Re: [dwm] ii best practices

2008-04-02 Thread pancake
i think scrollz already have this.

 Anselm R. Garbe [EMAIL PROTECTED] wrote:

 On Tue, Apr 01, 2008 at 03:24:44PM +0200, Matthias-Christian Ott wrote:
  I want to start using ii for irc. What discouraged me from doing so
 was
  a proper input method. Currently I know 2 methods to effectively input
  text to ii: vim and dinput. I don't like both for some reason.
  Do you have any best practices for using ii?

 echo(1), I'm not using ii but sic, but I usually create a fifo
 for sending input to sic using echo or sometimes cat(1), if I plan
 to have longer conversations than the usual hello.

 But this would require me to write something like each time I want to
 send something to the channel:

 $ echo 'Hello'  fifo
 # or
 $ cat  fifo
 Hello^D

 Of course you could do:
 $ while true; do cat  fifo; done

 But what about this:

   +---+
   |   |
   |   urxvt   |
   |tail -f fifo   |
   |   |
   |   |
   |...|
   |   input   |
   +---+

 The basic idea is to embedd the urxvt window in another window which also
 embedds an input window. The input window could be urxvt running a small
 programme or small X11 programme reading input and writing to the fifo.
 This is much nicer for tiling and feels more like an irc client.
 You could do this with screen of course.

 Any other ideas?

 Regards
 Matthias-Christian







Re: [dwm] ii best practices

2008-04-02 Thread Matthias-Christian Ott
pancake [EMAIL PROTECTED] wrote:

 i think scrollz already have this.

But scrollz is an ununix irc client.

Regards
Matthias-Christian



Re: [dwm] ii best practices

2008-04-02 Thread pancake
 pancake [EMAIL PROTECTED] wrote:

 i think scrollz already have this.

 But scrollz is an ununix irc client.

but scrollz is composed by two programs

 - a terminal control with the described functionality
 - an irc client that uses the terminal control tool

imho the terminal control is what you described in the other mail





Re: [dwm] ntile layout for dwm hg tip

2008-04-02 Thread Matthias-Christian Ott
Nibble [EMAIL PROTECTED] wrote:

 Hi there,

  And I can say that for xinerama we will probably need to adapt it to
  support also vertical tiling. Using horitzontal it's mostly a waste of
  space on big screens.

 Yes, I think you are right. Furthermore, horizontal tiling (Anselm's 
 notation) 
 could be useful for some other geoms (eg bottom).

  I would like to see something like vntile and hntile or just a toggle for
  it. What do you think about this?

 It is a good idea. In the patch there are now four layouts called:
   -|=, ntilevv
   -||, ntilevh
   ||=, ntilehv
   |||, ntilehh
 to fit all the cases: 2 small screens, 2 big screens, 1 small and/or 1 big 
 screen. People can safely remove from the patch those ones they don't 
 need.

You could simplify the code by introducing to two global variables that
indicate whether the master or tiled area is horizontal.
So switching to |= would simply set tileh = True and leave masterh
unchanged and switchting to -|= would set tileh = True and masterh = True.
So you could simple toggle the behaviour (e.g. MOD-v - masterh = False,
MOD-Shift-v - tileh = False) and don't need four layouts.

Anyway I think some kind of nmaster patch should go into dwm hg tip.

Regards
Matthias-Christian



Re: [dwm] A rather radical thought

2008-04-02 Thread Sander van Dijk
On Wed, Apr 2, 2008 at 2:59 PM, Anselm R. Garbe [EMAIL PROTECTED] wrote:
 What about getting rid of supporting the selection of multiple
  tags at a time and celebrating the revival of the column layout
  of wmii again in dwm 5.0? What about getting rid of all the tiled
  layout flavors in favor for the column layout?

You're one day late!

If you're serious though, what exactly do you mean? wmii-3/acme like
(as in: dynamic amount of columns, and three modes per column)?

Gr. Sander.



Re: [dwm] A rather radical thought

2008-04-02 Thread hiro
it would be hilariously great.
i've tried it, but never saw the point of this not-so-revolutionary concept.



