From mb at bese.it  Fri Jul  1 10:16:50 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 01 Jul 2005 12:16:50 +0200
Subject: [Bese-devel] broken links on arnesi page
In-Reply-To: <51e1bc3005062816534cd73cb9@mail.gmail.com> (Eric Lavigne's
	message of "Tue, 28 Jun 2005 23:53:13 +0000")
References: <51e1bc3005062816534cd73cb9@mail.gmail.com>
Message-ID: <m2zmt6omp9.fsf@soma.local>

Eric Lavigne <lavigne.eric at gmail.com> writes:

> The download links at the bottom of the arnesi page are broken.

fixed. thanks!

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From jamie at jborder.com  Fri Jul  1 18:02:33 2005
From: jamie at jborder.com (Jamie Border)
Date: Fri, 1 Jul 2005 19:02:33 +0100
Subject: [Bese-devel] UCW dies while SLIME is connected
Message-ID: <01f001c57e67$0ebf0c60$0100a8c0@JANET>

UCW is 0.3.9
Portable ASERVE is 0.0.0
SLIME is 0.0.0
EMACS is 0.0.0

Problem:

1) Navigate to 127.0.0.1:8080/ucw/examples/index.ucw
2) Click "Signal an Error"
3) Be impressed by pretty output
4) Start emacs
5) M-x slime-connect
6) Click "Signal an Error"
7)See the different restarts in Emacs.  Be impressed.
8) Pick one.  See 'restart returned nil' in minibuffer.
9) Repeat 8 and usually/eventually EMACS locks up

What am I missing?  When I read my ucw.dribble logfile, I see that I have hit the maximum error depth.

Any takers?

Jamie

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050701/240a0181/attachment.html>

From mb at bese.it  Mon Jul  4 07:13:26 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 04 Jul 2005 09:13:26 +0200
Subject: [Bese-devel] UCW dies while SLIME is connected
References: <01f001c57e67$0ebf0c60$0100a8c0@JANET>
Message-ID: <m2ll4nrqll.fsf@soma.local>

"Jamie Border" <jamie at jborder.com> writes:

> 1) Navigate to 127.0.0.1:8080/ucw/examples/index.ucw
> 2) Click "Signal an Error"
> 3) Be impressed by pretty output
> 4) Start emacs
> 5) M-x slime-connect
> 6) Click "Signal an Error"
> 7)See the different restarts in Emacs.  Be impressed.
> 8) Pick one.  See 'restart returned nil' in minibuffer.

does this happen not matter what restart you choose?

> 9) Repeat 8 and usually/eventually EMACS locks up

what's in *slime-events*?

> What am I missing?  When I read my ucw.dribble logfile, I see that I have hit
> the maximum error depth.

could you send a cut 'n paste of this.

> Any takers?

not at the moment, sorry.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Mon Jul  4 07:32:05 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 04 Jul 2005 09:32:05 +0200
Subject: [Bese-devel] UCW in a magazin
In-Reply-To: <m2br5pqy27.fsf@soma.local> (Marco Baringer's message of "Wed,
	29 Jun 2005 12:04:00 +0200")
References: <20050623141123.GA6123@melkor.elder-gods.org>
	<m2ll4yefim.fsf@soma.local>
	<20050628173554.GA27858@melkor.elder-gods.org>
	<m2br5pqy27.fsf@soma.local>
Message-ID: <874qbbqb62.fsf_-_@flarge.here>

Just if someone might be interested (and better is capable or reading
german).
There is a very long article in the newsest Toolbox magazine. Jochen
Schmidt writes about UCW and implements a web-log with UCW. 

Regards
Friedrich


From frido at q-software-solutions.de  Mon Jul  4 15:36:03 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 04 Jul 2005 17:36:03 +0200
Subject: [Bese-devel] A bug?
In-Reply-To: <01f001c57e67$0ebf0c60$0100a8c0@JANET> (Jamie Border's message
	of "Fri, 1 Jul 2005 19:02:33 +0100")
References: <01f001c57e67$0ebf0c60$0100a8c0@JANET>
Message-ID: <87d5pyoa70.fsf@flarge.here>

while running the examples. I activated JavaScript (which I dislike
extremly), but the completion example does not work. I type in
something but nothing happens while using the Up/Down Arrows.

System is a DEBIAN/AMD64, glibc 2.3.x, sbcl 9.2.x

Regards
Friedrich


From mb at bese.it  Mon Jul  4 15:38:53 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 04 Jul 2005 17:38:53 +0200
Subject: [Bese-devel] A bug?
In-Reply-To: <87d5pyoa70.fsf@flarge.here> (Friedrich Dominicus's message of
	"Mon, 04 Jul 2005 17:36:03 +0200")
References: <01f001c57e67$0ebf0c60$0100a8c0@JANET> <87d5pyoa70.fsf@flarge.here>
Message-ID: <m2zmt237jm.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> while running the examples. I activated JavaScript (which I dislike
> extremly), but the completion example does not work. I type in
> something but nothing happens while using the Up/Down Arrows.
>
> System is a DEBIAN/AMD64, glibc 2.3.x, sbcl 9.2.x

what browser?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Mon Jul  4 15:58:02 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 04 Jul 2005 17:58:02 +0200
Subject: [Bese-devel] A bug?
In-Reply-To: <m2zmt237jm.fsf@soma.local> (Marco Baringer's message of "Mon,
	04 Jul 2005 17:38:53 +0200")
References: <01f001c57e67$0ebf0c60$0100a8c0@JANET>
	<87d5pyoa70.fsf@flarge.here> <m2zmt237jm.fsf@soma.local>
Message-ID: <878y0mo96d.fsf@flarge.here>

Mozilla 1.7.7

Regards
Friedrich


From mb at bese.it  Mon Jul  4 19:13:12 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 04 Jul 2005 21:13:12 +0200
Subject: [Bese-devel] going darcs
Message-ID: <m21x6eo053.fsf@soma.local>


hi,

  it really (really) pains me to do this, but, in the words of the
  cech playwrite whose nome i forget, i'm mad as hell and i'm not
  going to take it anymore.

  ucw (and arnesi, yaclml and (eventually) fiveam) are going to stop
  using arch and start using darcs.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From sketerpot at gmail.com  Mon Jul  4 21:34:46 2005
From: sketerpot at gmail.com (Peter Scott)
Date: Mon, 4 Jul 2005 16:34:46 -0500
Subject: [Bese-devel] A bug?
In-Reply-To: <87d5pyoa70.fsf@flarge.here>
References: <01f001c57e67$0ebf0c60$0100a8c0@JANET> <87d5pyoa70.fsf@flarge.here>
Message-ID: <7e267a9205070414347e257072@mail.gmail.com>

On 7/4/05, Friedrich Dominicus <frido at q-software-solutions.de> wrote:
> while running the examples. I activated JavaScript (which I dislike
> extremly), but the completion example does not work. I type in
> something but nothing happens while using the Up/Down Arrows.

Look in the HTML source for the page and get the URL of the JavaScript
source file that the completion widget links to, and try to view it in
your browser. If your experience is anything like mine, you'll get a
404 error. Apparently the file is not being served by whatever backend
you're using. I have this problem using the mod_lisp and araneida
backends, by the way.

-Peter


From frido at q-software-solutions.de  Tue Jul  5 06:42:03 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 08:42:03 +0200
Subject: [Bese-devel] going darcs
In-Reply-To: <m21x6eo053.fsf@soma.local> (Marco Baringer's message of "Mon,
	04 Jul 2005 21:13:12 +0200")
References: <m21x6eo053.fsf@soma.local>
Message-ID: <87zmt1wy84.fsf@flarge.here>

"Marco Baringer" <mb at bese.it> writes:

> hi,
>
>   it really (really) pains me to do this, but, in the words of the
>   cech playwrite whose nome i forget, i'm mad as hell and i'm not
>   going to take it anymore.
I can full understand you desparation. I never have seen such a
convulted user interface as tla suggest. Ever tried to run it on
Windows? 
>
>   ucw (and arnesi, yaclml and (eventually) fiveam) are going to stop
>   using arch and start using darcs.
Don't forget to read
Tools and instructions for migrating arch repositories to darcs are
provided on the darcs community website:
http://darcs.net/DarcsWiki/ConvertingFromArchhttp://darcs.net/DarcsWiki/ConvertingFromArch

Regards
Friedrich


From frido at q-software-solutions.de  Tue Jul  5 06:42:47 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 08:42:47 +0200
Subject: [Bese-devel] A bug?
In-Reply-To: <7e267a9205070414347e257072@mail.gmail.com> (Peter Scott's
	message of "Mon, 4 Jul 2005 16:34:46 -0500")
References: <01f001c57e67$0ebf0c60$0100a8c0@JANET> <87d5pyoa70.fsf@flarge.here>
	<7e267a9205070414347e257072@mail.gmail.com>
Message-ID: <87slytwy6w.fsf@flarge.here>

Peter Scott <sketerpot at gmail.com> writes:

>
> Look in the HTML source for the page and get the URL of the JavaScript
> source file that the completion widget links to, and try to view it in
> your browser. If your experience is anything like mine, you'll get a
> 404 error. Apparently the file is not being served by whatever backend
> you're using. I have this problem using the mod_lisp and araneida
> backends, by the way.
Ok that explains it to me. I'm using araneida. 

Regards
Friedrich


From der_julian at web.de  Tue Jul  5 13:22:05 2005
From: der_julian at web.de (Julian Stecklina)
Date: Tue, 5 Jul 2005 15:22:05 +0200
Subject: [Bese-devel] going darcs
In-Reply-To: <m21x6eo053.fsf@soma.local>
References: <m21x6eo053.fsf@soma.local>
Message-ID: <20050705152205.18c98955@localhost>

On Mon, 04 Jul 2005 21:13:12 +0200
"Marco Baringer" <mb at bese.it> wrote:

> 
> hi,
> 
>   it really (really) pains me to do this, but, in the words of the
>   cech playwrite whose nome i forget, i'm mad as hell and i'm not
>   going to take it anymore.
> 
>   ucw (and arnesi, yaclml and (eventually) fiveam) are going to stop
>   using arch and start using darcs.

I never figured out what I needed to do to get arch to perform something
useful, but I was able to use darcs in about 10 minutes...

Regards,
-- 
Julian Stecklina

LISP has survived for 21 years because it is an approximate local
optimum in the space of programming languages. - John McCarthy (1980)




From frido at q-software-solutions.de  Tue Jul  5 13:46:17 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 15:46:17 +0200
Subject: [Bese-devel] next question
Message-ID: <878y0lv00m.fsf@flarge.here>

Does anyone see a difference between:
(defcomponent download-locations (simple-window-component address-manipulator-mixin)
  ()
  (:default-initargs
      :title "Download locations " :stylesheet "/css/qss.css"))

(defcomponent get-address (simple-window-component address-manipulator-mixin)
  ((message :accessor message :initarg :message :initform nil))
  (:default-initargs
      :title "Downloader " :stylesheet "/css/qss.css"))



with 

(defmethod render-on ((res response) (s download-locations))
  (let ((stream (content-stream res)))
    (html-stream 
     stream
       (write-main-menu :active +product-index+ :stream stream)
  (html-stream stream
        (:h1 "QSS Downloads")
        (:p
....


and

(defmethod render-on ((res response) (g get-address))
   (let ((stream (content-stream res)))
     (write-main-menu :active +product-index+ :stream stream)

I can't just the first generates the following code:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/transitional.dtd">
<html
  ><head
    ><meta content="text/html" http-equiv="Content-Type"
    /><title
      >NIL</title
    ></head
  ><body
    ><table class="main-menu" width="100%" cellpadding="0"><tr><th
class="main-menu" colspan="6"><img alt="" src="/images/qlogo.gif"
/></th></tr><tr><td clas

and the second:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/transitional.dtd">
<html
  ><head
    ><meta content="text/html" http-equiv="Content-Type"
    /><title
      >Downloader </title
    ><link href="/css/qss.css" rel="stylesheet" type="text/css"
/></head

Why does one see the title and link ref in the second example but not
in the first one?

Regards
Friedrich


From frido at q-software-solutions.de  Tue Jul  5 14:01:46 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 16:01:46 +0200
Subject: [Bese-devel] going darcs
In-Reply-To: <20050705152205.18c98955@localhost> (Julian Stecklina's message
	of "Tue, 5 Jul 2005 15:22:05 +0200")
References: <m21x6eo053.fsf@soma.local> <20050705152205.18c98955@localhost>
Message-ID: <87vf3ptkqd.fsf@flarge.here>

Julian Stecklina <der_julian at web.de> writes:

>
> I never figured out what I needed to do to get arch to perform something
> useful, but I was able to use darcs in about 10 minutes...
Well you are not alone with it, you can do a (incf
does-not-get-archs-guy) for me ;-)

Regards
Friedrich


From frido at q-software-solutions.de  Tue Jul  5 14:41:47 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 16:41:47 +0200
Subject: [Bese-devel] Araneid backend question
Message-ID: <87r7edtivo.fsf@flarge.here>

Well I wonder what I'm supposed to do to cleanly shutdown 
an Araneida Backend. In start.lisp one finds at the end:

(ucw:startup-server ucw:*default-server*)

well the obvious counterpart is
(ucw:shutdown-server ucw:*default-server*)

Well I call it and I call separatly:
(ucw::shutdown-backend (server.backend ucw:*default-server*))

the log says:
 (ucw:shutdown-server ucw:*default-server*)
2005-07-05T16:40.11 IT.BESE.ARNESI:+INFO+ IT.BESE.UCW::UCW.BACKEND: Stopping ARANEIDA backend #<ARANEIDA-BACKEND {1005397801}> on #<HTTP-URL "http://127.0.0.1:8080/" {10041B0681}>.
#<ARANEIDA-BACKEND {1005397801}>

fine so I can restart it or not?
Socket error in "bind": EADDRINUSE (Address already in use)
   [Condition of type SB-BSD-SOCKETS:ADDRESS-IN-USE-ERROR]

Restarts:
  0: [ABORT] Abort handling SLIME request.
  1: [ABORT] Exit debugger, returning to top level.

No, I can't. 

restart-server does not help a bit because it does the same I tried
shutdown the server and restart it. 

So what do I have to do to convince Araneidea to be completely
shut-down?

Regards
Friedrich


From frido at q-software-solutions.de  Tue Jul  5 14:42:49 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 16:42:49 +0200
Subject: [Bese-devel] next question
In-Reply-To: <878y0lv00m.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 05 Jul 2005 15:46:17 +0200")
References: <878y0lv00m.fsf@flarge.here>
Message-ID: <87k6k5tity.fsf@flarge.here>

I found the answer. It's not in this declarations but in the call to
it. There I called the old stuff instead of the new one. So I guess I
found it myself...

Regards
Friedrich


From frido at q-software-solutions.de  Tue Jul  5 15:19:12 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 17:19:12 +0200
Subject: [Bese-devel] Araneid backend question
In-Reply-To: <87r7edtivo.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 05 Jul 2005 16:41:47 +0200")
References: <87r7edtivo.fsf@flarge.here>
Message-ID: <877jg5th5b.fsf@flarge.here>

Ok I found the bug here's a patch for it

--- araneida.lisp.old	2005-07-05 17:16:04.632461139 +0200
+++ araneida.lisp	2005-07-05 17:12:46.537285991 +0200
@@ -57,7 +57,8 @@
   (ucw.backend.info "Stopping ARANEIDA backend ~S on ~A."
 		    backend
 		    (default-url backend))
-   backend) 
+  (araneida:stop-listening (listener backend))
+  backend) 
 
 (defmethod register-url-handler ((backend araneida-backend) url handler)
   (ucw.backend.dribble "Registering handler ~S at url ~S." handler url backend)


Probably it would be even better to call the stop-listening and check
thereafter, but at least this patch makes it possible to restart the
Araneida backend.

Regards
Friedrich


From mb at bese.it  Tue Jul  5 15:39:15 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 05 Jul 2005 17:39:15 +0200
Subject: [Bese-devel] Araneid backend question
In-Reply-To: <877jg5th5b.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 05 Jul 2005 17:19:12 +0200")
References: <87r7edtivo.fsf@flarge.here> <877jg5th5b.fsf@flarge.here>
Message-ID: <m2oe9hl0t8.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> --- araneida.lisp.old	2005-07-05 17:16:04.632461139 +0200
> +++ araneida.lisp	2005-07-05 17:12:46.537285991 +0200
> @@ -57,7 +57,8 @@
>    (ucw.backend.info "Stopping ARANEIDA backend ~S on ~A."
>  		    backend
>  		    (default-url backend))
> -   backend) 
> +  (araneida:stop-listening (listener backend))
> +  backend) 
>  
>  (defmethod register-url-handler ((backend araneida-backend) url handler)
>    (ucw.backend.dribble "Registering handler ~S at url ~S." handler url backend)

cool. i'm currently working on moving over to darcs so you might have
to wait a few days before i commit this.

> Probably it would be even better to call the stop-listening and check
> thereafter, but at least this patch makes it possible to restart the
> Araneida backend.

what do you mean?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Tue Jul  5 16:07:39 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 05 Jul 2005 18:07:39 +0200
Subject: [Bese-devel] Araneid backend question
In-Reply-To: <m2oe9hl0t8.fsf@soma.local> (Marco Baringer's message of "Tue,
	05 Jul 2005 17:39:15 +0200")
References: <87r7edtivo.fsf@flarge.here> <877jg5th5b.fsf@flarge.here>
	<m2oe9hl0t8.fsf@soma.local>
Message-ID: <87slyts0c4.fsf@flarge.here>

"Marco Baringer" <mb at bese.it> writes:

>> Probably it would be even better to call the stop-listening and check
>> thereafter, but at least this patch makes it possible to restart the
>> Araneida backend.
>
> what do you mean?
checking if the socket really is closed. IIRC there are some
escalation level in the port collection which does this. But for the
time beeing the patch allowd to stop the araneida backend and
thereafter start it again.

Regards
Friedrich


From dirk at dirkgerrits.com  Tue Jul  5 20:36:58 2005
From: dirk at dirkgerrits.com (Dirk Gerrits)
Date: Tue, 05 Jul 2005 22:36:58 +0200
Subject: [Bese-devel] going darcs
In-Reply-To: <87vf3ptkqd.fsf@flarge.here>
References: <m21x6eo053.fsf@soma.local> <20050705152205.18c98955@localhost>
	<87vf3ptkqd.fsf@flarge.here>
Message-ID: <42CAEF6A.4050904@dirkgerrits.com>

Friedrich Dominicus wrote:
> Julian Stecklina <der_julian at web.de> writes:
> 
> 
>>I never figured out what I needed to do to get arch to perform something
>>useful, but I was able to use darcs in about 10 minutes...
> 
> Well you are not alone with it, you can do a (incf
> does-not-get-archs-guy) for me ;-)

With Arch, Erlisp was a somewhat lifeless project.  Now with darcs it's 
sponsored by Google in its Summer of Code!

(Okay maybe this had absolutely nothing to do with darcs, but still. ;P)

Kind regards,

Dirk Gerrits, darcs enthusiast


From mb at bese.it  Wed Jul  6 13:42:42 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 06 Jul 2005 15:42:42 +0200
Subject: [Bese-devel] darcs
Message-ID: <m2k6k4giel.fsf@soma.local>


ucw, arnesi and yaclml are now under darcs. the repositories are
located at:

http://common-lisp.net/project/ucw/repos/ucw/
http://common-lisp.net/project/bese/repos/arnesi/
http://common-lisp.net/project/bese/repos/yaclml/

the history has been lost, so it goes.

comments welcome.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From rm at fabula.de  Wed Jul  6 14:26:28 2005
From: rm at fabula.de (rm at fabula.de)
Date: Wed, 6 Jul 2005 16:26:28 +0200
Subject: [Bese-devel] darcs
In-Reply-To: <m2k6k4giel.fsf@soma.local>
References: <m2k6k4giel.fsf@soma.local>
Message-ID: <20050706142628.GE22420@seid-online.de>

On Wed, Jul 06, 2005 at 03:42:42PM +0200, Marco Baringer wrote:
> 
> ucw, arnesi and yaclml are now under darcs. the repositories are
> located at:
> 
> http://common-lisp.net/project/ucw/repos/ucw/
> http://common-lisp.net/project/bese/repos/arnesi/
> http://common-lisp.net/project/bese/repos/yaclml/
> 
> the history has been lost, so it goes.
> 
> comments welcome.

First:  Thanks! Finaly a VCS i don't need to struggle with.
Second: There seems to be a few buglets with the debian package
        support. There's a command to clean the source tree
	before packaging:
	
	find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm -r

	that should either be removed or changed to

	-find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm -r

	The debian/rules file needs to be executable.
	Why does yaclml _depend_ on cl-fiveam? Wouldn't a simple
	'suggests' be more appropriate?


 Cheers Ralf Mattes

> -- 
> -Marco
> Ring the bells that still can ring.
> Forget the perfect offering.
> There is a crack in everything.
> That's how the light gets in.
> 	-Leonard Cohen
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel


From mb at bese.it  Wed Jul  6 14:35:49 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 06 Jul 2005 16:35:49 +0200
Subject: [Bese-devel] darcs
In-Reply-To: <20050706142628.GE22420@seid-online.de> (rm@fabula.de's message
	of "Wed, 6 Jul 2005 16:26:28 +0200")
References: <m2k6k4giel.fsf@soma.local> <20050706142628.GE22420@seid-online.de>
Message-ID: <m2y88kf1dm.fsf@soma.local>

rm at fabula.de writes:

> On Wed, Jul 06, 2005 at 03:42:42PM +0200, Marco Baringer wrote:
>> 
>> ucw, arnesi and yaclml are now under darcs. the repositories are
>> located at:
>> 
>> http://common-lisp.net/project/ucw/repos/ucw/
>> http://common-lisp.net/project/bese/repos/arnesi/
>> http://common-lisp.net/project/bese/repos/yaclml/
>> 
>> the history has been lost, so it goes.
>> 
>> comments welcome.
>
> First:  Thanks! Finaly a VCS i don't need to struggle with.
> Second: There seems to be a few buglets with the debian package
>         support. There's a command to clean the source tree
> 	before packaging:
> 	
> 	find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm -r
>
> 	that should either be removed or changed to
>
> 	-find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm -r

i removed it.

> 	The debian/rules file needs to be executable.

how can i tell darcs to version the executable bit on files?

