From ales.guzik at gmail.com  Mon Oct  1 13:02:41 2012
From: ales.guzik at gmail.com (Ales Guzik)
Date: Mon, 1 Oct 2012 16:02:41 +0300
Subject: [clfswm-devel] Packaged CLFSWM
Message-ID: <CADT2y-147T7Opz-k6+uZiv2HJUJ_owRZNAvn9RcjZWNwr1Mznw@mail.gmail.com>

Hi,

Are there CLFSWM packages for debian/ubuntu/redhat/suse available in
official distro repositories? Is it available in ubuntu software
center?
Basically, I think we need to improve clfswm availability to end-users.

-- 
 Ales Guzik <me at aguzik.net>
 skype: ales-guzik
 tel: +375 29 5751103



From pbrochard at common-lisp.net  Mon Oct  1 13:53:30 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Mon, 01 Oct 2012 15:53:30 +0200
Subject: [clfswm-devel] Packaged CLFSWM
In-Reply-To: <CADT2y-147T7Opz-k6+uZiv2HJUJ_owRZNAvn9RcjZWNwr1Mznw@mail.gmail.com>
	(Ales Guzik's message of "Mon, 1 Oct 2012 16:02:41 +0300")
References: <CADT2y-147T7Opz-k6+uZiv2HJUJ_owRZNAvn9RcjZWNwr1Mznw@mail.gmail.com>
Message-ID: <87d312uwmd.fsf@common-lisp.net>

Ales Guzik writes:

> Hi,
>
Hi Ales,

> Are there CLFSWM packages for debian/ubuntu/redhat/suse available in
> official distro repositories? Is it available in ubuntu software
> center?
>
Desmond O. Chang is the maintainer for the Debian package. I think
CLFSWM go automatically in the Ubuntu repository.
Xavier Maillard has made a PKGBUILD for Slackware (but I have had no
news for a while). There is one for Gentoo too.
And I maintain the one for AUR in Archlinux (I need to reinstall it to
make the update).

I don't know for Redhat, Suse and others.

In the past, I've build a Linux only binary (based on clisp or
sbcl). But I think it's more easy  for end-user to use the distro
package system. And I think developers knows or can learn how to build
CLFSWM.


> Basically, I think we need to improve clfswm availability to end-users.
>
I think so. But as with Wikipedia: alone, I'm judge and jury.

Another thing, I used to announce CLFSWM news in a  french Linux forum
(linuxfr.org). It has some visibility but it's in french.
Where do you think we can share CLFSWM news? Some forum talks about it
as something to unearth[1] :-)
Maybe on usenet c.l.l?

Regards,

Philippe

[1] http://crunchbanglinux.org/forums/topic/19404/clfswm-unearthing-another-hidden-gem/




From spacefrogg-clfswm at meterriblecrew.net  Sun Oct  7 18:46:45 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Sun, 07 Oct 2012 20:46:45 +0200
Subject: [clfswm-devel] Bug, in case no XINERAMA extension is present
Message-ID: <1349635048-sup-6984@leandros>

Hi,

in case no XINERAMA extension is present the sizes list
in (place-frames-from-xinerama-infos) is empty and
the function fails to establish the root frame.

I fixed it and request you to pull from my repository on github.

git://github.com/spacefrogg/clfswm.git
Branch: 'spacefrogg'
Commit: 980487d4d70ad6ede5c921e8614a62afa55e3b52

In general, how does development process work?
Are pull request via this list okay? 


Michael




From pbrochard at common-lisp.net  Sun Oct  7 20:02:20 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sun, 07 Oct 2012 22:02:20 +0200
Subject: [clfswm-devel] Bug, in case no XINERAMA extension is present
In-Reply-To: <1349635048-sup-6984@leandros> (Michael Raitza's message of "Sun, 
	07 Oct 2012 20:46:45 +0200")
References: <1349635048-sup-6984@leandros>
Message-ID: <87sj9qxd83.fsf@common-lisp.net>

Michael Raitza writes:

> Hi,
>
Hi, welcome on this list!

> in case no XINERAMA extension is present the sizes list
> in (place-frames-from-xinerama-infos) is empty and
> the function fails to establish the root frame.
>
Thanks a lot, I've seen this with a FreeBSD test. I'm fixing it another
way because if someone want to use another default-init-hook (ie:
without requesting xinerama infos from place-frames-from-xinerama-infos),
the root structure must be the same as when detected by this function.

> I fixed it and request you to pull from my repository on github.
>
> git://github.com/spacefrogg/clfswm.git
> Branch: 'spacefrogg'
> Commit: 980487d4d70ad6ede5c921e8614a62afa55e3b52
>
Many thanks for your patch and your quick reaction!!


> In general, how does development process work?
>
As you've done!

> Are pull request via this list okay? 
>
Yes perfectly okay.

Thanks again,

Regards,

Philippe



From z_axis at 163.com  Mon Oct  8 05:48:04 2012
From: z_axis at 163.com (z_axis)
Date: Mon, 08 Oct 2012 13:48:04 +0800
Subject: [clfswm-devel] How to use swank using clisp ?
Message-ID: <op.wluiefvc1uofcn@redirector.netnames.net>

In CCL+stumpwm, i use `(load  "~/.vim/slime/start-swank.lisp")` to start  
swank server, it works great.  But clisp + stumpwm doesnot work.

Does anybody here use swank in clisp+clfswm successfully ?


Sincerely!




From spacefrogg-clfswm at meterriblecrew.net  Mon Oct  8 17:02:45 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Mon, 08 Oct 2012 19:02:45 +0200
Subject: [clfswm-devel] Fixes for reorder-* and frame-select-*-child
Message-ID: <1349714802-sup-5699@leandros>

Hi,

I fixed reorder-brother and reorder-child. They missed some frames when
rotating in positive direction. The new versions are even simpler.
I also improved n-rotate-list.

I will attach patches as commit hashes won't be of any use. I am constantly
rebasing my branch to follow the development.

As a third patch, I changed the semantics of the frame-select-*-child
functions to wrap around the end of the list of frames. (Seems logical
for me.) Maybe we should introduce an optional argument for that, or
have separate functions.

Regards,


Michael
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patches.diff
Type: application/octet-stream
Size: 3520 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121008/3e77ac20/attachment.obj>

From pbrochard at common-lisp.net  Mon Oct  8 20:13:58 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Mon, 08 Oct 2012 22:13:58 +0200
Subject: [clfswm-devel] Bug, in case no XINERAMA extension is present
In-Reply-To: <mailman.617.1349713780.2246.clfswm-devel@common-lisp.net>
	(clfswm-devel-bounces@common-lisp.net's message of "Mon, 08 Oct 2012
	09:29:40 -0700")
References: <mailman.617.1349713780.2246.clfswm-devel@common-lisp.net>
Message-ID: <87y5jgoh6h.fsf@common-lisp.net>

clfswm-devel-bounces at common-lisp.net writes:

> The attached message has been automatically discarded.
>
> From: Michael Raitza <m.raitza at gmx.net>
> Subject: Re: [clfswm-devel] Bug, in case no XINERAMA extension is present
> To: clfswm-devel <clfswm-devel at common-lisp.net>
> Date: Mon, 08 Oct 2012 18:34:04 +0200
> User-Agent: Sup/git
>
> Hi,
>
>> Thanks a lot, I've seen this with a FreeBSD test. I'm fixing it another
>> way because if someone want to use another default-init-hook (ie:
>> without requesting xinerama infos from place-frames-from-xinerama-infos),
>> the root structure must be the same as when detected by this function.
>> 
> this I don't understand. In the case someone uses
> place-frames-from-xinerama-infos with XINERAMA disabled (as I did) this function
> just breaks directly, because its input values do not depend on anything
> else but the requested parameters from xinerama.
>
> Is there a fix for that already?
>
Please, try with the last commit (bbada27). It works here with XINERAMA
active or not. Does it crash (and where) for you?

The important part (I think) is src/clfswm.lisp default-init-hook.
If there is some XINERAMA infos, setup the screen from that. Else ensure
that there is at least one root and build the default structure.

If someone wants to use a different init-hook not based on XINERAMA, It
just has to respect the root structure as in ensure-at-least-one-root.


> Regards,
>
> Michael
>
Regards,

Philippe



From pbrochard at common-lisp.net  Mon Oct  8 21:09:24 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Mon, 08 Oct 2012 23:09:24 +0200
Subject: [clfswm-devel] Fixes for reorder-* and frame-select-*-child
In-Reply-To: <1349714802-sup-5699@leandros> (Michael Raitza's message of "Mon, 
	08 Oct 2012 19:02:45 +0200")
References: <1349714802-sup-5699@leandros>
Message-ID: <87txu4oem3.fsf@common-lisp.net>

Michael Raitza writes:

> Hi,
>
Hi,

> I fixed reorder-brother and reorder-child. They missed some frames when
> rotating in positive direction. The new versions are even simpler.
> I also improved n-rotate-list.
>
> I will attach patches as commit hashes won't be of any use. I am constantly
> rebasing my branch to follow the development.
>
> As a third patch, I changed the semantics of the frame-select-*-child
> functions to wrap around the end of the list of frames. (Seems logical
> for me.) Maybe we should introduce an optional argument for that, or
> have separate functions.
>
Hey!! I can't apply your patch! you broke an important CLFSWM feature:
when you rotate children in a frame and release the alt-tab key, the
next time the previous child will be the next child.

 A B C D E F -> rotate 3 times -> D A B C E F

Pressing only one time the Alt-Tab key enable to switch between two
children. See how other (stacking) window managers works.

