#10801: In zipfile, support different encodings for the header and the...

#10801: In zipfile, support different encodings for the header and the filenames.  Patch by MvL, test by Eli Bendersky.

Lib/test/test_zipfile.py

@@ -6,6 +6,7 @@ except ImportError:
 
 import io
 import os
+import sys
 import imp
 import time
 import shutil
@@ -23,6 +24,7 @@ from test.support import TESTFN, run_unittest, findfile, unlink
 TESTFN2 = TESTFN + "2"
 TESTFNDIR = TESTFN + "d"
 FIXEDTEST_SIZE = 1000
+DATAFILES_DIR = 'zipfile_datafiles'
 
 SMALL_TEST_DATA = [('_ziptest1', '1q2w3e4r5t'),
                    ('ziptest2dir/_ziptest2', 'qawsedrftg'),
@@ -487,6 +489,18 @@ class TestsWithSourceFile(unittest.TestCase):
         except zipfile.BadZipFile:
             self.assertTrue(zipfp2.fp is None, 'zipfp is not closed')
 
+    def test_unicode_filenames(self):
+        if __name__ == '__main__':
+            myfile = sys.argv[0]
+        else:
+            myfile = __file__
+
+        mydir = os.path.dirname(myfile) or os.curdir
+        fname = os.path.join(mydir, 'zip_cp437_header.zip')
+
+        with zipfile.ZipFile(fname) as zipfp:
+            zipfp.extractall()
+
     def tearDown(self):
         unlink(TESTFN)
         unlink(TESTFN2)

Lib/zipfile.py

@@ -930,7 +930,13 @@ class ZipFile:
         if fheader[_FH_EXTRA_FIELD_LENGTH]:
             zef_file.read(fheader[_FH_EXTRA_FIELD_LENGTH])
 
-        if fname != zinfo.orig_filename.encode("utf-8"):
+        if zinfo.flag_bits & 0x800:
+            # UTF-8 filename
+            fname_str = fname.decode("utf-8")
+        else:
+            fname_str = fname.decode("cp437")
+
+        if fname_str != zinfo.orig_filename:
             if not self._filePassed:
                 zef_file.close()
             raise BadZipFile(

Misc/NEWS

@@ -20,6 +20,9 @@ Core and Builtins
 Library
 -------
 
+- Issue #10801: In zipfile, support different encodings for the header and
+  the filenames.
+
 - Issue #6285: IDLE no longer crashes on missing help file; patch by Scott
   David Daniels.