Compare commits

...

5 Commits

Author SHA1 Message Date
eabdullin d0c83e320b import CS python3-3.6.8-62.el8 2024-05-23 15:19:49 +00:00
eabdullin c51dc1346c import UBI python3-3.6.8-59.el8 2024-05-22 14:38:54 +00:00
eabdullin a37af4d525 import UBI python3-3.6.8-56.el8_9.3 2024-01-15 19:59:02 +00:00
eabdullin 346df28398 import UBI python3-3.6.8-56.el8_9.2 2024-01-10 14:24:29 +00:00
eabdullin 1563e29a86 import UBI python3-3.6.8-56.el8_9 2023-11-14 20:00:14 +00:00
9 changed files with 6011 additions and 5 deletions

File diff suppressed because it is too large Load Diff

View File

@ -0,0 +1,143 @@
From c563f409ea30bcb0623d785428c9257917371b76 Mon Sep 17 00:00:00 2001
From: "Miss Islington (bot)"
<31488909+miss-islington@users.noreply.github.com>
Date: Thu, 23 Jan 2020 06:49:19 -0800
Subject: [PATCH] bpo-39421: Fix posible crash in heapq with custom comparison
operators (GH-18118) (GH-18146)
(cherry picked from commit 79f89e6e5a659846d1068e8b1bd8e491ccdef861)
Co-authored-by: Pablo Galindo <Pablogsal@gmail.com>
---
Lib/test/test_heapq.py | 31 ++++++++++++++++
.../2020-01-22-15-53-37.bpo-39421.O3nG7u.rst | 2 ++
Modules/_heapqmodule.c | 35 ++++++++++++++-----
3 files changed, 59 insertions(+), 9 deletions(-)
create mode 100644 Misc/NEWS.d/next/Core and Builtins/2020-01-22-15-53-37.bpo-39421.O3nG7u.rst
diff --git a/Lib/test/test_heapq.py b/Lib/test/test_heapq.py
index 2f8c648d84a58..7c3fb0210f69b 100644
--- a/Lib/test/test_heapq.py
+++ b/Lib/test/test_heapq.py
@@ -414,6 +414,37 @@ def test_heappop_mutating_heap(self):
with self.assertRaises((IndexError, RuntimeError)):
self.module.heappop(heap)
+ def test_comparison_operator_modifiying_heap(self):
+ # See bpo-39421: Strong references need to be taken
+ # when comparing objects as they can alter the heap
+ class EvilClass(int):
+ def __lt__(self, o):
+ heap.clear()
+ return NotImplemented
+
+ heap = []
+ self.module.heappush(heap, EvilClass(0))
+ self.assertRaises(IndexError, self.module.heappushpop, heap, 1)
+
+ def test_comparison_operator_modifiying_heap_two_heaps(self):
+
+ class h(int):
+ def __lt__(self, o):
+ list2.clear()
+ return NotImplemented
+
+ class g(int):
+ def __lt__(self, o):
+ list1.clear()
+ return NotImplemented
+
+ list1, list2 = [], []
+
+ self.module.heappush(list1, h(0))
+ self.module.heappush(list2, g(0))
+
+ self.assertRaises((IndexError, RuntimeError), self.module.heappush, list1, g(1))
+ self.assertRaises((IndexError, RuntimeError), self.module.heappush, list2, h(1))
class TestErrorHandlingPython(TestErrorHandling, TestCase):
module = py_heapq
diff --git a/Misc/NEWS.d/next/Core and Builtins/2020-01-22-15-53-37.bpo-39421.O3nG7u.rst b/Misc/NEWS.d/next/Core and Builtins/2020-01-22-15-53-37.bpo-39421.O3nG7u.rst
new file mode 100644
index 0000000000000..bae008150ee12
--- /dev/null
+++ b/Misc/NEWS.d/next/Core and Builtins/2020-01-22-15-53-37.bpo-39421.O3nG7u.rst
@@ -0,0 +1,2 @@
+Fix possible crashes when operating with the functions in the :mod:`heapq`
+module and custom comparison operators.
diff --git a/Modules/_heapqmodule.c b/Modules/_heapqmodule.c
index b499e1f668aae..0fb35ffe5ec48 100644
--- a/Modules/_heapqmodule.c
+++ b/Modules/_heapqmodule.c
@@ -29,7 +29,11 @@ siftdown(PyListObject *heap, Py_ssize_t startpos, Py_ssize_t pos)
while (pos > startpos) {
parentpos = (pos - 1) >> 1;
parent = arr[parentpos];
+ Py_INCREF(newitem);
+ Py_INCREF(parent);
cmp = PyObject_RichCompareBool(newitem, parent, Py_LT);
+ Py_DECREF(parent);
+ Py_DECREF(newitem);
if (cmp < 0)
return -1;
if (size != PyList_GET_SIZE(heap)) {
@@ -71,10 +75,13 @@ siftup(PyListObject *heap, Py_ssize_t pos)
/* Set childpos to index of smaller child. */
childpos = 2*pos + 1; /* leftmost child position */
if (childpos + 1 < endpos) {
- cmp = PyObject_RichCompareBool(
- arr[childpos],
- arr[childpos + 1],
- Py_LT);
+ PyObject* a = arr[childpos];
+ PyObject* b = arr[childpos + 1];
+ Py_INCREF(a);
+ Py_INCREF(b);
+ cmp = PyObject_RichCompareBool(a, b, Py_LT);
+ Py_DECREF(a);
+ Py_DECREF(b);
if (cmp < 0)
return -1;
childpos += ((unsigned)cmp ^ 1); /* increment when cmp==0 */
@@ -229,7 +236,10 @@ heappushpop(PyObject *self, PyObject *args)
return item;
}
- cmp = PyObject_RichCompareBool(PyList_GET_ITEM(heap, 0), item, Py_LT);
+ PyObject* top = PyList_GET_ITEM(heap, 0);
+ Py_INCREF(top);
+ cmp = PyObject_RichCompareBool(top, item, Py_LT);
+ Py_DECREF(top);
if (cmp < 0)
return NULL;
if (cmp == 0) {
@@ -383,7 +393,11 @@ siftdown_max(PyListObject *heap, Py_ssize_t startpos, Py_ssize_t pos)
while (pos > startpos) {
parentpos = (pos - 1) >> 1;
parent = arr[parentpos];
+ Py_INCREF(parent);
+ Py_INCREF(newitem);
cmp = PyObject_RichCompareBool(parent, newitem, Py_LT);
+ Py_DECREF(parent);
+ Py_DECREF(newitem);
if (cmp < 0)
return -1;
if (size != PyList_GET_SIZE(heap)) {
@@ -425,10 +439,13 @@ siftup_max(PyListObject *heap, Py_ssize_t pos)
/* Set childpos to index of smaller child. */
childpos = 2*pos + 1; /* leftmost child position */
if (childpos + 1 < endpos) {
- cmp = PyObject_RichCompareBool(
- arr[childpos + 1],
- arr[childpos],
- Py_LT);
+ PyObject* a = arr[childpos + 1];
+ PyObject* b = arr[childpos];
+ Py_INCREF(a);
+ Py_INCREF(b);
+ cmp = PyObject_RichCompareBool(a, b, Py_LT);
+ Py_DECREF(a);
+ Py_DECREF(b);
if (cmp < 0)
return -1;
childpos += ((unsigned)cmp ^ 1); /* increment when cmp==0 */

