python-Django/CVE-2024-41991.patch

From 772a73f70c3d249c99c23012849e66276b7b0715 Mon Sep 17 00:00:00 2001
From: Mariusz Felisiak <felisiak.mariusz@gmail.com>
Date: Wed, 10 Jul 2024 20:30:12 +0200
Subject: [PATCH 3/4] [4.2.x] Fixed CVE-2024-41991 -- Prevented potential ReDoS
 in django.utils.html.urlize() and AdminURLFieldWidget.

Thanks Seokchan Yoon for the report.

Co-authored-by: Sarah Boyce <42296566+sarahboyce@users.noreply.github.com>
---
 django/contrib/admin/widgets.py |  2 +-
 django/utils/html.py            | 10 ++++++++--
 tests/admin_widgets/tests.py    |  7 ++++++-
 tests/utils_tests/test_html.py  | 13 +++++++++++++
 5 files changed, 35 insertions(+), 4 deletions(-)

diff --git a/django/contrib/admin/widgets.py b/django/contrib/admin/widgets.py
index 5e3416bc28..3d11a40efe 100644
--- a/django/contrib/admin/widgets.py
+++ b/django/contrib/admin/widgets.py
@@ -383,7 +383,7 @@ class AdminURLFieldWidget(forms.URLInput):
         context["current_label"] = _("Currently:")
         context["change_label"] = _("Change:")
         context["widget"]["href"] = (
-            smart_urlquote(context["widget"]["value"]) if value else ""
+            smart_urlquote(context["widget"]["value"]) if url_valid else ""
         )
         context["url_valid"] = url_valid
         return context
diff --git a/django/utils/html.py b/django/utils/html.py
index dd52f1f7fe..23575d3c11 100644
--- a/django/utils/html.py
+++ b/django/utils/html.py
@@ -13,6 +13,8 @@ from django.utils.regex_helper import _lazy_re_compile
 from django.utils.safestring import SafeData, SafeString, mark_safe
 from django.utils.text import normalize_newlines
 
+MAX_URL_LENGTH = 2048
+
 
 @keep_lazy(SafeString)
 def escape(text):
@@ -300,9 +302,9 @@ class Urlizer:
             # Make URL we want to point to.
             url = None
             nofollow_attr = ' rel="nofollow"' if nofollow else ""
-            if self.simple_url_re.match(middle):
+            if len(middle) <= MAX_URL_LENGTH and self.simple_url_re.match(middle):
                 url = smart_urlquote(html.unescape(middle))
-            elif self.simple_url_2_re.match(middle):
+            elif len(middle) <= MAX_URL_LENGTH and self.simple_url_2_re.match(middle):
                 url = smart_urlquote("http://%s" % html.unescape(middle))
             elif ":" not in middle and self.is_email_simple(middle):
                 local, domain = middle.rsplit("@", 1)
@@ -417,6 +419,10 @@ class Urlizer:
         except ValueError:
             # value contains more than one @.
             return False
+        # Max length for domain name labels is 63 characters per RFC 1034.
+        # Helps to avoid ReDoS vectors in the domain part.
+        if len(p2) > 63:
+            return False
         # Dot must be in p2 (e.g. example.com)
         if "." not in p2 or p2.startswith("."):
             return False
diff --git a/tests/admin_widgets/tests.py b/tests/admin_widgets/tests.py
index 0e20206048..4281ed07c6 100644
--- a/tests/admin_widgets/tests.py
+++ b/tests/admin_widgets/tests.py
@@ -461,7 +461,12 @@ class AdminSplitDateTimeWidgetTest(SimpleTestCase):
 class AdminURLWidgetTest(SimpleTestCase):
     def test_get_context_validates_url(self):
         w = widgets.AdminURLFieldWidget()
-        for invalid in ["", "/not/a/full/url/", 'javascript:alert("Danger XSS!")']:
+        for invalid in [
+            "",
+            "/not/a/full/url/",
+            'javascript:alert("Danger XSS!")',
+            "http://" + "한.글." * 1_000_000 + "com",
+        ]:
             with self.subTest(url=invalid):
                 self.assertFalse(w.get_context("name", invalid, {})["url_valid"])
         self.assertTrue(w.get_context("name", "http://example.com", {})["url_valid"])
diff --git a/tests/utils_tests/test_html.py b/tests/utils_tests/test_html.py
index c45e0dfac1..83ebe4334b 100644
--- a/tests/utils_tests/test_html.py
+++ b/tests/utils_tests/test_html.py
@@ -328,6 +328,15 @@ class TestUtilsHtml(SimpleTestCase):
                 'Search for <a href="http://google.com/?q=">google.com/?q=</a>!',
             ),
             ("foo@example.com", '<a href="mailto:foo@example.com">foo@example.com</a>'),
+            (
+                "test@" + "한.글." * 15 + "aaa",
+                '<a href="mailto:test@'
+                + "xn--6q8b.xn--bj0b." * 15
+                + 'aaa">'
+                + "test@"
+                + "한.글." * 15
+                + "aaa</a>",
+            ),
         )
         for value, output in tests:
             with self.subTest(value=value):
@@ -336,6 +345,10 @@ class TestUtilsHtml(SimpleTestCase):
     def test_urlize_unchanged_inputs(self):
         tests = (
             ("a" + "@a" * 50000) + "a",  # simple_email_re catastrophic test
+            # Unicode domain catastrophic tests.
+            "a@" + "한.글." * 1_000_000 + "a",
+            "http://" + "한.글." * 1_000_000 + "com",
+            "www." + "한.글." * 1_000_000 + "com",
             ("a" + "." * 1000000) + "a",  # trailing_punctuation catastrophic test
             "foo@",
             "@foo.com",
-- 
2.34.1