> 	Why does yaclml _depend_ on cl-fiveam? Wouldn't a simple
> 	'suggests' be more appropriate?

definetly.

for what it's worth i didn't write the debianiziation of yaclml and
have never used it, any other hints/suggestions wolud be appreciated.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From rm at mh-freiburg.de  Wed Jul  6 14:46:57 2005
From: rm at mh-freiburg.de (Le grand pinguin)
Date: Wed, 6 Jul 2005 16:46:57 +0200
Subject: [Bese-devel] darcs
In-Reply-To: <m2y88kf1dm.fsf@soma.local>
References: <m2k6k4giel.fsf@soma.local> <20050706142628.GE22420@seid-online.de>
	<m2y88kf1dm.fsf@soma.local>
Message-ID: <20050706144657.GA8589@forte.mh-freiburg.de>

On Wed, Jul 06, 2005 at 04:35:49PM +0200, Marco Baringer wrote:
> rm at fabula.de writes:
> 
> > On Wed, Jul 06, 2005 at 03:42:42PM +0200, Marco Baringer wrote:
> >> 
> >> ucw, arnesi and yaclml are now under darcs. the repositories are
> >> located at:
> >> 
> >> http://common-lisp.net/project/ucw/repos/ucw/
> >> http://common-lisp.net/project/bese/repos/arnesi/
> >> http://common-lisp.net/project/bese/repos/yaclml/
> >> 
> >> the history has been lost, so it goes.
> >> 
> >> comments welcome.
> >
> > First:  Thanks! Finaly a VCS i don't need to struggle with.
> > Second: There seems to be a few buglets with the debian package
> >         support. There's a command to clean the source tree
> > 	before packaging:
> > 	
> > 	find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm -r
> >
> > 	that should either be removed or changed to
> >
> > 	-find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm -r
> 
> i removed it.
> 
> > 	The debian/rules file needs to be executable.
> 
> how can i tell darcs to version the executable bit on files?
> 
> > 	Why does yaclml _depend_ on cl-fiveam? Wouldn't a simple
> > 	'suggests' be more appropriate?
> 
> definetly.

I just changed it locally and rebuilt the package. Seems to work
flawless. Wow, that's make migration of my app so simple. Just
a (asdf:oos 'asdf:load-op :ucw) - no more twidling with the repository
pathes ... cool.

> for what it's worth i didn't write the debianiziation of yaclml and
> have never used it, any other hints/suggestions wolud be appreciated.

Hmm, i'll have a closer look at them - maybe i'll find some time to
provide a debian/ dir for ucw as well.

 Cheers 

    RalfD
> 
> -- 
> -Marco
> Ring the bells that still can ring.
> Forget the perfect offering.
> There is a crack in everything.
> That's how the light gets in.
> 	-Leonard Cohen
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel


From luca at pca.it  Wed Jul  6 14:33:35 2005
From: luca at pca.it (Luca Capello)
Date: Wed, 06 Jul 2005 16:33:35 +0200
Subject: [Bese-devel] darcs
In-Reply-To: <20050706142628.GE22420@seid-online.de> (rm@fabula.de's message
	of "Wed, 6 Jul 2005 16:26:28 +0200")
References: <m2k6k4giel.fsf@soma.local> <20050706142628.GE22420@seid-online.de>
Message-ID: <87br5gypfk.fsf@gismo.pca.it>

Hello!

On Wed 06 Jul 2005 16:26 +0200, 
> Second: There seems to be a few buglets with the debian package
>         support. There's a command to clean the source tree
> 	before packaging:
> 	
> 	find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm -r
>
> 	that should either be removed or changed to
>
> 	-find debian/$(debpkg)/$(clc-files) -name .arch-ids | xargs rm
> 	-r

Yep, Marco, can you safely remove that line?

> 	The debian/rules file needs to be executable.

Exactly.

> 	Why does yaclml _depend_ on cl-fiveam? Wouldn't a simple
> 	'suggests' be more appropriate?

I discussed this with Marco when I created the Debian patch, but I
don't remember the exact reason (because the dependency was necessary,
not 'suggested' ;-) ).

Anyway, there's a discussion on the Common-Listp Debian mailing-list
about Debian native packages or not and it seems that the second will
be the decision for the future. So, my suggestion is to keep the
situation as today until a correct decision will be taken.

Thx, bye,
Gismo / Luca

PS, I'm sorry I couldn't have finished the preparation of an ucw
Debian package, but I'm a bit busy at University ATM.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 188 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050706/749cb728/attachment.sig>

From ats at acm.org  Wed Jul  6 15:02:35 2005
From: ats at acm.org (Alan Shutko)
Date: Wed, 06 Jul 2005 10:02:35 -0500
Subject: [Bese-devel] darcs
In-Reply-To: <m2k6k4giel.fsf@soma.local> (Marco Baringer's message of "Wed, 06
	Jul 2005 15:42:42 +0200")
References: <m2k6k4giel.fsf@soma.local>
Message-ID: <87br5gj7uc.fsf@vera.springies.com>

"Marco Baringer" <mb at bese.it> writes:

> ucw, arnesi and yaclml are now under darcs. the repositories are
> located at:

And I'd actually figured out tla, mostly.  8^(  CVS sucked, but it
seems it sucks less than everything else.

-- 
Alan Shutko <ats at acm.org> - I am the rocks.
Clothes make the man.


From digash at gmail.com  Wed Jul  6 21:57:39 2005
From: digash at gmail.com (Dimitry Gashinsky)
Date: Wed, 6 Jul 2005 17:57:39 -0400
Subject: [Bese-devel] darcs
In-Reply-To: <m2y88kf1dm.fsf@soma.local>
References: <m2k6k4giel.fsf@soma.local> <20050706142628.GE22420@seid-online.de>
	<m2y88kf1dm.fsf@soma.local>
Message-ID: <e2c76ba50507061457691e66c8@mail.gmail.com>

On 7/6/05, Marco Baringer <mb at bese.it> wrote:
> rm at fabula.de writes:
> 
> > On Wed, Jul 06, 2005 at 03:42:42PM +0200, Marco Baringer wrote:
> >>
> >> ucw, arnesi and yaclml are now under darcs. the repositories are
> >> located at:
> >>
> >> http://common-lisp.net/project/ucw/repos/ucw/
> >> http://common-lisp.net/project/bese/repos/arnesi/
> >> http://common-lisp.net/project/bese/repos/yaclml/
> >>
> >> the history has been lost, so it goes.
> >>
> >> comments welcome.

I think tla is not that difficult. I've understood it after using
darcs for some time :)
I noticed you used older version of tla. Maybe that was the cause of
your problems? I would love to hear a postmortem analysis since I am
pretty happy with arch and use it for a lot of projects.

> how can i tell darcs to version the executable bit on files?

You can't.
http://darcs.net/DarcsWiki/FrequentlyAskedQuestions#head-f6690ca8f72ee57532b9cf5e74147ae92f0be62a

Regards,
DiG
-- 
((lambda (lambda) `(,lambda ',lambda)) '(lambda (lambda) `(,lambda ',lambda)))


From drewc at tech.coop  Thu Jul  7 19:48:37 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Thu, 07 Jul 2005 12:48:37 -0700
Subject: [Bese-devel] darcs
In-Reply-To: <m2k6k4giel.fsf@soma.local>
References: <m2k6k4giel.fsf@soma.local>
Message-ID: <42CD8715.30108@tech.coop>

Marco Baringer wrote:
> ucw, arnesi and yaclml are now under darcs.



Yay! I set up darcs.cgi for my branches, which are currently identical 
to upstream :) check:

http://versions.tech.coop/cgi-bin/darcs.cgi

Going to http://versions.tech.coop/ucw/ will also allow you to browse 
the source as it is on disk.

I plan on always keeping "fairly stable" versions of the packages there. 
In any case it will be the branches i use.

Enjoy!

drewc


  the repositories are
> located at:
> 
> http://common-lisp.net/project/ucw/repos/ucw/
> http://common-lisp.net/project/bese/repos/arnesi/
> http://common-lisp.net/project/bese/repos/yaclml/
> 
> the history has been lost, so it goes.
> 
> comments welcome.
> 



From frido at q-software-solutions.de  Thu Jul  7 12:27:15 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Thu, 07 Jul 2005 14:27:15 +0200
Subject: [Bese-devel] integer-range-field
Message-ID: <87vf3mby3g.fsf@flarge.here>

Is it on purpose that it's not exported from ucw or is it
a bug?

Regards
Friedrich


From frido at q-software-solutions.de  Thu Jul  7 12:46:30 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Thu, 07 Jul 2005 14:46:30 +0200
Subject: [Bese-devel] purpuse of :www-root
Message-ID: <87r7eabx7d.fsf@flarge.here>

as initargument. The sources do not tell anything about it.

Regards
Friedrich


From frido at q-software-solutions.de  Thu Jul  7 13:53:24 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Thu, 07 Jul 2005 15:53:24 +0200
Subject: [Bese-devel] Next trouble striking
Message-ID: <87irzmbu3v.fsf@flarge.here>

Again a simple problem. See the following code
(defcomponent foo-bar (form-component)
  ((num :accessor num
        :component (integer-field)))
  (:metaclass standard-component-class))

(defclass product ()
          ((tag :initarg :tag)
           (description :initarg :description)))
          

(defmethod render-on ((res response)(foo foo-bar))
  (let ((stream (content-stream res)))
    (write-main-menu :active +home-index+ :stream stream)
    (let ((product (make-instance 'product 
                                  :tag "foo"
                                  :description "foo-bar")))
      (<:h1 "Just testing")
      
      (<ucw:form :action (submit foo)
                 (<:table
                  (<:tr  
                   (<:td  (render-on res (num foo-bar)))
                   (<:td (<:as-html (tag product)))))))
    (write-footer :stream stream)))

(defentry-point "index.ucw" (:application *shop-application*) ()
  (call 'foo-bar))




(defparameter *shop-application*
  (make-instance 'cookie-session-application
                 :url-prefix "/ucw/shop/"
                 :tal-generator (make-instance 'yaclml:file-system-generator
                                               :cachep t
                                               :root-directories (list *ucw-tal-root*))))



Now I expected some output while doing getting to "index.ucw"

However I got:
(defentry-point "index.ucw" (:application *shop-application*) ()
  (call 'foo-bar))




(defparameter *shop-application*
  (make-instance 'cookie-session-application
                 :url-prefix "/ucw/shop/"
                 :tal-generator (make-instance 'yaclml:file-system-generator
                                               :cachep t
                                               :root-directories (list *ucw-tal-root*))))


What am I doing wrong?


What I want is:

Getting an start page with an 
drop-down box containing numbers from 1-20 and 
second a short rendering of let's say the tag part of the product
class.

After that I probably want to do something and after that I want to
switch to asking for address details, payment details etc.

In short I want to use UCW for implementing a simply e-shop solution. 

If anyone has used UCW in that area please give me a hand. I'm getting
stuck too often..

Regards
Friedrich


From mb at bese.it  Thu Jul  7 15:20:06 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 07 Jul 2005 17:20:06 +0200
Subject: [Bese-devel] integer-range-field
In-Reply-To: <87vf3mby3g.fsf@flarge.here> (Friedrich Dominicus's message of
	"Thu, 07 Jul 2005 14:27:15 +0200")
References: <87vf3mby3g.fsf@flarge.here>
Message-ID: <m28y0id4nt.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> Is it on purpose that it's not exported from ucw or is it
> a bug?

bug. 

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul  7 15:20:31 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 07 Jul 2005 17:20:31 +0200
Subject: [Bese-devel] purpuse of :www-root
In-Reply-To: <87r7eabx7d.fsf@flarge.here> (Friedrich Dominicus's message of
	"Thu, 07 Jul 2005 14:46:30 +0200")
References: <87r7eabx7d.fsf@flarge.here>
Message-ID: <m24qb6d4n4.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> as initargument. The sources do not tell anything about it.

a docstring is on its way.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Fri Jul  8 09:18:57 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Fri, 08 Jul 2005 11:18:57 +0200
Subject: [Bese-devel] Next trouble striking
In-Reply-To: <87irzmbu3v.fsf@flarge.here> (Friedrich Dominicus's message of
	"Thu, 07 Jul 2005 15:53:24 +0200")
References: <87irzmbu3v.fsf@flarge.here>
Message-ID: <87y88hfyf2.fsf@flarge.here>

Ok found that I had to add simpl-window-component to the parents
list. But now I do not understand why that is needed. 

Regards
Friedrich


From mb at bese.it  Fri Jul  8 19:09:15 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 08 Jul 2005 21:09:15 +0200
Subject: [Bese-devel] Next trouble striking
References: <87irzmbu3v.fsf@flarge.here> <87y88hfyf2.fsf@flarge.here>
Message-ID: <m2d5pt6rok.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> Ok found that I had to add simpl-window-component to the parents
> list. But now I do not understand why that is needed. 

my guess (but it's just a guess) is that without the
<html><head>...</head><body> and </body></html> tags, which are
rendered by the simple-window-componet's render-on method, your
browser didn't know what to make of the html and gave you a white
page.

could you compare the html actualy generated by the two setups?



-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Fri Jul  8 19:22:04 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 08 Jul 2005 21:22:04 +0200
Subject: [Bese-devel] call/cc interpreter
Message-ID: <m2k6k15cir.fsf@soma.local>


hi,

  i've put a darcs repo containing the current common-lisp+call/cc
  interpreter up at:

http://common-lisp.net/project/bese/repos/arnesi_cps/

(there's no documentation as of yet, the interesting files are
walk.lisp, cps.lisp and lexenv.lisp, the api hasn't changed (much) but
the implementation is completely different).

i haven't tried running ucw with this code yet, though it should be
almost ready. i'm going to be away from lisp, internet and computers
in general for a while, if anyway wants to have a look at it and try
running ucw with it...

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Sat Jul  9 06:09:56 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 09 Jul 2005 08:09:56 +0200
Subject: [Bese-devel] Next trouble striking
In-Reply-To: <m2d5pt6rok.fsf@soma.local> (Marco Baringer's message of "Fri,
	08 Jul 2005 21:09:15 +0200")
References: <87irzmbu3v.fsf@flarge.here> <87y88hfyf2.fsf@flarge.here>
	<m2d5pt6rok.fsf@soma.local>
Message-ID: <87ll4gfr2j.fsf@flarge.here>

"Marco Baringer" <mb at bese.it> writes:

> Friedrich Dominicus <frido at q-software-solutions.de> writes:
>
>> Ok found that I had to add simpl-window-component to the parents
>> list. But now I do not understand why that is needed. 
>
> my guess (but it's just a guess) is that without the
> <html><head>...</head><body> and </body></html> tags, which are
> rendered by the simple-window-componet's render-on method, your
> browser didn't know what to make of the html and gave you a white
> page.
That seems to be a reasonable guess. 


>
> could you compare the html actualy generated by the two setups?
Well I checked it makes a difference. 

It's a difficult to know all the relationships of the classes for me
the -component suggest that both share some common parent. 

They do partly but that's was not sufficient.

Regards
Friedrich



From a_bakic at yahoo.com  Sat Jul  9 23:54:08 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sat, 9 Jul 2005 16:54:08 -0700 (PDT)
Subject: [Bese-devel] "GET /ucw/js/UCWInputCompleter.js HTTP/1.1" 404 -1
Message-ID: <20050709235408.27024.qmail@web40613.mail.yahoo.com>

Hi,

I am getting the error in the subject while using SBCL and aserve backend.

Regards,
Alex


		
____________________________________________________
Sell on Yahoo! Auctions ? no fees. Bid on great items.  
http://auctions.yahoo.com/


From a_bakic at yahoo.com  Sun Jul 10 23:06:56 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sun, 10 Jul 2005 16:06:56 -0700 (PDT)
Subject: [Bese-devel] ucw-toplevel in make-image.lisp
Message-ID: <20050710230657.5413.qmail@web40627.mail.yahoo.com>

Hi,

The dumped core works only if lisp is later started in the same directory. This
is because *default-pathname-defaults* is not captured, being special. You need
something like this:

(let ((dpd *default-pathname-defaults*))
  (defun ucw-toplevel ()
    (setf *ucw-tal-root* (merge-pathnames "./wwwroot/" dpd))
    ... ; also merge with dpd whenever truename is called

Alex


		
____________________________________________________
Sell on Yahoo! Auctions ? no fees. Bid on great items.  
http://auctions.yahoo.com/


From frido at q-software-solutions.de  Tue Jul 12 08:42:31 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 12 Jul 2005 10:42:31 +0200
Subject: [Bese-devel] Usage of <ucw:select?
Message-ID: <871x64jtzc.fsf@flarge.here>

Well that's the point. What can be fed to that tag and how it is
supposed to be used?

Regards
Friedrich


From frido at q-software-solutions.de  Tue Jul 12 09:13:42 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 12 Jul 2005 11:13:42 +0200
Subject: [Bese-devel] Probably I have to be more specific
Message-ID: <87wtnwidyx.fsf@flarge.here>

Ok here we go. 
how to use 
ucw:select 

with Radiobuttons?

AFAIU the supported think is the "options" input way. Is that correct?

Regards
Friedrich


From frido at q-software-solutions.de  Tue Jul 12 16:36:09 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 12 Jul 2005 18:36:09 +0200
Subject: [Bese-devel] Probably I have to be more specific
In-Reply-To: <87wtnwidyx.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 12 Jul 2005 11:13:42 +0200")
References: <87wtnwidyx.fsf@flarge.here>
Message-ID: <87k6jwgex2.fsf@flarge.here>

Ok, I got a bit further. And am now quite sure that there is no 
special component for Radio Buttons. Now it seems to me that having
this would be a good thing. I found out that I can help myself while
using this ugly piece of code:
(ucw::get-parameter (ucw::context.request *context*) 'download)))

It's ugly because it uses the internal stuff. And now I wonder why
e.g get-parameter not exported? If there is not component available
then it seems to be a good idea to give uses an alternative
(documented way).

However I wonder how a radio box component might look. My idea would be
it can be put into a div or a span 
the layout could be vertical or horizontal
it could be grouped by a border

It seems it shares a lot with the select-field, but it's not fully the
same. 

Now my question is if someone had an idea on how to implement such
component.

I would like to know for what the following slots are used:
  ((options :accessor options :initform '() :initarg :options)
   (key :accessor key :initform #'identity :initarg :key)
   (test :accessor test :initform #'eql :initarg :test)
   (option-map :accessor option-map :initform (make-array 10 :adjustable t :fill-pointer 0))
   (option-writer :accessor option-writer :initform #'princ-to-string))

I imagine options is for an alist or hash table or the like so maybe
this would be ok to fill into this slot

(list (identifier-of-the-radio-button (list "download" "CD-ROM"
"other")))
....

test probably could be eql as here 
but what's the option map?
and how to access the choosen value? 

Of course with get-paramter I can get the value. But it seems some
kind of accessor would be a better idea?

Has anyone written any component by him/herself? 

If yes, can you give me a helping hand on how to approach this?

Regards
Friedrich



From frido at q-software-solutions.de  Wed Jul 13 06:28:57 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Wed, 13 Jul 2005 08:28:57 +0200
Subject: [Bese-devel] Probably I have to be more specific
In-Reply-To: <87ll4bsjr7.fsf@matley.muppetslab.org> (Luigi Panzeri's message
	of "Wed, 13 Jul 2005 01:10:52 +0200")
References: <87wtnwidyx.fsf@flarge.here> <87k6jwgex2.fsf@flarge.here>
	<87ll4bsjr7.fsf@matley.muppetslab.org>
Message-ID: <87y88bte1i.fsf@flarge.here>

Luigi Panzeri <matley at muppetslab.org> writes:

> Friedrich Dominicus <frido at q-software-solutions.de> writes:
>
>> Ok, I got a bit further. And am now quite sure that there is no 
>> special component for Radio Buttons. Now it seems to me that having
>> this would be a good thing. I found out that I can help myself while
>> using this ugly piece of code:
>> (ucw::get-parameter (ucw::context.request *context*) 'download)))
>>
>
> Mmmm, but for radio buttons, dont you have to use <ucw:input :type
> "radio" as in html?
Yes that's what one has to use. However I used:
    (<:input :type "radio" :name 'download :value "Download"
"Download")



>
> For further documentation on using custom field, see
> src/yaclml/ucw-tags.lisp
Well I better check this file I guess.

> But maybe i dont understand well your question, because perhaps you
> want a library component for radio buttons, as for select-field and
> option-dialog. That is a nice idea.
Yes that was my idea.

>
> (defclass radio-button-select (form-element)
> ...
> ...
> )
>
> (defmethod render-on ((res response) (r radio-button-select))
>  (iterate
>   (for o in (options r))
>   (for index upfrom 0)
>   (vector-push-extend o option-map)
>   (<ucw:input :type "radio" :value index 
>                             :name (make-new-callback (context.current-frame *context*)
>                                      (lambda (v) (setf (client-value r) v))))
>   (funcall (option-writer o) o)))
>
> (defmethod read-client-value ((r radio-button-select))
>   ...
>   (setf lisp-value (aref option-map (parse-integer (client-value r)))))
>
>
> </untested>
>
>
>
> I hope to had given you an help. Sorry for my terrible english, but
>i am tired now.
Thanks for taking the time, will see if and how I can make use of it.

Regards
Friedrich


From lispweb at mac.e4ward.com  Wed Jul 13 22:24:50 2005
From: lispweb at mac.e4ward.com (Mac Chan)
Date: Wed, 13 Jul 2005 15:24:50 -0700
Subject: [Bese-devel] UCW file upload example
Message-ID: <E1Dspfs-0002Qm-EN@e4ward.com>

Hi,

I was unable to run the file upload example. Upon playing around
with the sample code for a while I figured it out.

Please see the attached diff.

I'm building a site for teachers to upload multimedia contents which
can be upward of 100mb+ files. The way ucw stores uploaded content in
memory cannot cut it.

TBNL implemented a hack (write-content-to-file t) where they put a
hook in cl-rfc2388 to allow writing uploaded content to file directly.
Since it's a hack (TBNL manages the file creation, uploaded contents
itself) for TBNL only, it is not usable for other frameworks.

Even if it does, I do not have a good idea of how to incorporate 
this into ucw cleanly.

Right now code in UCW can access the file content with slot-value
directly, which is very simple but it doesn't scale for large files.

If we put a pathname there instead, then everyone that access the slot
will have to examine the tag first (much like the bug fix I submit
below), which complicates things.

Another drawback is when session expires, those files need to be
cleaned up. Right now GC handles that and if we put the pathname in
there, ucw has to do the cleanup.

I'm soliciting opinions how to architech this as I'm not all too
familiar with the underneath framework.

(Araneida right now doesn't support multipart MIME form post.
aserve however do support this.)


BTW There are certain things that can be done in a more straightforward way
in aserve / TBNL / araneida.

I think that UCW cannot share objects with Araneida or other backend directly.
Can we hack both and so that they share a unique session id and some hooks
to access objects created in both sides?

In that case I might be able to handle the boring file upload in
aserve and let ucw handle all the other cool stuff :-)

Does anyone have any experience to share, regarding using both
UCW and other backends in your app?

I heard that drewc runs cliki & ucw in the same image. I wonder
how that works.

