From 255449c1ee61c14778658caae8c430fa4d76afd6 Mon Sep 17 00:00:00 2001
From: Erik Romijn <eromijn@solidlinks.nl>
Date: Mon, 12 May 2014 07:38:39 -0400
Subject: [PATCH] Added additional checks in is_safe_url to account for
 flexible parsing.

This is a security fix. Disclosure following shortly.
---
 django/contrib/auth/tests/test_views.py | 12 ++++++----
 django/utils/http.py                    | 12 ++++++++++
 tests/utils_tests/test_http.py          | 30 +++++++++++++++++++++++++
 3 files changed, 50 insertions(+), 4 deletions(-)

diff --git a/django/contrib/auth/tests/test_views.py b/django/contrib/auth/tests/test_views.py
index a5be9ffff1..dcfda61ca9 100644
--- a/django/contrib/auth/tests/test_views.py
+++ b/django/contrib/auth/tests/test_views.py
@@ -482,8 +482,10 @@ class LoginTest(AuthViewsTestCase):
 
         # Those URLs should not pass the security check
         for bad_url in ('http://example.com',
+                        'http:///example.com',
                         'https://example.com',
                         'ftp://exampel.com',
+                        '///example.com',
                         '//example.com',
                         'javascript:alert("XSS")'):
 
@@ -505,8 +507,8 @@ class LoginTest(AuthViewsTestCase):
                          '/view/?param=https://example.com',
                          '/view?param=ftp://exampel.com',
                          'view/?param=//example.com',
-                         'https:///',
-                         'HTTPS:///',
+                         'https://testserver/',
+                         'HTTPS://testserver/',
                          '//testserver/',
                          '/url%20with%20spaces/'):  # see ticket #12534
             safe_url = '%(url)s?%(next)s=%(good_url)s' % {
@@ -743,8 +745,10 @@ class LogoutTest(AuthViewsTestCase):
 
         # Those URLs should not pass the security check
         for bad_url in ('http://example.com',
+                        'http:///example.com',
                         'https://example.com',
                         'ftp://exampel.com',
+                        '///example.com',
                         '//example.com',
                         'javascript:alert("XSS")'):
             nasty_url = '%(url)s?%(next)s=%(bad_url)s' % {
@@ -764,8 +768,8 @@ class LogoutTest(AuthViewsTestCase):
                          '/view/?param=https://example.com',
                          '/view?param=ftp://exampel.com',
                          'view/?param=//example.com',
-                         'https:///',
-                         'HTTPS:///',
+                         'https://testserver/',
+                         'HTTPS://testserver/',
                          '//testserver/',
                          '/url%20with%20spaces/'):  # see ticket #12534
             safe_url = '%(url)s?%(next)s=%(good_url)s' % {
diff --git a/django/utils/http.py b/django/utils/http.py
index e3365776f3..28245d23f4 100644
--- a/django/utils/http.py
+++ b/django/utils/http.py
@@ -272,6 +272,18 @@ def is_safe_url(url, host=None):
     """
     if not url:
         return False
+    # Chrome treats \ completely as /
+    url = url.replace('\\', '/')
+    # Chrome considers any URL with more than two slashes to be absolute, but
+    # urlaprse is not so flexible. Treat any url with three slashes as unsafe.
+    if url.startswith('///'):
+        return False
     url_info = urlparse(url)
+    # Forbid URLs like http:///example.com - with a scheme, but without a hostname.
+    # In that URL, example.com is not the hostname but, a path component. However,
+    # Chrome will still consider example.com to be the hostname, so we must not
+    # allow this syntax.
+    if not url_info.netloc and url_info.scheme:
+        return False
     return ((not url_info.netloc or url_info.netloc == host) and
             (not url_info.scheme or url_info.scheme in ['http', 'https']))
diff --git a/tests/utils_tests/test_http.py b/tests/utils_tests/test_http.py
index edd0fde44f..eec439a11c 100644
--- a/tests/utils_tests/test_http.py
+++ b/tests/utils_tests/test_http.py
@@ -89,6 +89,36 @@ class TestUtilsHttp(unittest.TestCase):
             self.assertEqual(http.int_to_base36(n), b36)
             self.assertEqual(http.base36_to_int(b36), n)
 
+    def test_is_safe_url(self):
+        for bad_url in ('http://example.com',
+                        'http:///example.com',
+                        'https://example.com',
+                        'ftp://exampel.com',
+                        r'\\example.com',
+                        r'\\\example.com',
+                        r'/\\/example.com',
+                        r'\\\example.com',
+                        r'\\example.com',
+                        r'\\//example.com',
+                        r'/\/example.com',
+                        r'\/example.com',
+                        r'/\example.com',
+                        'http:///example.com',
+                        'http:/\//example.com',
+                        'http:\/example.com',
+                        'http:/\example.com',
+                        'javascript:alert("XSS")'):
+            self.assertFalse(http.is_safe_url(bad_url, host='testserver'), "%s should be blocked" % bad_url)
+        for good_url in ('/view/?param=http://example.com',
+                     '/view/?param=https://example.com',
+                     '/view?param=ftp://exampel.com',
+                     'view/?param=//example.com',
+                     'https://testserver/',
+                     'HTTPS://testserver/',
+                     '//testserver/',
+                     '/url%20with%20spaces/'):
+            self.assertTrue(http.is_safe_url(good_url, host='testserver'), "%s should be allowed" % good_url)
+
 
 class ETagProcessingTests(unittest.TestCase):
     def testParsing(self):