Re: [g-a-devel]Tooltips and accessibility
- From: Marc Mulcahy <marc mulcahy sun com>
- To: Michael Meeks <michael ximian com>
- Cc: "Padraig O'Briain" <Padraig Obriain sun com>, accessibility mailing list <gnome-accessibility-devel gnome org>
- Subject: Re: [g-a-devel]Tooltips and accessibility
- Date: Thu, 28 Mar 2002 10:03:05 -0700
Yes, not exposing unmapped widgets makes sense to me. What does this mean
from an implementation point of view-- does it mean that we change the way
GailContainer works? When we count children and do refAccessibleChild we
just have to check if the children are mapped? Or am I being daft?
Marc
At 12:03 PM 3/28/2002 +0000, Michael Meeks wrote:
Hi Marc,
On Wed, 2002-03-27 at 21:13, Marc Mulcahy wrote:
> I like the idea of making the AtkObject for the tooltip a child of the
> object which threw it up. padraig, is this feasible?
Sounds like it might be horribly difficult to do to me; the way
tooltips are handled is not going to be extremely easy to do anything
with IMHO - since we keep a single persistent GtkWindow around to map /
unmap where we want it as necessary - which [ due to the fact that
AtkObjects are always exposed regardless of whether windows are mapped ]
means that the AtkObject will stay in the hierarchy.
My personal feeling is [ cf. the huge hidden menu hierarchy in
nautilus
being exposed ] that we should not expose unmapped GtkWidget's
accessibles at all; since by definition you can't see them - and
everyone assumes that therefore you can't use them either :-)
Regards,
Michael.
--
mmeeks gnu org <><, Pseudo Engineer, itinerant idiot
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]