diff --git a/SOURCES/00435-gh-121650-encode-newlines-in-headers-and-verify-headers-are-sound-gh-122233.patch b/SOURCES/00435-gh-121650-encode-newlines-in-headers-and-verify-headers-are-sound-gh-122233.patch new file mode 100644 index 0000000..432920d --- /dev/null +++ b/SOURCES/00435-gh-121650-encode-newlines-in-headers-and-verify-headers-are-sound-gh-122233.patch @@ -0,0 +1,356 @@ +From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 +From: Petr Viktorin +Date: Wed, 31 Jul 2024 00:19:48 +0200 +Subject: [PATCH] 00435: gh-121650: Encode newlines in headers, and verify + headers are sound (GH-122233) + +Per RFC 2047: + +> [...] these encoding schemes allow the +> encoding of arbitrary octet values, mail readers that implement this +> decoding should also ensure that display of the decoded data on the +> recipient's terminal will not cause unwanted side-effects + +It seems that the "quoted-word" scheme is a valid way to include +a newline character in a header value, just like we already allow +undecodable bytes or control characters. +They do need to be properly quoted when serialized to text, though. + +This should fail for custom fold() implementations that aren't careful +about newlines. + +(cherry picked from commit 097633981879b3c9de9a1dd120d3aa585ecc2384) + +Co-authored-by: Petr Viktorin +Co-authored-by: Bas Bloemsaat +Co-authored-by: Serhiy Storchaka +--- + Doc/library/email.errors.rst | 6 ++ + Doc/library/email.policy.rst | 18 ++++++ + Doc/whatsnew/3.9.rst | 12 ++++ + Lib/email/_header_value_parser.py | 12 +++- + Lib/email/_policybase.py | 8 +++ + Lib/email/errors.py | 4 ++ + Lib/email/generator.py | 13 +++- + Lib/test/test_email/test_generator.py | 62 +++++++++++++++++++ + Lib/test/test_email/test_policy.py | 26 ++++++++ + ...-07-27-16-10-41.gh-issue-121650.nf6oc9.rst | 5 ++ + 10 files changed, 162 insertions(+), 4 deletions(-) + create mode 100644 Misc/NEWS.d/next/Library/2024-07-27-16-10-41.gh-issue-121650.nf6oc9.rst + +diff --git a/Doc/library/email.errors.rst b/Doc/library/email.errors.rst +index f4b9f52509..878c09bb04 100644 +--- a/Doc/library/email.errors.rst ++++ b/Doc/library/email.errors.rst +@@ -59,6 +59,12 @@ The following exception classes are defined in the :mod:`email.errors` module: + :class:`~email.mime.image.MIMEImage`). + + ++.. exception:: HeaderWriteError() ++ ++ Raised when an error occurs when the :mod:`~email.generator` outputs ++ headers. ++ ++ + Here is the list of the defects that the :class:`~email.parser.FeedParser` + can find while parsing messages. Note that the defects are added to the message + where the problem was found, so for example, if a message nested inside a +diff --git a/Doc/library/email.policy.rst b/Doc/library/email.policy.rst +index bf53b9520f..57a75ce452 100644 +--- a/Doc/library/email.policy.rst ++++ b/Doc/library/email.policy.rst +@@ -229,6 +229,24 @@ added matters. To illustrate:: + + .. versionadded:: 3.6 + ++ ++ .. attribute:: verify_generated_headers ++ ++ If ``True`` (the default), the generator will raise ++ :exc:`~email.errors.HeaderWriteError` instead of writing a header ++ that is improperly folded or delimited, such that it would ++ be parsed as multiple headers or