Compare commits

...

4 Commits
c8-beta ... c9

10 changed files with 891 additions and 943 deletions

View File

@ -1,7 +1,7 @@
From 5cedde59cde3f05af798a7cb5bc722cb0deb4835 Mon Sep 17 00:00:00 2001
From 43ce74d971fad62db6ccd723fe6b01da9c7ff407 Mon Sep 17 00:00:00 2001
From: Charalampos Stratakis <cstratak@redhat.com>
Date: Thu, 12 Dec 2019 16:58:31 +0100
Subject: [PATCH 1/6] Expose blake2b and blake2s hashes from OpenSSL
Subject: [PATCH 1/5] Expose blake2b and blake2s hashes from OpenSSL
These aren't as powerful as Python's own implementation, but they can be
used under FIPS.
@ -251,13 +251,13 @@ index fb61a44..1e42b87 100644
-/*[clinic end generated code: output=b339e255db698147 input=a9049054013a1b77]*/
+/*[clinic end generated code: output=1d988d457a8beebe input=a9049054013a1b77]*/
--
2.43.0
2.45.0
From 2a12baa9e201f54560ec99ad5ee1fa5b0006aa39 Mon Sep 17 00:00:00 2001
From 6872b634078a2c69644235781ebffb07f8edcb83 Mon Sep 17 00:00:00 2001
From: Petr Viktorin <pviktori@redhat.com>
Date: Thu, 25 Jul 2019 17:19:06 +0200
Subject: [PATCH 2/6] Disable Python's hash implementations in FIPS mode,
Subject: [PATCH 2/5] Disable Python's hash implementations in FIPS mode,
forcing OpenSSL
---
@ -445,10 +445,10 @@ index a8bad9d..1b1d937 100644
+ if (_Py_hashlib_fips_error(exc, name)) return NULL; \
+} while (0)
diff --git a/configure.ac b/configure.ac
index 1876f77..1875d1e 100644
index 65ad1c2..b5f9ab5 100644
--- a/configure.ac
+++ b/configure.ac
@@ -7439,7 +7439,8 @@ PY_STDLIB_MOD([_sha2],
@@ -7463,7 +7463,8 @@ PY_STDLIB_MOD([_sha2],
PY_STDLIB_MOD([_sha3], [test "$with_builtin_sha3" = yes])
PY_STDLIB_MOD([_blake2],
[test "$with_builtin_blake2" = yes], [],
@ -459,13 +459,13 @@ index 1876f77..1875d1e 100644
PY_STDLIB_MOD([_crypt],
[], [test "$ac_cv_crypt_crypt" = yes],
--
2.43.0
2.45.0
From bca05b7fdb8dcab21ef80db1d59dd5daa835d84b Mon Sep 17 00:00:00 2001
From f904abdd7a607282c2cdfd18288045cedfa28414 Mon Sep 17 00:00:00 2001
From: Charalampos Stratakis <cstratak@redhat.com>
Date: Fri, 29 Jan 2021 14:16:21 +0100
Subject: [PATCH 3/6] Use python's fall back crypto implementations only if we
Subject: [PATCH 3/5] Use python's fall back crypto implementations only if we
are not in FIPS mode
---
@ -552,13 +552,13 @@ index dd61a9a..6031b02 100644
get_builtin_constructor = getattr(hashlib,
'__get_builtin_constructor')
--
2.43.0
2.45.0
From c9a79f0aafd28677e3e0b8a1f6410105a71ff071 Mon Sep 17 00:00:00 2001
From 9bf0a53b7831409613c44fd7feecb56476f5e5e7 Mon Sep 17 00:00:00 2001
From: Charalampos Stratakis <cstratak@redhat.com>
Date: Wed, 31 Jul 2019 15:43:43 +0200
Subject: [PATCH 4/6] Test equivalence of hashes for the various digests with
Subject: [PATCH 4/5] Test equivalence of hashes for the various digests with
usedforsecurity=True/False
---
@ -712,13 +712,13 @@ index 6031b02..5bd5297 100644
class KDFTests(unittest.TestCase):
--
2.43.0
2.45.0
From e972a838729ea84a0f2e0ca8e88ae1bfc129e7d8 Mon Sep 17 00:00:00 2001
From 8a76571515a64a57b4ea0586ae8376cf2ef0ac60 Mon Sep 17 00:00:00 2001
From: Petr Viktorin <pviktori@redhat.com>
Date: Mon, 26 Aug 2019 19:39:48 +0200
Subject: [PATCH 5/6] Guard against Python HMAC in FIPS mode
Subject: [PATCH 5/5] Guard against Python HMAC in FIPS mode
---
Lib/hmac.py | 13 +++++++++----
@ -726,7 +726,7 @@ Subject: [PATCH 5/6] Guard against Python HMAC in FIPS mode
2 files changed, 19 insertions(+), 4 deletions(-)
diff --git a/Lib/hmac.py b/Lib/hmac.py
index 8b4f920..20ef96c 100644
index 8b4eb2f..e8e4864 100644
--- a/Lib/hmac.py
+++ b/Lib/hmac.py
@@ -16,8 +16,9 @@ else:
@ -773,7 +773,7 @@ index 8b4f920..20ef96c 100644
digest_cons = digestmod
elif isinstance(digestmod, str):
diff --git a/Lib/test/test_hmac.py b/Lib/test/test_hmac.py
index a39a2c4..b7b24ab 100644
index 1502fba..7997073 100644
--- a/Lib/test/test_hmac.py
+++ b/Lib/test/test_hmac.py
@@ -5,6 +5,7 @@ import hashlib
@ -812,7 +812,7 @@ index a39a2c4..b7b24ab 100644
@unittest.skipUnless(sha256_module is not None, 'need _sha256')
def test_with_sha256_module(self):
h = hmac.HMAC(b"key", b"hash this!", digestmod=sha256_module.sha256)
@@ -481,6 +489,7 @@ class SanityTestCase(unittest.TestCase):
@@ -489,6 +497,7 @@ class UpdateTestCase(unittest.TestCase):
class CopyTestCase(unittest.TestCase):
@ -820,7 +820,7 @@ index a39a2c4..b7b24ab 100644
@hashlib_helper.requires_hashdigest('sha256')
def test_attributes_old(self):
# Testing if attributes are of same type.
@@ -492,6 +501,7 @@ class CopyTestCase(unittest.TestCase):
@@ -500,6 +509,7 @@ class CopyTestCase(unittest.TestCase):
self.assertEqual(type(h1._outer), type(h2._outer),
"Types of outer don't match.")
@ -829,270 +829,5 @@ index a39a2c4..b7b24ab 100644
def test_realcopy_old(self):
# Testing if the copy method created a real copy.
--
2.43.0
From b12202196a78b877dcd32cfea273051b60038a41 Mon Sep 17 00:00:00 2001
From: Petr Viktorin <encukou@gmail.com>
Date: Wed, 25 Aug 2021 16:44:43 +0200
Subject: [PATCH 6/6] Disable hash-based PYCs in FIPS mode
If FIPS mode is on, we can't use siphash-based HMAC
(_Py_KeyedHash), so:
- Unchecked hash PYCs can be imported, but not created
- Checked hash PYCs can not be imported nor created
- The default mode is timestamp-based PYCs, even if
SOURCE_DATE_EPOCH is set.
If FIPS mode is off, there are no changes in behavior.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1835169
---
Lib/py_compile.py | 4 +++-
Lib/test/support/__init__.py | 14 +++++++++++++
Lib/test/test_cmd_line_script.py | 2 ++
Lib/test/test_compileall.py | 11 +++++++++-
.../test_importlib/source/test_file_loader.py | 6 ++++++
Lib/test/test_py_compile.py | 11 ++++++++--
Lib/test/test_zipimport.py | 2 ++
Python/import.c | 20 +++++++++++++++++++
8 files changed, 66 insertions(+), 4 deletions(-)
diff --git a/Lib/py_compile.py b/Lib/py_compile.py
index 388614e..fd9a139 100644
--- a/Lib/py_compile.py
+++ b/Lib/py_compile.py
@@ -70,7 +70,9 @@ class PycInvalidationMode(enum.Enum):
def _get_default_invalidation_mode():
- if os.environ.get('SOURCE_DATE_EPOCH'):
+ import _hashlib
+ if (os.environ.get('SOURCE_DATE_EPOCH') and not
+ _hashlib.get_fips_mode()):
return PycInvalidationMode.CHECKED_HASH
else:
return PycInvalidationMode.TIMESTAMP
diff --git a/Lib/test/support/__init__.py b/Lib/test/support/__init__.py
index fd9265c..fcd1ea7 100644
--- a/Lib/test/support/__init__.py
+++ b/Lib/test/support/__init__.py
@@ -2346,6 +2346,20 @@ def sleeping_retry(timeout, err_msg=None, /,
delay = min(delay * 2, max_delay)
+def fails_in_fips_mode(expected_error):
+ import _hashlib
+ if _hashlib.get_fips_mode():
+ def _decorator(func):
+ def _wrapper(self, *args, **kwargs):
+ with self.assertRaises(expected_error):
+ func(self, *args, **kwargs)
+ return _wrapper
+ else:
+ def _decorator(func):
+ return func
+ return _decorator
+
+
@contextlib.contextmanager
def adjust_int_max_str_digits(max_digits):
"""Temporarily change the integer string conversion length limit."""
diff --git a/Lib/test/test_cmd_line_script.py b/Lib/test/test_cmd_line_script.py
