Re: [evolution-patches] patch for bug 50392

[Not Zed <notzed ximian com>]

Thats a very odd thing to suggest Mike.

Why the added stack frame, for no other beneifit than to give the compiler more symbols to play with?


On Sat, 2003-11-01 at 01:16, Mike Kestner wrote:
> On Fri, 2003-10-31 at 00:44, Bolian Yin wrote:
> > Hi,
> > 
> > provide more a11y implementation for e-table.
> > http://bugzilla.ximian.com/show_bug.cgi?id=50392
>
> > -	GalA11yETableItem *a11y = GAL_A11Y_E_TABLE_ITEM (accessible);
> > +	AtkGObjectAccessible *atk_gobj;
> > +	GObject *g_obj;
> > +	GalA11yETableItem *a11y;
> > +
> > +	g_return_val_if_fail (GAL_A11Y_IS_E_TABLE_ITEM (accessible), NULL);
> > +	atk_gobj = ATK_GOBJECT_ACCESSIBLE (accessible);
> > +	g_obj = atk_gobject_accessible_get_object (atk_gobj);
> > +	if (g_obj == NULL)
> > +		/* Object is defunct */
> > +		return NULL;
> > +
> > +	a11y = GAL_A11Y_E_TABLE_ITEM (accessible);
>
> This code repeats constantly throughout the patch.  Please put this is a
> function.  Something like:
>
> static GObject *
> get_gobject (AtkObject accessible)
> {
>         return atk_gobject_accessible_get_object (ATK_GOBJECT_ACCESSIBLE
> (accessible));
> }
>
> So you can just do:
>
>        GObject *g_obj = get_gobject (accessible);
>
> Otherwise, it looks fine.  Please update the patch to include the above
> and commit to trunk.  Thanks for the patch.