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.

Author: tonghuaroot

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)

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:

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.