[4.2.x] Fixed CVE-2026-25673 -- Simplified URLField scheme detection.

This simplicaftion mitigates a potential DoS in URLField on Windows. The
usage of `urlsplit()` in `URLField.to_python()` was replaced with
`str.partition(":")` for URL scheme detection. On Windows, `urlsplit()`
performs Unicode normalization which is slow for certain characters,
making `URLField` vulnerable to DoS via specially crafted POST payloads.

Thanks Seokchan Yoon for the report, and Jake Howard and Shai Berger
for the review.

Refs #36923.

Co-authored-by: Jacob Walls <jacobtylerwalls@gmail.com>

Backport of 951ffb3832cd83ba672c1e3deae2bda128eb9cca from main.

1 files changed, 16 insertions, 26 deletions

diff --git a/django/forms/fields.py b/django/forms/fields.py
index e62417f552..5c2e35091b 100644
--- a/django/forms/fields.py
+++ b/django/forms/fields.py
@@ -12,7 +12,6 @@ import re
 import uuid
 from decimal import Decimal, DecimalException
 from io import BytesIO
-from urllib.parse import urlsplit, urlunsplit
 
 from django.core import validators
 from django.core.exceptions import ValidationError
@@ -757,33 +756,24 @@ class URLField(CharField):
         super().__init__(strip=True, **kwargs)
 
     def to_python(self, value):
-        def split_url(url):
-            """
-            Return a list of url parts via urlparse.urlsplit(), or raise
-            ValidationError for some malformed URLs.
-            """
-            try:
-                return list(urlsplit(url))
-            except ValueError:
-                # urlparse.urlsplit can raise a ValueError with some
-                # misformatted URLs.
-                raise ValidationError(self.error_messages["invalid"], code="invalid")
-
         value = super().to_python(value)
         if value:
-            url_fields = split_url(value)
-            if not url_fields[0]:
-                # If no URL scheme given, assume http://
-                url_fields[0] = "http"
-            if not url_fields[1]:
-                # Assume that if no domain is provided, that the path segment
-                # contains the domain.
-                url_fields[1] = url_fields[2]
-                url_fields[2] = ""
-                # Rebuild the url_fields list, since the domain segment may now
-                # contain the path too.
-                url_fields = split_url(urlunsplit(url_fields))
-            value = urlunsplit(url_fields)
+            # Detect scheme via partition to avoid calling urlsplit() on
+            # potentially large or slow-to-normalize inputs.
+            scheme, sep, _ = value.partition(":")
+            if (
+                not sep
+                or not scheme
+                or not scheme[0].isascii()
+                or not scheme[0].isalpha()
+                or "/" in scheme
+            ):
+                # No valid scheme found -- prepend the assumed scheme. Handle
+                # scheme-relative URLs ("//example.com") separately.
+                if value.startswith("//"):
+                    value = "http:" + value
+                else:
+                    value = "http://" + value
         return value