Re: [PATCH 0/4] Native dockapp support
- From: Teika Kazura <teika lavabit com>
- To: sawfish-list gnome org
- Subject: Re: [PATCH 0/4] Native dockapp support
- Date: Sat, 12 Jun 2010 14:14:11 +0900 (JST)
On Tue, 8 Jun 2010 11:05:31 +0400, "Alexey I. Froloff" wrote:
> Sawfish just doesn't have place to draw application icon.
You're right.
> I think it's pretty well conforms to ICCCM specification, section
> 4. Main window in WithdrawnState => ignored by WM, icon supplied
> by client as window => display this window instead of pixmap.
No I've read ICCCM sec 4. It says that the window's icon is drawn when
the state is iconic. So it's better to introduce a new user option
in Sawfish which allows to assume that "Withdrawn + icon window ->
draw that icon".
In the meantime, if Window Maker has been doing so for a decade, and
it's ok, then it must be virtually ok.
(A window manager doesn't have to draw icons. If it does, it's a
feature.)
But you don't have to care about such new option. I can take that
part, so you can concentrate on icon.
> I shouldn't replace application window with it's icon_window. My
> C part needs to be rewritten.
Strictly speaking, you're right; icon windows are quite different,
whose most properties are ignored. But in our case, it's a workaround,
so adding some hacks may suffice.
> I really doubt that such thin exists in "documentation" form.
> There is libdockapp library (unmaintained, or in other words
> "rock stable") and wmgeneral.c source that being copied from one
> dockapp to another.
Together with ICCCM, now I'm sure that it's better to collect all lisp
codes into one file, wm.state.dockapp or wm.ext.dockapp.
For reference use of this message, let me point out that if
_NET_WM_HANDLED_ICON is set to a window, then the wm doesn't have to
provide an icon for that window.
Regards,
Teika (Teika kazura)
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]