Re: Dual-head with one head rotated?

Quoting Daniel German <dmg uvic ca> on Wed, Jul 23 18:36:
> (define (window-avoided-p w)
[ snip ]
> create a sawfish function that moves a window to the area you need
> (using move-resize-window-to) and set its attributes to sticky, avoid
> and ignore using window-put. That should be enough.

Thanks for the idea, but that seems like a huge kludge.  It seems to me
that sawfish should be able to figure that out by itself.  Another
problem with the avoided window is that it will not keep the mouse from
traveling into the dead space.  

Sawfish already knows about the dimensions of the heads:
> sawfish-client:
user> (current-head-dimensions )
(1680 . 1050)
[ move window to other head ]
user> (current-head-dimensions )
(1024 . 1280)

It also knows that the dead space in the ascii art below isn't actually
in a head:
user> (find-head 0 1049)
user> (find-head 0 1050)

Am I the only person trying to run sawfish with mis-matched screens?

> On Wed, Jul 23, 2008 at 6:05 PM, Omen Wild <omen+sawfish mandarb com> wrote:
> I have two questions about using sawfish with dual monitors, but with one
> rotated.  I have a 1680x1050 LCD hooked up to the DVI port of an Intel
> 965Q on a Dell Optiplex system.
> I have a 1280x1024 LCD hooked up to the VGA port.  I added 'Virtual 2704
> 1280' to the SubSection "Display" part of my xorg.conf and ran the
> following command:
> xrandr --output VGA --auto --rotate left --right-of TMDS-1
> This pulled up the second display rotated (woohoo!) and the speed of
> the rotated display is even pretty fast (double woohoo!).
> The problem is that sawfish thinks the screen is a square of 2704x1280
> pixels, and will happily place and allow windows in the dead zone in
> the bottom left, i.e. in the xxx's of this bad ascii art:
>  __________________
> |           . 1024 |
> | 1680 x    . 1280 |
> | 1050      .      |
> |           .      |
> |___________.      |
>  xxxxxxxxxxx|      |
>  xxxxxxxxxxx|______|
> Is there some way to convince sawfish that the x'd out part really does
> not exist and it cannot use it?  I would even be willing to hand edit
> some source and custom compile if it would fix the issue.
> Second question, is there a way to make the edge between the two screen
> (the vertial part with the dots) act as if it is the edge of a window,
> so when you move windows (or send windows with the 'Corner upper right'
> command in <>) they stop at the
> intersection of the two screens?
> The OS is Debian/unstable fully up to date:
> ii  xserver-xorg             1:7.3+14                 the X.Org X server
> ii  xserver-xorg-core        2:1.4.2-2                Xorg X server - core server
> ii  xserver-xorg-video-intel 2:2.3.2-2                X.Org X server -- Intel i8xx, i9xx display driver

Never judge a man by his taglines.

Attachment: signature.asc
Description: Digital signature

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]