-- nsr

Index: httpd.lisp
===================================================================
--- httpd.lisp  (revision 90)
+++ httpd.lisp  (revision 91)
@@ -507,12 +507,16 @@
                      :element-type (array-element-type line)))))

 (defmethod get-parameter ((request httpd-request) name)
-  (let ((cons (assoc name (parameters request) :test #'string=)))
-    (when cons (copy-seq (cdr cons)))))
+  (let* ((cons (assoc name (parameters request) :test #'string=))
+         (value (cdr cons)))
+    (when cons (copy-seq (if (mime-part-p value)
+                             (mime-part-body value) value)))))

 (defmethod map-parameters ((request httpd-request) lambda)
  (dolist* ((name . value) (parameters request))
-    (funcall lambda name (copy-seq value))))
+    (funcall lambda name (copy-seq
+                          (if (mime-part-p value)
+                              (mime-part-body value) value)))))

 (defun read-request-headers (backend stream)
  (iterate
------------------------------------------->8--------------------------------------

Index: examples/examples.lisp
===================================================================
--- examples/examples.lisp      (revision 91)
+++ examples/examples.lisp      (revision 94)
@@ -36,6 +36,8 @@
      (add-component 'counter)
      (add-component 'transaction-example)
      (add-component 'example-form)
+      (add-component 'file-upload-example)
+      (add-component 'file-upload-viewer)
      (add-component 'example-welcome)
      (add-component 'sum)
      (add-component 'presentations-example)
@@ -49,6 +51,7 @@
        (<:li (example-link 'counter "Counter"))
       (<:li (example-link 'transaction-example "Component Transaction"))
       (<:li (example-link 'example-form "Form demo"))
+       (<:li (example-link 'file-upload-example "Form upload example"))
       (<:li (example-link 'presentations-example "Presentations Example"))
       (<:li (example-link 'sum "Add some numbers."))
       (<:li (<ucw:a :action (error "FOO") "Signal an error")))))
Index: examples/forms.lisp
===================================================================
--- examples/forms.lisp (revision 91)
+++ examples/forms.lisp (revision 94)
@@ -7,7 +7,7 @@
 (defclass file-upload-example (widget-component template-component)
  ((file :accessor file-upload-example.file :initform ""))
  (:metaclass standard-component-class)
-  (:default-initargs :template-name "upload.tal")
+  (:default-initargs :template-name "ucw/examples/upload.tal")
  (:documentation "Form for uploading a file."))


From frido at q-software-solutions.de  Sat Jul 16 08:17:21 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 16 Jul 2005 10:17:21 +0200
Subject: [Bese-devel] Extension to SBCL 0.9.2 sucking memory.
Message-ID: <87pstjp3la.fsf@flarge.here>

It seems sbcl can not grok this code, see below:

(file xml-name-rune-p.lisp)

system AMD64
Debian/Linux 2.6.7
SBCL 0.9.2.35

The memory concumption explodes and well the CPU is nearly monoplized
;-(

Regards
Friedrich



(in-package :xml)

#.(funcall 
   (compile 
    nil
    '(lambda ()
      (let ((*max* #xD800))
        (labels
            ((name-start-rune-p (rune)
               (or (letter-rune-p rune)
                   (= #.(char-code #\_) rune)
                   (= #.(char-code #\:) rune)))

             (name-rune-p (rune)
               (or (letter-rune-p rune)
                   (digit-rune-p* rune)
                   (= rune #.(char-code #\.))
                   (= rune #.(char-code #\-))
                   (= rune #.(char-code #\_))
                   (= rune #.(char-code #\:))
                   (combining-rune-p rune)
                   (extender-rune-p rune)))

             (letter-rune-p (rune)
               (or (base-rune-p rune)
                   (ideographic-rune-p rune)))

             (digit-rune-p* (rune)
               (or (<= 48 rune 57)
                   (<= 1632 rune 1641)
                   (<= 1776 rune 1785)
                   (<= 2406 rune 2415)
                   (<= 2534 rune 2543)
                   (<= 2662 rune 2671)
                   (<= 2790 rune 2799)
                   (<= 2918 rune 2927)
                   (<= 3047 rune 3055)
                   (<= 3174 rune 3183)
                   (<= 3302 rune 3311)
                   (<= 3430 rune 3439)
                   (<= 3664 rune 3673)
                   (<= 3792 rune 3801)
                   (<= 3872 rune 3881)))


             (combining-rune-p (rune)
               (or (<= 768 rune 837)
                   (<= 864 rune 865)
                   (<= 1155 rune 1158)
                   (<= 1425 rune 1441)
                   (<= 1443 rune 1465)
                   (<= 1467 rune 1469)
                   (= 1471 rune)
                   (<= 1473 rune 1474)
                   (= 1476 rune)
                   (<= 1611 rune 1618)
                   (= 1648 rune)
                   (<= 1750 rune 1756)
                   (<= 1757 rune 1759)
                   (<= 1760 rune 1764)
                   (<= 1767 rune 1768)
                   (<= 1770 rune 1773)
                   (<= 2305 rune 2307)
                   (= 2364 rune)
                   (<= 2366 rune 2380)
                   (= 2381 rune)
                   (<= 2385 rune 2388)
                   (<= 2402 rune 2403)
                   (<= 2433 rune 2435)
                   (= 2492 rune)
                   (= 2494 rune)
                   (= 2495 rune)
                   (<= 2496 rune 2500)
                   (<= 2503 rune 2504)
                   (<= 2507 rune 2509)
                   (= 2519 rune)
                   (<= 2530 rune 2531)
                   (= 2562 rune)
                   (= 2620 rune)
                   (= 2622 rune)
                   (= 2623 rune)
                   (<= 2624 rune 2626)
                   (<= 2631 rune 2632)
                   (<= 2635 rune 2637)
                   (<= 2672 rune 2673)
                   (<= 2689 rune 2691)
                   (= 2748 rune)
                   (<= 2750 rune 2757)
                   (<= 2759 rune 2761)
                   (<= 2763 rune 2765)
                   (<= 2817 rune 2819)
                   (= 2876 rune)
                   (<= 2878 rune 2883)
                   (<= 2887 rune 2888)
                   (<= 2891 rune 2893)
                   (<= 2902 rune 2903)
                   (<= 2946 rune 2947)
                   (<= 3006 rune 3010)
                   (<= 3014 rune 3016)
                   (<= 3018 rune 3021)
                   (= 3031 rune)
                   (<= 3073 rune 3075)
                   (<= 3134 rune 3140)
                   (<= 3142 rune 3144)
                   (<= 3146 rune 3149)
                   (<= 3157 rune 3158)
                   (<= 3202 rune 3203)
                   (<= 3262 rune 3268)
                   (<= 3270 rune 3272)
                   (<= 3274 rune 3277)
                   (<= 3285 rune 3286)
                   (<= 3330 rune 3331)
                   (<= 3390 rune 3395)
                   (<= 3398 rune 3400)
                   (<= 3402 rune 3405)
                   (= 3415 rune)
                   (= 3633 rune)
                   (<= 3636 rune 3642)
                   (<= 3655 rune 3662)
                   (= 3761 rune)
                   (<= 3764 rune 3769)
                   (<= 3771 rune 3772)
                   (<= 3784 rune 3789)
                   (<= 3864 rune 3865)
                   (= 3893 rune)
                   (= 3895 rune)
                   (= 3897 rune)
                   (= 3902 rune)
                   (= 3903 rune)
                   (<= 3953 rune 3972)
                   (<= 3974 rune 3979)
                   (<= 3984 rune 3989)
                   (= 3991 rune)
                   (<= 3993 rune 4013)
                   (<= 4017 rune 4023)
                   (= 4025 rune)
                   (<= 8400 rune 8412)
                   (= 8417 rune)
                   (<= 12330 rune 12335)
                   (= 12441 rune)
                   (= 12442 rune)))

             (extender-rune-p (rune)
               (or
                (= 183 rune)
                (= 720 rune)
                (= 721 rune)
                (= 903 rune)
                (= 1600 rune)
                (= 3654 rune)
                (= 3782 rune)
                (= 12293 rune)
                (<= 12337 rune 12341)
                (<= 12445 rune 12446)
                (<= 12540 rune 12542)))

             (base-rune-p (rune)
               (or 
                (<= 65 rune 90) (<= 97 rune 122) (<= 192 rune 214) (<= 216 rune 246) (<= 248 rune 255) (<= 256 rune 305)
                (<= 308 rune 318) (<= 321 rune 328) (<= 330 rune 382) (<= 384 rune 451) (<= 461 rune 496) (<= 500 rune 501)
                (<= 506 rune 535) (<= 592 rune 680) (<= 699 rune 705) (= 902 rune) (<= 904 rune 906) (= 908 rune)
                (<= 910 rune 929) (<= 931 rune 974) (<= 976 rune 982) (= 986 rune) (= 988 rune) (= 990 rune) (= 992 rune)
                (<= 994 rune 1011) (<= 1025 rune 1036) (<= 1038 rune 1103) (<= 1105 rune 1116) (<= 1118 rune 1153)
                (<= 1168 rune 1220) (<= 1223 rune 1224) (<= 1227 rune 1228) (<= 1232 rune 1259) (<= 1262 rune 1269)
                (<= 1272 rune 1273) (<= 1329 rune 1366) (= 1369 rune) (<= 1377 rune 1414) (<= 1488 rune 1514)
                (<= 1520 rune 1522) (<= 1569 rune 1594) (<= 1601 rune 1610) (<= 1649 rune 1719) (<= 1722 rune 1726)
                (<= 1728 rune 1742) (<= 1744 rune 1747) (= 1749 rune) (<= 1765 rune 1766) (<= 2309 rune 2361) (= 2365 rune)
                (<= 2392 rune 2401) (<= 2437 rune 2444) (<= 2447 rune 2448) (<= 2451 rune 2472) (<= 2474 rune 2480)
                (= 2482 rune) (<= 2486 rune 2489) (<= 2524 rune 2525) (<= 2527 rune 2529) (<= 2544 rune 2545)
                (<= 2565 rune 2570) (<= 2575 rune 2576) (<= 2579 rune 2600) (<= 2602 rune 2608) (<= 2610 rune 2611)
                (<= 2613 rune 2614) (<= 2616 rune 2617) (<= 2649 rune 2652) (= 2654 rune) (<= 2674 rune 2676)
                (<= 2693 rune 2699) (= 2701 rune) (<= 2703 rune 2705) (<= 2707 rune 2728) (<= 2730 rune 2736)
                (<= 2738 rune 2739) (<= 2741 rune 2745) (= 2749 rune) (= 2784 rune) (<= 2821 rune 2828) (<= 2831 rune 2832)
                (<= 2835 rune 2856) (<= 2858 rune 2864) (<= 2866 rune 2867) (<= 2870 rune 2873) (= 2877 rune)
                (<= 2908 rune 2909) (<= 2911 rune 2913) (<= 2949 rune 2954) (<= 2958 rune 2960) (<= 2962 rune 2965)
                (<= 2969 rune 2970) (= 2972 rune) (<= 2974 rune 2975) (<= 2979 rune 2980) (<= 2984 rune 2986)
                (<= 2990 rune 2997) (<= 2999 rune 3001) (<= 3077 rune 3084) (<= 3086 rune 3088) (<= 3090 rune 3112)
                (<= 3114 rune 3123) (<= 3125 rune 3129) (<= 3168 rune 3169) (<= 3205 rune 3212) (<= 3214 rune 3216)
                (<= 3218 rune 3240) (<= 3242 rune 3251) (<= 3253 rune 3257) (= 3294 rune) (<= 3296 rune 3297)
                (<= 3333 rune 3340) (<= 3342 rune 3344) (<= 3346 rune 3368) (<= 3370 rune 3385) (<= 3424 rune 3425)
                (<= 3585 rune 3630) (= 3632 rune) (<= 3634 rune 3635) (<= 3648 rune 3653) (<= 3713 rune 3714) (= 3716 rune)
                (<= 3719 rune 3720) (= 3722 rune) (= 3725 rune) (<= 3732 rune 3735) (<= 3737 rune 3743) (<= 3745 rune 3747)
                (= 3749 rune) (= 3751 rune) (<= 3754 rune 3755) (<= 3757 rune 3758) (= 3760 rune) (<= 3762 rune 3763) (= 3773 rune)
                (<= 3776 rune 3780) (<= 3904 rune 3911) (<= 3913 rune 3945) (<= 4256 rune 4293) (<= 4304 rune 4342)
                (= 4352 rune) (<= 4354 rune 4355) (<= 4357 rune 4359) (= 4361 rune) (<= 4363 rune 4364) (<= 4366 rune 4370)
                (= 4412 rune) (= 4414 rune) (= 4416 rune) (= 4428 rune) (= 4430 rune) (= 4432 rune) (<= 4436 rune 4437) (= 4441 rune)
                (<= 4447 rune 4449) (= 4451 rune) (= 4453 rune) (= 4455 rune) (= 4457 rune) (<= 4461 rune 4462) (<= 4466 rune 4467)
                (= 4469 rune) (= 4510 rune) (= 4520 rune) (= 4523 rune) (<= 4526 rune 4527) (<= 4535 rune 4536) (= 4538 rune)
                (<= 4540 rune 4546) (= 4587 rune) (= 4592 rune) (= 4601 rune) (<= 7680 rune 7835) (<= 7840 rune 7929)
                (<= 7936 rune 7957) (<= 7960 rune 7965) (<= 7968 rune 8005) (<= 8008 rune 8013) (<= 8016 rune 8023)
                (= 8025 rune) (= 8027 rune) (= 8029 rune) (<= 8031 rune 8061) (<= 8064 rune 8116) (<= 8118 rune 8124) (= 8126 rune)
                (<= 8130 rune 8132) (<= 8134 rune 8140) (<= 8144 rune 8147) (<= 8150 rune 8155) (<= 8160 rune 8172)
                (<= 8178 rune 8180) (<= 8182 rune 8188) (= 8486 rune) (<= 8490 rune 8491) (= 8494 rune) (<= 8576 rune 8578)
                (<= 12353 rune 12436) (<= 12449 rune 12538) (<= 12549 rune 12588) (<= 44032 rune 55203)))

             (ideographic-rune-p (rune)
               (or (<= 19968 rune 40869) (= 12295 rune) (<= 12321 rune 12329)))


             (predicate-to-bv (p)
               (let ((r (make-array *max* :element-type 'bit :initial-element 0)))
                 (dotimes (i #x10000 r)
                   (when (funcall p i)
                     (setf (aref r i) 1))))) )

          `(progn
             (DEFSUBST NAME-RUNE-P (RUNE)
               (AND (<= 0 RUNE ,*max*)
                    (LOCALLY (DECLARE (OPTIMIZE (SAFETY 0) (SPEED 3)))
                             (= 1 (SBIT ',(predicate-to-bv #'name-rune-p)
                                        (THE FIXNUM RUNE))))))
             (DEFSUBST NAME-START-RUNE-P (RUNE)
               (AND (<= 0 RUNE ,*MAX*)
                    (LOCALLY (DECLARE (OPTIMIZE (SAFETY 0) (SPEED 3)))
                             (= 1 (SBIT ',(predicate-to-bv #'name-start-rune-p)
                                        (THE FIXNUM RUNE)))))))) ))))


From frido at q-software-solutions.de  Sat Jul 16 13:16:18 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 16 Jul 2005 15:16:18 +0200
Subject: [Bese-devel] Extension to SBCL 0.9.2 sucking memory.
In-Reply-To: <87pstjp3la.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sat, 16 Jul 2005 10:17:21 +0200")
References: <87pstjp3la.fsf@flarge.here>
Message-ID: <878y06q4bh.fsf@flarge.here>

Sorry, obviously the wrong newgroup I'm sorry.

Regards
Friedrich


From aml at gia.ist.utl.pt  Sat Jul  9 10:03:55 2005
From: aml at gia.ist.utl.pt (Antonio Menezes Leitao)
Date: Sat, 09 Jul 2005 11:03:55 +0100
Subject: [Bese-devel] Bug in aserve backend
Message-ID: <87mzowthx0.fsf@gia.ist.utl.pt>

Hi,

I noticed that, in ucw's aserve backend, the implementation of the
getter and setter of the response headers is:

(defmethod get-header ((response aserve-response) header-name)
  (net.aserve:reply-header-slot-value (request response) header-name))

(defmethod (setf get-header) (value (response aserve-response) header-name)
  (setf (net.aserve:reply-header-slot-value (request response) header-name)
	value))

We call these accessors as, e.g.,

  (setf (get-header response "Status") "500"
        (get-header response "Content-Type") "text/html")

However, according to the aserve documentation (also valid for
reply-header-slot-value):

(header-slot-value request header-name)

    Return the value given in the request for the given header-name (a
    keyword symbol).  If the header wasn't present in this request
    then nil will be returned.  header-slot-value is a macro that will
    expand into a fast accessor if the header-name is a constant
    naming a known header slot
    
    In older versions of aserve the header-name was a string..

It is not very well explained but the code is clear: the function uses
an assoc-list to find the header, but searches using #'eq.  This means
that (setf get-header) will not redefine an existing header but will
accumulate identical headers, instead.

However, it's even worse: some of the headers (the so called
'fast-reply-headers') are not stored in the assoc-list but in the
request object itself, namely 'date', 'content-type' and
'content-length'. Even more worse (if possible) is the fact that
net.aserve:reply-header-slot-value is a macro that can only work
properly with the fast-reply-headers if they are visible at macro
expansion time.

To solve this problem, I suggest that the methods:

(defmethod get-header ((response aserve-response) header-name)
  (net.aserve:reply-header-slot-value (request response) header-name))

(defmethod (setf get-header) (value (response aserve-response) header-name)
  (setf (net.aserve:reply-header-slot-value (request response) header-name)
        value))

should be replaced by something like:

(defmethod get-header ((response aserve-response) header-name)
  (switch (header-name :test #'string-equal)
    ("Date"             (net.aserve:reply-header-slot-value (request response) :date))
    ("Content-Type"     (net.aserve:reply-header-slot-value (request response) :content-type))
    ("Content-Length"   (net.aserve:reply-header-slot-value (request response) :content-length))
    (t (net.aserve:reply-header-slot-value
	(request response)
	(read-from-string (format nil ":~A" header-name))))))

(defmethod (setf get-header) (value (response aserve-response) header-name)
  (switch (header-name :test #'string-equal)
    ("Date"             (error "Can't update header Date"))
    ("Content-Type"     (setf (net.aserve:reply-header-slot-value (request response) :content-type)   value))
    ("Content-Length"   (setf (net.aserve:reply-header-slot-value (request response) :content-length) value))
    (t (setf (net.aserve:reply-header-slot-value
	      (request response)
	      (read-from-string (format nil ":~A" header-name)))
	     value))))

The identical methods for aserve-request should require a similar
correction.

Best regards,

Ant?nio Leit?o.


From aml at gia.ist.utl.pt  Sat Jul  9 11:16:18 2005
From: aml at gia.ist.utl.pt (Antonio Menezes Leitao)
Date: Sat, 09 Jul 2005 12:16:18 +0100
Subject: [Bese-devel] Bug in aserve backend for redirect component
Message-ID: <878y0gtekd.fsf@gia.ist.utl.pt>

Hi,

Once again, aserve macros (this time with-http-response) require more
information at macro expansion time than they should.  This affects
UCW's redirect-component as it attempts to set the Status header to
"302" but aserve doesn't use it because the macro expansion requires
that information as an explicit :response parameter with corresponding
argument given by, e.g., *response-moved-permanently*,
*response-found*, etc.  The function code-to-response returns the
appropriate value for a given status.

I replaced the method 

(defmethod shutdown ((response aserve-response))
  (net.aserve:with-http-response ((request response) (entity response))
    (net.aserve:with-http-body ((request response) (entity response))
      (write-string (get-output-stream-string (html-stream response)) 
		    net.aserve::*html-stream*))))

with 

(defmethod shutdown ((response aserve-response))
  (net.aserve:with-http-response ((request response) (entity response)
				  :response (if (get-header response "Status")
					      (net.aserve::code-to-response
					       (parse-integer
						(get-header response "Status")))
					      nil))
    (net.aserve:with-http-body ((request response) (entity response))
      (write-string (get-output-stream-string (html-stream response)) 
		    net.aserve::*html-stream*))))

and it now works OK. But it looks ugly :-(


Best regards,

Ant?nio Leit?o.


From mb at bese.it  Sun Jul 17 19:00:07 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 17 Jul 2005 21:00:07 +0200
Subject: [Bese-devel] Bug in aserve backend
In-Reply-To: <87mzowthx0.fsf@gia.ist.utl.pt> (Antonio Menezes Leitao's message
	of "Sat, 09 Jul 2005 11:03:55 +0100")
References: <87mzowthx0.fsf@gia.ist.utl.pt>
Message-ID: <m2k6jpcl6w.fsf@soma.local>

Antonio Menezes Leitao <aml at gia.ist.utl.pt> writes:

> To solve this problem, I suggest that the methods:

[snip]

> should be replaced by something like:

[snip]

> The identical methods for aserve-request should require a similar
> correction.

applied. i didn't test it much but it looks like it works as expected.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Sun Jul 17 19:05:20 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 17 Jul 2005 21:05:20 +0200
Subject: [Bese-devel] Bug in aserve backend for redirect component
In-Reply-To: <878y0gtekd.fsf@gia.ist.utl.pt> (Antonio Menezes Leitao's message
	of "Sat, 09 Jul 2005 12:16:18 +0100")
References: <878y0gtekd.fsf@gia.ist.utl.pt>
Message-ID: <m2fyudcky7.fsf@soma.local>

Antonio Menezes Leitao <aml at gia.ist.utl.pt> writes:

> and it now works OK. But it looks ugly :-(

applied. we'll with 'working,' leave pretty for later.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Sun Jul 17 19:12:42 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 17 Jul 2005 21:12:42 +0200
Subject: [Bese-devel] Usage of <ucw:select?
In-Reply-To: <871x64jtzc.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 12 Jul 2005 10:42:31 +0200")
References: <871x64jtzc.fsf@flarge.here>
Message-ID: <m28y05cklx.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> Well that's the point. What can be fed to that tag and how it is
> supposed to be used?

like this:

(<ucw:select :accessor (my-accessor)
  (<ucw:option :value :value1 "the first value")
  (<ucw:option :value :value2 "the second value")
  (dolist (v other-values)
    (<ucw:option :value v (<:as-html v))))

where V can be any lisp object, the select and options tags work
together to generate a hash table containing the keys whcih are used
to map between the server side objects and the strings which are sent
back and forth between the client.

does this answer your question?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Sun Jul 17 19:21:04 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 17 Jul 2005 21:21:04 +0200
Subject: [Bese-devel] ucw-toplevel in make-image.lisp
In-Reply-To: <20050710230657.5413.qmail@web40627.mail.yahoo.com> (Aleksandar
	Bakic's message of "Sun, 10 Jul 2005 16:06:56 -0700 (PDT)")
References: <20050710230657.5413.qmail@web40627.mail.yahoo.com>
Message-ID: <m24qatck7z.fsf@soma.local>

Aleksandar Bakic <a_bakic at yahoo.com> writes:

> The dumped core works only if lisp is later started in the same directory. This
> is because *default-pathname-defaults* is not captured, being special. You need
> something like this:

good catch. i created a new variable (*ucw-install-root*) and used it
exactly as you suggested.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Sun Jul 17 19:24:55 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 17 Jul 2005 21:24:55 +0200
Subject: [Bese-devel] "GET /ucw/js/UCWInputCompleter.js HTTP/1.1" 404 -1
In-Reply-To: <20050709235408.27024.qmail@web40613.mail.yahoo.com> (Aleksandar
	Bakic's message of "Sat, 9 Jul 2005 16:54:08 -0700 (PDT)")
References: <20050709235408.27024.qmail@web40613.mail.yahoo.com>
Message-ID: <m2zmslb5h4.fsf@soma.local>

Aleksandar Bakic <a_bakic at yahoo.com> writes:

> Hi,
>
> I am getting the error in the subject while using SBCL and aserve backend.

