From pbrochard at common-lisp.net  Tue Mar  4 22:53:28 2014
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Tue, 04 Mar 2014 22:53:28 +0000
Subject: Problem with macro with-timer
In-Reply-To: <878uspzvvj.fsf@argh.homenet.telecomitalia.it> (Andrea De
	Michele's message of "Tue, 04 Mar 2014 19:01:04 +0100")
References: <87ioruyrdz.fsf@argh.homenet.telecomitalia.it>
	<87k3cakm2s.fsf@common-lisp.net>
	<878uspzvvj.fsf@argh.homenet.telecomitalia.it>
Message-ID: <87fvmxlgnr.fsf@common-lisp.net>

Andrea De Michele writes:


[...]

> I think that gensym it is not created only once at compile time, but gensym as
> default value is evaluated each time before macroexpansion. 
>
> If I evaluate several time this form:
>
> (with-timer (2) (focus-frame-by (find-frame-by-name "notification")))
>
> the error is each time different:
>
> CLFSWM> (with-timer (2) (focus-frame-by (find-frame-by-name "notification")))
> ; Evaluation aborted on #<UNBOUND-VARIABLE G1462 {1003A88ED3}>.
> CLFSWM> (with-timer (2) (focus-frame-by (find-frame-by-name "notification")))
> ; Evaluation aborted on #<UNBOUND-VARIABLE G1463 {1003B1B463}>.
>
> gensym is evaluated each time and its return value is each time
> different (#:G1462 and #:G1463 in this case)
>
The symbol is created at compile time (macro expansion) and is new each
time we encounter a with-timer. When you quote it, you make (gensym)
available at evaluation time.

A test:

(defmacro test (&optional (x '(gensym)) (y (gensym)) (z (gensym)))
  `(list ,x ,y ',z))

(defun test1 ()
  (print (macroexpand-1 '(test)))
  (print (macroexpand-1 '(test)))
  (print (macroexpand-1 '(test))))

=>
(LIST (GENSYM) #:G3242 '#:G3243) 
(LIST (GENSYM) #:G3244 '#:G3245) 
(LIST (GENSYM) #:G3246 '#:G3247)

The error comes from the fact that (gensym) create a new symbol but
#G:3242 is a symbol not a bounded variable. And the list function
try to evaluate #:G3242. 
So we can write it as the x or z parts to prevent the error.


> I have found an old blog post that explains this strange macro parameter
> behaviour:
>
> http://www.findinglisp.com/blog/2005/01/keyword-parameters-in-macro-expansions.html
>
Thanks for the blog post. Those type of bugs are pretty hard to
spot. And the only place we used it is with a named id in a contrib
module.
I haven't tested sufficiently this macro :-(

> PS: thank you very much for this wonderful windows manager.
>
Thanks a lot for using it! And welcome on this list if I haven't wish
you this before. (I'm not accustomed to the new mailing list system to
know new members).



From andrea.demichele at gmail.com  Tue Mar  4 18:01:04 2014
From: andrea.demichele at gmail.com (Andrea De Michele)
Date: Tue, 04 Mar 2014 19:01:04 +0100
Subject: Problem with macro with-timer
In-Reply-To: <87k3cakm2s.fsf@common-lisp.net> (Philippe Brochard's message of
	"Tue, 04 Mar 2014 15:41:47 +0000")
References: <87ioruyrdz.fsf@argh.homenet.telecomitalia.it>
	<87k3cakm2s.fsf@common-lisp.net>
Message-ID: <878uspzvvj.fsf@argh.homenet.telecomitalia.it>

Philippe Brochard <pbrochard at common-lisp.net> writes:

> Andrea De Michele writes:
>
>> I tried to call the macro with-timer like this:
>>
>> (with-timer (2) (focus-frame-by (find-frame-by-name "notification")))
>>
>> But sbcl give me the following error:
>>
>> The variable #:G1442 is unbound.
>>
>> If I call the macro with the optional parameter id like this all works:
>>
>> (with-timer (2 (gensym)) (focus-frame-by (find-frame-by-name "notification")))
>>
>> I think (but I'm not sure) that the problem is that the default value
>> for the optional parameter id is evaluated before macroexpansion.
>>
>> If I change the definition of with-timer from:
>>
>> (defmacro with-timer ((delay &optional (id (gensym))) &body body)
>>   "Same thing as add-timer but with syntaxic sugar"
>>   `(add-timer ,delay
>> 	      (lambda ()
>> 		, at body)
>> 	      ,id))
>>
>> to:
>>
>> (defmacro with-timer ((delay &optional (id '(gensym))) &body body)
>>   "Same thing as add-timer but with syntaxic sugar"
>>   `(add-timer ,delay
>> 	      (lambda ()
>> 		, at body)
>> 	      ,id))
>>
>> both the way to call the macro (with and without the optional id parameters) works.
>>
> Hi, thanks for the report. You're right, the gensym is created only
> once at compilation time. Your second version is the right thing to do
> since a gensym is created each time we call add-timer.
> The fix is applied in your name.
>
> Regards,
>
> Philippe
>

I think that gensym it is not created only once at compile time, but gensym as
default value is evaluated each time before macroexpansion. 

If I evaluate several time this form:

(with-timer (2) (focus-frame-by (find-frame-by-name "notification")))

the error is each time different:

CLFSWM> (with-timer (2) (focus-frame-by (find-frame-by-name "notification")))
; Evaluation aborted on #<UNBOUND-VARIABLE G1462 {1003A88ED3}>.
CLFSWM> (with-timer (2) (focus-frame-by (find-frame-by-name "notification")))
; Evaluation aborted on #<UNBOUND-VARIABLE G1463 {1003B1B463}>.

gensym is evaluated each time and its return value is each time
different (#:G1462 and #:G1463 in this case)

I have found an old blog post that explains this strange macro parameter
behaviour:

http://www.findinglisp.com/blog/2005/01/keyword-parameters-in-macro-expansions.html

PS: thank you very much for this wonderful windows manager.
-- 
Andrea De Michele


From pbrochard at common-lisp.net  Tue Mar  4 15:41:47 2014
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Tue, 04 Mar 2014 15:41:47 +0000
Subject: Problem with macro with-timer
In-Reply-To: <87ioruyrdz.fsf@argh.homenet.telecomitalia.it> (Andrea De
	Michele's message of "Tue, 04 Mar 2014 15:23:20 +0100")
References: <87ioruyrdz.fsf@argh.homenet.telecomitalia.it>
Message-ID: <87k3cakm2s.fsf@common-lisp.net>

Andrea De Michele writes:

> I tried to call the macro with-timer like this:
>
> (with-timer (2) (focus-frame-by (find-frame-by-name "notification")))
>
> But sbcl give me the following error:
>
> The variable #:G1442 is unbound.
>
> If I call the macro with the optional parameter id like this all works:
>
> (with-timer (2 (gensym)) (focus-frame-by (find-frame-by-name "notification")))
>
> I think (but I'm not sure) that the problem is that the default value
> for the optional parameter id is evaluated before macroexpansion.
>
> If I change the definition of with-timer from:
>
> (defmacro with-timer ((delay &optional (id (gensym))) &body body)
>   "Same thing as add-timer but with syntaxic sugar"
>   `(add-timer ,delay
> 	      (lambda ()
> 		, at body)
> 	      ,id))
>
> to:
>
> (defmacro with-timer ((delay &optional (id '(gensym))) &body body)
>   "Same thing as add-timer but with syntaxic sugar"
>   `(add-timer ,delay
> 	      (lambda ()
> 		, at body)
> 	      ,id))
>
> both the way to call the macro (with and without the optional id parameters) works.
>
Hi, thanks for the report. You're right, the gensym is created only
once at compilation time. Your second version is the right thing to do
since a gensym is created each time we call add-timer.
The fix is applied in your name.

