fix CVE-2023-24329

(cherry picked from commit f33a7c54a67b7c7e040bdc34fd78ed90e3a5b769)

backport-Make-urllib.parse.urlparse-enforce-that-a-scheme-mus.patch

@@ -0,0 +1,73 @@
+From 439b9cfaf43080e91c4ad69f312f21fa098befc7 Mon Sep 17 00:00:00 2001
+From: Ben Kallus <49924171+kenballus@users.noreply.github.com>
+Date: Sun, 13 Nov 2022 18:25:55 +0000
+Subject: [PATCH] gh-99418: Make urllib.parse.urlparse enforce that a scheme
+ must begin with an alphabetical ASCII character. (#99421)
+
+Prevent urllib.parse.urlparse from accepting schemes that don't begin with an alphabetical ASCII character.
+
+RFC 3986 defines a scheme like this: `scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )`
+RFC 2234 defines an ALPHA like this: `ALPHA = %x41-5A / %x61-7A`
+
+The WHATWG URL spec defines a scheme like this:
+`"A URL-scheme string must be one ASCII alpha, followed by zero or more of ASCII alphanumeric, U+002B (+), U+002D (-), and U+002E (.)."`
+---
+ Lib/test/test_urlparse.py                      | 18 ++++++++++++++++++
+ Lib/urllib/parse.py                            |  2 +-
+ ...22-11-12-15-45-51.gh-issue-99418.FxfAXS.rst |  2 ++
+ 3 files changed, 21 insertions(+), 1 deletion(-)
+ create mode 100644 Misc/NEWS.d/next/Library/2022-11-12-15-45-51.gh-issue-99418.FxfAXS.rst
+
+diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
+index 31943f3..f42ed9b 100644
+--- a/Lib/test/test_urlparse.py
++++ b/Lib/test/test_urlparse.py
+@@ -665,6 +665,24 @@ 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
+diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
+index b7965fe..bd59852 100644
+--- a/Lib/urllib/parse.py
++++ b/Lib/urllib/parse.py
+@@ -470,7 +470,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
+         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
+diff --git a/Misc/NEWS.d/next/Library/2022-11-12-15-45-51.gh-issue-99418.FxfAXS.rst b/Misc/NEWS.d/next/Library/2022-11-12-15-45-51.gh-issue-99418.FxfAXS.rst
+new file mode 100644
+index 0000000..0a06e7c
+--- /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.
+-- 
+2.33.0
+

fix-CVE-2023-24329.patch

@@ -0,0 +1,44 @@
+From 1bad5b2ebc2f3cb663ce425b9979b4ec4dce27b2 Mon Sep 17 00:00:00 2001
+From: shixuantong <shixuantong1@huawei.com>
+Date: Thu, 6 Apr 2023 03:30:44 +0000
+Subject: [PATCH] fix CVE-2023-24329
+
+---
+ Lib/test/test_urlparse.py | 7 +++++++
+ Lib/urllib/parse.py       | 2 +-
+ 2 files changed, 8 insertions(+), 1 deletion(-)
+
+diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
+index f42ed9b..b310017 100644
+--- a/Lib/test/test_urlparse.py
++++ b/Lib/test/test_urlparse.py
+@@ -683,6 +683,13 @@ class UrlParseTestCase(unittest.TestCase):
+                         else:
+                             self.assertEqual(p.scheme, "")
+ 
++    def test_attributes_bad_scheme_CVE_2023_24329(self):
++        """Check handling of invalid schemes that starts with blank characters."""
++        for parse in (urllib.parse.urlsplit, urllib.parse.urlparse):
++            url = " https://www.example.net"
++            p = parse(url)
++            self.assertEqual(p.scheme, "https")
++
+     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
+diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
+index bd59852..7eb3ad8 100644
+--- a/Lib/urllib/parse.py
++++ b/Lib/urllib/parse.py
+@@ -454,7 +454,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
+ 
+     Note that % escapes are not expanded.
+     """
+-
++    url = url.lstrip()
+     url, scheme, _coerce_result = _coerce_args(url, scheme)
+ 
+     for b in _UNSAFE_URL_BYTES_TO_REMOVE:
+-- 
+2.33.0
+

python3.spec

@@ -3,7 +3,7 @@ Summary: Interpreter of the Python3 programming language
 URL: https://www.python.org/
 
 Version: 3.9.9
-Release: 23
+Release: 24
 License: Python-2.0
 
 %global branchversion 3.9
@@ -103,11 +103,13 @@ Patch6009: backport-bpo-42146-Unify-cleanup-in-subprocess_fork_exec-GH-2.patch
 Patch6010: backport-CVE-2022-42919.patch
 Patch6011: backport-CVE-2022-45061.patch
 Patch6012: backport-CVE-2022-37454.patch
+Patch6013: backport-Make-urllib.parse.urlparse-enforce-that-a-scheme-mus.patch
 
 Patch9000: add-the-sm3-method-for-obtaining-the-salt-value.patch
 Patch9001: python3-Add-sw64-architecture.patch
 Patch9002: Add-loongarch-support.patch
 Patch9003: avoid-usage-of-md5-in-multiprocessing.patch
+Patch9004: fix-CVE-2023-24329.patch
 
 Provides: python%{branchversion} = %{version}-%{release}
 Provides: python(abi) = %{branchversion}
@@ -204,11 +206,13 @@ rm -r Modules/expat
 %patch6010 -p1
 %patch6011 -p1
 %patch6012 -p1
+%patch6013 -p1
 
 %patch9000 -p1
 %patch9001 -p1
 %patch9002 -p1
 %patch9003 -p1
+%patch9004 -p1
 
 rm Lib/ensurepip/_bundled/*.whl
 rm configure pyconfig.h.in
@@ -832,6 +836,12 @@ export BEP_GTDLIST="$BEP_GTDLIST_TMP"
 %{_mandir}/*/*
 
 %changelog
+* Thu Apr 06 2023 shixuantong <shixuantong1@huawei.com>- 3.9.9-24
+- Type:CVE
+- CVE:CVE-2023-24329
+- SUG:NA
+- DESC:fix CVE-2023-24329
+
 * Mon Mar 13 2023 Chenxi Mao <chenxi.mao@suse.com> - 3.9.9-23
 - Type:enhancement
 - CVE:NA