I know it's complicated (I've had an headache on it) but it's very
important (at least for me).
Try pressing Alt-tab multiple times. What you want is
select-next-child-simple (see src/clfswm-circulate-mode.lisp). 

BTW, I like your n-rotate-list implementation (faster and less memory
expansive). Commited in your name. Many thanks.

> Regards,
>
Regards,

Philippe



From pbrochard at common-lisp.net  Mon Oct  8 21:15:03 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Mon, 08 Oct 2012 23:15:03 +0200
Subject: [clfswm-devel] How to use swank using clisp ?
In-Reply-To: <op.wluiefvc1uofcn@redirector.netnames.net> (z. axis's message of
	"Mon, 08 Oct 2012 13:48:04 +0800")
References: <op.wluiefvc1uofcn@redirector.netnames.net>
Message-ID: <87pq4soeco.fsf@common-lisp.net>

z. axis writes:

> In CCL+stumpwm, i use `(load  "~/.vim/slime/start-swank.lisp")` to
> start swank server, it works great.  But clisp + stumpwm doesnot work.
>
> Does anybody here use swank in clisp+clfswm successfully ?
>
Yes, but with Emacs. You seem to use a non threaded clisp version. I
think clisp hang until Emacs (or another editor) come in.

> Sincerely!
>
Regards,

Philippe

PS: Here it's the CLFSWM list :-)


> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel



From spacefrogg-clfswm at meterriblecrew.net  Tue Oct  9 19:09:51 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Tue, 09 Oct 2012 21:09:51 +0200
Subject: [clfswm-devel] Fixes for reorder-* and frame-select-*-child
In-Reply-To: <87txu4oem3.fsf@common-lisp.net>
References: <1349714802-sup-5699@leandros> <87txu4oem3.fsf@common-lisp.net>
Message-ID: <1349809116-sup-2951@leandros>

Sorry,

I was totally misguided in these cases. I missed the fact that the
functions are to be used in circulate mode. I used them directly.

Nevertheless I added a select-previous-child-simple function. It was
missing somehow. :) Patch attached.

One more thing. Concerning the circulation I have noticed that, when
changing root frames, the message box appears on the frame that was just
left and not on the entered frame. It seems to be due to the fact that
the window and the garbage collector are allocated when entering the
circulate mode. I had no good idea to fix it but to rewrite that stuff.

Regards,


Michael
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patches.diff
Type: application/octet-stream
Size: 571 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121009/be720820/attachment.obj>

From spacefrogg-clfswm at meterriblecrew.net  Thu Oct 11 20:04:52 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Thu, 11 Oct 2012 22:04:52 +0200
Subject: [clfswm-devel] Some patches
Message-ID: <1349985159-sup-5215@leandros>

Hi,

I prepared a series of patches mostly improvements.

The first patch adds select-previous-child-simple, this function was
missing somehow.

The second patch modifies the frame-select-*-child functions to wrap
around the end of the list of children. Isn't that the intended
behaviour? I found it much more pleasing this way.

The third patch adds set-layout-simple. The other set-layout* functions
were to be called from second mode if I understood correctly. I want to
do it from main mode. Is there a general way to get rid of these second
mode / main mode differences?

The fourth patch adds query-backspace-clear. To clear query input to the
beginning of the line. (I have bound it to C-u like in vim and some
shells.)

In the fifth patch I have refactored run-program-from-query-string. This
function could not be run from main mode because it binds the ? the
function list to be run when leaving the second mode. So it was never
called before entering and leaving second mode somewhen later. This
patch introduces the function run-program-from-query-string-fun that
does the actual query and generates the function.
run-program-from-query-string uses its output. It also introduces
run-program-from-query-string-main to be called from main mode.

The third and fifth patch seem to be related, because they are causes by
the differences of the functions called in second mode from functions
providing the same functionality in main mode. Can't we consolidate
these two concepts?

Regards,


Michael
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Added-select-previous-child-simple.patch
Type: application/octet-stream
Size: 972 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121011/a5b8c0d7/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-Changed-frame-select-child-to-wrap-around.patch
Type: application/octet-stream
Size: 1311 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121011/a5b8c0d7/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0003-Added-set-layout-simple.patch
Type: application/octet-stream
Size: 845 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121011/a5b8c0d7/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0004-Added-query-backspace-clear.patch
Type: application/octet-stream
Size: 1022 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121011/a5b8c0d7/attachment-0003.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0005-Refactored-run-program-from-query-string.patch
Type: application/octet-stream
Size: 1770 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121011/a5b8c0d7/attachment-0004.obj>

From pbrochard at common-lisp.net  Thu Oct 11 20:25:30 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Thu, 11 Oct 2012 22:25:30 +0200
Subject: [clfswm-devel] Some patches
In-Reply-To: <1349985159-sup-5215@leandros> (Michael Raitza's message of "Thu, 
	11 Oct 2012 22:04:52 +0200")
References: <1349985159-sup-5215@leandros>
Message-ID: <87txu03ged.fsf@common-lisp.net>

Hi Michael,

I haven't time now. I'll take a look at your patch this week-end.
But seems ok at first look.

Regards,

Philippe

Michael Raitza writes:

> Hi,
>
> I prepared a series of patches mostly improvements.
>
> The first patch adds select-previous-child-simple, this function was
> missing somehow.
>
> The second patch modifies the frame-select-*-child functions to wrap
> around the end of the list of children. Isn't that the intended
> behaviour? I found it much more pleasing this way.
>
> The third patch adds set-layout-simple. The other set-layout* functions
> were to be called from second mode if I understood correctly. I want to
> do it from main mode. Is there a general way to get rid of these second
> mode / main mode differences?
>
> The fourth patch adds query-backspace-clear. To clear query input to the
> beginning of the line. (I have bound it to C-u like in vim and some
> shells.)
>
> In the fifth patch I have refactored run-program-from-query-string. This
> function could not be run from main mode because it binds the ? the
> function list to be run when leaving the second mode. So it was never
> called before entering and leaving second mode somewhen later. This
> patch introduces the function run-program-from-query-string-fun that
> does the actual query and generates the function.
> run-program-from-query-string uses its output. It also introduces
> run-program-from-query-string-main to be called from main mode.
>
> The third and fifth patch seem to be related, because they are causes by
> the differences of the functions called in second mode from functions
> providing the same functionality in main mode. Can't we consolidate
> these two concepts?
>
> Regards,
>
>
> Michael
>
>
>
>
>
> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel



From spacefrogg-clfswm at meterriblecrew.net  Thu Oct 11 22:44:30 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Fri, 12 Oct 2012 00:44:30 +0200
Subject: [clfswm-devel] New query input completion
Message-ID: <1349995252-sup-5333@leandros>

Hi again,

I've written another completion function that not only partially
completes the input but cycles through the list of possible completions
and fully completes each entry in turn.

Regards,


Michael
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Added-query-mode-complete-suggest-and-helpers.patch
Type: application/octet-stream
Size: 3545 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121012/7fdb8613/attachment.obj>

From michael at cadilhac.name  Thu Oct 11 23:30:12 2012
From: michael at cadilhac.name (=?ISO-8859-1?Q?Micha=EBl_Cadilhac?=)
Date: Thu, 11 Oct 2012 19:30:12 -0400
Subject: [clfswm-devel] New query input completion
In-Reply-To: <1349995252-sup-5333@leandros>
References: <1349995252-sup-5333@leandros>
Message-ID: <CADt3fpOwj8q8+h7dyScnzOuTfecbcQVoVuaDrhffsva0qQ7QNg@mail.gmail.com>

On Thu, Oct 11, 2012 at 6:44 PM, Michael Raitza
<spacefrogg-clfswm at meterriblecrew.net> wrote:
> Hi again,
>
> I've written another completion function that not only partially
> completes the input but cycles through the list of possible completions
> and fully completes each entry in turn.

Michael,

Just FYI, this is much appreciated by the prospective users; I'm one of them.
I've only tried CLFSWM, but it lacked --- apart from stability --- such
features.  I've been waiting to see them implemented, so, congrats :-) The
more appealing CLFSWM is in a first place, the more people will contribute;
a virtuous circle for sure!

M.



From z_axis at 163.com  Fri Oct 12 05:45:11 2012
From: z_axis at 163.com (z_axis)
Date: Fri, 12 Oct 2012 13:45:11 +0800
Subject: [clfswm-devel] Can we use complete REPL in clfswm directly ?
Message-ID: <op.wl1wxlko1uofcn@redirector.netnames.net>

The standard way is use slime to connect to clfswm.  However, it's  
sometimes convenient for us to use WM built in CL directly.
In stumpwm, we press <Ctrl+t, :> to do evaluation which has many  
limitation. It is not a complete REPL.


Sincerely!




From pbrochard at common-lisp.net  Fri Oct 12 21:34:34 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Fri, 12 Oct 2012 23:34:34 +0200
Subject: [clfswm-devel] Some patches
In-Reply-To: <1349985159-sup-5215@leandros> (Michael Raitza's message of "Thu, 
	11 Oct 2012 22:04:52 +0200")
References: <1349985159-sup-5215@leandros>
Message-ID: <87wqyvwf11.fsf@common-lisp.net>

Michael Raitza writes:

> Hi,
>
Hi,

> I prepared a series of patches mostly improvements.
>
Thanks a lot. I've begin to apply some of them. I'll look others later.

> The first patch adds select-previous-child-simple, this function was
> missing somehow.
>
Commited

> The second patch modifies the frame-select-*-child functions to wrap
> around the end of the list of children. Isn't that the intended
> behaviour? I found it much more pleasing this way.
>
Commited. But this one break CLFSWM when there is no children in the
current frame (fixed).

