From bobstopper at bobturf.org  Mon Aug  1 05:12:29 2005
From: bobstopper at bobturf.org (Robert Marlow)
Date: Mon,  1 Aug 2005 13:12:29 +0800
Subject: [Bese-devel] darcs patch: Lisp startable UCW
Message-ID: <20050801051134.D342FCB9ED@spengler.bobturf.org>

Mon Aug  1 13:07:20 WST 2005  Robert Marlow <bobstopper at bobturf.org>
  * Lisp startable UCW
  
  This should allow UCW to be easily startable from within lisp rather than 
  from loading lisp start.lisp. The function for doing this is named
  create-server. ucwctl has been changed to make use of the new functionality
  but should work as it did before.
  
  defentry-point can now be used before its application has been attached 
  to a server - when the application is attached to a server its entry 
  points will then automatically be added. This makes it possible to load 
  up system definitions containing defentry-points without having a running 
  server beforehand. Thus lisp images can then be dumped containing full
  applications rather than just UCW.
  
  A get-backend method has been defined which creates a backend according
  to its arguments. It is also able to automatically choose a sensible
  default from currently loaded backends. It can also accept an existing,
  running server as a backend in which case it simply wraps it with a
  sensible UCW backend and returns it. This allows UCW applications to be
  attached to already running HTTP servers rather than requiring fresh
  servers be started by UCW. So far it's only been implemented for Araneida.
  
  I've only really tested the araneida and (briefly) HTTPD backends but the 
  method of starting the backends should be similar to how start.lisp worked
  so I'm hoping I haven't broken anything. 
  
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/x-darcs-patch
Size: 13306 bytes
Desc: A darcs patch for your repository!
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050801/996feebd/attachment.bin>

From mb at bese.it  Mon Aug  1 10:27:02 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 01 Aug 2005 12:27:02 +0200
Subject: [Bese-devel] darcs patch: Lisp startable UCW
In-Reply-To: <20050801051134.D342FCB9ED@spengler.bobturf.org> (Robert Marlow's
	message of "Mon, 1 Aug 2005 13:12:29 +0800")
References: <20050801051134.D342FCB9ED@spengler.bobturf.org>
Message-ID: <m2pssykl55.fsf@soma.local>

Robert Marlow <bobstopper at bobturf.org> writes:

first off: you rock!

now, the comments:

1) is there a particularl reason you moved a bunch of configuration
   stuff into ucwctl? i have a few apps which don't use ucwctl (but
   have their own 'start.lisp' scripts) so i'll need to rewrite the
   code you've added to ucwctl.

2) mind if i rename get-backend to 'make-backend'? i know this may
   seem trivial but all of our functions which create new stuff are
   called 'make-XYZ', while the get-XYZ always returns an already
   existisng object.

3) i'm still looking through it so i may come up with other questions.

> ] {
> hunk ./bin/ucwctl 12
> -VARROOT=$UCWROOT/var
> -LOGROOT=$UCWROOT/logs
> +VARROOT="$UCWROOT/var"
> +LOGROOT="$UCWROOT/logs"
> +LOGLEVEL="INFO"
> +HOST="127.0.0.1"
> +PORT="8080"
> +DEBUGGER="T"
> +INSPECTOR="NIL"
> +BACKEND="httpd"
> +#BACKEND="mod-lisp"
> +#BACKEND="aserve"
> +#BACKEND="araneida"
> +SYSTEMS="(:ucw :ucw.$BACKEND :ucw.admin :ucw.examples)"
> +APPLICATIONS="(list it.bese.ucw-user::*example-application*
> +                    ucw::*admin-application*)"
> hunk ./bin/ucwctl 40
> -                  $LISP_LOAD $UCWROOT/bin/start.lisp
> +                  $LISP_LOAD \
> +	          --eval "(defmethod print-object ((c asdf::missing-component) s)
> +                            (format s \"Unable to find the system for ~S.
> +asdf:*central-registry* is ~S.
> +Are the symlinks and asdf:*central-registry* properly setup?\" \
> +		                   (asdf::missing-requires c) \
> +		                   asdf:*central-registry*))" \
> +                  --eval "(mapc (lambda (name) (asdf:oos 'asdf:load-op name)) '$SYSTEMS)" \
> +		  --eval "(ucw:start-swank)" \
> +                  --eval "(ucw:create-server :backend :$BACKEND
> +                                             :applications $APPLICATIONS
> +                                             :debugger $DEBUGGER
> +                                             :inspector $INSPECTOR
> +                                             :log-root-directory #P\"$LOGROOT/\"
> +                                             :log-level it.bese.arnesi:+$LOGLEVEL+
> +                                             :host \"$HOST\"
> +                                             :port $PORT)"
> hunk ./bin/ucwctl 60
> -                  --eval "(load \"$UCWROOT/bin/stop.lisp\")"
> +                  --eval "(ucw:shutdown-server ucw:*default-server*)" \
> +	          --eval "(quit)"
> hunk ./src/backend/araneida.lisp 12
> -  ((listener :accessor listener :initarg :listener)
> +  ((listener :accessor listener :initarg :listener :initform nil)
> hunk ./src/backend/araneida.lisp 38
> -				       &key (port 8080) (host "127.0.0.1"))  
> -  (setf (listener backend) (make-instance (listener-class backend) :port port)
> -	(default-url backend) (araneida:make-url :scheme "http" :host host :port port)))
> +				       &key (port 8080) (host "127.0.0.1"))
> +  (unless (listener backend)
> +    (setf (listener backend) (make-instance (listener-class backend) :port port)
> +	  (default-url backend) (araneida:make-url :scheme "http" :host host :port port)
> +	  (araneida::http-listener-default-hostname (listener backend)) host
> +	  araneida::*default-url-defaults* (default-url backend))))
> hunk ./src/backend/araneida.lisp 56
> -  #+clisp (araneida:host-serve-events)) 
> +  #+clisp (araneida:host-serve-events))
> hunk ./src/backend/araneida.lisp 195
> +
> +
> +(defmethod get-backend ((backend araneida:http-listener) &optional host port)
> +  (declare (ignore host port))
> +  (make-instance 'ucw:araneida-backend
> +		 :listener backend
> +		 :default-url (araneida:make-url :scheme "http" :host host :port port)))

this is a particularly good idea :)

