_ Git - youtube-dl/blob - youtube_dl/extractor/toutv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     js_to_json,
  10     urlencode_postdata,
  11     extract_attributes,
  12     smuggle_url,
  13 )
  14
  15
  16 class TouTvIE(InfoExtractor):
  17     _NETRC_MACHINE = 'toutv'
  18     IE_NAME = 'tou.tv'
  19     _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/S[0-9]+E[0-9]+)?)'
  20     _access_token = None
  21     _claims = None
  22
  23     _TESTS = [{
  24         'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
  25         'info_dict': {
  26             'id': '122017',
  27             'ext': 'mp4',
  28             'title': 'Saison 2015 Épisode 17',
  29             'description': 'La photo de famille 2',
  30             'upload_date': '20100717',
  31         },
  32         'params': {
  33             # m3u8 download
  34             'skip_download': True,
  35         },
  36         'skip': '404 Not Found',
  37     }, {
  38         'url': 'http://ici.tou.tv/hackers',
  39         'only_matching': True,
  40     }]
  41
  42     def _real_initialize(self):
  43         email, password = self._get_login_info()
  44         if email is None:
  45             return
  46         state = 'http://ici.tou.tv/'
  47         webpage = self._download_webpage(state, None, 'Downloading homepage')
  48         toutvlogin = self._parse_json(self._search_regex(
  49             r'(?s)toutvlogin\s*=\s*({.+?});', webpage, 'toutvlogin'), None, js_to_json)
  50         authorize_url = toutvlogin['host'] + '/auth/oauth/v2/authorize'
  51         login_webpage = self._download_webpage(
  52             authorize_url, None, 'Downloading login page', query={
  53                 'client_id': toutvlogin['clientId'],
  54                 'redirect_uri': 'https://ici.tou.tv/login/loginCallback',
  55                 'response_type': 'token',
  56                 'scope': 'media-drmt openid profile email id.write media-validation.read.privileged',
  57                 'state': state,
  58             })
  59
  60         def extract_form_url_and_data(wp, default_form_url, form_spec_re=''):
  61             form, form_elem = re.search(
  62                 r'(?s)((<form[^>]+?%s[^>]*?>).+?</form>)' % form_spec_re, wp).groups()
  63             form_data = self._hidden_inputs(form)
  64             form_url = extract_attributes(form_elem).get('action') or default_form_url
  65             return form_url, form_data
  66
  67         post_url, form_data = extract_form_url_and_data(
  68             login_webpage,
  69             'https://services.radio-canada.ca/auth/oauth/v2/authorize/login',
  70             r'(?:id|name)="Form-login"')
  71         form_data.update({
  72             'login-email': email,
  73             'login-password': password,
  74         })
  75         consent_webpage = self._download_webpage(
  76             post_url, None, 'Logging in', data=urlencode_postdata(form_data))
  77         post_url, form_data = extract_form_url_and_data(
  78             consent_webpage,
  79             'https://services.radio-canada.ca/auth/oauth/v2/authorize/consent')
  80         _, urlh = self._download_webpage_handle(
  81             post_url, None, 'Following Redirection',
  82             data=urlencode_postdata(form_data))
  83         self._access_token = self._search_regex(
  84             r'access_token=([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
  85             urlh.geturl(), 'access token')
  86         self._claims = self._download_json(
  87             'https://services.radio-canada.ca/media/validation/v2/getClaims',
  88             None, 'Extracting Claims', query={
  89                 'token': self._access_token,
  90                 'access_token': self._access_token,
  91             })['claims']
  92
  93     def _real_extract(self, url):
  94         path = self._match_id(url)
  95         metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
  96         # IsDrm does not necessarily mean the video is DRM protected (see
  97         # https://github.com/rg3/youtube-dl/issues/13994).
  98         if metadata.get('IsDrm'):
  99             self.report_warning('This video is probably DRM protected.', path)
 100         video_id = metadata['IdMedia']
 101         details = metadata['Details']
 102         title = details['OriginalTitle']
 103         video_url = 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id)
 104         if self._access_token and self._claims:
 105             video_url = smuggle_url(video_url, {
 106                 'access_token': self._access_token,
 107                 'claims': self._claims,
 108             })
 109
 110         return {
 111             '_type': 'url_transparent',
 112             'url': video_url,
 113             'id': video_id,
 114             'title': title,
 115             'thumbnail': details.get('ImageUrl'),
 116             'duration': int_or_none(details.get('LengthInSeconds')),
 117         }