From a284d69de1d1a42714576d4a9562145a94e62127 Mon Sep 17 00:00:00 2001 From: Ben Kallus Date: Sat, 12 Nov 2022 15:43:33 -0500 Subject: [PATCH 1/2] gh-99418: Prevent urllib.parse.urlparse from accepting schemes that don't begin with an alphabetical ASCII character. --- Lib/test/test_urlparse.py | 18 ++++++++++ Lib/urllib/parse.py | 2 - Misc/NEWS.d/next/Library/2022-11-12-15-45-51.gh-issue-99418.FxfAXS.rst | 2 + 3 files changed, 21 insertions(+), 1 deletion(-) --- a/Lib/test/test_urlparse.py +++ b/Lib/test/test_urlparse.py @@ -668,6 +668,24 @@ class UrlParseTestCase(unittest.TestCase with self.assertRaises(ValueError): p.port + def test_attributes_bad_scheme(self): + """Check handling of invalid schemes.""" + for bytes in (False, True): + for parse in (urllib.parse.urlsplit, urllib.parse.urlparse): + for scheme in (".", "+", "-", "0", "http&", "६http"): + with self.subTest(bytes=bytes, parse=parse, scheme=scheme): + url = scheme + "://www.example.net" + if bytes: + if url.isascii(): + url = url.encode("ascii") + else: + continue + p = parse(url) + if bytes: + self.assertEqual(p.scheme, b"") + else: + self.assertEqual(p.scheme, "") + def test_attributes_without_netloc(self): # This example is straight from RFC 3261. It looks like it # should allow the username, hostname, and port to be filled --- a/Lib/urllib/parse.py +++ b/Lib/urllib/parse.py @@ -469,7 +469,7 @@ def urlsplit(url, scheme='', allow_fragm clear_cache() netloc = query = fragment = '' i = url.find(':') - if i > 0: + if i > 0 and url[0].isascii() and url[0].isalpha(): for c in url[:i]: if c not in scheme_chars: break --- /dev/null +++ b/Misc/NEWS.d/next/Library/2022-11-12-15-45-51.gh-issue-99418.FxfAXS.rst @@ -0,0 +1,2 @@ +Fix bug in :func:`urllib.parse.urlparse` that causes URL schemes that begin +with a digit, a plus sign, or a minus sign to be parsed incorrectly.