Re: [g-a-devel]Tooltips and accessibility

[Bill Haneman <bill haneman sun com>]

Michael Meeks wrote:
> 
> On Tue, 2002-03-26 at 17:06, Marc Mulcahy wrote:
> > I think the description might not be enough in the case of magnification or
> > some other application which wants to draw attention to the tooltip, since
> > the description wouldn't give the tooltip's location.  i understand
> > Michael's discontentmet with showing a new GailToplvel when a tooltip is
> > displayed.  Do we have any other options between not giving any
> > notification, and notifying the tooltip creation as a new GailToplevel?
> 
>         Could we not make the tooltip window a child of the accessible that
> threw it up ? though that's most likely ugliness and pain, hmm.

Yep, both I suspect.  Though it's not an unreasonable suggestion.

The way Java handles this is with API; there is a "getTooltip" or
something.  If it proves to be useful we can add this
(atk_object_get_tooltip) to a later revision of ATK.

>         Sigh; possibly the problem with tooltips is not emitting
> children-changed on the application when a tooltip window disappears -
> or (quite possibly) the resources associated with the tooltip are not
> destroyed, just hidden thus the Atk objects hang around [ and I still
> very strongly feel that proxies for un-mapped GtkWidgets should not
> appear in the tree at all ].

I agree with you most of the time; one possible exception is for 
widgets which are within a scrollpane but have perhaps never actually
been mapped to the display (i.e. haven't been scrolled into view).

For the most part, we are taking the position that accessibility is
presenting the "view" rather than the "model" of the UI, and thus
showing things that aren't visible makes no sense.  The case of a
scrollpane is the most notable exception that comes to mind.  We don't
want the children to come and go as the user scrolls, yuck...

It's also true that accessible objects need not be associated with
a GUI element at all, they may just be "UI" elements (i.e. a CLI
might still implement accessibility via ATK, but none of the objects 
would implement AtkComponent).

-Bill

>         Regards,
> 
>                 Michael.
> 
> --
>  mmeeks gnu org  <><, Pseudo Engineer, itinerant idiot
> 
> _______________________________________________
> Gnome-accessibility-devel mailing list
> Gnome-accessibility-devel gnome org
> http://mail.gnome.org/mailman/listinfo/gnome-accessibility-devel