From 62efd2c49fe0489176975a6e38360d129220a215 Mon Sep 17 00:00:00 2001
From: tonghuaroot <tonghuaroot@gmail.com>
Date: Mon, 13 Jul 2026 10:37:03 +0800
Subject: [PATCH 1/2] gh-153679: Raise ZipImportError for an invalid UTF-8 file
 name in zipimport

zipimport.zipimporter() is documented to raise ZipImportError for an invalid
archive, but _read_directory() leaked a raw UnicodeDecodeError when a central
directory entry set the UTF-8 file name flag (0x800) but stored bytes that are
not valid UTF-8: the UTF-8 branch called name.decode() unguarded, while the
sibling non-UTF-8 branch already handles UnicodeDecodeError. Guard the UTF-8
decode and raise ZipImportError instead.
---
 Lib/test/test_zipimport.py                     | 18 ++++++++++++++++++
 Lib/zipimport.py                               |  7 ++++++-
 ...6-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst |  4 ++++
 3 files changed, 28 insertions(+), 1 deletion(-)
 create mode 100644 Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst

diff --git a/Lib/test/test_zipimport.py b/Lib/test/test_zipimport.py
index 76cd85709a63af..50d0440c4fa563 100644
--- a/Lib/test/test_zipimport.py
+++ b/Lib/test/test_zipimport.py
@@ -1074,6 +1074,24 @@ def testNotZipFile(self):
         fp.close()
         self.assertZipFailure(TESTMOD)
 
+    def testInvalidUTF8FileName(self):
+        # A central directory entry that sets the UTF-8 file name flag (0x800)
+        # but stores bytes that are not valid UTF-8 must raise ZipImportError
+        # rather than leaking a UnicodeDecodeError.
+        UTF8_FLAG = 0x800
+        name = b'\xff\xfe\xff'
+        cdh = b'PK\x01\x02' + struct.pack(
+            '<HHHHHHIIIHHHHHII',
+            20, 20, UTF8_FLAG, 0, 0, 0, 0, 50, 100,
+            len(name), 0, 0, 0, 0, 0, 0)
+        cdh += name
+        eocd = b'PK\x05\x06' + struct.pack(
+            '<HHHHIIH', 0, 0, 1, 1, len(cdh), 0, 0)
+        os_helper.unlink(TESTMOD)
+        with open(TESTMOD, 'wb') as fp:
+            fp.write(cdh + eocd)
+        self.assertZipFailure(TESTMOD)
+
     # XXX: disabled until this works on Big-endian machines
     def _testBogusZipFile(self):
         os_helper.unlink(TESTMOD)
diff --git a/Lib/zipimport.py b/Lib/zipimport.py
index 19279d1c2bea36..c232b39e4cef04 100644
--- a/Lib/zipimport.py
+++ b/Lib/zipimport.py
@@ -427,7 +427,12 @@ def _read_directory(archive):
 
                 if flags & 0x800:
                     # UTF-8 file names extension
-                    name = name.decode()
+                    try:
+                        name = name.decode()
+                    except UnicodeDecodeError:
+                        raise ZipImportError(
+                            f"can't decode file name in Zip file: {archive!r}",
+                            path=archive)
                 else:
                     # Historical ZIP filename encoding
                     try:
diff --git a/Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst b/Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst
new file mode 100644
index 00000000000000..f1da233ab9d79d
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst
@@ -0,0 +1,4 @@
+:class:`zipimport.zipimporter` now raises :exc:`zipimport.ZipImportError`
+instead of :exc:`UnicodeDecodeError` when a central directory entry sets the
+UTF-8 file name flag but stores a file name that is not valid UTF-8.
+Patch by tonghuaroot.

From 4382c89acd001c366d692148d43e97c296610be6 Mon Sep 17 00:00:00 2001
From: tonghuaroot <tonghuaroot@gmail.com>
Date: Tue, 14 Jul 2026 10:30:16 +0800
Subject: [PATCH 2/2] gh-153679: Trim the test comment and NEWS wording

---
 Lib/test/test_zipimport.py                                   | 4 +---
 .../Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst   | 5 ++---
 2 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/Lib/test/test_zipimport.py b/Lib/test/test_zipimport.py
index 50d0440c4fa563..c3c43f39a58e85 100644
--- a/Lib/test/test_zipimport.py
+++ b/Lib/test/test_zipimport.py
@@ -1075,9 +1075,7 @@ def testNotZipFile(self):
         self.assertZipFailure(TESTMOD)
 
     def testInvalidUTF8FileName(self):
-        # A central directory entry that sets the UTF-8 file name flag (0x800)
-        # but stores bytes that are not valid UTF-8 must raise ZipImportError
-        # rather than leaking a UnicodeDecodeError.
+        # A UTF-8-flagged file name that is not valid UTF-8 raises ZipImportError.
         UTF8_FLAG = 0x800
         name = b'\xff\xfe\xff'
         cdh = b'PK\x01\x02' + struct.pack(
diff --git a/Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst b/Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst
index f1da233ab9d79d..d612903f225021 100644
--- a/Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst
+++ b/Misc/NEWS.d/next/Library/2026-07-13-00-10-00.gh-issue-153679.RJ2WGB.rst
@@ -1,4 +1,3 @@
 :class:`zipimport.zipimporter` now raises :exc:`zipimport.ZipImportError`
-instead of :exc:`UnicodeDecodeError` when a central directory entry sets the
-UTF-8 file name flag but stores a file name that is not valid UTF-8.
-Patch by tonghuaroot.
+instead of :exc:`UnicodeDecodeError` for a UTF-8-flagged file name that is not
+valid UTF-8. Patch by tonghuaroot.