joined with adjacent data. ++ Such headers can be generated by custom header classes or bugs ++ in the ``email`` module. ++ ++ As it's a security feature, this defaults to ``True`` even in the ++ :class:`~email.policy.Compat32` policy. ++ For backwards compatible, but unsafe, behavior, it must be set to ++ ``False`` explicitly. ++ ++ .. versionadded:: 3.9.20 ++ ++ + The following :class:`Policy` method is intended to be called by code using + the email library to create policy instances with custom settings: + +diff --git a/Doc/whatsnew/3.9.rst b/Doc/whatsnew/3.9.rst +index 1756a37338..eeda4e6955 100644 +--- a/Doc/whatsnew/3.9.rst ++++ b/Doc/whatsnew/3.9.rst +@@ -1625,3 +1625,15 @@ ipaddress + + * Fixed ``is_global`` and ``is_private`` behavior in ``IPv4Address``, + ``IPv6Address``, ``IPv4Network`` and ``IPv6Network``. ++ ++email ++----- ++ ++* Headers with embedded newlines are now quoted on output. ++ ++ The :mod:`~email.generator` will now refuse to serialize (write) headers ++ that are improperly folded or delimited, such that they would be parsed as ++ multiple headers or joined with adjacent data. ++ If you need to turn this safety feature off, ++ set :attr:`~email.policy.Policy.verify_generated_headers`. ++ (Contributed by Bas Bloemsaat and Petr Viktorin in :gh:`121650`.) +diff --git a/Lib/email/_header_value_parser.py b/Lib/email/_header_value_parser.py +index 8a8fb8bc42..e394cfd2e1 100644 +--- a/Lib/email/_header_value_parser.py ++++ b/Lib/email/_header_value_parser.py +@@ -92,6 +92,8 @@ TOKEN_ENDS = TSPECIALS | WSP + ASPECIALS = TSPECIALS | set("*'%") + ATTRIBUTE_ENDS = ASPECIALS | WSP + EXTENDED_ATTRIBUTE_ENDS = ATTRIBUTE_ENDS - set('%') ++NLSET = {'\n', '\r'} ++SPECIALSNL = SPECIALS | NLSET + + def quote_string(value): + return '"'+str(value).replace('\\', '\\\\').replace('"', r'\"')+'"' +@@ -2778,9 +2780,13 @@ def _refold_parse_tree(parse_tree, *, policy): + wrap_as_ew_blocked -= 1 + continue + tstr = str(part) +- if part.token_type == 'ptext' and set(tstr) & SPECIALS: +- # Encode if tstr contains special characters. +- want_encoding = True ++ if not want_encoding: ++ if part.token_type == 'ptext': ++ # Encode if tstr contains special characters. ++ want_encoding = not SPECIALSNL.isdisjoint(tstr) ++ else: ++ # Encode if tstr contains newlines. ++ want_encoding = not NLSET.isdisjoint(tstr) + try: + tstr.encode(encoding) + charset = encoding +diff --git a/Lib/email/_policybase.py b/Lib/email/_policybase.py +index c9cbadd2a8..d1f48211f9 100644 +--- a/Lib/email/_policybase.py ++++ b/Lib/email/_policybase.py +@@ -157,6 +157,13 @@ class Policy(_PolicyBase, metaclass=abc.ABCMeta): + message_factory -- the class to use to create new message objects. + If the value is None, the default is Message. + ++ verify_generated_headers ++ -- if true, the generator verifies that each header ++ they are properly folded, so that a parser won't ++ treat it as multiple headers, start-of-body, or ++ part of another header. ++ This is a check against custom Header & fold() ++ implementations. + """ + + raise_on_defect = False +@@ -165,6 +172,7 @@ class Policy(_PolicyBase, metaclass=abc.ABCMeta): + max_line_length = 78 + mangle_from_ = False + message_factory = None ++ verify_generated_headers = True + + def handle_defect(self, obj, defect): + """Based on policy, either raise defect or call register_defect. +diff --git a/Lib/email/errors.py b/Lib/email/errors.py +index d28a680010..1a0d5c63e6 100644 +--- a/Lib/email/errors.py ++++ b/Lib/email/errors.py +@@ -29,6 +29,10 @@ class CharsetError(MessageError): + """An illegal charset was given.""" + + ++class HeaderWriteError(MessageError): ++ """Error while writing headers.""" ++ ++ + # These are parsing defects which the parser was able to work around. + class MessageDefect(ValueError): + """Base class for a message defect.""" +diff --git a/Lib/email/generator.py b/Lib/email/generator.py +index c9b121624e..89224ae41c 100644 +--- a/Lib/email/generator.py ++++ b/Lib/email/generator.py +@@ -14,12 +14,14 @@ import random + from copy import deepcopy + from io import StringIO, BytesIO + from email.utils import _has_surrogates ++from email.errors import HeaderWriteError + + UNDERSCORE = '_' + NL = '\n' # XXX: no longer used by the code below. + + NLCRE = re.compile(r'\r\n|\r|\n') + fcre = re.compile(r'^From ', re.MULTILINE) ++NEWLINE_WITHOUT_FWSP = re.compile(r'\r\n[^ \t]|\r[^ \n\t]|\n[^ \t]') + + + +@@ -223,7 +225,16 @@ class Generator: + + def _write_headers(self, msg): + for h, v in msg.raw_items(): +- self.write(self.policy.fold(h, v)) ++ folded = self.policy.fold(h, v) ++ if self.policy.verify_generated_headers: ++ linesep = self.policy.linesep ++ if not folded.endswith(self.policy.linesep): ++ raise HeaderWriteError( ++ f'folded header does not end with {linesep!r}: {folded!r}') ++ if NEWLINE_WITHOUT_FWSP.search(folded.removesuffix(linesep)): ++ raise HeaderWriteError( ++ f'folded header contains newline: {folded!r}') ++ self.write(folded) + # A blank line always separates headers from body + self.write(self._NL) + +diff --git a/Lib/test/test_email/test_generator.py b/Lib/test/test_email/test_generator.py +index 89e7edeb63..d29400f0ed 100644 +--- a/Lib/test/test_email/test_generator.py ++++ b/Lib/test/test_email/test_generator.py +@@ -6,6 +6,7 @@ from email.message import EmailMessage + from email.generator import Generator, BytesGenerator + from email.headerregistry import Address + from email import policy ++import email.errors + from test.test_email import TestEmailBase, parameterize + + +@@ -216,6 +217,44 @@ class TestGeneratorBase: + g.flatten(msg) + self.assertEqual(s.getvalue(), self.typ(expected)) + ++ def test_keep_encoded_newlines(self): ++ msg = self.msgmaker(self.typ(textwrap.dedent("""\ ++ To: nobody ++ Subject: Bad subject=?UTF-8?Q?=0A?=Bcc: injection@example.com ++ ++ None ++ """))) ++ expected = textwrap.dedent("""\ ++ To: nobody ++ Subject: Bad subject=?UTF-8?Q?=0A?=Bcc: injection@example.com ++ ++ None ++ """) ++ s = self.ioclass() ++ g = self.genclass(s, policy=self.policy.clone(max_line_length=80)) ++ g.flatten(msg) ++ self.assertEqual(s.getvalue(), self.typ(expected)) ++ ++ def test_keep_long_encoded_newlines(self): ++ msg = self.msgmaker(self.typ(textwrap.dedent("""\ ++ To: nobody ++ Subject: Bad subject=?UTF-8?Q?=0A?=Bcc: injection@example.com ++ ++ None ++ """))) ++ expected = textwrap.dedent("""\ ++ To: nobody ++ Subject: Bad subject ++ =?utf-8?q?