Switch codebase to use sanitized_Request instead of

[youtube-dl] / youtube_dl / extractor / udemy.py

diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py
index cf7bd0ae5e8e3b9879066da7c30e07937471b438..825172806ab62b2edc28c205a299809023297281 100644 (file)
--- a/youtube_dl/extractor/udemy.py
+++ b/youtube_dl/extractor/udemy.py
@@ -3,17 +3,21 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
+)
+from ..utils import (
     ExtractorError,
+    sanitized_Request,
 )
 
 
 class UdemyIE(InfoExtractor):
     IE_NAME = 'udemy'
     _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
-    _LOGIN_URL = 'https://www.udemy.com/join/login-submit/'
+    _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
+    _ORIGIN_URL = 'https://www.udemy.com'
     _NETRC_MACHINE = 'udemy'
 
     _TESTS = [{
@@ -40,12 +44,7 @@ class UdemyIE(InfoExtractor):
                 error_str += ' - %s' % error_data.get('formErrors')
             raise ExtractorError(error_str, expected=True)
 
-    def _download_json(self, url, video_id, note='Downloading JSON metadata'):
-        response = super(UdemyIE, self)._download_json(url, video_id, note)
-        self._handle_error(response)
-        return response
-
-    def _download_json_cookies(self, url, video_id, note):
+    def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
         headers = {
             'X-Udemy-Snail-Case': 'true',
             'X-Requested-With': 'XMLHttpRequest',
@@ -55,8 +54,16 @@ class UdemyIE(InfoExtractor):
                 headers['X-Udemy-Client-Id'] = cookie.value
             elif cookie.name == 'access_token':
                 headers['X-Udemy-Bearer-Token'] = cookie.value
-        request = compat_urllib_request.Request(url, headers=headers)
-        return self._download_json(request, video_id, note)
+
+        if isinstance(url_or_request, compat_urllib_request.Request):
+            for header, value in headers.items():
+                url_or_request.add_header(header, value)
+        else:
+            url_or_request = sanitized_Request(url_or_request, headers=headers)
+
+        response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
+        self._handle_error(response)
+        return response
 
     def _real_initialize(self):
         self._login()
@@ -64,43 +71,45 @@ class UdemyIE(InfoExtractor):
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
-            raise ExtractorError(
-                'Udemy account is required, use --username and --password options to provide account credentials.',
-                expected=True)
+            self.raise_login_required('Udemy account is required')
 
         login_popup = self._download_webpage(
-            'https://www.udemy.com/join/login-popup?displayType=ajax&showSkipButton=1', None,
-            'Downloading login popup')
+            self._LOGIN_URL, None, 'Downloading login popup')
 
-        if login_popup == '<div class="run-command close-popup redirect" data-url="https://www.udemy.com/"></div>':
+        def is_logged(webpage):
+            return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
+
+        # already logged in
+        if is_logged(login_popup):
             return
 
-        csrf = self._html_search_regex(
-            r'<input type="hidden" name="csrf" value="(.+?)"',
-            login_popup, 'csrf token')
+        login_form = self._form_hidden_inputs('login-form', login_popup)
 
-        login_form = {
-            'email': username,
-            'password': password,
-            'csrf': csrf,
-            'displayType': 'json',
-            'isSubmitted': '1',
-        }
-        request = compat_urllib_request.Request(
-            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
-        response = self._download_json(
-            request, None, 'Logging in as %s' % username)
+        login_form.update({
+            'email': username.encode('utf-8'),
+            'password': password.encode('utf-8'),
+        })
 
-        if 'returnUrl' not in response:
-            raise ExtractorError('Unable to log in')
+        request = sanitized_Request(
+            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+        request.add_header('Referer', self._ORIGIN_URL)
+        request.add_header('Origin', self._ORIGIN_URL)
 
+        response = self._download_webpage(
+            request, None, 'Logging in as %s' % username)
 
+        if not is_logged(response):
+            error = self._html_search_regex(
+                r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
+                response, 'error message', default=None)
+            if error:
+                raise ExtractorError('Unable to login: %s' % error, expected=True)
+            raise ExtractorError('Unable to log in')
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        lecture_id = mobj.group('id')
+        lecture_id = self._match_id(url)
 
-        lecture = self._download_json_cookies(
+        lecture = self._download_json(
             'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id,
             lecture_id, 'Downloading lecture JSON')
 
@@ -164,7 +173,7 @@ class UdemyCourseIE(UdemyIE):
         mobj = re.match(self._VALID_URL, url)
         course_path = mobj.group('coursepath')
 
-        response = self._download_json_cookies(
+        response = self._download_json(
             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
             course_path, 'Downloading course JSON')
 
@@ -180,7 +189,7 @@ class UdemyCourseIE(UdemyIE):
         elif self._ALREADY_ENROLLED in webpage:
             self.to_screen('%s: Already enrolled in' % course_id)
 
-        response = self._download_json_cookies(
+        response = self._download_json(
             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
             course_id, 'Downloading course curriculum')