Re: [evolution-patches] Patch about a11y bugs in widgets directory

[JP Rosevear <jpr novell com>]

On Wed, 2004-09-29 at 17:52 +0800, Li Yuan wrote:
> Hi, JP and Rodrigo
> 
> Now I have a patch in widgets directory for cvs HEAD. It  fixes some 
> a11y bugs. The patch has passed our internel review.
> Please help me to review it. Thanks a lot.

@@ -179,6 +181,10 @@
 						gnome_canvas_widget_get_type (),
 						"widget", button,
 						NULL);
+	a11y = gtk_widget_get_accessible (button);
+	if (a11y != NULL) {
+		atk_object_set_name (a11y, _("Previous Button"));
+	}

Also, why do we check for a non-null a11y object?  The gtk color
selector

Its usual not to wrap a single statement in {}.


Index: misc/e-combo-button.c
===================================================================
RCS file: /cvs/gnome/evolution/widgets/misc/e-combo-button.c,v
retrieving revision 1.18
diff -u -r1.18 e-combo-button.c
--- misc/e-combo-button.c	17 May 2004 17:10:39 -0000	1.18
+++ misc/e-combo-button.c	29 Sep 2004 09:36:24 -0000
@@ -390,6 +390,12 @@
 	}
 }
 
+static void
+impl_activated (GtkButton *button)
+{
+	gtk_signal_emit (GTK_OBJECT (button), signals[ACTIVATE_DEFAULT]);
+}
+

 static void
 class_init (GtkObjectClass *object_class)
@@ -408,6 +414,7 @@
 
 	button_class = GTK_BUTTON_CLASS (object_class);
 	button_class->released = impl_released;
+	button_class->activate = impl_activated;
 
 	signals[ACTIVATE_DEFAULT] = gtk_signal_new ("activate_default",
 						    GTK_RUN_FIRST,

Why is this emitting the activated signal now? in impl_released we
explicity override the default activate behaviour.

-JP
-- 
JP Rosevear <jpr novell com>
Novell, Inc.