00444: pythongh-103848: Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format (pythonGH-103849) (python#104349)

pythongh-103848: Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format (pythonGH-103849)

* Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format

---------

(cherry picked from commit 29f348e)

Co-authored-by: JohnJamesUtley <[email protected]>
Co-authored-by: Gregory P. Smith <[email protected]>

Author: 3 people

Lib/test/test_urlparse.py

@@ -1076,6 +1076,32 @@ def test_issue14072(self):
         self.assertEqual(p2.scheme, 'tel')
         self.assertEqual(p2.path, '+31641044153')
 
+    def test_invalid_bracketed_hosts(self):
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[192.0.2.146]/Path?Query')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[important.com:8000]/Path?Query')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[v123r.IP]/Path?Query')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[v12ae]/Path?Query')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[v.IP]/Path?Query')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[v123.]/Path?Query')
+        self.assertRaises(ValueError, urllib.parse.urlsplit, 'Scheme://user@[v]/Path?Query')
+        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')
+
+    def test_splitting_bracketed_hosts(self):
+        p1 = urllib.parse.urlsplit('scheme://user@[v6a.ip]/path?query')
+        self.assertEqual(p1.hostname, 'v6a.ip')
+        self.assertEqual(p1.username, 'user')
+        self.assertEqual(p1.path, '/path')
+        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')
+        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')
+        self.assertEqual(p3.path, '/path')
+
     def test_telurl_params(self):
         p1 = urllib.parse.urlparse('tel:123-4;phone-context=+1-650-516')
         self.assertEqual(p1.scheme, 'tel')

Lib/urllib/parse.py

@@ -33,7 +33,13 @@
 
 import re
 import sys
+<<<<<<< HEAD
 import collections
+=======
+import types
+import warnings
+import ipaddress
+>>>>>>> b2171a2fd41 ([3.11] gh-103848: Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format (GH-103849) (#104349))
 
 __all__ = ["urlparse", "urlunparse", "urljoin", "urldefrag",
            "urlsplit", "urlunsplit", "urlencode", "parse_qs",
@@ -425,6 +431,17 @@ def _remove_unsafe_bytes_from_url(url):
         url = url.replace(b, "")
     return url
 
+# 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):
+    if hostname.startswith('v'):
+        if not re.match(r"\Av[a-fA-F0-9]+\..+\Z", hostname):
+            raise ValueError(f"IPvFuture address is invalid")
+    else:
+        ip = ipaddress.ip_address(hostname) # Throws Value Error if not IPv6 or IPv4
+        if isinstance(ip, ipaddress.IPv4Address):
+            raise ValueError(f"An IPv4 address cannot be in brackets")
+
 def urlsplit(url, scheme='', allow_fragments=True):
     """Parse a URL into 5 components:
     <scheme>://<netloc>/<path>?<query>#<fragment>
@@ -468,18 +485,25 @@ def urlsplit(url, scheme='', allow_fragments=True):
             if c not in scheme_chars:
                 break
         else:
+<<<<<<< HEAD
             # make sure "url" is not actually a port number (in which case
             # "scheme" is really part of the path)
             rest = url[i+1:]
             if not rest or any(c not in '0123456789' for c in rest):
                 # not a port number
                 scheme, url = url[:i].lower(), rest
 
+=======
+            scheme, url = url[:i].lower(), url[i+1:]
+>>>>>>> b2171a2fd41 ([3.11] gh-103848: Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format (GH-103849) (#104349))
     if url[:2] == '//':
         netloc, url = _splitnetloc(url, 2)
         if (('[' in netloc and ']' not in netloc) or
                 (']' 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)
     if allow_fragments and '#' in url:
         url, fragment = url.split('#', 1)
     if '?' in url:

Misc/NEWS.d/next/Library/2023-04-26-09-54-25.gh-issue-103848.aDSnpR.rst

@@ -0,0 +1,2 @@
+Add checks to ensure that ``[`` bracketed ``]`` hosts found by
+:func:`urllib.parse.urlsplit` are of IPv6 or IPvFuture format.