View File

@ -0,0 +1,560 @@
From cec66eefbee76ee95f252a52de0f5abc1b677f5b Mon Sep 17 00:00:00 2001
From: Lumir Balhar <lbalhar@redhat.com>
Date: Tue, 12 Dec 2023 13:03:42 +0100
Subject: [PATCH] [3.6] bpo-42103: Improve validation of Plist files.
(GH-22882) (GH-23118)
* Prevent some possible DoS attacks via providing invalid Plist files
with extremely large number of objects or collection sizes.
* Raise InvalidFileException for too large bytes and string size instead of returning garbage.
* Raise InvalidFileException instead of ValueError for specific invalid datetime (NaN).
* Raise InvalidFileException instead of TypeError for non-hashable dict keys.
* Add more tests for invalid Plist files..
(cherry picked from commit 34637a0ce21e7261b952fbd9d006474cc29b681f)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
---
Lib/plistlib.py | 34 +-
Lib/test/test_plistlib.py | 394 +++++++++++++++---
.../2020-10-23-19-20-14.bpo-42103.C5obK2.rst | 3 +
.../2020-10-23-19-19-30.bpo-42103.cILT66.rst | 2 +
4 files changed, 366 insertions(+), 67 deletions(-)
create mode 100644 Misc/NEWS.d/next/Library/2020-10-23-19-20-14.bpo-42103.C5obK2.rst
create mode 100644 Misc/NEWS.d/next/Security/2020-10-23-19-19-30.bpo-42103.cILT66.rst
diff --git a/Lib/plistlib.py b/Lib/plistlib.py
index a918643..df1f346 100644
--- a/Lib/plistlib.py
+++ b/Lib/plistlib.py
@@ -626,7 +626,7 @@ class _BinaryPlistParser:
return self._read_object(top_object)
except (OSError, IndexError, struct.error, OverflowError,
- UnicodeDecodeError):
+ ValueError):
raise InvalidFileException()
def _get_size(self, tokenL):
@@ -642,7 +642,7 @@ class _BinaryPlistParser:
def _read_ints(self, n, size):
data = self._fp.read(size * n)
if size in _BINARY_FORMAT:
- return struct.unpack('>' + _BINARY_FORMAT[size] * n, data)
+ return struct.unpack(f'>{n}{_BINARY_FORMAT[size]}', data)
else:
if not size or len(data) != size * n:
raise InvalidFileException()
@@ -701,19 +701,25 @@ class _BinaryPlistParser:
elif tokenH == 0x40: # data
s = self._get_size(tokenL)
- if self._use_builtin_types:
- result = self._fp.read(s)
- else:
- result = Data(self._fp.read(s))
+ result = self._fp.read(s)
+ if len(result) != s:
+ raise InvalidFileException()
+ if not self._use_builtin_types:
+ result = Data(result)
elif tokenH == 0x50: # ascii string
s = self._get_size(tokenL)
- result = self._fp.read(s).decode('ascii')
- result = result
+ data = self._fp.read(s)
+ if len(data) != s:
+ raise InvalidFileException()
+ result = data.decode('ascii')
elif tokenH == 0x60: # unicode string
- s = self._get_size(tokenL)
- result = self._fp.read(s * 2).decode('utf-16be')
+ s = self._get_size(tokenL) * 2
+ data = self._fp.read(s)
+ if len(data) != s:
+ raise InvalidFileException()
+ result = data.decode('utf-16be')
# tokenH == 0x80 is documented as 'UID' and appears to be used for
# keyed-archiving, not in plists.
@@ -737,9 +743,11 @@ class _BinaryPlistParser:
obj_refs = self._read_refs(s)
result = self._dict_type()
self._objects[ref] = result
- for k, o in zip(key_refs, obj_refs):
- result[self._read_object(k)] = self._read_object(o)
-
+ try:
+ for k, o in zip(key_refs, obj_refs):
+ result[self._read_object(k)] = self._read_object(o)
+ except TypeError:
+ raise InvalidFileException()
else:
raise InvalidFileException()
diff --git a/Lib/test/test_plistlib.py b/Lib/test/test_plistlib.py
index d47c607..f71245d 100644
--- a/Lib/test/test_plistlib.py
+++ b/Lib/test/test_plistlib.py
@@ -1,5 +1,6 @@
# Copyright (C) 2003-2013 Python Software Foundation
+import struct
import unittest
import plistlib
import os
@@ -90,6 +91,284 @@ TESTDATA={
xQHHAsQC0gAAAAAAAAIBAAAAAAAAADkAAAAAAAAAAAAAAAAAAALs'''),
}
+INVALID_BINARY_PLISTS = [
+ ('too short data',
+ b''
+ ),
+ ('too large offset_table_offset and offset_size = 1',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x2a'
+ ),
+ ('too large offset_table_offset and nonstandard offset_size',
+ b'\x00\x00\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x03\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x2c'
+ ),
+ ('integer overflow in offset_table_offset',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\xff\xff\xff\xff\xff\xff\xff\xff'
+ ),
+ ('too large top_object',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('integer overflow in top_object',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\xff\xff\xff\xff\xff\xff\xff\xff'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('too large num_objects and offset_size = 1',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\xff'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('too large num_objects and nonstandard offset_size',
+ b'\x00\x00\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x03\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\xff'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('extremally large num_objects (32 bit)',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x7f\xff\xff\xff'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('extremally large num_objects (64 bit)',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\xff\xff\xff\xff\xff'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('integer overflow in num_objects',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\xff\xff\xff\xff\xff\xff\xff\xff'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('offset_size = 0',
+ b'\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('ref_size = 0',
+ b'\xa1\x01\x00\x08\x0a'
+ b'\x00\x00\x00\x00\x00\x00\x01\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0b'
+ ),
+ ('too large offset',
+ b'\x00\x2a'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('integer overflow in offset',
+ b'\x00\xff\xff\xff\xff\xff\xff\xff\xff'
+ b'\x00\x00\x00\x00\x00\x00\x08\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x09'
+ ),
+ ('too large array size',
+ b'\xaf\x00\x01\xff\x00\x08\x0c'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0d'
+ ),
+ ('extremally large array size (32-bit)',
+ b'\xaf\x02\x7f\xff\xff\xff\x01\x00\x08\x0f'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x10'
+ ),
+ ('extremally large array size (64-bit)',
+ b'\xaf\x03\x00\x00\x00\xff\xff\xff\xff\xff\x01\x00\x08\x13'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x14'
+ ),
+ ('integer overflow in array size',
+ b'\xaf\x03\xff\xff\xff\xff\xff\xff\xff\xff\x01\x00\x08\x13'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x14'
+ ),
+ ('too large reference index',
+ b'\xa1\x02\x00\x08\x0a'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0b'
+ ),
+ ('integer overflow in reference index',
+ b'\xa1\xff\xff\xff\xff\xff\xff\xff\xff\x00\x08\x11'
+ b'\x00\x00\x00\x00\x00\x00\x01\x08'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x12'
+ ),
+ ('too large bytes size',
+ b'\x4f\x00\x23\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0c'
+ ),
+ ('extremally large bytes size (32-bit)',
+ b'\x4f\x02\x7f\xff\xff\xff\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0f'
+ ),
+ ('extremally large bytes size (64-bit)',
+ b'\x4f\x03\x00\x00\x00\xff\xff\xff\xff\xff\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x13'
+ ),
+ ('integer overflow in bytes size',
+ b'\x4f\x03\xff\xff\xff\xff\xff\xff\xff\xff\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x13'
+ ),
+ ('too large ASCII size',
+ b'\x5f\x00\x23\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0c'
+ ),
+ ('extremally large ASCII size (32-bit)',
+ b'\x5f\x02\x7f\xff\xff\xff\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0f'
+ ),
+ ('extremally large ASCII size (64-bit)',
+ b'\x5f\x03\x00\x00\x00\xff\xff\xff\xff\xff\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x13'
+ ),
+ ('integer overflow in ASCII size',
+ b'\x5f\x03\xff\xff\xff\xff\xff\xff\xff\xff\x41\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x13'
+ ),
+ ('invalid ASCII',
+ b'\x51\xff\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0a'
+ ),
+ ('too large UTF-16 size',
+ b'\x6f\x00\x13\x20\xac\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0e'
+ ),
+ ('extremally large UTF-16 size (32-bit)',
+ b'\x6f\x02\x4f\xff\xff\xff\x20\xac\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x11'
+ ),
+ ('extremally large UTF-16 size (64-bit)',
+ b'\x6f\x03\x00\x00\x00\xff\xff\xff\xff\xff\x20\xac\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x15'
+ ),
+ ('integer overflow in UTF-16 size',
+ b'\x6f\x03\xff\xff\xff\xff\xff\xff\xff\xff\x20\xac\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x15'
+ ),
+ ('invalid UTF-16',
+ b'\x61\xd8\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0b'
+ ),
+ ('non-hashable key',
+ b'\xd1\x01\x01\xa0\x08\x0b'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x02'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x0c'
+ ),
+ ('too large datetime (datetime overflow)',
+ b'\x33\x42\x50\x00\x00\x00\x00\x00\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x11'
+ ),
+ ('too large datetime (timedelta overflow)',
+ b'\x33\x42\xe0\x00\x00\x00\x00\x00\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x11'
+ ),
+ ('invalid datetime (Infinity)',
+ b'\x33\x7f\xf0\x00\x00\x00\x00\x00\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x11'
+ ),
+ ('invalid datetime (NaN)',
+ b'\x33\x7f\xf8\x00\x00\x00\x00\x00\x00\x08'
+ b'\x00\x00\x00\x00\x00\x00\x01\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x11'
+ ),
+]
class TestPlistlib(unittest.TestCase):
@@ -447,6 +726,21 @@ class TestPlistlib(unittest.TestCase):
class TestBinaryPlistlib(unittest.TestCase):
+ @staticmethod
+ def decode(*objects, offset_size=1, ref_size=1):
+ data = [b'bplist00']
+ offset = 8
+ offsets = []
+ for x in objects:
+ offsets.append(offset.to_bytes(offset_size, 'big'))
+ data.append(x)
+ offset += len(x)
+ tail = struct.pack('>6xBBQQQ', offset_size, ref_size,
+ len(objects), 0, offset)
+ data.extend(offsets)
+ data.append(tail)
+ return plistlib.loads(b''.join(data), fmt=plistlib.FMT_BINARY)
+
def test_nonstandard_refs_size(self):
# Issue #21538: Refs and offsets are 24-bit integers
data = (b'bplist00'
@@ -461,7 +755,7 @@ class TestBinaryPlistlib(unittest.TestCase):
def test_dump_duplicates(self):
# Test effectiveness of saving duplicated objects
- for x in (None, False, True, 12345, 123.45, 'abcde', b'abcde',
+ for x in (None, False, True, 12345, 123.45, 'abcde', 'абвгд', b'abcde',
datetime.datetime(2004, 10, 26, 10, 33, 33),
plistlib.Data(b'abcde'), bytearray(b'abcde'),
[12, 345], (12, 345), {'12': 345}):
@@ -500,6 +794,20 @@ class TestBinaryPlistlib(unittest.TestCase):
b = plistlib.loads(plistlib.dumps(a, fmt=plistlib.FMT_BINARY))
self.assertIs(b['x'], b)
+ def test_deep_nesting(self):
+ for N in [300, 100000]:
+ chunks = [b'\xa1' + (i + 1).to_bytes(4, 'big') for i in range(N)]
+ try:
+ result = self.decode(*chunks, b'\x54seed', offset_size=4, ref_size=4)
+ except RecursionError:
+ pass
+ else:
+ for i in range(N):
+ self.assertIsInstance(result, list)
+ self.assertEqual(len(result), 1)
+ result = result[0]
+ self.assertEqual(result, 'seed')
+
def test_large_timestamp(self):
# Issue #26709: 32-bit timestamp out of range
for ts in -2**31-1, 2**31:
@@ -509,55 +817,37 @@ class TestBinaryPlistlib(unittest.TestCase):
data = plistlib.dumps(d, fmt=plistlib.FMT_BINARY)
self.assertEqual(plistlib.loads(data), d)
+ def test_load_singletons(self):
+ self.assertIs(self.decode(b'\x00'), None)
+ self.assertIs(self.decode(b'\x08'), False)
+ self.assertIs(self.decode(b'\x09'), True)
+ self.assertEqual(self.decode(b'\x0f'), b'')
+
+ def test_load_int(self):
+ self.assertEqual(self.decode(b'\x10\x00'), 0)
+ self.assertEqual(self.decode(b'\x10\xfe'), 0xfe)
+ self.assertEqual(self.decode(b'\x11\xfe\xdc'), 0xfedc)
+ self.assertEqual(self.decode(b'\x12\xfe\xdc\xba\x98'), 0xfedcba98)
+ self.assertEqual(self.decode(b'\x13\x01\x23\x45\x67\x89\xab\xcd\xef'),
+ 0x0123456789abcdef)
+ self.assertEqual(self.decode(b'\x13\xfe\xdc\xba\x98\x76\x54\x32\x10'),
+ -0x123456789abcdf0)
+
+ def test_unsupported(self):
+ unsupported = [*range(1, 8), *range(10, 15),
+ 0x20, 0x21, *range(0x24, 0x33), *range(0x34, 0x40)]
+ for i in [0x70, 0x90, 0xb0, 0xc0, 0xe0, 0xf0]:
+ unsupported.extend(i + j for j in range(16))
+ for token in unsupported:
+ with self.subTest(f'token {token:02x}'):
+ with self.assertRaises(plistlib.InvalidFileException):
+ self.decode(bytes([token]) + b'\x00'*16)
+
def test_invalid_binary(self):
- for data in [
- # too short data
- b'',
- # too large offset_table_offset and nonstandard offset_size
- b'\x00\x08'
- b'\x00\x00\x00\x00\x00\x00\x03\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x00'
- b'\x00\x00\x00\x00\x00\x00\x00\x2a',
- # integer overflow in offset_table_offset
- b'\x00\x08'
- b'\x00\x00\x00\x00\x00\x00\x01\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x00'
- b'\xff\xff\xff\xff\xff\xff\xff\xff',
- # offset_size = 0
- b'\x00\x08'
- b'\x00\x00\x00\x00\x00\x00\x00\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x00'
- b'\x00\x00\x00\x00\x00\x00\x00\x09',
- # ref_size = 0
- b'\xa1\x01\x00\x08\x0a'
- b'\x00\x00\x00\x00\x00\x00\x01\x00'
- b'\x00\x00\x00\x00\x00\x00\x00\x02'
- b'\x00\x00\x00\x00\x00\x00\x00\x00'
- b'\x00\x00\x00\x00\x00\x00\x00\x0b',
- # integer overflow in offset
- b'\x00\xff\xff\xff\xff\xff\xff\xff\xff'
- b'\x00\x00\x00\x00\x00\x00\x08\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x00'
- b'\x00\x00\x00\x00\x00\x00\x00\x09',
- # invalid ASCII
- b'\x51\xff\x08'
- b'\x00\x00\x00\x00\x00\x00\x01\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x00'
- b'\x00\x00\x00\x00\x00\x00\x00\x0a',
- # invalid UTF-16
- b'\x61\xd8\x00\x08'
- b'\x00\x00\x00\x00\x00\x00\x01\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x01'
- b'\x00\x00\x00\x00\x00\x00\x00\x00'
- b'\x00\x00\x00\x00\x00\x00\x00\x0b',
- ]:
- with self.assertRaises(plistlib.InvalidFileException):
- plistlib.loads(b'bplist00' + data, fmt=plistlib.FMT_BINARY)
+ for name, data in INVALID_BINARY_PLISTS:
+ with self.subTest(name):
+ with self.assertRaises(plistlib.InvalidFileException):
+ plistlib.loads(b'bplist00' + data, fmt=plistlib.FMT_BINARY)
class TestPlistlibDeprecated(unittest.TestCase):
@@ -655,9 +945,5 @@ class MiscTestCase(unittest.TestCase):
support.check__all__(self, plistlib, blacklist=blacklist)
-def test_main():
- support.run_unittest(TestPlistlib, TestPlistlibDeprecated, MiscTestCase)
-
-
if __name__ == '__main__':
- test_main()
+ unittest.main()
diff --git a/Misc/NEWS.d/next/Library/2020-10-23-19-20-14.bpo-42103.C5obK2.rst b/Misc/NEWS.d/next/Library/2020-10-23-19-20-14.bpo-42103.C5obK2.rst
new file mode 100644
index 0000000..4eb694c
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2020-10-23-19-20-14.bpo-42103.C5obK2.rst
@@ -0,0 +1,3 @@
+:exc:`~plistlib.InvalidFileException` and :exc:`RecursionError` are now
+the only errors caused by loading malformed binary Plist file (previously
+ValueError and TypeError could be raised in some specific cases).
diff --git a/Misc/NEWS.d/next/Security/2020-10-23-19-19-30.bpo-42103.cILT66.rst b/Misc/NEWS.d/next/Security/2020-10-23-19-19-30.bpo-42103.cILT66.rst
new file mode 100644
index 0000000..15d7b65
--- /dev/null
+++ b/Misc/NEWS.d/next/Security/2020-10-23-19-19-30.bpo-42103.cILT66.rst
@@ -0,0 +1,2 @@
+Prevented potential DoS attack via CPU and RAM exhaustion when processing
+malformed Apple Property List files in binary format.
--
2.43.0

