[10596] in The GTK GIMP ToolKit mailing list archive

home help back first fref pref prev next nref lref last post

[gtk-list] RE: GtkList behaviour once more

daemon@ATHENA.MIT.EDU (Andreas Tille)
Wed Dec 2 09:21:02 1998

Date: Wed, 2 Dec 1998 15:20:29 +0100 (MET)
From: Andreas Tille <tille@physik.uni-halle.de>
To: "'gtk-list@redhat.com'" <gtk-list@redhat.com>
cc: gtk-devel-list@redhat.com
In-Reply-To: <99AA2270B1E6D111BCE10000F805F17F010A17E1@emss35m02.owg.fs.lmco.com>
Resent-From: gtk-list@redhat.com
Reply-To: gtk-list@redhat.com

On Wed, 2 Dec 1998, Rostedt, Steven wrote:

> I've stated earlier that there should be a mask that
> determines the button(s) that makes selections in the
> lists.  I've already updated my (now forked) code
> to handle this and I would love to share it.
> 
> Maybe we should send a note to the gtk-devel-list
> to get some attention.  This is a development 
> topic.  
Yes in dead.  I thought so and sended a mail to this list
one week ago but didn't got any answer.  That's why I CC-ed
my mail to this list while knowing that this breaks the
list policy.  Sorry for this behaviour but this topic is
really importand for me and I need a clear solution.  I
would like to do the work if someone of the developers would
tell me what the prefer and break the silence.

> At least have, as a minimum, a compiler
> option of -DMULTI_SELECT and use
> 
> #ifndef MULTI_SELECT
>    if (event->button != 1)
>       return FALSE;
> #endif
I think that isn't enough, because someone can't relay on the
compilation on different machines.  I want my program to behave
in the same way on each machine.

Yes a
   if ( !(event->button & GTK_LIST_SELECTION_MASK) )
      return FALSE;
together with
   gtk_list_get_selection_mask(...)
   gtk_list_set_selection_mask(...)
should fit all needs. (First plan)

May be we find a way to include this mechanism in
   gtk_list_set_selection_mode()  (Second plan)
or the best way I could think about would be to define
*different* selection_modes for different buttonsi (Third plan).

I never dealt with CVS stuff.  If there isn't anybody who
wants to implement this I would like to do this and post the
diffs here (or somebody can point me a better way to include
this).
Are there any objections against these plans?
Which plan should be prefered?
(In fact plan 0 would be to leave out the check for the button
and could be done very quickly.  May be you would prefer this one?)

Hopefully GTK+1.1.6 will give me a working as described GtkList
widget.

Kind regards

      Andreas.

-- 
To unsubscribe: mail -s unsubscribe gtk-list-request@redhat.com < /dev/null


home help back first fref pref prev next nref lref last post