RE: [g-a-devel]Notebook's selection interface ...
- From: "Padraig O'Briain" <Padraig Obriain Sun COM>
- To: mukund rajagopalan wipro com, michael ximian com
- Cc: gnome-accessibility-devel gnome org
- Subject: RE: [g-a-devel]Notebook's selection interface ...
- Date: Mon, 25 Mar 2002 10:51:31 +0000 (GMT)
I would like to get view from Bill and Marc on how we should deal with double
clicking on a cell in a GtkTreeView. Should we define an "activate" action on
the cell or how should we emulate the user action some other way?
In Michael's original mail he referred to clicking on one of the rows. I believe
this is equivalent to adding row selection.
The remainder of this mail discusses AT-SPI so Hopefully Bill or Marc will
comment as soon as they have got through their in-boxes.
Padraig
> Subject: RE: [g-a-devel]Notebook's selection interface ...
> To: mukund rajagopalan wipro com
> Cc: "Padraig O'Briain" <Padraig Obriain Sun COM>, accessibility mailing list
<gnome-accessibility-devel gnome org>
> Content-Transfer-Encoding: 7bit
> Mime-Version: 1.0
>
> Hi Mukund,
>
> On Sat, 2002-03-23 at 11:19, Mukund wrote:
> > To summarize (skip, if you already got it:)
> > (1) Selection interface on tree-table (parent) seems to be a read-only
> > interface - neither does it allow to select (focus) the cells nor click
> > them.
>
> Quite I tend to think using the selection interface for that is
> confusing and inadequate and we should have the 'row_activated' signal
> (action) instead.
>
> > (2) Table-interface: Though add_row_selection is implemented in gail for
> > GailTreeView, there is no complimentary method in CSPI. When done, this
> > should allow the sub-rows to be selected (focused).
>
> This needs to be in Bugzilla and dealt with soon, we can still add
> at-spi now if it's done quickly & approved.
>
> > (3) Action (click) is implemented for table cells that has sub-rows.
> > AccessibleAction_doAction, for these cells, expands and collapses the
> > children (view).
>
> We should IMHO have an expand and collapse action [ perhaps ], but
> that's fine for those cells at least.
>
> Also, the following APIs need to be registered in Buzilla as needing an
> SPIBoolean return type, since we can do that API change at the same time
> perhaps:
>
> void
> AccessibleStateSet_add (AccessibleStateSet *obj,
> AccessibleState state);
> void
> AccessibleStateSet_remove (AccessibleStateSet *obj,
> AccessibleState state);
> void
> AccessibleSelection_clearSelection (AccessibleSelection *obj);
>
> void
> AccessibleImage_getImageSize (AccessibleImage *obj,
> long int *width,
> long int *height);
>
> void
> AccessibleImage_getImagePosition (AccessibleImage *obj,
> long int *x,
> long int *y,
> AccessibleCoordType ctype);
>
> void
> AccessibleImage_getImageExtents (AccessibleImage *obj,
> long int *x,
> long int *y,
> long int *width,
> long int *height,
> AccessibleCoordType ctype);
>
> void
> AccessibleHyperlink_getIndexRange (AccessibleHyperlink *obj,
> long int *startIndex,
> long int *endIndex);
>
> void AccessibleComponent_grabFocus (AccessibleComponent *obj);
> void AccessibleComponent_getExtents (AccessibleComponent *obj,
> long int *x,
> long int *y,
> long int *width,
> long int *height,
> AccessibleCoordType ctype);
> void AccessibleComponent_getPosition (AccessibleComponent *obj,
> long int *x,
> long int *y,
> AccessibleCoordType ctype);
> void AccessibleComponent_getSize (AccessibleComponent *obj,
> long int *width,
> long int *height);
>
>
> It'd be great to file those too,
>
> Regards,
>
> Michael.
>
>
>
> --
> mmeeks gnu org <><, Pseudo Engineer, itinerant idiot
>
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]