index 1b58882..d6caff1 100644
--- a/Lib/test/test_cmd_line_script.py
+++ b/Lib/test/test_cmd_line_script.py
@@ -286,6 +286,7 @@ class CmdLineTest(unittest.TestCase):
self._check_script(zip_name, run_name, zip_name, zip_name, '',
zipimport.zipimporter)
+ @support.fails_in_fips_mode(ImportError)
def test_zipfile_compiled_checked_hash(self):
with os_helper.temp_dir() as script_dir:
script_name = _make_test_script(script_dir, '__main__')
@@ -296,6 +297,7 @@ class CmdLineTest(unittest.TestCase):
self._check_script(zip_name, run_name, zip_name, zip_name, '',
zipimport.zipimporter)
+ @support.fails_in_fips_mode(ImportError)
def test_zipfile_compiled_unchecked_hash(self):
with os_helper.temp_dir() as script_dir:
script_name = _make_test_script(script_dir, '__main__')
diff --git a/Lib/test/test_compileall.py b/Lib/test/test_compileall.py
index 9cd92ad..4ec29a1 100644
--- a/Lib/test/test_compileall.py
+++ b/Lib/test/test_compileall.py
@@ -806,14 +806,23 @@ class CommandLineTestsBase:
out = self.assertRunOK('badfilename')
self.assertRegex(out, b"Can't list 'badfilename'")
- def test_pyc_invalidation_mode(self):
+ @support.fails_in_fips_mode(AssertionError)
+ def test_pyc_invalidation_mode_checked(self):
script_helper.make_script(self.pkgdir, 'f1', '')
pyc = importlib.util.cache_from_source(
os.path.join(self.pkgdir, 'f1.py'))
+
self.assertRunOK('--invalidation-mode=checked-hash', self.pkgdir)
with open(pyc, 'rb') as fp:
data = fp.read()
self.assertEqual(int.from_bytes(data[4:8], 'little'), 0b11)
+
+ @support.fails_in_fips_mode(AssertionError)
+ def test_pyc_invalidation_mode_unchecked(self):
+ script_helper.make_script(self.pkgdir, 'f1', '')
+ pyc = importlib.util.cache_from_source(
+ os.path.join(self.pkgdir, 'f1.py'))
+
self.assertRunOK('--invalidation-mode=unchecked-hash', self.pkgdir)
with open(pyc, 'rb') as fp:
data = fp.read()
diff --git a/Lib/test/test_importlib/source/test_file_loader.py b/Lib/test/test_importlib/source/test_file_loader.py
index f35adec..62087c6 100644
--- a/Lib/test/test_importlib/source/test_file_loader.py
+++ b/Lib/test/test_importlib/source/test_file_loader.py
@@ -16,6 +16,7 @@ import types
import unittest
import warnings
+from test import support
from test.support.import_helper import make_legacy_pyc, unload
from test.test_py_compile import without_source_date_epoch
@@ -237,6 +238,7 @@ class SimpleTest(abc.LoaderTests):
loader.load_module('bad name')
@util.writes_bytecode_files
+ @support.fails_in_fips_mode(ImportError)
def test_checked_hash_based_pyc(self):
with util.create_modules('_temp') as mapping:
source = mapping['_temp']
@@ -268,6 +270,7 @@ class SimpleTest(abc.LoaderTests):
)
@util.writes_bytecode_files
+ @support.fails_in_fips_mode(ImportError)
def test_overridden_checked_hash_based_pyc(self):
with util.create_modules('_temp') as mapping, \
unittest.mock.patch('_imp.check_hash_based_pycs', 'never'):
@@ -293,6 +296,7 @@ class SimpleTest(abc.LoaderTests):
self.assertEqual(mod.state, 'old')
@util.writes_bytecode_files
+ @support.fails_in_fips_mode(ImportError)
def test_unchecked_hash_based_pyc(self):
with util.create_modules('_temp') as mapping:
source = mapping['_temp']
@@ -323,6 +327,7 @@ class SimpleTest(abc.LoaderTests):
)
@util.writes_bytecode_files
+ @support.fails_in_fips_mode(ImportError)
def test_overridden_unchecked_hash_based_pyc(self):
with util.create_modules('_temp') as mapping, \
unittest.mock.patch('_imp.check_hash_based_pycs', 'always'):
@@ -432,6 +437,7 @@ class BadBytecodeTest:
del_source=del_source)
test('_temp', mapping, bc_path)
+ @support.fails_in_fips_mode(ImportError)
def _test_partial_hash(self, test, *, del_source=False):
with util.create_modules('_temp') as mapping:
bc_path = self.manipulate_bytecode(
diff --git a/Lib/test/test_py_compile.py b/Lib/test/test_py_compile.py
index c4e6551..81fd962 100644
--- a/Lib/test/test_py_compile.py
+++ b/Lib/test/test_py_compile.py
@@ -141,13 +141,16 @@ class PyCompileTestsBase:
importlib.util.cache_from_source(bad_coding)))
def test_source_date_epoch(self):
+ import _hashlib
py_compile.compile(self.source_path, self.pyc_path)
self.assertTrue(os.path.exists(self.pyc_path))
self.assertFalse(os.path.exists(self.cache_path))
with open(self.pyc_path, 'rb') as fp:
flags = importlib._bootstrap_external._classify_pyc(
fp.read(), 'test', {})
- if os.environ.get('SOURCE_DATE_EPOCH'):
+ if _hashlib.get_fips_mode():
+ expected_flags = 0b00
+ elif os.environ.get('SOURCE_DATE_EPOCH'):
expected_flags = 0b11
else:
expected_flags = 0b00
@@ -178,7 +181,8 @@ class PyCompileTestsBase:
# Specifying optimized bytecode should lead to a path reflecting that.
self.assertIn('opt-2', py_compile.compile(self.source_path, optimize=2))
- def test_invalidation_mode(self):
+ @support.fails_in_fips_mode(ImportError)
+ def test_invalidation_mode_checked(self):
py_compile.compile(
self.source_path,
invalidation_mode=py_compile.PycInvalidationMode.CHECKED_HASH,
@@ -187,6 +191,9 @@ class PyCompileTestsBase:
flags = importlib._bootstrap_external._classify_pyc(
fp.read(), 'test', {})
self.assertEqual(flags, 0b11)
+
+ @support.fails_in_fips_mode(ImportError)
+ def test_invalidation_mode_unchecked(self):
py_compile.compile(
self.source_path,
invalidation_mode=py_compile.PycInvalidationMode.UNCHECKED_HASH,
diff --git a/Lib/test/test_zipimport.py b/Lib/test/test_zipimport.py
index 14c1971..bcd1466 100644
--- a/Lib/test/test_zipimport.py
+++ b/Lib/test/test_zipimport.py
@@ -190,6 +190,7 @@ class UncompressedZipImportTestCase(ImportHooksBaseTestCase):
TESTMOD + pyc_ext: (NOW, test_pyc)}
self.doTest(pyc_ext, files, TESTMOD)
+ @support.fails_in_fips_mode(ImportError)
def testUncheckedHashBasedPyc(self):
source = b"state = 'old'"
source_hash = importlib.util.source_hash(source)
@@ -204,6 +205,7 @@ class UncompressedZipImportTestCase(ImportHooksBaseTestCase):
self.assertEqual(mod.state, 'old')
self.doTest(None, files, TESTMOD, call=check)
+ @support.fails_in_fips_mode(ImportError)
@unittest.mock.patch('_imp.check_hash_based_pycs', 'always')
def test_checked_hash_based_change_pyc(self):
source = b"state = 'old'"
diff --git a/Python/import.c b/Python/import.c
index 54232a1..236786b 100644
--- a/Python/import.c
+++ b/Python/import.c
@@ -3829,6 +3829,26 @@ static PyObject *
_imp_source_hash_impl(PyObject *module, long key, Py_buffer *source)
/*[clinic end generated code: output=edb292448cf399ea input=9aaad1e590089789]*/
{
+ PyObject *_hashlib = PyImport_ImportModule("_hashlib");
+ if (_hashlib == NULL) {
+ return NULL;
+ }
+ PyObject *fips_mode_obj = PyObject_CallMethod(_hashlib, "get_fips_mode", NULL);
+ Py_DECREF(_hashlib);
+ if (fips_mode_obj == NULL) {
+ return NULL;
+ }
+ int fips_mode = PyObject_IsTrue(fips_mode_obj);
+ Py_DECREF(fips_mode_obj);
+ if (fips_mode < 0) {
+ return NULL;
+ }
+ if (fips_mode) {
+ PyErr_SetString(
+ PyExc_ImportError,
+ "hash-based PYC validation (siphash24) not available in FIPS mode");
+ return NULL;
+ };
union {
uint64_t x;
char data[sizeof(uint64_t)];
--
2.43.0
2.45.0

