Re: [dwm] remembering client order in each tag

From: Sander van Dijk <a.h.vandijk_AT_gmail.com>
Date: Tue, 5 Sep 2006 22:54:28 +0200

On 9/5/06, Seth Graham <sether_AT_tru7h.org> wrote:
> I'm 100% not sure this is related to the discussion, but it seems to be
> sort of the same problem:
>
> z-ordering within a single tag doesn't even work consistently. I use dwm
> exclusively in float mode, and use mod1+mouse1 to raise windows to the
> top of the stack, in addition to mod1+tab.
>
> This invites all sorts of random behavior, whether one tag or more is
> being displayed.
>
> Open three xterms in a single tag, overlay them in a cascade and bring
> each one to the top of the stack to witness it. It only gets worse when
> more clients are open, and multiple tags are being viewed.
>
> In most window managers I have used, bringing a client to the top
> effects only that client. Bring a second client to the top, and the same
> thing happens. With dwm, bringing a client forward can send the old top
> level client all the way to the back, somewhere in the middle, or
> directly beneath the new top level client. It may bring other clients to
> the top, ones nowhere near where the action is taking place.

I was just writing a mail about that myself :-)
This is indeed related to the discussion, as it happens for the exact
same reason as the symptom Anthony described.
The reason is that dwm restacks all clients on focus change (mainly to
make sure that floating clients are _always_ above tiled clients). I
guess it would be better if dwm would only make sure that _all_
floating clients are above _all_ tiled clients, and leave the
z-ordering between the floating clients alone (except for putting the
newly focused one on top if it's a floating one of course), but
looking at restack() I see no easy way to do that so far...

Btw. if you're using dwm in floating mode only, you might want to take
a look at evilwm; chances are you'll like it I think.

Greetings, Sander.
Received on Tue Sep 05 2006 - 22:54:30 UTC

This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 14:30:54 UTC