Re: [dwm] A rather radical thought

2008-04-02 Thread Szabolcs Nagy
On 4/2/08, Anselm R. Garbe [EMAIL PROTECTED] wrote:
 What about getting rid of supporting the selection of multiple

\o/

 tags at a time and celebrating the revival of the column layout
 of wmii again in dwm 5.0? What about getting rid of all the tiled
 layout flavors in favor for the column layout?

that's radical indeed
but sounds cleaner than the current solution so
+1



Re: [dwm] A rather radical thought

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 03:23:39PM +0200, Sander van Dijk wrote:
 On Wed, Apr 2, 2008 at 2:59 PM, Anselm R. Garbe [EMAIL PROTECTED] wrote:
  What about getting rid of supporting the selection of multiple
   tags at a time and celebrating the revival of the column layout
   of wmii again in dwm 5.0? What about getting rid of all the tiled
   layout flavors in favor for the column layout?

 If you're serious though, what exactly do you mean? wmii-3/acme like
 (as in: dynamic amount of columns, and three modes per column)?

Exactly.

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] A rather radical thought

2008-04-02 Thread Matthias-Christian Ott
Anselm R. Garbe [EMAIL PROTECTED] wrote:

 What about getting rid of supporting the selection of multiple
 tags at a time and celebrating the revival of the column layout
 of wmii again in dwm 5.0? What about getting rid of all the tiled
 layout flavors in favor for the column layout?

I haven't tested wmii for a long time, but if you mean want to turn dwm
into something more acme-like, I vote against it.
I'm using dwm on a fairly small screen (17'', 1024x768), but it works
for me with 2 windows in the master area:

  ++--+
  |$   |  |
  ||  |
  ||  |
  ++  |
  |$   |  |
  ||  |
  ||  |
  ++--+

But I would prefer something like this:

  ++--+
  ||  |
  ||  |
  ++--+
  ||  |
  ||  |
  +--+-+--+
  |  | |  |
  |  | |  |
  +---+

Additionally I would like to be able to stack multiple tags. So if I
select 1, 2 and 3 each one is a separated layer which you can raise or
lower. This is much like maximised windows that you cycle through. So
a tag in this layout is much more like a group.
It would be also nice in this layout to lower individual windows so
that by lowering them they become member of tag behind the current tag
in the stack.
If you want to allow raising that way you would of course need
transparency.
Maybe the status bar should also the tags, so that you can raise a certain
tag by clicking on its name.

So the idea of tag layout should be worth considering (image a big screen
with floating resizeable tags).

Regards
Matthias-Christian



Re: [dwm] A rather radical thought

2008-04-02 Thread Marc Andre Tanner

Anselm R. Garbe wrote:

What about getting rid of supporting the selection of multiple
tags at a time and celebrating the revival of the column layout
of wmii again in dwm 5.0? What about getting rid of all the tiled
layout flavors in favor for the column layout?


Could we please stop rewriting dwm for the n-th time and focus
on st instead!

Thanks,
Marc

--
 Marc Andre Tanner  http://www.brain-dump.org/  GPG key: CF7D56C0



Re: [dwm] A rather radical thought

2008-04-02 Thread hiro
I am still eager to sea a really acme-like *interface* additionally to
only it's layout. But i assume this is not your ambition.



Re: [dwm] A rather radical thought

2008-04-02 Thread Sander van Dijk
On Wed, Apr 2, 2008 at 3:34 PM, Anselm R. Garbe [EMAIL PROTECTED] wrote:
 On Wed, Apr 02, 2008 at 03:23:39PM +0200, Sander van Dijk wrote:
  If you're serious though, what exactly do you mean? wmii-3/acme like
   (as in: dynamic amount of columns, and three modes per column)?

  Exactly.

Hm, would that mean you'd have to introduce titlebars for stack mode?



Re: [dwm] A rather radical thought

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 03:53:10PM +0200, Sander van Dijk wrote:
 On Wed, Apr 2, 2008 at 3:34 PM, Anselm R. Garbe [EMAIL PROTECTED] wrote:
  On Wed, Apr 02, 2008 at 03:23:39PM +0200, Sander van Dijk wrote:
   If you're serious though, what exactly do you mean? wmii-3/acme like
(as in: dynamic amount of columns, and three modes per column)?
 
   Exactly.
 
 Hm, would that mean you'd have to introduce titlebars for stack mode?

No, I would only consider columns with usual tiling and the
possibility to grow/shrink the current window resp. column.

No modes in a column.

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] A rather radical thought

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 03:37:49PM +0200, Marc Andre Tanner wrote:
 Anselm R. Garbe wrote:
 What about getting rid of supporting the selection of multiple
 tags at a time and celebrating the revival of the column layout
 of wmii again in dwm 5.0? What about getting rid of all the tiled
 layout flavors in favor for the column layout?
 