View File

@ -0,0 +1,88 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Serhiy Storchaka <storchaka@gmail.com>
Date: Sun, 11 Feb 2024 12:08:39 +0200
Subject: [PATCH] 00422: gh-115133: Fix tests for XMLPullParser with Expat
2.6.0
Feeding the parser by too small chunks defers parsing to prevent
CVE-2023-52425. Future versions of Expat may be more reactive.
(cherry picked from commit 4a08e7b3431cd32a0daf22a33421cd3035343dc4)
---
Lib/test/test_xml_etree.py | 58 ++++++++++++-------
...-02-08-14-21-28.gh-issue-115133.ycl4ko.rst | 2 +
2 files changed, 38 insertions(+), 22 deletions(-)
create mode 100644 Misc/NEWS.d/next/Library/2024-02-08-14-21-28.gh-issue-115133.ycl4ko.rst
diff --git a/Lib/test/test_xml_etree.py b/Lib/test/test_xml_etree.py
index b50898f1d1..6fb888cb21 100644
--- a/Lib/test/test_xml_etree.py
+++ b/Lib/test/test_xml_etree.py
@@ -1400,28 +1400,37 @@ class XMLPullParserTest(unittest.TestCase):
self.assertEqual([(action, elem.tag) for action, elem in events],
expected)
- def test_simple_xml(self):
- for chunk_size in (None, 1, 5):
- with self.subTest(chunk_size=chunk_size):
- parser = ET.XMLPullParser()
- self.assert_event_tags(parser, [])
- self._feed(parser, "<!-- comment -->\n", chunk_size)
- self.assert_event_tags(parser, [])
- self._feed(parser,
- "<root>\n <element key='value'>text</element",
- chunk_size)
- self.assert_event_tags(parser, [])
- self._feed(parser, ">\n", chunk_size)
- self.assert_event_tags(parser, [('end', 'element')])
- self._feed(parser, "<element>text</element>tail\n", chunk_size)
- self._feed(parser, "<empty-element/>\n", chunk_size)
- self.assert_event_tags(parser, [
- ('end', 'element'),
- ('end', 'empty-element'),
- ])
- self._feed(parser, "</root>\n", chunk_size)
- self.assert_event_tags(parser, [('end', 'root')])
- self.assertIsNone(parser.close())
+ def test_simple_xml(self, chunk_size=None):
+ parser = ET.XMLPullParser()
+ self.assert_event_tags(parser, [])
+ self._feed(parser, "<!-- comment -->\n", chunk_size)
+ self.assert_event_tags(parser, [])
+ self._feed(parser,
+ "<root>\n <element key='value'>text</element",
+ chunk_size)
+ self.assert_event_tags(parser, [])
+ self._feed(parser, ">\n", chunk_size)
+ self.assert_event_tags(parser, [('end', 'element')])
+ self._feed(parser, "<element>text</element>tail\n", chunk_size)
+ self._feed(parser, "<empty-element/>\n", chunk_size)
+ self.assert_event_tags(parser, [
+ ('end', 'element'),
+ ('end', 'empty-element'),
+ ])
+ self._feed(parser, "</root>\n", chunk_size)
+ self.assert_event_tags(parser, [('end', 'root')])
+ self.assertIsNone(parser.close())
+
+ @unittest.expectedFailure
+ def test_simple_xml_chunk_1(self):
+ self.test_simple_xml(chunk_size=1)
+
+ @unittest.expectedFailure
+ def test_simple_xml_chunk_5(self):
+ self.test_simple_xml(chunk_size=5)
+
+ def test_simple_xml_chunk_22(self):
+ self.test_simple_xml(chunk_size=22)
def test_feed_while_iterating(self):
parser = ET.XMLPullParser()
diff --git a/Misc/NEWS.d/next/Library/2024-02-08-14-21-28.gh-issue-115133.ycl4ko.rst b/Misc/NEWS.d/next/Library/2024-02-08-14-21-28.gh-issue-115133.ycl4ko.rst
new file mode 100644
index 0000000000..6f1015235c
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2024-02-08-14-21-28.gh-issue-115133.ycl4ko.rst
@@ -0,0 +1,2 @@
+Fix tests for :class:`~xml.etree.ElementTree.XMLPullParser` with Expat
+2.6.0.

View File

@ -0,0 +1,345 @@
From 7a25b2f511054dd2011308275bb24e914e1977af Mon Sep 17 00:00:00 2001
From: Petr Viktorin <encukou@gmail.com>
Date: Tue, 6 Aug 2024 19:07:19 +0200
Subject: [PATCH] gh-121650: Encode newlines in headers, and verify headers are
sound (GH-122233) (#122599)
* gh-121650: Encode newlines in headers, and verify headers are sound (GH-122233)
- Encode header parts that contain newlines
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.
- Verify that email headers are well-formed
This should fail for custom fold() implementations that aren't careful
about newlines.
Co-authored-by: Bas Bloemsaat <bas@bloemsaat.org>
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
(cherry picked from commit 097633981879b3c9de9a1dd120d3aa585ecc2384)
* Document changes as made in 3.12.5
---
Doc/library/email.errors.rst | 7 +++
Doc/library/email.policy.rst | 18 ++++++
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 ++
9 files changed, 151 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 56aea65..27b0481 100644
--- a/Doc/library/email.errors.rst
+++ b/Doc/library/email.errors.rst
@@ -58,6 +58,13 @@ The following exception classes are defined in the :mod:`email.errors` module:
:class:`~email.mime.nonmultipart.MIMENonMultipart` (e.g.
:class:`~email.mime.image.MIMEImage`).
+
+.. exception:: HeaderWriteError()
+
+ Raised when an error occurs when the :mod:`~email.generator` outputs
+ headers.
+
+
.. exception:: MessageDefect()
This is the base class for all defects found when parsing email messages.
diff --git a/Doc/library/email.policy.rst b/Doc/library/email.policy.rst
index fd47dd0..6ec6e4d 100644
--- a/Doc/library/email.policy.rst
+++ b/Doc/library/email.policy.rst
@@ -230,6 +230,24 @@ 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:: 3.12.5
+
+
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/Lib/email/_header_value_parser.py b/Lib/email/_header_value_parser.py
index 0d6bd81..362edc5 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'\"')+'"'
@@ -2776,9 +2778,13 @@ 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 c9cbadd..d1f4821 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 3ad0056..02aa5ec 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 7ccbe10..ea87ad2 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]')
class Generator:
@@ -222,7 +224,16 @@ 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 89e7ede..d29400f 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 e87c275..ff1ddf7 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 0000000..83dd28d
--- /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`.)
--
2.45.2

View File

@ -0,0 +1,121 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: "Miss Islington (bot)"
<31488909+miss-islington@users.noreply.github.com>
Date: Mon, 12 Aug 2024 02:35:17 +0200
Subject: [PATCH] 00436: [CVE-2024-8088] gh-122905: Sanitize names in
zipfile.Path.
---
Lib/test/test_zipfile/_path/test_path.py | 17 +++++
Lib/zipfile/_path/__init__.py | 64 ++++++++++++++++++-
...-08-11-14-08-04.gh-issue-122905.7tDsxA.rst | 1 +
3 files changed, 81 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/_path/test_path.py b/Lib/test/test_zipfile/_path/test_path.py
index 06d5aab69b..90885dbbe3 100644
--- a/Lib/test/test_zipfile/_path/test_path.py
+++ b/Lib/test/test_zipfile/_path/test_path.py
@@ -577,3 +577,20 @@ def test_getinfo_missing(self, alpharep):
zipfile.Path(alpharep)
with self.assertRaises(KeyError):
alpharep.getinfo('does-not-exist')
+
+ 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',
+ ]
diff --git a/Lib/zipfile/_path/__init__.py b/Lib/zipfile/_path/__init__.py
index 78c413563b..42f9fded21 100644
--- a/Lib/zipfile/_path/__init__.py
+++ b/Lib/zipfile/_path/__init__.py
@@ -83,7 +83,69 @@ def __setstate__(self, state):
super().__init__(*args, **kwargs)
-class CompleteDirs(InitializedState, zipfile.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(InitializedState, SanitizedNames, zipfile.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.

