Re: TreeModel HAS_CHILD return

[Torsten Schoenfeld <kaffeetisch gmx de>]

Kevin Ryde wrote:

> gtk2perl_tree_model_iter_has_child looks like it uses POPi.  Could it
> test the return with sv_true (or whichever test it is that correctly
> handles magic)?

Yes, I think that would make sense.  Patch attached.  Does this work in
your case?

-- 
Bye,
-Torsten

Index: t/GtkTreeModelIface.t
===================================================================
RCS file: /cvsroot/gtk2-perl/gtk2-perl-xs/Gtk2/t/GtkTreeModelIface.t,v
retrieving revision 1.5
diff -u -d -p -r1.5 GtkTreeModelIface.t
--- t/GtkTreeModelIface.t	5 Oct 2005 19:13:07 -0000	1.5
+++ t/GtkTreeModelIface.t	4 May 2008 13:11:48 -0000
@@ -180,7 +180,7 @@ sub ITER_HAS_CHILD {
 	isa_ok ($list, "CustomList", "ITER_HAS_CHILD");
 	ok ($iter->[0] == $list->{stamp}, "ITER_HAS_CHILD");
 
-	return FALSE;
+	return 'asdf';
 }
 
 sub ITER_N_CHILDREN {
@@ -374,7 +374,7 @@ isa_ok ($iter = $model->iter_children(un
 isa_ok ($path = $model->get_path ($iter), "Gtk2::TreePath");
 is_deeply ([$path->get_indices], [0]);
 
-is ($model->iter_has_child ($iter), FALSE);
+is ($model->iter_has_child ($iter), TRUE);
 is ($model->iter_n_children ($iter), 0);
 
 isa_ok ($iter = $model->iter_nth_child (undef, 7), "Gtk2::TreeIter");
Index: xs/GtkTreeModel.xs
===================================================================
RCS file: /cvsroot/gtk2-perl/gtk2-perl-xs/Gtk2/xs/GtkTreeModel.xs,v
retrieving revision 1.52
diff -u -d -p -r1.52 GtkTreeModel.xs
--- xs/GtkTreeModel.xs	7 Jan 2008 20:23:29 -0000	1.52
+++ xs/GtkTreeModel.xs	4 May 2008 13:11:48 -0000
@@ -313,11 +313,13 @@ static gboolean
 gtk2perl_tree_model_iter_has_child (GtkTreeModel *tree_model,
                                     GtkTreeIter  *iter)
 {
+	SV *sv;
 	gboolean ret;
 	PREP (tree_model);
 	XPUSHs (sv_2mortal (sv_from_iter (iter)));
 	CALL ("ITER_HAS_CHILD", G_SCALAR);
-	ret = POPi;
+	sv = POPs;
+	ret = sv_2bool (sv);
 	FINISH;
 	return ret;
 }