cpython: 7c78279afc30 (original) (raw)

Mercurial > cpython

changeset 97346:7c78279afc30

Issue #20059: urllib.parse raises ValueError on all invalid ports. Patch by Martin Panter. [#20059]

line wrap: on

line diff

--- a/Doc/library/urllib.parse.rst +++ b/Doc/library/urllib.parse.rst @@ -115,8 +115,9 @@ or on combining URL components into a UR | | | if present | | +------------------+-------+--------------------------+----------------------+

• See section :ref:urlparse-result-object for more information on the result
• object.

• Reading the :attr:port attribute will raise a :exc:ValueError if
• an invalid port is specified in the URL. See section
• :ref:urlparse-result-object for more information on the result object. .. versionchanged:: 3.2 Added IPv6 URL parsing capabilities. @@ -126,6 +127,10 @@ or on combining URL components into a UR false), in accordance with :rfc:3986. Previously, a whitelist of schemes that support fragments existed.
• .. versionchanged:: 3.6

•  Out-of-range port numbers now raise :exc:`ValueError`, instead of[](#l1.20)

•  returning :const:`None`.[](#l1.21)

+ .. function:: parse_qs(qs, keep_blank_values=False, strict_parsing=False, encoding='utf-8', errors='replace') @@ -228,8 +233,13 @@ or on combining URL components into a UR | | | if present | | +------------------+-------+-------------------------+----------------------+

• See section :ref:urlparse-result-object for more information on the result
• object.

• Reading the :attr:port attribute will raise a :exc:ValueError if
• an invalid port is specified in the URL. See section
• :ref:urlparse-result-object for more information on the result object. +
• .. versionchanged:: 3.6

•  Out-of-range port numbers now raise :exc:`ValueError`, instead of[](#l1.37)

•  returning :const:`None`.[](#l1.38)

.. function:: urlunsplit(parts)

--- a/Doc/whatsnew/3.6.rst +++ b/Doc/whatsnew/3.6.rst @@ -162,7 +162,10 @@ that may require changes to your code. Changes in the Python API ------------------------- -* None yet. +* Reading the :attr:~urllib.parse.SplitResult.port attribute of

• :func:urllib.parse.urlsplit and :func:~urllib.parse.urlparse results
• now raises :exc:ValueError for out-of-range values, rather than
• returning :const:None. See :issue:20059. Changes in the C API

--- a/Lib/test/test_urlparse.py +++ b/Lib/test/test_urlparse.py @@ -554,29 +554,27 @@ class UrlParseTestCase(unittest.TestCase self.assertEqual(p.port, 80) self.assertEqual(p.geturl(), url)

•    # Verify an illegal port is returned as None[](#l3.7)

•    # Verify an illegal port raises ValueError[](#l3.8)
     url = b"HTTP://WWW.PYTHON.ORG:65536/doc/#frag"[](#l3.9)
     p = urllib.parse.urlsplit(url)[](#l3.10)

•    self.assertEqual(p.port, None)[](#l3.11)

•    with self.assertRaisesRegex(ValueError, "out of range"):[](#l3.12)

•        p.port[](#l3.13)

def test_attributes_bad_port(self):

•    """Check handling of non-integer ports."""[](#l3.16)

•    p = urllib.parse.urlsplit("http://www.example.net:foo")[](#l3.17)

•    self.assertEqual(p.netloc, "www.example.net:foo")[](#l3.18)

•    self.assertRaises(ValueError, lambda: p.port)[](#l3.19)

-

•    p = urllib.parse.urlparse("http://www.example.net:foo")[](#l3.21)

•    self.assertEqual(p.netloc, "www.example.net:foo")[](#l3.22)

•    self.assertRaises(ValueError, lambda: p.port)[](#l3.23)

-

•    # Once again, repeat ourselves to test bytes[](#l3.25)

•    p = urllib.parse.urlsplit(b"http://www.example.net:foo")[](#l3.26)

•    self.assertEqual(p.netloc, b"www.example.net:foo")[](#l3.27)

•    self.assertRaises(ValueError, lambda: p.port)[](#l3.28)

-

•    p = urllib.parse.urlparse(b"http://www.example.net:foo")[](#l3.30)

•    self.assertEqual(p.netloc, b"www.example.net:foo")[](#l3.31)

•    self.assertRaises(ValueError, lambda: p.port)[](#l3.32)

•    """Check handling of invalid ports."""[](#l3.33)

•    for bytes in (False, True):[](#l3.34)

•        for parse in (urllib.parse.urlsplit, urllib.parse.urlparse):[](#l3.35)

•            for port in ("foo", "1.5", "-1", "0x10"):[](#l3.36)

•                with self.subTest(bytes=bytes, parse=parse, port=port):[](#l3.37)

•                    netloc = "www.example.net:" + port[](#l3.38)

•                    url = "http://" + netloc[](#l3.39)

•                    if bytes:[](#l3.40)

•                        netloc = netloc.encode("ascii")[](#l3.41)

•                        url = url.encode("ascii")[](#l3.42)

•                    p = parse(url)[](#l3.43)

•                    self.assertEqual(p.netloc, netloc)[](#l3.44)

•                    with self.assertRaises(ValueError):[](#l3.45)

•                        p.port[](#l3.46)

def test_attributes_without_netloc(self): # This example is straight from RFC 3261. It looks like it

--- a/Lib/urllib/parse.py +++ b/Lib/urllib/parse.py @@ -156,9 +156,8 @@ class _NetlocResultMixinBase(object): port = self._hostinfo[1] if port is not None: port = int(port, 10)

•        # Return None on an illegal port[](#l4.7)
         if not ( 0 <= port <= 65535):[](#l4.8)

•            return None[](#l4.9)

•            raise ValueError("Port out of range 0-65535")[](#l4.10)
     return port[](#l4.11)

--- a/Misc/NEWS +++ b/Misc/NEWS @@ -15,6 +15,9 @@ Core and Builtins Library ------- +- Issue #20059: urllib.parse raises ValueError on all invalid ports.

• Patch by Martin Panter. +

• Issue #24824: Signatures of codecs.encode() and codecs.decode() now are compatible with pydoc.