View File

@ -0,0 +1,245 @@
From 4eaf4891c12589e3c7bdad5f5b076e4c8392dd06 Mon Sep 17 00:00:00 2001
From: "Miss Islington (bot)"
<31488909+miss-islington@users.noreply.github.com>
Date: Sun, 1 Sep 2024 00:35:24 +0200
Subject: [PATCH] [3.12] gh-121285: Remove backtracking when parsing tarfile
headers (GH-121286) (GH-123543)
gh-121285: Remove backtracking when parsing tarfile headers (GH-121286)
* Remove backtracking when parsing tarfile headers
* Rewrite PAX header parsing to be stricter
* Optimize parsing of GNU extended sparse headers v0.0
(cherry picked from commit 34ddb64d088dd7ccc321f6103d23153256caa5d4)
Co-authored-by: Seth Michael Larson <seth@python.org>
Co-authored-by: Kirill Podoprigora <kirill.bast9@mail.ru>
Co-authored-by: Gregory P. Smith <greg@krypto.org>
---
Lib/tarfile.py | 103 ++++++++++++------
Lib/test/test_tarfile.py | 42 +++++++
...-07-02-13-39-20.gh-issue-121285.hrl-yI.rst | 2 +
3 files changed, 112 insertions(+), 35 deletions(-)
create mode 100644 Misc/NEWS.d/next/Security/2024-07-02-13-39-20.gh-issue-121285.hrl-yI.rst
diff --git a/Lib/tarfile.py b/Lib/tarfile.py
index e1487e3864d44b..0a0f31eca06c04 100755
--- a/Lib/tarfile.py
+++ b/Lib/tarfile.py
@@ -843,6 +843,9 @@ def data_filter(member, dest_path):
# Sentinel for replace() defaults, meaning "don't change the attribute"
_KEEP = object()
+# Header length is digits followed by a space.
+_header_length_prefix_re = re.compile(br"([0-9]{1,20}) ")
+
class TarInfo(object):
"""Informational class which holds the details about an
archive member given by a tar header block.
@@ -1412,37 +1415,59 @@ def _proc_pax(self, tarfile):
else:
pax_headers = tarfile.pax_headers.copy()
- # Check if the pax header contains a hdrcharset field. This tells us
- # the encoding of the path, linkpath, uname and gname fields. Normally,
- # these fields are UTF-8 encoded but since POSIX.1-2008 tar
- # implementations are allowed to store them as raw binary strings if
- # the translation to UTF-8 fails.
- match = re.search(br"\d+ hdrcharset=([^\n]+)\n", buf)
- if match is not None:
- pax_headers["hdrcharset"] = match.group(1).decode("utf-8")
-
- # For the time being, we don't care about anything other than "BINARY".
- # The only other value that is currently allowed by the standard is
- # "ISO-IR 10646 2000 UTF-8" in other words UTF-8.
- hdrcharset = pax_headers.get("hdrcharset")
- if hdrcharset == "BINARY":
- encoding = tarfile.encoding
- else:
- encoding = "utf-8"
-
# Parse pax header information. A record looks like that:
# "%d %s=%s\n" % (length, keyword, value). length is the size
# of the complete record including the length field itself and
- # the newline. keyword and value are both UTF-8 encoded strings.
- regex = re.compile(br"(\d+) ([^=]+)=")
+ # the newline.
pos = 0
- while match := regex.match(buf, pos):
- length, keyword = match.groups()
- length = int(length)
- if length == 0:
+ encoding = None
+ raw_headers = []
+ while len(buf) > pos and buf[pos] != 0x00:
+ if not (match := _header_length_prefix_re.match(buf, pos)):
+ raise InvalidHeaderError("invalid header")
+ try:
+ length = int(match.group(1))
+ except ValueError:
+ raise InvalidHeaderError("invalid header")
+ # Headers must be at least 5 bytes, shortest being '5 x=\n'.
+ # Value is allowed to be empty.
+ if length < 5:
+ raise InvalidHeaderError("invalid header")
+ if pos + length > len(buf):
+ raise InvalidHeaderError("invalid header")
+
+ header_value_end_offset = match.start(1) + length - 1 # Last byte of the header
+ keyword_and_value = buf[match.end(1) + 1:header_value_end_offset]
+ raw_keyword, equals, raw_value = keyword_and_value.partition(b"=")
+
+ # Check the framing of the header. The last character must be '\n' (0x0A)
+ if not raw_keyword or equals != b"=" or buf[header_value_end_offset] != 0x0A:
raise InvalidHeaderError("invalid header")
- value = buf[match.end(2) + 1:match.start(1) + length - 1]
+ raw_headers.append((length, raw_keyword, raw_value))
+
+ # Check if the pax header contains a hdrcharset field. This tells us
+ # the encoding of the path, linkpath, uname and gname fields. Normally,
+ # these fields are UTF-8 encoded but since POSIX.1-2008 tar
+ # implementations are allowed to store them as raw binary strings if
+ # the translation to UTF-8 fails. For the time being, we don't care about
+ # anything other than "BINARY". The only other value that is currently
+ # allowed by the standard is "ISO-IR 10646 2000 UTF-8" in other words UTF-8.
+ # Note that we only follow the initial 'hdrcharset' setting to preserve
+ # the initial behavior of the 'tarfile' module.
+ if raw_keyword == b"hdrcharset" and encoding is None:
+ if raw_value == b"BINARY":
+ encoding = tarfile.encoding
+ else: # This branch ensures only the first 'hdrcharset' header is used.
+ encoding = "utf-8"
+ pos += length
+
+ # If no explicit hdrcharset is set, we use UTF-8 as a default.
+ if encoding is None:
+ encoding = "utf-8"
+
+ # After parsing the raw headers we can decode them to text.
+ for length, raw_keyword, raw_value in raw_headers:
# Normally, we could just use "utf-8" as the encoding and "strict"
# as the error handler, but we better not take the risk. For
# example, GNU tar <= 1.23 is known to store filenames it cannot
@@ -1450,17 +1475,16 @@ def _proc_pax(self, tarfile):
# hdrcharset=BINARY header).
# We first try the strict standard encoding, and if that fails we
# fall back on the user's encoding and error handler.
- keyword = self._decode_pax_field(keyword, "utf-8", "utf-8",
+ keyword = self._decode_pax_field(raw_keyword, "utf-8", "utf-8",
tarfile.errors)
if keyword in PAX_NAME_FIELDS:
- value = self._decode_pax_field(value, encoding, tarfile.encoding,
+ value = self._decode_pax_field(raw_value, encoding, tarfile.encoding,
tarfile.errors)
else:
- value = self._decode_pax_field(value, "utf-8", "utf-8",
+ value = self._decode_pax_field(raw_value, "utf-8", "utf-8",
tarfile.errors)
pax_headers[keyword] = value
- pos += length
# Fetch the next header.
try:
@@ -1475,7 +1499,7 @@ def _proc_pax(self, tarfile):
elif "GNU.sparse.size" in pax_headers:
# GNU extended sparse format version 0.0.
- self._proc_gnusparse_00(next, pax_headers, buf)
+ self._proc_gnusparse_00(next, raw_headers)
elif pax_headers.get("GNU.sparse.major") == "1" and pax_headers.get("GNU.sparse.minor") == "0":
# GNU extended sparse format version 1.0.
@@ -1497,15 +1521,24 @@ def _proc_pax(self, tarfile):
return next
- def _proc_gnusparse_00(self, next, pax_headers, buf):
+ def _proc_gnusparse_00(self, next, raw_headers):
"""Process a GNU tar extended sparse header, version 0.0.
"""
offsets = []
- for match in re.finditer(br"\d+ GNU.sparse.offset=(\d+)\n", buf):
- offsets.append(int(match.group(1)))
numbytes = []
- for match in re.finditer(br"\d+ GNU.sparse.numbytes=(\d+)\n", buf):
- numbytes.append(int(match.group(1)))
+ for _, keyword, value in raw_headers:
+ if keyword == b"GNU.sparse.offset":
+ try:
+ offsets.append(int(value.decode()))
+ except ValueError:
+ raise InvalidHeaderError("invalid header")
+
+ elif keyword == b"GNU.sparse.numbytes":
+ try:
+ numbytes.append(int(value.decode()))
+ except ValueError:
+ raise InvalidHeaderError("invalid header")
+
next.sparse = list(zip(offsets, numbytes))
def _proc_gnusparse_01(self, next, pax_headers):
diff --git a/Lib/test/test_tarfile.py b/Lib/test/test_tarfile.py
index 3fbd25e742b181..e28d0311826e2b 100644
--- a/Lib/test/test_tarfile.py
+++ b/Lib/test/test_tarfile.py
@@ -1237,6 +1237,48 @@ def test_pax_number_fields(self):
finally:
tar.close()
+ def test_pax_header_bad_formats(self):
+ # The fields from the pax header have priority over the
+ # TarInfo.
+ pax_header_replacements = (
+ b" foo=bar\n",
+ b"0 \n",
+ b"1 \n",
+ b"2 \n",
+ b"3 =\n",
+ b"4 =a\n",
+ b"1000000 foo=bar\n",
+ b"0 foo=bar\n",
+ b"-12 foo=bar\n",
+ b"000000000000000000000000036 foo=bar\n",
+ )
+ pax_headers = {"foo": "bar"}
+
+ for replacement in pax_header_replacements:
+ with self.subTest(header=replacement):
+ tar = tarfile.open(tmpname, "w", format=tarfile.PAX_FORMAT,
+ encoding="iso8859-1")
+ try:
+ t = tarfile.TarInfo()
+ t.name = "pax" # non-ASCII
+ t.uid = 1
+ t.pax_headers = pax_headers
+ tar.addfile(t)
+ finally:
+ tar.close()
+
+ with open(tmpname, "rb") as f:
+ data = f.read()
+ self.assertIn(b"11 foo=bar\n", data)
+ data = data.replace(b"11 foo=bar\n", replacement)
+
+ with open(tmpname, "wb") as f:
+ f.truncate()
+ f.write(data)
+
+ with self.assertRaisesRegex(tarfile.ReadError, r"method tar: ReadError\('invalid header'\)"):
+ tarfile.open(tmpname, encoding="iso8859-1")
+
class WriteTestBase(TarTest):
# Put all write tests in here that are supposed to be tested
diff --git a/Misc/NEWS.d/next/Security/2024-07-02-13-39-20.gh-issue-121285.hrl-yI.rst b/Misc/NEWS.d/next/Security/2024-07-02-13-39-20.gh-issue-121285.hrl-yI.rst
new file mode 100644
index 00000000000000..81f918bfe2b255
--- /dev/null
+++ b/Misc/NEWS.d/next/Security/2024-07-02-13-39-20.gh-issue-121285.hrl-yI.rst
@@ -0,0 +1,2 @@
+Remove backtracking from tarfile header parsing for ``hdrcharset``, PAX, and
+GNU sparse headers.