:(

1) could you trace ucw:publish-directory and make sure that
   ucw/wwwroot is getting published?

2) if so, what's the value of:

(ucw:application.www-roots it.bese.ucw-user::*example-application*) ?



-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From a_bakic at yahoo.com  Sun Jul 17 22:48:26 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sun, 17 Jul 2005 15:48:26 -0700 (PDT)
Subject: [Bese-devel] "GET /ucw/js/UCWInputCompleter.js HTTP/1.1" 404 -1
In-Reply-To: <m2zmslb5h4.fsf@soma.local>
Message-ID: <20050717224826.7801.qmail@web40601.mail.yahoo.com>

> 1) could you trace ucw:publish-directory and make sure that
>    ucw/wwwroot is getting published?

I don't think it's published. In the meantime,  I figured out that the
following helps (in start.lisp and make-image.lisp/ucw-toplevel):

    (publish-directory (server.backend *default-server*)
                       (truename (merge-pathnames "./wwwroot/ucw/js/" dpd))
                       "/ucw/js/")

> 2) if so, what's the value of:
> 
> (ucw:application.www-roots it.bese.ucw-user::*example-application*) ?

It is #P".../wwwroot/ucw/examples/".

Alex


		
__________________________________ 
Yahoo! Mail for Mobile 
Take Yahoo! Mail with you! Check email on your mobile phone. 
http://mobile.yahoo.com/learn/mail 


From drewc at tech.coop  Mon Jul 18 17:58:47 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Mon, 18 Jul 2005 10:58:47 -0700
Subject: [Bese-devel] XMLS and sbcl 0.9.0
Message-ID: <42DBEDD7.4000906@tech.coop>

I'm using TAL for the first time in a while, and i noticed that YACLML's 
  XMLS is broken under my unicode enabled sbcl.

There is a patch in my yaclml repository that fixes the bug in yaclml : 
'darcs pull http://versions.tech.coop/yaclml'

I'm (supposed) to be maintaining XMLS, and i do plan on making a release 
of it shortly, so it would be ideal if we could use one XMLS everywhere :)


drewc



From pupeno at pupeno.com  Mon Jul 18 03:46:57 2005
From: pupeno at pupeno.com (Pupeno)
Date: Mon, 18 Jul 2005 00:46:57 -0300
Subject: [Bese-devel] Corrections
Message-ID: <200507180047.01535.pupeno@pupeno.com>

I have corrected the information regarding the versions of the dependencies 
and how to get them. Here's the patch.
-- 
Pupeno <pupeno at pupeno.com> (http://pupeno.com)
Reading ? Science Fiction ? http://sfreaders.com.ar

PS: Marco, I am CCing to you because today I sent other mails to the mailing 
list that I didn't get back.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: deps3.patch
Type: text/x-diff
Size: 3604 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050718/80c574db/attachment.patch>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050718/80c574db/attachment.sig>

From pupeno at pupeno.com  Sun Jul 17 21:36:56 2005
From: pupeno at pupeno.com (Pupeno)
Date: Sun, 17 Jul 2005 18:36:56 -0300
Subject: [Bese-devel] Dependencies [PATCH]
Message-ID: <200507171836.59662.pupeno@pupeno.com>

Hello,
A little patch stating some facts about UCW's dependencies. As I find some 
more correction I'll be sendind them. These are for iterate and rfc2388.
-- 
Pupeno <pupeno at pupeno.com> (http://pupeno.com)
Reading ? Science Fiction ? http://sfreaders.com.ar
-------------- next part --------------
A non-text attachment was scrubbed...
Name: deps.patch
Type: text/x-diff
Size: 1610 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050717/c4f8cc8c/attachment.patch>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050717/c4f8cc8c/attachment.sig>

From pupeno at pupeno.com  Sun Jul 17 22:15:59 2005
From: pupeno at pupeno.com (Pupeno)
Date: Sun, 17 Jul 2005 19:15:59 -0300
Subject: [Bese-devel] Mero on dependencies
Message-ID: <200507171916.01720.pupeno@pupeno.com>

Here I added a bit more on dependencies.
-- 
Pupeno <pupeno at pupeno.com> (http://pupeno.com)
Reading ? Science Fiction ? http://sfreaders.com.ar
-------------- next part --------------
A non-text attachment was scrubbed...
Name: deps2.patch
Type: text/x-diff
Size: 2608 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050717/0cf0644c/attachment.patch>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050717/0cf0644c/attachment.sig>

From mb at bese.it  Mon Jul 18 06:05:57 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 18 Jul 2005 08:05:57 +0200
Subject: [Bese-devel] XMLS and sbcl 0.9.0
In-Reply-To: <42DBEDD7.4000906@tech.coop> (Drew Crampsie's message of "Mon,
	18 Jul 2005 10:58:47 -0700")
References: <42DBEDD7.4000906@tech.coop>
Message-ID: <m2u0isbqd6.fsf@soma.local>

Drew Crampsie <drewc at tech.coop> writes:

> I'm using TAL for the first time in a while, and i noticed that
> YACLML's XMLS is broken under my unicode enabled sbcl.
>
> There is a patch in my yaclml repository that fixes the bug in yaclml
> : 'darcs pull http://versions.tech.coop/yaclml'

applied. 

currently yaclml uses a slightly modified version of xmls, when you
put out a new release i think i'd be best to remove xmls from within
yaclml and add the dependncy on your version.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Mon Jul 18 06:09:23 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 18 Jul 2005 08:09:23 +0200
Subject: [Bese-devel] Re: Corrections
In-Reply-To: <200507180047.01535.pupeno@pupeno.com> (pupeno@pupeno.com's
	message of "Mon, 18 Jul 2005 00:46:57 -0300")
References: <200507180047.01535.pupeno@pupeno.com>
Message-ID: <m2pstgbq7g.fsf@soma.local>

Pupeno <pupeno at pupeno.com> writes:

> I have corrected the information regarding the versions of the dependencies 
> and how to get them. Here's the patch.

applied, thanks.

> PS: Marco, I am CCing to you because today I sent other mails to the mailing 
> list that I didn't get back.

for some odd reason you weren't subscribed to the list anymore. this
is probably due to cl.net's crash a few weeks ago.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Mon Jul 18 07:00:50 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 18 Jul 2005 09:00:50 +0200
Subject: [Bese-devel] Usage of <ucw:select?
In-Reply-To: <m28y05cklx.fsf@soma.local> (Marco Baringer's message of "Sun,
	17 Jul 2005 21:12:42 +0200")
References: <871x64jtzc.fsf@flarge.here> <m28y05cklx.fsf@soma.local>
Message-ID: <874qasmwd9.fsf@flarge.here>

"Marco Baringer" <mb at bese.it> writes:

> Friedrich Dominicus <frido at q-software-solutions.de> writes:
>
>> Well that's the point. What can be fed to that tag and how it is
>> supposed to be used?
>
> like this:
>
> (<ucw:select :accessor (my-accessor)
>   (<ucw:option :value :value1 "the first value")
>   (<ucw:option :value :value2 "the second value")
>   (dolist (v other-values)
>     (<ucw:option :value v (<:as-html v))))
Thanks I'll check it. I've worked around it a bit different and I
think I do like it  a bit more ;-)

Regards
Friedrich


From mb at bese.it  Mon Jul 18 17:36:42 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 18 Jul 2005 19:36:42 +0200
Subject: [Bese-devel] component class redefinition
Message-ID: <m27jfot3rp.fsf@soma.local>


ucw has, for way way way too long, had the following issue:

(defcomponent foo ()
  ((a)))

(defcomponent foo ()
  ((a :component (some-random-component))))

after these two forms the A slot will _not_ be properly initialized
when creating foo components, the only work around was to (setf
(find-class 'foo) nil) and then redefine all the methods and
subclasses. this is at best ugly and at worst flat out wrong. i have
just applied a patch which should allow component class redefinition
to do what it should do. It has been tested on openmcl only, since the
various MOP implementations differ i would like to warn everyone that,
if code which previously worked fine magically stops working, this
could be the reason.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Tue Jul 19 05:46:13 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 19 Jul 2005 07:46:13 +0200
Subject: [Bese-devel] secure connections?
In-Reply-To: <m27jfot3rp.fsf@soma.local> (Marco Baringer's message of "Mon,
	18 Jul 2005 19:36:42 +0200")
References: <m27jfot3rp.fsf@soma.local>
Message-ID: <87zmsjcpqy.fsf@flarge.here>

I wonder which backend allows for secure connections. At the moment my
idea is to let Apache do that kind of work. However AFAIKT the only
backend really supporting secure connections seems to be
AllegroServer.

Am I righr on this?

Regards
Friedrich


From mb at bese.it  Tue Jul 19 07:14:39 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 19 Jul 2005 09:14:39 +0200
Subject: [Bese-devel] secure connections?
In-Reply-To: <87zmsjcpqy.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 19 Jul 2005 07:46:13 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
Message-ID: <m2br4zmfmo.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> I wonder which backend allows for secure connections. At the moment my
> idea is to let Apache do that kind of work. However AFAIKT the only
> backend really supporting secure connections seems to be
> AllegroServer.
>
> Am I righr on this?

yes. though many people use araneida behind a reverse proxy such as
pound, which does support https. even portableaserve supports https
only on AllegroCL and lispworks (afair)

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Tue Jul 19 17:07:39 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 19 Jul 2005 19:07:39 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <87zmsjcpqy.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 19 Jul 2005 07:46:13 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
Message-ID: <878y02d8ro.fsf_-_@flarge.here>

Ok, another round of queries. How can one intermix the both things 
Let's say at the top of the page I like generated html and below some
statical stuff or other things.

e.g assume I have the following 
(defcomponent trans-info (simple-window-component template-component)
  ((products-in-basket :initarg :products-in-basket 
                       :accessor products-in-basket)
   (payment-terms :accessor payment-terms)
   (trans-price :accessor trans-price :initarg :trans-price :initform 0.0))
  (:default-initargs 
      :title "Your shopping basket"
    :stylesheet "/css/qss.css"))


;;     :template-name "/home/frido/lib/cl/web-app/qss-web/wwwroot/get-user-info.tal"))

Now I want to use the get-user-info.tal file, which looks like this at
the moment
<!-- -*- html -*- -->


<html xmlns:tal="http://common-lisp.net/project/bese/tal/core"
      xmlns:ucw="http://common-lisp.net/project/ucw/core"
      tal:in-package="qss.web.ucw">
      <head>
        <title>Please provide information</title>
    <link href="/css/qss.css" rel="stylesheet"/>
  </head>
    <span <tal:include="/global/menu-store.html"> />
    
      <h1>User information</h1>
      
      <h2>Your shopping basket</h2>
      <span
        <tal:lisp>(render-on (context.response *context*) $component)
          </tal:lisp> >
        Your shopping basket.
      </span>
      </p>
    <span <tal:include="/global/footer.html"> />


How do I call this tal file and after that how can I use a
user-written render-on function to render the beginning?

Are there any suggestions on when to use what approach. When to use
TAL and when to use yaclml?


Regards
Friedrich


From joseph at mammalia.net  Thu Jul 21 05:07:59 2005
From: joseph at mammalia.net (R. Joseph Wright)
Date: Wed, 20 Jul 2005 22:07:59 -0700
Subject: [Bese-devel] stylesheets.css and mod-lisp
Message-ID: <D50E3D3F-5D97-4004-A0C6-965A4C9BB41D@mammalia.net>

I'm testing out UCW with mod-lisp, and I haven't been able to get the  
stylesheet to work in the examples directory without creating a  
symlink to *ucw-tal-root* in Apache's DocumentRoot.  Is this the  
standard way?  I have the following Location directive in http.conf,  
and I thought if that was there then lisp would hand whichever  
documents necessary under that directory to apache.

<Location "/ucw/examples/">
    SetHandler lisp-handler
</Location>

(\_/)  Joseph
(o,o)  Those are my principles.  If you don't like them I have others.
()_()   --Groucho Marx
  " "



From jdz at dir.lv  Thu Jul 21 12:06:37 2005
From: jdz at dir.lv (Janis Dzerins)
Date: Thu, 21 Jul 2005 15:06:37 +0300
Subject: [Bese-devel] Bug in aserve backend
In-Reply-To: <87mzowthx0.fsf@gia.ist.utl.pt>
References: <87mzowthx0.fsf@gia.ist.utl.pt>
Message-ID: <42DF8FCD.2050406@dir.lv>

Antonio Menezes Leitao wrote:

> To solve this problem, I suggest that the methods:
> 
> (defmethod get-header ((response aserve-response) header-name)
>   (net.aserve:reply-header-slot-value (request response) header-name))
> 
> (defmethod (setf get-header) (value (response aserve-response) header-name)
>   (setf (net.aserve:reply-header-slot-value (request response) header-name)
>         value))
> 
> should be replaced by something like:
> 
> (defmethod get-header ((response aserve-response) header-name)
>   (switch (header-name :test #'string-equal)
>     ("Date"             (net.aserve:reply-header-slot-value (request response) :date))
>     ("Content-Type"     (net.aserve:reply-header-slot-value (request response) :content-type))
>     ("Content-Length"   (net.aserve:reply-header-slot-value (request response) :content-length))
>     (t (net.aserve:reply-header-slot-value
> 	(request response)
> 	(read-from-string (format nil ":~A" header-name))))))
> 
> (defmethod (setf get-header) (value (response aserve-response) header-name)
>   (switch (header-name :test #'string-equal)
>     ("Date"             (error "Can't update header Date"))
>     ("Content-Type"     (setf (net.aserve:reply-header-slot-value (request response) :content-type)   value))
>     ("Content-Length"   (setf (net.aserve:reply-header-slot-value (request response) :content-length) value))
>     (t (setf (net.aserve:reply-header-slot-value
> 	      (request response)
> 	      (read-from-string (format nil ":~A" header-name)))
> 	     value))))

Why not like this:


(defmethod get-header ((response aserver-response) header-name)
  (net.aserve:reply-header-slot-value
   (request response)
   (intern (string-upcase header-name) :keyword)))

(defmethod (setf get-header) (value (response aserver-response) header-name)
  (let ((header (intern (string-upcase header-name) :keyword)))
    (if (eq :date header)
        (error "Can't update Date header.")
        (setf (net.aserve:reply-header-slot-value (request response)
                                                  header)
              value))))

?

(Ok, there most probably is a problem with Allegro's "modern" mode, but your
version is not really better in this respect.)

--
Janis Dzerins


From mb at bese.it  Thu Jul 21 14:29:56 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 21 Jul 2005 16:29:56 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <878y02d8ro.fsf_-_@flarge.here> (Friedrich Dominicus's message of
	"Tue, 19 Jul 2005 19:07:39 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here>
Message-ID: <m2zmsgi657.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> How do I call this tal file and after that how can I use a
> user-written render-on function to render the beginning?

i believe the method you are looking for is ucw::render-template (see
ucw/src/yaclml/yaclml.lisp). you can call that from within any
render-on method, the only thing you'll need to do is create a TAL
environement object (a list of alists, objects or hash tables).

if you're trying to embed render-on methods within TAL files you can
use the ucw:render-component tag, it takes one attribute ucw:component
whcih will be evaluated and must return a component object (usually
somthing like "(current-pane $component)" is sufficent).

of course ucw:render-component does nothing more than call render-on,
so it's exactly the same as the example you pasted with embeddeb lisp
code.

if instead you have a common page layout whcih you want to reuse for
many components then you could use tal:include. The only 'interesting'
characteristc about tal:include is that it's generally used inside
out. each specific page will include the defalut layout and pass it
the code to put in the various parts of the page. example:

my-site-layout.tal:
------------------------------------------------------------------------
<!-- -*- html -*- -->
<html xmlns="http://www.w3.org/1999/xhtml"
      xmlns:tal="http://common-lisp.net/project/bese/tal/core"
      tal:in-package="idi">
  <head>
    <title tal:content="$title">TITLE</title>
  </head>

  <body>
    <h1 tal:content="$title">TITLE</h1>
    <div id="body" tal:content="$body" tal:escape-html="nil">BODY</div>
  </body>
</html>
------------------------------------------------------------------------

index.tal:
------------------------------------------------------------------------
<tal:include tal:name="my-site-layout.tal"
             xmlns="http://www.w3.org/1999/xhtml"
             xmlns:tal="http://common-lisp.net/project/bese/tal/core"
             xmlns:param="http://common-lisp.net/project/bese/tal/params">
  <param:title>Commercial Agency</param:title>
  <param:body>
    ...
  </param:body>
</tal:include>
------------------------------------------------------------------------

does this help any?

> Are there any suggestions on when to use what approach. When to use
> TAL and when to use yaclml?

i use yaclml as much as i can, falling back to TAL when someone else
takes care of the graphic work, which is the case in most of my
projects.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul 21 14:32:28 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 21 Jul 2005 16:32:28 +0200
Subject: [Bese-devel] Bug in aserve backend
In-Reply-To: <42DF8FCD.2050406@dir.lv> (Janis Dzerins's message of "Thu,
	21 Jul 2005 15:06:37 +0300")
References: <87mzowthx0.fsf@gia.ist.utl.pt> <42DF8FCD.2050406@dir.lv>
Message-ID: <m2vf34i60z.fsf@soma.local>

Janis Dzerins <jdz at dir.lv> writes:

> Why not like this:
>
> (defmethod get-header ((response aserver-response) header-name)
>   (net.aserve:reply-header-slot-value
>    (request response)
>    (intern (string-upcase header-name) :keyword)))
>
> (defmethod (setf get-header) (value (response aserver-response) header-name)
>   (let ((header (intern (string-upcase header-name) :keyword)))
>     (if (eq :date header)
>         (error "Can't update Date header.")
>         (setf (net.aserve:reply-header-slot-value (request response)
>                                                   header)
>               value))))
>
> ?
>
> (Ok, there most probably is a problem with Allegro's "modern" mode, but your
> version is not really better in this respect.)

interning and symbol name case mapping issues aside, isn't the point
of this patch that reply-header-slot-value only works if certain
arguments are compile time constant keywords? am i getting confused?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul 21 14:34:55 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 21 Jul 2005 16:34:55 +0200
Subject: [Bese-devel] "GET /ucw/js/UCWInputCompleter.js HTTP/1.1" 404 -1
In-Reply-To: <20050717224826.7801.qmail@web40601.mail.yahoo.com> (Aleksandar
	Bakic's message of "Sun, 17 Jul 2005 15:48:26 -0700 (PDT)")
References: <20050717224826.7801.qmail@web40601.mail.yahoo.com>
Message-ID: <m2r7dsi5ww.fsf@soma.local>

Aleksandar Bakic <a_bakic at yahoo.com> writes:

>> 1) could you trace ucw:publish-directory and make sure that
>>    ucw/wwwroot is getting published?
>
> I don't think it's published. In the meantime,  I figured out that the
> following helps (in start.lisp and make-image.lisp/ucw-toplevel):

ok, this is a bug in the examples themselves. the issue is this:

whenever we startup an application we publish the wwwroot of the
application object at the url-prefix of the app, in the case of the
example app this means that "ucw/wwwroot/ucw/examples/" gets published
as "host:port/ucw/examples/", unfortunetly there is no application
which ever publishes the ucw/js directory. i'll move the js stuff into
the example directory (which is where it should be anyway).

thanks for the bug report.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul 21 14:45:05 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 21 Jul 2005 16:45:05 +0200
Subject: [Bese-devel] UCW file upload example
In-Reply-To: <E1Dspfs-0002Qm-EN@e4ward.com> (Mac Chan's message of "Wed,
	13 Jul 2005 15:24:50 -0700")
References: <E1Dspfs-0002Qm-EN@e4ward.com>
Message-ID: <m21x5si5fy.fsf@soma.local>

Mac Chan <lispweb at mac.e4ward.com> writes:

> Hi,
>
> I was unable to run the file upload example. Upon playing around
> with the sample code for a while I figured it out.
>
> Please see the attached diff.

thakn you.

> I'm building a site for teachers to upload multimedia contents which
> can be upward of 100mb+ files. The way ucw stores uploaded content in
> memory cannot cut it.

no, it definetly can't. but even if it could wouldn't it be better to
use something like ftp (or any protocol which allows resumes) for
files that big?

> TBNL implemented a hack (write-content-to-file t) where they put a
> hook in cl-rfc2388 to allow writing uploaded content to file directly.
> Since it's a hack (TBNL manages the file creation, uploaded contents
> itself) for TBNL only, it is not usable for other frameworks.
>
> Even if it does, I do not have a good idea of how to incorporate 
> this into ucw cleanly.

just change the mime-part object (i took a quick walk through the
backend code it shouldn't be very hard at all).

> Right now code in UCW can access the file content with slot-value
> directly, which is very simple but it doesn't scale for large files.
>
> If we put a pathname there instead, then everyone that access the slot
> will have to examine the tag first (much like the bug fix I submit
> below), which complicates things.

only a little bit, and it makes something possible which is currently
impossible. looks like a worthy tradeoff to me.

what about putting a stream (tied to a file on disk) in the slot?

> Another drawback is when session expires, those files need to be
> cleaned up. Right now GC handles that and if we put the pathname in
> there, ucw has to do the cleanup.

this is not a big deal, stick all the generated pathnames in the
session object and define an :after method on ucw:expire-session (that
function exists for the very purpose of allowing this type of clean up
to happen). you have no guarntee over exactly when expire-session will
be called but you can sure that it will be called (assuming the
machine doesn't crash).

> I'm soliciting opinions how to architech this as I'm not all too
> familiar with the underneath framework.
>
> (Araneida right now doesn't support multipart MIME form post.
> aserve however do support this.)

would it be too hard to add mime support to araneida?

> BTW There are certain things that can be done in a more straightforward way
> in aserve / TBNL / araneida.

like what? why aren't these straight forward in ucw?

> I think that UCW cannot share objects with Araneida or other backend directly.
> Can we hack both and so that they share a unique session id and some hooks
> to access objects created in both sides?

UCW currently calls the generic function FIND-SESSION on the current
application object, it gets the passed the application and the
context, since the context contains backend specific request objects
there's no reason why you could not create your own application class
which uses an araneida specific session object (or mod_lisp's
HTTPS_SESSION_ID).

the reason ucw has a RERL protocol is to make this kind of stuff easy,
i've tried to put in as many hooks as was reasonably possible. if
you're missing something i'd be happy to see about adding it.

> In that case I might be able to handle the boring file upload in
> aserve and let ucw handle all the other cool stuff :-)
>
> Does anyone have any experience to share, regarding using both
> UCW and other backends in your app?
>
> I heard that drewc runs cliki & ucw in the same image. I wonder
> how that works.

afaik the now run in the same image but are completly independant.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul 21 15:15:02 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 21 Jul 2005 17:15:02 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <m2zmsgi657.fsf@soma.local> (Marco Baringer's message of "Thu,
	21 Jul 2005 16:29:56 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here> <m2zmsgi657.fsf@soma.local>
Message-ID: <m2vf34fax5.fsf@soma.local>

"Marco Baringer" <mb at bese.it> writes:

> Friedrich Dominicus <frido at q-software-solutions.de> writes:
>
>> How do I call this tal file and after that how can I use a
>> user-written render-on function to render the beginning?
>
> i believe the method you are looking for is ucw::render-template (see
> ucw/src/yaclml/yaclml.lisp). you can call that from within any
> render-on method, the only thing you'll need to do is create a TAL
> environement object (a list of alists, objects or hash tables).

that last part would should be: a list of binding objects, a binding
object con be either a CLOS instance, an alist or a hash table.