View File

@ -0,0 +1,88 @@
From 0d02ff99721f7650e39ba4c7d8fe06f412bbb591 Mon Sep 17 00:00:00 2001
From: Victor Stinner <vstinner@python.org>
Date: Wed, 13 Dec 2023 11:50:26 +0100
Subject: [PATCH] bpo-46623: Skip two test_zlib tests on s390x (GH-31096)
Skip test_pair() and test_speech128() of test_zlib on s390x since
they fail if zlib uses the s390x hardware accelerator.
---
Lib/test/test_zlib.py | 32 +++++++++++++++++++
.../2022-02-03-09-45-26.bpo-46623.vxzuhV.rst | 2 ++
2 files changed, 34 insertions(+)
create mode 100644 Misc/NEWS.d/next/Tests/2022-02-03-09-45-26.bpo-46623.vxzuhV.rst
diff --git a/Lib/test/test_zlib.py b/Lib/test/test_zlib.py
index b7170b4..770a425 100644
--- a/Lib/test/test_zlib.py
+++ b/Lib/test/test_zlib.py
@@ -1,6 +1,7 @@
import unittest
from test import support
import binascii
+import os
import pickle
import random
import sys
@@ -15,6 +16,35 @@ requires_Decompress_copy = unittest.skipUnless(
hasattr(zlib.decompressobj(), "copy"),
'requires Decompress.copy()')
+# bpo-46623: On s390x, when a hardware accelerator is used, using different
+# ways to compress data with zlib can produce different compressed data.
+# Simplified test_pair() code:
+#
+# def func1(data):
+# return zlib.compress(data)
+#
+# def func2(data)
+# co = zlib.compressobj()
+# x1 = co.compress(data)
+# x2 = co.flush()
+# return x1 + x2
+#
+# On s390x if zlib uses a hardware accelerator, func1() creates a single
+# "final" compressed block whereas func2() produces 3 compressed blocks (the
+# last one is a final block). On other platforms with no accelerator, func1()
+# and func2() produce the same compressed data made of a single (final)
+# compressed block.
+#
+# Only the compressed data is different, the decompression returns the original
+# data:
+#
+# zlib.decompress(func1(data)) == zlib.decompress(func2(data)) == data
+#
+# Make the assumption that s390x always has an accelerator to simplify the skip
+# condition. Windows doesn't have os.uname() but it doesn't support s390x.
+skip_on_s390x = unittest.skipIf(hasattr(os, 'uname') and os.uname().machine == 's390x',
+ 'skipped on s390x')
+
class VersionTestCase(unittest.TestCase):
@@ -174,6 +204,7 @@ class CompressTestCase(BaseCompressTestCase, unittest.TestCase):
bufsize=zlib.DEF_BUF_SIZE),
HAMLET_SCENE)
+ @skip_on_s390x
def test_speech128(self):
# compress more data
data = HAMLET_SCENE * 128
@@ -225,6 +256,7 @@ class CompressTestCase(BaseCompressTestCase, unittest.TestCase):
class CompressObjectTestCase(BaseCompressTestCase, unittest.TestCase):
# Test compression object
+ @skip_on_s390x
def test_pair(self):
# straightforward compress/decompress objects
datasrc = HAMLET_SCENE * 128
diff --git a/Misc/NEWS.d/next/Tests/2022-02-03-09-45-26.bpo-46623.vxzuhV.rst b/Misc/NEWS.d/next/Tests/2022-02-03-09-45-26.bpo-46623.vxzuhV.rst
new file mode 100644
index 0000000..be085c0
--- /dev/null
+++ b/Misc/NEWS.d/next/Tests/2022-02-03-09-45-26.bpo-46623.vxzuhV.rst
@@ -0,0 +1,2 @@
+Skip test_pair() and test_speech128() of test_zlib on s390x since they fail
+if zlib uses the s390x hardware accelerator. Patch by Victor Stinner.
--
2.43.0