 Could we please stop rewriting dwm for the n-th time and focus
 on st instead!

It won't be a rewrite, just a change in the basic concepts.
As long as dwm does not feel perfect to me, I continue changing
it. Of course st is one of my really high priorities.

Regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



[dwm] setmfact improvements

2008-04-02 Thread Peter Hartlich
Hi,

Attached are two small patches to setmfact. The first one removes some
redundancy, saving 5 LOC; the second one reintroduces the MFACT define
that was present (as MWFACT) before, so that you can do setmfact(NULL)
to restore your initial master factor.

Taken together, they add 1 LOC.

Regards,
Peter
diff -r 3ac7eb240b52 dwm.c
--- a/dwm.c Tue Apr 01 15:45:37 2008 +0100
+++ b/dwm.c Wed Apr 02 15:25:44 2008 +0200
@@ -1467,21 +1467,16 @@
 
 void
 setmfact(const char *arg) {
-   double delta;
+   double d;
 
if(!arg || lt-isfloating)
return;
-   delta = strtod(arg, NULL);
-   if(arg[0] == '-' || arg[0] == '+') {
-   if(mfact + delta  0.1 || mfact + delta  0.9)
-   return;
-   mfact += delta;
-   }
-   else {
-   if(delta  0.1 || delta  0.9)
-   return;
-   mfact = delta;
-   }
+   d = strtod(arg, NULL);
+   if(arg[0] == '-' || arg[0] == '+')
+   d += mfact;
+   if(d  0.1 || d  0.9)
+   return;
+   mfact = d;
setgeom(geom-symbol);
 }
 
diff -r 3c8ba4a4092d config.def.h
--- a/config.def.h  Wed Apr 02 15:55:43 2008 +0200
+++ b/config.def.h  Wed Apr 02 15:59:51 2008 +0200
@@ -20,7 +20,7 @@
 
 /* geometries, s{x,y,w,h} and bh are already initualized here */
 /*   func name  bx  by  bw  wx  wy  ww wh  mx  my   mw  mh tx  ty  
   tw  th  mox moy mow moh */
-double mfact = 0.55;
+#define MFACT  0.55/* master width factor [0.1 .. 0.9] */
 DEFGEOM(single,  0,  0, sw,  0, bh, sw, sh-bh, wx, wy, mfact*sw, wh, mx+mw, 
wy, ww-mw, wh,  wx, wy, ww, wh)
 DEFGEOM(dual,0,  0,1280, 0, bh, ww, wh-bh, wx, wy, 1280,800-bh,  1280,  0, 
ww-mw, sh,  mx, my, mw, mh)
 
diff -r 3c8ba4a4092d dwm.c
--- a/dwm.c Wed Apr 02 15:55:43 2008 +0200
+++ b/dwm.c Wed Apr 02 15:59:51 2008 +0200
@@ -212,6 +212,7 @@
 int screen, sx, sy, sw, sh;
 int (*xerrorxlib)(Display *, XErrorEvent *);
 int bx, by, bw, bh, blw, bgw, mx, my, mw, mh, mox, moy, mow, moh, tx, ty, tw, 
th, wx, wy, ww, wh;
+double mfact;
 unsigned int numlockmask = 0;
 void (*handler[LASTEvent]) (XEvent *) = {
[ButtonPress] = buttonpress,
@@ -1469,14 +1470,18 @@
 setmfact(const char *arg) {
double d;
 
-   if(!arg || lt-isfloating)
+   if(lt-isfloating)
return;
-   d = strtod(arg, NULL);
-   if(arg[0] == '-' || arg[0] == '+')
-   d += mfact;
-   if(d  0.1 || d  0.9)
-   return;
-   mfact = d;
+   if(!arg)
+   mfact = MFACT;
+   else {
+   d = strtod(arg, NULL);
+   if(arg[0] == '-' || arg[0] == '+')
+   d += mfact;
+   if(d  0.1 || d  0.9)
+   return;
+   mfact = d;
+   }
setgeom(geom-symbol);
 }
 
@@ -1496,6 +1501,7 @@
sw = DisplayWidth(dpy, screen);
sh = DisplayHeight(dpy, screen);
bh = dc.font.height + 2;
+   mfact = MFACT;
geom = geoms[0];
geom-apply();
 


Re: [dwm] A rather radical thought

2008-04-02 Thread Matthias-Christian Ott
Anselm R. Garbe [EMAIL PROTECTED] wrote:

 Here is what I have exactly in mind from a user perspective:

 Initially there is 1 column, a new client is inserted below the
 currently selected window in the column, similiar to the stack
 in tiled layout.