> addfile ./src/control.lisp
> hunk ./src/control.lisp 1
> +(in-package :ucw)
> +
> +(defun threaded-lisp-p ()
> +  #+(and sbcl sb-thread) t
> +  #+(and cmu mp) t
> +  #+openmcl t)
> +
> +
> +(defun start-swank ()
> +  (if (threaded-lisp-p)
> +      (swank:create-server :dont-close t)
> +      (warn "Can't start slime server due to lack of threads.")))
> +
> +
> +(defun create-server (&key (backend :default) applications debugger
> +		      inspector host port (start-p t)
> +		      (server-class 'standard-server) log-root-directory
> +		      (log-level +info+))
> +  "Creates and returns a UCW server according to SERVER-CLASS, HOST and
> +PORT. Affects *DEFAULT-SERVER*.
> +
> +BACKEND may be :HTTPD, :MOD-LISP, :ASERVE, :ARANEIDA, an existing backend,
> +an existing UCW server backend or :DEFAULT in which case it attempts to
> +return a sane default from the UCW backends loaded and available.
> +
> +APPLICATIONS is a list of defined applications to be loaded into the
> +server.
> +
> +DEBUGGER and INSPECTOR are booleans which affect *DEBUG-ON-ERROR* and
> +*INSPECT-COMPONENTS* respectively.
> +
> +Logs are generated in verbosity defined by LOG-LEVEL and directed to
> +LOG-ROOT-DIRECTORY if defined."
> +  (setf *debug-on-error* (or *debug-on-error* debugger)
> +	*inspect-components* (or *inspect-components* inspector))
> +
> +  (when log-root-directory
> +    (let ((ucw-logger (get-logger 'ucw::ucw-logger))
> +	  (ucw.backend (get-logger 'ucw::ucw.backend))
> +	  (ucw.log (merge-pathnames "ucw.log" log-root-directory))
> +	  (ucw.backend.log (merge-pathnames "ucw-backend.log" log-root-directory))
> +	  (console-appender (make-stream-log-appender)))
> +      
> +      (setf (appenders ucw-logger) (list console-appender
> +					 (make-file-log-appender ucw.log))
> +	    (log.level ucw-logger) log-level
> +	    (appenders ucw.backend) (list console-appender
> +					  (make-file-log-appender ucw.backend.log)))))
> +
> +  (let ((server nil))
> +    
> +    (restart-case
> +	(when *default-server*
> +	  (error "*DEFAULT-SERVER* already defined as ~A.
> +Create another server anyway?" *default-server*))
> +      (replace ()
> +	:report "Replace *DEFAULT-SERVER* with a new server instance"
> +	(shutdown-server *default-server*)
> +	(setf *default-server* nil))
> +      (continue ()
> +	:report "Create an additional server"))
> +    (setf server (make-instance server-class))
> +    (unless *default-server*
> +      (setf *default-server* server))
> +    (setf (server.backend server)
> +	  (get-backend backend host port))
> +    (mapc (lambda (app)
> +	    (register-application server app))
> +	  applications)
> +    (when start-p
> +      (startup-server server))
> +    server))
> +  
> +
> +(defgeneric get-backend (backend &optional host port)
> +  (:documentation "Returns a UCW server backend as requested by the
> +functional arguments. BACKEND may be :HTTPD, :MOD-LISP, :ASERVE,
> +:ARANEIDA, an existing backend, an existing UCW server backend or
> +:DEFAULT in which case it attempts to return a sane default from
> +the UCW backends loaded and available."))
> +
> +
> +(defmethod get-backend ((backend (eql :httpd)) &optional host port)
> +  (make-instance (if (threaded-lisp-p)
> +		     'ucw:multithread-httpd-backend
> +		     'ucw:httpd-backend)
> +		 :host host
> +		 :port port))
> +
> +
> +(defmethod get-backend ((backend (eql :mod-lisp)) &optional host port)
> +  (make-instance (if (threaded-lisp-p)
> +		     'ucw:multithread-mod-lisp-backend
> +		     'ucw:mod-lisp-backend)
> +		 :host host
> +		 :port 8080))
> +
> +
> +(defmethod get-backend ((backend (eql :aserve)) &optional host port)
> +  (declare (ignore host))
> +  (make-instance 'ucw:aserve-backend :port 8080))
> +
> +
> +(defmethod get-backend ((backend (eql :araneida)) &optional host port)
> +  (make-instance 'ucw:araneida-backend
> +		 :host host
> +		 :port port))
> +
> +
> +(defmethod get-backend ((backend (eql :default)) &optional host port)
> +  (cond
> +    ((find-class 'mod-lisp-backend nil)
> +     (get-backend :mod-lisp host port))
> +    ((find-class 'httpd-backend nil)
> +     (get-backend :httpd host port))
> +    ((find-class 'aserve-backend nil)
> +     (get-backend :aserve host port))
> +    ((find-class 'araneida-backend nil)
> +     (get-backend :araneida host port))
> +    (t (error "No backends loaded and ready for use"))))
> +
> +
> +(defmethod get-backend ((backend backend) &optional host port)
> +  (declare (ignore host port))
> +  backend)
> +
> +
> +(defmethod get-backend (backend &optional host port)
> +  (error "Unacceptable UCW backend ~A" backend))
> +
> hunk ./src/packages.lisp 231
> -   #:time-difference))
> +   #:time-difference
> +
> +   ;; Control utilities
> +   #:start-swank
> +   #:create-server))
> hunk ./src/rerl/standard-application.lisp 41
> -  (register-entry-point (application.server app)
> -			(strcat (application.url-prefix app) url)
> -			entry-point))
> +  (when (application.server app)
> +    (register-entry-point (application.server app)
> +			  (strcat (application.url-prefix app) url)
> +			  entry-point)))
> hunk ./src/rerl/standard-classes.lisp 37
> +   (started :accessor server.started :initform nil :initarg :started)
> hunk ./src/rerl/standard-classes.lisp 80
> +
> hunk ./src/rerl/standard-server.lisp 16
> +  (setf (server.started server) t)
> hunk ./src/rerl/standard-server.lisp 40
> -			 (application.url-prefix app)))))
> +			 (application.url-prefix app)))
> +    (maphash (lambda (url entry-point)
> +	       (register-entry-point server
> +				     (strcat (application.url-prefix app) url)
> +				     entry-point))
> +	     (application.entry-points app))))
> hunk ./src/vars.lisp 7
> -(defvar *debug-on-error*)
> +(defvar *debug-on-error* nil)
> hunk ./src/vars.lisp 9
> -(defvar *ucw-tal-root*)
> +(defvar *ucw-tal-root*
> +  (merge-pathnames (make-pathname :directory '(:relative "wwwroot"))
> +		   (asdf:component-pathname (asdf:find-system :ucw))))
> hunk ./src/vars.lisp 13
> -(defvar *inspect-components*)
> +(defvar *inspect-components* nil)
> hunk ./ucw.asd 92
> -                  :depends-on ("packages" :rerl)))))
> +                  :depends-on ("packages" :rerl))
> +		 (:file "control"
> +		  :depends-on ("packages" :backend :rerl)))))
> }
>
> Context:
>
> [bash -> sh
> Julian Stecklina <der_julian at web.de>**20050731145119] 
> [When creating the string for request paramteers use the same element-type as the raw-uri array
> Marco Baringer <mb at bese.it>**20050730114429] 
> [Use defcomponent instead of defclass for simple-template-component
> Marco Baringer <mb at bese.it>**20050726095406] 
> [Instead of looking for a property named VERSIONS we now call it :FEATURES. Updated the error message accordingly
> Marco Baringer <mb at bese.it>**20050729103158] 
> [Renamed 'version' to 'feature' in asdf property check. added check for cps-interpreter
> Marco Baringer <mb at bese.it>**20050728120322] 
> [Changes to the stylesheet
> Marco Baringer <mb at bese.it>**20050728120129] 
> [template-component-environment methods must always return a fresh list
> Marco Baringer <mb at bese.it>**20050725121551] 
> [Make the wiki example use the show-window macro
> Marco Baringer <mb at bese.it>**20050725121447] 
> [Added SHOW and SHOW-WINDOW macros
> Marco Baringer <mb at bese.it>**20050725121421] 
> [Post method for login component
> matley at innerloop.it**20050724142609] 
> [Added initarg to client-value (form-component)
> matley at innerloop.it**20050723142419] 
> [Little typo bugfix
> matley at innerloop.it**20050722200835] 
> [Only use copy-seq on the request parameters if thay are strings (as opposed to mime-part objects)
> Marco Baringer <mb at bese.it>**20050721154259] 
> [Export the mime-part accessors
> Marco Baringer <mb at bese.it>**20050721154124] 
> [Update component slots and backtrack slots when reinitializing component classes
> Marco Baringer <mb at bese.it>**20050718172834] 
> [Allow defcomponent slot forms to be symbols (some effect as with defclass)
> Marco Baringer <mb at bese.it>**20050718171831] 
> [More accurate information regarding versions and how to get the software.
> Jos?? Pablo Ezequiel Fern??ndez <pupeno at pupeno.com>**20050718033950] 
> [More changes regarding dependencies.
> Jos?? Pablo Ezequiel Fern??ndez <pupeno at pupeno.com>**20050717214720] 
> [Some clear statements about the dependencies (more is still neede).
> Jos?? Pablo Ezequiel Fern??ndez <pupeno at pupeno.com>**20050717213407] 
> [Make ucw's installation directory an explicit configuration variable in make-image.lisp
> Marco Baringer <mb at bese.it>**20050717191936] 
> [Remove useless debugging code from ucw-tags.lisp
> Marco Baringer <mb at bese.it>**20050717191301] 
> [Fix handling of HTTR response status codes in aserve backend (Patch by: Antonio Menezes Leitao <aml at gia.ist.utl.pt>)
> Marco Baringer <mb at bese.it>**20050717190254] 
> [Fix handling of HTTP request/respons headers in aserve backend. (Patch by: Antonio Menezes Leitao <aml at gia.ist.utl.pt>)
> Marco Baringer <mb at bese.it>**20050717185826] 
> [Initial Import from arch
> Marco Baringer <mb at bese.it>**20050706133829
>  This patch is exactly equal to ucw-2004 at common-lisp.net/ucw--dev--0.3--patch-426 it simply
>  represents the move over to darcs.
> ] 
> [Added boring file
> Marco Baringer <mb at bese.it>**20050706132641] 
> Patch bundle hash:
> 544613ec9e5780e48c348da7472bba4343554ed4
>
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel


From marco_simon at web.de  Mon Aug  1 10:33:17 2005
From: marco_simon at web.de (Marco Simon)
Date: Mon, 01 Aug 2005 12:33:17 +0200
Subject: [Bese-devel] how to set ucw-documentroot ?
Message-ID: <42EDFA6D.207@web.de>

Hi list,

I've just set up my ucw as follows:

ucw's bin's are in /home/ucw/ucw/bin
all libs for ucw are in /home/ucw/*
the ucw-package ist in /hme/ucw/ucw/*

I've set up apache 1.33 with mod_lisp and the following
httpd.conf - modifications:

LoadModule lisp_module /usr/lib/apache/1.3/mod_lisp.so
AddModule mod_lisp.c

LispServer 127.0.0.1 3000 "mylisp"
<Location /asp>
 SetHandler lisp-handler
</Location>

Beside that I changed /home/ucw/ucw/bin/utils.lisp to use
the port 3000 (instead of 3001) for the mod-lisp-backend
and activated the mod-lisp-backend in /home/ucw/ucw/bin/start.lisp
by commenting-out the following lines:
;(use-httpd-backend)
(use-mod-lisp-backend)
;(use-aserve-backend)
;(use-araneida-backend)

After this modifications the ucw seems to run and listen on port 3000.
Whatever, if I try to connect to myWebServerHost:80/asp/ I get an normal
autoindex returned from my webserver. That's ok so far.
But as soon as I try to connect to myWebServerHost:80/asp/* (e.g. 
/asp/myfile.lips)
I always get an ucw-created error retuned: /asp/something_existing.lisp 
not found.


I guess I've got some failure in my configuration settings, but I 
haven'nt any idea
how to fix it. Could please give me some hints how to set the path to my 
lisp wwwroot
correctly ?

Thanks a lot in advance,
 best regards
 Marco






From bobstopper at bobturf.org  Mon Aug  1 11:08:23 2005
From: bobstopper at bobturf.org (Robert Marlow)
Date: Mon, 01 Aug 2005 19:08:23 +0800
Subject: [Bese-devel] darcs patch: Lisp startable UCW
In-Reply-To: <m2pssykl55.fsf@soma.local>
References: <20050801051134.D342FCB9ED@spengler.bobturf.org>
	<m2pssykl55.fsf@soma.local>
Message-ID: <87oe8hncd4.wl%bobstopper@bobturf.org>

Aw, it was nuthin :)

1. The intention wasn't so much to move configuration to ucwctl... it
   was more to make use of create-server which accepts the
   configuration stuff as its arguments. I don't think my changes
   should break the existing start.lisp... I haven't checked
   though. If you don't use ucwctl perhaps your start.lisp files will
   continue to work? Lemme know what it breaks and I'll see if I can
   fix what I've broken. 

   On a related note, would it be more useful to have the config stuff
   put into a separate ucw.conf file? I've written config-file asdf
   loaders before to handle that kinda stuff and it shouldn't be too
   difficult to do the same for UCW though I'd imagine such things
   possibly belong in final applications rather than UCW itself.

2. Go right ahead. Someone else has pointed out get-backend was a crap
   name. I wasn't really thinking too hard about the name when I wrote
   it.


At Mon, 01 Aug 2005 12:27:02 +0200,
Marco Baringer wrote:
> 
> Robert Marlow <bobstopper at bobturf.org> writes:
> 
> first off: you rock!
> 
> now, the comments:
> 
> 1) is there a particularl reason you moved a bunch of configuration
>    stuff into ucwctl? i have a few apps which don't use ucwctl (but
>    have their own 'start.lisp' scripts) so i'll need to rewrite the
>    code you've added to ucwctl.
> 
> 2) mind if i rename get-backend to 'make-backend'? i know this may
>    seem trivial but all of our functions which create new stuff are
>    called 'make-XYZ', while the get-XYZ always returns an already
>    existisng object.
> 
> 3) i'm still looking through it so i may come up with other questions.


From mb at bese.it  Mon Aug  1 12:45:46 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 01 Aug 2005 14:45:46 +0200
Subject: [Bese-devel] how to set ucw-documentroot ?
In-Reply-To: <42EDFA6D.207@web.de> (Marco Simon's message of "Mon,
	01 Aug 2005 12:33:17 +0200")
References: <42EDFA6D.207@web.de>
Message-ID: <m2u0i9yged.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> After this modifications the ucw seems to run and listen on port 3000.
> Whatever, if I try to connect to myWebServerHost:80/asp/ I get an normal
> autoindex returned from my webserver. That's ok so far.
> But as soon as I try to connect to myWebServerHost:80/asp/*
> (e.g. /asp/myfile.lips)
> I always get an ucw-created error retuned:
> /asp/something_existing.lisp not found.

how are you setting up the somethi_existing.lisp entry point? 

does /asp/ucw/examples/index.ucw work? (NB: for this to work you will
have to change tthe url-prefix from "/ucw/examples/" to
"/asp/ucw/examples/" (look in ucw/examples/examples.lisp:9))

-- 
-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 smoof-ra at elder-gods.org  Mon Aug  1 18:50:33 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Mon, 1 Aug 2005 14:50:33 -0400
Subject: [Bese-devel] declarations
In-Reply-To: <20050727151240.GA7447@melkor.elder-gods.org>
References: <20050726170217.GA17394@melkor.elder-gods.org>
	<m2ll3syg30.fsf@soma.local>
	<20050727151240.GA7447@melkor.elder-gods.org>
Message-ID: <20050801185033.GA26811@melkor.elder-gods.org>

* Larry D'Anna (smoof-ra at elder-gods.org) [050727 11:13]:
> (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!

