[3.12] gh-82951: Fix serializing by name in pickle protocols < 4 (GH-122149) (GH-122265)

miss-islington · serhiy-storchaka · web-flow · commit 9d65ea14ce5b · 2024-07-25T13:44:25.000+03:00
Serializing objects with complex __qualname__ (such as unbound methods and nested classes) by name no longer involves serializing parent objects by value in pickle protocols < 4. (cherry picked from commit dc07f65) Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
diff --git a/Lib/pickle.py b/Lib/pickle.py
@@ -1108,11 +1108,35 @@ def save_global(self, obj, name=None):
             self.save(module_name)
             self.save(name)
             write(STACK_GLOBAL)
-        elif parent is not module:
-            self.save_reduce(getattr, (parent, lastname))
-        elif self.proto >= 3:
-            write(GLOBAL + bytes(module_name, "utf-8") + b'\n' +
-                  bytes(name, "utf-8") + b'\n')
+        elif '.' in name:
+            # In protocol < 4, objects with multi-part __qualname__
+            # are represented as
+            # getattr(getattr(..., attrname1), attrname2).
+            dotted_path = name.split('.')
+            name = dotted_path.pop(0)
+            save = self.save
+            for attrname in dotted_path:
+                save(getattr)
+                if self.proto < 2:
+                    write(MARK)
+            self._save_toplevel_by_name(module_name, name)
+            for attrname in dotted_path:
+                save(attrname)
+                if self.proto < 2:
+                    write(TUPLE)
+                else:
+                    write(TUPLE2)
+                write(REDUCE)
+        else:
+            self._save_toplevel_by_name(module_name, name)
+
+        self.memoize(obj)
+
+    def _save_toplevel_by_name(self, module_name, name):
+        if self.proto >= 3:
+            # Non-ASCII identifiers are supported only with protocols >= 3.
+            self.write(GLOBAL + bytes(module_name, "utf-8") + b'\n' +
+                       bytes(name, "utf-8") + b'\n')
         else:
             if self.fix_imports:
                 r_name_mapping = _compat_pickle.REVERSE_NAME_MAPPING
@@ -1122,15 +1146,13 @@ def save_global(self, obj, name=None):
                 elif module_name in r_import_mapping:
                     module_name = r_import_mapping[module_name]
             try:
-                write(GLOBAL + bytes(module_name, "ascii") + b'\n' +
-                      bytes(name, "ascii") + b'\n')
+                self.write(GLOBAL + bytes(module_name, "ascii") + b'\n' +
+                           bytes(name, "ascii") + b'\n')
             except UnicodeEncodeError:
                 raise PicklingError(
                     "can't pickle global identifier '%s.%s' using "
                     "pickle protocol %i" % (module, name, self.proto)) from None
 
-        self.memoize(obj)
-
     def save_type(self, obj):
         if obj is type(None):
             return self.save_reduce(type, (None,), obj=obj)
diff --git a/Lib/test/pickletester.py b/Lib/test/pickletester.py
@@ -2802,6 +2802,18 @@ class Recursive:
                 self.assertIs(unpickled, Recursive)
         del Recursive.mod # break reference loop
 
+    def test_recursive_nested_names2(self):
+        global Recursive
+        class Recursive:
+            pass
+        Recursive.ref = Recursive
+        Recursive.__qualname__ = 'Recursive.ref'
+        for proto in range(pickle.HIGHEST_PROTOCOL + 1):
+            with self.subTest(proto=proto):
+                unpickled = self.loads(self.dumps(Recursive, proto))
+                self.assertIs(unpickled, Recursive)
+        del Recursive.ref # break reference loop
+
     def test_py_methods(self):
         global PyMethodsTest
         class PyMethodsTest:
diff --git a/Misc/NEWS.d/next/Library/2024-07-23-09-14-44.gh-issue-82951.-F5p5A.rst b/Misc/NEWS.d/next/Library/2024-07-23-09-14-44.gh-issue-82951.-F5p5A.rst
@@ -0,0 +1,3 @@
+Serializing objects with complex ``__qualname__`` (such as unbound methods
+and nested classes) by name no longer involves serializing parent objects by
+value in pickle protocols < 4.
diff --git a/Modules/_pickle.c b/Modules/_pickle.c
@@ -3651,7 +3651,6 @@ save_global(PickleState *st, PicklerObject *self, PyObject *obj,
     PyObject *module = NULL;
     PyObject *parent = NULL;
     PyObject *dotted_path = NULL;
-    PyObject *lastname = NULL;
     PyObject *cls;
     int status = 0;
 
