[3.11] [3.12] pythongh-117084: Fix ZIP file extraction for directory …

…entry names with backslashes on Windows (pythonGH-117129) (pythonGH-117162) (cherry picked from commit f3fee23) Co-authored-by: Serhiy Storchaka <[email protected]> (cherry picked from commit 567ab3b)
serhiy-storchaka · Mar 22, 2024 · c23d242 · c23d242
1 parent da5efeb
commit c23d242
Show file tree

Hide file tree

Showing 4 changed files with 27 additions and 1 deletion.
diff --git a/Lib/test/test_zipfile.py b/Lib/test/test_zipfile.py
@@ -2882,6 +2882,22 @@ def test_bug_6050(self):
         os.mkdir(os.path.join(TESTFN2, "a"))
         self.test_extract_dir()
 
+    def test_extract_dir_backslash(self):
+        zfname = findfile("zipdir_backslash.zip")
+        with zipfile.ZipFile(zfname) as zipf:
+            zipf.extractall(TESTFN2)
+        if os.name == 'nt':
+            self.assertTrue(os.path.isdir(os.path.join(TESTFN2, "a")))
+            self.assertTrue(os.path.isdir(os.path.join(TESTFN2, "a", "b")))
+            self.assertTrue(os.path.isfile(os.path.join(TESTFN2, "a", "b", "c")))
+            self.assertTrue(os.path.isdir(os.path.join(TESTFN2, "d")))
+            self.assertTrue(os.path.isdir(os.path.join(TESTFN2, "d", "e")))
+        else:
+            self.assertTrue(os.path.isfile(os.path.join(TESTFN2, "a\\b\\c")))
+            self.assertTrue(os.path.isfile(os.path.join(TESTFN2, "d\\e\\")))
+            self.assertFalse(os.path.exists(os.path.join(TESTFN2, "a")))
+            self.assertFalse(os.path.exists(os.path.join(TESTFN2, "d")))
+
     def test_write_dir(self):
         dirpath = os.path.join(TESTFN2, "x")
         os.mkdir(dirpath)

diff --git a/Lib/test/zipdir_backslash.zip b/Lib/test/zipdir_backslash.zip
diff --git a/Lib/zipfile.py b/Lib/zipfile.py
@@ -559,7 +559,15 @@ def from_file(cls, filename, arcname=None, *, strict_timestamps=True):
 
     def is_dir(self):
         """Return True if this archive member is a directory."""
-        return self.filename[-1] == '/'
+        if self.filename.endswith('/'):
+            return True
+        # The ZIP format specification requires to use forward slashes
+        # as the directory separator, but in practice some ZIP files
+        # created on Windows can use backward slashes.  For compatibility
+        # with the extraction code which already handles this:
+        if os.path.altsep:
+            return self.filename.endswith((os.path.sep, os.path.altsep))
+        return False
 
 
 # ZIP encryption uses the CRC32 one-byte primitive for scrambling some

diff --git a/Misc/NEWS.d/next/Library/2024-03-21-17-07-38.gh-issue-117084.w1mTpT.rst b/Misc/NEWS.d/next/Library/2024-03-21-17-07-38.gh-issue-117084.w1mTpT.rst
@@ -0,0 +1,2 @@
+Fix :mod:`zipfile` extraction for directory entries with the name containing
+backslashes on Windows.