PEP 489: Multi-phase extension module initialization

Known limitations of the current implementation:

- documentation changes are incomplete
- there's a reference leak I haven't tracked down yet

The leak is most visible by running:

  ./python -m test -R3:3 test_importlib

However, you can also see it by running:

  ./python -X showrefcount

Importing the array or _testmultiphase modules, and
then deleting them from both sys.modules and the local
namespace shows significant increases in the total
number of active references each cycle. By contrast,
with _testcapi (which continues to use single-phase
initialisation) the global refcounts stabilise after
a couple of cycles.

Modules/arraymodule.c

@@ -2981,34 +2981,17 @@ static PyMethodDef a_methods[] = {
     {NULL, NULL, 0, NULL}        /* Sentinel */
 };
 
-static struct PyModuleDef arraymodule = {
-    PyModuleDef_HEAD_INIT,
-    "array",
-    module_doc,
-    -1,
-    a_methods,
-    NULL,
-    NULL,
-    NULL,
-    NULL
-};
-
-
-PyMODINIT_FUNC
-PyInit_array(void)
+static int
+array_modexec(PyObject *m)
 {
-    PyObject *m;
     char buffer[Py_ARRAY_LENGTH(descriptors)], *p;
     PyObject *typecodes;
     Py_ssize_t size = 0;
     struct arraydescr *descr;
 
     if (PyType_Ready(&Arraytype) < 0)
-        return NULL;
+        return -1;
     Py_TYPE(&PyArrayIter_Type) = &PyType_Type;
-    m = PyModule_Create(&arraymodule);
-    if (m == NULL)
-        return NULL;
 
     Py_INCREF((PyObject *)&Arraytype);
     PyModule_AddObject(m, "ArrayType", (PyObject *)&Arraytype);
@@ -3031,5 +3014,30 @@ PyInit_array(void)
         Py_DECREF(m);
         m = NULL;
     }
-    return m;
+    return 0;
+}
+
+static PyModuleDef_Slot arrayslots[] = {
+    {Py_mod_exec, array_modexec},
+    {0, NULL}
+};
+
+
+static struct PyModuleDef arraymodule = {
+    PyModuleDef_HEAD_INIT,
+    "array",
+    module_doc,
+    0,
+    a_methods,
+    arrayslots,
+    NULL,
+    NULL,
+    NULL
+};
+
+
+PyMODINIT_FUNC
+PyInit_array(void)
+{
+    return PyModuleDef_Init(&arraymodule);
 }