Regards,

Philippe


From michael at cadilhac.name  Fri Mar 14 12:26:03 2014
From: michael at cadilhac.name (=?UTF-8?Q?Micha=C3=ABl_Cadilhac?=)
Date: Fri, 14 Mar 2014 13:26:03 +0100
Subject: clisp can't find build-lisp-image
In-Reply-To: <87zjl9k1dr.fsf@common-lisp.net>
References: <CADt3fpMHgp5V2mzw7XrJCgNeSYK-NDmyKPj+vZ_2W6e3-Rgekw@mail.gmail.com>
	<87zjl9k1dr.fsf@common-lisp.net>
Message-ID: <CADt3fpNa1fFCSWpqF+UPvfqFnb6Nt4SVnd2uxW-dy4G30q7wgQ@mail.gmail.com>

Philippe;

Thanks for your answer!

On Sat, Mar 1, 2014 at 11:19 PM, Philippe Brochard
<pbrochard at common-lisp.net> wrote:

> Here is what I've done to run clfswm under ubuntu 12.04 and 13.10:
>
> sudo apt-get install git sbcl
> git clone git://common-lisp.net/projects/clfswm/clfswm.git
> cd clfswm
> make    (accept to load clx)

This failed on my system with:

> CLX not found. Do you want to download it from git://github.com/sharplispers/clx.git ? (y or n) y
>
> debugger invoked on a SB-INT:C-STRING-DECODING-ERROR in thread
> #<THREAD "initial thread" RUNNING {AAF2931}>:
>  c-string decoding error (:external-format :UTF-8):
>    the octet sequence 3 cannot be decoded.

because of an environment variable the adminsys set which had some
miscoding.  Fixing the variable, plus the line quit>exit, everything
went fine, thanks a bunch!

With the latest Git, I still have the occasional crashes with the fast
switch (a resounding CLFSWM Error: The value 3 is not of type (OR NULL
SIMPLE-VECTOR)..), and I still have the focus problems with the popup
term.  Should you need any more infos on that, I can provide them.

Happy hacking;
M.


From renaud at casenave-pere.fr  Fri Mar 28 04:57:55 2014
From: renaud at casenave-pere.fr (=?iso-8859-1?Q?Renaud_Casenave-P=E9r=E9?=)
Date: Fri, 28 Mar 2014 13:57:55 +0900
Subject: Hidden features
Message-ID: <87txaizzqk.fsf@sandalphon.casenave.fr>

Hi,

I'm trying to find new ways to use clfswm and as I looked around in the menu I
tried to understand how I could use the command "Remove the current child from
its parent frame". When I use it for example on a frame containing conkeror, the
frame and conkeror are indeed removed from the parent frame but there seems to
be no way to get conkeror back although it is still running in the background.

How is this command best used?

Thanks,

-- 
Renaud Casenave-P?r?


From pbrochard at common-lisp.net  Sat Mar  1 22:19:44 2014
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 01 Mar 2014 22:19:44 +0000
Subject: clisp can't find build-lisp-image
In-Reply-To: <CADt3fpMHgp5V2mzw7XrJCgNeSYK-NDmyKPj+vZ_2W6e3-Rgekw@mail.gmail.com>
	(=?iso-8859-1?Q?=22Micha=EBl?= Cadilhac"'s message of "Wed, 26 Feb 2014
 15:40:31 +0100")
References: <CADt3fpMHgp5V2mzw7XrJCgNeSYK-NDmyKPj+vZ_2W6e3-Rgekw@mail.gmail.com>
Message-ID: <87zjl9k1dr.fsf@common-lisp.net>

Micha?l Cadilhac writes:

> Hi there;
>
Hi, sorry for the lag but I have had to boot under ubuntu for tests
(grub is a very nice piece of software for this with its ability to
boot on iso files).