> The third patch adds set-layout-simple. The other set-layout* functions
> were to be called from second mode if I understood correctly. I want to
> do it from main mode. Is there a general way to get rid of these second
> mode / main mode differences?
>
Commited but how do you use it?

> The fourth patch adds query-backspace-clear. To clear query input to the
> beginning of the line. (I have bound it to C-u like in vim and some
> shells.)
>
Commited. I've bound it to C-u and S-C-Backspace.

> In the fifth patch I have refactored run-program-from-query-string. This
> function could not be run from main mode because it binds the ? the
> function list to be run when leaving the second mode. So it was never
> called before entering and leaving second mode somewhen later. This
> patch introduces the function run-program-from-query-string-fun that
> does the actual query and generates the function.
> run-program-from-query-string uses its output. It also introduces
> run-program-from-query-string-main to be called from main mode.
>
I'll take a look at this one tomorrow.

> The third and fifth patch seem to be related, because they are causes by
> the differences of the functions called in second mode from functions
> providing the same functionality in main mode. Can't we consolidate
> these two concepts?
>
Same thing.


Many thanks for your patch!

While your are motivated to go deeply in the CLFSWM source code. There
is a thing CLFSWM will always lake if nobody other than me have a look
at it: CLFSWM communicate poorly with the outside: it is not fully netwm
compliant. You can have a look at src/netwm-util.lisp and at some
functions like netwm-update-desktop-property. 
I think there may be a time when users will want to use CLFSWM with
Gnome or KDE. netwm-util.lisp is here for this but this is really not a
priority for me. 

> Regards,
>
Regards,

Philippe



From pbrochard at common-lisp.net  Fri Oct 12 21:45:16 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Fri, 12 Oct 2012 23:45:16 +0200
Subject: [clfswm-devel] New query input completion
In-Reply-To: <CADt3fpOwj8q8+h7dyScnzOuTfecbcQVoVuaDrhffsva0qQ7QNg@mail.gmail.com>
	(=?iso-8859-1?Q?=22Micha=EBl?= Cadilhac"'s message of "Thu, 11 Oct 2012
	19:30:12 -0400")
References: <1349995252-sup-5333@leandros>
	<CADt3fpOwj8q8+h7dyScnzOuTfecbcQVoVuaDrhffsva0qQ7QNg@mail.gmail.com>
Message-ID: <87sj9jwej7.fsf@common-lisp.net>

Micha?l Cadilhac writes:

> On Thu, Oct 11, 2012 at 6:44 PM, Michael Raitza
> <spacefrogg-clfswm at meterriblecrew.net> wrote:
>> Hi again,
>>
>> I've written another completion function that not only partially
>> completes the input but cycles through the list of possible completions
>> and fully completes each entry in turn.
>
> Michael,
>
> Just FYI, this is much appreciated by the prospective users; I'm one of them.
> I've only tried CLFSWM, but it lacked --- apart from stability --- such
> features.  I've been waiting to see them implemented, so, congrats :-) The
> more appealing CLFSWM is in a first place, the more people will contribute;
> a virtuous circle for sure!
>
Yes, many thanks to him. And I hope in the virtuous circle.
But please, stop whining about stability and start your involvement in
CLFSWM with backtrace/bug report/patch...

Since your last remark about stability, I've spotted 2 cases which causes
CLFSWM freeze (one with the new version of Gimp and one with the
Virtualbox GUI). I've fixed them as soon I've identified them. 

So: Where is your *feedback*? Does CLFSWM always freeze for you?

There is still a bug with the new code for root geometry change when we
plug a new monitor. Can you spot it? Indeed, it'll be fixed at the end
of the week-end.
But we really need feedback and bug reports for other problems...

Regards,

Philippe



From pbrochard at common-lisp.net  Fri Oct 12 21:51:45 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Fri, 12 Oct 2012 23:51:45 +0200
Subject: [clfswm-devel] Can we use complete REPL in clfswm directly ?
In-Reply-To: <op.wl1wxlko1uofcn@redirector.netnames.net> (z. axis's message of
	"Fri, 12 Oct 2012 13:45:11 +0800")
References: <op.wl1wxlko1uofcn@redirector.netnames.net>
Message-ID: <87obk7we8e.fsf@common-lisp.net>

z. axis writes:

> The standard way is use slime to connect to clfswm.  However, it's
> sometimes convenient for us to use WM built in CL directly.
> In stumpwm, we press <Ctrl+t, :> to do evaluation which has many
> limitation. It is not a complete REPL.
>
In CLFSWM it's <M-t :>. This REPL has symbol completion and key binding
is nearly the same as a shell. And it's improved by Michael (patch
applied soon).

What do you lack in the CLFSWM REPL?

BTW there is also a server/client method in contrib. But I'd like to
implement one by X property change. This let the user with a REPL and
script communication not blocked by Xlib event timeout.

Cheers,

Philippe



From z_axis at 163.com  Sat Oct 13 00:19:28 2012
From: z_axis at 163.com (z_axis)
Date: Sat, 13 Oct 2012 08:19:28 +0800
Subject: [clfswm-devel] Can we use complete REPL in clfswm directly ?
In-Reply-To: <87obk7we8e.fsf@common-lisp.net>
References: <op.wl1wxlko1uofcn@redirector.netnames.net>
	<87obk7we8e.fsf@common-lisp.net>
Message-ID: <op.wl3cirnj1uofcn@redirector.netnames.net>

Can CLFSWM REPl output multiple line result ?

(defun div618 (p1 p2)
     (let ((ratio '(0. 0.191 0.236 0.382 0.5 0.618 0.809 1.))
           (price #'(lambda (r) (if (<= p1 p2) (+ p1 (* (- p2 p1) r)) (- p1  
(* (- p1 p2) r))))))
           (if (<= p1 p2)
             (dolist (r (reverse ratio)) (format t "-------~3$    
~$-------~%" r (funcall price r)))
             (dolist (r ratio) (format t "-------~3$  ~$-------~%" r  
(funcall price r))))))

The stumpwm REPL cannot output the result because div618 just return nil.


? Sat, 13 Oct 2012 05:51:45 +0800?Philippe Brochard  
<pbrochard at common-lisp.net> ??:

> What do you lack in the CLFSWM REPL?




From pbrochard at common-lisp.net  Sat Oct 13 06:06:00 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 13 Oct 2012 08:06:00 +0200
Subject: [clfswm-devel] Can we use complete REPL in clfswm directly ?
In-Reply-To: <op.wl3cirnj1uofcn@redirector.netnames.net> (z. axis's message of
	"Sat, 13 Oct 2012 08:19:28 +0800")
References: <op.wl1wxlko1uofcn@redirector.netnames.net>
	<87obk7we8e.fsf@common-lisp.net>
	<op.wl3cirnj1uofcn@redirector.netnames.net>
Message-ID: <87fw5ikit3.fsf@common-lisp.net>

z. axis writes:

> Can CLFSWM REPl output multiple line result ?
>
> (defun div618 (p1 p2)
>     (let ((ratio '(0. 0.191 0.236 0.382 0.5 0.618 0.809 1.))
>           (price #'(lambda (r) (if (<= p1 p2) (+ p1 (* (- p2 p1) r))
> (- p1 (* (- p1 p2) r))))))
>           (if (<= p1 p2)
>             (dolist (r (reverse ratio)) (format t "-------~3$
> ~$-------~%" r (funcall price r)))
>             (dolist (r ratio) (format t "-------~3$  ~$-------~%" r
> (funcall price r))))))
>
> The stumpwm REPL cannot output the result because div618 just return nil.
>
Is this ok for you:

  http://common-lisp.net/project/clfswm/z_axis-REPL-screen.png

(You can have tested by yourself also...)

>
> ? Sat, 13 Oct 2012 05:51:45 +0800?Philippe Brochard
> <pbrochard at common-lisp.net> ??:
>
>> What do you lack in the CLFSWM REPL?
>
>
> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel



From z_axis at 163.com  Sat Oct 13 08:06:16 2012
From: z_axis at 163.com (z_axis)
Date: Sat, 13 Oct 2012 16:06:16 +0800
Subject: [clfswm-devel] Can we use complete REPL in clfswm directly ?
In-Reply-To: <87fw5ikit3.fsf@common-lisp.net>
References: <op.wl1wxlko1uofcn@redirector.netnames.net>
	<87obk7we8e.fsf@common-lisp.net>
	<op.wl3cirnj1uofcn@redirector.netnames.net>
	<87fw5ikit3.fsf@common-lisp.net>
Message-ID: <op.wl3x4rfb1uofcn@redirector.netnames.net>

That's great!  I have just tested it .  I donot understand why stumpwm  
cannot do it ?

