Start moving to ytdl-org
[youtube-dl] / youtube_dl / extractor / toutv.py
index f1ab91cf2bc3b4b7f29627f19c2c1163a7f9e85c..8cc57b919bcecbdd49a0adf76b839e2c923d0bb4 100644 (file)
@@ -1,14 +1,14 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
+import json
 
 from .radiocanada import RadioCanadaIE
+from ..compat import compat_HTTPError
 from ..utils import (
-    extract_attributes,
+    ExtractorError,
     int_or_none,
     merge_dicts,
-    urlencode_postdata,
 )
 
 
@@ -38,54 +38,37 @@ class TouTvIE(RadioCanadaIE):
         'url': 'https://ici.tou.tv/l-age-adulte/S01C501',
         'only_matching': True,
     }]
+    _CLIENT_KEY = '4dd36440-09d5-4468-8923-b6d91174ad36'
 
     def _real_initialize(self):
         email, password = self._get_login_info()
         if email is None:
             return
-        login_webpage = self._download_webpage(
-            'https://services.radio-canada.ca/auth/oauth/v2/authorize',
-            None, 'Downloading login page', query={
-                'client_id': '4dd36440-09d5-4468-8923-b6d91174ad36',
-                'redirect_uri': 'https://ici.tou.tv/logincallback',
-                'response_type': 'token',
-                'scope': 'id.write media-validation.read',
-                'state': '/',
-            })
-
-        def extract_form_url_and_data(wp, default_form_url, form_spec_re=''):
-            form, form_elem = re.search(
-                r'(?s)((<form[^>]+?%s[^>]*?>).+?</form>)' % form_spec_re, wp).groups()
-            form_data = self._hidden_inputs(form)
-            form_url = extract_attributes(form_elem).get('action') or default_form_url
-            return form_url, form_data
-
-        post_url, form_data = extract_form_url_and_data(
-            login_webpage,
-            'https://services.radio-canada.ca/auth/oauth/v2/authorize/login',
-            r'(?:id|name)="Form-login"')
-        form_data.update({
-            'login-email': email,
-            'login-password': password,
-        })
-        consent_webpage = self._download_webpage(
-            post_url, None, 'Logging in', data=urlencode_postdata(form_data))
-        post_url, form_data = extract_form_url_and_data(
-            consent_webpage,
-            'https://services.radio-canada.ca/auth/oauth/v2/authorize/consent')
-        _, urlh = self._download_webpage_handle(
-            post_url, None, 'Following Redirection',
-            data=urlencode_postdata(form_data))
-        self._access_token = self._search_regex(
-            r'access_token=([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
-            urlh.geturl(), 'access token')
+        try:
+            self._access_token = self._download_json(
+                'https://services.radio-canada.ca/toutv/profiling/accounts/login',
+                None, 'Logging in', data=json.dumps({
+                    'ClientId': self._CLIENT_KEY,
+                    'ClientSecret': '34026772-244b-49b6-8b06-317b30ac9a20',
+                    'Email': email,
+                    'Password': password,
+                    'Scope': 'id.write media-validation.read',
+                }).encode(), headers={
+                    'Authorization': 'client-key ' + self._CLIENT_KEY,
+                    'Content-Type': 'application/json;charset=utf-8',
+                })['access_token']
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                error = self._parse_json(e.cause.read().decode(), None)['Message']
+                raise ExtractorError(error, expected=True)
+            raise
         self._claims = self._call_api('validation/v2/getClaims')['claims']
 
     def _real_extract(self, url):
         path = self._match_id(url)
         metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
         # IsDrm does not necessarily mean the video is DRM protected (see
-        # https://github.com/rg3/youtube-dl/issues/13994).
+        # https://github.com/ytdl-org/youtube-dl/issues/13994).
         if metadata.get('IsDrm'):
             self.report_warning('This video is probably DRM protected.', path)
         video_id = metadata['IdMedia']