 Each window can be moved left- or rightwards (Mod1-Shift-{h,l})
 which may result in a new column if there is none.  A
 new column gets half the size of the current column.

 Each window can also be moved up- or downwards in the column
 itself (Mod1-Shift-{j,k}).

 Columns can be grown/shrinked using Mod1-Control-{h,l}.
 Windows can be grown/shrinked using Mod1-Control-{j,k}.

 The navigation is rather straight-forward:

 Focussing the previous/next window in the current column is done
 with Mod1-{j,k}, focussing the previous/next column is done with
 Mod1-{h,l}.

 So the whole layout concept consists of basically 4 keys with 3
 kinds of modifiers.

But this wouldn't exclude the tag concept of dwm, it just means giving
up this tiled and master area concept, or did I get it wrong?

Is your concept limited to horizontal columns or are vertical columns or
both simultaneously possible?

Regards
Matthias-Christian



Re: [dwm] A rather radical thought

2008-04-02 Thread Fabio Scotoni

On Wed, 2 Apr 2008, Anselm R. Garbe wrote:

What about getting rid of supporting the selection of multiple
tags at a time and celebrating the revival of the column layout
of wmii again in dwm 5.0? What about getting rid of all the tiled
layout flavors in favor for the column layout?


I would like it :) But i need modes per column in this case.
Stack (That would need titlebars...), maximized and tiled.
But are the modes are even possible in dwm with only 2000 lines of code?

Regards,
Fabio Scotoni




Re: [dwm] A rather radical thought

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 04:37:47PM +0200, Matthias Kirschner wrote:
 * Anselm R. Garbe [EMAIL PROTECTED] [2008-04-02 16:11:54 +0200]:
 
  Initially there is 1 column, a new client is inserted below the
  currently selected window in the column, similiar to the stack
  in tiled layout.
 
 If I open a new window, it would be more handy for me if the default is
 to open it in a new column. Because for most programs I use it is more
 important to see more of the vertical than the horizontal part (e.g. ls,
 editing text). If the default is to open a new row that would be a
 waste of place (even on my 12 screen). This was the reason for me not
 to use wmii.
 
 So I would like to know what the reason is for the default open in the
 same column in row is. (Perhaps I can improve my way of working :) .)
 
 Perhaps it is possible to have the option between those two ways?

Well what you describe was actually a reason why I disliked the
column layout as well. One could consider a global ncol value
which takes preference being reached, before columns are filled.

But would this really be worth the effort?

 And two more questions:
 - would there be no maximise / monocle? 

There should be some kind of maximise.

 - would there be no floating?

I'm not sure about a floating layout in this regard, but a
floating layer is a must, esp. for transients.

But all the thoughts might never be realized, I just want to
know what the community thinks.

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] A rather radical thought

2008-04-02 Thread Matthias-Christian Ott
Anselm R. Garbe [EMAIL PROTECTED] wrote:

 On Wed, Apr 02, 2008 at 04:27:34PM +0200, Matthias-Christian Ott wrote:
  Anselm R. Garbe [EMAIL PROTECTED] wrote:
  
   Here is what I have exactly in mind from a user perspective:
  
   Initially there is 1 column, a new client is inserted below the
   currently selected window in the column, similiar to the stack
   in tiled layout.
  
   Each window can be moved left- or rightwards (Mod1-Shift-{h,l})
   which may result in a new column if there is none.  A
   new column gets half the size of the current column.
  
   Each window can also be moved up- or downwards in the column
   itself (Mod1-Shift-{j,k}).
  
   Columns can be grown/shrinked using Mod1-Control-{h,l}.
   Windows can be grown/shrinked using Mod1-Control-{j,k}.
  
   The navigation is rather straight-forward:
  
   Focussing the previous/next window in the current column is done
   with Mod1-{j,k}, focussing the previous/next column is done with
   Mod1-{h,l}.
  
   So the whole layout concept consists of basically 4 keys with 3
   kinds of modifiers.
  