? Sat, 13 Oct 2012 14:06:00 +0800?Philippe Brochard  
<pbrochard at common-lisp.net> ??:

> z. axis writes:
>
>> Can CLFSWM REPl output multiple line result ?
>>
>> (defun div618 (p1 p2)
>>     (let ((ratio '(0. 0.191 0.236 0.382 0.5 0.618 0.809 1.))
>>           (price #'(lambda (r) (if (<= p1 p2) (+ p1 (* (- p2 p1) r))
>> (- p1 (* (- p1 p2) r))))))
>>           (if (<= p1 p2)
>>             (dolist (r (reverse ratio)) (format t "-------~3$
>> ~$-------~%" r (funcall price r)))
>>             (dolist (r ratio) (format t "-------~3$  ~$-------~%" r
>> (funcall price r))))))
>>
>> The stumpwm REPL cannot output the result because div618 just return  
>> nil.
>>
> Is this ok for you:
>
>   http://common-lisp.net/project/clfswm/z_axis-REPL-screen.png
>
> (You can have tested by yourself also...)
>
>>
>> ? Sat, 13 Oct 2012 05:51:45 +0800?Philippe Brochard
>> <pbrochard at common-lisp.net> ??:
>>
>>> What do you lack in the CLFSWM REPL?
>>
>>
>> _______________________________________________
>> clfswm-devel mailing list
>> clfswm-devel at common-lisp.net
>> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel
>
> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel


-- 
?? Opera ????????????: http://www.opera.com/mail/




From pbrochard at common-lisp.net  Sat Oct 13 11:16:21 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 13 Oct 2012 13:16:21 +0200
Subject: [clfswm-devel] Can we use complete REPL in clfswm directly ?
In-Reply-To: <op.wl3x4rfb1uofcn@redirector.netnames.net> (z. axis's message of
	"Sat, 13 Oct 2012 16:06:16 +0800")
References: <op.wl1wxlko1uofcn@redirector.netnames.net>
	<87obk7we8e.fsf@common-lisp.net>
	<op.wl3cirnj1uofcn@redirector.netnames.net>
	<87fw5ikit3.fsf@common-lisp.net>
	<op.wl3x4rfb1uofcn@redirector.netnames.net>
Message-ID: <87ehl2aagq.fsf@common-lisp.net>

z. axis writes:

> That's great!  I have just tested it .  I donot understand why stumpwm
> cannot do it ?
>
Simply because nobody need it at the point to code it...
BTW the CLFSWM code is free, you can use it in Stumpwm!
But it lacks the *, ** and ***. (I'll fix this soon).

> ? Sat, 13 Oct 2012 14:06:00 +0800?Philippe Brochard
> <pbrochard at common-lisp.net> ??:
>
>> z. axis writes:
>>
>>> Can CLFSWM REPl output multiple line result ?
>>>
>>> (defun div618 (p1 p2)
>>>     (let ((ratio '(0. 0.191 0.236 0.382 0.5 0.618 0.809 1.))
>>>           (price #'(lambda (r) (if (<= p1 p2) (+ p1 (* (- p2 p1) r))
>>> (- p1 (* (- p1 p2) r))))))
>>>           (if (<= p1 p2)
>>>             (dolist (r (reverse ratio)) (format t "-------~3$
>>> ~$-------~%" r (funcall price r)))
>>>             (dolist (r ratio) (format t "-------~3$  ~$-------~%" r
>>> (funcall price r))))))
>>>
>>> The stumpwm REPL cannot output the result because div618 just
>>> return nil.
>>>
>> Is this ok for you:
>>
>>   http://common-lisp.net/project/clfswm/z_axis-REPL-screen.png
>>
>> (You can have tested by yourself also...)
>>
>>>
>>> ? Sat, 13 Oct 2012 05:51:45 +0800?Philippe Brochard
>>> <pbrochard at common-lisp.net> ??:
>>>
>>>> What do you lack in the CLFSWM REPL?
>>>
>>>
>>> _______________________________________________
>>> clfswm-devel mailing list
>>> clfswm-devel at common-lisp.net
>>> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel
>>
>> _______________________________________________
>> clfswm-devel mailing list
>> clfswm-devel at common-lisp.net
>> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel
>
>
> --
> ?? Opera ????????????: http://www.opera.com/mail/
>
>
> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel



From pbrochard at common-lisp.net  Sat Oct 13 13:01:03 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 13 Oct 2012 15:01:03 +0200
Subject: [clfswm-devel] Some patches
In-Reply-To: <87wqyvwf11.fsf@common-lisp.net> (Philippe Brochard's message of
	"Fri, 12 Oct 2012 23:34:34 +0200")
References: <1349985159-sup-5215@leandros> <87wqyvwf11.fsf@common-lisp.net>
Message-ID: <87fw5ibk6o.fsf@common-lisp.net>

Philippe Brochard writes:

>> Michael Raitza writes:
>
>> Hi,
>>
> Hi,
>
Hi again,

>> I prepared a series of patches mostly improvements.
>>

[... Patches already applied ...]

>> In the fifth patch I have refactored run-program-from-query-string. This
>> function could not be run from main mode because it binds the ? the
>> function list to be run when leaving the second mode. So it was never
>> called before entering and leaving second mode somewhen later. This
>> patch introduces the function run-program-from-query-string-fun that
>> does the actual query and generates the function.
>> run-program-from-query-string uses its output. It also introduces
>> run-program-from-query-string-main to be called from main mode.
>>
> I'll take a look at this one tomorrow.
>
I've commited, in your name, a slightly simpler version. You have the
*in-second-mode* variable which indicates in which mode you are (main or
second).

>> The third and fifth patch seem to be related, because they are causes by
>> the differences of the functions called in second mode from functions
>> providing the same functionality in main mode. Can't we consolidate
>> these two concepts?
>>
> Same thing.
>
You can use the *in-second-mode* to know in which mode you're and the
leave-second-mode function can be called from the main mode.
So, you can build some functions made for both the main mode and the
second mode.

>
> Many thanks for your patch!
>
> While your are motivated to go deeply in the CLFSWM source code. There
> is a thing CLFSWM will always lake if nobody other than me have a look
                                lack (this one is hilarious, sorry)
> at it: CLFSWM communicate poorly with the outside: it is not fully netwm
> compliant. You can have a look at src/netwm-util.lisp and at some
> functions like netwm-update-desktop-property. 
> I think there may be a time when users will want to use CLFSWM with
> Gnome or KDE. netwm-util.lisp is here for this but this is really not a
> priority for me. 
>
Another request by Valentin Plechinger is the copy/paste method for
query string. I've little time to implement it since I'm working on
simplifying the CLFSWM build system (leave ./confugure && make && make
install) and enhance the load.lisp.  

Regards,

Philippe



From z_axis at 163.com  Sun Oct 14 02:04:50 2012
From: z_axis at 163.com (z_axis)
Date: Sun, 14 Oct 2012 10:04:50 +0800
Subject: [clfswm-devel] Can i use WIN key as mod-1 instead of ALT key ?
Message-ID: <op.wl5b2cf71uofcn@redirector.netnames.net>

The ALT key is a hard working key as many applications use it as hot key.  
So i want to use WIN key instead.
In ~/.config/clfswm/clfswm, i define:

(defun my-binding ()
   (define-shell ("g" :mod-4) b-start-eweiqi "cd $HOME && wine \"c:/Program  
Files/eweiqi/LiveBaduk.exe\"")
   (define-shell ("p" :mod-4) b-start-dmenu "cd $HOME && exec  
./bin/dmenu.sh")
   (define-shell ("t" :mod-4) b-start-xterm "cd $HOME && exec xterm")
   (define-shell ("w" :mod-4) b-start-opera "cd $HOME && exec opera"))

(add-hook *binding-hook* 'my-binding)

But it doesnot work!

Sincerely!
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: tmp.txt
URL: <https://mailman.common-lisp.net/pipermail/clfswm-devel/attachments/20121014/45ec7610/attachment.txt>

From z_axis at 163.com  Sun Oct 14 06:46:41 2012
From: z_axis at 163.com (z_axis)
Date: Sun, 14 Oct 2012 14:46:41 +0800
Subject: [clfswm-devel] Cannot display the result of clisp shell function ?
Message-ID: <op.wl5o34jb1uofcn@redirector.netnames.net>

In the eval window, (shell "ls ~") will output nothing.

Regards!




From pbrochard at common-lisp.net  Sun Oct 14 11:50:54 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sun, 14 Oct 2012 13:50:54 +0200
Subject: [clfswm-devel] Can i use WIN key as mod-1 instead of ALT key ?
In-Reply-To: <op.wl5b2cf71uofcn@redirector.netnames.net> (z. axis's message of
	"Sun, 14 Oct 2012 10:04:50 +0800")
References: <op.wl5b2cf71uofcn@redirector.netnames.net>
Message-ID: <87bog5uva9.fsf@common-lisp.net>

z. axis writes:

> The ALT key is a hard working key as many applications use it as hot
> key. So i want to use WIN key instead.
> In ~/.config/clfswm/clfswm, i define:
>
> (defun my-binding ()
>   (define-shell ("g" :mod-4) b-start-eweiqi "cd $HOME && wine
> \"c:/Program Files/eweiqi/LiveBaduk.exe\"")
>   (define-shell ("p" :mod-4) b-start-dmenu "cd $HOME && exec
> ./bin/dmenu.sh")
>   (define-shell ("t" :mod-4) b-start-xterm "cd $HOME && exec xterm")
>   (define-shell ("w" :mod-4) b-start-opera "cd $HOME && exec opera"))
>
> (add-hook *binding-hook* 'my-binding)
>
> But it doesnot work!
>
> Sincerely!
>
Hi, sure you can use MOD-4 instead of MOD-1.

Use 'binding-substitute-modifier' at the end of your binding hook.
Something like this:

--------------------------------------------------
(defun my-bindiung ()
  ... bla bla ...
  (binding-substitute-modifier :mod-4 :mod-1))

(add-hook *binding-hook* 'my-binding)
--------------------------------------------------

As always, feedback is welcome if this works or not for you.

And if you have the courage, you can add an entry in the trick and tips
or FAQ wiki page:  http://trac.common-lisp.net/clfswm/. Thanks.

Regards,

Philippe



From pbrochard at common-lisp.net  Sun Oct 14 12:05:07 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sun, 14 Oct 2012 14:05:07 +0200
Subject: [clfswm-devel] Cannot display the result of clisp shell
	function ?
References: <op.wl5o34jb1uofcn@redirector.netnames.net>
Message-ID: <87d30lfe1l.fsf@common-lisp.net>

z. axis writes:

> In the eval window, (shell "ls ~") will output nothing.
>
No, it output nothing because the command is passed to the underlying
shell and is not captured by the lisp.