View File

@ -0,0 +1,750 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Victor Stinner <vstinner@python.org>
Date: Fri, 15 Dec 2023 16:10:40 +0100
Subject: [PATCH] 00415: [CVE-2023-27043] gh-102988: Reject malformed addresses
in email.parseaddr() (#111116)
Detect email address parsing errors and return empty tuple to
indicate the parsing error (old API). Add an optional 'strict'
parameter to getaddresses() and parseaddr() functions. Patch by
Thomas Dwyer.
Co-Authored-By: Thomas Dwyer <github@tomd.tel>
---
Doc/library/email.utils.rst | 19 +-
Lib/email/utils.py | 151 ++++++++++++-
Lib/test/test_email/test_email.py | 204 +++++++++++++++++-
...-10-20-15-28-08.gh-issue-102988.dStNO7.rst | 8 +
4 files changed, 361 insertions(+), 21 deletions(-)
create mode 100644 Misc/NEWS.d/next/Library/2023-10-20-15-28-08.gh-issue-102988.dStNO7.rst
diff --git a/Doc/library/email.utils.rst b/Doc/library/email.utils.rst
index 63fae2ab84..d1e1898591 100644
--- a/Doc/library/email.utils.rst
+++ b/Doc/library/email.utils.rst
@@ -60,13 +60,18 @@ of the new API.
begins with angle brackets, they are stripped off.
-.. function:: parseaddr(address)
+.. function:: parseaddr(address, *, strict=True)
Parse address -- which should be the value of some address-containing field such
as :mailheader:`To` or :mailheader:`Cc` -- into its constituent *realname* and
*email address* parts. Returns a tuple of that information, unless the parse
fails, in which case a 2-tuple of ``('', '')`` is returned.
+ If *strict* is true, use a strict parser which rejects malformed inputs.
+
+ .. versionchanged:: 3.13
+ Add *strict* optional parameter and reject malformed inputs by default.
+
.. function:: formataddr(pair, charset='utf-8')
@@ -84,12 +89,15 @@ of the new API.
Added the *charset* option.
-.. function:: getaddresses(fieldvalues)
+.. function:: getaddresses(fieldvalues, *, strict=True)
This method returns a list of 2-tuples of the form returned by ``parseaddr()``.
*fieldvalues* is a sequence of header field values as might be returned by
- :meth:`Message.get_all <email.message.Message.get_all>`. Here's a simple
- example that gets all the recipients of a message::
+ :meth:`Message.get_all <email.message.Message.get_all>`.
+
+ If *strict* is true, use a strict parser which rejects malformed inputs.
+
+ Here's a simple example that gets all the recipients of a message::
from email.utils import getaddresses
@@ -99,6 +107,9 @@ of the new API.
resent_ccs = msg.get_all('resent-cc', [])
all_recipients = getaddresses(tos + ccs + resent_tos + resent_ccs)
+ .. versionchanged:: 3.13
+ Add *strict* optional parameter and reject malformed inputs by default.
+
.. function:: parsedate(date)
diff --git a/Lib/email/utils.py b/Lib/email/utils.py
index 39c2240607..f83b7e5d7e 100644
--- a/Lib/email/utils.py
+++ b/Lib/email/utils.py
@@ -48,6 +48,7 @@ TICK = "'"
specialsre = re.compile(r'[][\\()<>@,:;".]')
escapesre = re.compile(r'[\\"]')
+
def _has_surrogates(s):
"""Return True if s contains surrogate-escaped binary data."""
# This check is based on the fact that unless there are surrogates, utf8
@@ -106,12 +107,127 @@ def formataddr(pair, charset='utf-8'):
return address
+def _iter_escaped_chars(addr):
+ pos = 0
+ escape = False
+ for pos, ch in enumerate(addr):
+ if escape:
+ yield (pos, '\\' + ch)
+ escape = False
+ elif ch == '\\':
+ escape = True
+ else:
+ yield (pos, ch)
+ if escape:
+ yield (pos, '\\')
-def getaddresses(fieldvalues):
- """Return a list of (REALNAME, EMAIL) for each fieldvalue."""
- all = COMMASPACE.join(fieldvalues)
- a = _AddressList(all)
- return a.addresslist
+
+def _strip_quoted_realnames(addr):
+ """Strip real names between quotes."""
+ if '"' not in addr:
+ # Fast path
+ return addr
+
+ start = 0
+ open_pos = None
+ result = []
+ for pos, ch in _iter_escaped_chars(addr):
+ if ch == '"':
+ if open_pos is None:
+ open_pos = pos
+ else:
+ if start != open_pos:
+ result.append(addr[start:open_pos])
+ start = pos + 1
+ open_pos = None
+
+ if start < len(addr):
+ result.append(addr[start:])
+
+ return ''.join(result)
+
+
+supports_strict_parsing = True
+
+def getaddresses(fieldvalues, *, strict=True):
+ """Return a list of (REALNAME, EMAIL) or ('','') for each fieldvalue.
+
+ When parsing fails for a fieldvalue, a 2-tuple of ('', '') is returned in
+ its place.
+
+ If strict is true, use a strict parser which rejects malformed inputs.
+ """
+
+ # If strict is true, if the resulting list of parsed addresses is greater
+ # than the number of fieldvalues in the input list, a parsing error has
+ # occurred and consequently a list containing a single empty 2-tuple [('',
+ # '')] is returned in its place. This is done to avoid invalid output.
+ #
+ # Malformed input: getaddresses(['alice@example.com <bob@example.com>'])
+ # Invalid output: [('', 'alice@example.com'), ('', 'bob@example.com')]
+ # Safe output: [('', '')]
+
+ if not strict:
+ all = COMMASPACE.join(str(v) for v in fieldvalues)
+ a = _AddressList(all)
+ return a.addresslist
+
+ fieldvalues = [str(v) for v in fieldvalues]
+ fieldvalues = _pre_parse_validation(fieldvalues)
+ addr = COMMASPACE.join(fieldvalues)
+ a = _AddressList(addr)
+ result = _post_parse_validation(a.addresslist)
+
+ # Treat output as invalid if the number of addresses is not equal to the
+ # expected number of addresses.
+ n = 0
+ for v in fieldvalues:
+ # When a comma is used in the Real Name part it is not a deliminator.
+ # So strip those out before counting the commas.
+ v = _strip_quoted_realnames(v)
+ # Expected number of addresses: 1 + number of commas
+ n += 1 + v.count(',')
+ if len(result) != n:
+ return [('', '')]
+
+ return result
+
+
+def _check_parenthesis(addr):
+ # Ignore parenthesis in quoted real names.
+ addr = _strip_quoted_realnames(addr)
+
+ opens = 0
+ for pos, ch in _iter_escaped_chars(addr):
+ if ch == '(':
+ opens += 1
+ elif ch == ')':
+ opens -= 1
+ if opens < 0:
+ return False
+ return (opens == 0)
+
+
+def _pre_parse_validation(email_header_fields):
+ accepted_values = []
+ for v in email_header_fields:
+ if not _check_parenthesis(v):
+ v = "('', '')"
+ accepted_values.append(v)
+
+ return accepted_values
+
+
+def _post_parse_validation(parsed_email_header_tuples):
+ accepted_values = []
+ # The parser would have parsed a correctly formatted domain-literal
+ # The existence of an [ after parsing indicates a parsing failure
+ for v in parsed_email_header_tuples:
+ if '[' in v[1]:
+ v = ('', '')
+ accepted_values.append(v)
+
+ return accepted_values
@@ -214,16 +330,33 @@ def parsedate_to_datetime(data):
tzinfo=datetime.timezone(datetime.timedelta(seconds=tz)))
-def parseaddr(addr):
+def parseaddr(addr, *, strict=True):
"""
Parse addr into its constituent realname and email address parts.
Return a tuple of realname and email address, unless the parse fails, in
which case return a 2-tuple of ('', '').
+
+ If strict is True, use a strict parser which rejects malformed inputs.
"""
- addrs = _AddressList(addr).addresslist
- if not addrs:
- return '', ''
+ if not strict:
+ addrs = _AddressList(addr).addresslist
+ if not addrs:
+ return ('', '')
+ return addrs[0]
+
+ if isinstance(addr, list):
+ addr = addr[0]
+
+ if not isinstance(addr, str):
+ return ('', '')
+
+ addr = _pre_parse_validation([addr])[0]
+ addrs = _post_parse_validation(_AddressList(addr).addresslist)
+
+ if not addrs or len(addrs) > 1:
+ return ('', '')
+
return addrs[0]
diff --git a/Lib/test/test_email/test_email.py b/Lib/test/test_email/test_email.py
index e4e40b612f..ce36efc1b1 100644
--- a/Lib/test/test_email/test_email.py
+++ b/Lib/test/test_email/test_email.py
@@ -19,6 +19,7 @@ except ImportError:
import email
import email.policy
+import email.utils
from email.charset import Charset
from email.header import Header, decode_header, make_header
@@ -3207,15 +3208,154 @@ Foo
[('Al Person', 'aperson@dom.ain'),
('Bud Person', 'bperson@dom.ain')])
+ def test_getaddresses_comma_in_name(self):
+ """GH-106669 regression test."""
+ self.assertEqual(
+ utils.getaddresses(
+ [
+ '"Bud, Person" <bperson@dom.ain>',
+ 'aperson@dom.ain (Al Person)',
+ '"Mariusz Felisiak" <to@example.com>',
+ ]
+ ),
+ [
+ ('Bud, Person', 'bperson@dom.ain'),
+ ('Al Person', 'aperson@dom.ain'),
+ ('Mariusz Felisiak', 'to@example.com'),
+ ],
+ )
+
+ def test_parsing_errors(self):
+ """Test for parsing errors from CVE-2023-27043 and CVE-2019-16056"""
+ alice = 'alice@example.org'
+ bob = 'bob@example.com'
+ empty = ('', '')
+
+ # Test utils.getaddresses() and utils.parseaddr() on malformed email
+ # addresses: default behavior (strict=True) rejects malformed address,
+ # and strict=False which tolerates malformed address.
+ for invalid_separator, expected_non_strict in (
+ ('(', [(f'<{bob}>', alice)]),
+ (')', [('', alice), empty, ('', bob)]),
+ ('<', [('', alice), empty, ('', bob), empty]),
+ ('>', [('', alice), empty, ('', bob)]),
+ ('[', [('', f'{alice}[<{bob}>]')]),
+ (']', [('', alice), empty, ('', bob)]),
+ ('@', [empty, empty, ('', bob)]),
+ (';', [('', alice), empty, ('', bob)]),
+ (':', [('', alice), ('', bob)]),
+ ('.', [('', alice + '.'), ('', bob)]),
+ ('"', [('', alice), ('', f'<{bob}>')]),
+ ):
+ address = f'{alice}{invalid_separator}<{bob}>'
+ with self.subTest(address=address):
+ self.assertEqual(utils.getaddresses([address]),
+ [empty])
+ self.assertEqual(utils.getaddresses([address], strict=False),
+ expected_non_strict)
+
+ self.assertEqual(utils.parseaddr([address]),
+ empty)
+ self.assertEqual(utils.parseaddr([address], strict=False),
+ ('', address))
+
+ # Comma (',') is treated differently depending on strict parameter.
+ # Comma without quotes.
+ address = f'{alice},<{bob}>'
+ self.assertEqual(utils.getaddresses([address]),
+ [('', alice), ('', bob)])
+ self.assertEqual(utils.getaddresses([address], strict=False),
+ [('', alice), ('', bob)])
+ self.assertEqual(utils.parseaddr([address]),
+ empty)
+ self.assertEqual(utils.parseaddr([address], strict=False),
+ ('', address))
+
+ # Real name between quotes containing comma.
+ address = '"Alice, alice@example.org" <bob@example.com>'
+ expected_strict = ('Alice, alice@example.org', 'bob@example.com')
+ self.assertEqual(utils.getaddresses([address]), [expected_strict])
+ self.assertEqual(utils.getaddresses([address], strict=False), [expected_strict])
+ self.assertEqual(utils.parseaddr([address]), expected_strict)
+ self.assertEqual(utils.parseaddr([address], strict=False),
+ ('', address))
+
+ # Valid parenthesis in comments.
+ address = 'alice@example.org (Alice)'
+ expected_strict = ('Alice', 'alice@example.org')
+ self.assertEqual(utils.getaddresses([address]), [expected_strict])
+ self.assertEqual(utils.getaddresses([address], strict=False), [expected_strict])
+ self.assertEqual(utils.parseaddr([address]), expected_strict)
+ self.assertEqual(utils.parseaddr([address], strict=False),
+ ('', address))
+
+ # Invalid parenthesis in comments.
+ address = 'alice@example.org )Alice('
+ self.assertEqual(utils.getaddresses([address]), [empty])
+ self.assertEqual(utils.getaddresses([address], strict=False),
+ [('', 'alice@example.org'), ('', ''), ('', 'Alice')])
+ self.assertEqual(utils.parseaddr([address]), empty)
+ self.assertEqual(utils.parseaddr([address], strict=False),
+ ('', address))
+
+ # Two addresses with quotes separated by comma.
+ address = '"Jane Doe" <jane@example.net>, "John Doe" <john@example.net>'
+ self.assertEqual(utils.getaddresses([address]),
+ [('Jane Doe', 'jane@example.net'),
+ ('John Doe', 'john@example.net')])
+ self.assertEqual(utils.getaddresses([address], strict=False),
+ [('Jane Doe', 'jane@example.net'),
+ ('John Doe', 'john@example.net')])
+ self.assertEqual(utils.parseaddr([address]), empty)
+ self.assertEqual(utils.parseaddr([address], strict=False),
+ ('', address))
+
+ # Test email.utils.supports_strict_parsing attribute
+ self.assertEqual(email.utils.supports_strict_parsing, True)
+
def test_getaddresses_nasty(self):
- eq = self.assertEqual
- eq(utils.getaddresses(['foo: ;']), [('', '')])
- eq(utils.getaddresses(
- ['[]*-- =~$']),
- [('', ''), ('', ''), ('', '*--')])
- eq(utils.getaddresses(
- ['foo: ;', '"Jason R. Mastaler" <jason@dom.ain>']),
- [('', ''), ('Jason R. Mastaler', 'jason@dom.ain')])
+ for addresses, expected in (
+ (['"Sürname, Firstname" <to@example.com>'],
+ [('Sürname, Firstname', 'to@example.com')]),
+