> As you may recall, I'm not at all versed in lisp machinery (which is
> evil, let it be noted).  I just arrived at my new job (on a Ubuntu
> 12.04) and wanted to show off my CLFSWM skills, but I can't seem to be
> able to install it.  The debian package seems outdated (at least, it
> doesn't look at my config) hence I went on to compile my own.  It gets
> stuck with:
>
>    undefined function BUILD-LISP-IMAGE
>
> My meager knowledge of this led me to believe that I had to install
> the cl-acl-compat package, which was cumbersome enough as it required
> to install an older version of asdf, which conflicted with the one
> CLFSWM installs.  Conclusion: still no build-lisp-image to be found.
>
> A hand would be greatly appreciated.
>
Here is what I've done to run clfswm under ubuntu 12.04 and 13.10:

sudo apt-get install git sbcl
git clone git://common-lisp.net/projects/clfswm/clfswm.git
cd clfswm
make    (accept to load clx)
echo xterm > ~/.xsession
Control + Alt + F1 (switch to console)
startx -- :1
cd clfswm
./clfswm

For the sbcl in 12.04, you have to change the line 788 in src/tools.lisp
from  #+sbcl (sb-ext:exit) to   #+sbcl (sb-ext:quit)
you can also comment this line.

It works also with clisp (both in 13.10 and 12.04).

I haven't tried the tarball (I'll test now) but if it fails it'll be
time to make a new one.

Regards,

Philippe



From pbrochard at common-lisp.net  Fri Mar 14 22:43:51 2014
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Fri, 14 Mar 2014 22:43:51 +0000
Subject: clisp can't find build-lisp-image
In-Reply-To: <CADt3fpNa1fFCSWpqF+UPvfqFnb6Nt4SVnd2uxW-dy4G30q7wgQ@mail.gmail.com>
	(=?iso-8859-1?Q?=22Micha=EBl?= Cadilhac"'s message of "Fri, 14 Mar 2014
 13:26:03 +0100")
References: <CADt3fpMHgp5V2mzw7XrJCgNeSYK-NDmyKPj+vZ_2W6e3-Rgekw@mail.gmail.com>
	<87zjl9k1dr.fsf@common-lisp.net>
	<CADt3fpNa1fFCSWpqF+UPvfqFnb6Nt4SVnd2uxW-dy4G30q7wgQ@mail.gmail.com>
Message-ID: <8761ng9za0.fsf@common-lisp.net>

Micha?l Cadilhac writes:

> Philippe;
>
> Thanks for your answer!
>
> On Sat, Mar 1, 2014 at 11:19 PM, Philippe Brochard
> <pbrochard at common-lisp.net> wrote:
>
>> Here is what I've done to run clfswm under ubuntu 12.04 and 13.10:
>>
>> sudo apt-get install git sbcl
>> git clone git://common-lisp.net/projects/clfswm/clfswm.git
>> cd clfswm
>> make    (accept to load clx)
>
> This failed on my system with:
>
>> CLX not found. Do you want to download it from git://github.com/sharplispers/clx.git ? (y or n) y
>>
>> debugger invoked on a SB-INT:C-STRING-DECODING-ERROR in thread
>> #<THREAD "initial thread" RUNNING {AAF2931}>:
>>  c-string decoding error (:external-format :UTF-8):
>>    the octet sequence 3 cannot be decoded.
>
> because of an environment variable the adminsys set which had some
> miscoding.  Fixing the variable, plus the line quit>exit, everything
> went fine, thanks a bunch!
>
Ok. Maybe, which variable is the cause of this bug? We can maybe prevent
this. 

> With the latest Git, I still have the occasional crashes with the fast
> switch (a resounding CLFSWM Error: The value 3 is not of type (OR NULL
> SIMPLE-VECTOR)..), and I still have the focus problems with the popup
> term.  Should you need any more infos on that, I can provide them.
>
Oh, yes! If you can reproduce it or see when it happen, this will be
greatly appreciated. Any feedback is welcome.

> Happy hacking;
>
Thanks :-)

> M.
>
Philippe


From renaud at casenave-pere.fr  Sun Mar  2 03:21:21 2014
From: renaud at casenave-pere.fr (=?iso-8859-1?Q?Renaud_Casenave-P=E9r=E9?=)
Date: Sun, 02 Mar 2014 12:21:21 +0900
Subject: clisp can't find build-lisp-image
In-Reply-To: <8761nx1qfr.fsf@common-lisp.net> (Philippe Brochard's message of
	"Sat, 01 Mar 2014 22:53:28 +0000")
References: <CADt3fpMHgp5V2mzw7XrJCgNeSYK-NDmyKPj+vZ_2W6e3-Rgekw@mail.gmail.com>
	<87zjl9k1dr.fsf@common-lisp.net> <8761nx1qfr.fsf@common-lisp.net>
Message-ID: <87txbhmgjy.fsf@zahikel.casenave.fr>

On Sun, Mar 02 2014, Philippe Brochard wrote:

[...]