=0A?=Bcc: ++ injection@example.com ++ ++ None ++ """) ++ s = self.ioclass() ++ g = self.genclass(s, policy=self.policy.clone(max_line_length=30)) ++ g.flatten(msg) ++ self.assertEqual(s.getvalue(), self.typ(expected)) ++ + + class TestGenerator(TestGeneratorBase, TestEmailBase): + +@@ -224,6 +263,29 @@ class TestGenerator(TestGeneratorBase, TestEmailBase): + ioclass = io.StringIO + typ = str + ++ def test_verify_generated_headers(self): ++ """gh-121650: by default the generator prevents header injection""" ++ class LiteralHeader(str): ++ name = 'Header' ++ def fold(self, **kwargs): ++ return self ++ ++ for text in ( ++ 'Value\r\nBad Injection\r\n', ++ 'NoNewLine' ++ ): ++ with self.subTest(text=text): ++ message = message_from_string( ++ "Header: Value\r\n\r\nBody", ++ policy=self.policy, ++ ) ++ ++ del message['Header'] ++ message['Header'] = LiteralHeader(text) ++ ++ with self.assertRaises(email.errors.HeaderWriteError): ++ message.as_string() ++ + + class TestBytesGenerator(TestGeneratorBase, TestEmailBase): + +diff --git a/Lib/test/test_email/test_policy.py b/Lib/test/test_email/test_policy.py +index e87c275549..ff1ddf7d7a 100644 +--- a/Lib/test/test_email/test_policy.py ++++ b/Lib/test/test_email/test_policy.py +@@ -26,6 +26,7 @@ class PolicyAPITests(unittest.TestCase): + 'raise_on_defect': False, + 'mangle_from_': True, + 'message_factory': None, ++ 'verify_generated_headers': True, + } + # These default values are the ones set on email.policy.default. + # If any of these defaults change, the docs must be updated. +@@ -277,6 +278,31 @@ class PolicyAPITests(unittest.TestCase): + with self.assertRaises(email.errors.HeaderParseError): + policy.fold("Subject", subject) + ++ def test_verify_generated_headers(self): ++ """Turning protection off allows header injection""" ++ policy = email.policy.default.clone(verify_generated_headers=False) ++ for text in ( ++ 'Header: Value\r\nBad: Injection\r\n', ++ 'Header: NoNewLine' ++ ): ++ with self.subTest(text=text): ++ message = email.message_from_string( ++ "Header: Value\r\n\r\nBody", ++ policy=policy, ++ ) ++ class LiteralHeader(str): ++ name = 'Header' ++ def fold(self, **kwargs): ++ return self ++ ++ del message['Header'] ++ message['Header'] = LiteralHeader(text) ++ ++ self.assertEqual( ++ message.as_string(), ++ f"{text}\nBody", ++ ) ++ + # XXX: Need subclassing tests. + # For adding subclassed objects, make sure the usual rules apply (subclass + # wins), but that the order still works (right overrides left). +diff --git a/Misc/NEWS.d/next/Library/2024-07-27-16-10-41.gh-issue-121650.nf6oc9.rst b/Misc/NEWS.d/next/Library/2024-07-27-16-10-41.gh-issue-121650.nf6oc9.rst +new file mode 100644 +index 0000000000..83dd28d4ac +--- /dev/null ++++ b/Misc/NEWS.d/next/Library/2024-07-27-16-10-41.gh-issue-121650.nf6oc9.rst +@@ -0,0 +1,5 @@ ++:mod:`email` headers with embedded newlines are now quoted on output. The ++:mod:`~email.generator` will now refuse to serialize (write) headers that ++are unsafely folded or delimited; see ++:attr:`~email.policy.Policy.verify_generated_headers`. (Contributed by Bas ++Bloemsaat and Petr Viktorin in :gh:`121650`.) diff --git