gh-96035: Make urllib.parse.urlparse reject non-numeric ports (#98273)

Co-authored-by: Jelle Zijlstra <jelle.zijlstra@gmail.com>

diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
index 81d6018bd1a4953f0510f98758bcb1ccdd06bc7b..59a601d9e85b08dd40b1b0e14c728620ad0b0024 100644 (file)
--- a/Lib/test/test_urlparse.py
+++ b/Lib/test/test_urlparse.py
@@ -653,13 +653,16 @@ class UrlParseTestCase(unittest.TestCase):
         """Check handling of invalid ports."""
         for bytes in (False, True):
             for parse in (urllib.parse.urlsplit, urllib.parse.urlparse):
-                for port in ("foo", "1.5", "-1", "0x10"):
+                for port in ("foo", "1.5", "-1", "0x10", "-0", "1_1", " 1", "1 ", "६"):
                     with self.subTest(bytes=bytes, parse=parse, port=port):
                         netloc = "www.example.net:" + port
                         url = "http://" + netloc
                         if bytes:
-                            netloc = netloc.encode("ascii")
-                            url = url.encode("ascii")
+                            if netloc.isascii() and port.isascii():
+                                netloc = netloc.encode("ascii")
+                                url = url.encode("ascii")
+                            else:
+                                continue
                         p = parse(url)
                         self.assertEqual(p.netloc, netloc)
                         with self.assertRaises(ValueError):
@@ -1199,6 +1202,7 @@ class Utility_Tests(unittest.TestCase):
         self.assertEqual(splitnport('127.0.0.1', 55), ('127.0.0.1', 55))
         self.assertEqual(splitnport('parrot:cheese'), ('parrot', None))
         self.assertEqual(splitnport('parrot:cheese', 55), ('parrot', None))
+        self.assertEqual(splitnport('parrot: +1_0 '), ('parrot', None))
 
     def test_splitquery(self):
         # Normal cases are exercised by other tests; ensure that we also

diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
index 3734c73948c69549afa491b79a26996490f767ca..9a3102afd63b9cc910d9250f0482aca501cfc23e 100644 (file)
--- a/Lib/urllib/parse.py
+++ b/Lib/urllib/parse.py
@@ -167,12 +167,11 @@ class _NetlocResultMixinBase(object):
     def port(self):
         port = self._hostinfo[1]
         if port is not None:
-            try:
-                port = int(port, 10)
-            except ValueError:
-                message = f'Port could not be cast to integer value as {port!r}'
-                raise ValueError(message) from None
-            if not ( 0 <= port <= 65535):
+            if port.isdigit() and port.isascii():
+                port = int(port)
+            else:
+                raise ValueError(f"Port could not be cast to integer value as {port!r}")
+            if not (0 <= port <= 65535):
                 raise ValueError("Port out of range 0-65535")
         return port
 
@@ -1132,15 +1131,15 @@ def splitnport(host, defport=-1):
 def _splitnport(host, defport=-1):
     """Split host and port, returning numeric port.
     Return given default port if no ':' found; defaults to -1.
-    Return numerical port if a valid number are found after ':'.
+    Return numerical port if a valid number is found after ':'.
     Return None if ':' but not a valid number."""
     host, delim, port = host.rpartition(':')
     if not delim:
         host = port
     elif port:
-        try:
+        if port.isdigit() and port.isascii():
             nport = int(port)
-        except ValueError:
+        else:
             nport = None
         return host, nport
     return host, defport

diff --git a/Misc/NEWS.d/next/Library/2022-10-14-19-57-37.gh-issue-96035.0xcX-p.rst b/Misc/NEWS.d/next/Library/2022-10-14-19-57-37.gh-issue-96035.0xcX-p.rst
new file mode 100644 (file)
index 0000000..f04a0fd
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2022-10-14-19-57-37.gh-issue-96035.0xcX-p.rst
@@ -0,0 +1,3 @@
+Fix bug in :func:`urllib.parse.urlparse` that causes certain port numbers
+containing whitespace, underscores, plus and minus signs, or non-ASCII digits to be
+incorrectly accepted.