Hi,

I myself have had to fiddle with load.lisp to be able to run it and I
was thinking maybe you could use quicklisp for the loading and compiling
process. What do you think?

-- 
Renaud Casenave-P?r?


From pbrochard at common-lisp.net  Sat Mar 29 22:22:25 2014
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 29 Mar 2014 22:22:25 +0000
Subject: Hidden features
In-Reply-To: <87txaizzqk.fsf@sandalphon.casenave.fr> ("Renaud
 \=\?iso-8859-1\?Q\?Casenave-P\=E9r\=E9\=22's\?\=
	message of "Fri, 28 Mar 2014 13:57:55 +0900")
References: <87txaizzqk.fsf@sandalphon.casenave.fr>
Message-ID: <87wqfcy7a6.fsf@common-lisp.net>

Renaud Casenave-P?r? a ?crit :

> Hi,
>
Hi,

> I'm trying to find new ways to use clfswm and as I looked around in the menu I
> tried to understand how I could use the command "Remove the current child from
> its parent frame". When I use it for example on a frame containing conkeror, the
> frame and conkeror are indeed removed from the parent frame but there seems to
> be no way to get conkeror back although it is still running in the background.
>
> How is this command best used?
>
As the documentation says, this function "remove the current child from
its parent frame". It is supposed to be used when you have copied the
current child in more than one frame. This is useful when you want
multiple view of a child (for example, with a layout per frame and the
same children reorganized in each).
But there is no guard: if you remove the last view of the child, you'll
have no way to retrieve it.
Maybe we must add this guard to prevent to remove the last instance of a
child. Tell me if you think so.

> Thanks,
>
Hope this help.

Philippe


From pbrochard at common-lisp.net  Sun Mar  2 09:10:36 2014
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sun, 02 Mar 2014 09:10:36 +0000
Subject: clisp can't find build-lisp-image
In-Reply-To: <87txbhmgjy.fsf@zahikel.casenave.fr> ("Renaud
 \=\?iso-8859-1\?Q\?Casenave-P\=E9r\=E9\=22's\?\=
	message of "Sun, 02 Mar 2014 12:21:21 +0900")
References: <CADt3fpMHgp5V2mzw7XrJCgNeSYK-NDmyKPj+vZ_2W6e3-Rgekw@mail.gmail.com>
	<87zjl9k1dr.fsf@common-lisp.net> <8761nx1qfr.fsf@common-lisp.net>
	<87txbhmgjy.fsf@zahikel.casenave.fr>
Message-ID: <87r46lc6er.fsf@common-lisp.net>

Renaud Casenave-P?r? a ?crit :

> On Sun, Mar 02 2014, Philippe Brochard wrote:
>
> [...]
>
> Hi,
>
> I myself have had to fiddle with load.lisp to be able to run it and I
> was thinking maybe you could use quicklisp for the loading and compiling
> process. What do you think?
>
Well, you can already use quicklisp to load clfswm if you want and
you'll get the last git version:

(ql:quickload "clfswm")
(clfswm:main)

I'd like to stay with something with as less dependency as possible. And
I don't want to force someone to use quicklisp. But it's possible if it
wants to.

I have greatly simplified the building process since the old
contrib/clfswm bash file. Now it's just a make which build a lisp
executable image. 
Bug report on this part is very appreciated (as always it works for me :-)).
In fact the make process is just a way to choose a lisp available to load
load.lisp.

Maybe we can have a separate load-with-quicklisp.lisp file which use
quicklisp instead of a home made process.
I don't feel a need for it but if you want it can be relatively simple
to write it. The documentation will have to be clear on the choice of
load.lisp or load-with-quicklisp.lisp.

