On int to the negative integral power, let float handle it instead of

raising an error.  This was one of the two issues that the VPython
folks were particularly problematic for their students.  (The other
one was integer division...)  This implements (my) SF patch #440487.

1 files changed, 5 insertions, 7 deletions

diff --git a/Objects/intobject.c b/Objects/intobject.c
index de28156..b0ed82a 100644
--- a/Objects/intobject.c
+++ b/Objects/intobject.c
@@ -510,13 +510,11 @@ int_pow(PyIntObject *v, PyIntObject *w, PyIntObject *z)
 	CONVERT_TO_LONG(v, iv);
 	CONVERT_TO_LONG(w, iw);
 	if (iw < 0) {
-		if (iv)
-			PyErr_SetString(PyExc_ValueError,
-					"cannot raise integer to a negative power");
-		else
-			PyErr_SetString(PyExc_ZeroDivisionError,
-					"cannot raise 0 to a negative power");
-		return NULL;
+		/* Return a float.  This works because we know that
+		   this calls float_pow() which converts its
+		   arguments to double. */
+		return PyFloat_Type.tp_as_number->nb_power(
+			(PyObject *)v, (PyObject *)w, (PyObject *)z);
 	}
  	if ((PyObject *)z != Py_None) {
 		CONVERT_TO_LONG(z, iz);