What you want can be done by redirecting the shell output to a stream:

(let ((output (run-shell-command "ls ~" :output :stream))) (loop for line = (read-line output nil nil) while line do (print line)))

For what you want, you'd better use the present-clfswm-terminal function
bound to Control+F12 by default. Or an external program like guake.

But if you have often a use for this in lisp, we can add a function to
do this. Just request.

> Regards!
>
Regards,

Philippe



From z_axis at 163.com  Mon Oct 15 00:28:25 2012
From: z_axis at 163.com (z_axis)
Date: Mon, 15 Oct 2012 08:28:25 +0800
Subject: [clfswm-devel] Cannot display the result of clisp shell
 function ?
In-Reply-To: <87d30lfe1l.fsf@common-lisp.net>
References: <op.wl5o34jb1uofcn@redirector.netnames.net>
	<87d30lfe1l.fsf@common-lisp.net>
Message-ID: <op.wl619opr1uofcn@redirector.netnames.net>

I see now. thanks!
? Sun, 14 Oct 2012 20:05:07 +0800?Philippe Brochard  
<pbrochard at common-lisp.net> ??:

> z. axis writes:
>
>> In the eval window, (shell "ls ~") will output nothing.
>>
> No, it output nothing because the command is passed to the underlying
> shell and is not captured by the lisp.
>
> What you want can be done by redirecting the shell output to a stream:
>
> (let ((output (run-shell-command "ls ~" :output :stream))) (loop for  
> line = (read-line output nil nil) while line do (print line)))
>
> For what you want, you'd better use the present-clfswm-terminal function
> bound to Control+F12 by default. Or an external program like guake.
>
> But if you have often a use for this in lisp, we can add a function to
> do this. Just request.
>
>> Regards!
>>
> Regards,
>
> Philippe
>
> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel


-- 
?? Opera ????????????: http://www.opera.com/mail/




From z_axis at 163.com  Mon Oct 15 00:31:51 2012
From: z_axis at 163.com (z_axis)
Date: Mon, 15 Oct 2012 08:31:51 +0800
Subject: [clfswm-devel] Can i use WIN key as mod-1 instead of ALT key ?
In-Reply-To: <87bog5uva9.fsf@common-lisp.net>
References: <op.wl5b2cf71uofcn@redirector.netnames.net>
	<87bog5uva9.fsf@common-lisp.net>
Message-ID: <op.wl62fdgq1uofcn@redirector.netnames.net>

I am a clfswm newbie. Maybe later i will have courage.

thanks!

? Sun, 14 Oct 2012 19:50:54 +0800?Philippe Brochard  
<pbrochard at common-lisp.net> ??:

> courage


-- 
?? Opera ????????????: http://www.opera.com/mail/

From ddeus.clfswm at mailnull.com  Mon Oct 15 21:36:00 2012
From: ddeus.clfswm at mailnull.com (Oleksandr Kozachuk)
Date: Mon, 15 Oct 2012 23:36:00 +0200
Subject: [clfswm-devel] Problems with MPlayer
Message-ID: <86391f5sfz.fsf@nonexistent.info>

Hi!

I have a problem with MPlayer, if try to start it in fullscreen, it
imidiatly exits with return code 0. Otherwise it works without problems.

$> mplayer -fs file.avi ; echo $?
....
Exiting... (Quit)
0

I am using newest CLFSWM from GIT.

Best regards,
Alex.



From pbrochard at common-lisp.net  Mon Oct 15 22:13:45 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Tue, 16 Oct 2012 00:13:45 +0200
Subject: [clfswm-devel] Problems with MPlayer
In-Reply-To: <86391f5sfz.fsf@nonexistent.info> (Oleksandr Kozachuk's message
	of "Mon, 15 Oct 2012 23:36:00 +0200")
References: <86391f5sfz.fsf@nonexistent.info>
Message-ID: <87d30jpena.fsf@common-lisp.net>

Oleksandr Kozachuk writes:

> Hi!
>
Hi, welcome on this list.

> I have a problem with MPlayer, if try to start it in fullscreen, it
> imidiatly exits with return code 0. Otherwise it works without problems.
>
> $> mplayer -fs file.avi ; echo $?
> ....
> Exiting... (Quit)
> 0
>
All is ok here. Please, can you test with another video with smaller
sizes? Can you also test with VLC? Does CLFSWM reports anything?

> I am using newest CLFSWM from GIT.
>
What is the video size and what is your screen size?

> Best regards,
>
Cheers,

Philippe



From ddeus.clfswm at mailnull.com  Wed Oct 17 19:18:26 2012
From: ddeus.clfswm at mailnull.com (ddeus.clfswm at mailnull.com)
Date: Wed, 17 Oct 2012 21:18:26 +0200
Subject: [clfswm-devel] Problems with MPlayer
In-Reply-To: <383bcca1d93634b14caa7c4748e77268@kozachuk.info> (Oleksandr
	Kozachuk's message of "Wed, 17 Oct 2012 21:09:31 +0200")
References: <mailman.1.1350414012.10704.clfswm-devel@common-lisp.net>
	<383bcca1d93634b14caa7c4748e77268@kozachuk.info>
Message-ID: <86lif452m5.fsf@nonexistent.info>


> $> mplayer /tmp/vidio.mkv
Oh, i had some copy paste problem :-)

Correct command was:
--- SNIP ---
$> mplayer -fs /tmp/video.mkv
MPlayer SVN-r32219-snapshot-4.5.3 (C) 2000-2010 MPlayer Team
159 audio & 348 video codecs

Playing /tmp/video.mkv
libavformat file format detected.
[matroska,webm @ 0x7f7ff77cc010] Estimating duration from bitrate,
this may be inaccurate
[lavf] stream 0: video (h264), -vid 0
[lavf] stream 1: audio (ac3), -aid 0
[lavf] stream 2: subtitle (unknown), -sid 0
VIDEO:  [H264]  704x480  0bpp  23.976 fps    0.0 kbps ( 0.0 kbyte/s)
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffh264] vfm: ffmpeg (FFmpeg H.264)
==========================================================================
==========================================================================
Opening audio decoder: [ffmpeg] FFmpeg/libavcodec audio decoders
AUDIO: 48000 Hz, 2 ch, s16le, 448.0 kbit/29.17% (ratio: 56000->192000)
Selected audio codec: [ffac3] afm: ffmpeg (FFmpeg AC-3)
==========================================================================
AO: [oss] 48000Hz 2ch s16le (2 bytes per sample)
Starting playback...
Movie-Aspect is 1.33:1 - prescaling to correct movie aspect.
VO: [xv] 704x480 => 704x528 Planar YV12  [fs]
A:   0.2 V:   0.0 A-V:  0.192 ct:  0.000   0/  0 ??% ??% ??,?% 0 0

Exiting... (Quit)
--- SNIP ---

Best regards,
Alex.



From ddeus.clfswm at mailnull.com  Wed Oct 17 21:54:41 2012
From: ddeus.clfswm at mailnull.com (Oleksandr Kozachuk)
Date: Wed, 17 Oct 2012 23:54:41 +0200
Subject: [clfswm-devel] Problems with MPlayer
In-Reply-To: <86bog04x9k.fsf@nonexistent.info> (Oleksandr Kozachuk's message
	of "Wed, 17 Oct 2012 23:13:59 +0200")
References: <mailman.1.1350414012.10704.clfswm-devel@common-lisp.net>
	<383bcca1d93634b14caa7c4748e77268@kozachuk.info>
	<86lif452m5.fsf@nonexistent.info> <86bog04x9k.fsf@nonexistent.info>
Message-ID: <864nls4vdq.fsf@nonexistent.info>

For some reason, my full answer was not sent to the mailing list (i
think i have used wrong from address), so here the full answer again
(corrected). Sorry for spamming the mailing list a bit ....

>> Hi!
> Hi, welcome on this list.
Thank you :-)

> Please, can you test with another video with smaller sizes?
I have tried with different sizes and different formats, here full
output of a mplayer run:

--- SNIP ---
$> mplayer -fs /tmp/video.mkv
MPlayer SVN-r32219-snapshot-4.5.3 (C) 2000-2010 MPlayer Team
159 audio & 348 video codecs

Playing /tmp/video.mkv
libavformat file format detected.
[matroska,webm @ 0x7f7ff77cc010] Estimating duration from bitrate,
this may be inaccurate
[lavf] stream 0: video (h264), -vid 0
[lavf] stream 1: audio (ac3), -aid 0
[lavf] stream 2: subtitle (unknown), -sid 0
VIDEO:  [H264]  704x480  0bpp  23.976 fps    0.0 kbps ( 0.0 kbyte/s)
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffh264] vfm: ffmpeg (FFmpeg H.264)
==========================================================================
==========================================================================
Opening audio decoder: [ffmpeg] FFmpeg/libavcodec audio decoders
AUDIO: 48000 Hz, 2 ch, s16le, 448.0 kbit/29.17% (ratio: 56000->192000)
Selected audio codec: [ffac3] afm: ffmpeg (FFmpeg AC-3)
==========================================================================
AO: [oss] 48000Hz 2ch s16le (2 bytes per sample)
Starting playback...
Movie-Aspect is 1.33:1 - prescaling to correct movie aspect.
VO: [xv] 704x480 => 704x528 Planar YV12  [fs]
A:   0.2 V:   0.0 A-V:  0.192 ct:  0.000   0/  0 ??% ??% ??,?% 0 0

Exiting... (Quit)
--- SNIP ---

I have tried very different sizes and formats, it does not work.