From mb at bese.it  Thu Jul 21 15:17:56 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 21 Jul 2005 17:17:56 +0200
Subject: [Bese-devel] UCW file upload example
In-Reply-To: <E1Dspfs-0002Qm-EN@e4ward.com> (Mac Chan's message of "Wed,
	13 Jul 2005 15:24:50 -0700")
References: <E1Dspfs-0002Qm-EN@e4ward.com>
Message-ID: <m2r7dsfasb.fsf@soma.local>

Mac Chan <lispweb at mac.e4ward.com> writes:

>  (defmethod get-parameter ((request httpd-request) name)
> -  (let ((cons (assoc name (parameters request) :test #'string=)))
> -    (when cons (copy-seq (cdr cons)))))
> +  (let* ((cons (assoc name (parameters request) :test #'string=))
> +         (value (cdr cons)))
> +    (when cons (copy-seq (if (mime-part-p value)
> +                             (mime-part-body value) value)))))
>
>  (defmethod map-parameters ((request httpd-request) lambda)
>   (dolist* ((name . value) (parameters request))
> -    (funcall lambda name (copy-seq value))))
> +    (funcall lambda name (copy-seq
> +                          (if (mime-part-p value)
> +                              (mime-part-body value) value)))))
>
>  (defun read-request-headers (backend stream)
>   (iterate
> ------------------------------------------->8--------------------------------------

actually, this is wrong. in particular this makes it impossible to get
at any of the mime-headers associated with the parameter. in the case
of file uploads we (i at least) often need the file's name and
content-type. 

i think i'm going to leave get-parameter as it is and fix the code in
the file upload example to deal with mime-part object. you (or anyone
else) have anything agaisnt this?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul 21 16:00:30 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 21 Jul 2005 18:00:30 +0200
Subject: [Bese-devel] UCW file upload example
In-Reply-To: <E1Dspfs-0002Qm-EN@e4ward.com> (Mac Chan's message of "Wed,
	13 Jul 2005 15:24:50 -0700")
References: <E1Dspfs-0002Qm-EN@e4ward.com>
Message-ID: <m2br4wf8td.fsf@soma.local>

Mac Chan <lispweb at mac.e4ward.com> writes:

> I was unable to run the file upload example. Upon playing around
> with the sample code for a while I figured it out.

here's what i ended up doing:

if the form is sumbitted as a regular form
(application/x-www-form-urlencoded) then all parameters as passed from
the backend as strings, nothing new here. if the form is multipart
(multipart/form-data) then a parameter is passed as string iff it has
exactly one header named "name", otherwise we create a wrap it and the
headers in a mime-part object and pass those on. map-parameters, and
get-parameter have been updated to call copy-seq only when the
parameter is a string. this change only affects the httpd and mod_lisp
backends, patches for araneida (if possible) and aserve are more than
welcome (but those backends aren't priporities for me atm).

this is an incompatible change and has, therefore, been put in the
darcs repo at:

http://common-lisp.net/project/ucw/repos/ucw_dev

it will sit there until i've tested it on a few of may apps and then
i'll move the changes over to the regular repo. of course, the
previous code was seriously broken so this probably won't affect
anyone.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From jdz at dir.lv  Thu Jul 21 15:35:18 2005
From: jdz at dir.lv (Janis Dzerins)
Date: Thu, 21 Jul 2005 18:35:18 +0300
Subject: [Bese-devel] Bug in aserve backend
In-Reply-To: <m2vf34i60z.fsf@soma.local>
References: <87mzowthx0.fsf@gia.ist.utl.pt> <42DF8FCD.2050406@dir.lv>
	<m2vf34i60z.fsf@soma.local>
Message-ID: <42DFC0B6.4080602@dir.lv>

Marco Baringer wrote:
> Janis Dzerins <jdz at dir.lv> writes:
> 
> 
>>Why not like this:
>>
>>(defmethod get-header ((response aserver-response) header-name)
>>  (net.aserve:reply-header-slot-value
>>   (request response)
>>   (intern (string-upcase header-name) :keyword)))
>>
>>(defmethod (setf get-header) (value (response aserver-response) header-name)
>>  (let ((header (intern (string-upcase header-name) :keyword)))
>>    (if (eq :date header)
>>        (error "Can't update Date header.")
>>        (setf (net.aserve:reply-header-slot-value (request response)
>>                                                  header)
>>              value))))
>>
>>?
>>
>>(Ok, there most probably is a problem with Allegro's "modern" mode, but your
>>version is not really better in this respect.)
> 
> 
> interning and symbol name case mapping issues aside, isn't the point
> of this patch that reply-header-slot-value only works if certain
> arguments are compile time constant keywords? am i getting confused?

Ok, that clears my confusion now.  I hope I did not confuse anybody.

-- 
Janis Dzerins

   Common Lisp -- you get more than what you see.



From a_bakic at yahoo.com  Thu Jul 21 18:06:07 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Thu, 21 Jul 2005 11:06:07 -0700 (PDT)
Subject: [Bese-devel] toplevel-repl and sbcl
Message-ID: <20050721180607.6210.qmail@web40629.mail.yahoo.com>

Hi,

In make-image.lisp, #+sbcl (sb-impl::toplevel-repl) works for me only if I add
an argument, t. That is, #+sbcl (sb-impl::toplevel-repl t). I use :sb-thread,
not sure if that matters.

Alex


		
____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs 
 


From a_bakic at yahoo.com  Fri Jul 22 12:01:53 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Fri, 22 Jul 2005 05:01:53 -0700 (PDT)
Subject: [Bese-devel] wall-time
Message-ID: <20050722120153.56622.qmail@web40601.mail.yahoo.com>

How about renaming the ucw functions a little, for a start? (Later, perhaps a
reader macro can be used to decide whether to use clsql's definitions, if
present, or ucw's definitions.)

Alex

>> Which means that I end up patching ucw and make it depend on clsql
>> instead of using its own wall-time.  What do you guys do?  It seems
>> you can't store a ucw::wall-time into a clsql view seamlessly.>
>yeah, that's a bit of ugliness i should probably work around. what i
>wanted to avoid was adding a dependency on clsql just for its
>wall-time implementation. on the other hand i have no intention of
>implementing my own time implmentation, nor using universal-time. i'm
>open to suggestions.


		
____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs 
 


From mb at bese.it  Fri Jul 22 12:19:08 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 22 Jul 2005 14:19:08 +0200
Subject: [Bese-devel] wall-time
In-Reply-To: <20050722120153.56622.qmail@web40601.mail.yahoo.com> (Aleksandar
	Bakic's message of "Fri, 22 Jul 2005 05:01:53 -0700 (PDT)")
References: <20050722120153.56622.qmail@web40601.mail.yahoo.com>
Message-ID: <m2oe8vc9tv.fsf@soma.local>

Aleksandar Bakic <a_bakic at yahoo.com> writes:

> How about renaming the ucw functions a little, for a start? (Later, perhaps a
> reader macro can be used to decide whether to use clsql's definitions, if
> present, or ucw's definitions.)

could, but it looks like drew's lisp-on-lines which (afair) contianst
the presentation stuff (and a dependncy on clsql) is almost ready for
release. as soon that happens i'm just going to remove the
presentation stuff from ucw and tell people to use LoL (which doesn't
have this problem since it uses clsql:wall-time directly). this sholud
fix both problems. am i missing something?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From asbjxrn at bjxrnstad.net  Fri Jul 22 15:40:57 2005
From: asbjxrn at bjxrnstad.net (=?ISO-8859-1?Q?Asbj=F8rn_Bj=F8rnstad?=)
Date: Fri, 22 Jul 2005 23:40:57 +0800
Subject: [Bese-devel] automated testing of ucw apps
Message-ID: <2E5A457E-F900-47E6-985C-C82B5964F6F9@bjxrnstad.net>


Is there a nice easy way of doing automated testing of ucw apps?
I guess you'd need a testing framework that would handle the  
continuations/sessions.


-- 
   -asbjxrn



From mb at bese.it  Fri Jul 22 17:49:37 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 22 Jul 2005 19:49:37 +0200
Subject: [Bese-devel] automated testing of ucw apps
In-Reply-To: <2E5A457E-F900-47E6-985C-C82B5964F6F9@bjxrnstad.net>
	(=?iso-8859-1?Q?Asbj=F8rn_Bj=F8rnstad's?= message of "Fri, 22 Jul 2005
	23:40:57 +0800")
References: <2E5A457E-F900-47E6-985C-C82B5964F6F9@bjxrnstad.net>
Message-ID: <m2mzoeafym.fsf@soma.local>

Asbj?rn Bj?rnstad <asbjxrn at bjxrnstad.net> writes:

> Is there a nice easy way of doing automated testing of ucw apps?
> I guess you'd need a testing framework that would handle the
> continuations/sessions.

i think defining your own response object and seperate render-on
methods for each component would go a _long_ way to making it easy to
test ucw apps.

something like this for window components:

(defmethod render-on :before ((res test-response) (w window-component))
  (princ *yaclml-stream*
         (list (session.id (context.session *context*))
               (frame.id (session.current-frame ...)))))

and then you colud use something like this as a generic render-on
method:

(defmethod render-on ((res test-response) (c standard-component))
  (nth-value 1 (make-load-form c nil)))

basically what would this accomplish: every time you sent in a 'test'
request you'd get an easily parsable (READ sholud be enough) object
which describes the current componet without having to worry about
parsing xhtml. the only thing (that i can think of atm) left is a
machine friendly way of outputing the action links, this would allow
you to setup a script which gets a page, inspects the component and
then calls the actions and then makes sure the proper component comes
back. doing this would be a "simple" matter of using a function like
this:

(defmarco <ucw:test-a (&key action)
  (format *yaclml-stream*
          (list :action-link 
                (print-uri-to-string
                 (compute-url *current-component*
                              :action-id (make-new-action (context.current-frame *context*)
                                                          (lambda () ,action)))

again, this isn't fundamentally different than <ucw:a but has the
advantage being READ'able.

other than these "minor" changes to the rendering out i can't think of
anything which could done to make testing ucw apps easier (thoguh
suggestions are welcome).

should you ever write anything and feel like releasing it i'd be vory
vory happy to include it in ucw.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Sat Jul 23 12:07:20 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 23 Jul 2005 14:07:20 +0200
Subject: [Bese-devel] new cps interprater based ucw ready
Message-ID: <m2u0il4tfr.fsf@soma.local>


hi,

  i've gotten to the point where the new interpter is able to run
  ucw's examples and admin app. you'lle need to grab the patches to
  ucw from here:

http://common-lisp.net/project/ucw/repos/ucw_new_cps_integration

  and the interpreter from here:

http://common-lisp.net/project/bese/repos/arnesi_cps

  i'd be very very happy if people would test this code out and make
  sure it handles all the weird corner cases which come up. if you had
  code which didn't work using the old cps transformer try using that
  code as well, it should work now.

why should you care about this new interpreter?

1) you can now define action and defun/cc's in any order. you alse
   redefine things as you please and it will do the Right Thing(TM).

2) multiple-values are supported. 

3) no more exponential code growth.

4) error messages report the actual, original, source code where the
   error occured.

5) as long as you don't attempt to change the values of bindings
   outside of the lexical scope of the with-call/cc form the
   continuations are now serializable. writing code to actually do the
   serialization (and then store entire session objects in a memcache)
   is on the TODO list (assuming nothing really really bad happens
   with this new code).

6) this paves the way to defining a special library of cl's high order
   functions which will allow things like:

   (remove-if (lambda (object)
                (call 'yes-or-n-p :object object))
              *objects*)

happy hacking.
-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From darkman_spam at yahoo.fr  Sat Jul 23 18:29:22 2005
From: darkman_spam at yahoo.fr (drkm)
Date: Sat, 23 Jul 2005 20:29:22 +0200
Subject: [Bese-devel] Error while installing
Message-ID: <uirz1jrzx.fsf@yahoo.fr>

  Hi

  After having read very good things about UCW on c.l.lisp, I
tried to install it.  I downloaded latest versions for all
packages, as adviced on CLiki.  Here is what I have in my
personnal ASDF repository:

    ~/.clc/systems $ ls -ol
    lrwxrwxrwx  1 drkm   20 arnesi.asd -> ../arnesi/arnesi.asd
    lrwxrwxrwx  1 drkm   24 puri.asd -> ../puri-1.3.1.2/puri.asd
    lrwxrwxrwx  1 drkm   22 rfc2388.asd -> ../rfc2388/rfc2388.asd
    drwxr-xr-x  2 drkm 4096 src
    lrwxrwxrwx  1 drkm   28 swank.asd -> ../../.elisp/slime/swank.asd
    lrwxrwxrwx  1 drkm   18 ucw.asd -> ../ucw_dev/ucw.asd
    lrwxrwxrwx  1 drkm   20 yaclml.asd -> ../yaclml/yaclml.asd

  Then I 'M-x slime' and evaluate the following:

    (asdf:oos 'asdf:load-op 'iterate)
    (asdf:oos 'asdf:load-op 'arnesi)
    (asdf:oos 'asdf:load-op 'rfc2388)
    (asdf:oos 'asdf:load-op 'yaclml)
    (asdf:oos 'asdf:load-op 'puri)
    (asdf:oos 'asdf:load-op 'ucw)

  All evaluate without error, but the last one.  I reproduce the
error at the end of this email.  I use GNU CLISP on Debian.

  Regards,

--drkm

(DEFCLASS IT.BESE.UCW:STANDARD-COMPONENT-CLASS): superclass #1=#<STRUCTURE-CLASS
     STANDARD-CLASS> should be of class STANDARD-CLASS
   [Condition of type SIMPLE-ERROR]

Restarts:
  0: [RETRY] Retry performing #<ASDF:COMPILE-OP NIL #x000333DFE218> on #<ASDF:CL-SOURCE-FILE "standard-component-class" #x000333DF7310>.
  1: [ACCEPT] Continue, treating #<ASDF:COMPILE-OP NIL #x000333DFE218> on #<ASDF:CL-SOURCE-FILE "standard-component-class" #x000333DF7310> as having been successful.
  2: [ABORT] Abort handling SLIME request.

Backtrace:
  0: frame binding variables (~ = dynamically):
  | ~ SYSTEM::*FASOUTPUT-STREAM* <--> NIL
  1: EVAL frame for form (CLOS::ENSURE-CLASS 'IT.BESE.UCW:STANDARD-COMPONENT-CLASS :DIRECT-SUPERCLASSES (LIST (OR (FIND-CLASS 'STANDARD-CLASS NIL) 'STANDARD-CLASS)) :DIRECT-SLOTS (LIST (LIST :NAME 'IT.BESE.UCW::BACKTRACK-SLOTS :ACCESSORS '(IT.BESE.UCW::COMPONENT-CLASS.BACKTRACK-SLOTS) :READERS '(IT.BESE.UCW::COMPONENT-CLASS.BACKTRACK-SLOTS) :WRITERS '(#) :INITER ...) (LIST :NAME 'IT.BESE.UCW::COMPONENT-SLOTS :ACCESSORS '(IT.BESE.UCW::COMPONENT-CLASS.COMPONENT-SLOTS) :READERS '(IT.BESE.UCW::COMPONENT-CLASS.COMPONENT-SLOTS) :WRITERS '(#) :INITER ...) (LIST :NAME 'IT.BESE.UCW::PLACE-SLOT :ACCESSORS '(IT.BESE.UCW::COMPONENT-CLASS.PLACE-SLOT) :READERS '(IT.BESE.UCW::COMPONENT-CLASS.PLACE-SLOT) :WRITERS '(#) :INITARGS ...)) :DOCUMENTATION '"The meta class for standard-components.

Classes defined with this meta-class have extra slot options, see
the class STANDARD-COMPONENT-DIRECT-SLOT for details.")
  2: EVAL frame for form (PROGN (CLOS::ENSURE-CLASS 'IT.BESE.UCW:STANDARD-COMPONENT-CLASS :DIRECT-SUPERCLASSES (LIST (OR # 'STANDARD-CLASS)) :DIRECT-SLOTS (LIST (LIST :NAME 'IT.BESE.UCW::BACKTRACK-SLOTS :ACCESSORS '# :READERS '# :WRITERS '# :INITER ...) (LIST :NAME 'IT.BESE.UCW::COMPONENT-SLOTS :ACCESSORS '# :READERS '# :WRITERS '# :INITER ...) (LIST :NAME 'IT.BESE.UCW::PLACE-SLOT :ACCESSORS '# :READERS '# :WRITERS '# :INITARGS ...)) :DOCUMENTATION '"The meta class for standard-components.

Classes defined with this meta-class have extra slot options, see
the class STANDARD-COMPONENT-DIRECT-SLOT for details."))
  3: EVAL frame for form (ASDF:OOS 'ASDF:LOAD-OP 'UCW)
  4: EVAL frame for form (SWANK:LISTENER-EVAL "(asdf:oos 'asdf:load-op 'ucw)
")
  5: EVAL frame for form (SWANK:START-SERVER "/tmp/slime.3725" :EXTERNAL-FORMAT :ISO-LATIN-1-UNIX)
 --more--




From mb at bese.it  Sat Jul 23 22:44:55 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 24 Jul 2005 00:44:55 +0200
Subject: [Bese-devel] Error while installing
In-Reply-To: <uirz1jrzx.fsf@yahoo.fr> (drkm's message of "Sat,
	23 Jul 2005 20:29:22 +0200")
References: <uirz1jrzx.fsf@yahoo.fr>
Message-ID: <m2wtnh2lco.fsf@soma.local>

drkm <darkman_spam at yahoo.fr> writes:

>   All evaluate without error, but the last one.  I reproduce the
> error at the end of this email.  I use GNU CLISP on Debian.

ucw requires a fairly recent version of clisp (one that has a fairly
complient MOP). what version of clisp are you using?

that said, ucw is nicer to work with if the underlying lisp is
multi-threaded, why don't you try with sbcl or cumcl?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From svg at surnet.ru  Sat Jul 23 22:52:20 2005
From: svg at surnet.ru (Vladimir Sekissov)
Date: Sun, 24 Jul 2005 04:52:20 +0600 (YEKST)
Subject: [Bese-devel] Error while installing
In-Reply-To: <uirz1jrzx.fsf@yahoo.fr>
References: <uirz1jrzx.fsf@yahoo.fr>
Message-ID: <20050724.045220.99731082.svg@surnet.ru>

Good day,

darkman_spam>   After having read very good things about UCW on c.l.lisp, I
darkman_spam> tried to install it.  I downloaded latest versions for all
darkman_spam> packages, as adviced on CLiki.  Here is what I have in my
darkman_spam> personnal ASDF repository:

darkman_spam>   All evaluate without error, but the last one.  I reproduce the
darkman_spam> error at the end of this email.  I use GNU CLISP on Debian.

Probaly you have too old version of CLISP. CLISP versions prior the
current CLISP 2.34.* had non-standard MOP implementation.

Try to run UCW with latest CLISP (it was released last week and
 probably not yet in Debian) or with other lisps (CMUCL or SBCL).

Debian has excellent CMUCL package. SBCL has unicode but is not as
stable as CMUCL.

Best Regards,
Vladimir Sekissov


From a_bakic at yahoo.com  Sun Jul 24 08:21:38 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sun, 24 Jul 2005 01:21:38 -0700 (PDT)
Subject: [Bese-devel] wall-time
In-Reply-To: <m2oe8vc9tv.fsf@soma.local>
Message-ID: <20050724082138.34286.qmail@web40621.mail.yahoo.com>

> could, but it looks like drew's lisp-on-lines which (afair) contianst
> the presentation stuff (and a dependncy on clsql) is almost ready for
> release. as soon that happens i'm just going to remove the
> presentation stuff from ucw and tell people to use LoL (which doesn't
> have this problem since it uses clsql:wall-time directly). this sholud
> fix both problems. am i missing something?

There seem to be a few darcs repositories on the web with possibly different
versions of UCW (including ucw-dev and/or ucw-dist, not sure). I would like to
be able to use both UCW (and try new stuff such as the interpreter) and LoL.
For now, I suppose it is best to get everything from Drew's site, even if their
UCW is maybe not the same as yours. What should I do?

Thanks,
Alex


		
____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs 
 


From mb at bese.it  Sun Jul 24 09:57:47 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 24 Jul 2005 11:57:47 +0200
Subject: [Bese-devel] wall-time
In-Reply-To: <20050724082138.34286.qmail@web40621.mail.yahoo.com> (Aleksandar
	Bakic's message of "Sun, 24 Jul 2005 01:21:38 -0700 (PDT)")
References: <20050724082138.34286.qmail@web40621.mail.yahoo.com>
Message-ID: <m2ackc34ro.fsf@soma.local>

Aleksandar Bakic <a_bakic at yahoo.com> writes:

>> could, but it looks like drew's lisp-on-lines which (afair) contianst
>> the presentation stuff (and a dependncy on clsql) is almost ready for
>> release. as soon that happens i'm just going to remove the
>> presentation stuff from ucw and tell people to use LoL (which doesn't
>> have this problem since it uses clsql:wall-time directly). this sholud
>> fix both problems. am i missing something?
>
> There seem to be a few darcs repositories on the web with possibly different
> versions of UCW (including ucw-dev and/or ucw-dist, not sure). I would like to
> be able to use both UCW (and try new stuff such as the interpreter) and LoL.
> For now, I suppose it is best to get everything from Drew's site, even if their
> UCW is maybe not the same as yours. What should I do?

the arch stuff (ucw--dev and ucw--dist in the repository
ucw-2004 at common-lisp.net) is fairly recent but won't be updated
anymore. we currently have these darcs repositories:

http://common-lisp.net/project/ucw/repos/ucw

this is the 'stable' repository. i'll move patches over here once
they've been tested and ready for public consumpition. this repo
hasn't changed for a while and basically exists so that newbies have
some place they can download code from and have everything 'Just
Work (TM)' [in so far as anything ever just works]

http://common-lisp.net/project/ucw/repos/ucw_dev

this is the 'development' repository. this may or may not work, may
contain incompatable changes, may make your computer catch fire,
etc. as soon as the new cps stuff is more or less tested i'll move it
over to here. (this is probably the ucw repo you want to follow)

http://common-lisp.net/project/ucw/repos/ucw_new_cps_integration

this repository, which won't be around very long, contians those
patches to ucw which are neccessary to run ucw with the new cps
interpreter. i'm using this, as opposed to ucw_dev, because i'd like
to make sure things more or less work before breaking everything.

http://common-lisp.net/project/bese/repos/arnesi

this is the main arnesi repository. it currently cantians the _old_
cps transformer and a few other trivial patches.

http://common-lisp.net/project/bese/repos/arnesi_cps

this repository has the new cps interpreter and works with the code in
ucw_new_cps_integration. as soon as things have settled a bit and i'm
fairly confident everything works i'm going to move the code over to
the regular arnesi repository and delete this one.

so basically if you want to test out the new cps interpreter you need
to get/pull from arnesi_cps and ucw_new_cps_integration (the examples
currently all work, don't know about other code).

hope this helps.


