Re: move/resize handling again
- From: Dominik Vogt <dominik vogt gmx de>
- To: wm-spec-list gnome org
- Subject: Re: move/resize handling again
- Date: Thu, 14 Nov 2002 00:49:47 +0100
On Wed, Nov 13, 2002 at 02:57:25PM -0600, Ben Jansens wrote:
> On Wed, Nov 13, 2002 at 07:23:44PM +0100, Dominik Vogt wrote:
> > On Wed, Nov 13, 2002 at 06:30:52PM +0100, Dominik Vogt wrote:
> > > On Wed, Nov 13, 2002 at 11:47:00AM -0500, Havoc Pennington wrote:
> > > > On Wed, Nov 13, 2002 at 10:48:56AM +0100, Dominik Vogt wrote:
> > > > > The legacy way is to ignore win_gravity for ConfigureRequest and
> > > > > always use StaticGravity.
> > > >
> > > > Which WMs historically do that?
> >
> > I did a quick tests of many of the window managers shipping with
> > Debian 3.0. I fired up a mozilla 1.1 window and pressed F11 to
> > resize it to full screen mode and back. Results:
> >
> > x = Crashes, does not move the window or shows very strange
> > behaviour
> > i = ICCCM compliant
> > n,w,nw,se = Window travels in the given direction
> > - = not tested
>
> Well, just to tip the scales :) I did this in my own.
Hehe, I just tested the ones with "wm" in their names plus a few
others that I remembered :-) And I left out the wms that we
already know how they work.
> > On a "legacy" wm, the window should travel nw. In no particular
> > order:
[snip]
> > 8 x
> > 8 i
> 9 i
> > 5 nw
> > 4 n/se/w
> > 3 -
> > 28 total
I should have elaborated on the 'x' and 'n/se/w' factions a bit.
Most of them work the "legacy" way when making the window full
screen: they place the top left corner of the client window at
0 0 and hide the wm frame. But when you press F11 again to get
back the old geometry, they do funny things: shift the window
to one of the directions, keep the full screen size, place it at
the pointer position, hang, crash, etc.
To sum it up: there are lots of wms representing either school
(and lots that were not even tested). On average, the ICCCM
compliant wms are less buggy. There are a number of well written
wms of the "legacy" school, but a whole lot more that are totally
buggy. To be honest I am a bit surprised that there are at least
8 different ways wms treat COnfigureRequests :-( Horrible.
Bye
Dominik ^_^ ^_^
--
Dominik Vogt, dominik vogt gmx de
Reply-To: dominik vogt gmx de
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]