Backport r61458:

Issue 2321: reduce memory usage (increase the memory that is returned
to the system) by using pymalloc for the data of unicode objects.

2 files changed, 17 insertions, 11 deletions

diff --git a/Misc/NEWS b/Misc/NEWS
index 43bc504b82..efbfc5f2f3 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -11,6 +11,8 @@ What's New in Python 2.5.3?
 
 Core and builtins
 -----------------
+- Issue #2321: use pymalloc for unicode object string data to reduce
+  memory usage in some circumstances.
 
 Library
 -------
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index f2582398a8..c749ac51a7 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -200,7 +200,8 @@ int unicode_resize(register PyUnicodeObject *unicode,
        it contains). */
 
     oldstr = unicode->str;
-    PyMem_RESIZE(unicode->str, Py_UNICODE, length + 1);
+    unicode->str = PyObject_REALLOC(unicode->str,
+				    sizeof(Py_UNICODE) * (length + 1));
     if (!unicode->str) {
 	unicode->str = (Py_UNICODE *)oldstr;
         PyErr_NoMemory();
@@ -249,20 +250,23 @@ PyUnicodeObject *_PyUnicode_New(Py_ssize_t length)
 	       never downsize it. */
 	    if ((unicode->length < length) &&
                 unicode_resize(unicode, length) < 0) {
-		PyMem_DEL(unicode->str);
+		PyObject_DEL(unicode->str);
 		goto onError;
 	    }
 	}
         else {
-	    unicode->str = PyMem_NEW(Py_UNICODE, length + 1);
+	    size_t new_size = sizeof(Py_UNICODE) * ((size_t)length + 1);
+	    unicode->str = (Py_UNICODE*) PyObject_MALLOC(new_size);
         }
         PyObject_INIT(unicode, &PyUnicode_Type);
     }
     else {
+	size_t new_size;
         unicode = PyObject_New(PyUnicodeObject, &PyUnicode_Type);
         if (unicode == NULL)
             return NULL;
-	unicode->str = PyMem_NEW(Py_UNICODE, length + 1);
+	new_size = sizeof(Py_UNICODE) * ((size_t)length + 1);
+	unicode->str = (Py_UNICODE*) PyObject_MALLOC(new_size);
     }
 
     if (!unicode->str) {
@@ -296,7 +300,7 @@ void unicode_dealloc(register PyUnicodeObject *unicode)
 	unicode_freelist_size < MAX_UNICODE_FREELIST_SIZE) {
         /* Keep-Alive optimization */
 	if (unicode->length >= KEEPALIVE_SIZE_LIMIT) {
-	    PyMem_DEL(unicode->str);
+	    PyObject_DEL(unicode->str);
 	    unicode->str = NULL;
 	    unicode->length = 0;
 	}
@@ -310,7 +314,7 @@ void unicode_dealloc(register PyUnicodeObject *unicode)
         unicode_freelist_size++;
     }
     else {
-	PyMem_DEL(unicode->str);
+	PyObject_DEL(unicode->str);
 	Py_XDECREF(unicode->defenc);
 	unicode->ob_type->tp_free((PyObject *)unicode);
     }
@@ -7129,8 +7133,8 @@ unicode_subscript(PyUnicodeObject* self, PyObject* item)
             return PyUnicode_FromUnicode(NULL, 0);
         } else {
             source_buf = PyUnicode_AS_UNICODE((PyObject*)self);
-            result_buf = (Py_UNICODE *)PyMem_MALLOC(slicelength*
-                                                    sizeof(Py_UNICODE));
+            result_buf = (Py_UNICODE *)PyObject_MALLOC(slicelength*
+                                                       sizeof(Py_UNICODE));
 	    
 	    if (result_buf == NULL)
 		    return PyErr_NoMemory();
@@ -7140,7 +7144,7 @@ unicode_subscript(PyUnicodeObject* self, PyObject* item)
             }
 
             result = PyUnicode_FromUnicode(result_buf, slicelength);
-            PyMem_FREE(result_buf);
+            PyObject_FREE(result_buf);
             return result;
         }
     } else {
@@ -7948,7 +7952,7 @@ unicode_subtype_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
 		Py_DECREF(tmp);
 		return NULL;
 	}
-	pnew->str = PyMem_NEW(Py_UNICODE, n+1);
+	pnew->str = (Py_UNICODE*) PyObject_MALLOC(sizeof(Py_UNICODE) * (n+1));
 	if (pnew->str == NULL) {
 		_Py_ForgetReference((PyObject *)pnew);
 		PyObject_Del(pnew);
@@ -8075,7 +8079,7 @@ _PyUnicode_Fini(void)
 	PyUnicodeObject *v = u;
 	u = *(PyUnicodeObject **)u;
 	if (v->str)
-	    PyMem_DEL(v->str);
+	    PyObject_DEL(v->str);
 	Py_XDECREF(v->defenc);
 	PyObject_Del(v);
     }