@@ -3692,10 +3691,7 @@ save_global(PickleState *st, PicklerObject *self, PyObject *obj,
                      obj, module_name);
         goto error;
     }
-    lastname = Py_NewRef(PyList_GET_ITEM(dotted_path,
-                         PyList_GET_SIZE(dotted_path) - 1));
     cls = get_deep_attribute(module, dotted_path, &parent);
-    Py_CLEAR(dotted_path);
     if (cls == NULL) {
         PyErr_Format(st->PicklingError,
                      "Can't pickle %R: attribute lookup %S on %S failed",
@@ -3783,7 +3779,10 @@ save_global(PickleState *st, PicklerObject *self, PyObject *obj,
     else {
   gen_global:
         if (parent == module) {
-            Py_SETREF(global_name, Py_NewRef(lastname));
+            Py_SETREF(global_name,
+                Py_NewRef(PyList_GET_ITEM(dotted_path,
+                                          PyList_GET_SIZE(dotted_path) - 1)));
+            Py_CLEAR(dotted_path);
         }
         if (self->proto >= 4) {
             const char stack_global_op = STACK_GLOBAL;
@@ -3796,20 +3795,30 @@ save_global(PickleState *st, PicklerObject *self, PyObject *obj,
             if (_Pickler_Write(self, &stack_global_op, 1) < 0)
                 goto error;
         }
-        else if (parent != module) {
-            PyObject *reduce_value = Py_BuildValue("(O(OO))",
-                                        st->getattr, parent, lastname);
-            if (reduce_value == NULL)
-                goto error;
-            status = save_reduce(st, self, reduce_value, NULL);
-            Py_DECREF(reduce_value);
-            if (status < 0)
-                goto error;
-        }
         else {
             /* Generate a normal global opcode if we are using a pickle
                protocol < 4, or if the object is not registered in the
-               extension registry. */
+               extension registry.
+
+               Objects with multi-part __qualname__ are represented as
+               getattr(getattr(..., attrname1), attrname2). */
+            const char mark_op = MARK;
+            const char tupletwo_op = (self->proto < 2) ? TUPLE : TUPLE2;
+            const char reduce_op = REDUCE;
+            Py_ssize_t i;
+            if (dotted_path) {
+                if (PyList_GET_SIZE(dotted_path) > 1) {
+                    Py_SETREF(global_name, Py_NewRef(PyList_GET_ITEM(dotted_path, 0)));
+                }
+                for (i = 1; i < PyList_GET_SIZE(dotted_path); i++) {
+                    if (save(st, self, st->getattr, 0) < 0 ||
+                        (self->proto < 2 && _Pickler_Write(self, &mark_op, 1) < 0))
+                    {
+                        goto error;
+                    }
+                }
+            }
+
             PyObject *encoded;
             PyObject *(*unicode_encoder)(PyObject *);
 
@@ -3871,6 +3880,17 @@ save_global(PickleState *st, PicklerObject *self, PyObject *obj,
             Py_DECREF(encoded);
             if (_Pickler_Write(self, "\n", 1) < 0)
                 goto error;
+
+            if (dotted_path) {
+                for (i = 1; i < PyList_GET_SIZE(dotted_path); i++) {
+                    if (save(st, self, PyList_GET_ITEM(dotted_path, i), 0) < 0 ||
+                        _Pickler_Write(self, &tupletwo_op, 1) < 0 ||
+                        _Pickler_Write(self, &reduce_op, 1) < 0)
+                    {
+                        goto error;
+                    }
+                }
+            }
         }
         /* Memoize the object. */
         if (memo_put(st, self, obj) < 0)
@@ -3886,7 +3906,6 @@ save_global(PickleState *st, PicklerObject *self, PyObject *obj,
     Py_XDECREF(module);
     Py_XDECREF(parent);
     Py_XDECREF(dotted_path);
-    Py_XDECREF(lastname);
 
     return status;
 }

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,3 @@`
	`1`	+Serializing objects with complex ``__qualname__`` (such as unbound methods
	`2`	`+and nested classes) by name no longer involves serializing parent objects by`
	`3`	`+value in pickle protocols < 4.`