  But this wouldn't exclude the tag concept of dwm, it just means giving
  up this tiled and master area concept, or did I get it wrong?

 The tagging concept would stay, except the ability of viewing
 more than 1 tag at a time (though a specific window could be
 sticky because it contains all tags).

Is there a particular reason for this? Probably most people consider
viewing multiple tags as a necessary feature.

  Is your concept limited to horizontal columns or are vertical columns or
  both simultaneously possible?

 Theoretically one could do rows from columns, but in practice I
 believe that columns just work out very well.

For me columns won't work, because I can't display two 80x24 terminals
at the same time on my monitor. Rows are essential for me, because
otherwise I have to switch tags very often or use a floating layout.

 What won't be possible is having rows and columns mixed
 together -- at least I won't support this officially.

But this would actually make the layout I previously described possible.

Regards
Matthias-Christian



[dwm] A rather radical thought

2008-04-02 Thread Michael
Anselm R. Garbe wrote:
What about getting rid of supporting the selection of multiple
tags at a time

Please, don't do it, personaly, I use this feature very much.
Don't you think that eliminating it will approach dwm to workspace model?



Re: [dwm] A rather radical thought

2008-04-02 Thread Ross Mohn
On Wed, Apr 02, 2008 at 02:59:24PM +0200, Anselm R. Garbe wrote:
 What about getting rid of supporting the selection of multiple
 tags at a time and celebrating the revival of the column layout
 of wmii again in dwm 5.0? What about getting rid of all the tiled
 layout flavors in favor for the column layout?

Maybe. Are you going to publish dwm 4.9 shortly? It has some changes I
want, but I'd rather not grab the tip and wait for the release.
Thanks! -Ross




Re: [dwm] A rather radical thought

2008-04-02 Thread Antoni Grzymala
Marc Andre Tanner dixit (2008-04-02, 15:37):

 Anselm R. Garbe wrote:
 What about getting rid of supporting the selection of multiple
 tags at a time and celebrating the revival of the column layout
 of wmii again in dwm 5.0? What about getting rid of all the tiled
 layout flavors in favor for the column layout?

 Could we please stop rewriting dwm for the n-th time and focus
 on st instead!

Agreed. I reckon hg tip number 1122 was pretty much perfect. Everything
afterwards is pretty much a waste of time (with all due respect :)).

-- 
[a]


signature.asc
Description: Digital signature


Re: [dwm] A rather radical thought

2008-04-02 Thread John Grahor

I never display multiple tags at a time.  The idea, though very elegant,
doesn't work for me.  I usually work by adding terminals, etc to a
single tag until there are too many to be useful then throw one or more
over to another tag and keep going.

For me a column layout would be appealing, although if you aren't going
to support column modes, I don't see how it would be much different than
the htile layout.

On Wed, 2 Apr 2008, [EMAIL PROTECTED] wrote:

Date: Wed, 2 Apr 2008 14:59:24 +0200
From: Anselm R. Garbe [EMAIL PROTECTED]
Subject: [dwm] A rather radical thought
To: dwm@suckless.org
Message-ID: [EMAIL PROTECTED]
Content-Type: text/plain; charset=us-ascii

What about getting rid of supporting the selection of multiple
tags at a time and celebrating the revival of the column layout
of wmii again in dwm 5.0? What about getting rid of all the tiled
layout flavors in favor for the column layout?

Kind regards,
--
Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361




Re: [dwm] A rather radical thought

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 07:43:06PM +0400, Michael wrote:
 Anselm R. Garbe wrote:
 What about getting rid of supporting the selection of multiple
 tags at a time
 
 Please, don't do it, personaly, I use this feature very much.
 Don't you think that eliminating it will approach dwm to workspace model?

Well if it is not possible to view several tags at a time, two
tagall/dropall functions could be considered to apply/drop a certain
tag to all visible clients for convenience reasons.

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] dwm Digest, Vol 22, Issue 5

2008-04-02 Thread John Grahor

I really like this, though I seem to be in a minority.

Also looks like you need way more state to implement this.

I wonder, Anselm, if you looked at my optimal tile patch.  It adds
more rows and columns as needed, is fairly simple, and doesn't require
more state.