> Can you also test with VLC?
No, i have not installed it jet, but i will try it latter.

> Does CLFSWM reports anything?
No, should I activate some debug output or similar?

> What is the video size and what is your screen size?

--- SNIP ---
$> xrandr
Screen 0: minimum 320 x 200, current 1920 x 1080, maximum 1920 x 1920
VGA-0 connected 1920x1080+0+0 (normal left inverted right x axis y
axis) 521mm x 293mm
   1920x1080      60.0*+
--- SNIP ---

Best regards,
Alex.



From ddeus.clfswm at mailnull.com  Wed Oct 17 22:15:41 2012
From: ddeus.clfswm at mailnull.com (Oleksandr Kozachuk)
Date: Thu, 18 Oct 2012 00:15:41 +0200
Subject: [clfswm-devel] Create frames on start
Message-ID: <86y5j43fua.fsf@nonexistent.info>

Hi again!

I would like to create and configure some frames at startup and have now
this code in my clfswmrc for it:

;; -----------------------------------------------------------------------------

(defun set-frame-layout-fast (layout &optional (frame (current-child)))
  (when (frame-p frame)
    (set-frame-default frame)
    (setf (frame-layout frame) layout)
    (show-all-children frame)))

(defun create-bounded-frame (name parent layout &optional (slot nil))
  (let ((frame (add-frame (create-frame :name name) parent)))
    (set-frame-layout-fast layout frame)
    (when slot (bind-child-on-slot slot frame))
    frame))

(defun ok-init-hook ()
  (let ((root (create-bounded-frame "Root" *root-frame* #'tile-space-layout)))
    (let* ((default   (create-bounded-frame "Default"   root #'tile-space-layout 0))
	   (devel     (create-bounded-frame "Devel"     root #'tile-space-layout 1))
	   (www       (create-bounded-frame "WWW"       root #'tile-space-layout 2))
	   (unmanaged (create-bounded-frame "Unmanaged" root #'tile-left-layout  3)))
      (create-bounded-frame "Devel P2" devel #'tile-left-layout 4)
      (create-bounded-frame "Devel P1" devel #'tile-left-layout 5)
      (setf (frame-managed-type unmanaged) nil))
    (define-as-root root 0 0 (xlib:screen-width *screen*) (xlib:screen-height *screen*))
    (focus-all-children root root)))

(setf *init-hook* '(ok-init-hook))

;; -----------------------------------------------------------------------------

It works, but i do not know if it is a goot way to do it, because i
replace the initialization function fully and i am not sure about root
frames. How are you doing that?

Other thing is my "bind-child-on-slot" function, it would be nice to
have additional optional parameter in the normal "bind-on-slot" function
which is bound per default to (current-child), instead of using
(current-child) directly.

Best regards,
Alex.



From pbrochard at common-lisp.net  Sat Oct 20 07:08:41 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 20 Oct 2012 09:08:41 +0200
Subject: [clfswm-devel] Problems with MPlayer
In-Reply-To: <864nls4vdq.fsf@nonexistent.info> (Oleksandr Kozachuk's message
	of "Wed, 17 Oct 2012 23:54:41 +0200")
References: <mailman.1.1350414012.10704.clfswm-devel@common-lisp.net>
	<383bcca1d93634b14caa7c4748e77268@kozachuk.info>
	<86lif452m5.fsf@nonexistent.info> <86bog04x9k.fsf@nonexistent.info>
	<864nls4vdq.fsf@nonexistent.info>
Message-ID: <87ipa5oc1y.fsf@common-lisp.net>

[Sorry for the lag]

Oleksandr Kozachuk writes:

> For some reason, my full answer was not sent to the mailing list (i
> think i have used wrong from address), so here the full answer again
> (corrected). Sorry for spamming the mailing list a bit ....
>
>>> Hi!
>> Hi, welcome on this list.
> Thank you :-)
>
>> Please, can you test with another video with smaller sizes?
> I have tried with different sizes and different formats, here full
> output of a mplayer run:
>
> --- SNIP ---
> $> mplayer -fs /tmp/video.mkv
> MPlayer SVN-r32219-snapshot-4.5.3 (C) 2000-2010 MPlayer Team
> 159 audio & 348 video codecs
>
> Playing /tmp/video.mkv
> libavformat file format detected.
> [matroska,webm @ 0x7f7ff77cc010] Estimating duration from bitrate,
> this may be inaccurate
> [lavf] stream 0: video (h264), -vid 0
> [lavf] stream 1: audio (ac3), -aid 0
> [lavf] stream 2: subtitle (unknown), -sid 0
> VIDEO:  [H264]  704x480  0bpp  23.976 fps    0.0 kbps ( 0.0 kbyte/s)
> ==========================================================================
> Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
> Selected video codec: [ffh264] vfm: ffmpeg (FFmpeg H.264)
> ==========================================================================
> ==========================================================================
> Opening audio decoder: [ffmpeg] FFmpeg/libavcodec audio decoders
> AUDIO: 48000 Hz, 2 ch, s16le, 448.0 kbit/29.17% (ratio: 56000->192000)
> Selected audio codec: [ffac3] afm: ffmpeg (FFmpeg AC-3)
> ==========================================================================
> AO: [oss] 48000Hz 2ch s16le (2 bytes per sample)
> Starting playback...
> Movie-Aspect is 1.33:1 - prescaling to correct movie aspect.
> VO: [xv] 704x480 => 704x528 Planar YV12  [fs]
> A:   0.2 V:   0.0 A-V:  0.192 ct:  0.000   0/  0 ??% ??% ??,?% 0 0
>
> Exiting... (Quit)
> --- SNIP ---
>
> I have tried very different sizes and formats, it does not work.
>
I've no clue on this. I'm using a more recent mplayer than yours
(r34540). I don't know if this is related.

Some things you can try:

  1) Start mplayer in a floating frame : open a new frame and tell it to
  not manage any children (Second key-> f -> w -> u).
  And try different window format to change the aspect ratio.

  2) Start mplayer in fullscreen mode or with different drivers 
  (mplayer -vo help).

If something works, please tell it here to add an entry in the wiki.

>> Can you also test with VLC?
> No, i have not installed it jet, but i will try it latter.
>
>> Does CLFSWM reports anything?
> No, should I activate some debug output or similar?
>
No, CLFSWM is not bugging here. I think it's mplayer who don't like to
be resized by CLFSWM.

>> What is the video size and what is your screen size?
>
> --- SNIP ---
> $> xrandr
> Screen 0: minimum 320 x 200, current 1920 x 1080, maximum 1920 x 1920
> VGA-0 connected 1920x1080+0+0 (normal left inverted right x axis y
> axis) 521mm x 293mm
>    1920x1080      60.0*+
> --- SNIP ---
>
Thanks.

> Best regards,
> Alex.
>
Cheers,

Philippe



From pbrochard at common-lisp.net  Sat Oct 20 12:14:48 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 20 Oct 2012 14:14:48 +0200
Subject: [clfswm-devel] Create frames on start
In-Reply-To: <86y5j43fua.fsf@nonexistent.info> (Oleksandr Kozachuk's message
	of "Thu, 18 Oct 2012 00:15:41 +0200")
References: <86y5j43fua.fsf@nonexistent.info>
Message-ID: <878vb1nxvr.fsf@common-lisp.net>

[Sorry for the lag again]

Oleksandr Kozachuk writes:

> Hi again!
>
> I would like to create and configure some frames at startup and have now
> this code in my clfswmrc for it:
>
> ;; -----------------------------------------------------------------------------
>
> (defun set-frame-layout-fast (layout &optional (frame (current-child)))
>   (when (frame-p frame)
>     (set-frame-default frame)
>     (setf (frame-layout frame) layout)
>     (show-all-children frame)))
>
> (defun create-bounded-frame (name parent layout &optional (slot nil))
>   (let ((frame (add-frame (create-frame :name name) parent)))
>     (set-frame-layout-fast layout frame)
>     (when slot (bind-child-on-slot slot frame))
>     frame))
>
> (defun ok-init-hook ()
>   (let ((root (create-bounded-frame "Root" *root-frame* #'tile-space-layout)))
>     (let* ((default   (create-bounded-frame "Default"   root #'tile-space-layout 0))
> 	   (devel     (create-bounded-frame "Devel"     root #'tile-space-layout 1))
> 	   (www       (create-bounded-frame "WWW"       root #'tile-space-layout 2))
> 	   (unmanaged (create-bounded-frame "Unmanaged" root #'tile-left-layout  3)))
>       (create-bounded-frame "Devel P2" devel #'tile-left-layout 4)
>       (create-bounded-frame "Devel P1" devel #'tile-left-layout 5)
>       (setf (frame-managed-type unmanaged) nil))
>     (define-as-root root 0 0 (xlib:screen-width *screen*) (xlib:screen-height *screen*))
>     (focus-all-children root root)))
>
> (setf *init-hook* '(ok-init-hook))
>
> ;; -----------------------------------------------------------------------------
>
> It works, but i do not know if it is a goot way to do it, because i
> replace the initialization function fully and i am not sure about root
> frames. How are you doing that?
>
This is perfectly valid code but it'll not work if you plug another
monitor. A better way would be to let CLFSWM make its root
initialization and add your own init hook at the end of the default init
hook and use the CLFSWM structure.

