_ Git - youtube-dl/blob - 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     urlencode_postdata,
   9     extract_attributes,
  10     smuggle_url,
  11 )
  12
  13
  14 class TouTvIE(InfoExtractor):
  15     _NETRC_MACHINE = 'toutv'
  16     IE_NAME = 'tou.tv'
  17     _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/S[0-9]+E[0-9]+)?)'
  18     _access_token = None
  19     _claims = None
  20
  21     _TESTS = [{
  22         'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
  23         'info_dict': {
  24             'id': '122017',
  25             'ext': 'mp4',
  26             'title': 'Saison 2015 Épisode 17',
  27             'description': 'La photo de famille 2',
  28             'upload_date': '20100717',
  29         },
  30         'params': {
  31             # m3u8 download
  32             'skip_download': True,
  33         },
  34         'skip': '404 Not Found',
  35     }, {
  36         'url': 'http://ici.tou.tv/hackers',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_initialize(self):
  41         email, password = self._get_login_info()
  42         if email is None:
  43             return
  44         state = 'http://ici.tou.tv//'
  45         webpage = self._download_webpage(state, None, 'Downloading homepage')
  46         toutvlogin = self._parse_json(self._search_regex(
  47             r'(?s)toutvlogin\s*=\s*({.+?});', webpage, 'toutvlogin'), None, js_to_json)
  48         authorize_url = toutvlogin['host'] + '/auth/oauth/v2/authorize'
  49         login_webpage = self._download_webpage(
  50             authorize_url, None, 'Downloading login page', query={
  51                 'client_id': toutvlogin['clientId'],
  52                 'redirect_uri': 'https://ici.tou.tv/login/loginCallback',
  53                 'response_type': 'token',
  54                 'scope': 'media-drmt openid profile email id.write media-validation.read.privileged',
  55                 'state': state,
  56             })
  57         login_form = self._search_regex(
  58             r'(?s)(<form[^>]+(?:id|name)="Form-login".+?</form>)', login_webpage, 'login form')
  59         form_data = self._hidden_inputs(login_form)
  60         form_data.update({
  61             'login-email': email,
  62             'login-password': password,
  63         })
  64         post_url = extract_attributes(login_form).get('action') or authorize_url
  65         _, urlh = self._download_webpage_handle(
  66             post_url, None, 'Logging in', data=urlencode_postdata(form_data))
  67         self._access_token = self._search_regex(
  68             r'access_token=([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
  69             urlh.geturl(), 'access token')
  70         self._claims = self._download_json(
  71             'https://services.radio-canada.ca/media/validation/v2/getClaims',
  72             None, 'Extracting Claims', query={
  73                 'token': self._access_token,
  74                 'access_token': self._access_token,
  75             })['claims']
  76
  77     def _real_extract(self, url):
  78         path = self._match_id(url)
  79         metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
  80         # IsDrm does not necessarily mean the video is DRM protected (see
  81         # https://github.com/rg3/youtube-dl/issues/13994).
  82         if metadata.get('IsDrm'):
  83             self.report_warning('This video is probably DRM protected.', path)
  84         video_id = metadata['IdMedia']
  85         details = metadata['Details']
  86         title = details['OriginalTitle']
  87         video_url = 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id)
  88         if self._access_token and self._claims:
  89             video_url = smuggle_url(video_url, {
  90                 'access_token': self._access_token,
  91                 'claims': self._claims,
  92             })
  93
  94         return {
  95             '_type': 'url_transparent',
  96             'url': video_url,
  97             'id': video_id,
  98             'title': title,
  99             'thumbnail': details.get('ImageUrl'),
 100             'duration': int_or_none(details.get('LengthInSeconds')),
 101         }