-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Mon Jul 25 09:49:15 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 25 Jul 2005 11:49:15 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <m2zmsgi657.fsf@soma.local> (Marco Baringer's message of "Thu,
	21 Jul 2005 16:29:56 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here> <m2zmsgi657.fsf@soma.local>
Message-ID: <87k6jfgqqs.fsf@flarge.here>

Well you have explained how things work but I'm missing one thing. 
How to call a tal file from within a defaction.

at the moment the code is
defaction
...
                               (make-instance 'trans-product
                                              :product-id (id (product item))
                                              :tag (tag (product item))
                                              :num-copies copies
                                              :distribution ""
                                              :usage ""
                                              :price price))))
                  ;; (inspect obj)
                (push obj prod-list)))))

    (call 'trans-info
          :products-in-basket (nreverse prod-list))))


Now if I cal 'trans-info then in the end I call the render-on
method. But what do I have to do "call" a tal template file?

Regards
Friedrich


From mb at bese.it  Mon Jul 25 10:10:19 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 25 Jul 2005 12:10:19 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <87k6jfgqqs.fsf@flarge.here> (Friedrich Dominicus's message of
	"Mon, 25 Jul 2005 11:49:15 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here> <m2zmsgi657.fsf@soma.local>
	<87k6jfgqqs.fsf@flarge.here>
Message-ID: <m2fyu3mc1g.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> Well you have explained how things work but I'm missing one thing. 
> How to call a tal file from within a defaction.
>
> at the moment the code is
> defaction
> ...
>                                (make-instance 'trans-product
>                                               :product-id (id (product item))
>                                               :tag (tag (product item))
>                                               :num-copies copies
>                                               :distribution ""
>                                               :usage ""
>                                               :price price))))
>                   ;; (inspect obj)
>                 (push obj prod-list)))))
>
>     (call 'trans-info
>           :products-in-basket (nreverse prod-list))))
>
>
> Now if I cal 'trans-info then in the end I call the render-on
> method. But what do I have to do "call" a tal template file?

i've been thinking about a 'show' macro (and accompanying component)
for quite a while. this is as good as a time as any to write one
(untested but conceptually correct):

(defcomponent show-template-component (template-component)
  ((env :accessor env :initarg :env)))

(defmethod template-component-environment nconc
    ((component show-template-component))
  (env component))

(defmacro show (page-name &rest env)
  `(call 'show-template-component :template-tame ,page-name
                                  :env (tal-env , at env)))

you'd then use this in an action like this:

(defaction 
  ...
  (show "whatever.tal"
        :products-in-basket (mapcar (lambda (product)
                                      (tal-env 'item product))
                                    prod-list)))

and then you'd access the prod-list with something like:

<ul>
  <li tal:dolist="$:products-in-basket">
    <b tal:content="$item">ITEM GOES HERE</b>
  </li>
</ul>

avoiding the ugly $:products-in-basket is fairly easy, just pass
'products-in-basket ta show instead of :products-in-basket and make
sure you add a tal:in-package attribute somewhere in the tal file.

since this is a component the show macro won't return until your
template calls answer, this can be done easily by adding a like whcih
calls the OK action (all components can do this and it very simply
calls answer):

<ucw:a action="(ok $component)">Done.</a>

(the $component tal environment variable is setup by the
template-component class and, the symbol itself, is exported from the
ucw package).

hth.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Mon Jul 25 10:12:54 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 25 Jul 2005 12:12:54 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <m2fyu3mc1g.fsf@soma.local> (Marco Baringer's message of "Mon,
	25 Jul 2005 12:10:19 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here> <m2zmsgi657.fsf@soma.local>
	<87k6jfgqqs.fsf@flarge.here> <m2fyu3mc1g.fsf@soma.local>
Message-ID: <m24qajmbx5.fsf@soma.local>

"Marco Baringer" <mb at bese.it> writes:

> (defmethod template-component-environment nconc
>     ((component show-template-component))
>   (env component))

this is incorrect. try this instead:

(defmethod template-component-environment nconc
    ((component show-template-component))
  (copy-list (env component)))

(nconc method combination is free to modify the value returned by the
methods, most template components build up a new environment object
each time so this was never a problem).


-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Mon Jul 25 12:18:24 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 25 Jul 2005 14:18:24 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <87k6jfgqqs.fsf@flarge.here> (Friedrich Dominicus's message of
	"Mon, 25 Jul 2005 11:49:15 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here> <m2zmsgi657.fsf@soma.local>
	<87k6jfgqqs.fsf@flarge.here>
Message-ID: <m2ackbkrjj.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> How to call a tal file from within a defaction.

the show macro (and a corresponding show-window macro for window
components) has been added to ucw's ucw_dev branch. i changed the wiki
to use show-window to provide an example.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Mon Jul 25 15:15:19 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 25 Jul 2005 17:15:19 +0200
Subject: [Bese-devel] TAL und yaclml
In-Reply-To: <m2ackbkrjj.fsf@soma.local> (Marco Baringer's message of "Mon,
	25 Jul 2005 14:18:24 +0200")
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here> <m2zmsgi657.fsf@soma.local>
	<87k6jfgqqs.fsf@flarge.here> <m2ackbkrjj.fsf@soma.local>
Message-ID: <87ackahq7s.fsf@flarge.here>

Thanks, I probably will make good (or worse ;-) use of it.

But one step forward and a new question arises.
- I do dislike JavaScript as one can imagine.

Now if I use a ucw:button I will be left with JavaScript. This is not
a good thing for me. So I wonder if there is another way to associate
a button with an action. For the time beeing I use this "ugly" code:

in the form I write something along this lines:
              (<ucw:form 
               :action (user-details prods)
               (<:p
                (<:input :type "submit" :value "Yes"
                         :name "am-customer")
                (<:input :type "submit" :value "No"
                         :name "")))))
      


and in the action I write:
(defaction user-details ((transaction trans-info))
  ;; (inspect (ucw::all-params (ucw::context.request *context*)))
  (let ((btn-text (ucw::get-parameter (ucw::context.request *context*) "am-customer")))
    (if btn-text
      (call 'shop-login :message "You customer data please"))
    (call 'info-message :message "Not yet a customer")))

What I dislike about is accessing internal symbols and the global
*context* is there a cleaner way without JavaScript?

Regards
Friedrich


From frido at q-software-solutions.de  Mon Jul 25 15:17:47 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 25 Jul 2005 17:17:47 +0200
Subject: [Bese-devel] a button panel?
Message-ID: <8764uyhq3o.fsf@flarge.here>

I think having such a thing would be very nice. I would prefer using
it instead of the options-dialog. Has anyone written such a beast? 

Regards
Friedrich


From billclem at gmail.com  Mon Jul 25 22:46:34 2005
From: billclem at gmail.com (Bill Clementson)
Date: Mon, 25 Jul 2005 15:46:34 -0700
Subject: [Bese-devel] Bug in new cps ucw examples
Message-ID: <8757cb490507251546406cbd2b@mail.gmail.com>

This has probably already been reported; however, when testing the
examples with the new cps ucw (with new cps arnesi), I found that most
of the OK buttons result in an error. Below is an example of what I
get when I press the OK button on the first page of the presentations
example.

Cheers,
Bill

No applicable method for args:
 (#<STANDARD-SESSION 0 NIL #x683CCF6>
  ("BeljIKBccvwCitjKTLHG" "BeljIKBccvwCitjKTLHG"))
 to #<STANDARD-GENERIC-FUNCTION IT.BESE.UCW::FIND-FRAME #x6634F46>
   [Condition of type SIMPLE-ERROR]

Restarts:
  0: [SERVER-ERROR] Send the client an internal server error page.
  1: [ABORT-BREAK] Reset this process
  2: [ABORT] Kill this process

Backtrace:
  0: (#<STANDARD-METHOD SWANK-BACKEND:COMPUTE-BACKTRACE (T T)> 0 20)
  1: (SWANK:BACKTRACE 1 40)
  2: (CCL::CALL-CHECK-REGS 'SWANK:BACKTRACE)
  3: (#<Anonymous Function #x6575A1E> '(SWANK:BACKTRACE 1 40) 11 'NIL)
  4: (#<STANDARD-METHOD SWANK-BACKEND:CALL-WITH-DEBUGGER-HOOK (T T)>
#<Compiled-function SWANK:SWANK-DEBUGGER-HOOK #x658811E>
#<COMPILED-LEXICAL-CLOSURE #x689D946>)
  5: (FUNCALL 'SWANK::EVAL-FOR-EMACS)
  6: (SWANK::SLDB-LOOP '(T 200 T NIL 10 T
#S(CCL::PPRINT-DISPATCH-TABLE :CONSES-WITH-CARS #<HASH-TABLE :TEST EQ
size 60/60 #x63971CE> :STRUCTURES NIL :OTHERS (# #)) NIL T 10 ...))
  7: (#<STANDARD-METHOD SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT
(T)> #<Anonymous Function #x6587ABE>)
  8: (SWANK::CALL-WITH-BINDINGS '((*PRINT-PRETTY*) (*PRINT-LEVEL* . 4)
(*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*)
(*PRINT-PPRINT-DISPATCH* . #S(CCL::PPRINT-DISPATCH-TABLE
:CONSES-WITH-CARS #<HASH-TABLE :TEST EQ size 60/60 #x63971CE>
:STRUCTURES NIL :OTHERS #)) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10)
(*PRINT-RADIX*) (*PRINT-ARRAY* . T) ...) #<Anonymous Function
#x6587AFE>)
  9: (SWANK::DEBUG-IN-EMACS #<SIMPLE-ERROR #x683C616>)
 10: (#<Anonymous Function #x65613C6> #<COMPILED-LEXICAL-CLOSURE #x683C5DE>)
 11: (SWANK::CALL-WITH-REDIRECTED-IO #<COMPILED-LEXICAL-CLOSURE
#x683C5C6> #<SIMPLE-ERROR #x683C616>)
 12: (SWANK::CALL-WITH-CONNECTION #<CONNECTION  #x63D0D2E>
#<COMPILED-LEXICAL-CLOSURE #x683C5DE>)
 13: (#<Anonymous Function #x66051D6> #<SIMPLE-ERROR #x683C616>)
 14: (SIGNAL #<SIMPLE-ERROR #x683C616>)


From mb at bese.it  Tue Jul 26 10:49:01 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 26 Jul 2005 12:49:01 +0200
Subject: [Bese-devel] ucw movie
Message-ID: <m2iryxc06a.fsf@soma.local>


yesterday i made a small (20min) movie of ucw. it was a lot of fun and
i think i'll do it again, but first i'd like to get some feed back on
it (i didn't really have much of a plan when doing it, i just sat down
and started talking, and it shows). so, does anybody feel like
downloading a huge (233MB) quicktime movie and sending me some
feedback? or, is there who'd bother to compress a huge quicktime movie
should i make another one?

the movie starts up ucw and then walks through the creation a simple
hello world application.

http://common-lisp.net/project/ucw/docs/hello-world.mov

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Tue Jul 26 10:53:49 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 26 Jul 2005 12:53:49 +0200
Subject: [Bese-devel] ucw movie
References: <m2iryxc06a.fsf@soma.local>
Message-ID: <m264uxbzya.fsf@soma.local>

"Marco Baringer" <mb at bese.it> writes:

> yesterday i made a small (20min) movie of ucw. it was a lot of fun and
                     ^^^^^

that should read short, not small.

> i think i'll do it again, but first i'd like to get some feed back on
> it (i didn't really have much of a plan when doing it, i just sat down
> and started talking, and it shows). so, does anybody feel like
> downloading a huge (233MB) quicktime movie and sending me some
> feedback? or, is there who'd bother to compress a huge quicktime movie
> should i make another one?
>
> the movie starts up ucw and then walks through the creation a simple
> hello world application.
>
> http://common-lisp.net/project/ucw/docs/hello-world.mov

p.s. - given the size of this and the quality i do plan on redoing it
shortly, so don't announce this to the entire world (says he who posts
about it on a public, google spidered, mailing list...)

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Tue Jul 26 10:10:40 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 26 Jul 2005 12:10:40 +0200
Subject: [Bese-devel] small patch for XHTML strict
Message-ID: <874qah513z.fsf@flarge.here>

It seems to validate the generated code as XHTML1.1 strict one has to
add a <p> around the input fields. 

Here's a small diff you may want to apply (or forget)

--- ucw-tags.lisp	2005-07-26 12:03:06.000000000 +0200
+++ ucw-tags.lisp.old	2005-07-26 12:07:32.000000000 +0200
@@ -52,9 +52,8 @@
                                                           (lambda () ,action)))))
        (<:form :action (print-uri-to-string-sans-query ,url)
                , at others
-        (<:p :class "hidden-fields"
          (dolist (,query (uri.query ,url))
-           (<:input :type "hidden" :name (car ,query) :value (cdr ,query))))
+           (<:input :type "hidden" :name (car ,query) :value (cdr ,query)))
          , at body))))
 
 (deftag-macro <ucw::%select (&attribute multiple writer accessor (test '#'eql) (key '#'identity) on-change


Another thing I've changed is extending the window-component to be
even more paremetrized. If anyone is intersted, just drop me a note.

Regards
Friedrich



From mb at bese.it  Tue Jul 26 11:35:18 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 26 Jul 2005 13:35:18 +0200
Subject: [Bese-devel] small patch for XHTML strict
In-Reply-To: <874qah513z.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 26 Jul 2005 12:10:40 +0200")
References: <874qah513z.fsf@flarge.here>
Message-ID: <m21x5lby15.fsf@soma.local>

Friedrich Dominicus <frido at q-software-solutions.de> writes:

> It seems to validate the generated code as XHTML1.1 strict one has to
> add a <p> around the input fields. 
>
> Here's a small diff you may want to apply (or forget)

thanks! i'll apply this later this evening.

> Another thing I've changed is extending the window-component to be
> even more paremetrized. If anyone is intersted, just drop me a note.

i'm interested.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From darkman_spam at yahoo.fr  Tue Jul 26 16:05:02 2005
From: darkman_spam at yahoo.fr (drkm)
Date: Tue, 26 Jul 2005 18:05:02 +0200 (CEST)
Subject: [Bese-devel] Error while installing
In-Reply-To: <m2wtnh2lco.fsf@soma.local>
Message-ID: <20050726160503.13420.qmail@web25803.mail.ukl.yahoo.com>

Marco Baringer writes:

> that said, ucw is nicer to work with if the underlying lisp is
> multi-threaded, why don't you try with sbcl or cumcl?

  CLisp was the only Common Lisp implementation available trough the
'aptitude(8)' Debian package manager.  I'll install SBCL and retest
installing UCW.

  Thanks,

--drkm



















	

	
		
___________________________________________________________________________ 
Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger 
T?l?chargez cette version sur http://fr.messenger.yahoo.com


From frido at q-software-solutions.de  Tue Jul 26 16:23:02 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 26 Jul 2005 18:23:02 +0200
Subject: [Bese-devel] small patch for XHTML strict
In-Reply-To: <m21x5lby15.fsf@soma.local> (Marco Baringer's message of "Tue,
	26 Jul 2005 13:35:18 +0200")
References: <874qah513z.fsf@flarge.here> <m21x5lby15.fsf@soma.local>
Message-ID: <87r7dl35ax.fsf@flarge.here>

"Marco Baringer" <mb at bese.it> writes:

>
>> Another thing I've changed is extending the window-component to be
>> even more paremetrized. If anyone is intersted, just drop me a note.
>
> i'm interested.
Please be aware: It's very much influx. I'm tying to wrap around my
brain for "UCW" programming. This is a first protyp which works for me
but won't for you because you are lacking a few of the writer
functions. 

If I will go to the point where the applicaton I try to implement will
run here, I may think about more parametrization of all the things,
that others will be able to use it. 

I'm thinking that one should use a separate "write-header"
function, but I did not do it yet. 

I think the documentation in window.lisp is outdated. I cite:

"Subclass of simple-window-component should implement the
render-body method, not render-on."))"

There's no where a render method to find.




(in-package :qss.web.ucw)


(defclass qss-window (window-component)
  ((title :accessor title
          :initarg :title
          :initform "Q Software Solutions" )
   (html-version :accessor html-version
                 :initarg :html-version
                 :initform +xhtml1.0-strict+)
                 
   (meta-tags :accessor meta-tags
              :initarg :meta-tags
              :initform nil)
   (menu-index :accessor menu-index
               :initarg :menu-index)
   (write-header :initform nil)
   (write-menu-fun :accessor write-menu-fun
               :initarg :write-menu-fun
               :initform #'(lambda (stream index)
                             (write-main-menu :stream stream
                                              :active index)))
   (write-footer-fun :accessor write-footer-fun
                     :initarg :write-footer-fun
                     :initform #'(lambda (stream)
                                   (write-footer :stream stream)))
                                      
   (stylesheet :accessor stylesheet
               :initarg :stylesheet
               :initform "/css/qss.css"
               :documentation "The URL of the css file to use as a stylesheet for this window.")
   (content-type :accessor content-type
                 :initarg :content-type
                 :initform "text/html"
                 :documentation "The Content-Type header for the
                 http response (also used in the meta tag)"))
  (:metaclass standard-component-class)
  (:documentation "The QSS standard page class"))



                                
(defmethod render-on :wrapping ((res response) (window qss-window))
  "This convience method assumes: 1) the stylesheet is
external (as opposed to inlined) or is not used; 2) no JavaScript at all."
  (let ((stream (content-stream res)))

    (<:as-is (html-version window))
    (<:html
     (<:head
      (dolist (meta-tag (meta-tags window))
        (<:as-is (format nil "<meta ~a=\"~a\" content=\"~a\" />~%" 
                (first meta-tag) (second meta-tag) (third meta-tag))))

      (<:title (<:as-is (title window)))
      (when (stylesheet window)
        (<:link :rel "stylesheet"
                :href (stylesheet window)
                :type "text/css")))
     (<:body
      (funcall (write-menu-fun window) stream (menu-index window))
      (call-next-method)
      (funcall (write-footer-fun window) stream)))))

       
       
       

;; test page


#|
(defclass simple-qss-window (qss-window)
  ((menu-index :initform +home-index+)
   (meta-tags :initform +no-cache+)
   (title :initform "Simple page"))
  (:metaclass standard-component-class))



(defmethod render-on ((res response) (window simple-qss-window))
  (declare (ignore res window))
  (<:h1 "Just one header in the body"))
  
    
 (defentry-point "qss-window.ucw" (:application *shop*) ()
  (call 'simple-qss-window))
|#


I'm using this code like this at the moment:
(defcomponent customer-input-page (qss-window)
  ((menu-index :initform +store-index+)
   (customer-data :initarg :customer-data :accessor customer-data)
   (address :initarg :address :accessor address))
  (:metaclass standard-component-class)
  (:default-initargs :title "You personal data"))


(defmethod render-on ((res response) (address qss.db::shipping-address))
  (<:table 
   (<:tr
    (<:td "Street")
    (<:td (<:as-is (slot-value address 'street))))
   (<:tr 
    (<:td "ZIP")
    (<:td (<:as-is (slot-value address 'zip))))))

The output is like this:

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ucw-snapshot1.png
Type: image/png
Size: 65565 bytes
Desc: Snapshot of the modified standard page
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050726/fe82e27c/attachment.png>
-------------- next part --------------


Regards
Friedrich

From smoof-ra at elder-gods.org  Tue Jul 26 17:02:17 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Tue, 26 Jul 2005 13:02:17 -0400
Subject: [Bese-devel] declarations
Message-ID: <20050726170217.GA17394@melkor.elder-gods.org>


Marco: could you have walk-form not throw away declarations?  Or would
you accept a patch to that effect?

    --larry 


From ran at setq.net  Tue Jul 26 17:58:49 2005
From: ran at setq.net (ran at setq.net)
Date: Tue, 26 Jul 2005 17:58:49 +0000
Subject: [Bese-devel] TAL und yaclml (Form actions and javascript)
In-Reply-To: <87ackahq7s.fsf@flarge.here>
References: <m27jfot3rp.fsf@soma.local> <87zmsjcpqy.fsf@flarge.here>
	<878y02d8ro.fsf_-_@flarge.here> <m2zmsgi657.fsf@soma.local>
	<87k6jfgqqs.fsf@flarge.here> <m2ackbkrjj.fsf@soma.local>
	<87ackahq7s.fsf@flarge.here>
Message-ID: <20050726175849.GA11543@setq.net>

Greetings,

Friedrich Dominicus wrote:
> Now if I use a ucw:button I will be left with JavaScript. This is not
[...]

It seems that <button> and <input type="button"> do not cause form submission,
and really are to be used with javascript, but your sample uses a submit button.

Here is a way to have form actions without javascript, by patching <ucw:input
and adding a new function (setf get-parameter) to whatever backend you are using.

It works by setting the action parameter, ("a"), to what it would be set to by javascript,
in a callback for the parameter of the input tag's name attribute.
This means that you can not use the :name attribute with <ucw:input

I would write your form like this:
(<ucw:form
 (<:p (<ucw:input :type "submit" :value "Yes" :action (call 'shop-login :message "You...")
      (<ucw:input :type "submit" :value "No" :action (call 'info-message :message "Not...")))

I guess this is what you would already be doing if you liked javascript.

Below is the patch and (setf get-parameter) methods for araneida and mod-lisp/httpd
Wash hands thoroughly after applying.
It might be cleaner if the action could be selected through means other than
the "a" parameter.
for example allowing callbacks to select the action, like:
(make-new-callback (context.current-frame *context*)
  (lambda (value)
    (push action (context.requested-actions *context*)))

Ryan

;; Sorry, no patches for (setf get-parameter)
;; araneida: src/backend/araneida.lisp
;; WARNING: this will only set a POST parameter
(defmethod (setf get-parameter) (value (request araneida-request) parameter-name)
  (let ((cons (assoc parameter-name (araneida:request-body (request request)) :test #'string-equal)))
    (when cons (setf (cadr cons) value))))

;; httpd/mod-lisp: src/backend/httpd.lisp
(defmethod (setf get-parameter) (value (request httpd-request) name)
  (let ((cons (assoc name (parameters request) :test #'string=)))
    (when cons (setf (cdr cons) (copy-seq value)))))

diff -ur /src/marco/ucw/src/yaclml/ucw-tags.lisp ./src/yaclml/ucw-tags.lisp
--- /src/marco/ucw/src/yaclml/ucw-tags.lisp	2005-07-20 15:06:49.000000000 +0900
+++ ./src/yaclml/ucw-tags.lisp	2005-07-27 01:12:52.000000000 +0900
@@ -31,7 +31,7 @@
 ;;;; *** Form tags
 
 (deftag-macro <ucw:button (&attribute action &allow-other-attributes others &body body)
-  "Presents an butten tag which executes ACTION when clicked."
+  "Presents a button tag which executes ACTION when clicked."
   (unless action
     (error "Must specify :ACTION attribute in button tags (or use a simple <:button tag)"))  
   `(<:button :onclick (concatenate 'string
@@ -149,13 +149,22 @@
 			 `(:checked ,value))
 		     , at others)))))
     (action
-     `(<:input , at others
-               :onclick (concatenate 'string
-                                     "javascript: "
-                                     "this.form." +action-parameter-name+ ".value ='"
-                                     (make-new-action (context.current-frame *context*)
-                                                      (lambda () ,action)) "';"
-                                     "return true;")))
+     `(let ((action (make-new-action (context.current-frame *context*)
+				      (lambda () ,action))))
+	 (<:input , at others
+		  :name (make-new-callback (context.current-frame *context*)
+					   (lambda (value)
+					     (declare (ignore value))
+					     (setf (get-parameter
+						    (context.request *context*)
+						    +action-parameter-name+)
+						   action)))
+		  :onclick (concatenate 'string
+					"javascript: "
+					"this.form." +action-parameter-name+ ".value ='"
+					action
+					"';"
+					"return true;"))))
     (t
      (error "Must specify either :ACTION or :ACCESSOR (or use a regular <:input tag.)"))))
 


