youtube_dl/extractor/tf1.py

   1 # coding: utf-8
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8 class TF1IE(InfoExtractor):
   9     """TF1 uses the wat.tv player."""
  10     _VALID_URL = r'http://videos.tf1.fr/.*-(.*?).html'
  11     _TEST = {
  12         u'url': u'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html',
  13         u'file': u'10635995.mp4',
  14         u'md5': u'2e378cc28b9957607d5e88f274e637d8',
  15         u'info_dict': {
  16             u'title': u'Citroën Grand C4 Picasso 2013 : présentation officielle',
  17             u'description': u'Vidéo officielle du nouveau Citroën Grand C4 Picasso, lancé à l\'automne 2013.',
  18         },
  19         u'skip': u'Sometimes wat serves the whole file with the --test option',
  20     }
  21
  22     def _real_extract(self, url):
  23         mobj = re.match(self._VALID_URL, url)
  24         id = mobj.group(1)
  25         webpage = self._download_webpage(url, id)
  26         embed_url = self._html_search_regex(r'"(https://www.wat.tv/embedframe/.*?)"',
  27                                 webpage, 'embed url')
  28         embed_page = self._download_webpage(embed_url, id, u'Downloading embed player page')
  29         wat_id = self._search_regex(r'UVID=(.*?)&', embed_page, 'wat id')
  30         wat_info = self._download_webpage('http://www.wat.tv/interface/contentv3/%s' % wat_id, id, u'Downloading Wat info')
  31         wat_info = json.loads(wat_info)['media']
  32         wat_url = wat_info['url']
  33         return self.url_result(wat_url, 'Wat')