View File

@ -16,7 +16,6 @@ LEVELS = (None, 1, 2)
# list of globs of test and other files that we expect not to have bytecode
not_compiled = [
'/usr/bin/*',
'/usr/lib/rpm/redhat/*',
'*/test/badsyntax_*.py',
'*/tokenizedata/bad_coding.py',
'*/tokenizedata/bad_coding2.py',

View File

@ -1,171 +0,0 @@
'''Script to perform import of each module given to %%py_check_import
'''
import argparse
import importlib
import fnmatch
import os
import re
import site
import sys
from contextlib import contextmanager
from pathlib import Path
def read_modules_files(file_paths):
'''Read module names from the files (modules must be newline separated).
Return the module names list or, if no files were provided, an empty list.
'''
if not file_paths:
return []
modules = []
for file in file_paths:
file_contents = file.read_text()
modules.extend(file_contents.split())
return modules
def read_modules_from_cli(argv):
'''Read module names from command-line arguments (space or comma separated).
Return the module names list.
'''
if not argv:
return []
# %%py3_check_import allows to separate module list with comma or whitespace,
# we need to unify the output to a list of particular elements
modules_as_str = ' '.join(argv)
modules = re.split(r'[\s,]+', modules_as_str)
# Because of shell expansion in some less typical cases it may happen
# that a trailing space will occur at the end of the list.
# Remove the empty items from the list before passing it further
modules = [m for m in modules if m]
return modules
def filter_top_level_modules_only(modules):
'''Filter out entries with nested modules (containing dot) ie. 'foo.bar'.
Return the list of top-level modules.
'''
return [module for module in modules if '.' not in module]
def any_match(text, globs):
'''Return True if any of given globs fnmatchcase's the given text.'''
return any(fnmatch.fnmatchcase(text, g) for g in globs)
def exclude_unwanted_module_globs(globs, modules):
'''Filter out entries which match the either of the globs given as argv.
Return the list of filtered modules.
'''
return [m for m in modules if not any_match(m, globs)]
def read_modules_from_all_args(args):
'''Return a joined list of modules from all given command-line arguments.
'''
modules = read_modules_files(args.filename)
modules.extend(read_modules_from_cli(args.modules))
if args.exclude:
modules = exclude_unwanted_module_globs(args.exclude, modules)
if args.top_level:
modules = filter_top_level_modules_only(modules)
# Error when someone accidentally managed to filter out everything
if len(modules) == 0:
raise ValueError('No modules to check were left')
return modules
def import_modules(modules):
'''Procedure to perform import check for each module name from the given list of modules.
'''
for module in modules:
print('Check import:', module, file=sys.stderr)
importlib.import_module(module)
def argparser():
parser = argparse.ArgumentParser(
description='Generate list of all importable modules for import check.'
)
parser.add_argument(
'modules', nargs='*',
help=('Add modules to check the import (space or comma separated).'),
)
parser.add_argument(
'-f', '--filename', action='append', type=Path,
help='Add importable module names list from file.',
)
parser.add_argument(
'-t', '--top-level', action='store_true',
help='Check only top-level modules.',
)
parser.add_argument(
'-e', '--exclude', action='append',
help='Provide modules globs to be excluded from the check.',
)
return parser
@contextmanager
def remove_unwanteds_from_sys_path():
'''Remove cwd and this script's parent from sys.path for the import test.
Bring the original contents back after import is done (or failed)
'''
cwd_absolute = Path.cwd().absolute()
this_file_parent = Path(__file__).parent.absolute()
old_sys_path = list(sys.path)
for path in old_sys_path:
if Path(path).absolute() in (cwd_absolute, this_file_parent):
sys.path.remove(path)
try:
yield
finally:
sys.path = old_sys_path
def addsitedirs_from_environ():
'''Load directories from the _PYTHONSITE environment variable (separated by :)
and load the ones already present in sys.path via site.addsitedir()
to handle .pth files in them.
This is needed to properly import old-style namespace packages with nspkg.pth files.
See https://bugzilla.redhat.com/2018551 for a more detailed rationale.'''
for path in os.getenv('_PYTHONSITE', '').split(':'):
if path in sys.path:
site.addsitedir(path)
def main(argv=None):
cli_args = argparser().parse_args(argv)
if not cli_args.modules and not cli_args.filename:
raise ValueError('No modules to check were provided')
modules = read_modules_from_all_args(cli_args)
with remove_unwanteds_from_sys_path():
addsitedirs_from_environ()
import_modules(modules)
if __name__ == '__main__':
main()