More something like that:

;; -----------------------------------------------------------------------------
(defun my-init-hook ()
  (let* ((parent    (first (frame-child *root-frame*)))
         (default   (first (frame-child parent)))
         (devel     (create-bounded-frame "Devel"     parent #'tile-left-layout  1))
         (www       (create-bounded-frame "WWW"       parent #'tile-space-layout 2))
         (unmanaged (create-bounded-frame "Unmanaged" parent #'tile-left-layout  3)))
    (dbg (child-fullname parent))     ; To see where we are
    (setf (frame-layout parent) #'tile-left-layout
          (frame-name default) "Default"
          (frame-layout default) #'tile-left-layout
          (frame-managed-type unmanaged) nil)
    (bind-on-slot 0 default)
    (create-bounded-frame "Devel P2" devel #'tile-left-layout 4)
    (create-bounded-frame "Devel P1" devel #'tile-left-layout 5)
    (setf (current-child) default)
    (focus-all-children (current-child) parent)
    ;; We can show all children only once at the end of initialization
    (show-all-children)))

(add-hook *init-hook* 'my-init-hook)
;; -----------------------------------------------------------------------------



> Other thing is my "bind-child-on-slot" function, it would be nice to
> have additional optional parameter in the normal "bind-on-slot" function
> which is bound per default to (current-child), instead of using
> (current-child) directly.
>
You're right. This is fixed with the commit c31ef42.

> Best regards,
> Alex.
>
Best regards,

Philippe



From pbrochard at common-lisp.net  Sun Oct 21 18:33:00 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sun, 21 Oct 2012 20:33:00 +0200
Subject: [clfswm-devel] Create frames on start
In-Reply-To: <mailman.808.1350834011.2246.clfswm-devel@common-lisp.net>
	(clfswm-devel-bounces@common-lisp.net's message of "Sun, 21 Oct 2012
	08:40:11 -0700")
References: <mailman.808.1350834011.2246.clfswm-devel@common-lisp.net>
Message-ID: <87r4orfzfn.fsf@common-lisp.net>

>> Philippe Brochard <pbrochard at common-lisp.net> writes:
>> ;; -----------------------------------------------------------------------------
>> (defun my-init-hook ()
>>   (let* ((parent    (first (frame-child *root-frame*)))
>>          (default   (first (frame-child parent)))
>>          (devel     (create-bounded-frame "Devel"     parent #'tile-left-layout  1))
>>          (www       (create-bounded-frame "WWW"       parent #'tile-space-layout 2))
>>          (unmanaged (create-bounded-frame "Unmanaged" parent #'tile-left-layout  3)))
>>     (dbg (child-fullname parent))     ; To see where we are
>>     (setf (frame-layout parent) #'tile-left-layout
>>           (frame-name default) "Default"
>>           (frame-layout default) #'tile-left-layout
>>           (frame-managed-type unmanaged) nil)
>>     (bind-on-slot 0 default)
>>     (create-bounded-frame "Devel P2" devel #'tile-left-layout 4)
>>     (create-bounded-frame "Devel P1" devel #'tile-left-layout 5)
>>     (setf (current-child) default)
>>     (focus-all-children (current-child) parent)
>>     ;; We can show all children only once at the end of initialization
>>     (show-all-children)))
>>
>> (add-hook *init-hook* 'my-init-hook)
>> ;; -----------------------------------------------------------------------------
> Thanks, it works very well.
>
>>> Other thing is my "bind-child-on-slot" function, it would be nice to
>>> have additional optional parameter in the normal "bind-on-slot" function
>>> which is bound per default to (current-child), instead of using
>>> (current-child) directly.
>> You're right. This is fixed with the commit c31ef42.
> Super! Thank you!
>
Ok, good.

Best regards,

Philippe



From pbrochard at common-lisp.net  Sun Oct 21 21:12:09 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sun, 21 Oct 2012 23:12:09 +0200
Subject: [clfswm-devel] Cannot display the result of clisp shell
	function ?
In-Reply-To: <op.wl5o34jb1uofcn@redirector.netnames.net> (z. axis's message of
	"Sun, 14 Oct 2012 14:46:41 +0800")
References: <op.wl5o34jb1uofcn@redirector.netnames.net>
Message-ID: <878vazedhy.fsf@common-lisp.net>

z. axis writes:

> In the eval window, (shell "ls ~") will output nothing.
>
I have added a 'do-shell-output' function which return a list of shell
command output.

> Regards!
>
Cheers,

Philippe



From pbrochard at common-lisp.net  Sun Oct 21 21:26:35 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sun, 21 Oct 2012 23:26:35 +0200
Subject: [clfswm-devel] ./configure && make && make install?
Message-ID: <874nlnectw.fsf@common-lisp.net>

Hi,

I have deeply changed the load.lisp file so now it builds a clfswm
executable image in the current directory. It can then install it in a
different location. Also, if it can't find ASDF or CLX, it can ask to
download them from there web site.
So building the CLFSWM binary from scratch is as simple as this for each
implementation. We just need a lisp on our system: 

clisp load.lisp
sbcl --load load.lisp
cmucl -load load.lisp
ccl -l load.lisp
ecl -load load.lisp

(At the time of writing, CMUCL doesn't link it. And ECL can build it but
it doesn't works as expected).

So the question is: Is there somebody who use the ./configure && make
&& make install method? Or do you use your distro packages or quicklisp?

If nobody uses the ./configure script, I'd like to remove it from the
source directory.

What do you think about this?
And indeed test and feedback are very appreciated. 

Regards,

Philippe 



From z_axis at 163.com  Mon Oct 22 00:11:21 2012
From: z_axis at 163.com (z_axis)
Date: Mon, 22 Oct 2012 08:11:21 +0800
Subject: [clfswm-devel] Cannot display the result of clisp shell
 function ?
In-Reply-To: <878vazedhy.fsf@common-lisp.net>
References: <op.wl5o34jb1uofcn@redirector.netnames.net>
	<878vazedhy.fsf@common-lisp.net>
Message-ID: <op.wmjz480k1uofcn@redirector.netnames.net>

thanks!
? Mon, 22 Oct 2012 05:12:09 +0800?Philippe Brochard  
<pbrochard at common-lisp.net> ??:

> z. axis writes:
>
>> In the eval window, (shell "ls ~") will output nothing.
>>
> I have added a 'do-shell-output' function which return a list of shell
> command output.
>
>> Regards!
>>
> Cheers,
>
> Philippe
>
> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel


-- 
?? Opera ????????????: http://www.opera.com/mail/




From spacefrogg-clfswm at meterriblecrew.net  Wed Oct 24 18:54:02 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Wed, 24 Oct 2012 20:54:02 +0200
Subject: [clfswm-devel] Some patches
In-Reply-To: <87wqyvwf11.fsf@common-lisp.net>
References: <1349985159-sup-5215@leandros> <87wqyvwf11.fsf@common-lisp.net>
Message-ID: <1351104458-sup-9334@leandros>

> 
> > Hi,
> >
> Hi,
> 
> > The second patch modifies the frame-select-*-child functions to wrap
> > around the end of the list of children. Isn't that the intended
> > behaviour? I found it much more pleasing this way.
> >
> Commited. But this one break CLFSWM when there is no children in the
> current frame (fixed).
> 
I noticed this myself already. :)
> > The third patch adds set-layout-simple. The other set-layout* functions
> > were to be called from second mode if I understood correctly. I want to
> > do it from main mode. Is there a general way to get rid of these second
> > mode / main mode differences?
> >
> Commited but how do you use it?
> 
I have bound some keys to different fixed layouts. I get the impression
that I use the second mode much less than is intended. I have bound
mod-4 to caps lock so I have a modifier to the window manager's
descretion. So I can fill my keys with lots of short-cuts without
interference to other software.
> 
> Many thanks for your patch!
> 
Thanks the the window manager in the first place!

> While your are motivated to go deeply in the CLFSWM source code. There
> is a thing CLFSWM will always lake if nobody other than me have a look
> at it: CLFSWM communicate poorly with the outside: it is not fully netwm
> compliant. You can have a look at src/netwm-util.lisp and at some
> functions like netwm-update-desktop-property. 
> I think there may be a time when users will want to use CLFSWM with
> Gnome or KDE. netwm-util.lisp is here for this but this is really not a
> priority for me. 

I'll look at it soon as time permits...

Regards,

Michael




From spacefrogg-clfswm at meterriblecrew.net  Wed Oct 24 19:05:13 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Wed, 24 Oct 2012 21:05:13 +0200
Subject: [clfswm-devel] Problems with MPlayer
In-Reply-To: <87ipa5oc1y.fsf@common-lisp.net>
References: <mailman.1.1350414012.10704.clfswm-devel@common-lisp.net>
	<383bcca1d93634b14caa7c4748e77268@kozachuk.info>
	<86lif452m5.fsf@nonexistent.info> <86bog04x9k.fsf@nonexistent.info>
	<864nls4vdq.fsf@nonexistent.info> <87ipa5oc1y.fsf@common-lisp.net>
Message-ID: <1351105408-sup-6525@leandros>

I can confirm that mplayer quits if switched to fullscreen. Haven't
looked into it. Maybe some hint from the window manager is missing?

