python/CVE-2024-0450-zipfile-avoid-quoted-overlap-zipbomb.patch

From d8877aaabe9aa5d9b9904c222c552f3c6a85017c Mon Sep 17 00:00:00 2001
From: Serhiy Storchaka <storchaka@gmail.com>
Date: Wed, 17 Jan 2024 15:41:50 +0200
Subject: [PATCH] [CVE-2024-0450] Protect zipfile from "quoted-overlap" zipbomb

Raise BadZipFile when try to read an entry that overlaps with
other entry or central directory.
(cherry picked from commit 66363b9a7b9fe7c99eba3a185b74c5fdbf842eba)

From-PR: gh#python/cpython!110016
Fixes: gh#python/cpython#109858
Patch: CVE-2024-0450-zipfile-avoid-quoted-overlap-zipbomb.patch
---
 Lib/test/test_zipfile.py                                                |   60 ++++++++++
 Lib/zipfile.py                                                          |   12 ++
 Misc/NEWS.d/next/Library/2023-09-28-13-15-51.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

--- a/Lib/test/test_zipfile.py
+++ b/Lib/test/test_zipfile.py
@@ -1377,6 +1377,66 @@ class TestsWithRandomBinaryFiles(unittes
         with open(TESTFN, "wb") as fp:
             fp.write(self.data)
 
+    @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)
--- a/Lib/zipfile.py
+++ b/Lib/zipfile.py
@@ -305,6 +305,7 @@ class ZipInfo (object):
             'compress_size',
             'file_size',
             '_raw_time',
+            '_end_offset',
         )
 
     def __init__(self, filename="NoName", date_time=(1980,1,1,0,0,0)):
@@ -343,6 +344,7 @@ 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
@@ -891,6 +893,12 @@ class ZipFile(object):
             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."""
@@ -1002,6 +1010,10 @@ class ZipFile(object):
                         'File name in directory "%s" and header "%s" differ.' % (
                             zinfo.orig_filename, fname)
 
+            if (zinfo._end_offset is not None and
+                zef_file.tell() + zinfo.compress_size > zinfo._end_offset):
+                raise BadZipFile("Overlapped entries: {!r} (possible zip bomb)".format(zinfo.orig_filename))
+
             # check for encrypted flag & handle password
             is_encrypted = zinfo.flag_bits & 0x1
             zd = None
--- /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.