[g-a-devel]Re: atkgobjectaccessible bug ...

["Padraig O'Briain" <Padraig Obriain Sun COM>]

Michael,

My expectation is that the first time this function is called 
quark_accessible_object will have the value 0. By the time the call to 
g_object_set_data() is reached I would expect that quark_accessible_object would 
be non-zero as atk_gobject_accessible_class_init() would have been called.

It seems to me that the accessible object which has been created does not derive 
from AtkGObjectAccessible. Is this what you intend should happen?

I could add a call to g_qurak_from_static_string() in this function but I would 
like to understand why atk_gobject_accessible_class_init() was not called before 
the call to g_object_set_data() was reached.

Padraig

> Subject: atkgobjectaccessible bug ...
> To: "Padraig O'Briain" <Padraig Obriain sun com>
> Cc: accessibility mailing list <gnome-accessibility-devel gnome org>
> Content-Transfer-Encoding: 7bit
> Mime-Version: 1.0
> 
> 
> 	It is sensible to call:
> 
> AtkObject*
> atk_gobject_accessible_for_object (GObject *obj)
> {
>   AtkObject* accessible;
> 
>   g_return_val_if_fail (G_IS_OBJECT (obj), NULL);
>   /* See if we have a cached accessible for this object */
> 
>   accessible = g_object_get_qdata (obj,
> 				   quark_accessible_object);
> ...
> }
> 
> 	Before instantiating a AtkGObjectAccessible. Sadly that means that the
> quark_accessible_object is not initialized:
> 
> (lt-nautilus:9153): GLib-GObject-CRITICAL **: file gobject.c: line 1339
> (g_object_set_qdata): assertion `quark > 0' failed
> 
> 	Be good to see it fixed.
> 
> 	Regards,
> 
> 		Michael.
> 	
> 
> 
> -- 
>  mmeeks gnu org  <><, Pseudo Engineer, itinerant idiot
>