[toutv] add support for authentication(closes #10669)

youtube_dl/extractor/radiocanada.py

@@ -13,6 +13,7 @@
     xpath_element,
     ExtractorError,
     determine_protocol,
+    unsmuggle_url,
 )
 
 
@@ -35,28 +36,51 @@ class RadioCanadaIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
         app_code, video_id = re.match(self._VALID_URL, url).groups()
 
-        device_types = ['ipad', 'android']
+        metadata = self._download_xml(
+            'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
+            video_id, note='Downloading metadata XML', query={
+                'appCode': app_code,
+                'idMedia': video_id,
+            })
+
+        def get_meta(name):
+            el = find_xpath_attr(metadata, './/Meta', 'name', name)
+            return el.text if el is not None else None
+
+        if get_meta('protectionType'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        device_types = ['ipad']
         if app_code != 'toutv':
             device_types.append('flash')
+        if not smuggled_data:
+            device_types.append('android')
 
         formats = []
         # TODO: extract f4m formats
         # f4m formats can be extracted using flashhd device_type but they produce unplayable file
         for device_type in device_types:
-            v_data = self._download_xml(
-                'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
-                video_id, note='Downloading %s XML' % device_type, query={
-                    'appCode': app_code,
-                    'idMedia': video_id,
-                    'connectionType': 'broadband',
-                    'multibitrate': 'true',
-                    'deviceType': device_type,
+            validation_url = 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx'
+            query = {
+                'appCode': app_code,
+                'idMedia': video_id,
+                'connectionType': 'broadband',
+                'multibitrate': 'true',
+                'deviceType': device_type,
+            }
+            if smuggled_data:
+                validation_url = 'https://services.radio-canada.ca/media/validation/v2/'
+                query.update(smuggled_data)
+            else:
+                query.update({
                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
                     'paysJ391wsHjbOJwvCs26toz': 'CA',
                     'bypasslock': 'NZt5K62gRqfc',
-                }, fatal=False)
+                })
+            v_data = self._download_xml(validation_url, video_id, note='Downloading %s XML' % device_type, query=query, fatal=False)
             v_url = xpath_text(v_data, 'url')
             if not v_url:
                 continue
@@ -101,17 +125,6 @@ def _real_extract(self, url):
                                 f4m_id='hds', fatal=False))
         self._sort_formats(formats)
 
-        metadata = self._download_xml(
-            'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
-            video_id, note='Downloading metadata XML', query={
-                'appCode': app_code,
-                'idMedia': video_id,
-            })
-
-        def get_meta(name):
-            el = find_xpath_attr(metadata, './/Meta', 'name', name)
-            return el.text if el is not None else None
-
         return {
             'id': video_id,
             'title': get_meta('Title'),

youtube_dl/extractor/toutv.py

@@ -2,12 +2,22 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    ExtractorError,
+    urlencode_postdata,
+    extract_attributes,
+    smuggle_url,
+)
 
 
 class TouTvIE(InfoExtractor):
+    _NETRC_MACHINE = 'toutv'
     IE_NAME = 'tou.tv'
     _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+/S[0-9]+E[0-9]+)'
+    _access_token = None
+    _claims = None
 
     _TEST = {
         'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
@@ -22,18 +32,64 @@ class TouTvIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
+        'skip': '404 Not Found',
     }
 
+    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={
+                'client_id': toutvlogin['clientId'],
+                'redirect_uri': 'https://ici.tou.tv/login/loginCallback',
+                'response_type': 'token',
+                'scope': 'media-drmt openid profile email id.write media-validation.read.privileged',
+                'state': state,
+            })
+        login_form = self._search_regex(
+            r'(?s)(<form[^>]+id="Form-login".+?</form>)', login_webpage, 'login form')
+        form_data = self._hidden_inputs(login_form)
+        form_data.update({
+            'login-email': email,
+            'login-password': password,
+        })
+        post_url = extract_attributes(login_form).get('action') or authorize_url
+        _, urlh = self._download_webpage_handle(
+            post_url, None, 'Logging in', 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')
+        self._claims = self._download_json(
+            '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)
         metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
+        if metadata.get('IsDrm'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
         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 {
             '_type': 'url_transparent',
-            'url': 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id),
+            'url': video_url,
             'id': video_id,
             'title': title,
             'thumbnail': details.get('ImageUrl'),