Re: Minutes of the GTK+ Team Meeting - 2008-09-23
- From: Mike Kestner <mkestner gmail com>
- To: Michael Natterer <mitch gimp org>
- Cc: Gtk+ Developers <gtk-devel-list gnome org>, Alexander Larsson <alexl redhat com>, Morten Welinder <mortenw gnome org>
- Subject: Re: Minutes of the GTK+ Team Meeting - 2008-09-23
- Date: Thu, 25 Sep 2008 13:47:10 -0500
On Thu, 2008-09-25 at 19:24 +0200, Michael Natterer wrote:
> Currently 90-something-ish% of all children have to be packed into
> boxes using gtk_box_pack_foo(box, child, FALSE, FALSE, pad).
> That's totally redundant clutter and keeps people from using
> the generic gtk_container_add() API. If the default values
> were sane, only the expanding element would have to be
> packed using such a verbose API.
Your something-ish% statistic fills me with confidence in the
research. ;-)
I've packed a ton of widgets into boxes, and I couldn't have told you
that expand was the default. Changing the default isn't going to make
for any more confidence over what the defaults are. Likely result:
people will still explicitly specify their packing desires. Add in 2.0
to 3.0 compatibility and it will be even more important to explicitly
specify them.
I don't think it's insane for a linear layout container to default to
sharing the extra allocation amongst its members. I would put this in
the "things could be more beautiful" realm which doesn't rise to the
threshold of breaking compat.
--
Mike Kestner <mkestner gmail com>
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]