On Wed, 2 Apr 2008, [EMAIL PROTECTED] wrote:

Date: Wed, 2 Apr 2008 16:11:54 +0200
From: Anselm R. Garbe [EMAIL PROTECTED]
Subject: Re: [dwm] A rather radical thought
To: dwm@suckless.org
Message-ID: [EMAIL PROTECTED]
Content-Type: text/plain; charset=us-ascii

Here is what I have exactly in mind from a user perspective:

Initially there is 1 column, a new client is inserted below the
currently selected window in the column, similiar to the stack
in tiled layout.

Each window can be moved left- or rightwards (Mod1-Shift-{h,l})
which may result in a new column if there is none.  A
new column gets half the size of the current column.

Each window can also be moved up- or downwards in the column
itself (Mod1-Shift-{j,k}).

Columns can be grown/shrinked using Mod1-Control-{h,l}.
Windows can be grown/shrinked using Mod1-Control-{j,k}.

The navigation is rather straight-forward:

Focussing the previous/next window in the current column is done
with Mod1-{j,k}, focussing the previous/next column is done with
Mod1-{h,l}.

So the whole layout concept consists of basically 4 keys with 3
kinds of modifiers.

Kind regards,
--
Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361




Re: [dwm] A rather radical thought

2008-04-02 Thread Michael
On Wed, Apr 02, 2008 at 06:04:26PM +0200, Anselm R. Garbe wrote:
 Well if it is not possible to view several tags at a time, two
 tagall/dropall functions could be considered to apply/drop a certain
 tag to all visible clients for convenience reasons.

But then after switching to another tag, I would still see those window windows,
which i marked in previous view, current approach is good because of lacking
such behaviour.

Let me be more specific, I use this feature when I want mix firefox and OOO, I
just apply for the current view two tags: doc and www, which is very logical.
I don't want to apply doc tag for firefox window every time I need to look for
something in web page, it will require too many actions, and i think it's just
not correct: I want only *view* specific tags, I don't want apply any of them.

Dwm is excellent manager, it has simple and powerful system of views and tags,
and you want to elliminate one of the degree of freedom in it, please, don't do
it.



Re: [dwm] A rather radical thought

2008-04-02 Thread Richard Pöttler
On Wed, Apr 02, 2008 at 02:59:24PM +0200, Anselm R. Garbe wrote:
 What about getting rid of supporting the selection of multiple
 tags 

I never used it, but it is more because most of the time I don't think about it,
and simply apply the same tag to two windows and switch to this one, instead of
selecting two tags. It was the same with alt+tab, but since it made it's way
into my brain, and now  it's the combination I use most.

 at a time and celebrating the revival of the column layout
 of wmii again in dwm 5.0? What about getting rid of all the tiled
 layout flavors in favor for the column layout?

I like the two column layout of dwm and if I would like to have more, I would
use wmii. I would more prefer to be able to apply different column withs to
different tags.

bye
richi
-- 
quoting guide: http://www.xs4all.nl/~hanb/documents/quotingguide.html


pgpwzTMg4QGJs.pgp
Description: PGP signature


Re: [dwm] A rather radical thought

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 08:27:11PM +0400, Michael wrote:
 On Wed, Apr 02, 2008 at 06:04:26PM +0200, Anselm R. Garbe wrote:
  Well if it is not possible to view several tags at a time, two
  tagall/dropall functions could be considered to apply/drop a certain
  tag to all visible clients for convenience reasons.
 
 But then after switching to another tag, I would still see those window 
 windows,
 which i marked in previous view, current approach is good because of lacking
 such behaviour.
 
 Let me be more specific, I use this feature when I want mix firefox and OOO, I
 just apply for the current view two tags: doc and www, which is very logical.
 I don't want to apply doc tag for firefox window every time I need to look for
 something in web page, it will require too many actions, and i think it's just
 not correct: I want only *view* specific tags, I don't want apply any of them.
 
 Dwm is excellent manager, it has simple and powerful system of views and tags,
 and you want to elliminate one of the degree of freedom in it, please, don't 
 do
 it.

Well the reason I started this discussion was actually to just
discuss this idea. We did several times in the past, and always
we came to the conclusion that dwm is right already.

Though, I see a lot of potential to improve the basic layout
algorithm to something which is similiar capable to the wmii column
layout. 

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



[dwm] simplify grabbuttons()

2008-04-02 Thread Johannes Hofmann
Hello,

below is a patch that shortens grabbuttons() a bit.