From der_julian at web.de  Tue Jul 26 21:09:33 2005
From: der_julian at web.de (Julian Stecklina)
Date: Tue, 26 Jul 2005 23:09:33 +0200
Subject: [Bese-devel] ucw movie
In-Reply-To: <m2iryxc06a.fsf@soma.local>
References: <m2iryxc06a.fsf@soma.local>
Message-ID: <20050726230933.2163d4cb@localhost>

On Tue, 26 Jul 2005 12:49:01 +0200
"Marco Baringer" <mb at bese.it> wrote:

> 
> yesterday i made a small (20min) movie of ucw. it was a lot of fun and
> i think i'll do it again, but first i'd like to get some feed back on
> it (i didn't really have much of a plan when doing it, i just sat down
> and started talking, and it shows). so, does anybody feel like
> downloading a huge (233MB) quicktime movie and sending me some
> feedback? or, is there who'd bother to compress a huge quicktime movie
> should i make another one?
> 
> the movie starts up ucw and then walks through the creation a simple
> hello world application.

The video is *really* nice and I think it will attract more people to
UCW. And btw, I want a Mac. ;-)

I am currently reencoding it as XVID AVI, but I think the quality will
be to low to be really useful. If someone tells me how to efficiently
encode videos with lots of stills, this would be very helpful. :)

Regards,
-- 
Julian Stecklina

LISP has survived for 21 years because it is an approximate local
optimum in the space of programming languages. - John McCarthy (1980)




From der_julian at web.de  Tue Jul 26 21:41:11 2005
From: der_julian at web.de (Julian Stecklina)
Date: Tue, 26 Jul 2005 23:41:11 +0200
Subject: [Bese-devel] ucw movie
In-Reply-To: <20050726230933.2163d4cb@localhost>
References: <m2iryxc06a.fsf@soma.local>
	<20050726230933.2163d4cb@localhost>
Message-ID: <20050726234111.2931e411@localhost>

On Tue, 26 Jul 2005 23:09:33 +0200
Julian Stecklina <der_julian at web.de> wrote:

> On Tue, 26 Jul 2005 12:49:01 +0200
> "Marco Baringer" <mb at bese.it> wrote:
> 
> > 
> > yesterday i made a small (20min) movie of ucw. it was a lot of fun
> > and i think i'll do it again, but first i'd like to get some feed
> > back on it (i didn't really have much of a plan when doing it, i
> > just sat down and started talking, and it shows). so, does anybody
> > feel like downloading a huge (233MB) quicktime movie and sending me
> > some feedback? or, is there who'd bother to compress a huge
> > quicktime movie should i make another one?
> > 
> > the movie starts up ucw and then walks through the creation a simple
> > hello world application.
> 
> The video is *really* nice and I think it will attract more people to
> UCW. And btw, I want a Mac. ;-)
> 
> I am currently reencoding it as XVID AVI, but I think the quality will
> be to low to be really useful. If someone tells me how to efficiently
> encode videos with lots of stills, this would be very helpful. :)

I have it down to 48 MB with close to optimal quality. I'd be happy to
upload it somewhere.

Regards,
-- 
Julian Stecklina

LISP has survived for 21 years because it is an approximate local
optimum in the space of programming languages. - John McCarthy (1980)




From asf at boinkor.net  Tue Jul 26 22:19:34 2005
From: asf at boinkor.net (Andreas Fuchs)
Date: Wed, 27 Jul 2005 00:19:34 +0200
Subject: [Bese-devel] ucw movie
In-Reply-To: <20050726234111.2931e411@localhost>
References: <m2iryxc06a.fsf@soma.local> <20050726230933.2163d4cb@localhost>
	<20050726234111.2931e411@localhost>
Message-ID: <87d5p543d5.wl%asf@boinkor.net>

On 2005-07-26, Julian Stecklina <der_julian at web.de> wrote:
>> I am currently reencoding it as XVID AVI, but I think the quality
>> will be to low to be really useful. If someone tells me how to
>> efficiently encode videos with lots of stills, this would be very
>> helpful. :)
>
> I have it down to 48 MB with close to optimal quality. I'd be happy
> to upload it somewhere.

The purpose of the bittorrent tracker at
http://boinkor.net:6969/announce is to help distribute good lisp
porn. Go wild (-:

(with the vanilla bittorrent distribution, making a torrent is as
simple as running:

$ btmakemetafile.py hello-world.avi http://boinkor.net:6969/announce
# upload the resulting .torrent file somewhere
$ btdownloadcurses.py --max_upload_rate <whatever> --url http://example.com/hello-world.avi.torrent
# tell other people where to find the .torrent file
)

Have fun,
-- 
Andreas Fuchs, <asf at boinkor.net>, asf at jabber.at, antifuchs


From der_julian at web.de  Tue Jul 26 23:04:50 2005
From: der_julian at web.de (Julian Stecklina)
Date: Wed, 27 Jul 2005 01:04:50 +0200
Subject: [Bese-devel] ucw movie
In-Reply-To: <87d5p543d5.wl%asf@boinkor.net>
References: <m2iryxc06a.fsf@soma.local> <20050726230933.2163d4cb@localhost>
	<20050726234111.2931e411@localhost> <87d5p543d5.wl%asf@boinkor.net>
Message-ID: <20050727010450.5144bce4@localhost>

On Wed, 27 Jul 2005 00:19:34 +0200
Andreas Fuchs <asf at boinkor.net> wrote:

> On 2005-07-26, Julian Stecklina <der_julian at web.de> wrote:
> >> I am currently reencoding it as XVID AVI, but I think the quality
> >> will be to low to be really useful. If someone tells me how to
> >> efficiently encode videos with lots of stills, this would be very
> >> helpful. :)
> >
> > I have it down to 48 MB with close to optimal quality. I'd be happy
> > to upload it somewhere.
> 
> The purpose of the bittorrent tracker at
> http://boinkor.net:6969/announce is to help distribute good lisp
> porn. Go wild (-:
> 
> (with the vanilla bittorrent distribution, making a torrent is as
> simple as running:
> 
> $ btmakemetafile.py hello-world.avi http://boinkor.net:6969/announce
> # upload the resulting .torrent file somewhere
> $ btdownloadcurses.py --max_upload_rate <whatever> --url
> http://example.com/hello-world.avi.torrent # tell other people where
> to find the .torrent file )

http://www.inf.tu-dresden.de/~s1054849/ucw.hello.world.avi.torrent

Regards,
-- 
Julian Stecklina

LISP has survived for 21 years because it is an approximate local
optimum in the space of programming languages. - John McCarthy (1980)




From der_julian at web.de  Tue Jul 26 23:34:35 2005
From: der_julian at web.de (Julian Stecklina)
Date: Wed, 27 Jul 2005 01:34:35 +0200
Subject: [Bese-devel] ucw movie
In-Reply-To: <20050727010450.5144bce4@localhost>
References: <m2iryxc06a.fsf@soma.local> <20050726230933.2163d4cb@localhost>
	<20050726234111.2931e411@localhost> <87d5p543d5.wl%asf@boinkor.net>
	<20050727010450.5144bce4@localhost>
Message-ID: <20050727013435.45ca1698@localhost>

On Wed, 27 Jul 2005 01:04:50 +0200
Julian Stecklina <der_julian at web.de> wrote:

> http://www.inf.tu-dresden.de/~s1054849/ucw.hello.world.avi.torrent

You'll also find it at:

http://www.fasl.info/ucw.hello.world.avi.torrent
http://www.noctua.co.uk/test/ucw.hello.world.avi.torrent

Thank the nice people in #lisp

Regards,
-- 
Julian Stecklina

LISP has survived for 21 years because it is an approximate local
optimum in the space of programming languages. - John McCarthy (1980)




From mb at bese.it  Wed Jul 27 05:26:43 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 27 Jul 2005 07:26:43 +0200
Subject: [Bese-devel] declarations
References: <20050726170217.GA17394@melkor.elder-gods.org>
Message-ID: <m2ll3syg30.fsf@soma.local>

Larry D'Anna <smoof-ra at elder-gods.org> writes:

> Marco: could you have walk-form not throw away declarations?  Or would
> you accept a patch to that effect?

no problem, where do you want me to put them? is a new slot on
implicit-progn enough or should be go all the way and store them along
with the variables/functions?

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From frido at q-software-solutions.de  Wed Jul 27 08:57:07 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Wed, 27 Jul 2005 10:57:07 +0200
Subject: [Bese-devel] ucw movie
In-Reply-To: <20050726230933.2163d4cb@localhost> (Julian Stecklina's message
	of "Tue, 26 Jul 2005 23:09:33 +0200")
References: <m2iryxc06a.fsf@soma.local> <20050726230933.2163d4cb@localhost>
Message-ID: <874qagsk2k.fsf@flarge.here>

Unfortunatly it just shows the easy part. It does not reveal how
action, tal-files etc come into play. It does not give a good
development approach or an idea on how to develop new components. 
It's IMHO in the same league as the Squeak Seaside tutorial..

 

Regards
Friedrich


From mb at bese.it  Wed Jul 27 09:34:17 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 27 Jul 2005 11:34:17 +0200
Subject: [Bese-devel] uncommon web introduction movie: hello-world.mov
Message-ID: <m2ack8ego6.fsf@soma.local>


http://common-lisp.net/project/ucw/docs/hello-world.mov

The contents of the movie are basically the same (no need to watch it
if you've already seen the other one), the quality is 100% but the
resolution has been dropped down to 800x600 and the file size is
"only" 51MB. As with before this a quicktime encoded mpeg-4.

enjoy.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From asbjxrn at bjxrnstad.net  Wed Jul 27 10:35:32 2005
From: asbjxrn at bjxrnstad.net (=?ISO-8859-1?Q?Asbj=F8rn_Bj=F8rnstad?=)
Date: Wed, 27 Jul 2005 18:35:32 +0800
Subject: [Bese-devel] ucw movie
In-Reply-To: <874qagsk2k.fsf@flarge.here>
References: <m2iryxc06a.fsf@soma.local> <20050726230933.2163d4cb@localhost>
	<874qagsk2k.fsf@flarge.here>
Message-ID: <79508712-1F2C-446C-B224-FAEC169A4273@bjxrnstad.net>


On 27 Jul 2005, at 4:57 PM, Friedrich Dominicus wrote:

> Unfortunatly it just shows the easy part. It does not reveal how
> action, tal-files etc come into play. It does not give a good
> development approach or an idea on how to develop new components.
> It's IMHO in the same league as the Squeak Seaside tutorial..

I'd say it's a good first step. It gives some insight to how
the different parts of ucw fits together. (calling actions
that return values and the loop example.) Don't think of it as
shallow, but as a basis to build more complex examples later.

-- 
   -asbjxrn



From mb at bese.it  Wed Jul 27 10:37:36 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 27 Jul 2005 12:37:36 +0200
Subject: [Bese-devel] uncommon web introduction movie: hello-world.mov
In-Reply-To: <m2ack8ego6.fsf@soma.local> (Marco Baringer's message of "Wed,
	27 Jul 2005 11:34:17 +0200")
References: <m2ack8ego6.fsf@soma.local>
Message-ID: <m2mzo8cz67.fsf@soma.local>

"Marco Baringer" <mb at bese.it> writes:

> http://common-lisp.net/project/ucw/docs/hello-world.mov

torrent available at:

http://common-lisp.net/project/ucw/docs/hello-world.torrent

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From pdf23ds at gmail.com  Wed Jul 27 02:03:23 2005
From: pdf23ds at gmail.com (Chris Capel)
Date: Tue, 26 Jul 2005 21:03:23 -0500
Subject: [Bese-devel] ucw movie
In-Reply-To: <20050727013435.45ca1698@localhost>
References: <m2iryxc06a.fsf@soma.local> <20050726230933.2163d4cb@localhost>
	<20050726234111.2931e411@localhost> <87d5p543d5.wl%asf@boinkor.net>
	<20050727010450.5144bce4@localhost>
	<20050727013435.45ca1698@localhost>
Message-ID: <737b61f30507261903a395e14@mail.gmail.com>

It's not supported on a mac (not even players, I don't think), but
Camtasia (http://www.techsmith.com/) produces some incredibly small
videos for most desktop stuff. Its compression algorithm is tailored
to the capture of computer sessions. Maybe there's something similar
yet cross-platform. But compression algorithms designed for
full-motion video are of course going to do a horrible job on screen
capture.

Chris Capel
-- 
"What is it like to be a bat? What is it like to bat a bee? What is it
like to be a bee being batted? What is it like to be a batted bee?"
-- The Mind's I (Hofstadter, Dennet)


From smoof-ra at elder-gods.org  Wed Jul 27 15:07:32 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 27 Jul 2005 11:07:32 -0400
Subject: [Bese-devel] call/cc and unwind-protect
Message-ID: <20050727150732.GA12047@melkor.elder-gods.org>


Is call/cc fundamentally incompatible with unwind-protect?  Is there
any way unwind-protect can be weakened to be compatible with call/cc
but still do The Right Thing?


    --larry 


From smoof-ra at elder-gods.org  Wed Jul 27 15:12:40 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 27 Jul 2005 11:12:40 -0400
Subject: [Bese-devel] declarations
In-Reply-To: <m2ll3syg30.fsf@soma.local>
References: <20050726170217.GA17394@melkor.elder-gods.org>
	<m2ll3syg30.fsf@soma.local>
Message-ID: <20050727151240.GA7447@melkor.elder-gods.org>

(sorry for sending this to you twice marco)

* Marco Baringer (mb at bese.it) [050727 01:26]:
> Larry D'Anna <smoof-ra at elder-gods.org> writes:
> 
> > Marco: could you have walk-form not throw away declarations?  Or would
> > you accept a patch to that effect?
> 
> no problem, where do you want me to put them? is a new slot on
> implicit-progn enough or should be go all the way and store them along
> with the variables/functions?

A new slot on implicit-progn is fine for what I'm doing.  Thanks!

      --larry 


From smoof-ra at elder-gods.org  Wed Jul 27 15:26:32 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 27 Jul 2005 11:26:32 -0400
Subject: [Bese-devel] trivial patch
Message-ID: <20050727152632.GA13509@melkor.elder-gods.org>


what's the preffered way of sending in patches?  send them to the list
or just "darcs send"?  anyway here's a trivial one.
-------------- next part --------------

New patches:

[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] {
hunk ./src/walk.lisp 331
-                     (push (make-instance 'allow-other-keys-function-argument
+                     (push (make-instance 'allow-other-keys-function-argument-form
}

Context:

[defmethod arguments should be ignorable, not ignore (openmcl warns whenever you ignore a specialized argument)
Marco Baringer <mb at bese.it>**20050726090308] 
[Typo in extract-argument-names
Marco Baringer <mb at bese.it>**20050726090256] 
[Fix generation of defmethod froms from defmethod/cc; added tests
Marco Baringer <mb at bese.it>**20050726085226] 
[Fix handling of optional arguments in apply-cps-lambda
Marco Baringer <mb at bese.it>**20050726085155] 
[More tests
Marco Baringer <mb at bese.it>**20050723133158] 
[Fix a bug in the handling of the case when LOOKUP finds a value for a name but the value is NIL
Marco Baringer <mb at bese.it>**20050723133106] 
[Export the symbol KALL
Marco Baringer <mb at bese.it>**20050723133052] 
[Change the test for constant-form'ness in walk.lisp
Marco Baringer <mb at bese.it>**20050723113019] 
[Extend the walker to handle #'(setf foo) function names
Marco Baringer <mb at bese.it>**20050723104431] 
[Fix bug in the lambda generated for method forms
Marco Baringer <mb at bese.it>**20050720144450] 
[Added ignore declarations to defun/cc and defmethod/cc to make the compiler happy
Marco Baringer <mb at bese.it>**20050720110112] 
[Temporarily comment out the log tests
Marco Baringer <mb at bese.it>**20050720092312] 
[Rename (optional|keyword|rest)-argument-form classes to include the -form suffix
Marco Baringer <mb at bese.it>**20050720092124] 
[Allow defun/cc defined functions te be called outside of a with-call/cc
Marco Baringer <mb at bese.it>**20050720091826] 
[Added support for &optional arguments in cps closures
Marco Baringer <mb at bese.it>**20050720091658] 
[Added defgeneric/cc
Marco Baringer <mb at bese.it>**20050719153441] 
[Move the error checking code for lambda arguments into apply, not lambda (in cps interpreter)
Marco Baringer <mb at bese.it>**20050719153121] 
[More cps tests
Marco Baringer <mb at bese.it>**20050719152327] 
[Fix openmcl's definition of lexical-variables to deal with ccl::definition-environment
Marco Baringer <mb at bese.it>**20050719152230] 
[Added support to the cps interpreter forl communicating with lexical variables
Marco Baringer <mb at bese.it>**20050707094608] 
[walk.lisp depends on lexenv.lisp
Marco Baringer <mb at bese.it>**20050707093140] 
[added support for walking local varibales in the lexical environment
Marco Baringer <mb at bese.it>**20050707093027
 
 this applies to all those variables defined the envorinment object
 grabbed via an &environment macro argument. 
 
] 
[mistyped #+openmcl in lexenv.lisp
Marco Baringer <mb at bese.it>**20050707092959] 
[added src/lexenv.lisp to arnesi.asd
Marco Baringer <mb at bese.it>**20050707091127] 
[Rename src/environment.lisp
Marco Baringer <mb at bese.it>**20050707091114] 
[Initial version of environment.lisp
Marco Baringer <mb at bese.it>**20050707091019] 
[Minor docstring fixup for with-call/cc
Marco Baringer <mb at bese.it>**20050707090619] 
[Big patch including all the cps interpreter stuff upto now
Marco Baringer <mb at bese.it>**20050707083739] 
[Fix bug in handling of defclass-struct's :predicate option
Marco Baringer <mb at bese.it>**20050706105324] 
[Initial Import
Marco Baringer <mb at bese.it>**20050706101657
 This import moves arnesi from arch to darcs. Nothing has actually changed since 
 bese-2004 at common-lisp.net/arnesi--dev--1.4--patch-14
] 
[added arch stuff to boring file
Marco Baringer <mb at bese.it>**20050706101630] 
[Setup boring file
Marco Baringer <mb at bese.it>**20050706100535] 
Patch bundle hash:
d22a875966e88fe2666683277bc92822b7cee4b8

From smoof-ra at elder-gods.org  Wed Jul 27 15:38:02 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 27 Jul 2005 11:38:02 -0400
Subject: [Bese-devel] trivial patch
In-Reply-To: <20050727152632.GA13509@melkor.elder-gods.org>
References: <20050727152632.GA13509@melkor.elder-gods.org>
Message-ID: <20050727153802.GA14285@melkor.elder-gods.org>

* Larry D'Anna (smoof-ra at elder-gods.org) [050727 11:26]:
> 
> what's the preffered way of sending in patches?  send them to the list
> or just "darcs send"?  anyway here's a trivial one.

Oops that didn't quite do it.  
-------------- next part --------------

New patches:

[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] {
hunk ./src/walk.lisp 331
-                     (push (make-instance 'allow-other-keys-function-argument
+                     (push (make-instance 'allow-other-keys-function-argument-form
}

[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] {
hunk ./src/walk.lisp 319
-                 (setf env (register env :let (name argument) argument))))
+		 (unless (typep argument 'allow-other-keys-function-argument-form)
+		   (setf env (register env :let (name argument) argument)))))
}

Context:

[defmethod arguments should be ignorable, not ignore (openmcl warns whenever you ignore a specialized argument)
Marco Baringer <mb at bese.it>**20050726090308] 
[Typo in extract-argument-names
Marco Baringer <mb at bese.it>**20050726090256] 
[Fix generation of defmethod froms from defmethod/cc; added tests
Marco Baringer <mb at bese.it>**20050726085226] 
[Fix handling of optional arguments in apply-cps-lambda
Marco Baringer <mb at bese.it>**20050726085155] 
[More tests
Marco Baringer <mb at bese.it>**20050723133158] 
[Fix a bug in the handling of the case when LOOKUP finds a value for a name but the value is NIL
Marco Baringer <mb at bese.it>**20050723133106] 
[Export the symbol KALL
Marco Baringer <mb at bese.it>**20050723133052] 
[Change the test for constant-form'ness in walk.lisp
Marco Baringer <mb at bese.it>**20050723113019] 
[Extend the walker to handle #'(setf foo) function names
Marco Baringer <mb at bese.it>**20050723104431] 
[Fix bug in the lambda generated for method forms
Marco Baringer <mb at bese.it>**20050720144450] 
[Added ignore declarations to defun/cc and defmethod/cc to make the compiler happy
Marco Baringer <mb at bese.it>**20050720110112] 
[Temporarily comment out the log tests
Marco Baringer <mb at bese.it>**20050720092312] 
[Rename (optional|keyword|rest)-argument-form classes to include the -form suffix
Marco Baringer <mb at bese.it>**20050720092124] 
[Allow defun/cc defined functions te be called outside of a with-call/cc
Marco Baringer <mb at bese.it>**20050720091826] 
[Added support for &optional arguments in cps closures
Marco Baringer <mb at bese.it>**20050720091658] 
[Added defgeneric/cc
Marco Baringer <mb at bese.it>**20050719153441] 
[Move the error checking code for lambda arguments into apply, not lambda (in cps interpreter)
Marco Baringer <mb at bese.it>**20050719153121] 
[More cps tests
Marco Baringer <mb at bese.it>**20050719152327] 
[Fix openmcl's definition of lexical-variables to deal with ccl::definition-environment
Marco Baringer <mb at bese.it>**20050719152230] 
[Added support to the cps interpreter forl communicating with lexical variables
Marco Baringer <mb at bese.it>**20050707094608] 
[walk.lisp depends on lexenv.lisp
Marco Baringer <mb at bese.it>**20050707093140] 
[added support for walking local varibales in the lexical environment
Marco Baringer <mb at bese.it>**20050707093027
 
 this applies to all those variables defined the envorinment object
 grabbed via an &environment macro argument. 
 