Here's a patch.

       --larry 

-------------- next part --------------

New patches:

[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] {
hunk ./src/walk.lisp 117
+(defclass implicit-progn-with-declare-mixin (implicit-progn-mixin)
+  ((declares :accessor declares :initarg :declares)))
+
hunk ./src/walk.lisp 123
+(defmacro multiple-value-setf (places form)
+  (let ((names (loop for place in places collect (gensym))))
+    `(multiple-value-bind ,names ,form
+       ,@(loop for name in names for place in places if (null place) collect `(declare (ignore ,name)))
+	 (setf ,@(loop
+		    for name in names
+		    for place in places
+		    if place collect place
+		    if place collect name)))))
+
hunk ./src/walk.lisp 134
-  (let ((documentation nil))
+  (let ((documentation nil) 
+	(decl nil)
+	(decls nil))
hunk ./src/walk.lisp 138
-             (return-from split-body (values body env documentation))))
+             (return-from split-body (values body env documentation decls))))
hunk ./src/walk.lisp 147
-                            (setf env (parse-declaration dec env))))
+                            (multiple-value-setf (env decl) (parse-declaration dec env))
+			    (when decl
+			      (push decl decls))))
hunk ./src/walk.lisp 170
+(defclass declaration-form (form)
+  ())
+
hunk ./src/walk.lisp 204
-  environment)
+  (values environment declaration))
hunk ./src/walk.lisp 207
-  (multiple-value-bind (body env docstring)
+  (multiple-value-bind (body env docstring declarations)
hunk ./src/walk.lisp 212
-            docstring)))
+            docstring
+	    declarations)))
hunk ./src/walk.lisp 306
-(defclass lambda-function-form (function-form implicit-progn-mixin)
+(defclass lambda-function-form (function-form implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 363
-      (setf (body func) (walk-implict-progn func (cddr form) env :declare t))
+      (multiple-value-setf ((body func) nil (declares func)) (walk-implict-progn func (cddr form) env :declare t))
hunk ./src/walk.lisp 503
-(defclass function-binding-form (form binding-form-mixin implicit-progn-mixin)
+(defclass function-binding-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 522
-      (setf (body flet) (walk-implict-progn flet
-                                            body
-                                            (loop
-                                               with env = env
-                                               for (name . lambda) in (binds flet)
-                                               do (setf env (register env :flet name lambda))
-                                               finally (return env))
-                                            :declare t)))))
+      (multiple-value-setf ((body flet) nil (declares flet))
+			   (walk-implict-progn flet
+					       body
+					       (loop
+						  with env = env
+						  for (name . lambda) in (binds flet)
+						  do (setf env (register env :flet name lambda))
+						  finally (return env))
+					       :declare t)))))
hunk ./src/walk.lisp 562
-(defclass variable-binding-form (form binding-form-mixin implicit-progn-mixin)
+(defclass variable-binding-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 578
-    (setf (body let) (walk-implict-progn let (cddr form) env :declare t))))
+    (multiple-value-setf ((body let) nil (declares let)) (walk-implict-progn let (cddr form) env :declare t))))
hunk ./src/walk.lisp 588
-    (setf (binds let*) (nreverse (binds let*))
-          (body let*) (walk-implict-progn let* (cddr form) env :declare t))))
+    (setf (binds let*) (nreverse (binds let*)))
+    (multiple-value-setf ((body let*) nil (declares let*)) (walk-implict-progn let* (cddr form) env :declare t))))
hunk ./src/walk.lisp 593
-(defclass locally-form (form implicit-progn-mixin)
+(defclass locally-form (form implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 598
-    (setf (body locally) (walk-implict-progn locally (cdr form) env :declare t))))
+    (multiple-value-setf ((body locally) nil (declares locally)) (walk-implict-progn locally (cdr form) env :declare t))))
hunk ./src/walk.lisp 602
-(defclass macrolet-form (form binding-form-mixin implicit-progn-mixin)
+(defclass macrolet-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 612
-    (setf (binds macrolet) (nreverse (binds macrolet))
-          (body macrolet) (walk-implict-progn macrolet (cddr form) env :declare t))))
+    (setf (binds macrolet) (nreverse (binds macrolet)))
+    (multiple-value-setf ((body macrolet) nil (declares macrolet)) (walk-implict-progn macrolet (cddr form) env :declare t))))
hunk ./src/walk.lisp 685
-(defclass symbol-macrolet-form (form binding-form-mixin implicit-progn-mixin)
+(defclass symbol-macrolet-form (form binding-form-mixin implicit-progn-with-declare-mixin)
}

Context:

[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
27538f2839b9eae10fc6c01ed7e15e455a4ffdfe

From smoof-ra at elder-gods.org  Mon Aug  1 18:58:10 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Mon, 1 Aug 2005 14:58:10 -0400
Subject: [Bese-devel] declarations
In-Reply-To: <20050801185033.GA26811@melkor.elder-gods.org>
References: <20050726170217.GA17394@melkor.elder-gods.org>
	<m2ll3syg30.fsf@soma.local>
	<20050727151240.GA7447@melkor.elder-gods.org>
	<20050801185033.GA26811@melkor.elder-gods.org>
Message-ID: <20050801185810.GA27407@melkor.elder-gods.org>

* Larry D'Anna (smoof-ra at elder-gods.org) [050801 14:50]:
> * Larry D'Anna (smoof-ra at elder-gods.org) [050727 11:13]:
> > (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!
> 
> Here's a patch.

I forgot a line on that patch, here's the right one.

  --larry 
-------------- next part --------------

New patches:

[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] {
hunk ./src/walk.lisp 117
+(defclass implicit-progn-with-declare-mixin (implicit-progn-mixin)
+  ((declares :accessor declares :initarg :declares)))
+
hunk ./src/walk.lisp 123
+(defmacro multiple-value-setf (places form)
+  (let ((names (loop for place in places collect (gensym))))
+    `(multiple-value-bind ,names ,form
+       ,@(loop for name in names for place in places if (null place) collect `(declare (ignore ,name)))
+	 (setf ,@(loop
+		    for name in names
+		    for place in places
+		    if place collect place
+		    if place collect name)))))
+
hunk ./src/walk.lisp 134
-  (let ((documentation nil))
+  (let ((documentation nil) 
+	(decl nil)
+	(decls nil))
hunk ./src/walk.lisp 138
-             (return-from split-body (values body env documentation))))
+             (return-from split-body (values body env documentation decls))))
hunk ./src/walk.lisp 147
-                            (setf env (parse-declaration dec env))))
+                            (multiple-value-setf (env decl) (parse-declaration dec env))
+			    (when decl
+			      (push decl decls))))
hunk ./src/walk.lisp 170
+(defclass declaration-form (form)
+  ())
+
hunk ./src/walk.lisp 204
-  environment)
+  (values environment declaration))
hunk ./src/walk.lisp 207
-  (multiple-value-bind (body env docstring)
+  (multiple-value-bind (body env docstring declarations)
hunk ./src/walk.lisp 212
-            docstring)))
+            docstring
+	    declarations)))
hunk ./src/walk.lisp 306
-(defclass lambda-function-form (function-form implicit-progn-mixin)
+(defclass lambda-function-form (function-form implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 363
-      (setf (body func) (walk-implict-progn func (cddr form) env :declare t))
+      (multiple-value-setf ((body func) nil (declares func)) (walk-implict-progn func (cddr form) env :declare t))
hunk ./src/walk.lisp 503
-(defclass function-binding-form (form binding-form-mixin implicit-progn-mixin)
+(defclass function-binding-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 522
-      (setf (body flet) (walk-implict-progn flet
-                                            body
-                                            (loop
-                                               with env = env
-                                               for (name . lambda) in (binds flet)
-                                               do (setf env (register env :flet name lambda))
-                                               finally (return env))
-                                            :declare t)))))
+      (multiple-value-setf ((body flet) nil (declares flet))
+			   (walk-implict-progn flet
+					       body
+					       (loop
+						  with env = env
+						  for (name . lambda) in (binds flet)
+						  do (setf env (register env :flet name lambda))
+						  finally (return env))
+					       :declare t)))))
hunk ./src/walk.lisp 562
-(defclass variable-binding-form (form binding-form-mixin implicit-progn-mixin)
+(defclass variable-binding-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 578
-    (setf (body let) (walk-implict-progn let (cddr form) env :declare t))))
+    (multiple-value-setf ((body let) nil (declares let)) (walk-implict-progn let (cddr form) env :declare t))))
hunk ./src/walk.lisp 588
-    (setf (binds let*) (nreverse (binds let*))
-          (body let*) (walk-implict-progn let* (cddr form) env :declare t))))
+    (setf (binds let*) (nreverse (binds let*)))
+    (multiple-value-setf ((body let*) nil (declares let*)) (walk-implict-progn let* (cddr form) env :declare t))))
hunk ./src/walk.lisp 593
-(defclass locally-form (form implicit-progn-mixin)
+(defclass locally-form (form implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 598
-    (setf (body locally) (walk-implict-progn locally (cdr form) env :declare t))))
+    (multiple-value-setf ((body locally) nil (declares locally)) (walk-implict-progn locally (cdr form) env :declare t))))
hunk ./src/walk.lisp 602
-(defclass macrolet-form (form binding-form-mixin implicit-progn-mixin)
+(defclass macrolet-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 612
-    (setf (binds macrolet) (nreverse (binds macrolet))
-          (body macrolet) (walk-implict-progn macrolet (cddr form) env :declare t))))
+    (setf (binds macrolet) (nreverse (binds macrolet)))
+    (multiple-value-setf ((body macrolet) nil (declares macrolet)) (walk-implict-progn macrolet (cddr form) env :declare t))))
hunk ./src/walk.lisp 685
-(defclass symbol-macrolet-form (form binding-form-mixin implicit-progn-mixin)
+(defclass symbol-macrolet-form (form binding-form-mixin implicit-progn-with-declare-mixin)
}

[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] {
hunk ./src/walk.lisp 204
-  (values environment declaration))
+  (values environment (make-instance 'declaration-form :source declaration)))
}

Context:

[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
79722b16767d5673b199a28db95457ea74d0a7bc

From smoof-ra at elder-gods.org  Mon Aug  1 19:32:59 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Mon, 1 Aug 2005 15:32:59 -0400
Subject: [Bese-devel] declarations
In-Reply-To: <20050801185810.GA27407@melkor.elder-gods.org>
References: <20050726170217.GA17394@melkor.elder-gods.org>
	<m2ll3syg30.fsf@soma.local>
	<20050727151240.GA7447@melkor.elder-gods.org>
	<20050801185033.GA26811@melkor.elder-gods.org>
	<20050801185810.GA27407@melkor.elder-gods.org>
Message-ID: <20050801193259.GA29802@melkor.elder-gods.org>

* Larry D'Anna (smoof-ra at elder-gods.org) [050801 14:58]:
> * Larry D'Anna (smoof-ra at elder-gods.org) [050801 14:50]:
> > * Larry D'Anna (smoof-ra at elder-gods.org) [050727 11:13]:
> > > (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!
> > 
> > Here's a patch.
> 
> I forgot a line on that patch, here's the right one.

