[xmonad] darcs patch: NoFrillsDecoration - most basic version of decoration ...

[Jan Vornberger]

Hey!

On Mon, Oct 12, 2009 at 03:15:07AM -0400, Adam Vogt wrote:
> * On Monday, October 12 2009, Jan Vornberger wrote:
> 
> >This is the decoration layout modifier I use for Bluetile's floating layout. I wrote it
> >because SimpleDecoration wasn't simple enough. :-)
> >
> >Mon Oct 12 00:05:12 CEST 2009  Jan Vornberger <jan.vornberger at informatik.uni-oldenburg.de>
> >  * NoFrillsDecoration - most basic version of decoration for windows
> >
> >PS: I since looked into the problem of patch bundles with unicode characters
> >    getting mangled when I use 'darcs send'. It is now a reported
> >    bug ( http://bugs.darcs.net/issue1644 ) and I'm using a workaround that will
> >    hopefully work for now.
> 
> When you reset the layout (M-S-space), all windows take a shorter size.
> 
> ex. with this ordering:
> 
>             (named "F" $ borderResize
>                        $ noFrillsDeco shrinkText defaultTheme
>                        $ windowArrange simplestFloat)
> 
> The window shrinks every time you resize it horizontally too.
> 
> Could you document what the correct ordering of these pieces is (and/or
> just reference a XMonad.Config.Bluetile)?

In my recent patch bundle containing Bluetile's floating layout, I also
now reference a typical usage example for BorderResize. I hope this
clears this up a little bit.

I never really tested it with simplestFloat much. The problems you
observed are probably not avoidable, because of the way simplestFloat
works. It makes calls with xlib to figure out the sizes of the windows.
If decoration comes into play, the windows shrink a little bit to make
room for the decoration. Those shrunken windows are then again the bases
for simplestFloat if you do a reset etc., which results in them
shrinking every time.

This problem was one of the reasons why I decided to create the
PositionStore, because relying on those xlib calls to provide size
information just doesn't seem like a very reliable way.

So the BorderResize documentation now mentions, that it has only been
tested with PositionStoreFloat.

Regards!

Jan