] 
[mistyped #+openmcl in lexenv.lisp
Marco Baringer <mb at bese.it>**20050707092959] 
[added src/lexenv.lisp to arnesi.asd
Marco Baringer <mb at bese.it>**20050707091127] 
[Rename src/environment.lisp
Marco Baringer <mb at bese.it>**20050707091114] 
[Initial version of environment.lisp
Marco Baringer <mb at bese.it>**20050707091019] 
[Minor docstring fixup for with-call/cc
Marco Baringer <mb at bese.it>**20050707090619] 
[Big patch including all the cps interpreter stuff upto now
Marco Baringer <mb at bese.it>**20050707083739] 
[Fix bug in handling of defclass-struct's :predicate option
Marco Baringer <mb at bese.it>**20050706105324] 
[Initial Import
Marco Baringer <mb at bese.it>**20050706101657
 This import moves arnesi from arch to darcs. Nothing has actually changed since 
 bese-2004 at common-lisp.net/arnesi--dev--1.4--patch-14
] 
[added arch stuff to boring file
Marco Baringer <mb at bese.it>**20050706101630] 
[Setup boring file
Marco Baringer <mb at bese.it>**20050706100535] 
Patch bundle hash:
96f932ae261d085ebff712d535a747dac5470595

From mb at bese.it  Wed Jul 27 15:43:46 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 27 Jul 2005 17:43:46 +0200
Subject: [Bese-devel] call/cc and unwind-protect
In-Reply-To: <20050727150732.GA12047@melkor.elder-gods.org> (Larry D'Anna's
	message of "Wed, 27 Jul 2005 11:07:32 -0400")
References: <20050727150732.GA12047@melkor.elder-gods.org>
Message-ID: <m2zms845l9.fsf@soma.local>

Larry D'Anna <smoof-ra at elder-gods.org> writes:

> Is call/cc fundamentally incompatible with unwind-protect?  Is there
> any way unwind-protect can be weakened to be compatible with call/cc
> but still do The Right Thing?

only if you allow the protected form of unwind-protect to be called
multiple times or if continuations can be invalidated.

http://www.nhplace.com/kent/PFAQ/unwind-protect-vs-continuations.html

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Wed Jul 27 15:55:22 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 27 Jul 2005 17:55:22 +0200
Subject: [Bese-devel] trivial patch
In-Reply-To: <20050727153802.GA14285@melkor.elder-gods.org> (Larry D'Anna's
	message of "Wed, 27 Jul 2005 11:38:02 -0400")
References: <20050727152632.GA13509@melkor.elder-gods.org>
	<20050727153802.GA14285@melkor.elder-gods.org>
Message-ID: <m2vf2w451x.fsf@soma.local>

Larry D'Anna <smoof-ra at elder-gods.org> writes:

> * Larry D'Anna (smoof-ra at elder-gods.org) [050727 11:26]:
>> 
>> what's the preffered way of sending in patches?  send them to the list
>> or just "darcs send"?  anyway here's a trivial one.

darcs send seems to morking good so far, but i'll take whatever's
easiest for you.

> Oops that didn't quite do it.  

applied. thanks.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From marco_simon at web.de  Thu Jul 28 21:44:28 2005
From: marco_simon at web.de (Marco Simon)
Date: Thu, 28 Jul 2005 23:44:28 +0200
Subject: [Bese-devel] How to get mod_lisp talking to cmucl ? where to find
	(start-apache-listener)
Message-ID: <42E951BC.9050104@web.de>

Hi list,

first of all: I'm quite a lisp beginner !

I'm acually trying to set up an web application
based on lisp and mod_lisp. This should be the
first step on my way to an UCweb environment.

So I set up my apache 1.3 with the mod_lisp module,
configured the httpd.conf and tried to get the apache
talking with my installed cmucl-lisp on my debian-sarge
system.

I followed the steps at http://lisp.t2100cdt.kippona.net/lispy/home.

What I didn't understand so far is:
Where does the (start-apache-listener) function come from ?
Because my cmucl-lisp doesn't seem to know this function.
I think this function is the missing step for getting my lips-process
listening on port 3000, where mod_lisp sends its request to.
 
Is this function part of any library, or where can I call it from ?

Thanks for any hints in advance,

best regards
 Marco  




From pansteth at yahoo.com  Wed Jul 27 22:50:34 2005
From: pansteth at yahoo.com (Philip Ansteth)
Date: Wed, 27 Jul 2005 15:50:34 -0700 (PDT)
Subject: [Bese-devel] thanks for the movie; pronunciation of "Apache"?
In-Reply-To: <20050726162817.0CCB3880DC@common-lisp.net>
Message-ID: <20050727225034.87847.qmail@web32902.mail.mud.yahoo.com>

Man, I wish I could talk, type, and think as
fast as you do!

Thanks.  I will study it for use of Emacs as well
as ucw.

By the way, a speaker of American English, like
myself,
would almost certainly pronounce Apache like the
Indian tribe.   The web page has some explanation:
http://httpd.apache.org/docs/1.3/misc/FAQ.html#name




		
____________________________________________________
Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs 
 


From jrfitzgerald at gmail.com  Wed Jul 27 20:47:46 2005
From: jrfitzgerald at gmail.com (John Fitzgerald)
Date: Wed, 27 Jul 2005 16:47:46 -0400
Subject: [Bese-devel] UCW without threads
Message-ID: <c311134705072713471d4ffbbc@mail.gmail.com>

Hi,
Is it possible to use UCW / Slime without threads? I'm using
Freebsd/SBCL, when I try starting up ucw, I can't connect to swank
since it won't start. When I try loading it from within slime, it's
incredibly slow / appears to hang.

The examples will run after loading outside of slime, but without
swank server I'm pretty much lost.

-- 
John R. Fitzgerald


From asbjxrn at bjxrnstad.net  Thu Jul 28 00:07:46 2005
From: asbjxrn at bjxrnstad.net (=?ISO-8859-1?Q?Asbj=F8rn_Bj=F8rnstad?=)
Date: Thu, 28 Jul 2005 08:07:46 +0800
Subject: [Bese-devel] UCW without threads
In-Reply-To: <c311134705072713471d4ffbbc@mail.gmail.com>
References: <c311134705072713471d4ffbbc@mail.gmail.com>
Message-ID: <26B2FB6F-7FD4-4E86-AD39-FB8CBC587735@bjxrnstad.net>


On 28 Jul 2005, at 4:47 AM, John Fitzgerald wrote:

> Hi,
> Is it possible to use UCW / Slime without threads? I'm using
> Freebsd/SBCL, when I try starting up ucw, I can't connect to swank
> since it won't start. When I try loading it from within slime, it's
> incredibly slow / appears to hang.
>
> The examples will run after loading outside of slime, but without
> swank server I'm pretty much lost.

You can try to use the araneida backend if possible:

22:27 <      segv> asbjxrn: i believe (though i haven't checked  
recently) that the araneida backend ties into serve-event and so,  
even on a single therad lisp, you get ucw+slime+repl

-- 
   -asbjxrn



From mb at bese.it  Thu Jul 28 06:23:24 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 28 Jul 2005 08:23:24 +0200
Subject: [Bese-devel] How to get mod_lisp talking to cmucl ? where to find
	(start-apache-listener)
In-Reply-To: <42E951BC.9050104@web.de> (Marco Simon's message of "Thu,
	28 Jul 2005 23:44:28 +0200")
References: <42E951BC.9050104@web.de>
Message-ID: <m2ack71mar.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> Hi list,
>
> first of all: I'm quite a lisp beginner !
>
> I'm acually trying to set up an web application
> based on lisp and mod_lisp. This should be the
> first step on my way to an UCweb environment.
>
> So I set up my apache 1.3 with the mod_lisp module,
> configured the httpd.conf and tried to get the apache
> talking with my installed cmucl-lisp on my debian-sarge
> system.

since you mention debian sarge i'd like to remind you that debian
often has very old versions installed. i'd suggest checking and, if
needed, upgrading cmucl. this may not change much now but will save
you much grief and confusion later on, and you'll get enough grief and
confusion as it is already...

> I followed the steps at http://lisp.t2100cdt.kippona.net/lispy/home.
>
> What I didn't understand so far is:
> Where does the (start-apache-listener) function come from ?
> Because my cmucl-lisp doesn't seem to know this function.
> I think this function is the missing step for getting my lips-process
> listening on port 3000, where mod_lisp sends its request to.
>  Is this function part of any library, or where can I call it from ?

http://lisp.t2100cdt.kippona.net/lispy/thiscode

you could also look at kmr's cl-modlisp library
(http://cl-modlisp.b9.com/) if this one doesn't work out. 

good luck!

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul 28 06:15:14 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 28 Jul 2005 08:15:14 +0200
Subject: [Bese-devel] UCW without threads
In-Reply-To: <26B2FB6F-7FD4-4E86-AD39-FB8CBC587735@bjxrnstad.net>
	(=?iso-8859-1?Q?Asbj=F8rn_Bj=F8rnstad's?= message of "Thu, 28 Jul 2005
	08:07:46 +0800")
References: <c311134705072713471d4ffbbc@mail.gmail.com>
	<26B2FB6F-7FD4-4E86-AD39-FB8CBC587735@bjxrnstad.net>
Message-ID: <m2k6jb1mod.fsf@soma.local>

Asbj?rn Bj?rnstad <asbjxrn at bjxrnstad.net> writes:

> You can try to use the araneida backend if possible:
>
> 22:27 <      segv> asbjxrn: i believe (though i haven't checked
> recently) that the araneida backend ties into serve-event and so,
> even on a single therad lisp, you get ucw+slime+repl

just so it's clear, my irc handle is segv.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From mb at bese.it  Thu Jul 28 06:19:33 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 28 Jul 2005 08:19:33 +0200
Subject: [Bese-devel] thanks for the movie; pronunciation of "Apache"?
In-Reply-To: <20050727225034.87847.qmail@web32902.mail.mud.yahoo.com> (Philip
	Ansteth's message of "Wed, 27 Jul 2005 15:50:34 -0700 (PDT)")
References: <20050727225034.87847.qmail@web32902.mail.mud.yahoo.com>
Message-ID: <m2ek9j1mh6.fsf@soma.local>

Philip Ansteth <pansteth at yahoo.com> writes:

> By the way, a speaker of American English, like myself, would almost
> certainly pronounce Apache like the Indian tribe.  The web page has
> some explanation:
> http://httpd.apache.org/docs/1.3/misc/FAQ.html#name

yes, you're completly correct and i sholud have been a bit more
carefull. my problem is that 99% of the times i pronounce apache i'm
doing it in italian to an italian and with an italian accent, it's
harder to switch languages than i thought (you'll notice i make
similar mistakes with the word request and a few others).

p.s. - thanks for the link, the apache tribe is far cooler than i
thought.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From digash at gmail.com  Thu Jul 28 15:54:20 2005
From: digash at gmail.com (Dimitry Gashinsky)
Date: Thu, 28 Jul 2005 11:54:20 -0400
Subject: [Bese-devel] thanks for the movie; pronunciation of "Apache"?
In-Reply-To: <m2ek9j1mh6.fsf@soma.local>
References: <20050727225034.87847.qmail@web32902.mail.mud.yahoo.com>
	<m2ek9j1mh6.fsf@soma.local>
Message-ID: <e2c76ba5050728085456f20643@mail.gmail.com>

Ah, Apache now I got it. I could not figure out what is that "apushe" thing is.
Thank for the movie. I really enjoyed it.

--DiG

On 7/28/05, Marco Baringer <mb at bese.it> wrote:
> Philip Ansteth <pansteth at yahoo.com> writes:
> 
> > By the way, a speaker of American English, like myself, would almost
> > certainly pronounce Apache like the Indian tribe.  The web page has
> > some explanation:
> > http://httpd.apache.org/docs/1.3/misc/FAQ.html#name
> 
> yes, you're completly correct and i sholud have been a bit more
> carefull. my problem is that 99% of the times i pronounce apache i'm
> doing it in italian to an italian and with an italian accent, it's
> harder to switch languages than i thought (you'll notice i make
> similar mistakes with the word request and a few others).
> 
> p.s. - thanks for the link, the apache tribe is far cooler than i
> thought.
> 
> --
> -Marco
> Ring the bells that still can ring.
> Forget the perfect offering.
> There is a crack in everything.
> That's how the light gets in.
>         -Leonard Cohen
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel
> 


-- 
((lambda (lambda) `(,lambda ',lambda)) '(lambda (lambda) `(,lambda ',lambda)))


From zbir at urbanape.com  Fri Jul 29 17:59:15 2005
From: zbir at urbanape.com (Zachery Bir)
Date: Fri, 29 Jul 2005 13:59:15 -0400
Subject: [Bese-devel] Question about my UCW setup - getting all 404s
Message-ID: <F7A4B449-D652-4162-9AD3-307855F16A4B@urbanape.com>

I've got UCW started up with an araneida backed, and have an image  
with Marco's *hello-world* app created and registered with the  
*default-server* but anything off http://127.0.0.1:8080/ results in a  
404 including the entry point for hello-world and the /ucw/admin and / 
ucw/examples/

Mac OS X 10.4.2, OpenMCL 0.14.3, Slime 1.2.1 (from CVS), ucw 0.3.9,  
and asdf-installed other packages (yaclml, arnesi, rfc2388, &c)

slime inspecting the *default-server* shows the three applications  
registered, but I'm not sure where to further debug this. it's  
logging to my terminal with the 404s.

Thanks,

Zac


From mb at bese.it  Fri Jul 29 19:30:11 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 29 Jul 2005 21:30:11 +0200
Subject: [Bese-devel] Question about my UCW setup - getting all 404s
In-Reply-To: <F7A4B449-D652-4162-9AD3-307855F16A4B@urbanape.com> (Zachery
	Bir's message of "Fri, 29 Jul 2005 13:59:15 -0400")
References: <F7A4B449-D652-4162-9AD3-307855F16A4B@urbanape.com>
Message-ID: <m2slxxqukc.fsf@soma.local>

Zachery Bir <zbir at urbanape.com> writes:

> I've got UCW started up with an araneida backed, and have an image
> with Marco's *hello-world* app created and registered with the
> *default-server* but anything off http://127.0.0.1:8080/ results in a
> 404 including the entry point for hello-world and the /ucw/admin and /
> ucw/examples/
>
> Mac OS X 10.4.2, OpenMCL 0.14.3, Slime 1.2.1 (from CVS), ucw 0.3.9,
> and asdf-installed other packages (yaclml, arnesi, rfc2388, &c)

i know you're gonna hate this answer, but:

what hoppens if you grab ucw, arnesi and yaclml from darcs? (i _think_
the issue you're runnig into is that ucw isn't automatically
registering the entrp points with araneida, what happens when you try
to access /ucw/examples/stylesheet.css?)

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From zbir at urbanape.com  Fri Jul 29 20:15:36 2005
From: zbir at urbanape.com (Zachery Bir)
Date: Fri, 29 Jul 2005 16:15:36 -0400
Subject: [Bese-devel] Question about my UCW setup - getting all 404s
In-Reply-To: <m2slxxqukc.fsf@soma.local>
References: <F7A4B449-D652-4162-9AD3-307855F16A4B@urbanape.com>
	<m2slxxqukc.fsf@soma.local>
Message-ID: <227EEC8F-E2CB-42C3-9272-542DFFE4277A@urbanape.com>

On Jul 29, 2005, at 3:30 PM, Marco Baringer wrote:

> Zachery Bir <zbir at urbanape.com> writes:
>
>> I've got UCW started up with an araneida backed, and have an image
>> with Marco's *hello-world* app created and registered with the
>> *default-server* but anything off http://127.0.0.1:8080/ results in a
>> 404 including the entry point for hello-world and the /ucw/admin  
>> and /
>> ucw/examples/
>>
>> Mac OS X 10.4.2, OpenMCL 0.14.3, Slime 1.2.1 (from CVS), ucw 0.3.9,
>> and asdf-installed other packages (yaclml, arnesi, rfc2388, &c)
>
> i know you're gonna hate this answer, but:
>
> what hoppens if you grab ucw, arnesi and yaclml from darcs?

Heh, no worries. I'm used to needing bleeding (or at least seriously  
bruised) edge stuff.

> (i _think_
> the issue you're runnig into is that ucw isn't automatically
> registering the entrp points with araneida, what happens when you try
> to access /ucw/examples/stylesheet.css?)

Post-darcs using SBCL with the image made:

zbir at gorilla $ sbcl --core ucw.core bin/start.lisp
This is SBCL 0.9.2, an implementation of ANSI Common Lisp.
More information about SBCL is available at <http://www.sbcl.org/>.

SBCL is free software, provided as is, with absolutely no warranty.
It is mostly in the public domain; some portions are provided under
BSD-style licenses.  See the CREDITS and COPYING files in the
distribution for more information.
STYLE-WARNING: Undefined alien: "fdatasync"
STYLE-WARNING: Undefined alien: "getresuid"
STYLE-WARNING: Undefined alien: "setfsuid"
STYLE-WARNING: Undefined alien: "setresuid"
STYLE-WARNING: Undefined alien: "getresgid"
STYLE-WARNING: Undefined alien: "setfsgid"
STYLE-WARNING: Undefined alien: "setresgid"
STYLE-WARNING: Undefined alien: "linux_no_threads_p"
2005-07-29T16:12.34 IT.BESE.ARNESI:+INFO+ IT.BESE.UCW::UCW-LOGGER:  
Starting up standard server #<IT.BESE.UCW:STANDARD-SERVER  
IT.BESE.UCW:ARANEIDA-BACKEND 2 {8580221}>.
2005-07-29T16:12.34 IT.BESE.ARNESI:+INFO+ IT.BESE.UCW::UCW.BACKEND:  
Starting up ARANEIDA backend #<IT.BESE.UCW:ARANEIDA-BACKEND {864E879} 
 > on http://127.0.0.1:8080/
;; Swank started at port: 4005.
; Adding fd handler: 7

debugger invoked on a SB-INT:SIMPLE-PROGRAM-ERROR in thread 2846:
   invalid number of arguments: 0

Type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL.

(no restarts: If you didn't do this on purpose, please report it as a  
bug.)

(SB-IMPL::TOPLEVEL-REPL)
0] Logged error: 404 Not Found NIL
Logged error: 404 Not Found NIL
Logged error: 404 Not Found NIL

This is on a fresh image - so without your *hello-world* app, but I'd  
expect the two other apps to be registered. I'll try with OpenMCL  
using the new fresh-from-darcs versions of yaclml, arnesi, and ucw...

Zac



From zbir at urbanape.com  Fri Jul 29 20:24:04 2005
From: zbir at urbanape.com (Zachery Bir)
Date: Fri, 29 Jul 2005 16:24:04 -0400
Subject: [Bese-devel] Question about my UCW setup - getting all 404s
In-Reply-To: <227EEC8F-E2CB-42C3-9272-542DFFE4277A@urbanape.com>
References: <F7A4B449-D652-4162-9AD3-307855F16A4B@urbanape.com>
	<m2slxxqukc.fsf@soma.local>
	<227EEC8F-E2CB-42C3-9272-542DFFE4277A@urbanape.com>
Message-ID: <CACB12D4-3FEE-4135-99E8-95B1B0905720@urbanape.com>

On Jul 29, 2005, at 4:15 PM, Zachery Bir wrote:

> This is on a fresh image - so without your *hello-world* app, but  
> I'd expect the two other apps to be registered. I'll try with  
> OpenMCL using the new fresh-from-darcs versions of yaclml, arnesi,  
> and ucw...

FWIW, same thing with a fresh OpenMCL:

zbir at gorilla $ openmcl -I ucw.image bin/start.lisp
2005-07-29T16:19.54 IT.BESE.ARNESI:+INFO+ IT.BESE.UCW::UCW-LOGGER:  
Starting up standard server #<STANDARD-SERVER IT.BESE.UCW:ARANEIDA- 
BACKEND 2 #x6546186>.
2005-07-29T16:19.54 IT.BESE.ARNESI:+INFO+ IT.BESE.UCW::UCW.BACKEND:  
Starting up ARANEIDA backend #<IT.BESE.UCW:ARANEIDA-BACKEND  
#x65926C6> on http://127.0.0.1:8080/
;; Swank started at port: 4005.
?

Same 404s for everything (even /ucw/examples/stylesheet.css), but  
nothing gets logged to this terminal...

I'm sure it's something simple I'm missing...

Zac



From jdz at dir.lv  Sat Jul 30 10:37:33 2005
From: jdz at dir.lv (Janis Dzerins)
Date: Sat, 30 Jul 2005 13:37:33 +0300
Subject: [Bese-devel] Patch for LispWorks
Message-ID: <42EB586D.1050505@dir.lv>

Hello list,

I just made UCW work with LispWorks (4.4.5, Personal).  Had to tweak a
few things, thought.  Patches attached.

Happy hacking,
--
Janis Dzerins
-------------- next part --------------
A non-text attachment was scrubbed...
Name: arnesi.diff
Type: text/x-patch
Size: 594 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050730/d695c899/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ucw.diff
Type: text/x-patch
Size: 2081 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050730/d695c899/attachment-0001.bin>

From mb at bese.it  Sat Jul 30 11:45:45 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 30 Jul 2005 13:45:45 +0200
Subject: [Bese-devel] Patch for LispWorks
In-Reply-To: <42EB586D.1050505@dir.lv> (Janis Dzerins's message of "Sat,
	30 Jul 2005 13:37:33 +0300")
References: <42EB586D.1050505@dir.lv>
Message-ID: <m2y87oms9i.fsf@soma.local>

Janis Dzerins <jdz at dir.lv> writes:

> Hello list,
>
> I just made UCW work with LispWorks (4.4.5, Personal).  Had to tweak a
> few things, thought.  Patches attached.

applied. thanks!

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen


From der_julian at web.de  Sun Jul 31 17:19:54 2005
From: der_julian at web.de (Julian Stecklina)
Date: Sun, 31 Jul 2005 19:19:54 +0200
Subject: [Bese-devel] bash -> sh
Message-ID: <20050731191954.4c9aa25d@localhost>

Hello,

the ucwctl script does not work on FreeBSD, because it depends on bash
being part of the base system (i.e. in /bin/). As the script happens to
be perfectly fine Bourne Shell code you may consider applying the
attached patch which changes /bin/bash to /bin/sh which most people on
most platforms should have. Another approach would be #!/usr/bin/env
bash or something. But the point is: No bash is actually needed. ;)

Regards,
-- 
Julian Stecklina

LISP has survived for 21 years because it is an approximate local
optimum in the space of programming languages. - John McCarthy (1980)


-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: bash.patch
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050731/357a4f9e/attachment.ksh>

From mb at bese.it  Sun Jul 31 21:15:45 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 31 Jul 2005 23:15:45 +0200
Subject: [Bese-devel] bash -> sh
In-Reply-To: <20050731191954.4c9aa25d@localhost> (Julian Stecklina's message
	of "Sun, 31 Jul 2005 19:19:54 +0200")
References: <20050731191954.4c9aa25d@localhost>
Message-ID: <m2d5oyoewu.fsf@soma.local>

Julian Stecklina <der_julian at web.de> writes:

> Hello,
>
> the ucwctl script does not work on FreeBSD, because it depends on bash
> being part of the base system (i.e. in /bin/). As the script happens to
> be perfectly fine Bourne Shell code you may consider applying the
> attached patch which changes /bin/bash to /bin/sh which most people on
> most platforms should have. Another approach would be #!/usr/bin/env
> bash or something. But the point is: No bash is actually needed. ;)

applied. thanks!

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen