GNOME.org
 

Re: Buttons and AtkValue

Padraig,

This all seems right to me.

Marc

At 06:16 PM 7/10/2001 +0100, Padraig O'Briain wrote:

Hi,

I am slowly getting my head around work stuff after too long off. Comments
requested on my proposal below.

I propose to do the following:

1) Remove the implementation of AtkValue from GailToggleButton.

2) Cause a property change for "accessible-state" to be generated when a
togglebutton is checked or unchecked either manually or programatically by
calling atk_action_do_action (..., 0).

3) If a togglebutton or radio button is checked its state set will include
ATK_STATE_CHECKED.

4) Remove the implementation of atk_object_ref_relation_set() from GailButton.
This implementation is incorrect as it specifies that all buttons in a
GtkButtonBox are in a MEMBER_OF relationship. It will be replaced by an
implementation for radio buttons so that all radio buttons which are in
the same Button Group are in a MEMBER_OF relationship.

Padraig

>
>
> Hi Padraig,
>
> > Are we saying that Java's implementation of AccessibleValue on
AbstractButton
> > is a mistake?
> >
> > I have no difficulty removing AtkValue from GaelToggleButton.
>
> More like overkill than a mistake... I don't think it's a problem, and again, 
I
> think it's necessary for tri-state check boxes.  But I don't see a good
argument
> for it being on AbstractButton.
>
> Peter
>
> > > Delivered-To: gnome-accessibility-list gnome org
> > > X-Accept-Language: en
> > > MIME-Version: 1.0
> > > To: gnome-accessibility-list gnome org
> > > Subject: Re: Buttons and AtkValue
> > > Content-Transfer-Encoding: 7bit
> > > X-BeenThere: gnome-accessibility-list gnome org
> > > X-Loop: gnome-accessibility-list gnome org
> > > X-Mailman-Version: 2.0beta5
> > > List-Id: GNOME accessibility development
<gnome-accessibility-list.gnome.org>
> > >
> > > Hi guys,
> > >
> > > > I agree with Bill here. I think the semantics of what it means to have 
a
> > > > value are important (I.E., the idea of having multiple values with a min 
> > > > and max).
> > > >   Exposing a button as having this goes against that idea.
> > >
> > > Let me join you in agreement, mostly...
> > >
> > > Two state buttons don't need values, and probably shouldn't implement
them.  A
> > > number of programs implement tri-state check boxes. Should those be done 
with
> > > AtkValue?  If not, what are the states exposed for such a check box
(certainly
> > > not "checked" and "unchecked"...).
> > >
> > > Peter
> > >
> > > > From: Bill Haneman <bill haneman ireland sun com>
> > > > To: "Padraig O'Briain" <Padraig Obriain Sun COM>
> > > > Cc: gnom        e-accessibility-list gnome org
> > > > Subject: Re: Buttons and AtkValue
> > > > Date: Thu, 26 Apr 2001 11:52:32 +0100
> > > > Padraig O'Briain wrote:
> > > >  >
> > > > > My excuse here is that I am following Java. AccessibleAbstractButton 
> > > > implements
> > > > > AccessibleValue to set a button to be selected (or not) and get the 
> > selected
> > > >  > state. I propose to follow the same convention for ATK.
> > > >
> > > > This still seems broken to me.
> > > >
> > > > For one thing, it complicates life for the AT since it makes it harder 
> > > > to detect "range controllers" as opposed to just plain
> > > > togglebuttons... a range controller with a range of 0 to 1 and no
> > > > intermediate values seems like a degenerate case.  OK, a *truly*
> > > > degenerate case would have min==max, we ignore that one ;-)
> > > >
> > > > It also seems totally unnecessary since a combination of get_state and 
> > > > do_action entirely duplicates the functionality of AtkValue for this
> > > > case.  Usually I like to keep consistent with the Java API but this
> > > > might be an exception.
> > > >
> > > > -Bill
> > > >
> > > >  > My understanding is that "selected" is equivalent to "checked". I
think
> > > > I lied
> > > >  > in my previous statement about not having an action "toggle". A
button
> > > > has one
> > > >  > action which is to click the button which is probably the same as
toggle.
> > > >  >
> > > >  > Padraig
> > > >  >
> > > >  > >
> > > >  > >
> > > >  > > >GtkToggleButton Notes:
> > > > > > >The comments on GtkButton and AtkAction interface also apply here; 
> > > > i.e. one
> > > >  > action on a button. I have
> > > > > > >GaelToggleButton implementing AtkValue interface to set the button 
to
> > > >  > > >active or
> > > >  > > >not so I do not have an action "toggle".
> > > >  > > >
> > > >  > > >***
> > > >  > > My understanding of AtkValue is that it is for things which can
have a
> > > > > > range of values, and that the "checked" state is for objects which 
only
> > > >  > > have two states.
> > > >  > > ***
> > > >
> > > > --
> > > > --------------
> > > > Bill Haneman
> > > > Gnome Accessibility / Batik SVG Toolkit
> > > > Sun Microsystems Ireland
> > > >
> > > > References:
> > > > Buttons and AtkValue
> > > > From: Padraig O'Briain
> > > > [
> > > > Date Prev][
> > > > Date Next]   [
> > > > Thread Prev][
> > > > Thread Next]   [
> > > > Thread Index] [
> > > > Date Index] [
> > > > Author Index]
> > > >
> > > >
> > > > _______________________________________________
> > > > Gnome-accessibility-list mailing list
> > > > Gnome-accessibility-list gnome org
> > > > http://mail.gnome.org/mailman/listinfo/gnome-accessibility-list
> > >
> > > _______________________________________________
> > > Gnome-accessibility-list mailing list
> > > Gnome-accessibility-list gnome org
> > > http://mail.gnome.org/mailman/listinfo/gnome-accessibility-list
> >
> > _______________________________________________
> > Gnome-accessibility-list mailing list
> > Gnome-accessibility-list gnome org
> > http://mail.gnome.org/mailman/listinfo/gnome-accessibility-list
>
>


_______________________________________________
gnome-accessibility-list mailing list
gnome-accessibility-list gnome org
http://mail.gnome.org/mailman/listinfo/gnome-accessibility-list
[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]