+ (['foo: ;'],
+ [('', '')]),
+
+ (['foo: ;', '"Jason R. Mastaler" <jason@dom.ain>'],
+ [('', ''), ('Jason R. Mastaler', 'jason@dom.ain')]),
+
+ ([r'Pete(A nice \) chap) <pete(his account)@silly.test(his host)>'],
+ [('Pete (A nice ) chap his account his host)', 'pete@silly.test')]),
+
+ (['(Empty list)(start)Undisclosed recipients :(nobody(I know))'],
+ [('', '')]),
+
+ (['Mary <@machine.tld:mary@example.net>, , jdoe@test . example'],
+ [('Mary', 'mary@example.net'), ('', ''), ('', 'jdoe@test.example')]),
+
+ (['John Doe <jdoe@machine(comment). example>'],
+ [('John Doe (comment)', 'jdoe@machine.example')]),
+
+ (['"Mary Smith: Personal Account" <smith@home.example>'],
+ [('Mary Smith: Personal Account', 'smith@home.example')]),
+
+ (['Undisclosed recipients:;'],
+ [('', '')]),
+
+ ([r'<boss@nil.test>, "Giant; \"Big\" Box" <bob@example.net>'],
+ [('', 'boss@nil.test'), ('Giant; "Big" Box', 'bob@example.net')]),
+ ):
+ with self.subTest(addresses=addresses):
+ self.assertEqual(utils.getaddresses(addresses),
+ expected)
+ self.assertEqual(utils.getaddresses(addresses, strict=False),
+ expected)
+
+ addresses = ['[]*-- =~$']
+ self.assertEqual(utils.getaddresses(addresses),
+ [('', '')])
+ self.assertEqual(utils.getaddresses(addresses, strict=False),
+ [('', ''), ('', ''), ('', '*--')])
def test_getaddresses_embedded_comment(self):
"""Test proper handling of a nested comment"""
@@ -3397,6 +3537,54 @@ multipart/report
m = cls(*constructor, policy=email.policy.default)
self.assertIs(m.policy, email.policy.default)
+ def test_iter_escaped_chars(self):
+ self.assertEqual(list(utils._iter_escaped_chars(r'a\\b\"c\\"d')),
+ [(0, 'a'),
+ (2, '\\\\'),
+ (3, 'b'),
+ (5, '\\"'),
+ (6, 'c'),
+ (8, '\\\\'),
+ (9, '"'),
+ (10, 'd')])
+ self.assertEqual(list(utils._iter_escaped_chars('a\\')),
+ [(0, 'a'), (1, '\\')])
+
+ def test_strip_quoted_realnames(self):
+ def check(addr, expected):
+ self.assertEqual(utils._strip_quoted_realnames(addr), expected)
+
+ check('"Jane Doe" <jane@example.net>, "John Doe" <john@example.net>',
+ ' <jane@example.net>, <john@example.net>')
+ check(r'"Jane \"Doe\"." <jane@example.net>',
+ ' <jane@example.net>')
+
+ # special cases
+ check(r'before"name"after', 'beforeafter')
+ check(r'before"name"', 'before')
+ check(r'b"name"', 'b') # single char
+ check(r'"name"after', 'after')
+ check(r'"name"a', 'a') # single char
+ check(r'"name"', '')
+
+ # no change
+ for addr in (
+ 'Jane Doe <jane@example.net>, John Doe <john@example.net>',
+ 'lone " quote',
+ ):
+ self.assertEqual(utils._strip_quoted_realnames(addr), addr)
+
+
+ def test_check_parenthesis(self):
+ addr = 'alice@example.net'
+ self.assertTrue(utils._check_parenthesis(f'{addr} (Alice)'))
+ self.assertFalse(utils._check_parenthesis(f'{addr} )Alice('))
+ self.assertFalse(utils._check_parenthesis(f'{addr} (Alice))'))
+ self.assertFalse(utils._check_parenthesis(f'{addr} ((Alice)'))
+
+ # Ignore real name between quotes
+ self.assertTrue(utils._check_parenthesis(f'")Alice((" {addr}'))
+
# Test the iterator/generators
class TestIterators(TestEmailBase):
diff --git a/Misc/NEWS.d/next/Library/2023-10-20-15-28-08.gh-issue-102988.dStNO7.rst b/Misc/NEWS.d/next/Library/2023-10-20-15-28-08.gh-issue-102988.dStNO7.rst
new file mode 100644
index 0000000000..3d0e9e4078
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2023-10-20-15-28-08.gh-issue-102988.dStNO7.rst
@@ -0,0 +1,8 @@
+:func:`email.utils.getaddresses` and :func:`email.utils.parseaddr` now
+return ``('', '')`` 2-tuples in more situations where invalid email
+addresses are encountered instead of potentially inaccurate values. Add
+optional *strict* parameter to these two functions: use ``strict=False`` to
+get the old behavior, accept malformed inputs.
+``getattr(email.utils, 'supports_strict_parsing', False)`` can be use to check
+if the *strict* paramater is available. Patch by Thomas Dwyer and Victor
+Stinner to improve the CVE-2023-27043 fix.
From 4df4fad359c280f2328b98ea9b4414f244624a58 Mon Sep 17 00:00:00 2001
From: Lumir Balhar <lbalhar@redhat.com>
Date: Mon, 18 Dec 2023 20:15:33 +0100
Subject: [PATCH] Make it possible to disable strict parsing in email module
---
Doc/library/email.utils.rst | 26 +++++++++++
Lib/email/utils.py | 54 ++++++++++++++++++++++-
Lib/test/test_email/test_email.py | 72 ++++++++++++++++++++++++++++++-
3 files changed, 149 insertions(+), 3 deletions(-)
diff --git a/Doc/library/email.utils.rst b/Doc/library/email.utils.rst
index d1e1898591..7aef773b5f 100644
--- a/Doc/library/email.utils.rst
+++ b/Doc/library/email.utils.rst
@@ -69,6 +69,19 @@ of the new API.
If *strict* is true, use a strict parser which rejects malformed inputs.
+ The default setting for *strict* is set to ``True``, but you can override
+ it by setting the environment variable ``PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING``
+ to non-empty string.
+
+ Additionally, you can permanently set the default value for *strict* to
+ ``False`` by creating the configuration file ``/etc/python/email.cfg``
+ with the following content:
+
+ .. code-block:: ini
+
+ [email_addr_parsing]
+ PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING = true
+
.. versionchanged:: 3.13
Add *strict* optional parameter and reject malformed inputs by default.
@@ -97,6 +110,19 @@ of the new API.
If *strict* is true, use a strict parser which rejects malformed inputs.
+ The default setting for *strict* is set to ``True``, but you can override
+ it by setting the environment variable ``PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING``
+ to non-empty string.
+
+ Additionally, you can permanently set the default value for *strict* to
+ ``False`` by creating the configuration file ``/etc/python/email.cfg``
+ with the following content:
+
+ .. code-block:: ini
+
+ [email_addr_parsing]
+ PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING = true
+
Here's a simple example that gets all the recipients of a message::
from email.utils import getaddresses
diff --git a/Lib/email/utils.py b/Lib/email/utils.py
index f83b7e5d7e..b8e90ceb8e 100644
--- a/Lib/email/utils.py
+++ b/Lib/email/utils.py
@@ -48,6 +48,46 @@ TICK = "'"
specialsre = re.compile(r'[][\\()<>@,:;".]')
escapesre = re.compile(r'[\\"]')
+_EMAIL_CONFIG_FILE = "/etc/python/email.cfg"
+_cached_strict_addr_parsing = None
+
+
+def _use_strict_email_parsing():
+ """"Cache implementation for _cached_strict_addr_parsing"""
+ global _cached_strict_addr_parsing
+ if _cached_strict_addr_parsing is None:
+ _cached_strict_addr_parsing = _use_strict_email_parsing_impl()
+ return _cached_strict_addr_parsing
+
+
+def _use_strict_email_parsing_impl():
+ """Returns True if strict email parsing is not disabled by
+ config file or env variable.
+ """
+ disabled = bool(os.environ.get("PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING"))
+ if disabled:
+ return False
+
+ try:
+ file = open(_EMAIL_CONFIG_FILE)
+ except FileNotFoundError:
+ pass
+ else:
+ with file:
+ import configparser
+ config = configparser.ConfigParser(
+ interpolation=None,
+ comment_prefixes=('#', ),
+
+ )
+ config.read_file(file)
+ disabled = config.getboolean('email_addr_parsing', "PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING", fallback=None)
+
+ if disabled:
+ return False
+
+ return True
+
def _has_surrogates(s):
"""Return True if s contains surrogate-escaped binary data."""
@@ -149,7 +189,7 @@ def _strip_quoted_realnames(addr):
supports_strict_parsing = True
-def getaddresses(fieldvalues, *, strict=True):
+def getaddresses(fieldvalues, *, strict=None):
"""Return a list of (REALNAME, EMAIL) or ('','') for each fieldvalue.
When parsing fails for a fieldvalue, a 2-tuple of ('', '') is returned in
@@ -158,6 +198,11 @@ def getaddresses(fieldvalues, *, strict=True):
If strict is true, use a strict parser which rejects malformed inputs.
"""
+ # If default is used, it's True unless disabled
+ # by env variable or config file.
+ if strict == None:
+ strict = _use_strict_email_parsing()
+
# If strict is true, if the resulting list of parsed addresses is greater
# than the number of fieldvalues in the input list, a parsing error has
# occurred and consequently a list containing a single empty 2-tuple [('',
@@ -330,7 +375,7 @@ def parsedate_to_datetime(data):
tzinfo=datetime.timezone(datetime.timedelta(seconds=tz)))
-def parseaddr(addr, *, strict=True):
+def parseaddr(addr, *, strict=None):
"""
Parse addr into its constituent realname and email address parts.
@@ -339,6 +384,11 @@ def parseaddr(addr, *, strict=True):
If strict is True, use a strict parser which rejects malformed inputs.
"""
+ # If default is used, it's True unless disabled
+ # by env variable or config file.
+ if strict == None:
+ strict = _use_strict_email_parsing()
+
if not strict:
addrs = _AddressList(addr).addresslist
if not addrs:
diff --git a/Lib/test/test_email/test_email.py b/Lib/test/test_email/test_email.py
index ce36efc1b1..05ea201b68 100644
--- a/Lib/test/test_email/test_email.py
+++ b/Lib/test/test_email/test_email.py
@@ -7,6 +7,9 @@ import time
import base64
import unittest
import textwrap
+import contextlib
+import tempfile
+import os
from io import StringIO, BytesIO
from itertools import chain
@@ -41,7 +44,7 @@ from email import iterators
from email import base64mime
from email import quoprimime
-from test.support import unlink, start_threads
+from test.support import unlink, start_threads, EnvironmentVarGuard, swap_attr
from test.test_email import openfile, TestEmailBase
# These imports are documented to work, but we are testing them using a
@@ -3313,6 +3316,73 @@ Foo
# Test email.utils.supports_strict_parsing attribute
self.assertEqual(email.utils.supports_strict_parsing, True)
+ def test_parsing_errors_strict_set_via_env_var(self):
+ address = 'alice@example.org )Alice('
+ empty = ('', '')
+
+ # Reset cached default value to make the function
+ # reload the config file provided below.
+ utils._cached_strict_addr_parsing = None
+
+ # Strict disabled via env variable, old behavior expected
+ with EnvironmentVarGuard() as environ:
+ environ["PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING"] = "1"
+
+ self.assertEqual(utils.getaddresses([address]),
+ [('', 'alice@example.org'), ('', ''), ('', 'Alice')])
+ self.assertEqual(utils.parseaddr([address]), ('', address))
+
+ # Clear cache again
+ utils._cached_strict_addr_parsing = None
+
+ # Default strict=True, empty result expected
+ self.assertEqual(utils.getaddresses([address]), [empty])
+ self.assertEqual(utils.parseaddr([address]), empty)
+
+ # Clear cache again
+ utils._cached_strict_addr_parsing = None
+
+ # Empty string in env variable = strict parsing enabled (default)
+ with EnvironmentVarGuard() as environ:
+ environ["PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING"] = ""
+
+ # Default strict=True, empty result expected
+ self.assertEqual(utils.getaddresses([address]), [empty])
+ self.assertEqual(utils.parseaddr([address]), empty)
+
+ @contextlib.contextmanager
+ def _email_strict_parsing_conf(self):
+ """Context for the given email strict parsing configured in config file"""
+ with tempfile.TemporaryDirectory() as tmpdirname:
+ filename = os.path.join(tmpdirname, 'conf.cfg')
+ with swap_attr(utils, "_EMAIL_CONFIG_FILE", filename):
+ with open(filename, 'w') as file:
+ file.write('[email_addr_parsing]\n')
+ file.write('PYTHON_EMAIL_DISABLE_STRICT_ADDR_PARSING = true')
+ utils._EMAIL_CONFIG_FILE = filename
+ yield
+
+ def test_parsing_errors_strict_disabled_via_config_file(self):
+ address = 'alice@example.org )Alice('
+ empty = ('', '')
+
+ # Reset cached default value to make the function
+ # reload the config file provided below.
+ utils._cached_strict_addr_parsing = None
+
+ # Strict disabled via config file, old results expected
+ with self._email_strict_parsing_conf():
+ self.assertEqual(utils.getaddresses([address]),
+ [('', 'alice@example.org'), ('', ''), ('', 'Alice')])
+ self.assertEqual(utils.parseaddr([address]), ('', address))
+
+ # Clear cache again
+ utils._cached_strict_addr_parsing = None
+
+ # Default strict=True, empty result expected
+ self.assertEqual(utils.getaddresses([address]), [empty])
+ self.assertEqual(utils.parseaddr([address]), empty)
+
def test_getaddresses_nasty(self):
for addresses, expected in (
(['"Sürname, Firstname" <to@example.com>'],
--
2.43.0

