[3.14] gh-133516: Raise ValueError when constants True, False or None are used as an identifier after NFKC normalization (GH-133523) (#133596)

Author: miss-islington

Lib/test/test_ast/test_ast.py

@@ -821,6 +821,17 @@ def test_constant_as_name(self):
             with self.assertRaisesRegex(ValueError, f"identifier field can't represent '{constant}' constant"):
                 compile(expr, "<test>", "eval")
 
+    def test_constant_as_unicode_name(self):
+        constants = [
+            ("True", b"Tru\xe1\xb5\x89"),
+            ("False", b"Fal\xc5\xbfe"),
+            ("None", b"N\xc2\xbane"),
+        ]
+        for constant in constants:
+            with self.assertRaisesRegex(ValueError,
+                f"identifier field can't represent '{constant[0]}' constant"):
+                ast.parse(constant[1], mode="eval")
+
     def test_precedence_enum(self):
         class _Precedence(enum.IntEnum):
             """Precedence table that originated from python grammar."""

Misc/NEWS.d/next/Core_and_Builtins/2025-05-06-15-01-41.gh-issue-133516.RqWVf2.rst

@@ -0,0 +1,2 @@
+Raise :exc:`ValueError` when constants ``True``, ``False`` or ``None`` are
+used as an identifier after NFKC normalization.

Parser/pegen.c

@@ -549,6 +549,21 @@ _PyPegen_new_identifier(Parser *p, const char *n)
         }
         id = id2;
     }
+    static const char * const forbidden[] = {
+        "None",
+        "True",
+        "False",
+        NULL
+    };
+    for (int i = 0; forbidden[i] != NULL; i++) {
+        if (_PyUnicode_EqualToASCIIString(id, forbidden[i])) {
+            PyErr_Format(PyExc_ValueError,
+                         "identifier field can't represent '%s' constant",
+                         forbidden[i]);
+            Py_DECREF(id);
+            goto error;
+        }
+    }
     PyInterpreterState *interp = _PyInterpreterState_GET();
     _PyUnicode_InternImmortal(interp, &id);
     if (_PyArena_AddPyObject(p->arena, id) < 0)