Regards,

Philippe


From andrea.demichele at gmail.com  Tue Mar  4 14:23:20 2014
From: andrea.demichele at gmail.com (Andrea De Michele)
Date: Tue, 04 Mar 2014 15:23:20 +0100
Subject: Problem with macro with-timer 
Message-ID: <87ioruyrdz.fsf@argh.homenet.telecomitalia.it>

I tried to call the macro with-timer like this:

(with-timer (2) (focus-frame-by (find-frame-by-name "notification")))

But sbcl give me the following error:

The variable #:G1442 is unbound.

If I call the macro with the optional parameter id like this all works:

(with-timer (2 (gensym)) (focus-frame-by (find-frame-by-name "notification")))

I think (but I'm not sure) that the problem is that the default value
for the optional parameter id is evaluated before macroexpansion.

If I change the definition of with-timer from:

(defmacro with-timer ((delay &optional (id (gensym))) &body body)
  "Same thing as add-timer but with syntaxic sugar"
  `(add-timer ,delay
	      (lambda ()
		, at body)
	      ,id))

to:

(defmacro with-timer ((delay &optional (id '(gensym))) &body body)
  "Same thing as add-timer but with syntaxic sugar"
  `(add-timer ,delay
	      (lambda ()
		, at body)
	      ,id))

both the way to call the macro (with and without the optional id parameters) works.
-- 
Andrea De Michele


From pbrochard at common-lisp.net  Sat Mar  1 22:53:28 2014
From: pbrochard at common-lisp.net (Philippe Brochard)
Date: Sat, 01 Mar 2014 22:53:28 +0000
Subject: clisp can't find build-lisp-image
In-Reply-To: <87zjl9k1dr.fsf@common-lisp.net> (Philippe Brochard's message of
	"Sat, 01 Mar 2014 22:19:44 +0000")
References: <CADt3fpMHgp5V2mzw7XrJCgNeSYK-NDmyKPj+vZ_2W6e3-Rgekw@mail.gmail.com>
	<87zjl9k1dr.fsf@common-lisp.net>
Message-ID: <8761nx1qfr.fsf@common-lisp.net>

Philippe Brochard writes:


[...]

> It works also with clisp (both in 13.10 and 12.04).
>
> I haven't tried the tarball (I'll test now) but if it fails it'll be
> time to make a new one.
>
Same thing with the tarball, you have to change the line as above in
src/tool.lisp for the 12.04 version and then type make. All seems to
works fine.

Regards,

Philippe


From renaud at casenave-pere.fr  Sun Mar 30 09:25:27 2014
From: renaud at casenave-pere.fr (=?iso-8859-1?Q?Renaud_Casenave-P=E9r=E9?=)
Date: Sun, 30 Mar 2014 18:25:27 +0900
Subject: [Clfswm-devel] Hidden features
In-Reply-To: <87wqfcy7a6.fsf@common-lisp.net> (Philippe Brochard's message of
 "Sat, 29 Mar 2014 22:22:25 +0000")
References: <87txaizzqk.fsf@sandalphon.casenave.fr>
 <87wqfcy7a6.fsf@common-lisp.net>
Message-ID: <87txagvy0o.fsf@zahikel.casenave.fr>

Hi,

On Sun, Mar 30 2014, Philippe Brochard wrote:

> As the documentation says, this function "remove the current child from
> its parent frame". It is supposed to be used when you have copied the
> current child in more than one frame. This is useful when you want
> multiple view of a child (for example, with a layout per frame and the
> same children reorganized in each).
> But there is no guard: if you remove the last view of the child, you'll
> have no way to retrieve it.
> Maybe we must add this guard to prevent to remove the last instance of a
> child. Tell me if you think so.

I see? Thanks for the explanation.

You could indeed foolproof this feature by preventing deletion of the
last child or you could put it in the list of hidden children (or another list)
to let the user retrieve it if removed by error, but this would be quite similar
to the "hide current child" feature?

-- 
Renaud Casenave-P?r?