Re: [wmii] wmii-4 fs proposal/discussion

From: Kris Maglione <>
Date: Fri, 9 Jun 2006 22:28:51 -0400

>N <x> <y> <w> <h> <class:instance:name>\n
>N is the global client index equivalent to /client/N/
>The number of lines defines the number of clients in the area. The selected
>client will be indicated with N==sel. <x> <y> <w> <h> indicates
>the geometry of this client in this area, the rest is equivalent
>to /client/N/props (however might be of use).
I agree with uriel that the floating area should not be represented as column
0 and I think that ~ is fine for consistency's sake.

I've got a few other problems, though. First, we shouldn't think of clients in
columns having a width so much as columns having widths. I think that tags
should have a file containing the widths of each column which should perhapps
be writable. Either way, we'd change the width of columns in the coltrol file,
not clients (floating clients excepted, of course).

As for adjusting height, that's another issue that's heavily entangled with
the new colmodes. I can think of two ways of dealing with it. One is to group
clients into associations with visible frames, with n frames existing at once
and 1 visible at a time. There would be a group file containing lines as:
"<group no> <visible client> <height> <title>" and perhaps a boolean of
whether to show 1 or all title bars. The other would be to give the index
file "<col|~> <client #> <height>". Floating clients would have x, y, w and h
in the index file either way.

I'm not really happy at all with either of these ideas. Please give me input.

It would be nice to read a brain dump of all that's currently known or decided
about the n clients per column change because I'm trying to deal with
something that's still somewhat amorpus in my mind.

Kris Maglione
Perscriptio in manibus tabellariorum est.
Received on Sat Jun 10 2006 - 04:29:25 UTC

This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 16:08:51 UTC