Sorry for the three emails.  I have one more line for you :-)

      --larry 
-------------- next part --------------

New patches:

[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] {
hunk ./src/walk.lisp 117
+(defclass implicit-progn-with-declare-mixin (implicit-progn-mixin)
+  ((declares :accessor declares :initarg :declares)))
+
hunk ./src/walk.lisp 123
+(defmacro multiple-value-setf (places form)
+  (let ((names (loop for place in places collect (gensym))))
+    `(multiple-value-bind ,names ,form
+       ,@(loop for name in names for place in places if (null place) collect `(declare (ignore ,name)))
+	 (setf ,@(loop
+		    for name in names
+		    for place in places
+		    if place collect place
+		    if place collect name)))))
+
hunk ./src/walk.lisp 134
-  (let ((documentation nil))
+  (let ((documentation nil) 
+	(decl nil)
+	(decls nil))
hunk ./src/walk.lisp 138
-             (return-from split-body (values body env documentation))))
+             (return-from split-body (values body env documentation decls))))
hunk ./src/walk.lisp 147
-                            (setf env (parse-declaration dec env))))
+                            (multiple-value-setf (env decl) (parse-declaration dec env))
+			    (when decl
+			      (push decl decls))))
hunk ./src/walk.lisp 170
+(defclass declaration-form (form)
+  ())
+
hunk ./src/walk.lisp 204
-  environment)
+  (values environment declaration))
hunk ./src/walk.lisp 207
-  (multiple-value-bind (body env docstring)
+  (multiple-value-bind (body env docstring declarations)
hunk ./src/walk.lisp 212
-            docstring)))
+            docstring
+	    declarations)))
hunk ./src/walk.lisp 306
-(defclass lambda-function-form (function-form implicit-progn-mixin)
+(defclass lambda-function-form (function-form implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 363
-      (setf (body func) (walk-implict-progn func (cddr form) env :declare t))
+      (multiple-value-setf ((body func) nil (declares func)) (walk-implict-progn func (cddr form) env :declare t))
hunk ./src/walk.lisp 503
-(defclass function-binding-form (form binding-form-mixin implicit-progn-mixin)
+(defclass function-binding-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 522
-      (setf (body flet) (walk-implict-progn flet
-                                            body
-                                            (loop
-                                               with env = env
-                                               for (name . lambda) in (binds flet)
-                                               do (setf env (register env :flet name lambda))
-                                               finally (return env))
-                                            :declare t)))))
+      (multiple-value-setf ((body flet) nil (declares flet))
+			   (walk-implict-progn flet
+					       body
+					       (loop
+						  with env = env
+						  for (name . lambda) in (binds flet)
+						  do (setf env (register env :flet name lambda))
+						  finally (return env))
+					       :declare t)))))
hunk ./src/walk.lisp 562
-(defclass variable-binding-form (form binding-form-mixin implicit-progn-mixin)
+(defclass variable-binding-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 578
-    (setf (body let) (walk-implict-progn let (cddr form) env :declare t))))
+    (multiple-value-setf ((body let) nil (declares let)) (walk-implict-progn let (cddr form) env :declare t))))
hunk ./src/walk.lisp 588
-    (setf (binds let*) (nreverse (binds let*))
-          (body let*) (walk-implict-progn let* (cddr form) env :declare t))))
+    (setf (binds let*) (nreverse (binds let*)))
+    (multiple-value-setf ((body let*) nil (declares let*)) (walk-implict-progn let* (cddr form) env :declare t))))
hunk ./src/walk.lisp 593
-(defclass locally-form (form implicit-progn-mixin)
+(defclass locally-form (form implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 598
-    (setf (body locally) (walk-implict-progn locally (cdr form) env :declare t))))
+    (multiple-value-setf ((body locally) nil (declares locally)) (walk-implict-progn locally (cdr form) env :declare t))))
hunk ./src/walk.lisp 602
-(defclass macrolet-form (form binding-form-mixin implicit-progn-mixin)
+(defclass macrolet-form (form binding-form-mixin implicit-progn-with-declare-mixin)
hunk ./src/walk.lisp 612
-    (setf (binds macrolet) (nreverse (binds macrolet))
-          (body macrolet) (walk-implict-progn macrolet (cddr form) env :declare t))))
+    (setf (binds macrolet) (nreverse (binds macrolet)))
+    (multiple-value-setf ((body macrolet) nil (declares macrolet)) (walk-implict-progn macrolet (cddr form) env :declare t))))
hunk ./src/walk.lisp 685
-(defclass symbol-macrolet-form (form binding-form-mixin implicit-progn-mixin)
+(defclass symbol-macrolet-form (form binding-form-mixin implicit-progn-with-declare-mixin)
}

[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] {
hunk ./src/walk.lisp 204
-  (values environment declaration))
+  (values environment (make-instance 'declaration-form :source declaration)))
}

[declares needs to be copied in the labels handler just like the other lambda-function-form slots
smoof-ra at elder-gods.org**20050801193107] {
hunk ./src/walk.lisp 557
-                  (arguments lambda) (arguments tmp-lambda)))
+                  (arguments lambda) (arguments tmp-lambda)
+		  (declares lambda) (declares tmp-lambda)))
}

Context:

[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
ba6c6ab51a00ba2bfaa4e4c352b6a3586596af42

From mb at bese.it  Mon Aug  1 21:37:38 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 01 Aug 2005 23:37:38 +0200
Subject: [Bese-devel] declarations
In-Reply-To: <20050801193259.GA29802@melkor.elder-gods.org> (Larry D'Anna's
	message of "Mon, 1 Aug 2005 15:32:59 -0400")
References: <20050726170217.GA17394@melkor.elder-gods.org>
	<m2ll3syg30.fsf@soma.local>
	<20050727151240.GA7447@melkor.elder-gods.org>
	<20050801185033.GA26811@melkor.elder-gods.org>
	<20050801185810.GA27407@melkor.elder-gods.org>
	<20050801193259.GA29802@melkor.elder-gods.org>
Message-ID: <m2ll3lqqxp.fsf@soma.local>

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

> Sorry for the three emails.  I have one more line for you :-)

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 carlos.ungil at bluewin.ch  Mon Aug  1 22:48:32 2005
From: carlos.ungil at bluewin.ch (Carlos Ungil)
Date: Tue, 2 Aug 2005 00:48:32 +0200
Subject: [Bese-devel] problem with actions
Message-ID: <f3cb1bc79ec99bff83efb831118949cf@bluewin.ch>

Hello,

I don't understand why the code below is not working. Well, I bet the 
code is fine and the problem is only in my expectations. I hope this 
will be obvious to someone who can give me a hint.

The test-list component (/test/index.ucw) shows a list of ten items 
0..9, and clicking on one of them won't show that number, as I would 
like, but all of them display the number 10. I think this is because 10 
is the value the of i variable after the loop, and somehow this is the 
only value used by the system.

How can I make the display-number actions see the different values of 
the variable, and not just the final one? Maybe my approach is 
completely wrong. In that case, what  is the right construct to get 
something like this? What I really want to do is to show a list of 
objects, displaying the object when clicking on link.

Thanks [for any help on this, and for making the ucw code available in 
the first place!],

Carlos

-=-=-=-=-=-=-=-=-=-

(in-package :it.bese.ucw-user)

(defvar *test-application*
   (make-instance 'cookie-session-application
                  :url-prefix "/test/"))

(register-application *default-server* *test-application*)

(defentry-point "index.ucw" (:application *test-application*) ()
   (call 'test-list))

(defcomponent test-list (simple-window-component) ())

(defcomponent test-display (simple-window-component)
   ((number :accessor number :initarg :number)))

(defaction display-number ((c test-list) number)
   (call 'test-display :number number))

(defmethod render-on ((res response) (app test-list))
   (<:ul (loop for i from 0 below 10
	      do (<:li (<ucw:a :action (display-number app i)
			       (<:as-html i))))))

(defmethod render-on ((res response) (app test-display))
   (<:as-html (number app)))



From carlos.ungil at bluewin.ch  Tue Aug  2 06:06:18 2005
From: carlos.ungil at bluewin.ch (Carlos Ungil)
Date: Tue, 2 Aug 2005 08:06:18 +0200
Subject: [Bese-devel] problem with actions
Message-ID: <b7cce589b70827e8848be17bc29a43fa@bluewin.ch>

Hello,

I don't understand why the code below is not working. Well, I bet the 
code is fine and the problem is only in my expectations. I hope this 
will be obvious to someone who can give me a hint.

The test-list component (/test/index.ucw) shows a list of ten items 
0..9, and clicking on one of them won't show that number, as I would 
like, but all of them display the number 10. I think this is because 10 
is the value the of i variable after the loop, and somehow this is the 
only value used by the system.

How can I make the display-number actions see the different values of 
the variable, and not just the final one? Maybe my approach is 
completely wrong. In that case, what  is the right construct to get 
something like this? What I really want to do is to show a list of 
objects, displaying the object when clicking on link.

Thanks [for any help on this, and for making the ucw code available in 
the first place!],

Carlos

-=-=-=-=-=-=-=-=-=-

(in-package :it.bese.ucw-user)

(defvar *test-application*
   (make-instance 'cookie-session-application
                  :url-prefix "/test/"))

(register-application *default-server* *test-application*)

(defentry-point "index.ucw" (:application *test-application*) ()
   (call 'test-list))

(defcomponent test-list (simple-window-component) ())

(defcomponent test-display (simple-window-component)
   ((number :accessor number :initarg :number)))

(defaction display-number ((c test-list) number)
   (call 'test-display :number number))

(defmethod render-on ((res response) (app test-list))
   (<:ul (loop for i from 0 below 10
	      do (<:li (<ucw:a :action (display-number app i)
			       (<:as-html i))))))

(defmethod render-on ((res response) (app test-display))
   (<:as-html (number app)))



From drewc at tech.coop  Tue Aug  2 22:17:04 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Tue, 02 Aug 2005 15:17:04 -0700
Subject: [Bese-devel] problem with actions
Message-ID: <42EFF0E0.1050409@tech.coop>

Carlos Ungil wrote:
> Hello,

> The test-list component (/test/index.ucw) shows a list of ten items 
> 0..9, and clicking on one of them won't show that number, as I would 
> like, but all of them display the number 10. I think this is because 10 
> is the value the of i variable after the loop, and somehow this is the 
> only value used by the system.

