Apply two changes, systematically: (1) Use PyErr_NewException("module.class", NULL, NULL) to create the exception object. (2) Remove all calls to Py_FatalError(); instead, return or ignore the errors -- the import code now checks PyErr_Occurred() after calling a module's init function, so it's no longer a fatal error for the initialization to fail. Also did some small cleanups, e.g. removed unnecessary test for "already initialized" from initfpectl(), and unified initposix()/initnt(). I haven't checked this very thoroughly, so while the changes are pretty trivial -- beware of untested code!

--- a/Modules/stdwinmodule.c	Tue Sep 30 22:14:39 1997 +0000
+++ b/Modules/stdwinmodule.c	Wed Oct 01 04:29:29 1997 +0000
@@ -2651,13 +2651,11 @@
 	d = PyModule_GetDict(m);
 	
 	/* Initialize stdwin.error exception */
-	StdwinError = PyString_FromString("stdwin.error");
+	StdwinError = PyErr_NewException("stdwin.error", NULL, NULL);
 	if (StdwinError == NULL ||
 	    PyDict_SetItemString(d, "error", StdwinError) != 0)
-		Py_FatalError("can't define stdwin.error");
+		return;
 #ifdef WITH_THREAD
 	StdwinLock = allocate_lock();
-	if (StdwinLock == NULL)
-		Py_FatalError("can't allocate stdwin lock");
 #endif
 }