[toutv] fix authentication(closes #16398)(closes #18700)

youtube_dl/extractor/radiocanada.py

@@ -4,12 +4,12 @@
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
     determine_ext,
     ExtractorError,
     int_or_none,
     unified_strdate,
-    unsmuggle_url,
 )
 
 
@@ -58,23 +58,35 @@ class RadioCanadaIE(InfoExtractor):
         }
     ]
     _GEO_COUNTRIES = ['CA']
+    _access_token = None
+    _claims = None
 
-    def _call_api(self, path, video_id, app_code, query):
+    def _call_api(self, path, video_id=None, app_code=None, query=None):
+        if not query:
+            query = {}
         query.update({
-            'appCode': app_code,
+            'client_key': '773aea60-0e80-41bb-9c7f-e6d7c3ad17fb',
-            'idMedia': video_id,
             'output': 'json',
         })
-        return self._download_json(
+        if video_id:
-            'https://services.radio-canada.ca/media/' + path, video_id, headers={
+            query.update({
-                'Authorization': 'Client-Key 773aea60-0e80-41bb-9c7f-e6d7c3ad17fb'
+                'appCode': app_code,
-            }, query=query)
+                'idMedia': video_id,
+            })
+        if self._access_token:
+            query['access_token'] = self._access_token
+        try:
+            return self._download_json(
+                'https://services.radio-canada.ca/media/' + path, video_id, query=query)
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 422):
+                data = self._parse_json(e.cause.read().decode(), None)
+                error = data.get('error_description') or data['errorMessage']['text']
+                raise ExtractorError(error, expected=True)
+            raise
 
-    def _real_extract(self, url):
+    def _extract_info(self, app_code, video_id):
-        url, smuggled_data = unsmuggle_url(url, {})
+        metas = self._call_api('meta/v1/index.ashx', video_id, app_code)['Metas']
-        app_code, video_id = re.match(self._VALID_URL, url).groups()
-
-        metas = self._call_api('meta/v1/index.ashx', video_id, app_code, {})['Metas']
 
         def get_meta(name):
             for meta in metas:
@@ -93,14 +105,16 @@ def get_meta(name):
             'deviceType': 'ipad',
             'multibitrate': 'true',
         }
-        if smuggled_data:
+        if self._claims:
-            query.update(smuggled_data)
+            query['claims'] = self._claims
         v_data = self._call_api('validation/v2/', video_id, app_code, query)
         v_url = v_data.get('url')
         if not v_url:
             error = v_data['message']
             if error == "Le contenu sélectionné n'est pas disponible dans votre pays":
                 raise self.raise_geo_restricted(error, self._GEO_COUNTRIES)
+            if error == 'Le contenu sélectionné est disponible seulement en premium':
+                self.raise_login_required(error)
             raise ExtractorError(
                 '%s said: %s' % (self.IE_NAME, error), expected=True)
         formats = self._extract_m3u8_formats(v_url, video_id, 'mp4')
@@ -128,6 +142,9 @@ def get_meta(name):
             'formats': formats,
         }
 
+    def _real_extract(self, url):
+        return self._extract_info(*re.match(self._VALID_URL, url).groups())
+
 
 class RadioCanadaAudioVideoIE(InfoExtractor):
     'radiocanada:audiovideo'

youtube_dl/extractor/toutv.py

@@ -3,22 +3,19 @@
 
 import re
 
-from .common import InfoExtractor
+from .radiocanada import RadioCanadaIE
 from ..utils import (
-    int_or_none,
-    js_to_json,
-    urlencode_postdata,
     extract_attributes,
-    smuggle_url,
+    int_or_none,
+    merge_dicts,
+    urlencode_postdata,
 )
 
 
-class TouTvIE(InfoExtractor):
+class TouTvIE(RadioCanadaIE):
     _NETRC_MACHINE = 'toutv'
     IE_NAME = 'tou.tv'
     _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/S[0-9]+[EC][0-9]+)?)'
-    _access_token = None
-    _claims = None
 
     _TESTS = [{
         'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
@@ -46,18 +43,14 @@ def _real_initialize(self):
         email, password = self._get_login_info()
         if email is None:
             return
-        state = 'http://ici.tou.tv/'
-        webpage = self._download_webpage(state, None, 'Downloading homepage')
-        toutvlogin = self._parse_json(self._search_regex(
-            r'(?s)toutvlogin\s*=\s*({.+?});', webpage, 'toutvlogin'), None, js_to_json)
-        authorize_url = toutvlogin['host'] + '/auth/oauth/v2/authorize'
         login_webpage = self._download_webpage(
-            authorize_url, None, 'Downloading login page', query={
+            'https://services.radio-canada.ca/auth/oauth/v2/authorize',
-                'client_id': toutvlogin['clientId'],
+            None, 'Downloading login page', query={
-                'redirect_uri': 'https://ici.tou.tv/login/loginCallback',
+                'client_id': '4dd36440-09d5-4468-8923-b6d91174ad36',
+                'redirect_uri': 'https://ici.tou.tv/logincallback',
                 'response_type': 'token',
-                'scope': 'media-drmt openid profile email id.write media-validation.read.privileged',
+                'scope': 'id.write media-validation.read',
-                'state': state,
+                'state': '/',
             })
 
         def extract_form_url_and_data(wp, default_form_url, form_spec_re=''):
@@ -86,12 +79,7 @@ def extract_form_url_and_data(wp, default_form_url, form_spec_re=''):
         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')
-        self._claims = self._download_json(
+        self._claims = self._call_api('validation/v2/getClaims')['claims']
-            'https://services.radio-canada.ca/media/validation/v2/getClaims',
-            None, 'Extracting Claims', query={
-                'token': self._access_token,
-                'access_token': self._access_token,
-            })['claims']
 
     def _real_extract(self, url):
         path = self._match_id(url)
@@ -102,19 +90,10 @@ def _real_extract(self, url):
             self.report_warning('This video is probably DRM protected.', path)
         video_id = metadata['IdMedia']
         details = metadata['Details']
-        title = details['OriginalTitle']
-        video_url = 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id)
-        if self._access_token and self._claims:
-            video_url = smuggle_url(video_url, {
-                'access_token': self._access_token,
-                'claims': self._claims,
-            })
 
-        return {
+        return merge_dicts({
-            '_type': 'url_transparent',
-            'url': video_url,
             'id': video_id,
-            'title': title,
+            'title': details.get('OriginalTitle'),
             'thumbnail': details.get('ImageUrl'),
             'duration': int_or_none(details.get('LengthInSeconds')),
-        }
+        }, self._extract_info(metadata.get('AppCode', 'toutv'), video_id))