> (defmethod render-on ((res response) (app test-list))
>   (<:ul (loop for i from 0 below 10
>           do (<:li (<ucw:a :action (display-number app i)
>                    (<:as-html i))))))

This is because your implementation is free to re-use the binding of i
within the loop, and the <ucw:a is creating a closure over that binding.
  consider the following code :

CL> (defparameter *closures*
      (loop for i from 0 below 10 collect (lambda () i)))
*CLOSURES*
CL> (loop for c in *closures* collect (funcall c))
(10 10 10 10 10 10 10 10 10 10)
CL> (defparameter *closures*
      (loop for i from 0 below 10 collect (let ((i i))
                                           (lambda () i))))
*CLOSURES*
CL> (loop for c in *closures* collect (funcall c))
(0 1 2 3 4 5 6 7 8 9)

This is a common CL gotcha. All you need to do is change your loop to
look like this :

(loop for i from 0 below 10
          do (let ((i i))
              (<:li (<ucw:a :action (display-number app i)
                   (<:as-html i)))))))

drewc




From astor at pvv.org  Tue Aug  2 07:21:12 2005
From: astor at pvv.org (Alexander Kjeldaas)
Date: Tue, 02 Aug 2005 09:21:12 +0200
Subject: [Bese-devel] problem with actions
In-Reply-To: <42EFF0E0.1050409@tech.coop>
References: <42EFF0E0.1050409@tech.coop>
Message-ID: <42EF1EE8.4050100@pvv.org>

Drew Crampsie wrote:

> This is because your implementation is free to re-use the binding of i
> within the loop, and the <ucw:a is creating a closure over that binding.
>  consider the following code :
>
[...]

Shouldn't ideally the macro-expansion of display-number take care of 
this?  Or are there situations where this behaviour is necessary?

astor



From drewc at tech.coop  Tue Aug  2 23:07:06 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Tue, 02 Aug 2005 16:07:06 -0700
Subject: [Bese-devel] problem with actions
Message-ID: <42EFFC9A.2090503@tech.coop>

Alexander Kjeldaas wrote:
> Drew Crampsie wrote:
> 
>> This is because your implementation is free to re-use the binding of i
>> within the loop, and the <ucw:a is creating a closure over that binding.
>>  consider the following code :
>>
> [...]
> 
> Shouldn't ideally the macro-expansion of display-number take care of 
> this?  Or are there situations where this behaviour is necessary?

How would the macro-expansion find out which lexicals is has to rebind?

This behavior is part of the ANSI specification (not specific to UCW), 
and while i've never found it useful, it is effecient from an 
implementation point-of-view.

Imagine a loop over a large set. If the implementation kept all those
bindings around it would require twice the memory to run the loop, and
then a massive amount of garbage to clean up afterwards (in the normal 
case, where the binding is not closed over). Not an ideal
situation.

drewc




From marco_simon at web.de  Tue Aug  2 08:28:07 2005
From: marco_simon at web.de (Marco Simon)
Date: Tue, 02 Aug 2005 10:28:07 +0200
Subject: [Bese-devel] Publish-Directory error 
In-Reply-To: <42EFFC9A.2090503@tech.coop>
References: <42EFFC9A.2090503@tech.coop>
Message-ID: <42EF2E97.3080104@web.de>

Does anybody know what the following error-message means, and what
I have to do against it ?

(+WARN+ 3331959573 IT.BESE.UCW::UCW.BACKEND "Attempting to publish 
#p\"/home/ucw/ucw/bin/../wwwroot/./ucw/examples/\" at \"/ucw/admin/\" 
but mod_lisp backend does not support publish-directory.")

Thanks in advance,
 Marco



From mb at bese.it  Tue Aug  2 08:47:36 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 02 Aug 2005 10:47:36 +0200
Subject: [Bese-devel] Publish-Directory error
In-Reply-To: <42EF2E97.3080104@web.de> (Marco Simon's message of "Tue,
	02 Aug 2005 10:28:07 +0200")
References: <42EFFC9A.2090503@tech.coop> <42EF2E97.3080104@web.de>
Message-ID: <m264uopvx3.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> Does anybody know what the following error-message means, and what
> I have to do against it ?

it's just informing you that ucw can't configure apache by itself but
you must manually setup the DocumentReet and/or Location tags in
httpd.conf to make sure apache sees the stylesheets and
what not. there's nothing you can do about it (you could disable the
warning if it worries you).

> (+WARN+ 3331959573 IT.BESE.UCW::UCW.BACKEND "Attempting to publish
> #p\"/home/ucw/ucw/bin/../wwwroot/./ucw/examples/\" at \"/ucw/admin/\"
> but mod_lisp backend does not support publish-directory.")

if this is misleading i'd be willing to change it to an +INFO+ message
(since there's nothing really wrong with this setup).

-- 
-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  Tue Aug  2 14:26:19 2005
From: marco_simon at web.de (Marco Simon)
Date: Tue, 02 Aug 2005 16:26:19 +0200
Subject: [Bese-devel] Several beginner's questions
Message-ID: <42EF828B.1040702@web.de>

Hi ucw-list,

I've been trying to set up and understand the ucw-system for
some days now, but I'm still fighting with some beginners problems.

First of all, the myhost/ucw/examples/index.ucw works fine.

What I don't understand so far:
There are several subdirectories within the /ucw - tree, and
I'm not sure if I got their function right so far.

/ucw/examples/ seems to be the application-directory for the
dynamic-part of the demo-application. This is the place where
I have to change the code if I want to modify the ucw-servers
webpage-output.

/ucw/wwwroot seems to hold some static files (like the stylesheet.css)
as well as some dynamic parts of the demo application (the counter-demo
doesn't work without this directory). Besides that, this directory must be
reachable by apache (or even better must be its document-root) for
serving static files like images and css-files, right ?

/ucw/bin/ holds the start-skript and initialisation-files.
/ucw/logs/ is the place for all ucw log files
/ucw/var/ holds the socket and pid files
/ucw/doc/ is more or less empty. Just keeps some issue and todo-infos.
/ucw/src/ mh - I'm not really clear about this directory's content.

After I got running the /ucw/examples/index.ucw I tried to follow your
great "hello-world" movie. But I'm stuck several times.

The very first problem: My emacs/slime/cmucl environment doesn't know
the it.bese.ucw-user package. So I tried just to do a
(load /home/ucw/src/packages.lisp)
After ignoring the following compilation-errors (via "make this package"):
Type-error in LISP::PAGAGE-OR-LOSE: "IT.BESE.ARNESI" is not of type
PACKAGE
Type-error in LISP::PAGAGE-OR-LOSE: "IT.BESE.YYCLML" is not of type
PACKAGE
Type-error in LISP::PAGAGE-OR-LOSE: "ITERATE" is not of type PACKAGE
I was able to change-package to IT.BESE.UCW-USER.
Next step was typing the first lines of your hello world demo in the
repl (for debugging reasons).
What made me write you this mail finally, was the following error:

CL-USER> (load "/home/ucw/ucw/src/packages.lisp")
; Loading #p"/home/ucw/ucw/src/packages.lisp".
T
CL-USER>
UCW-USER> (in-package :it.bese.ucw-user)
#<The IT.BESE.UCW-USER package, 0/9 internal, 0/2 external>
UCW-USER> (defvar *hello-world* (make-instance
'cookie-session-application :url-prefix "/hello/" ))

Error in function PCL::FIND-CLASS-FROM-CELL:
   No class named COOKIE-SESSION-APPLICATION.
   [Condition of type SIMPLE-ERROR]

Restarts:
  0: [ABORT] Abort handling SLIME request.
  1: [ABORT] Return to Top-Level.


I guess I've made some fundamental mistake at the very beginning and
I would be glad if you could give some hints and further explication.

Thanks a lot,
 best regards
  Marco Simon








From der_julian at web.de  Tue Aug  2 14:44:06 2005
From: der_julian at web.de (Julian Stecklina)
Date: Tue, 2 Aug 2005 16:44:06 +0200
Subject: [Bese-devel] Several beginner's questions
In-Reply-To: <42EF828B.1040702@web.de>
References: <42EF828B.1040702@web.de>
Message-ID: <20050802164406.13315ef1@localhost>

On Tue, 02 Aug 2005 16:26:19 +0200
Marco Simon <marco_simon at web.de> wrote:

> The very first problem: My emacs/slime/cmucl environment doesn't know
> the it.bese.ucw-user package. So I tried just to do a
> (load /home/ucw/src/packages.lisp)

(load "/home/ucw/bin/start.lisp") will do the job.

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 smoof-ra at elder-gods.org  Tue Aug  2 15:05:28 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Tue, 2 Aug 2005 11:05:28 -0400
Subject: [Bese-devel] progv
Message-ID: <20050802150528.GA7278@melkor.elder-gods.org>

here's a patch that adds progv to the code walker

       --larry 
-------------- next part --------------

New patches:

[labels can have declarations inside the body
smoof-ra at elder-gods.org**20050801193433] {
hunk ./src/walk.lisp 559
-      (setf (body labels) (walk-implict-progn labels body env)))))
+      (multiple-value-setf ((body labels) nil (declares labels)) (walk-implict-progn labels body env :declare t)))))
}

[progv
smoof-ra at elder-gods.org**20050802150342] {
hunk ./src/walk.lisp 744
+;;;; PROGV 
+
+(defclass progv-form (form implicit-progn-mixin)
+  ((vars-form :accessor vars-form :initarg :vars-form)
+   (values-form :accessor values-form :initarg :values-form)))
+
+(defwalker-handler progv (form parent env)
+  (with-form-object (progv progv-form :parent parent :source form)
+    (setf (vars-form progv)   (walk-form (cadr form) progv env))
+    (setf (values-form progv) (walk-form (caddr form) progv env))
+    (setf (body progv) (walk-implict-progn progv (cdddr form) env))
+    progv))
+    
+
+    
}

Context:

[declares needs to be copied in the labels handler just like the other lambda-function-form slots
smoof-ra at elder-gods.org**20050801193107] 
[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] 
[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] 
[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
b414dedf82963a6c40be64e8e9a44ab9e2bc88e3

From smoof-ra at elder-gods.org  Tue Aug  2 16:56:38 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Tue, 2 Aug 2005 12:56:38 -0400
Subject: [Bese-devel] a little patch
Message-ID: <20050802165638.GA14800@melkor.elder-gods.org>


this little patch modifies find-walker-handler so that if there is a
handler for a "special form" that isn't really a special form it will
use that handler instead of calling it an function call.  It shouldn't 
effect ucw behavior at all because all of the defwalker-handler's are 
real special forms, but it lets you shadow *walker-handlers* to add your 
own new special forms.