View File

@ -1,91 +0,0 @@
%__python3 /usr/bin/python3.12
%python3_pkgversion 3.12
# The following are macros from macros.python3 in Fedora that are newer/different than those in the python3-rpm-macros package in RHEL 8.
# These macros overwrite/supercede some of the macros in the python3-rpm-macros package in RHEL.
# nb: $RPM_BUILD_ROOT is not set when the macros are expanded (at spec parse time)
# so we set it manually (to empty string), making our Python prefer the correct install scheme location
# platbase/base is explicitly set to %%{_prefix} to support custom values, such as /app for flatpaks
%python3_sitelib %(RPM_BUILD_ROOT= %{__python3} -Ic "import sysconfig; print(sysconfig.get_path('purelib', vars={'platbase': '%{_prefix}', 'base': '%{_prefix}'}))")
%python3_sitearch %(RPM_BUILD_ROOT= %{__python3} -Ic "import sysconfig; print(sysconfig.get_path('platlib', vars={'platbase': '%{_prefix}', 'base': '%{_prefix}'}))")
%python3_version %(RPM_BUILD_ROOT= %{__python3} -Ic "import sys; sys.stdout.write('{0.major}.{0.minor}'.format(sys.version_info))")
%python3_version_nodots %(RPM_BUILD_ROOT= %{__python3} -Ic "import sys; sys.stdout.write('{0.major}{0.minor}'.format(sys.version_info))")
%python3_platform %(RPM_BUILD_ROOT= %{__python3} -Ic "import sysconfig; print(sysconfig.get_platform())")
%python3_platform_triplet %(RPM_BUILD_ROOT= %{__python3} -Ic "import sysconfig; print(sysconfig.get_config_var('MULTIARCH'))")
%python3_ext_suffix %(RPM_BUILD_ROOT= %{__python3} -Ic "import sysconfig; print(sysconfig.get_config_var('EXT_SUFFIX'))")
%python3_cache_tag %(RPM_BUILD_ROOT= %{__python3} -Ic "import sys; print(sys.implementation.cache_tag)")
%_py3_shebang_s s
%_py3_shebang_P %(RPM_BUILD_ROOT= %{__python3} -Ic "import sys; print('P' if hasattr(sys.flags, 'safe_path') else '')")
%py3_shbang_opts -%{?_py3_shebang_s}%{?_py3_shebang_P}
%py3_shebang_fix %{expand:\\\
if [ -z "%{?py3_shebang_flags}" ]; then
shebang_flags="-k"
else
shebang_flags="-ka%{py3_shebang_flags}"
fi
%{__python3} -B %{_rpmconfigdir}/redhat/pathfix_py3_12.py -pni %{__python3} $shebang_flags}
%py3_install() %{expand:\\\
CFLAGS="${CFLAGS:-${RPM_OPT_FLAGS}}" LDFLAGS="${LDFLAGS:-${RPM_LD_FLAGS}}"\\\
%{__python3} %{py_setup} %{?py_setup_args} install -O1 --skip-build --root %{buildroot} --prefix %{_prefix} %{?*}
rm -rfv %{buildroot}%{_bindir}/__pycache__
}
%py3_install_egg() %{expand:\\\
mkdir -p %{buildroot}%{python3_sitelib}
%{__python3} -m easy_install -m --prefix %{buildroot}%{_prefix} -Z dist/*-py%{python3_version}.egg %{?*}
rm -rfv %{buildroot}%{_bindir}/__pycache__
}
%py3_install_wheel() %{expand:\\\
%{__python3} -m pip install -I dist/%{1} --root %{buildroot} --prefix %{_prefix} --no-deps --no-index --no-warn-script-location
rm -rfv %{buildroot}%{_bindir}/__pycache__
for distinfo in %{buildroot}%{python3_sitelib}/*.dist-info %{buildroot}%{python3_sitearch}/*.dist-info; do
if [ -f ${distinfo}/direct_url.json ]; then
rm -fv ${distinfo}/direct_url.json
sed -i '/direct_url.json/d' ${distinfo}/RECORD
fi
done
}
# With $PATH and $PYTHONPATH set to the %%buildroot,
# try to import the Python 3 module(s) given as command-line args or read from file (-f).
# Respect the custom values of %%py3_shebang_flags or set nothing if it's undefined.
# Filter and check import on only top-level modules using -t flag.
# Exclude unwanted modules by passing their globs to -e option.
# Useful as a smoke test in %%check when running tests is not feasible.
# Use spaces or commas as separators if providing list directly.
# Use newlines as separators if providing list in a file.
%py3_check_import(e:tf:) %{expand:\\\
PATH="%{buildroot}%{_bindir}:$PATH"\\\
PYTHONPATH="${PYTHONPATH:-%{buildroot}%{python3_sitearch}:%{buildroot}%{python3_sitelib}}"\\\
_PYTHONSITE="%{buildroot}%{python3_sitearch}:%{buildroot}%{python3_sitelib}"\\\
PYTHONDONTWRITEBYTECODE=1\\\
%{lua:
local command = "%{__python3} "
if rpm.expand("%{?py3_shebang_flags}") ~= "" then
command = command .. "-%{py3_shebang_flags}"
end
command = command .. " %{_rpmconfigdir}/redhat/import_all_modules_py3_12.py "
-- handle multiline arguments correctly, see https://bugzilla.redhat.com/2018809
local args=rpm.expand('%{?**}'):gsub("[%s\\\\]*%s+", " ")
print(command .. args)
}
}
# Environment variables used by %%pytest, %%tox or standalone, e.g.:
# %%{py3_test_envvars} %%{python3} -m unittest
%py3_test_envvars %{expand:\\\
CFLAGS="${CFLAGS:-${RPM_OPT_FLAGS}}" LDFLAGS="${LDFLAGS:-${RPM_LD_FLAGS}}"\\\
PATH="%{buildroot}%{_bindir}:$PATH"\\\
PYTHONPATH="${PYTHONPATH:-%{buildroot}%{python3_sitearch}:%{buildroot}%{python3_sitelib}}"\\\
PYTHONDONTWRITEBYTECODE=1\\\
%{?__pytest_addopts:PYTEST_ADDOPTS="${PYTEST_ADDOPTS:-} %{__pytest_addopts}"}\\\
PYTEST_XDIST_AUTO_NUM_WORKERS=%{_smp_build_ncpus}}
# This is intended for Python 3 only, hence also no Python version in the name.
%__pytest /usr/bin/pytest-%{python3_version}
%pytest %py3_test_envvars %__pytest

View File

@ -1,199 +0,0 @@
#!/usr/bin/env python3
import sys
import os
from stat import *
import getopt
err = sys.stderr.write
dbg = err
rep = sys.stdout.write
new_interpreter = None
preserve_timestamps = False
create_backup = True
keep_flags = False
add_flags = b''
def main():
global new_interpreter
global preserve_timestamps
global create_backup
global keep_flags
global add_flags
usage = ('usage: %s -i /interpreter -p -n -k -a file-or-directory ...\n' %
sys.argv[0])
try:
opts, args = getopt.getopt(sys.argv[1:], 'i:a:kpn')
except getopt.error as msg:
err(str(msg) + '\n')
err(usage)
sys.exit(2)
for o, a in opts:
if o == '-i':
new_interpreter = a.encode()
if o == '-p':
preserve_timestamps = True
if o == '-n':
create_backup = False
if o == '-k':
keep_flags = True
if o == '-a':
add_flags = a.encode()
if b' ' in add_flags:
err("-a option doesn't support whitespaces")
sys.exit(2)
if not new_interpreter or not new_interpreter.startswith(b'/') or \
not args:
err('-i option or file-or-directory missing\n')
err(usage)
sys.exit(2)
bad = 0
for arg in args:
if os.path.isdir(arg):
if recursedown(arg): bad = 1
elif os.path.islink(arg):
err(arg + ': will not process symbolic links\n')
bad = 1
else:
if fix(arg): bad = 1
sys.exit(bad)
def ispython(name):
return name.endswith('.py')
def recursedown(dirname):
dbg('recursedown(%r)\n' % (dirname,))
bad = 0
try:
names = os.listdir(dirname)
except OSError as msg:
err('%s: cannot list directory: %r\n' % (dirname, msg))
return 1
names.sort()
subdirs = []
for name in names:
if name in (os.curdir, os.pardir): continue
fullname = os.path.join(dirname, name)
if os.path.islink(fullname): pass
elif os.path.isdir(fullname):
subdirs.append(fullname)
elif ispython(name):
if fix(fullname): bad = 1
for fullname in subdirs:
if recursedown(fullname): bad = 1
return bad
def fix(filename):
## dbg('fix(%r)\n' % (filename,))
try:
f = open(filename, 'rb')
except IOError as msg:
err('%s: cannot open: %r\n' % (filename, msg))
return 1
with f:
line = f.readline()
fixed = fixline(line)
if line == fixed:
rep(filename+': no change\n')
return
head, tail = os.path.split(filename)
tempname = os.path.join(head, '@' + tail)
try:
g = open(tempname, 'wb')
except IOError as msg:
err('%s: cannot create: %r\n' % (tempname, msg))
return 1
with g:
rep(filename + ': updating\n')
g.write(fixed)
BUFSIZE = 8*1024
while 1:
buf = f.read(BUFSIZE)
if not buf: break
g.write(buf)
# Finishing touch -- move files
mtime = None
atime = None
# First copy the file's mode to the temp file
try:
statbuf = os.stat(filename)
mtime = statbuf.st_mtime
atime = statbuf.st_atime
os.chmod(tempname, statbuf[ST_MODE] & 0o7777)
except OSError as msg:
err('%s: warning: chmod failed (%r)\n' % (tempname, msg))
# Then make a backup of the original file as filename~
if create_backup:
try:
os.rename(filename, filename + '~')
except OSError as msg:
err('%s: warning: backup failed (%r)\n' % (filename, msg))
else:
try:
os.remove(filename)
except OSError as msg:
err('%s: warning: removing failed (%r)\n' % (filename, msg))
# Now move the temp file to the original file
try:
os.rename(tempname, filename)
except OSError as msg:
err('%s: rename failed (%r)\n' % (filename, msg))
return 1
if preserve_timestamps:
if atime and mtime:
try:
os.utime(filename, (atime, mtime))
except OSError as msg:
err('%s: reset of timestamp failed (%r)\n' % (filename, msg))
return 1
# Return success
return 0
def parse_shebang(shebangline):
shebangline = shebangline.rstrip(b'\n')
start = shebangline.find(b' -')
if start == -1:
return b''
return shebangline[start:]
def populate_flags(shebangline):
old_flags = b''
if keep_flags:
old_flags = parse_shebang(shebangline)
if old_flags:
old_flags = old_flags[2:]
if not (old_flags or add_flags):
return b''
# On Linux, the entire string following the interpreter name
# is passed as a single argument to the interpreter.
# e.g. "#! /usr/bin/python3 -W Error -s" runs "/usr/bin/python3 "-W Error -s"
# so shebang should have single '-' where flags are given and
# flag might need argument for that reasons adding new flags is
# between '-' and original flags
# e.g. #! /usr/bin/python3 -sW Error
return b' -' + add_flags + old_flags
def fixline(line):
if not line.startswith(b'#!'):
return line
if b"python" not in line:
return line
flags = populate_flags(line)
return b'#! ' + new_interpreter + flags + b'\n'
if __name__ == '__main__':
main()

