bpo-39337: encodings.normalize_encoding() now ignores non-ASCII characters

Lib/encodings/__init__.py

@@ -61,7 +61,8 @@ def normalize_encoding(encoding):
         if c.isalnum() or c == '.':
             if punct and chars:
                 chars.append('_')
-            chars.append(c)
+            if c.isascii():
+                chars.append(c)
             punct = False
         else:
             punct = True

Lib/test/test_source_encoding.py

@@ -14,11 +14,11 @@ class MiscSourceEncodingTest(unittest.TestCase):
 
     def test_pep263(self):
         self.assertEqual(
-            "�����".encode("utf-8"),
+            "ðÉÔÏÎ".encode("utf-8"),
             b'\xd0\x9f\xd0\xb8\xd1\x82\xd0\xbe\xd0\xbd'
         )
         self.assertEqual(
-            "\�".encode("utf-8"),
+            "\ð".encode("utf-8"),
             b'\\\xd0\x9f'
         )
 
@@ -226,5 +226,18 @@ def check_script_output(self, src, expected):
         self.assertEqual(res.out.rstrip(), expected)
 
 
+class EncodingsTest(unittest.TestCase):
+
+    def test_bpo39337(self):
+        """
+        bpo-39337: similar to _Py_normalize_encoding(),
+        encodings.normalize_encoding() should ignore non-ASCII letters .
+        """
+        import encodings
+
+        out = encodings.normalize_encoding("кои-8")
+        self.assertEqual(out, '8')
+
+
 if __name__ == "__main__":
     unittest.main()

Misc/NEWS.d/next/Library/2020-09-13-02-02-18.bpo-39337.L3NXTt.rst

@@ -0,0 +1,2 @@
+similar to :c:func:`_Py_normalize_encoding`,
+:func:`encodings.normalize_encoding` should ignore non-ASCII letters.