gh-105704: Disallow square brackets ([ and ]) in domain names for parsed URLs (GH-129418)

* gh-105704: Disallow square brackets ( and ) in domain names for parsed URLs

* Use Sphinx references

Co-authored-by: Peter Bierma <[email protected]>

* Add mismatched bracket test cases, fix news format

* Add more test coverage for ports

---------

(cherry picked from commit d89a5f6)

Co-authored-by: Seth Michael Larson <[email protected]>
Co-authored-by: Peter Bierma <[email protected]>

Author: 2 people

Lib/test/test_urlparse.py

@@ -1273,16 +1273,51 @@ def test_invalid_bracketed_hosts(self):
         self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[0439:23af::2309::fae7:1234]/Path?Query')
         self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[0439:23af:2309::fae7:1234:2342:438e:192.0.2.146]/Path?Query')
         self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@]v6a.ip[/Path')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[v6a.ip]')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[v6a.ip].suffix')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[v6a.ip]/')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[v6a.ip].suffix/')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[v6a.ip]?')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[v6a.ip].suffix?')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[::1].suffix')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]/')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[::1].suffix/')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]?')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[::1].suffix?')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]:a')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[::1].suffix:a')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]:a1')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[::1].suffix:a1')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]:1a')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[::1].suffix:1a')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]:')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[::1].suffix:/')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[::1]:?')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://user@prefix.[v6a.ip]')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://user@[v6a.ip].suffix')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://[v6a.ip')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://v6a.ip]')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://]v6a.ip[')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://]v6a.ip')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://v6a.ip[')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix.[v6a.ip')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://v6a.ip].suffix')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix]v6a.ip[suffix')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://prefix]v6a.ip')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'scheme://v6a.ip[suffix')
 
     def test_splitting_bracketed_hosts(self):
-        p1 = urllib.parse.urlsplit('scheme://user@[v6a.ip]/path?query')
+        p1 = urllib.parse.urlsplit('scheme://user@[v6a.ip]:1234/path?query')
         self.assertEqual(p1.hostname, 'v6a.ip')
         self.assertEqual(p1.username, 'user')
         self.assertEqual(p1.path, '/path')
+        self.assertEqual(p1.port, 1234)
         p2 = urllib.parse.urlsplit('scheme://user@[0439:23af:2309::fae7%test]/path?query')
         self.assertEqual(p2.hostname, '0439:23af:2309::fae7%test')
         self.assertEqual(p2.username, 'user')
         self.assertEqual(p2.path, '/path')
+        self.assertIs(p2.port, None)
         p3 = urllib.parse.urlsplit('scheme://user@[0439:23af:2309::fae7:1234:192.0.2.146%test]/path?query')
         self.assertEqual(p3.hostname, '0439:23af:2309::fae7:1234:192.0.2.146%test')
         self.assertEqual(p3.username, 'user')

Lib/urllib/parse.py

@@ -436,6 +436,23 @@ def _checknetloc(netloc):
             raise ValueError("netloc '" + netloc + "' contains invalid " +
                              "characters under NFKC normalization")
 
+def _check_bracketed_netloc(netloc):
+    # Note that this function must mirror the splitting
+    # done in NetlocResultMixins._hostinfo().
+    hostname_and_port = netloc.rpartition('@')[2]
+    before_bracket, have_open_br, bracketed = hostname_and_port.partition('[')
+    if have_open_br:
+        # No data is allowed before a bracket.
+        if before_bracket:
+            raise ValueError("Invalid IPv6 URL")
+        hostname, _, port = bracketed.partition(']')
+        # No data is allowed after the bracket but before the port delimiter.
+        if port and not port.startswith(":"):
+            raise ValueError("Invalid IPv6 URL")
+    else:
+        hostname, _, port = hostname_and_port.partition(':')
+    _check_bracketed_host(hostname)
+
 # Valid bracketed hosts are defined in
 # https://www.rfc-editor.org/rfc/rfc3986#page-49 and https://url.spec.whatwg.org/
 def _check_bracketed_host(hostname):
@@ -496,8 +513,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
                 (']' in netloc and '[' not in netloc)):
             raise ValueError("Invalid IPv6 URL")
         if '[' in netloc and ']' in netloc:
-            bracketed_host = netloc.partition('[')[2].partition(']')[0]
-            _check_bracketed_host(bracketed_host)
+            _check_bracketed_netloc(netloc)
     if allow_fragments and '#' in url:
         url, fragment = url.split('#', 1)
     if '?' in url:

Misc/NEWS.d/next/Security/2025-01-28-14-08-03.gh-issue-105704.EnhHxu.rst

@@ -0,0 +1,4 @@
+When using :func:`urllib.parse.urlsplit` and :func:`urllib.parse.urlparse` host
+parsing would not reject domain names containing square brackets (``[`` and
+``]``). Square brackets are only valid for IPv6 and IPvFuture hosts according to
+`RFC 3986 Section 3.2.2 <https://www.rfc-editor.org/rfc/rfc3986#section-3.2.2>`__.