a/SOURCES/00436-cve-2024-8088-gh-122905-sanitize-names-in-zipfile-path.patch b/SOURCES/00436-cve-2024-8088-gh-122905-sanitize-names-in-zipfile-path.patch new file mode 100644 index 0000000..fed0497 --- /dev/null +++ b/SOURCES/00436-cve-2024-8088-gh-122905-sanitize-names-in-zipfile-path.patch @@ -0,0 +1,128 @@ +From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 +From: "Jason R. Coombs" +Date: Mon, 19 Aug 2024 19:28:20 -0400 +Subject: [PATCH] 00436: [CVE-2024-8088] gh-122905: Sanitize names in + zipfile.Path. + +Co-authored-by: Jason R. Coombs +--- + Lib/test/test_zipfile.py | 17 ++++++ + Lib/zipfile.py | 61 ++++++++++++++++++- + ...-08-11-14-08-04.gh-issue-122905.7tDsxA.rst | 1 + + 3 files changed, 78 insertions(+), 1 deletion(-) + create mode 100644 Misc/NEWS.d/next/Library/2024-08-11-14-08-04.gh-issue-122905.7tDsxA.rst + +diff --git a/Lib/test/test_zipfile.py b/Lib/test/test_zipfile.py +index 17e95eb862..9a72152357 100644 +--- a/Lib/test/test_zipfile.py ++++ b/Lib/test/test_zipfile.py +@@ -3054,6 +3054,23 @@ class TestPath(unittest.TestCase): + data = ['/'.join(string.ascii_lowercase + str(n)) for n in range(10000)] + zipfile.CompleteDirs._implied_dirs(data) + ++ def test_malformed_paths(self): ++ """ ++ Path should handle malformed paths. ++ """ ++ data = io.BytesIO() ++ zf = zipfile.ZipFile(data, "w") ++ zf.writestr("/one-slash.txt", b"content") ++ zf.writestr("//two-slash.txt", b"content") ++ zf.writestr("../parent.txt", b"content") ++ zf.filename = '' ++ root = zipfile.Path(zf) ++ assert list(map(str, root.iterdir())) == [ ++ 'one-slash.txt', ++ 'two-slash.txt', ++ 'parent.txt', ++ ] ++ + + if __name__ == "__main__": + unittest.main() +diff --git a/Lib/zipfile.py b/Lib/zipfile.py +index 95f95ee112..2e9b2868cd 100644 +--- a/Lib/zipfile.py ++++ b/Lib/zipfile.py +@@ -9,6 +9,7 @@ import io + import itertools + import os + import posixpath ++import re + import shutil + import stat + import struct +@@ -2177,7 +2178,65 @@ def _difference(minuend, subtrahend): + return itertools.filterfalse(set(subtrahend).__contains__, minuend) + + +-class CompleteDirs(ZipFile): ++class SanitizedNames: ++ """ ++ ZipFile mix-in to ensure names are sanitized. ++ """ ++ ++ def namelist(self): ++ return list(map(self._sanitize, super().namelist())) ++ ++ @staticmethod ++ def _sanitize(name): ++ r""" ++ Ensure a relative path with posix separators and no dot names. ++ Modeled after ++ https://github.com/python/cpython/blob/bcc1be39cb1d04ad9fc0bd1b9193d3972835a57c/Lib/zipfile/__init__.py#L1799-L1813 ++ but provides consistent cross-platform behavior. ++ >>> san = SanitizedNames._sanitize ++ >>> san('/foo/bar') ++ 'foo/bar' ++ >>> san('//foo.txt') ++ 'foo.txt' ++ >>> san('foo/.././bar.txt') ++ 'foo/bar.txt' ++ >>> san('foo../.bar.txt') ++ 'foo../.bar.txt' ++ >>> san('\\foo\\bar.txt') ++ 'foo/bar.txt' ++ >>> san('D:\\foo.txt') ++ 'D/foo.txt' ++ >>> san('\\\\server\\share\\file.txt') ++ 'server/share/file.txt' ++ >>> san('\\\\?\\GLOBALROOT\\Volume3') ++ '?/GLOBALROOT/Volume3' ++ >>> san('\\\\.\\PhysicalDrive1\\root') ++ 'PhysicalDrive1/root' ++ Retain any trailing slash. ++ >>> san('abc/') ++ 'abc/' ++ Raises a ValueError if the result is empty. ++ >>> san('../..') ++ Traceback (most recent call last): ++ ... ++ ValueError: Empty filename ++ """ ++ ++ def allowed(part): ++ return part and part not in {'..', '.'