View File

@ -0,0 +1,90 @@
From 87acab66e124912549fbc3151f27ca7fae76386c Mon Sep 17 00:00:00 2001
From: Serhiy Storchaka <storchaka@gmail.com>
Date: Tue, 23 Apr 2024 19:54:00 +0200
Subject: [PATCH] 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 | 53 +++++++++++--------
...-02-08-14-21-28.gh-issue-115133.ycl4ko.rst | 2 +
2 files changed, 33 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 acaa519..c01af47 100644
--- a/Lib/test/test_xml_etree.py
+++ b/Lib/test/test_xml_etree.py
@@ -1044,28 +1044,37 @@ 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 0000000..6f10152
--- /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.
--
2.44.0

View File

@ -0,0 +1,291 @@
From 82f1ea4b72be40f58fd0a9a37f8d8d2f7d16f9e0 Mon Sep 17 00:00:00 2001
From: Lumir Balhar <lbalhar@redhat.com>
Date: Wed, 24 Apr 2024 00:19:23 +0200
Subject: [PATCH] CVE-2023-6597
Co-authored-by: Søren Løvborg <sorenl@unity3d.com>
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
---
Lib/tempfile.py | 44 +++++++++-
Lib/test/test_tempfile.py | 166 +++++++++++++++++++++++++++++++++++---
2 files changed, 199 insertions(+), 11 deletions(-)
diff --git a/Lib/tempfile.py b/Lib/tempfile.py
index 2cb5434..d79b70c 100644
--- a/Lib/tempfile.py
+++ b/Lib/tempfile.py
@@ -276,6 +276,23 @@ def _mkstemp_inner(dir, pre, suf, flags, output_type):
"No usable temporary file name found")
+def _dont_follow_symlinks(func, path, *args):
+ # Pass follow_symlinks=False, unless not supported on this platform.
+ if func in _os.supports_follow_symlinks:
+ func(path, *args, follow_symlinks=False)
+ elif _os.name == 'nt' or not _os.path.islink(path):
+ func(path, *args)
+
+
+def _resetperms(path):
+ try:
+ chflags = _os.chflags
+ except AttributeError:
+ pass
+ else:
+ _dont_follow_symlinks(chflags, path, 0)
+ _dont_follow_symlinks(_os.chmod, path, 0o700)
+
# User visible interfaces.
def gettempprefix():
@@ -794,9 +811,32 @@ class TemporaryDirectory(object):
self, self._cleanup, self.name,
warn_message="Implicitly cleaning up {!r}".format(self))
+ @classmethod
+ def _rmtree(cls, name):
+ def onerror(func, path, exc_info):
+ if issubclass(exc_info[0], PermissionError):
+ try:
+ if path != name:
+ _resetperms(_os.path.dirname(path))
+ _resetperms(path)
+
+ try:
+ _os.unlink(path)
+ # PermissionError is raised on FreeBSD for directories
+ except (IsADirectoryError, PermissionError):
+ cls._rmtree(path)
+ except FileNotFoundError:
+ pass
+ elif issubclass(exc_info[0], FileNotFoundError):
+ pass
+ else:
+ raise
+
+ _shutil.rmtree(name, onerror=onerror)
+
@classmethod
def _cleanup(cls, name, warn_message):
- _shutil.rmtree(name)
+ cls._rmtree(name)
_warnings.warn(warn_message, ResourceWarning)
def __repr__(self):
@@ -810,4 +850,4 @@ class TemporaryDirectory(object):
def cleanup(self):
if self._finalizer.detach():
- _shutil.rmtree(self.name)
+ self._rmtree(self.name)
diff --git a/Lib/test/test_tempfile.py b/Lib/test/test_tempfile.py
index 710756b..c5560e1 100644
--- a/Lib/test/test_tempfile.py
+++ b/Lib/test/test_tempfile.py
@@ -1298,19 +1298,25 @@ class NulledModules:
class TestTemporaryDirectory(BaseTestCase):
"""Test TemporaryDirectory()."""
- def do_create(self, dir=None, pre="", suf="", recurse=1):
+ def do_create(self, dir=None, pre="", suf="", recurse=1, dirs=1, files=1):
if dir is None:
dir = tempfile.gettempdir()
tmp = tempfile.TemporaryDirectory(dir=dir, prefix=pre, suffix=suf)
self.nameCheck(tmp.name, dir, pre, suf)
- # Create a subdirectory and some files
- if recurse:
- d1 = self.do_create(tmp.name, pre, suf, recurse-1)
- d1.name = None
- with open(os.path.join(tmp.name, "test.txt"), "wb") as f:
- f.write(b"Hello world!")
+ self.do_create2(tmp.name, recurse, dirs, files)
return tmp
+ def do_create2(self, path, recurse=1, dirs=1, files=1):
+ # Create subdirectories and some files
+ if recurse:
+ for i in range(dirs):
+ name = os.path.join(path, "dir%d" % i)
+ os.mkdir(name)
+ self.do_create2(name, recurse-1, dirs, files)
+ for i in range(files):
+ with open(os.path.join(path, "test%d.txt" % i), "wb") as f:
+ f.write(b"Hello world!")
+
def test_mkdtemp_failure(self):
# Check no additional exception if mkdtemp fails
# Previously would raise AttributeError instead
@@ -1350,11 +1356,108 @@ class TestTemporaryDirectory(BaseTestCase):
"TemporaryDirectory %s exists after cleanup" % d1.name)
self.assertTrue(os.path.exists(d2.name),
"Directory pointed to by a symlink was deleted")
- self.assertEqual(os.listdir(d2.name), ['test.txt'],
+ self.assertEqual(os.listdir(d2.name), ['test0.txt'],
"Contents of the directory pointed to by a symlink "
"were deleted")
d2.cleanup()
+ @support.skip_unless_symlink
+ def test_cleanup_with_symlink_modes(self):
+ # cleanup() should not follow symlinks when fixing mode bits (#91133)
+ with self.do_create(recurse=0) as d2:
+ file1 = os.path.join(d2, 'file1')
+ open(file1, 'wb').close()
+ dir1 = os.path.join(d2, 'dir1')
+ os.mkdir(dir1)
+ for mode in range(8):
+ mode <<= 6
+ with self.subTest(mode=format(mode, '03o')):
+ def test(target, target_is_directory):
+ d1 = self.do_create(recurse=0)
+ symlink = os.path.join(d1.name, 'symlink')
+ os.symlink(target, symlink,
+ target_is_directory=target_is_directory)
+ try:
+ os.chmod(symlink, mode, follow_symlinks=False)
+ except NotImplementedError:
+ pass
+ try:
+ os.chmod(symlink, mode)
+ except FileNotFoundError:
+ pass
+ os.chmod(d1.name, mode)
+ d1.cleanup()
+ self.assertFalse(os.path.exists(d1.name))
+
+ with self.subTest('nonexisting file'):
+ test('nonexisting', target_is_directory=False)
+ with self.subTest('nonexisting dir'):
+ test('nonexisting', target_is_directory=True)
+
+ with self.subTest('existing file'):
+ os.chmod(file1, mode)
+ old_mode = os.stat(file1).st_mode
+ test(file1, target_is_directory=False)
+ new_mode = os.stat(file1).st_mode
+ self.assertEqual(new_mode, old_mode,
+ '%03o != %03o' % (new_mode, old_mode))
+
+ with self.subTest('existing dir'):
+ os.chmod(dir1, mode)
+ old_mode = os.stat(dir1).st_mode
+ test(dir1, target_is_directory=True)
+ new_mode = os.stat(dir1).st_mode
+ self.assertEqual(new_mode, old_mode,
+ '%03o != %03o' % (new_mode, old_mode))
+
+ @unittest.skipUnless(hasattr(os, 'chflags'), 'requires os.chflags')
+ @support.skip_unless_symlink
+ def test_cleanup_with_symlink_flags(self):
+ # cleanup() should not follow symlinks when fixing flags (#91133)
+ flags = stat.UF_IMMUTABLE | stat.UF_NOUNLINK
+ self.check_flags(flags)
+
+ with self.do_create(recurse=0) as d2:
+ file1 = os.path.join(d2, 'file1')
+ open(file1, 'wb').close()
+ dir1 = os.path.join(d2, 'dir1')
+ os.mkdir(dir1)
+ def test(target, target_is_directory):
+ d1 = self.do_create(recurse=0)
+ symlink = os.path.join(d1.name, 'symlink')
+ os.symlink(target, symlink,
+ target_is_directory=target_is_directory)
+ try:
+ os.chflags(symlink, flags, follow_symlinks=False)
+ except NotImplementedError:
+ pass
+ try:
+ os.chflags(symlink, flags)
+ except FileNotFoundError:
+ pass
+ os.chflags(d1.name, flags)
+ d1.cleanup()
+ self.assertFalse(os.path.exists(d1.name))
+
+ with self.subTest('nonexisting file'):
+ test('nonexisting', target_is_directory=False)
+ with self.subTest('nonexisting dir'):
+ test('nonexisting', target_is_directory=True)
+
+ with self.subTest('existing file'):
+ os.chflags(file1, flags)
+ old_flags = os.stat(file1).st_flags
+ test(file1, target_is_directory=False)
+ new_flags = os.stat(file1).st_flags
+ self.assertEqual(new_flags, old_flags)
+
+ with self.subTest('existing dir'):
+ os.chflags(dir1, flags)
+ old_flags = os.stat(dir1).st_flags
+ test(dir1, target_is_directory=True)
+ new_flags = os.stat(dir1).st_flags
+ self.assertEqual(new_flags, old_flags)
+
@support.cpython_only
def test_del_on_collection(self):
# A TemporaryDirectory is deleted when garbage collected
@@ -1385,7 +1488,7 @@ class TestTemporaryDirectory(BaseTestCase):
tmp2 = os.path.join(tmp.name, 'test_dir')
os.mkdir(tmp2)
- with open(os.path.join(tmp2, "test.txt"), "w") as f:
+ with open(os.path.join(tmp2, "test0.txt"), "w") as f:
f.write("Hello world!")
{mod}.tmp = tmp
@@ -1453,6 +1556,51 @@ class TestTemporaryDirectory(BaseTestCase):
self.assertEqual(name, d.name)
self.assertFalse(os.path.exists(name))
+ def test_modes(self):
+ for mode in range(8):
+ mode <<= 6
+ with self.subTest(mode=format(mode, '03o')):
+ d = self.do_create(recurse=3, dirs=2, files=2)
+ with d:
+ # Change files and directories mode recursively.
+ for root, dirs, files in os.walk(d.name, topdown=False):
+ for name in files:
+ os.chmod(os.path.join(root, name), mode)
+ os.chmod(root, mode)
+ d.cleanup()
+ self.assertFalse(os.path.exists(d.name))
+
+ def check_flags(self, flags):
+ # skip the test if these flags are not supported (ex: FreeBSD 13)
+ filename = support.TESTFN
+ try:
+ open(filename, "w").close()
+ try:
+ os.chflags(filename, flags)
+ except OSError as exc:
+ # "OSError: [Errno 45] Operation not supported"
+ self.skipTest(f"chflags() doesn't support flags "
+ f"{flags:#b}: {exc}")
+ else:
+ os.chflags(filename, 0)
+ finally:
+ support.unlink(filename)
+
+ @unittest.skipUnless(hasattr(os, 'chflags'), 'requires os.lchflags')
+ def test_flags(self):
+ flags = stat.UF_IMMUTABLE | stat.UF_NOUNLINK
+ self.check_flags(flags)
+
+ d = self.do_create(recurse=3, dirs=2, files=2)
+ with d:
+ # Change files and directories flags recursively.
+ for root, dirs, files in os.walk(d.name, topdown=False):
+ for name in files:
+ os.chflags(os.path.join(root, name), flags)
+ os.chflags(root, flags)
+ d.cleanup()
+ self.assertFalse(os.path.exists(d.name))
+
if __name__ == "__main__":
unittest.main()
--
2.44.0

