_ Git - youtube-dl/blob - 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     """
  10     TF1 uses the wat.tv player, currently it can only download videos with the
  11     html5 player enabled, it cannot download HD videos or the news.
  12     """
  13     _VALID_URL = r'http://videos.tf1.fr/.*-(.*?).html'
  14     _TEST = {
  15         u'url': u'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html',
  16         u'file': u'6bysb.mp4',
  17         u'md5': u'66789d3e91278d332f75e1feb7aea327',
  18         u'info_dict': {
  19             u"title": u"Citroën Grand C4 Picasso 2013 : présentation officielle"
  20         }
  21     }
  22
  23     def _real_extract(self, url):
  24         mobj = re.match(self._VALID_URL, url)
  25         id = mobj.group(1)
  26         webpage = self._download_webpage(url, id)
  27         embed_url = self._html_search_regex(r'"(https://www.wat.tv/embedframe/.*?)"',
  28                                 webpage, 'embed url')
  29         embed_page = self._download_webpage(embed_url, id, u'Downloading embed player page')
  30         wat_id = self._search_regex(r'UVID=(.*?)&', embed_page, 'wat id')
  31         wat_info = self._download_webpage('http://www.wat.tv/interface/contentv3/%s' % wat_id, id, u'Downloading Wat info')
  32         wat_info = json.loads(wat_info)['media']
  33         wat_url = wat_info['url']
  34         return self.url_result(wat_url, 'Wat')