Cheers,
Johannes

PS: I really appreciate the attitude to reconsider everything from
time to time - even though I'm quite satisfied with current dwm.


diff --git a/dwm.c b/dwm.c
--- a/dwm.c
+++ b/dwm.c
@@ -805,36 +805,18 @@
 
 void
 grabbuttons(Client *c, Bool focused) {
+   unsigned int Buttons[] = {Button1, Button2, Button3};
+   unsigned int Modifiers[] = {MODKEY, MODKEY|LockMask,
+   MODKEY|numlockmask, MODKEY|numlockmask|LockMask};
+   int i, j;   
+ 
XUngrabButton(dpy, AnyButton, AnyModifier, c-win);
 
-   if(focused) {
-   XGrabButton(dpy, Button1, MODKEY, c-win, False, BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button1, MODKEY|LockMask, c-win, False, 
BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button1, MODKEY|numlockmask, c-win, False, 
BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button1, MODKEY|numlockmask|LockMask, c-win, 
False, BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-
-   XGrabButton(dpy, Button2, MODKEY, c-win, False, BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button2, MODKEY|LockMask, c-win, False, 
BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button2, MODKEY|numlockmask, c-win, False, 
BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button2, MODKEY|numlockmask|LockMask, c-win, 
False, BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-
-   XGrabButton(dpy, Button3, MODKEY, c-win, False, BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button3, MODKEY|LockMask, c-win, False, 
BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button3, MODKEY|numlockmask, c-win, False, 
BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   XGrabButton(dpy, Button3, MODKEY|numlockmask|LockMask, c-win, 
False, BUTTONMASK,
-   GrabModeAsync, GrabModeSync, None, None);
-   }
+   if(focused)
+   for (i = 0; i  LENGTH(Buttons); i++)
+   for (j = 0; j  LENGTH(Modifiers); j++)
+   XGrabButton(dpy, Buttons[i], Modifiers[j], 
c-win, False,
+   BUTTONMASK, GrabModeAsync, 
GrabModeSync, None, None);
else
XGrabButton(dpy, AnyButton, AnyModifier, c-win, False, 
BUTTONMASK,
GrabModeAsync, GrabModeSync, None, None);



Re: [dwm] A rather radical thought

2008-04-02 Thread Kurt H Maier
On Wed, Apr 02, 2008 at 02:59:24PM +0200, Anselm R. Garbe wrote:

 What about getting rid of supporting the selection of multiple tags

In my estimation, without multiple tagging, there's no reason to use dwm

-- 
# Kurt H Maier



Re: [dwm] A rather radical thought

2008-04-02 Thread voltaic
I agree. For me, the major distinction between DWM and the other
window managers has been the use of tags instead of workspaces.

I also don't like the idea of moving windows around columns. I think
the zoom function is a sufficient mode of manipulating windows
(between master/slave areas). I feel like the navigation within the
columns would make the window manager less dynamic, as the user would
have to (or feel obligated to) keep rearranging windows.


On 4/2/08, Kurt H Maier [EMAIL PROTECTED] wrote:
 On Wed, Apr 02, 2008 at 02:59:24PM +0200, Anselm R. Garbe wrote:

   What about getting rid of supporting the selection of multiple tags


 In my estimation, without multiple tagging, there's no reason to use dwm

  --

 # Kurt H Maier





Re: [dwm] simplify grabbuttons()

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 07:01:11PM +0200, Johannes Hofmann wrote:
 below is a patch that shortens grabbuttons() a bit.

Of course this will make it to hg tip ;)

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



[dwm] there won't be a radical change

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 12:15:21PM -0400, John Grahor wrote:
 I really like this, though I seem to be in a minority.
 
 Also looks like you need way more state to implement this.
 
 I wonder, Anselm, if you looked at my optimal tile patch.  It adds
 more rows and columns as needed, is fairly simple, and doesn't require
 more state.

Well, I think it might go into this direction. I agree on most
people claiming to not do too major changes. So see this as a
statement to not expect a radical change ;)

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] A rather radical thought

2008-04-02 Thread andrew
On Wed, Apr 2, 2008 at 10:11 AM, Anselm R. Garbe [EMAIL PROTECTED] wrote:
 Here is what I have exactly in mind from a user perspective:
 ... snip several command explanations ...
  So the whole layout concept consists of basically 4 keys with 3
  kinds of modifiers.

4x3 = 12 (TWELVE!) combos to remember/learn to manage windows?
Currently I have (i think) only three: mwresize (left and right) and zoom.