View File

@ -0,0 +1,346 @@
From 066df4fd454d6ff9be66e80b2a65995b10af174f Mon Sep 17 00:00:00 2001
From: John Jolly <john.jolly@gmail.com>
Date: Tue, 30 Jan 2018 01:51:35 -0700
Subject: [PATCH] bpo-22908: Add seek and tell functionality to ZipExtFile
(GH-4966)
This allows for nested zip files, tar files within zip files, zip files within tar files, etc.
Contributed by: John Jolly
---
Doc/library/zipfile.rst | 6 +-
Lib/test/test_zipfile.py | 34 ++++++++
Lib/zipfile.py | 82 +++++++++++++++++++
.../2017-12-21-22-00-11.bpo-22908.cVm89I.rst | 2 +
4 files changed, 121 insertions(+), 3 deletions(-)
create mode 100644 Misc/NEWS.d/next/Library/2017-12-21-22-00-11.bpo-22908.cVm89I.rst
diff --git a/Doc/library/zipfile.rst b/Doc/library/zipfile.rst
index d58efe0b417516..7c9a8c80225491 100644
--- a/Doc/library/zipfile.rst
+++ b/Doc/library/zipfile.rst
@@ -246,9 +246,9 @@ ZipFile Objects
With *mode* ``'r'`` the file-like object
(``ZipExtFile``) is read-only and provides the following methods:
:meth:`~io.BufferedIOBase.read`, :meth:`~io.IOBase.readline`,
- :meth:`~io.IOBase.readlines`, :meth:`__iter__`,
- :meth:`~iterator.__next__`. These objects can operate independently of
- the ZipFile.
+ :meth:`~io.IOBase.readlines`, :meth:`~io.IOBase.seek`,
+ :meth:`~io.IOBase.tell`, :meth:`__iter__`, :meth:`~iterator.__next__`.
+ These objects can operate independently of the ZipFile.
With ``mode='w'``, a writable file handle is returned, which supports the
:meth:`~io.BufferedIOBase.write` method. While a writable file handle is open,
diff --git a/Lib/test/test_zipfile.py b/Lib/test/test_zipfile.py
index 94db858a1517c4..61c3e349a69ef4 100644
--- a/Lib/test/test_zipfile.py
+++ b/Lib/test/test_zipfile.py
@@ -1628,6 +1628,40 @@ def test_open_conflicting_handles(self):
self.assertEqual(zipf.read('baz'), msg3)
self.assertEqual(zipf.namelist(), ['foo', 'bar', 'baz'])
+ def test_seek_tell(self):
+ # Test seek functionality
+ txt = b"Where's Bruce?"
+ bloc = txt.find(b"Bruce")
+ # Check seek on a file
+ with zipfile.ZipFile(TESTFN, "w") as zipf:
+ zipf.writestr("foo.txt", txt)
+ with zipfile.ZipFile(TESTFN, "r") as zipf:
+ with zipf.open("foo.txt", "r") as fp:
+ fp.seek(bloc, os.SEEK_SET)
+ self.assertEqual(fp.tell(), bloc)
+ fp.seek(-bloc, os.SEEK_CUR)
+ self.assertEqual(fp.tell(), 0)
+ fp.seek(bloc, os.SEEK_CUR)
+ self.assertEqual(fp.tell(), bloc)
+ self.assertEqual(fp.read(5), txt[bloc:bloc+5])
+ fp.seek(0, os.SEEK_END)
+ self.assertEqual(fp.tell(), len(txt))
+ # Check seek on memory file
+ data = io.BytesIO()
+ with zipfile.ZipFile(data, mode="w") as zipf:
+ zipf.writestr("foo.txt", txt)
+ with zipfile.ZipFile(data, mode="r") as zipf:
+ with zipf.open("foo.txt", "r") as fp:
+ fp.seek(bloc, os.SEEK_SET)
+ self.assertEqual(fp.tell(), bloc)
+ fp.seek(-bloc, os.SEEK_CUR)
+ self.assertEqual(fp.tell(), 0)
+ fp.seek(bloc, os.SEEK_CUR)
+ self.assertEqual(fp.tell(), bloc)
+ self.assertEqual(fp.read(5), txt[bloc:bloc+5])
+ fp.seek(0, os.SEEK_END)
+ self.assertEqual(fp.tell(), len(txt))
+
def tearDown(self):
unlink(TESTFN)
unlink(TESTFN2)
diff --git a/Lib/zipfile.py b/Lib/zipfile.py
index f9db45f58a2bde..5df7b1bf75b9d9 100644
--- a/Lib/zipfile.py
+++ b/Lib/zipfile.py
@@ -696,6 +696,18 @@ def __init__(self, file, pos, close, lock, writing):
self._close = close
self._lock = lock
self._writing = writing
+ self.seekable = file.seekable
+ self.tell = file.tell
+
+ def seek(self, offset, whence=0):
+ with self._lock:
+ if self.writing():
+ raise ValueError("Can't reposition in the ZIP file while "
+ "there is an open writing handle on it. "
+ "Close the writing handle before trying to read.")
+ self._file.seek(self._pos)
+ self._pos = self._file.tell()
+ return self._pos
def read(self, n=-1):
with self._lock:
@@ -746,6 +758,9 @@ class ZipExtFile(io.BufferedIOBase):
# Read from compressed files in 4k blocks.
MIN_READ_SIZE = 4096
+ # Chunk size to read during seek
+ MAX_SEEK_READ = 1 << 24
+
def __init__(self, fileobj, mode, zipinfo, decrypter=None,
close_fileobj=False):
self._fileobj = fileobj
@@ -778,6 +793,17 @@ def __init__(self, fileobj, mode, zipinfo, decrypter=None,
else:
self._expected_crc = None
+ self._seekable = False
+ try:
+ if fileobj.seekable():
+ self._orig_compress_start = fileobj.tell()
+ self._orig_compress_size = zipinfo.compress_size
+ self._orig_file_size = zipinfo.file_size
+ self._orig_start_crc = self._running_crc
+ self._seekable = True
+ except AttributeError:
+ pass
+
def __repr__(self):
result = ['<%s.%s' % (self.__class__.__module__,
self.__class__.__qualname__)]
@@ -963,6 +989,62 @@ def close(self):
finally:
super().close()
+ def seekable(self):
+ return self._seekable
+
+ def seek(self, offset, whence=0):
+ if not self._seekable:
+ raise io.UnsupportedOperation("underlying stream is not seekable")
+ curr_pos = self.tell()
+ if whence == 0: # Seek from start of file
+ new_pos = offset
+ elif whence == 1: # Seek from current position
+ new_pos = curr_pos + offset
+ elif whence == 2: # Seek from EOF
+ new_pos = self._orig_file_size + offset
+ else:
+ raise ValueError("whence must be os.SEEK_SET (0), "
+ "os.SEEK_CUR (1), or os.SEEK_END (2)")
+
+ if new_pos > self._orig_file_size:
+ new_pos = self._orig_file_size
+
+ if new_pos < 0:
+ new_pos = 0
+
+ read_offset = new_pos - curr_pos
+ buff_offset = read_offset + self._offset
+
+ if buff_offset >= 0 and buff_offset < len(self._readbuffer):
+ # Just move the _offset index if the new position is in the _readbuffer
+ self._offset = buff_offset
+ read_offset = 0
+ elif read_offset < 0:
+ # Position is before the current position. Reset the ZipExtFile
+
+ self._fileobj.seek(self._orig_compress_start)
+ self._running_crc = self._orig_start_crc
+ self._compress_left = self._orig_compress_size
+ self._left = self._orig_file_size
+ self._readbuffer = b''
+ self._offset = 0
+ self._decompressor = zipfile._get_decompressor(self._compress_type)
+ self._eof = False
+ read_offset = new_pos
+
+ while read_offset > 0:
+ read_len = min(self.MAX_SEEK_READ, read_offset)
+ self.read(read_len)
+ read_offset -= read_len
+
+ return self.tell()
+
+ def tell(self):
+ if not self._seekable:
+ raise io.UnsupportedOperation("underlying stream is not seekable")
+ filepos = self._orig_file_size - self._left - len(self._readbuffer) + self._offset
+ return filepos
+
class _ZipWriteFile(io.BufferedIOBase):
def __init__(self, zf, zinfo, zip64):
diff --git a/Misc/NEWS.d/next/Library/2017-12-21-22-00-11.bpo-22908.cVm89I.rst b/Misc/NEWS.d/next/Library/2017-12-21-22-00-11.bpo-22908.cVm89I.rst
new file mode 100644
index 00000000000000..4f3cc0166019f1
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2017-12-21-22-00-11.bpo-22908.cVm89I.rst
@@ -0,0 +1,2 @@
+Added seek and tell to the ZipExtFile class. This only works if the file
+object used to open the zipfile is seekable.
From 55beb125db2942b5362454e05542e9661e964a65 Mon Sep 17 00:00:00 2001
From: Serhiy Storchaka <storchaka@gmail.com>
Date: Tue, 23 Apr 2024 14:29:31 +0200
Subject: [PATCH] gh-109858: Protect zipfile from "quoted-overlap" zipbomb
(GH-110016) (GH-113916)
Raise BadZipFile when try to read an entry that overlaps with other entry or
central directory.
(cherry picked from commit 66363b9a7b9fe7c99eba3a185b74c5fdbf842eba)
---
Lib/test/test_zipfile.py | 60 +++++++++++++++++++
Lib/zipfile.py | 12 ++++
...-09-28-13-15-51.gh-issue-109858.43e2dg.rst | 3 +
3 files changed, 75 insertions(+)
create mode 100644 Misc/NEWS.d/next/Library/2023-09-28-13-15-51.gh-issue-109858.43e2dg.rst
diff --git a/Lib/test/test_zipfile.py b/Lib/test/test_zipfile.py
index 7f82586..0379909 100644
--- a/Lib/test/test_zipfile.py
+++ b/Lib/test/test_zipfile.py
@@ -1644,6 +1644,66 @@ class OtherTests(unittest.TestCase):
fp.seek(0, os.SEEK_END)
self.assertEqual(fp.tell(), len(txt))
+ @requires_zlib
+ def test_full_overlap(self):
+ data = (
+ b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e'
+ b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00a\xed'
+ b'\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\d\x0b`P'
+ b'K\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2'
+ b'\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00aPK'
+ b'\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0lH\x05\xe2\x1e'
+ b'8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00\x00\x00\x00\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00bPK\x05'
+ b'\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00\x00/\x00\x00'
+ b'\x00\x00\x00'
+ )
+ with zipfile.ZipFile(io.BytesIO(data), 'r') as zipf:
+ self.assertEqual(zipf.namelist(), ['a', 'b'])
+ zi = zipf.getinfo('a')
+ self.assertEqual(zi.header_offset, 0)
+ self.assertEqual(zi.compress_size, 16)
+ self.assertEqual(zi.file_size, 1033)
+ zi = zipf.getinfo('b')
+ self.assertEqual(zi.header_offset, 0)
+ self.assertEqual(zi.compress_size, 16)
+ self.assertEqual(zi.file_size, 1033)
+ self.assertEqual(len(zipf.read('a')), 1033)
+ with self.assertRaisesRegex(zipfile.BadZipFile, 'File name.*differ'):
+ zipf.read('b')
+
+ @requires_zlib
+ def test_quoted_overlap(self):
+ data = (
+ b'PK\x03\x04\x14\x00\x00\x00\x08\x00\xa0lH\x05Y\xfc'
+ b'8\x044\x00\x00\x00(\x04\x00\x00\x01\x00\x00\x00a\x00'
+ b'\x1f\x00\xe0\xffPK\x03\x04\x14\x00\x00\x00\x08\x00\xa0l'
+ b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00'
+ b'\x00\x00b\xed\xc0\x81\x08\x00\x00\x00\xc00\xd6\xfbK\\'
+ b'd\x0b`PK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0'
+ b'lH\x05Y\xfc8\x044\x00\x00\x00(\x04\x00\x00\x01'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00'
+ b'\x00aPK\x01\x02\x14\x00\x14\x00\x00\x00\x08\x00\xa0l'
+ b'H\x05\xe2\x1e8\xbb\x10\x00\x00\x00\t\x04\x00\x00\x01\x00'
+ b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00$\x00\x00\x00'
+ b'bPK\x05\x06\x00\x00\x00\x00\x02\x00\x02\x00^\x00\x00'
+ b'\x00S\x00\x00\x00\x00\x00'
+ )
+ with zipfile.ZipFile(io.BytesIO(data), 'r') as zipf:
+ self.assertEqual(zipf.namelist(), ['a', 'b'])
+ zi = zipf.getinfo('a')
+ self.assertEqual(zi.header_offset, 0)
+ self.assertEqual(zi.compress_size, 52)
+ self.assertEqual(zi.file_size, 1064)
+ zi = zipf.getinfo('b')
+ self.assertEqual(zi.header_offset, 36)
+ self.assertEqual(zi.compress_size, 16)
+ self.assertEqual(zi.file_size, 1033)
+ with self.assertRaisesRegex(zipfile.BadZipFile, 'Overlapped entries'):
+ zipf.read('a')
+ self.assertEqual(len(zipf.read('b')), 1033)
+
def tearDown(self):
unlink(TESTFN)
unlink(TESTFN2)
diff --git a/Lib/zipfile.py b/Lib/zipfile.py
index 0ab9fac..e6d7676 100644
--- a/Lib/zipfile.py
+++ b/Lib/zipfile.py
@@ -338,6 +338,7 @@ class ZipInfo (object):
'compress_size',
'file_size',
'_raw_time',
+ '_end_offset',
)
def __init__(self, filename="NoName", date_time=(1980,1,1,0,0,0)):
@@ -376,6 +377,7 @@ class ZipInfo (object):
self.volume = 0 # Volume number of file header
self.internal_attr = 0 # Internal attributes
self.external_attr = 0 # External file attributes
+ self._end_offset = None # Start of the next local header or central directory
# Other attributes are set by class ZipFile:
# header_offset Byte offset to the file header
# CRC CRC-32 of the uncompressed file
@@ -1346,6 +1348,12 @@ class ZipFile:
if self.debug > 2:
print("total", total)
+ end_offset = self.start_dir
+ for zinfo in sorted(self.filelist,
+ key=lambda zinfo: zinfo.header_offset,
+ reverse=True):
+ zinfo._end_offset = end_offset
+ end_offset = zinfo.header_offset
def namelist(self):
"""Return a list of file names in the archive."""
@@ -1500,6 +1508,10 @@ class ZipFile:
'File name in directory %r and header %r differ.'
% (zinfo.orig_filename, fname))
+ if (zinfo._end_offset is not None and
+ zef_file.tell() + zinfo.compress_size > zinfo._end_offset):
+ raise BadZipFile(f"Overlapped entries: {zinfo.orig_filename!r} (possible zip bomb)")
+
# check for encrypted flag & handle password
is_encrypted = zinfo.flag_bits & 0x1
zd = None
diff --git a/Misc/NEWS.d/next/Library/2023-09-28-13-15-51.gh-issue-109858.43e2dg.rst b/Misc/NEWS.d/next/Library/2023-09-28-13-15-51.gh-issue-109858.43e2dg.rst
new file mode 100644
index 0000000..be279ca
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2023-09-28-13-15-51.gh-issue-109858.43e2dg.rst
@@ -0,0 +1,3 @@
+Protect :mod:`zipfile` from "quoted-overlap" zipbomb. It now raises
+BadZipFile when try to read an entry that overlaps with other entry or
+central directory.
--
2.44.0

