gh-93627: Align Python implementation of pickle with C implementation of pickle

Lib/pickle.py

@@ -396,6 +396,8 @@ def decode_long(data):
     return int.from_bytes(data, byteorder='little', signed=True)
 
 
+_NoValue = object()
+
 # Pickling machinery
 
 class _Pickler:
@@ -542,8 +544,8 @@ def save(self, obj, save_persistent_id=True):
             return
 
         rv = NotImplemented
-        reduce = getattr(self, "reducer_override", None)
-        if reduce is not None:
+        reduce = getattr(self, "reducer_override", _NoValue)
+        if reduce is not _NoValue:
             rv = reduce(obj)
 
         if rv is NotImplemented:
@@ -556,8 +558,8 @@ def save(self, obj, save_persistent_id=True):
 
             # Check private dispatch table if any, or else
             # copyreg.dispatch_table
-            reduce = getattr(self, 'dispatch_table', dispatch_table).get(t)
-            if reduce is not None:
+            reduce = getattr(self, 'dispatch_table', dispatch_table).get(t, _NoValue)
+            if reduce is not _NoValue:
                 rv = reduce(obj)
             else:
                 # Check for a class with a custom metaclass; treat as regular
@@ -567,12 +569,12 @@ def save(self, obj, save_persistent_id=True):
                     return
 
                 # Check for a __reduce_ex__ method, fall back to __reduce__
-                reduce = getattr(obj, "__reduce_ex__", None)
-                if reduce is not None:
+                reduce = getattr(obj, "__reduce_ex__", _NoValue)
+                if reduce is not _NoValue:
                     rv = reduce(self.proto)
                 else:
-                    reduce = getattr(obj, "__reduce__", None)
-                    if reduce is not None:
+                    reduce = getattr(obj, "__reduce__", _NoValue)
+                    if reduce is not _NoValue:
                         rv = reduce()
                     else:
                         raise PicklingError("Can't pickle %r object: %r" %
@@ -1610,7 +1612,7 @@ def load_binget(self):
         i = self.read(1)[0]
         try:
             self.append(self.memo[i])
-        except KeyError as exc:
+        except KeyError:
             msg = f'Memo value not found at index {i}'
             raise UnpicklingError(msg) from None
     dispatch[BINGET[0]] = load_binget
@@ -1619,7 +1621,7 @@ def load_long_binget(self):
         i, = unpack('<I', self.read(4))
         try:
             self.append(self.memo[i])
-        except KeyError as exc:
+        except KeyError:
             msg = f'Memo value not found at index {i}'
             raise UnpicklingError(msg) from None
     dispatch[LONG_BINGET[0]] = load_long_binget
@@ -1705,8 +1707,8 @@ def load_build(self):
         stack = self.stack
         state = stack.pop()
         inst = stack[-1]
-        setstate = getattr(inst, "__setstate__", None)
-        if setstate is not None:
+        setstate = getattr(inst, "__setstate__", _NoValue)
+        if setstate is not _NoValue:
             setstate(state)
             return
         slotstate = None

Lib/test/test_pickle.py

@@ -570,6 +570,22 @@ def test_multiprocessing_exceptions(self):
                                  ('multiprocessing.context', name))
 
 
+class PickleReductionMethodsTests(unittest.TestCase):
+
+    def test_pickle_invalid_reduction_method(self):
+        class C:
+            __reduce_ex__ = None
+        c = C()
+        with self.assertRaises(TypeError):
+            pickle.dumps(c)
+
+        class C:
+            __reduce__ = None
+        c = C()
+        with self.assertRaises(TypeError):
+            pickle.dumps(c)
+
+
 def load_tests(loader, tests, pattern):
     tests.addTest(doctest.DocTestSuite())
     return tests

Misc/NEWS.d/next/Core and Builtins/2023-03-26-19-11-10.gh-issue-93627.0UgwBL.rst

@@ -0,0 +1 @@
+Update the Python pickle module implementation to match the C implementation of the pickle module and update the implementation of the copy module to match the pickle module conventions. For all modules setting reduction methods like ``__reduce_ex__`` or ``__reduce__`` to ``None`` will result in a ``TypeError`` with pickling or copying an object.

Modules/_pickle.c

@@ -3987,7 +3987,7 @@ save_reduce(PickleState *st, PicklerObject *self, PyObject *args,
     size = PyTuple_Size(args);
     if (size < 2 || size > 6) {
         PyErr_SetString(st->PicklingError, "tuple returned by "
-                        "__reduce__ must contain 2 through 6 elements");
+                        "__reduce__ must contain 2 to 6 elements");
         return -1;
     }
 
@@ -4417,6 +4417,7 @@ save(PickleState *st, PicklerObject *self, PyObject *obj, int pers_save)
             Py_INCREF(reduce_func);
         }
     } else {
+        // check private dispatch table
         reduce_func = PyObject_GetItem(self->dispatch_table,
                                        (PyObject *)type);
         if (reduce_func == NULL) {
@@ -4447,8 +4448,7 @@ save(PickleState *st, PicklerObject *self, PyObject *obj, int pers_save)
             goto error;
         }
         if (reduce_func != NULL) {
-            PyObject *proto;
-            proto = PyLong_FromLong(self->proto);
+            PyObject *proto = PyLong_FromLong(self->proto);
             if (proto != NULL) {
                 reduce_value = _Pickle_FastCall(reduce_func, proto);
             }