Version used: SVN-r33094-4.5.3

Michael




From spacefrogg-clfswm at meterriblecrew.net  Wed Oct 24 19:30:37 2012
From: spacefrogg-clfswm at meterriblecrew.net (Michael Raitza)
Date: Wed, 24 Oct 2012 21:30:37 +0200
Subject: [clfswm-devel] ./configure && make && make install?
In-Reply-To: <874nlnectw.fsf@common-lisp.net>
References: <874nlnectw.fsf@common-lisp.net>
Message-ID: <1351105997-sup-2784@leandros>

Excerpts from Philippe Brochard's message of Sun Oct 21 23:26:35 +0200 2012:
> Hi,
> 
> I have deeply changed the load.lisp file so now it builds a clfswm
> executable image in the current directory. It can then install it in a
> different location. Also, if it can't find ASDF or CLX, it can ask to
> download them from there web site.
> So building the CLFSWM binary from scratch is as simple as this for each
> implementation. We just need a lisp on our system: 
> 
> clisp load.lisp
> sbcl --load load.lisp
> cmucl -load load.lisp
> ccl -l load.lisp
> ecl -load load.lisp
> 
> (At the time of writing, CMUCL doesn't link it. And ECL can build it but
> it doesn't works as expected).
> 
> So the question is: Is there somebody who use the ./configure && make
> && make install method? Or do you use your distro packages or quicklisp?
> 
> If nobody uses the ./configure script, I'd like to remove it from the
> source directory.
> 
> What do you think about this?
> And indeed test and feedback are very appreciated. 
> 
> Regards,
> 
> Philippe 
> 

I, for myself, see CLFSWM as a living system. I've modified my load.lisp
to start a SWANK listener on startup, so I can connect to my running
CLFSWM and modify it via EMACS while running it. As a consequence I
want CLFSWM to start from its sources and not to load any precompiled
image. So much for my usage model.

I think CLFSM can spare the configure/make trifold as this was developed
mostly for C-related projects in mind, but I am also not a fan of
over-automating things. Maybe we should provide a build.lisp that does
all the fanciness of finding and building the package, e.g. including
installing the binary and the sources (if somebody, like me, wished to)
somewhere meaningful.

But apart from that I would like to have a load.lisp such that I can
start CLFSWM directly from the sources.

For the building process... I would like to advocate XCVB. It's from the
same author as ASDF and is meant to be its successor. I shall be fairly
usable, but I haven't looked into it deeply. I think it's worth the try.

My 2cts,


Michael




From pbrochard at common-lisp.net  Wed Oct 24 19:39:52 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Wed, 24 Oct 2012 21:39:52 +0200
Subject: [clfswm-devel] Problems with MPlayer
In-Reply-To: <1351105408-sup-6525@leandros> (Michael Raitza's message of "Wed, 
	24 Oct 2012 21:05:13 +0200")
References: <mailman.1.1350414012.10704.clfswm-devel@common-lisp.net>
	<383bcca1d93634b14caa7c4748e77268@kozachuk.info>
	<86lif452m5.fsf@nonexistent.info> <86bog04x9k.fsf@nonexistent.info>
	<864nls4vdq.fsf@nonexistent.info> <87ipa5oc1y.fsf@common-lisp.net>
	<1351105408-sup-6525@leandros>
Message-ID: <87r4on4q2f.fsf@common-lisp.net>

Michael Raitza writes:

> I can confirm that mplayer quits if switched to fullscreen. Haven't
> looked into it. Maybe some hint from the window manager is missing?
>
> Version used: SVN-r33094-4.5.3
>
Isn't it related to this (from the man page):

--------------------------------------------------
keyboard control
...
ENTER
 Go forward in the playlist, even over the end.
...
--------------------------------------------------

Please, try playing with -fs and -nofs options. And with different
drivers. And with more than one video.
It seems that in SDL fullscreen mode the modifiers are not handled. 
So in this mode: Enter = Alt Enter = Alt Shift Enter = Leave mplayer if
at the end of the playlist.

To convince you, evaluate those two lines in the CLFSWM REPL:

--------------------------------------------------
(define-main-key ("a" :control) 'enter-frame)
(define-main-key ("a" :control :shift) 'leave-frame)
--------------------------------------------------

Test them without mplayer.

Start mplayer in fullscreen mode. Those two keys do nothing. But they
works as expected in windowed mode and change the aspect ratio without
any problem. 

So, I think this is more a SDL/mplayer bug than a CLFSWM bug.

Please, can you confirm?

Regards,

Philippe

PS: indeed, after this, evaluate

--------------------------------------------------
(undefine-main-key ("a" :control))
(undefine-main-key ("a" :control :shift))
--------------------------------------------------



From pbrochard at common-lisp.net  Wed Oct 24 20:05:56 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Wed, 24 Oct 2012 22:05:56 +0200
Subject: [clfswm-devel] ./configure && make && make install?
In-Reply-To: <1351105997-sup-2784@leandros> (Michael Raitza's message of "Wed, 
	24 Oct 2012 21:30:37 +0200")
References: <874nlnectw.fsf@common-lisp.net> <1351105997-sup-2784@leandros>
Message-ID: <87mwzb4ouz.fsf@common-lisp.net>

Michael Raitza writes:

> Excerpts from Philippe Brochard's message of Sun Oct 21 23:26:35 +0200 2012:
>> Hi,
>> 
>> I have deeply changed the load.lisp file so now it builds a clfswm
>> executable image in the current directory. It can then install it in a
>> different location. Also, if it can't find ASDF or CLX, it can ask to
>> download them from there web site.
>> So building the CLFSWM binary from scratch is as simple as this for each
>> implementation. We just need a lisp on our system: 
>> 
>> clisp load.lisp
>> sbcl --load load.lisp
>> cmucl -load load.lisp
>> ccl -l load.lisp
>> ecl -load load.lisp
>> 
>> (At the time of writing, CMUCL doesn't link it. And ECL can build it but
>> it doesn't works as expected).
>> 
>> So the question is: Is there somebody who use the ./configure && make
>> && make install method? Or do you use your distro packages or quicklisp?
>> 
>> If nobody uses the ./configure script, I'd like to remove it from the
>> source directory.
>> 
>> What do you think about this?
>> And indeed test and feedback are very appreciated. 
>> 
>> Regards,
>> 
>> Philippe 
>> 
>
> I, for myself, see CLFSWM as a living system. I've modified my load.lisp
> to start a SWANK listener on startup, so I can connect to my running
> CLFSWM and modify it via EMACS while running it. As a consequence I
> want CLFSWM to start from its sources and not to load any precompiled
> image. So much for my usage model.
>
> I think CLFSM can spare the configure/make trifold as this was developed
> mostly for C-related projects in mind, but I am also not a fan of
> over-automating things. Maybe we should provide a build.lisp that does
> all the fanciness of finding and building the package, e.g. including
> installing the binary and the sources (if somebody, like me, wished to)
> somewhere meaningful.
>
> But apart from that I would like to have a load.lisp such that I can
> start CLFSWM directly from the sources.
>
You can always do what you want with the actual load.lisp. Just comment
the line with :clfswm-build-image and uncomment the one with :clfswm-run
at the beginning of the file.
And as always, you can use asdf and clx installed manually (I do this
myself). 

I've just tried to ease the process for newcomers to use the source on a
fresh system.

If nobody complain, I'll remove the ./configure script as it is too C
related and complex to manage/enhance.


> For the building process... I would like to advocate XCVB. It's from the
> same author as ASDF and is meant to be its successor. I shall be fairly
> usable, but I haven't looked into it deeply. I think it's worth the try.
>
Interesting, but at first look, it seems more complicated than
ASDF (far more dependencies). But I keep it in mind. Thanks.

> My 2cts,
>
Cheers,

Philippe



From pbrochard at common-lisp.net  Mon Oct 29 16:35:48 2012
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Mon, 29 Oct 2012 17:35:48 +0100
Subject: [clfswm-devel] Problems with MPlayer
In-Reply-To: <87r4on4q2f.fsf@common-lisp.net> (Philippe Brochard's message of
	"Wed, 24 Oct 2012 21:39:52 +0200")
References: <mailman.1.1350414012.10704.clfswm-devel@common-lisp.net>
	<383bcca1d93634b14caa7c4748e77268@kozachuk.info>
	<86lif452m5.fsf@nonexistent.info> <86bog04x9k.fsf@nonexistent.info>
	<864nls4vdq.fsf@nonexistent.info> <87ipa5oc1y.fsf@common-lisp.net>
	<1351105408-sup-6525@leandros> <87r4on4q2f.fsf@common-lisp.net>
Message-ID: <878vapfd7f.fsf@common-lisp.net>

So, is this ok? Or is there something to fix in CLFSWM?


Philippe Brochard writes:

> Michael Raitza writes:
>
>> I can confirm that mplayer quits if switched to fullscreen. Haven't
>> looked into it. Maybe some hint from the window manager is missing?
>>
>> Version used: SVN-r33094-4.5.3
>>
> Isn't it related to this (from the man page):
>
> --------------------------------------------------
> keyboard control
> ...
> ENTER
>  Go forward in the playlist, even over the end.
> ...
> --------------------------------------------------
>
> Please, try playing with -fs and -nofs options. And with different
> drivers. And with more than one video.
> It seems that in SDL fullscreen mode the modifiers are not handled. 
> So in this mode: Enter = Alt Enter = Alt Shift Enter = Leave mplayer if
> at the end of the playlist.
>
> To convince you, evaluate those two lines in the CLFSWM REPL:
>
> --------------------------------------------------
> (define-main-key ("a" :control) 'enter-frame)
> (define-main-key ("a" :control :shift) 'leave-frame)
> --------------------------------------------------
>
> Test them without mplayer.
>
> Start mplayer in fullscreen mode. Those two keys do nothing. But they
> works as expected in windowed mode and change the aspect ratio without
> any problem. 
>
> So, I think this is more a SDL/mplayer bug than a CLFSWM bug.
>
> Please, can you confirm?
>
> Regards,
>
> Philippe
>
> PS: indeed, after this, evaluate
>
> --------------------------------------------------
> (undefine-main-key ("a" :control))
> (undefine-main-key ("a" :control :shift))
> --------------------------------------------------
>
> _______________________________________________
> clfswm-devel mailing list
> clfswm-devel at common-lisp.net
> http://lists.common-lisp.net/cgi-bin/mailman/listinfo/clfswm-devel