View File

@ -20,7 +20,7 @@ URL: https://www.python.org/
#global prerel ...
%global upstream_version %{general_version}%{?prerel}
Version: %{general_version}%{?prerel:~%{prerel}}
Release: 4%{?dist}
Release: 4%{?dist}.4
License: Python
@ -62,6 +62,7 @@ License: Python
# Whether to use RPM build wheels from the python-{pip,setuptools,wheel}-wheel packages
# Uses upstream bundled prebuilt wheels otherwise
%bcond_without rpmwheels
# If the rpmwheels condition is disabled, we use the bundled wheel packages
# from Python with the versions below.
# This needs to be manually updated when we update Python.
@ -313,11 +314,6 @@ Source1: %{url}ftp/python/%{general_version}/Python-%{upstream_version}.tar.xz.a
# The release manager for Python 3.12 is Thomas Wouters
Source2: https://github.com/Yhg1s.gpg
# Sources for the python3.12-rpm-macros
Source3: macros.python3.12
Source4: import_all_modules_py3_12.py
Source5: pathfix_py3_12.py
# A simple script to check timestamps of bytecode files
# Run in check section with Python that is currently being built
# Originally written by bkabrda
@ -389,6 +385,46 @@ Patch397: 00397-tarfile-filter.patch
# Thomas Dwyer.
Patch415: 00415-cve-2023-27043-gh-102988-reject-malformed-addresses-in-email-parseaddr-111116.patch
# 00422 # a353cebef737c41420dc7ae2469dd657371b8881
# gh-115133: Fix tests for XMLPullParser with Expat 2.6.0
#
# Feeding the parser by too small chunks defers parsing to prevent
# CVE-2023-52425. Future versions of Expat may be more reactive.
Patch422: 00422-gh-115133-fix-tests-for-xmlpullparser-with-expat-2-6-0.patch
# 00435 #
# Security fix for CVE-2024-6923
# gh-121650: Encode newlines in headers, and verify headers are sound
#
# Encode header parts that contain newlines
#
# 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.
#
# Verify that email headers are well-formed
#
# This should fail for custom fold() implementations that aren't careful about newlines.
# Tracking bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2302255
# Resolved upstream: https://github.com/python/cpython/issues/121650
Patch435: 00435-CVE-2024-6923.patch
# 00436 # c76cc2aa3a2c30375ade4859b732ada851cc89ed
# [CVE-2024-8088] gh-122905: Sanitize names in zipfile.Path.
Patch436: 00436-cve-2024-8088-gh-122905-sanitize-names-in-zipfile-path.patch
# CVE-2024-6232: Remove backtracking when parsing tarfile headers
# Resolved upstream: https://github.com/python/cpython/issues/121285
Patch437: 00437-CVE-2024-6232.patch
# (New patches go here ^^^)
#
# When adding new patches to "python" and "python3" in Fedora, EL, etc.,
@ -407,14 +443,6 @@ Patch415: 00415-cve-2023-27043-gh-102988-reject-malformed-addresses-in-email-par
# Descriptions, and metadata for subpackages
# ==========================================
# Require alternatives version that implements the --keep-foreign flag and fixes rhbz#2203820
Requires: alternatives >= 1.19.2-1
Requires(post): alternatives >= 1.19.2-1
Requires(postun): alternatives >= 1.19.2-1
# When the user tries to `yum install python`, yum will list this package among
# the possible alternatives
Provides: alternative-for(python)
%if %{with main_python}
# Description for the python3X SRPM only:
@ -489,7 +517,6 @@ Documentation for Python is provided in the %{pkgname}-docs package.
Packages containing additional libraries for Python are generally named with
the "%{pkgname}-" prefix.
For the unversioned "python" executable, see manual page "unversioned-python".
%if %{with main_python}
# https://fedoraproject.org/wiki/Changes/Move_usr_bin_python_into_separate_package
@ -570,12 +597,9 @@ Requires: (python3-rpm-macros if rpm-build)
# On Fedora, we keep this to avoid one additional round of %%generate_buildrequires.
%{!?rhel:Requires: (pyproject-rpm-macros if rpm-build)}
# Require alternatives version that implements the --keep-foreign flag and fixes rhbz#2203820
Requires(postun): alternatives >= 1.19.2-1
# python3.12 installs the alternatives master symlink to which we attach a slave
Requires(post): %{pkgname}
Requires(postun): %{pkgname}
# We provide the python3.12-rpm-macros here to make it possible to
# BuildRequire them in the same manner as RHEL8.
Provides: %{pkgname}-rpm-macros = %{version}-%{release}
%unversioned_obsoletes_of_python3_X_if_main devel
@ -624,13 +648,6 @@ Provides: idle = %{version}-%{release}
Provides: %{pkgname}-tools = %{version}-%{release}
Provides: %{pkgname}-tools%{?_isa} = %{version}-%{release}
# Require alternatives version that implements the --keep-foreign flag and fixes rhbz#2203820
Requires(postun): alternatives >= 1.19.2-1
# python3.12 installs the alternatives master symlink to which we attach a slave
Requires(post): %{pkgname}
Requires(postun): %{pkgname}
%description -n %{pkgname}-idle
IDLE is Pythons Integrated Development and Learning Environment.
@ -702,13 +719,6 @@ Requires: %{pkgname}-idle%{?_isa} = %{version}-%{release}
%unversioned_obsoletes_of_python3_X_if_main debug
# Require alternatives version that implements the --keep-foreign flag and fixes rhbz#2203820
Requires(postun): alternatives >= 1.19.2-1
# python3.12 installs the alternatives master symlink to which we attach a slave
Requires(post): %{pkgname}
Requires(postun): %{pkgname}
%description -n %{pkgname}-debug
python3-debug provides a version of the Python runtime with numerous debugging
features enabled, aimed at advanced Python users such as developers of Python
@ -727,24 +737,6 @@ The debug runtime additionally supports debug builds of C-API extensions
%endif # with debug_build
# We package the python3.12-rpm-macros in RHEL8 as to properly set the
# %%__python3 and %%python3_pkgversion macros as well as provide modern
# versions the current base macros.
%package -n %{pkgname}-rpm-macros
Summary: RPM macros for building RPMs with Python %{pybasever}
License: MIT
Provides: python-modular-rpm-macros == %{pybasever}
Conflicts: python-modular-rpm-macros > %{pybasever}
Requires: python3-rpm-macros
BuildArch: noarch
%description -n %{pkgname}-rpm-macros
RPM macros for building RPMs with Python %{pybasever} from the python%{pyshortver} module.
If you want to build an RPM against the python%{pyshortver} module, you need to add:
BuildRequire: %{pkgname}-rpm-macros.
# ======================================================
# The prep phase of the build:
# ======================================================
@ -753,6 +745,14 @@ If you want to build an RPM against the python%{pyshortver} module, you need to
%gpgverify -k2 -s1 -d0
%autosetup -S git_am -n Python-%{upstream_version}
# Verify the second level of bundled provides is up to date
# Arguably this should be done in %%check, but %%prep has a faster feedback loop
# setuptools.whl does not contain the vendored.txt files
if [ -f %{_rpmconfigdir}/pythonbundles.py ]; then