    --larry 
-------------- next part --------------

New patches:

[allow new special forms to be added to the walker by shadowing *walker-handlers*
smoof-ra at elder-gods.org**20050802165355] {
hunk ./src/walk.lisp 31
-      (case (car form)
-        ((block declare flet function go if labels let let*
-          macrolet progn quote return-from setq symbol-macrolet
-          tagbody unwind-protect catch multiple-value-call
-          multiple-value-prog1 throw load-time-value the
-          eval-when locally progv)
-         (or
-          (gethash (car form) *walker-handlers*)
-          (error "Sorry, No walker for the special operater ~S defined." (car form))))
-        (t (gethash 'application *walker-handlers*)))))
+      (aif (gethash (car form) *walker-handlers*)
+	   it
+	   (case (car form)
+	     ((block declare flet function go if labels let let*
+		     macrolet progn quote return-from setq symbol-macrolet
+		     tagbody unwind-protect catch multiple-value-call
+		     multiple-value-prog1 throw load-time-value the
+		     eval-when locally progv)
+	      (error "Sorry, No walker for the special operater ~S defined." (car form)))
+	     (t (gethash 'application *walker-handlers*))))))
}

Context:

[progv
smoof-ra at elder-gods.org**20050802150342] 
[labels can have declarations inside the body
smoof-ra at elder-gods.org**20050801193433] 
[declares needs to be copied in the labels handler just like the other lambda-function-form slots
smoof-ra at elder-gods.org**20050801193107] 
[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] 
[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] 
[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
b8d27e89545c5f3b9537124e8c8a5f4799f280c4

From marco_simon at web.de  Wed Aug  3 11:36:11 2005
From: marco_simon at web.de (Marco Simon)
Date: Wed, 03 Aug 2005 13:36:11 +0200
Subject: [Bese-devel] Several beginner's questions
In-Reply-To: <20050802164406.13315ef1@localhost>
References: <42EF828B.1040702@web.de> <20050802164406.13315ef1@localhost>
Message-ID: <42F0AC2B.3000402@web.de>

Julian Stecklina schrieb:

>On Tue, 02 Aug 2005 16:26:19 +0200
>Marco Simon <marco_simon at web.de> wrote:
>
>  
>
>>The very first problem: My emacs/slime/cmucl environment doesn't know
>>the it.bese.ucw-user package. So I tried just to do a
>>(load /home/ucw/src/packages.lisp)
>>    
>>
>
>(load "/home/ucw/bin/start.lisp") will do the job.
>
>Regards,
>  
>
Hi Julian,
thanks a lot for that hint - after some futher trying it made my
ucw-environment working.

Now I face a further minor problems with the demo-applications
"presentation" :
As soon as I click the "edit" button or the "Create" link in the
presentation-example, the following error is shown in my browser's
window:


  ERROR


    Error in KERNEL::UNBOUND-SYMBOL-ERROR-HANDLER: the variable
    IT.BESE.UCW:INSTANCE is unbound.
    <http://10.0.0.9/ucw/examples/index.ucw?s=iURaXsstWddoHinbExUBLcxSbMWGpoLXzUHfKBGu&f=zabBfbzkGenrYXbLJZei&a=vsSLSQIDoK>


    BACKTRACE

[ + ] (ERROR UNBOUND-VARIABLE :FUNCTION-NAME...

Description 	(ERROR UNBOUND-VARIABLE :FUNCTION-NAME (MAKE-INSTANCE 
IT.BESE.UCW-USER::PERSON-EDITOR IT.BESE.UCW:INSTANCE PCL::|.P0.|) :NAME 
...)


and so on...

The REPL/logs just shows:
2005-08-03T13:24.20 IT.BESE.ARNESI:+ERROR+ IT.BESE.UCW::UCW-LOGGER: 
Error #<UNBOUND-VARIABLE {58725D75}> while serving action.
2005-08-03T13:24.20 IT.BESE.ARNESI:+ERROR+ IT.BESE.UCW::UCW-LOGGER: 
Aborting action.


The other actions of that example seem to work without any problems.
Any ideas what the Problem might be ?


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050803/91314837/attachment.html>

From mb at bese.it  Wed Aug  3 11:46:09 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 03 Aug 2005 13:46:09 +0200
Subject: [Bese-devel] Several beginner's questions
References: <42EF828B.1040702@web.de>
Message-ID: <m2r7dbw8e6.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> Hi ucw-list,
>
> I've been trying to set up and understand the ucw-system for
> some days now, but I'm still fighting with some beginners problems.
>
> First of all, the myhost/ucw/examples/index.ucw works fine.
>
> What I don't understand so far:
> There are several subdirectories within the /ucw - tree, and
> I'm not sure if I got their function right so far.
>
> /ucw/examples/ seems to be the application-directory for the
> dynamic-part of the demo-application. This is the place where
> I have to change the code if I want to modify the ucw-servers
> webpage-output.

yes, but i think you're missing the point.

inside ucw/examples are a bunch of files which, when _loaded_, will
create some objects and functions in the lisp image which causes the
url '/ucw/examples/index.ucw' to do what it does. changing those files
is not enough (and, often, not neccessary), what's important is to
change the things those files define.

for example: load up ucw (using the start.lisp script or bobstopper's
new control.lisp stuff). then connect to the lisp and simply type:

(defmethod render-on ((res response) (welcome example-welcome))
  (<:as-html "Hello, World!"))

that will change the html generated by the home page. you needn't edit
or save or even have a copy of the example.lisp file which contains
the original definition of render-on.

> /ucw/wwwroot seems to hold some static files (like the stylesheet.css)
> as well as some dynamic parts of the demo application (the counter-demo
> doesn't work without this directory). Besides that, this directory must be
> reachable by apache (or even better must be its document-root) for
> serving static files like images and css-files, right ?

countre-demo doesn't work because i put static and tal files in the
same directory (this probably isn't such a great idea). the counter
component uses a TAL file (which is one of the many templating tools
for lisp) and so won't work without that file. had you defined a
render-on method for the counter component you wouldn't need that
directory.

> /ucw/bin/ holds the start-skript and initialisation-files.

yes.

> /ucw/logs/ is the place for all ucw log files

yes, but only because of this:

(setup-loggers (make-pathname :name nil :type nil
                              :directory (append (pathname-directory *load-truename*)
                                                 (list :up "logs"))
                              :defaults *load-truename*)
               +info+)

which is located in bin/start.lisp. you colud just as well do:

(setup-loggers "/var/log/ucw/" +info+)

and you'd no longer need ucw/logs/

> /ucw/var/ holds the socket and pid files

yes, but only because of what we pass to detachtty in ucwctl.

> /ucw/doc/ is more or less empty. Just keeps some issue and todo-infos.

correct.

> /ucw/src/ mh - I'm not really clear about this directory's content.

that directory holds the source code for uncommon web itself.

> After I got running the /ucw/examples/index.ucw I tried to follow your
> great "hello-world" movie. But I'm stuck several times.
>
> The very first problem: My emacs/slime/cmucl environment doesn't know
> the it.bese.ucw-user package. So I tried just to do a
> (load /home/ucw/src/packages.lisp)
> After ignoring the following compilation-errors (via "make this package"):
> Type-error in LISP::PAGAGE-OR-LOSE: "IT.BESE.ARNESI" is not of type
> PACKAGE
> Type-error in LISP::PAGAGE-OR-LOSE: "IT.BESE.YYCLML" is not of type
> PACKAGE
> Type-error in LISP::PAGAGE-OR-LOSE: "ITERATE" is not of type PACKAGE
> I was able to change-package to IT.BESE.UCW-USER.
> Next step was typing the first lines of your hello world demo in the
> repl (for debugging reasons).
> What made me write you this mail finally, was the following error:
>
> CL-USER> (load "/home/ucw/ucw/src/packages.lisp")
> ; Loading #p"/home/ucw/ucw/src/packages.lisp".
> T
> CL-USER>
> UCW-USER> (in-package :it.bese.ucw-user)
> #<The IT.BESE.UCW-USER package, 0/9 internal, 0/2 external>
> UCW-USER> (defvar *hello-world* (make-instance
> 'cookie-session-application :url-prefix "/hello/" ))
>
> Error in function PCL::FIND-CLASS-FROM-CELL:
>    No class named COOKIE-SESSION-APPLICATION.
>    [Condition of type SIMPLE-ERROR]
>
> Restarts:
>   0: [ABORT] Abort handling SLIME request.
>   1: [ABORT] Return to Top-Level.

loading the packages file defines the packages but does _not_ load up
the rest of the code. se while you've told lisp that there's a package
named :it.bese.ucw and that that packgae has an external symbol named
cookie-session-application, you haven't created the class and much
less bound it to the proper name.

you'll notice that one of the first things start.lisp does is this:

(asdf:oos 'asdf:load-op :ucw)

this causes the package file to be loaded up (first) and then it loads
up the rest of ucw itself. notice that in ucw.asd we have a list of
every file in ucw/src with their dependencies.

> I guess I've made some fundamental mistake at the very beginning and
> I would be glad if you could give some hints and further explication.

i think you're only mistako is not realizing how lisp code is loaded
and what the association between file and a running lisp is.

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  Wed Aug  3 12:25:03 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 03 Aug 2005 14:25:03 +0200
Subject: [Bese-devel] Several beginner's questions
In-Reply-To: <42F0AC2B.3000402@web.de> (Marco Simon's message of "Wed,
	03 Aug 2005 13:36:11 +0200")
References: <42EF828B.1040702@web.de> <20050802164406.13315ef1@localhost>
	<42F0AC2B.3000402@web.de>
Message-ID: <m2wtn3us0w.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> Hi Julian,
> thanks a lot for that hint - after some futher trying it made my
> ucw-environment working.
>
> Now I face a further minor problems with the demo-applications
> "presentation" :
> As soon as I click the "edit" button or the "Create" link in the
> presentation-example, the following error is shown in my browser's
> window:
 
would you mind sending a complete backtrace? (just cut 'n paste the
sldb buffer with all the locals displayed)

-- 
-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 smoof-ra at elder-gods.org  Wed Aug  3 22:31:18 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 3 Aug 2005 18:31:18 -0400
Subject: [Bese-devel] :lexical-flet
Message-ID: <20050803223118.GA16461@melkor.elder-gods.org>


Here's another walker patch.  It adds a :lexical-flet mark similar to
the :lexical-let one, so it can recognise flets from the surrounding
environment.  This patch only adds placeholder functions for cmucl and
openmcl, so it should have no effect on those. 		 

