From 29f348e232e82938ba2165843c448c2b291504c5 Mon Sep 17 00:00:00 2001 From: JohnJamesUtley <81572567+JohnJamesUtley@users.noreply.github.com> Date: Tue, 9 May 2023 20:18:35 -0400 Subject: gh-103848: Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format (#103849) * Adds checks to ensure that bracketed hosts found by urlsplit are of IPv6 or IPvFuture format --------- Co-authored-by: Gregory P. Smith --- Lib/test/test_urlparse.py | 26 ++++++++++++++++++++++ Lib/urllib/parse.py | 16 ++++++++++++- .../2023-04-26-09-54-25.gh-issue-103848.aDSnpR.rst | 2 ++ 3 files changed, 43 insertions(+), 1 deletion(-) create mode 100644 Misc/NEWS.d/next/Library/2023-04-26-09-54-25.gh-issue-103848.aDSnpR.rst diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py index dcdbb1c..e324bab 100644 --- a/Lib/test/test_urlparse.py +++ b/Lib/test/test_urlparse.py @@ -1042,6 +1042,32 @@ class UrlParseTestCase(unittest.TestCase): 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 index 777b7c5..da00739 100644 --- a/Lib/urllib/parse.py +++ b/Lib/urllib/parse.py @@ -33,6 +33,7 @@ import math import re import types import warnings +import ipaddress __all__ = ["urlparse", "urlunparse", "urljoin", "urldefrag", "urlsplit", "urlunsplit", "urlencode", "parse_qs", @@ -427,6 +428,17 @@ 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") + # typed=True avoids BytesWarnings being emitted during cache key # comparison since this API supports both bytes and str input. @functools.lru_cache(typed=True) @@ -466,12 +478,14 @@ 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 new file mode 100644 index 0000000..81e5904 --- /dev/null +++ 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. -- cgit v0.12