youtube_dl/extractor/tfo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     HEADRequest,
   9     ExtractorError,
  10     int_or_none,
  11 )
  12
  13
  14 class TFOIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?tfo\.org/(?:en|fr)/(?:[^/]+/){2}(?P<id>\d+)'
  16     _TEST = {
  17         'url': 'http://www.tfo.org/en/universe/tfo-247/100463871/video-game-hackathon',
  18         'md5': '47c987d0515561114cf03d1226a9d4c7',
  19         'info_dict': {
  20             'id': '100463871',
  21             'ext': 'mp4',
  22             'title': 'Video Game Hackathon',
  23             'description': 'md5:558afeba217c6c8d96c60e5421795c07',
  24             'upload_date': '20160212',
  25             'timestamp': 1455310233,
  26         }
  27     }
  28
  29     def _real_extract(self, url):
  30         video_id = self._match_id(url)
  31         self._request_webpage(HEADRequest('http://www.tfo.org/'), video_id)
  32         infos = self._download_json(
  33             'http://www.tfo.org/api/web/video/get_infos', video_id, data=json.dumps({
  34                 'product_id': video_id,
  35             }).encode(), headers={
  36                 'X-tfo-session': self._get_cookies('http://www.tfo.org/')['tfo-session'].value,
  37             })
  38         if infos.get('success') == 0:
  39             raise ExtractorError('%s said: %s' % (self.IE_NAME, infos['msg']), expected=True)
  40         video_data = infos['data']
  41
  42         return {
  43             '_type': 'url_transparent',
  44             'id': video_id,
  45             'url': 'limelight:media:' + video_data['llid'],
  46             'title': video_data['title'],
  47             'description': video_data.get('description'),
  48             'series': video_data.get('collection'),
  49             'season_number': int_or_none(video_data.get('season')),
  50             'episode_number': int_or_none(video_data.get('episode')),
  51             'duration': int_or_none(video_data.get('duration')),
  52             'ie_key': 'LimelightMedia',
  53         }