Thanks for this PR!

I am not much into Python API, but somehow I get your comment in contradiction with the changes in the code, respectively:

PyErr_SetString is a void function. It doesn't return NULL.

But then the changes replace:

		Py_INCREF(Py_None);
		return Py_None;

With a NULL return:

return NULL;

Is it that after use of PyErr_SetString(...); it is not needed to reference Py_None, because it was not returning a NULL (or the Py_None) object?

Since (I think), Python does not really have NULL inside, but the None keyword, what is the impact of this change returning NULL?

Anything particularly wrong without this patch?

Practically I am trying to figure out what exactly was wrong and this PR fixes, because the code is like this for rather long time.

—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.