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

JohnJamesUtley · gpshead · vstinner · commit a246131212cd · 2024-11-18T16:35:37.000+01:00
* Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format --------- Co-authored-by: Gregory P. Smith <greg@krypto.org> (cherry picked from commit 29f348e)
diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
@@ -1135,6 +1135,32 @@ 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_port_casting_failure_message(self):
         message = "Port could not be cast to integer value as 'oracle'"
         p1 = urllib.parse.urlparse('http://Server=sde; Service=sde:oracle')
diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
@@ -36,6 +36,7 @@
 import types
 import collections
 import warnings
+import ipaddress
 
 __all__ = ["urlparse", "urlunparse", "urljoin", "urldefrag",
            "urlsplit", "urlunsplit", "urlencode", "parse_qs",
@@ -442,6 +443,17 @@ def _checknetloc(netloc):
             raise ValueError("netloc '" + netloc + "' contains invalid " +
                              "characters under NFKC normalization")
 
+# 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>
@@ -488,12 +500,14 @@ def urlsplit(url, scheme='', allow_fragments=True):
                 break
         else:
             scheme, url = url[:i].lower(), url[i+1:]
-
     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:
diff --git a/Misc/NEWS.d/next/Library/2023-04-26-09-54-25.gh-issue-103848.aDSnpR.rst b/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.

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