	 --larry 

-------------- next part --------------

New patches:

[recognise flets from the lexical environment (on sbcl)
smoof-ra at elder-gods.org**20050803222732] {
hunk ./src/lexenv.lisp 27
+;fixme 
+#+openmcl
+(defun lexical-functions (environment)
+  (declare (ignore environment))
+  nil)
+
hunk ./src/lexenv.lisp 43
+#+sbcl
+(defun lexical-functions (environment)
+  (mapcar #'first (sb-c::lexenv-funs environment)))
+
+
hunk ./src/lexenv.lisp 57
+
+;fixme 
+#+cmu
+(defun lexical-functions (environment)
+  (declare (ignore environment))
+  nil)
+
hunk ./src/walk.lisp 17
-        (setf walk-env (register walk-env :lexical-let var t))))
+        (setf walk-env (register walk-env :lexical-let var t)))
+      (dolist (fun (lexical-functions lexical-env))
+	(setf walk-env (register walk-env :lexical-flet fun t))))
hunk ./src/walk.lisp 267
+(defclass lexical-application-form (application-form)
+  ())
+
hunk ./src/walk.lisp 291
-         (multiple-value-bind (expansion expanded)
-             (macroexpand-1 form nil)
-           (when expanded
-             (return (walk-form expansion parent env)))))
+	(multiple-value-bind (expansion expanded)
+	    (macroexpand-1 form nil)
+	  (when expanded
+	    (return (walk-form expansion parent env)))))
hunk ./src/walk.lisp 297
-                     (make-instance 'free-application-form))))
+                     (if (lookup env :lexical-flet op)
+			 (make-instance 'lexical-application-form)
+			 (make-instance 'free-application-form)))))
hunk ./src/walk.lisp 325
+(defclass lexical-function-object-form (function-object-form)
+  ())
+
hunk ./src/walk.lisp 336
-                         'free-function-object-form)
+                         (if (lookup env :lexical-flet (second form))
+			     'lexical-function-object-form
+			     'free-function-object-form))
}

Context:

[minor comment fixup
Marco Baringer <mb at bese.it>**20050803085322] 
[Moved defclass progv-form to keep the walker classes defined in alphabetical order 
Marco Baringer <mb at bese.it>**20050803085254] 
[Added walker class for THE forms
Marco Baringer <mb at bese.it>**20050803085210] 
[Minor spacing fixs to the previous patch
Marco Baringer <mb at bese.it>**20050802152421] 
[allow new special forms to be added to the walker by shadowing *walker-handlers*
smoof-ra at elder-gods.org**20050802165355] 
[progv
smoof-ra at elder-gods.org**20050802150342] 
[labels can have declarations inside the body
smoof-ra at elder-gods.org**20050801193433] 
[declares needs to be copied in the labels handler just like the other lambda-function-form slots
smoof-ra at elder-gods.org**20050801193107] 
[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] 
[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] 
[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
38bda27e2c3b2c53b126f6c592652f05c40c4240

From mb at bese.it  Thu Aug  4 07:09:36 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 04 Aug 2005 09:09:36 +0200
Subject: [Bese-devel] :lexical-flet
In-Reply-To: <20050803223118.GA16461@melkor.elder-gods.org> (Larry D'Anna's
	message of "Wed, 3 Aug 2005 18:31:18 -0400")
References: <20050803223118.GA16461@melkor.elder-gods.org>
Message-ID: <m2irym9o0f.fsf@soma.local>

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

> Here's another walker patch.  It adds a :lexical-flet mark similar to
> the :lexical-let one, so it can recognise flets from the surrounding
> environment.  This patch only adds placeholder functions for cmucl and
> openmcl, so it should have no effect on those. 		 

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 mb at bese.it  Thu Aug  4 12:07:29 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 04 Aug 2005 14:07:29 +0200
Subject: [Bese-devel] new interpter and iterate
Message-ID: <m2slxp52im.fsf@soma.local>


the new cps interepreter (which i'll rename soon since it doesn't
actually cps transform anything) has one major drawback: it doesn't
really work within iterate. the reason for this is simple, let's
pretend you have this code (which does in fact appear in ucw's source
code):

(iterate
  (for object in some-list)
  (let ((object object))
    (<ucw:a :action (do-something object) "Do something.")))

the problem is that with-call/cc will, when macroexpanded, attempt to
walk the (do-something object) form. unfortunetly iterate will do the
macroexpansion instead of the complier so when the compiler
environment passed to with-call/cc doesn't contain a reference to
OBJECT and the walker thinks OBJECT is a free variable, when in fact
it's a local. bad things follow.

we've basically got two options:

1) fix iterate.

2) warn everyone that, as of now, <ucw:input or <ucw:form or <ucw:a
   tags inside iterate forms are a VERY BADY idea. (though they may
   still accidentally work).

for now i'll take option #2:

As of now <ucw:input or <ucw:form or <ucw:a tags inside iterate forms
are a VERY BADY idea. (though they may still accidentally work).

-- 
-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 Aug  4 13:59:20 2005
From: mb at bese.it (Marco Baringer)
Date: Thu, 04 Aug 2005 15:59:20 +0200
Subject: [Bese-devel] new cps interpreter WARNING
Message-ID: <m2oe8dyf9j.fsf@soma.local>


I have just moved all the patches from ucw_new_cps_integration into
ucw_dev and arnesi_cps into arnesi.

your code may very well break. i'm sorry, in the end it's worth it.

should i delete the ucw_new_cps_integration and arnesi_cps directories
so people don't keep puling from them and wonder why no new patches
ever come through?

-- 
-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 whalliburton at gmail.com  Fri Aug  5 04:11:23 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Thu, 4 Aug 2005 23:11:23 -0500
Subject: [Bese-devel] Re: Backtracking help
Message-ID: <4e7bd29e050804211172092f42@mail.gmail.com>

I have refactored the previoius example to make simpler. Could someone
please explain why I need the (let (x x)) in the example and how to
make the backtracking work propery. I have version that uses
individual slots for each member of the array and it backtracks
propery.

Thank you,
William Halliburton

;;; tictactoe

(defcomponent tictactoe (widget-component)
  ((state :backtrack #'copy-seq
	  :initform (make-array 9 :element-type 'fixnum))))

(defaction tictactoe-select ((tictactoe tictactoe) index)
  (with-slots (state) tictactoe
     (setf (aref state index) 1)))

(defmethod render-on ((res response) (tictactoe tictactoe))
  (with-slots (state) tictactoe
       (<:pre 
	(dotimes (x 9) 
	  (let ((x x))
	    (case (aref state x)
	      (0 (<ucw:a :action (tictactoe-select tictactoe x) (<:as-is " ")))
	      (1 (<:as-is "X"))
	      (2 (<:as-is "O"))))
	  (if (or (= x 2) (= x 5))
	      (progn
		(<:as-is #\Newline)
		(<:as-html "---------")
		(<:as-is #\Newline))
	    (if (= x 8) (<:as-is #\Newline)
	      (<:as-html " | ")))))))


From whalliburton at gmail.com  Fri Aug  5 03:18:05 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Thu, 4 Aug 2005 22:18:05 -0500
Subject: [Bese-devel] Backtracking help
Message-ID: <4e7bd29e050804201867dc6de@mail.gmail.com>

Hello.

I am just beginning to use UCW. The following example does not
backtrack properly.
What would be the proper way to write this. 

Thank you.
William Halliburton


;;; tictactoe

(defun tictactoe-copy-state (state)
  (let ((array (make-array 9 :element-type 'fixnum)))
    (dotimes (x 9) (setf (aref array x) (aref state x)))
    array))

(defcomponent tictactoe (widget-component)
  ((state :backtrack #'tictactoe-copy-state
	  :initform (make-array 9 :element-type 'fixnum) :initarg :state)))

(defaction tictactoe-select ((tictactoe tictactoe) index)
  (with-slots (state) tictactoe
     (setf (aref state index) 1)))

(defmethod render-on ((res response) (tictactoe tictactoe))
  (with-slots (state) tictactoe
     (flet ((state-out (index)
       (let ((val (aref state index)))
	 (case val
	   (0 (<ucw:a :action (tictactoe-select tictactoe index) (<:as-is " ")))
	   (1 (<:as-is "X"))
	   (2 (<:as-is "O"))))))
       (<:pre 
	(dotimes (x 9) 
	  (state-out x)
	  (if (or (= x 2) (= x 5))
	      (progn
		(<:as-is #\Newline)
		(<:as-html "---------")
		(<:as-is #\Newline))
	    (if (= x 8) (<:as-is #\Newline)
	      (<:as-html " | "))))))))


From pjb at informatimago.com  Fri Aug  5 02:16:25 2005
From: pjb at informatimago.com (Pascal Bourguignon)
Date: Fri,  5 Aug 2005 04:16:25 +0200 (CEST)
Subject: [Bese-devel] :initform or :initarg ?
Message-ID: <20050805021625.E712E10FBF3@thalassa.informatimago.com>


In ucw-0.3.9, presentations.lisp:

(defcomponent list-presentation (presentation)
  ((slots :accessor slots :initarg :slots)
   (editablep :accessor editablep :initform t :initarg :editablep)
   (edit-label :accessor edit-label :initform "Edit")
   (deleteablep :accessor deleteablep :initform :deleteablep)
   (delete-label :accessor delete-label :initform "Delete")
   (instances :accessor instances)))

I'm wondering if you didn't want to write:

   (deleteablep :accessor deleteablep :initarg :deleteablep)

instead?


-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
You're always typing.
Well, let's see you ignore my
sitting on your hands.


From mb at bese.it  Fri Aug  5 07:31:30 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 05 Aug 2005 09:31:30 +0200
Subject: [Bese-devel] :initform or :initarg ?
In-Reply-To: <20050805021625.E712E10FBF3@thalassa.informatimago.com> (Pascal
	Bourguignon's message of "Fri, 5 Aug 2005 04:16:25 +0200 (CEST)")
References: <20050805021625.E712E10FBF3@thalassa.informatimago.com>
Message-ID: <m2vf2kon59.fsf@soma.local>

Pascal Bourguignon <pjb at informatimago.com> writes:

> In ucw-0.3.9, presentations.lisp:
>
> (defcomponent list-presentation (presentation)
>   ((slots :accessor slots :initarg :slots)
>    (editablep :accessor editablep :initform t :initarg :editablep)
>    (edit-label :accessor edit-label :initform "Edit")
>    (deleteablep :accessor deleteablep :initform :deleteablep)
>    (delete-label :accessor delete-label :initform "Delete")
>    (instances :accessor instances)))
>
> I'm wondering if you didn't want to write:
>
>    (deleteablep :accessor deleteablep :initarg :deleteablep)
>
> instead?

good catch. fixed.

-- 
-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 whalliburton at gmail.com  Fri Aug  5 21:51:09 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Fri, 5 Aug 2005 16:51:09 -0500
Subject: [Bese-devel] Backtracking question
Message-ID: <4e7bd29e0508051451143b63b8@mail.gmail.com>

Hello. I am trying to get the hang of the backtracking.