} ++ ++ # Remove the drive letter. ++ # Don't use ntpath.splitdrive, because that also strips UNC paths ++ bare = re.sub('^([A-Z]):', r'\1', name, flags=re.IGNORECASE) ++ clean = bare.replace('\\', '/') ++ parts = clean.split('/') ++ joined = '/'.join(filter(allowed, parts)) ++ if not joined: ++ raise ValueError("Empty filename") ++ return joined + '/' * name.endswith('/') ++ ++ ++class CompleteDirs(SanitizedNames, ZipFile): + """ + A ZipFile subclass that ensures that implied directories + are always included in the namelist. +diff --git a/Misc/NEWS.d/next/Library/2024-08-11-14-08-04.gh-issue-122905.7tDsxA.rst b/Misc/NEWS.d/next/Library/2024-08-11-14-08-04.gh-issue-122905.7tDsxA.rst +new file mode 100644 +index 0000000000..1be44c906c +--- /dev/null ++++ b/Misc/NEWS.d/next/Library/2024-08-11-14-08-04.gh-issue-122905.7tDsxA.rst +@@ -0,0 +1 @@ ++:class:`zipfile.Path` objects now sanitize names from the zipfile. diff --git a/SPECS/python3.9.spec b/SPECS/python3.9.spec index 18fb7a0..34b4fa6 100644 --- a/SPECS/python3.9.spec +++ b/SPECS/python3.9.spec @@ -17,7 +17,7 @@ URL: https://www.python.org/ #global prerel ... %global upstream_version %{general_version}%{?prerel} Version: %{general_version}%{?prerel:~%{prerel}} -Release: 3%{?dist}.3 +Release: 3%{?dist}.5 License: Python @@ -455,6 +455,29 @@ Patch427: 00427-CVE-2024-0450.patch # Upstream issue: https://github.com/python/cpython/issues/113171 Patch431: 00431-CVE-2024-4032.patch +# 00435 # f2924d30f4dd44804219c10410a57dd96764d297 +# gh-121650: Encode newlines in headers, and verify headers are sound (GH-122233) +# +# Per RFC 2047: +# +# > [...] these encoding schemes allow the +# > encoding of arbitrary octet values, mail readers that implement this +# > decoding should also ensure that display of the decoded data on the +# > recipient's terminal will not cause unwanted side-effects +# +# It seems that the "quoted-word" scheme is a valid way to include +# a newline character in a header value, just like we already allow +# undecodable bytes or control characters. +# They do need to be properly quoted when serialized to text, though. +# +# This should fail for custom fold() implementations that aren't careful +# about newlines. +Patch435: 00435-gh-121650-encode-newlines-in-headers-and-verify-headers-are-sound-gh-122233.patch + +# 00436 # 506dd77b7132f69ada7185b8bb91eba0e1296aa8 +# [CVE-2024-8088] gh-122905: Sanitize names in zipfile.Path. +Patch436: 00436-cve-2024-8088-gh-122905-sanitize-names-in-zipfile-path.patch + # (New patches go here ^^^) # # When adding new patches to "python" and "python3" in Fedora, EL, etc., @@ -1856,6 +1879,14 @@ CheckPython optimized # ====================================================== %changelog +* Fri Aug 23 2024 Charalampos Stratakis - 3.9.18-3.5 +- Security fix for CVE-2024-8088 +Resolves: RHEL-55968 + +* Tue Aug 13 2024 Lumír Balhar - 3.9.18-3.4 +- Security fix for CVE-2024-6923 +Resolves: RHEL-53044 + * Wed Jul 03 2024 Lumír Balhar - 3.9.18-3.3 - Security fix for CVE-2024-4032 Resolves: RHEL-44106