youtube_dl/extractor/toutv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     js_to_json,
   8     ExtractorError,
   9     urlencode_postdata,
  10     extract_attributes,
  11     smuggle_url,
  12 )
  13
  14
  15 class TouTvIE(InfoExtractor):
  16     _NETRC_MACHINE = 'toutv'
  17     IE_NAME = 'tou.tv'
  18     _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/S[0-9]+E[0-9]+)?)'
  19     _access_token = None
  20     _claims = None
  21
  22     _TESTS = [{
  23         'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
  24         'info_dict': {
  25             'id': '122017',
  26             'ext': 'mp4',
  27             'title': 'Saison 2015 Épisode 17',
  28             'description': 'La photo de famille 2',
  29             'upload_date': '20100717',
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35         'skip': '404 Not Found',
  36     }, {
  37         'url': 'http://ici.tou.tv/hackers',
  38         'only_matching': True,
  39     }]
  40
  41     def _real_initialize(self):
  42         email, password = self._get_login_info()
  43         if email is None:
  44             return
  45         state = 'http://ici.tou.tv//'
  46         webpage = self._download_webpage(state, None, 'Downloading homepage')
  47         toutvlogin = self._parse_json(self._search_regex(
  48             r'(?s)toutvlogin\s*=\s*({.+?});', webpage, 'toutvlogin'), None, js_to_json)
  49         authorize_url = toutvlogin['host'] + '/auth/oauth/v2/authorize'
  50         login_webpage = self._download_webpage(
  51             authorize_url, None, 'Downloading login page', query={
  52                 'client_id': toutvlogin['clientId'],
  53                 'redirect_uri': 'https://ici.tou.tv/login/loginCallback',
  54                 'response_type': 'token',
  55                 'scope': 'media-drmt openid profile email id.write media-validation.read.privileged',
  56                 'state': state,
  57             })
  58         login_form = self._search_regex(
  59             r'(?s)(<form[^>]+(?:id|name)="Form-login".+?</form>)', login_webpage, 'login form')
  60         form_data = self._hidden_inputs(login_form)
  61         form_data.update({
  62             'login-email': email,
  63             'login-password': password,
  64         })
  65         post_url = extract_attributes(login_form).get('action') or authorize_url
  66         _, urlh = self._download_webpage_handle(
  67             post_url, None, 'Logging in', data=urlencode_postdata(form_data))
  68         self._access_token = self._search_regex(
  69             r'access_token=([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
  70             urlh.geturl(), 'access token')
  71         self._claims = self._download_json(
  72             'https://services.radio-canada.ca/media/validation/v2/getClaims',
  73             None, 'Extracting Claims', query={
  74                 'token': self._access_token,
  75                 'access_token': self._access_token,
  76             })['claims']
  77
  78     def _real_extract(self, url):
  79         path = self._match_id(url)
  80         metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
  81         if metadata.get('IsDrm'):
  82             raise ExtractorError('This video is DRM protected.', expected=True)
  83         video_id = metadata['IdMedia']
  84         details = metadata['Details']
  85         title = details['OriginalTitle']
  86         video_url = 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id)
  87         if self._access_token and self._claims:
  88             video_url = smuggle_url(video_url, {
  89                 'access_token': self._access_token,
  90                 'claims': self._claims,
  91             })
  92
  93         return {
  94             '_type': 'url_transparent',
  95             'url': video_url,
  96             'id': video_id,
  97             'title': title,
  98             'thumbnail': details.get('ImageUrl'),
  99             'duration': int_or_none(details.get('LengthInSeconds')),
 100         }