From d147a8ebbdf28c17cafbbe2884f0bc57e2bf82e2 Mon Sep 17 00:00:00 2001
From: Sarah Boyce <42296566+sarahboyce@users.noreply.github.com>
Date: Mon, 12 Aug 2024 15:17:57 +0200
Subject: [PATCH] [4.2.x] Fixed CVE-2024-45230 -- Mitigated potential DoS in
 urlize and urlizetrunc template filters.

Thanks MProgrammer (https://hackerone.com/mprogrammer) for the report.
---
 django/utils/html.py                          | 17 ++++++++------
 docs/ref/templates/builtins.txt               | 11 ++++++++++
 docs/releases/4.2.16.txt                      |  7 +++++-
 .../filter_tests/test_urlize.py               | 22 +++++++++++++++++++
 tests/utils_tests/test_html.py                |  1 +
 5 files changed, 50 insertions(+), 8 deletions(-)

diff --git a/django/utils/html.py b/django/utils/html.py
index 23575d3c11..df38c20519 100644
--- a/django/utils/html.py
+++ b/django/utils/html.py
@@ -395,14 +395,17 @@ class Urlizer:
                 potential_entity = middle[amp:]
                 escaped = html.unescape(potential_entity)
                 if escaped == potential_entity or escaped.endswith(";"):
-                    rstripped = middle.rstrip(";")
-                    amount_stripped = len(middle) - len(rstripped)
-                    if amp > -1 and amount_stripped > 1:
-                        # Leave a trailing semicolon as might be an entity.
-                        trail = middle[len(rstripped) + 1 :] + trail
-                        middle = rstripped + ";"
+                    rstripped = middle.rstrip(self.trailing_punctuation_chars)
+                    trail_start = len(rstripped)
+                    amount_trailing_semicolons = len(middle) - len(middle.rstrip(";"))
+                    if amp > -1 and amount_trailing_semicolons > 1:
+                        # Leave up to most recent semicolon as might be an entity.
+                        recent_semicolon = middle[trail_start:].index(";")
+                        middle_semicolon_index = recent_semicolon + trail_start + 1
+                        trail = middle[middle_semicolon_index:] + trail
+                        middle = rstripped + middle[trail_start:middle_semicolon_index]
                     else:
-                        trail = middle[len(rstripped) :] + trail
+                        trail = middle[trail_start:] + trail
                         middle = rstripped
                     trimmed_something = True
 
diff --git a/docs/ref/templates/builtins.txt b/docs/ref/templates/builtins.txt
index 39aa398338..dda5b42655 100644
--- a/docs/ref/templates/builtins.txt
+++ b/docs/ref/templates/builtins.txt
@@ -2831,6 +2831,17 @@ Django's built-in :tfilter:`escape` filter. The default value for
     email addresses that contain single quotes (``'``), things won't work as
     expected. Apply this filter only to plain text.
 
+.. warning::
+
+    Using ``urlize`` or ``urlizetrunc`` can incur a performance penalty, which
+    can become severe when applied to user controlled values such as content
+    stored in a :class:`~django.db.models.TextField`. You can use
+    :tfilter:`truncatechars` to add a limit to such inputs:
+
+    .. code-block:: html+django
+
+        {{ value|truncatechars:500|urlize }}
+
 .. templatefilter:: urlizetrunc
 
 ``urlizetrunc``
diff --git a/docs/releases/4.2.16.txt b/docs/releases/4.2.16.txt
index 9853004386..2a84186867 100644
--- a/docs/releases/4.2.16.txt
+++ b/docs/releases/4.2.16.txt
@@ -7,4 +7,9 @@ Django 4.2.16 release notes
 Django 4.2.16 fixes one security issue with severity "moderate" and one
 security issue with severity "low" in 4.2.15.
 
-...
+CVE-2024-45230: Potential denial-of-service vulnerability in ``django.utils.html.urlize()``
+===========================================================================================
+
+:tfilter:`urlize` and :tfilter:`urlizetrunc` were subject to a potential
+denial-of-service attack via very large inputs with a specific sequence of
+characters.
diff --git a/tests/template_tests/filter_tests/test_urlize.py b/tests/template_tests/filter_tests/test_urlize.py
index abc227ba6a..e542802aae 100644
--- a/tests/template_tests/filter_tests/test_urlize.py
+++ b/tests/template_tests/filter_tests/test_urlize.py
@@ -305,6 +305,28 @@ class FunctionTests(SimpleTestCase):
             "http://testing.com/example</a>.,:;)&quot;!",
         )
 
+    def test_trailing_semicolon(self):
+        self.assertEqual(
+            urlize("http://example.com?x=&amp;", autoescape=False),
+            '<a href="http://example.com?x=" rel="nofollow">'
+            "http://example.com?x=&amp;</a>",
+        )
+        self.assertEqual(
+            urlize("http://example.com?x=&amp;;", autoescape=False),
+            '<a href="http://example.com?x=" rel="nofollow">'
+            "http://example.com?x=&amp;</a>;",
+        )
+        self.assertEqual(
+            urlize("http://example.com?x=&amp;;;", autoescape=False),
+            '<a href="http://example.com?x=" rel="nofollow">'
+            "http://example.com?x=&amp;</a>;;",
+        )
+        self.assertEqual(
+            urlize("http://example.com?x=&amp.;...;", autoescape=False),
+            '<a href="http://example.com?x=" rel="nofollow">'
+            "http://example.com?x=&amp</a>.;...;",
+        )
+
     def test_brackets(self):
         """
         #19070 - Check urlize handles brackets properly
diff --git a/tests/utils_tests/test_html.py b/tests/utils_tests/test_html.py
index 83ebe4334b..7ff5020fb6 100644
--- a/tests/utils_tests/test_html.py
+++ b/tests/utils_tests/test_html.py
@@ -364,6 +364,7 @@ class TestUtilsHtml(SimpleTestCase):
             "&:" + ";" * 100_000,
             "&.;" * 100_000,
             ".;" * 100_000,
+            "&" + ";:" * 100_000,
         )
         for value in tests:
             with self.subTest(value=value):