View File

@ -14,7 +14,7 @@ URL: https://www.python.org/
# WARNING When rebasing to a new Python version,
# remember to update the python3-docs package as well
Version: %{pybasever}.8
Release: 51%{?dist}.2
Release: 62%{?dist}
License: Python
@ -765,6 +765,19 @@ Patch387: 00387-cve-2020-10735-prevent-dos-by-very-large-int.patch
# the behavior to linear.
Patch394: 00394-cve-2022-45061-cpu-denial-of-service-via-inefficient-idna-decoder.patch
# 00397 #
# Add filters for tarfile extraction (CVE-2007-4559, PEP-706)
# The first patches in the file backport the upstream fix:
# - https://github.com/python/cpython/pull/104583
# (see the linked issue for merged backports)
# Next-to-last patch fixes determination of symlink targets, which were treated
# as relative to the root of the archive,
# rather than the directory containing the symlink.
# Not yet upstream as of this writing.
# The last patch is Red Hat configuration, see KB for documentation:
# - https://access.redhat.com/articles/7004769
Patch397: 00397-tarfile-filter.patch
# 00399 #
# CVE-2023-24329
#
@ -789,6 +802,78 @@ Patch399: 00399-cve-2023-24329.patch
# Backported from Python 3.8
Patch404: 00404-cve-2023-40217.patch
# 00408 #
# CVE-2022-48560
#
# Security fix for CVE-2022-48560: python3: use after free in heappushpop()
# of heapq module
# Resolved upstream: https://github.com/python/cpython/issues/83602
Patch408: 00408-CVE-2022-48560.patch
# 00413 #
# CVE-2022-48564
#
# DoS when processing malformed Apple Property List files in binary format
# Resolved upstream: https://github.com/python/cpython/commit/a63234c49b2fbfb6f0aca32525e525ce3d43b2b4
Patch413: 00413-CVE-2022-48564.patch
# 00414 #
#
# Skip test_pair() and test_speech128() of test_zlib on s390x since
# they fail if zlib uses the s390x hardware accelerator.
Patch414: 00414-skip_test_zlib_s390x.patch
# 00415 #
# [CVE-2023-27043] gh-102988: Reject malformed addresses in email.parseaddr() (#111116)
#
# Detect email address parsing errors and return empty tuple to
# indicate the parsing error (old API). Add an optional 'strict'
# parameter to getaddresses() and parseaddr() functions. Patch by
# Thomas Dwyer.
#
# Upstream PR: https://github.com/python/cpython/pull/111116
#
# Second patch implmenets the possibility to restore the old behavior via
# config file or environment variable.
Patch415: 00415-cve-2023-27043-gh-102988-reject-malformed-addresses-in-email-parseaddr-111116.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.
#
# Patch rebased because the CVE fix is backported to older expat in RHEL.
Patch422: 00422-gh-115133-fix-tests-for-xmlpullparser-with-expat-2-6-0.patch
# 426 #
# CVE-2023-6597
#
# Path traversal on tempfile.TemporaryDirectory
#
# Upstream: https://github.com/python/cpython/issues/91133
# Tracking bug: https://bugzilla.redhat.com/show_bug.cgi?id=CVE-2023-6597
#
# To backport the fix cleanly the patch contains also this rebased commit:
# Fix permission errors in TemporaryDirectory cleanup
# https://github.com/python/cpython/commit/e9b51c0ad81da1da11ae65840ac8b50a8521373c
Patch426: 00426-CVE-2023-6597.patch
# 427 #
# CVE-2024-0450
#
# The zipfile module is vulnerable to zip-bombs leading to denial of service.
#
# Upstream: https://github.com/python/cpython/issues/109858
# Tracking bug: https://bugzilla.redhat.com/show_bug.cgi?id=CVE-2024-0450
#
# To backport the fix cleanly also this change is backported:
# Add seek and tell functionality to ZipExtFile
# https://github.com/python/cpython/commit/066df4fd454d6ff9be66e80b2a65995b10af174f
#
# Patch rebased from 3.8.
Patch427: 00427-CVE-2024-0450.patch
# (New patches go here ^^^)
#
# When adding new patches to "python" and "python3" in Fedora, EL, etc.,
@ -1003,6 +1088,7 @@ configuration, browsers, and other dialogs.
%package tkinter
Summary: A GUI toolkit for Python
Requires: platform-python = %{version}-%{release}
Requires: %{name}-libs%{?_isa} = %{version}-%{release}
%description tkinter
The Tkinter (Tk interface) library is a graphical user interface toolkit for
@ -1134,8 +1220,16 @@ git apply %{PATCH351}
%patch386 -p1
%patch387 -p1
%patch394 -p1
%patch397 -p1
%patch399 -p1
%patch404 -p1
%patch408 -p1
%patch413 -p1
%patch414 -p1
%patch415 -p1
%patch422 -p1
%patch426 -p1
%patch427 -p1
# Remove files that should be generated by the build
# (This is after patching, so that we can use patches directly from upstream)
@ -2067,14 +2161,52 @@ fi
# ======================================================
%changelog
* Wed Sep 27 2023 Masahiro Matsuya <mmatsuya@redhat.com> - 3.6.8-51.2
- Security fix for CVE-2023-40217
Resolves: RHEL-2933
* Wed Apr 24 2024 Lumír Balhar <lbalhar@redhat.com> - 3.6.8-62
- Security fix for CVE-2024-0450
Resolves: RHEL-33683
* Wed May 31 2023 Charalampos Stratakis <cstratak@redhat.com> - 3.6.8-51.1
* Wed Apr 24 2024 Lumír Balhar <lbalhar@redhat.com> - 3.6.8-61
- Security fix for CVE-2023-6597
Resolves: RHEL-33671
* Wed Apr 24 2024 Lumír Balhar <lbalhar@redhat.com> - 3.6.8-60
- Fix build with expat with fixed CVE-2023-52425
Related: RHEL-33671
* Thu Jan 04 2024 Lumír Balhar <lbalhar@redhat.com> - 3.6.8-59
- Security fix for CVE-2023-27043
Resolves: RHEL-20610
* Tue Dec 12 2023 Lumír Balhar <lbalhar@redhat.com> - 3.6.8-58
- Security fix for CVE-2022-48564
Resolves: RHEL-16674
- Skip tests failing on s390x
Resolves: RHEL-19252
* Thu Nov 23 2023 Lumír Balhar <lbalhar@redhat.com> - 3.6.8-57
- Security fix for CVE-2022-48560
Resolves: RHEL-16707
* Thu Sep 07 2023 Charalampos Stratakis <cstratak@redhat.com> - 3.6.8-56
- Security fix for CVE-2023-40217
Resolves: RHEL-3041
* Wed Aug 09 2023 Petr Viktorin <pviktori@redhat.com> - 3.6.8-55
- Fix symlink handling in the fix for CVE-2007-4559
Resolves: rhbz#263261
* Fri Jul 07 2023 Charalampos Stratakis <cstratak@redhat.com> - 3.6.8-54
- Bump release for rebuild
Resolves: rhbz#2173917
* Fri Jun 30 2023 Charalampos Stratakis <cstratak@redhat.com> - 3.6.8-53
- Security fix for CVE-2023-24329
Resolves: rhbz#2173917
* Tue Jun 06 2023 Petr Viktorin <pviktori@redhat.com> - 3.6.8-52
- Add filters for tarfile extraction (CVE-2007-4559, PEP-706)
Resolves: rhbz#263261
* Tue Jan 24 2023 Charalampos Stratakis <cstratak@redhat.com> - 3.6.8-51
- Properly strip the LTO bytecode from python.o
Resolves: rhbz#2137707