Accepting request 1034962 from devel:languages:python:Factory

- Add CVE-2022-45061-DoS-by-IDNA-decode.patch to avoid
  CVE-2022-45061 (bsc#1205244) allowing DoS by IDNA decoding
  extremely long domain names.

OBS-URL: https://build.opensuse.org/request/show/1034962
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/python310?expand=0&rev=25

CVE-2022-45061-DoS-by-IDNA-decode.patch

@@ -0,0 +1,88 @@
+From b8f89940de09a51fdbd8fe4705d3d1d7f1bb0c6a Mon Sep 17 00:00:00 2001
+From: "Miss Islington (bot)"
+ <31488909+miss-islington@users.noreply.github.com>
+Date: Mon, 7 Nov 2022 18:57:10 -0800
+Subject: [PATCH] [3.11] gh-98433: Fix quadratic time idna decoding. (GH-99092)
+ (GH-99222)
+
+There was an unnecessary quadratic loop in idna decoding. This restores
+the behavior to linear.
+
+(cherry picked from commit d315722564927c7202dd6e111dc79eaf14240b0d)
+
+(cherry picked from commit a6f6c3a3d6f2b580f2d87885c9b8a9350ad7bf15)
+
+Co-authored-by: Miss Islington (bot) <31488909+miss-islington@users.noreply.github.com>
+Co-authored-by: Gregory P. Smith <greg@krypto.org>
+---
+ Lib/encodings/idna.py                                                   |   32 ++++------
+ Lib/test/test_codecs.py                                                 |    6 +
+ Misc/NEWS.d/next/Security/2022-11-04-09-29-36.gh-issue-98433.l76c5G.rst |    6 +
+ 3 files changed, 27 insertions(+), 17 deletions(-)
+ create mode 100644 Misc/NEWS.d/next/Security/2022-11-04-09-29-36.gh-issue-98433.l76c5G.rst
+
+--- a/Lib/encodings/idna.py
++++ b/Lib/encodings/idna.py
+@@ -39,23 +39,21 @@ def nameprep(label):
+ 
+     # Check bidi
+     RandAL = [stringprep.in_table_d1(x) for x in label]
+-    for c in RandAL:
+-        if c:
+-            # There is a RandAL char in the string. Must perform further
+-            # tests:
+-            # 1) The characters in section 5.8 MUST be prohibited.
+-            # This is table C.8, which was already checked
+-            # 2) If a string contains any RandALCat character, the string
+-            # MUST NOT contain any LCat character.
+-            if any(stringprep.in_table_d2(x) for x in label):
+-                raise UnicodeError("Violation of BIDI requirement 2")
+-
+-            # 3) If a string contains any RandALCat character, a
+-            # RandALCat character MUST be the first character of the
+-            # string, and a RandALCat character MUST be the last
+-            # character of the string.
+-            if not RandAL[0] or not RandAL[-1]:
+-                raise UnicodeError("Violation of BIDI requirement 3")
++    if any(RandAL):
++        # There is a RandAL char in the string. Must perform further
++        # tests:
++        # 1) The characters in section 5.8 MUST be prohibited.
++        # This is table C.8, which was already checked
++        # 2) If a string contains any RandALCat character, the string
++        # MUST NOT contain any LCat character.
++        if any(stringprep.in_table_d2(x) for x in label):
++            raise UnicodeError("Violation of BIDI requirement 2")
++        # 3) If a string contains any RandALCat character, a
++        # RandALCat character MUST be the first character of the
++        # string, and a RandALCat character MUST be the last
++        # character of the string.
++        if not RandAL[0] or not RandAL[-1]:
++            raise UnicodeError("Violation of BIDI requirement 3")
+ 
+     return label
+ 
+--- a/Lib/test/test_codecs.py
++++ b/Lib/test/test_codecs.py
+@@ -1534,6 +1534,12 @@ class IDNACodecTest(unittest.TestCase):
+         self.assertEqual("pyth\xf6n.org".encode("idna"), b"xn--pythn-mua.org")
+         self.assertEqual("pyth\xf6n.org.".encode("idna"), b"xn--pythn-mua.org.")
+ 
++    def test_builtin_decode_length_limit(self):
++        with self.assertRaisesRegex(UnicodeError, "too long"):
++            (b"xn--016c"+b"a"*1100).decode("idna")
++        with self.assertRaisesRegex(UnicodeError, "too long"):
++            (b"xn--016c"+b"a"*70).decode("idna")
++
+     def test_stream(self):
+         r = codecs.getreader("idna")(io.BytesIO(b"abc"))
+         r.read(3)
+--- /dev/null
++++ b/Misc/NEWS.d/next/Security/2022-11-04-09-29-36.gh-issue-98433.l76c5G.rst
+@@ -0,0 +1,6 @@
++The IDNA codec decoder used on DNS hostnames by :mod:`socket` or :mod:`asyncio`
++related name resolution functions no longer involves a quadratic algorithm.
++This prevents a potential CPU denial of service if an out-of-spec excessive
++length hostname involving bidirectional characters were decoded. Some protocols
++such as :mod:`urllib` http ``3xx`` redirects potentially allow for an attacker
++to supply such a name.

python310.changes

@@ -1,3 +1,10 @@
+-------------------------------------------------------------------
+Wed Nov  9 18:31:23 UTC 2022 - Matej Cepl <mcepl@suse.com>
+
+- Add CVE-2022-45061-DoS-by-IDNA-decode.patch to avoid
+  CVE-2022-45061 (bsc#1205244) allowing DoS by IDNA decoding
+  extremely long domain names.
+
 -------------------------------------------------------------------
 Thu Nov  3 21:35:28 UTC 2022 - Matej Cepl <mcepl@suse.com>
 

python310.spec

@@ -175,6 +175,9 @@ Patch38:        98437-sphinx.locale._-as-gettext-in-pyspecific.patch
 # PATCH-FIX-UPSTREAM CVE-2022-42919-loc-priv-mulitproc-forksrv.patch bsc#1204886 mcepl@suse.com
 # Avoid Linux specific local privilege escalation via the multiprocessing forkserver start method
 Patch39:        CVE-2022-42919-loc-priv-mulitproc-forksrv.patch
+# PATCH-FIX-UPSTREAM CVE-2022-45061-DoS-by-IDNA-decode.patch bsc#1205244 mcepl@suse.com
+# Avoid DoS by decoding IDNA for too long domain names
+Patch40:        CVE-2022-45061-DoS-by-IDNA-decode.patch
 BuildRequires:  autoconf-archive
 BuildRequires:  automake
 BuildRequires:  fdupes
@@ -446,6 +449,7 @@ other applications.
 %patch37 -p1
 %patch38 -p1
 %patch39 -p1
+%patch40 -p1
 
 # drop Autoconf version requirement
 sed -i 's/^AC_PREREQ/dnl AC_PREREQ/' configure.ac