_ 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         video_id = self._match_id(url)
  32         display_id = video_id.lstrip('-')
  33         webpage = self._download_webpage(url, display_id)
  34
  35         playerpage = self._download_webpage(
  36             'http://www.tagesschau.de/multimedia/video/video%s~player_autoplay-true.html' % video_id,
  37             display_id, 'Downloading player page')
  38
  39         medias = re.findall(
  40             r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"',
  41             playerpage)
  42
  43         formats = []
  44         for url, ext, res in medias:
  45             f = {
  46                 'format_id': res + '_' + ext,
  47                 'url': url,
  48                 'ext': ext,
  49             }
  50             f.update(self._FORMATS.get(res, {}))
  51             formats.append(f)
  52
  53         self._sort_formats(formats)
  54
  55         thumbnail = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1]
  56
  57         return {
  58             'id': display_id,
  59             'title': self._og_search_title(webpage).strip(),
  60             'thumbnail': 'http://www.tagesschau.de' + thumbnail,
  61             'formats': formats,
  62             'description': self._og_search_description(webpage).strip(),
  63         }