bpo-35474: Fix mimetypes.guess_all_extensions() potentially mutating list (GH-28286) (GH-28289)

miss-islington · serhiy-storchaka · web-flow · commit 06c26f4d2909 · 2021-09-13T19:24:29.000+02:00
* Calling guess_all_extensions() with strict=False potentially mutated types_map_inv. * Mutating the result of guess_all_extensions() mutated types_map_inv. (cherry picked from commit 97ea18e) Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
diff --git a/Lib/mimetypes.py b/Lib/mimetypes.py
@@ -175,7 +175,7 @@ def guess_all_extensions(self, type, strict=True):
         but non-standard types.
         """
         type = type.lower()
-        extensions = self.types_map_inv[True].get(type, [])
+        extensions = list(self.types_map_inv[True].get(type, []))
         if not strict:
             for ext in self.types_map_inv[False].get(type, []):
                 if ext not in extensions:
diff --git a/Lib/test/test_mimetypes.py b/Lib/test/test_mimetypes.py
@@ -113,20 +113,29 @@ def test_filename_with_url_delimiters(self):
         eq(self.db.guess_type(r" \"\`;b&b&c |.tar.gz"), gzip_expected)
 
     def test_guess_all_types(self):
-        eq = self.assertEqual
-        unless = self.assertTrue
         # First try strict.  Use a set here for testing the results because if
         # test_urllib2 is run before test_mimetypes, global state is modified
         # such that the 'all' set will have more items in it.
-        all = set(self.db.guess_all_extensions('text/plain', strict=True))
-        unless(all >= set(['.bat', '.c', '.h', '.ksh', '.pl', '.txt']))
+        all = self.db.guess_all_extensions('text/plain', strict=True)
+        self.assertTrue(set(all) >= {'.bat', '.c', '.h', '.ksh', '.pl', '.txt'})
+        self.assertEqual(len(set(all)), len(all))  # no duplicates
         # And now non-strict
         all = self.db.guess_all_extensions('image/jpg', strict=False)
-        all.sort()
-        eq(all, ['.jpg'])
+        self.assertEqual(all, ['.jpg'])
         # And now for no hits
         all = self.db.guess_all_extensions('image/jpg', strict=True)
-        eq(all, [])
+        self.assertEqual(all, [])
+        # And now for type existing in both strict and non-strict mappings.
+        self.db.add_type('test-type', '.strict-ext')
+        self.db.add_type('test-type', '.non-strict-ext', strict=False)
+        all = self.db.guess_all_extensions('test-type', strict=False)
+        self.assertEqual(all, ['.strict-ext', '.non-strict-ext'])
+        all = self.db.guess_all_extensions('test-type')
+        self.assertEqual(all, ['.strict-ext'])
+        # Test that changing the result list does not affect the global state
+        all.append('.no-such-ext')
+        all = self.db.guess_all_extensions('test-type')
+        self.assertNotIn('.no-such-ext', all)
 
     def test_encoding(self):
         getpreferredencoding = locale.getpreferredencoding
diff --git a/Misc/NEWS.d/next/Library/2021-09-11-10-45-12.bpo-35474.tEY3SD.rst b/Misc/NEWS.d/next/Library/2021-09-11-10-45-12.bpo-35474.tEY3SD.rst
@@ -0,0 +1,3 @@
+Calling :func:`mimetypes.guess_all_extensions` with ``strict=False`` no
+longer affects the result of the following call with ``strict=True``.
+Also, mutating the returned list no longer affects the global state.

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,3 @@`
	`1`	+Calling :func:`mimetypes.guess_all_extensions` with ``strict=False`` no
	`2`	+longer affects the result of the following call with ``strict=True``.
	`3`	`+Also, mutating the returned list no longer affects the global state.`