%{_rpmconfigdir}/pythonbundles.py <(unzip -p Lib/ensurepip/_bundled/pip-*.whl pip/_vendor/vendor.txt) --compare-with '%pip_bundled_provides'
%{_rpmconfigdir}/pythonbundles.py <(unzip -p Lib/test/wheel-*.whl wheel/vendored/vendor.txt) --compare-with '%wheel_bundled_provides'
fi
%if %{with rpmwheels}
rm Lib/ensurepip/_bundled/pip-%{pip_version}-py3-none-any.whl
rm Lib/test/setuptools-%{setuptools_version}-py3-none-any.whl
@ -1059,11 +1059,8 @@ done
# Switch all shebangs to refer to the specific Python version.
# This currently only covers files matching ^[a-zA-Z0-9_]+\.py$,
# so handle files named using other naming scheme separately.
# - RHEL 8 note: we use %%{SOURCE5} instead of pathfix.py, because in RHEL 8 we
# ship our own versioned pathfix_py3_12.py in this package, but during
# bootstrap it's not yet installed.
LD_LIBRARY_PATH=./build/optimized ./build/optimized/python \
%{SOURCE5} \
%{_rpmconfigdir}/redhat/pathfix.py \
-i "%{_bindir}/python%{pybasever}" -pn \
%{buildroot} \
%{buildroot}%{_bindir}/*%{pybasever}.py \
@ -1165,29 +1162,6 @@ for file in %{buildroot}%{pylibdir}/pydoc_data/topics.py $(grep --include='*.py'
rm ${directory}/{__pycache__/${module}.cpython-%{pyshortver}.opt-?.pyc,${module}.py}
done
# Python RPM macros for python3.12-rpm-macros
mkdir -p %{buildroot}%{rpmmacrodir}/
install -m 644 %{SOURCE3} \
%{buildroot}/%{rpmmacrodir}/
# Add scripts that are being used by python3.12-rpm-macros
mkdir -p %{buildroot}%{_rpmconfigdir}/redhat
install -m 644 %{SOURCE4} %{buildroot}%{_rpmconfigdir}/redhat/
install -m 644 %{SOURCE5} %{buildroot}%{_rpmconfigdir}/redhat/
# All ghost files controlled by alternatives need to exist for the files
# section check to succeed
# - Don't list /usr/bin/python as a ghost file so `yum install /usr/bin/python`
# doesn't install this package
touch %{buildroot}%{_bindir}/unversioned-python
touch %{buildroot}%{_mandir}/man1/python.1.gz
touch %{buildroot}%{_bindir}/python3
touch %{buildroot}%{_mandir}/man1/python3.1.gz
touch %{buildroot}%{_bindir}/pydoc3
touch %{buildroot}%{_bindir}/pydoc-3
touch %{buildroot}%{_bindir}/idle3
touch %{buildroot}%{_bindir}/python3-config
# ======================================================
# Checks for packaging issues
# ======================================================
@ -1272,119 +1246,10 @@ CheckPython optimized
%endif # with tests
# ======================================================
# Scriptlets for alternatives on rhel8
# ======================================================
%post
# Alternative for /usr/bin/python -> /usr/bin/python3 + man page
alternatives --install %{_bindir}/unversioned-python \
python \
%{_bindir}/python3 \
300 \
--slave %{_bindir}/python \
unversioned-python \
%{_bindir}/python3 \
--slave %{_mandir}/man1/python.1.gz \
unversioned-python-man \
%{_mandir}/man1/python3.1.gz
# Alternative for /usr/bin/python -> /usr/bin/python3.12 + man page
alternatives --install %{_bindir}/unversioned-python \
python \
%{_bindir}/python3.12 \
211 \
--slave %{_bindir}/python \
unversioned-python \
%{_bindir}/python3.12 \
--slave %{_mandir}/man1/python.1.gz \
unversioned-python-man \
%{_mandir}/man1/python3.12.1.gz
# Alternative for /usr/bin/python3 -> /usr/bin/python3.12 + related files
# Create only if it doesn't exist already
EXISTS=`alternatives --display python3 | \
grep -c "^/usr/bin/python3.12 - priority [0-9]*"`
if [ $EXISTS -eq 0 ]; then
alternatives --install %{_bindir}/python3 \
python3 \
%{_bindir}/python3.12 \
31200 \
--slave %{_mandir}/man1/python3.1.gz \
python3-man \
%{_mandir}/man1/python3.12.1.gz \
--slave %{_bindir}/pydoc3 \
pydoc3 \
%{_bindir}/pydoc3.12 \
--slave %{_bindir}/pydoc-3 \
pydoc-3 \
%{_bindir}/pydoc3.12
fi
%postun
# Do this only during uninstall process (not during update)
if [ $1 -eq 0 ]; then
alternatives --keep-foreign --remove python \
%{_bindir}/python3.12
alternatives --keep-foreign --remove python3 \
%{_bindir}/python3.12
# Remove link python → python3 if no other python3.* exists
if ! alternatives --display python3 > /dev/null; then
alternatives --keep-foreign --remove python \
%{_bindir}/python3
fi
fi
%post devel
alternatives --add-slave python3 %{_bindir}/python3.12 \
%{_bindir}/python3-config \
python3-config \
%{_bindir}/python3.12-config
%postun devel
# Do this only during uninstall process (not during update)
if [ $1 -eq 0 ]; then
alternatives --keep-foreign --remove-slave python3 %{_bindir}/python3.12 \
python3-config
fi
%post idle
alternatives --add-slave python3 %{_bindir}/python3.12 \
%{_bindir}/idle3 \
idle3 \
%{_bindir}/idle3.12
%postun idle
# Do this only during uninstall process (not during update)
if [ $1 -eq 0 ]; then
alternatives --keep-foreign --remove-slave python3 %{_bindir}/python3.12 \
idle3
fi
# ======================================================
# Files for each RPM (sub)package
# ======================================================
%files -n %{pkgname}-rpm-macros
%{rpmmacrodir}/macros.python%{pybasever}
%{_rpmconfigdir}/redhat/import_all_modules_py3_12.py
%{_rpmconfigdir}/redhat/pathfix_py3_12.py
%files -n %{pkgname}
%doc README.rst
# Alternatives
%ghost %{_bindir}/unversioned-python
%ghost %{_mandir}/man1/python.1.gz
%ghost %{_bindir}/python3
%ghost %{_mandir}/man1/python3.1.gz
%ghost %{_bindir}/pydoc3
%ghost %{_bindir}/pydoc-3
%if %{with main_python}
%{_bindir}/pydoc*
%{_bindir}/python3
@ -1676,9 +1541,6 @@ fi
%{_bindir}/python%{pybasever}-config
%{_bindir}/python%{LDVERSION_optimized}-config
%{_bindir}/python%{LDVERSION_optimized}-*-config
# Alternatives
%ghost %{_bindir}/python3-config
%{_libdir}/libpython%{LDVERSION_optimized}.so
%{_libdir}/pkgconfig/python-%{LDVERSION_optimized}.pc
%{_libdir}/pkgconfig/python-%{LDVERSION_optimized}-embed.pc
@ -1691,8 +1553,6 @@ fi
%{_bindir}/idle*
%else
%{_bindir}/idle%{pybasever}
# Alternatives
%ghost %{_bindir}/idle3
%endif
%{pylibdir}/idlelib
@ -1876,6 +1736,23 @@ fi
# ======================================================
%changelog
* Mon Oct 07 2024 Arti Agrawal <artagraw@redhat.com> - 3.12.1-4.4
- Security fix for CVE-2024-6232
Resolves: RHEL-57416
* Fri Aug 23 2024 Charalampos Stratakis <cstratak@redhat.com> - 3.12.1-4.3
- Security fix for CVE-2024-8088
Resolves: RHEL-55964
* Mon Aug 12 2024 Charalampos Stratakis <cstratak@redhat.com> - 3.12.1-4.2
- Security fix for CVE-2024-6923
Resolves: RHEL-53087
* Fri May 03 2024 Lumír Balhar <lbalhar@redhat.com> - 3.12.1-4.1
- Fix tests for XMLPullParser with Expat with fixed CVE
- Enable importing of hash-based .pyc files under FIPS mode
Resolves: RHEL-40773
* Mon Feb 19 2024 Charalampos Stratakis <cstratak@redhat.com> - 3.12.1-4
- Add Red Hat configuration for CVE-2007-4559
@ -1932,4 +1809,3 @@ fi
Ville Skyttä <ville.skytta@iki.fi>
Yaakov Selkowitz <yselkowi@redhat.com>
Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>