This patch adds a new builtin unistr() which behaves like str()

except that it always returns Unicode objects.

A new C API PyObject_Unicode() is also provided.

This closes patch #101664.

Written by Marc-Andre Lemburg. Copyright assigned to Guido van Rossum.

2 files changed, 48 insertions, 0 deletions

diff --git a/Objects/object.c b/Objects/object.c
index 3cad241..20950c1 100644
--- a/Objects/object.c
+++ b/Objects/object.c
@@ -568,6 +568,53 @@ get_inprogress_dict(void)
 	return inprogress;
 }
 
+PyObject *
+PyObject_Unicode(PyObject *v)
+{
+	PyObject *res;
+	
+	if (v == NULL)
+		res = PyString_FromString("<NULL>");
+	else if (PyUnicode_Check(v)) {
+		Py_INCREF(v);
+		return v;
+	}
+	else if (PyString_Check(v))
+	    	res = v;
+	else if (v->ob_type->tp_str != NULL)
+		res = (*v->ob_type->tp_str)(v);
+	else {
+		PyObject *func;
+		static PyObject *strstr;
+		if (strstr == NULL) {
+			strstr= PyString_InternFromString("__str__");
+			if (strstr == NULL)
+				return NULL;
+		}
+		if (!PyInstance_Check(v) ||
+		    (func = PyObject_GetAttr(v, strstr)) == NULL) {
+			PyErr_Clear();
+			res = PyObject_Repr(v);
+		}
+		else {
+		    	res = PyEval_CallObject(func, (PyObject *)NULL);
+			Py_DECREF(func);
+		}
+	}
+	if (res == NULL)
+		return NULL;
+	if (!PyUnicode_Check(res)) {
+		PyObject* str;
+		str = PyUnicode_FromObject(res);
+		Py_DECREF(res);
+		if (str)
+			res = str;
+		else
+		    	return NULL;
+	}
+	return res;
+}
+
 static PyObject *
 make_pair(PyObject *v, PyObject *w)
 {
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index a3678d5..c1f3d54 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -413,6 +413,7 @@ PyObject *PyUnicode_FromEncodedObject(register PyObject *obj,
     }
     else 
 	v = PyUnicode_Decode(s, len, encoding, errors);
+
  done:
     if (owned) {
 	Py_DECREF(obj);