I would reiterate that the whole appeal of dwm is to manage my windows
in a predictable way and get out of my way. All this capability to
move windows around is exactly why i abandoned wmii (not to mention
non-tiling WMs) in the first place, wasting to much time organizing
windows just so.

And I would like to echo that the ability to mix and match what tags I
am viewing is the real dynamic backbone in dwm. I separate
applications by tag and then view the tags I need for the job at hand,
whether thats a text editor and www for reference, web and chat on the
side, I pull in email periodically to check it and then dismiss it.
Without the ability to quickly bring in windows not in the current
view, whats the point of tags?

Regards,
-Andrew



Re: [dwm] A rather radical thought

2008-04-02 Thread Damjan Vrencur
On Wednesday 02 of April 2008 19:47:30 voltaic wrote:
 I agree. For me, the major distinction between DWM and the other
 window managers has been the use of tags instead of workspaces.

 I also don't like the idea of moving windows around columns. I think
 the zoom function is a sufficient mode of manipulating windows
 (between master/slave areas). I feel like the navigation within the
 columns would make the window manager less dynamic, as the user would
 have to (or feel obligated to) keep rearranging windows.

 On 4/2/08, Kurt H Maier [EMAIL PROTECTED] wrote:
  On Wed, Apr 02, 2008 at 02:59:24PM +0200, Anselm R. Garbe wrote:
What about getting rid of supporting the selection of multiple tags
 
  In my estimation, without multiple tagging, there's no reason to use dwm
 
   --
 
  # Kurt H Maier

Agree.
I use viewing of multiple tags regularly in my workflow and would miss it 
(too?) much.

-- 
 Damjan Vrenčur ~ http://lmmri.fri.uni-lj.si/damjan/ ~ GPG key: C6A3146F



[dwm] again, don't expect the change I discussed about happen

2008-04-02 Thread Anselm R. Garbe
See subject, this is related to the column layout discussion ;)

-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] experimental patch: escher

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 11:42:57AM -0700, Michael Elkins wrote:
 More refactoring:
 
 - remove boilerplate code by creating a macro to generate layout
   functions
 - replace tileh/tilev with the version that do nmaster
 
 It's interesting now that removing nmaster support doesn't really buy
 that much: just one function (setnmaster) and one if-statement.  LOC
 counts:
 
 dwm stock:
 $ wc -l dwm.c config.def.h
  1952 dwm.c
   102 config.def.h
  2054 total
 
 patched:
 $ wc -l dwm.c config.def.h dwindle.c 
  1884 dwm.c
   109 config.def.h
78 dwindle.c
  2071 total
 
 +17 LOC for nmaster and two spiral modes isn't shabby.
 
 me

Well that's a good result, but I will postpone a decision here.
I plan to refactor the layout stuff a little bit after dwm 4.9,
there is obviously potential to simplify the algorithms.

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] setmfact improvements

2008-04-02 Thread Anselm R. Garbe
On Wed, Apr 02, 2008 at 04:23:40PM +0200, Peter Hartlich wrote:
 Attached are two small patches to setmfact. The first one removes some
 redundancy, saving 5 LOC; the second one reintroduces the MFACT define
 that was present (as MWFACT) before, so that you can do setmfact(NULL)
 to restore your initial master factor.
 
 Taken together, they add 1 LOC.

Thanks, I applied the second one which re-introduces MFACT.

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



[dwm] final testing call for dwm-4.9

2008-04-02 Thread Anselm R. Garbe
Hi there,

   I kindly ask you to give hg tip a try. If there are no bugs,
   it will be dwm-4.9 tomorrow evening.

Kind regards,
-- 
 Anselm R. Garbe  http://www.suckless.org/  GPG key: 0D73F361



Re: [dwm] ntile layout for dwm hg tip

2008-04-02 Thread Nibble
 You could simplify the code by introducing to two global variables that
 indicate whether the master or tiled area is horizontal.
 So switching to |= would simply set tileh = True and leave masterh
 unchanged and switchting to -|= would set tileh = True and masterh = True.
 So you could simple toggle the behaviour (e.g. MOD-v - masterh = False,
 MOD-Shift-v - tileh = False) and don't need four layouts.

I was thinking over that solution, but I feel that setting four different 
layouts goes more in accordance with the dwm philosophy (eg. tileh and 
tilev). Furthermore, people usually need only one of the nmaster layouts, so 
they can safely remove the other ones.

Kind regards,
Nibble