': 'http://www.tagesschau.de' + thumbnail,
  67             'formats': formats,
  68             'description': self._og_search_description(webpage).strip(),
  69         }