I have a test component:

(defcomponent backtest (widget-component)
  ((val1 :backtrack t :initform nil :initarg :val1 :accessor backtest.val1)
   (val2 :backtrack #'copy-seq :initform '(nil) :initarg :val2
:accessor backtest.val2)))
   
(defaction backtest-toggle-1 ((backtest backtest))
  (setf (backtest.val1 backtest) (not (backtest.val1 backtest))))

(defaction backtest-toggle-2 ((backtest backtest))
  (setf (car (backtest.val2 backtest)) (not (car (backtest.val2 backtest)))))

(defmethod render-on ((res response) (backtest backtest))
  (<ucw:a :action (backtest-toggle-1 backtest) (<:as-html
(backtest.val1 backtest)))
  (<:as-is " ")
  (<ucw:a :action (backtest-toggle-2 backtest) (<:as-html (car
(backtest.val2 backtest)))))

Val1 works as expected. I can toggle it and it will not affect other
frames that have forked. Val2, on the other hand, does not work as
exprected. Toggling and then reloading the page will toggle again,
also forking the page and toggleing on one page will set the value in
all instances. I replaced copy-seq with my own function for copying
and it looks like it is never called. I am assuming that I have a
misunderstanding about how the backtracking works.

Does anyone have a good explanation on the backtracking and how to fix
this code.

Thank you,
William Halliburton


From whalliburton at gmail.com  Fri Aug  5 21:31:04 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Fri, 5 Aug 2005 16:31:04 -0500
Subject: [Bese-devel] code in inspector
Message-ID: <4e7bd29e050805143130172a84@mail.gmail.com>

The following hack makes the code output more readable. 

(defmethod render-on ((res response) (insp ucw-inspector))
  (multiple-value-bind (title content)
      (swank::inspect-for-emacs (slot-value insp 'datum)
(swank::make-default-inspector))
    (<:h2 (<:as-html title))
    (dolist (part content)
      (etypecase part
        (null nil)
        (string (if (< (length part) 100) (<:as-html part) (<:pre
(<:as-html part))))
        (cons (swank::destructure-case part
                ((:newline) (<:br))
                ((:value obj &optional (str (prin1-to-string obj)))
                 (<ucw:a :action (call-inspector insp obj)
                         (<:as-html str)))
                ((:action text lambda)
                 (<ucw:a :action (call-action insp lambda)
                         (<:as-html text)))))))
    (<:p (<ucw:a :action (ok insp)) "Ok.")))


From pjb at informatimago.com  Sat Aug  6 00:53:59 2005
From: pjb at informatimago.com (Pascal Bourguignon)
Date: Sat,  6 Aug 2005 02:53:59 +0200 (CEST)
Subject: [Bese-devel] Encodings?
Message-ID: <20050806005359.BB4E110FBF3@thalassa.informatimago.com>


What provisions exist to manage encodings?

I'm observing the following behavior:

With Mozilla 1.7, and with Safari, in a text area, I type (copy-paste
from emacs HELLO actually) random characters.  The text I receive
contains these characters encoded as &#9999; sequences.  

Content of attribute as received in clisp:

| Japanese (&#26085;&#26412;&#35486;)	&#12371;&#12435;&#12395;&#12385;&#12399;, &#65402;&#65437;&#65414;&#65409;&#65418;
| Chinese (&#20013;&#25991;,&#26222;&#36890;&#35805;,&#27721;&#35821;)	&#20320;&#22909;
| Cantonese (&#31925;&#35486;,&#24291;&#26481;&#35441;)	&#26089;&#26216;, &#20320;&#22909;
| Korean (&#54620;&#44544;)	&#50504;&#45397;&#54616;&#49464;&#50836;, &#50504;&#45397;&#54616;&#49901;&#45768;&#44620;
| 
| Difference among chinese characters in GB, JIS, KSC, BIG5:
| GB	&#20803;&#27668;  &#24320;&#21457;
| JIS	&#20803;&#27671;  &#38283;&#30330;
| KSC	&#20803;&#27683;  &#38283;&#30332;
| BIG5	&#20803;&#27683;  &#38283;&#30332;
 

But when I send back this text in the same textarea, the ampersands
get escaped:

Source of HTML received back:

<textarea cols="72" rows="4" name="EjJvepjqxS"
    >

Japanese (&amp;#26085;&amp;#26412;&amp;#35486;)	&amp;#12371;&amp;#12435;&amp;#12395;&amp;#12385;&amp;#12399;, &amp;#65402;&amp;#65437;&amp;#65414;&amp;#65409;&amp;#65418;
Chinese (&amp;#20013;&amp;#25991;,&amp;#26222;&amp;#36890;&amp;#35805;,&amp;#27721;&amp;#35821;)	&amp;#20320;&amp;#22909;
Cantonese (&amp;#31925;&amp;#35486;,&amp;#24291;&amp;#26481;&amp;#35441;)	&amp;#26089;&amp;#26216;, &amp;#20320;&amp;#22909;
Korean (&amp;#54620;&amp;#44544;)	&amp;#50504;&amp;#45397;&amp;#54616;&amp;#49464;&amp;#50836;, &amp;#50504;&amp;#45397;&amp;#54616;&amp;#49901;&amp;#45768;&amp;#44620;

Difference among chinese characters in GB, JIS, KSC, BIG5:
GB	&amp;#20803;&amp;#27668;  &amp;#24320;&amp;#21457;
JIS	&amp;#20803;&amp;#27671;  &amp;#38283;&amp;#30330;
KSC	&amp;#20803;&amp;#27683;  &amp;#38283;&amp;#30332;
BIG5	&amp;#20803;&amp;#27683;  &amp;#38283;&amp;#30332;

</textarea
  > 

So, it looks like either the text is not decoded on the way from the
browser to ucw, or is not encoded back from ucw to the browser.
Is this a bug or a feature?

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
Grace personified,
I leap into the window.
I meant to do that.


From mb at bese.it  Sat Aug  6 06:20:26 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 06 Aug 2005 08:20:26 +0200
Subject: [Bese-devel] code in inspector
In-Reply-To: <4e7bd29e050805143130172a84@mail.gmail.com> (William
	Halliburton's message of "Fri, 5 Aug 2005 16:31:04 -0500")
References: <4e7bd29e050805143130172a84@mail.gmail.com>
Message-ID: <m2u0i3r3h1.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> The following hack makes the code output more readable. 

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 mb at bese.it  Sat Aug  6 06:20:49 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 06 Aug 2005 08:20:49 +0200
Subject: [Bese-devel] Backtracking question
In-Reply-To: <4e7bd29e0508051451143b63b8@mail.gmail.com> (William
	Halliburton's message of "Fri, 5 Aug 2005 16:51:09 -0500")
References: <4e7bd29e0508051451143b63b8@mail.gmail.com>
Message-ID: <m2pssrr3ge.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> Does anyone have a good explanation on the backtracking and how to fix
> this code.

looks like a bug to me.

-- 
-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 Aug  6 09:10:21 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 06 Aug 2005 11:10:21 +0200
Subject: [Bese-devel] Backtracking question
In-Reply-To: <4e7bd29e0508051451143b63b8@mail.gmail.com> (William
	Halliburton's message of "Fri, 5 Aug 2005 16:51:09 -0500")
References: <4e7bd29e0508051451143b63b8@mail.gmail.com>
Message-ID: <m27jezjurm.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> Val1 works as expected. I can toggle it and it will not affect other
> frames that have forked. Val2, on the other hand, does not work as
> exprected. Toggling and then reloading the page will toggle again,
> also forking the page and toggleing on one page will set the value in
> all instances. I replaced copy-seq with my own function for copying
> and it looks like it is never called. I am assuming that I have a
> misunderstanding about how the backtracking works.

ok, this was something of a pain to track down but turned out to be a
bug in ucw's backtracking stuff. thanks for the report (and in
particular the test case).

to make a long story short we were copy'ing values where we didn't
need too (when creating the new backtrack list) and not doing it when
we should have (when restoring old backtracks after the user had used
the back butten). what ended up happening is that we'de save, for
example, (NIL) in the backtrack, you'do go forward a bit and then come
back. when you came back we used that exact list, instead of a copy,
and you'd change it (in the backtest-toggle-2 action). we'd then start
saving/copying that list, madness would ensue.

do a darcs pull from the ucw_dev repo to get this fix (the patch
you're interested in is called "Fix backtracking on non immediate
objects").

> Does anyone have a good explanation on the backtracking and how to fix
> this code.

how to fix it? the same way i fix everything! shoot marco! :)

-- 
-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 whalliburton at gmail.com  Sun Aug  7 00:14:30 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Sat, 6 Aug 2005 19:14:30 -0500
Subject: [Bese-devel] Another possible backtracking bug.
Message-ID: <4e7bd29e05080617146a0767d1@mail.gmail.com>

I am now using the latest ucw_dev.

Code:

(defcomponent calltest (widget-component)
  ((val :backtrack t :initform 0 :initarg :val :accessor calltest.val)))
   
(defaction calltest-increment ((calltest calltest))
  (incf (calltest.val calltest))
  (if (= (calltest.val calltest) 3) (call 'info-message :message "At
3" :ok-text nil)))

(defmethod render-on ((res response) (calltest calltest))
  (<ucw:a :action (calltest-increment calltest) (<:as-html
(calltest.val calltest))))


Starts at 0. Clicked to 1. Clicked to 2. Clicked to "At 3". Back to 2.
Reload send to "At 3".

Starts at 0. Clicked to 1. Clicked to 2. Clicked to "At 3". Back to 2.
Back to 1. Click sends to "At 3".

I will attempt to debug to get more familiar. 

Thanks,
Will


From whalliburton at gmail.com  Sun Aug  7 00:47:47 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Sat, 6 Aug 2005 19:47:47 -0500
Subject: [Bese-devel] Now a problem of a different sort.
Message-ID: <4e7bd29e050806174734256863@mail.gmail.com>

Hello.

My last message may not be a problem - it could be caused by using an
old arnesi without the new cps. I am unable to get ucw_dev to run.

I noticed that after moving to ucw_dev that the examples did not work
- KALL not found.

I darcs the latest arnesi and it contained KALL. For good mesure I got
a new ucw_dev. Reloaded ucw (sbcl + araneida). Now any request causes
an error. I am doing a (load "start.lisp") and then attempting to load
the examples/index.ucw.

 Am I using the proper versions? What is the requirements for ucw_dev?

Thank you,
Will

Backtrace:

 The value (SETF IT.BESE.UCW:COMPONENT.PLACE) is not of type SYMBOL.
   [Condition of type TYPE-ERROR]

Restarts:
  0: [SHOW-BACKTRACE] Send the client a backtrace page.
  1: [RETRY] Clearout the response and retry calling the action.
  2: [SERVER-ERROR] Send the client an internal server error page.
  3: [ABORT-RESPONSE] Abort this response and answer another request
  4: [ABORT] Exit debugger, returning to top level.

Backtrace:
  0: ((SB-PCL::FAST-METHOD
SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT (T)) #<unused argument>
#<unused argument> #<FUNCTION (LAMBDA NIL) {94852E5}>)
  1: (SWANK::CALL-WITH-BINDINGS ((*PRINT-PRETTY*) (*PRINT-LEVEL* . 4)
(*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*)
(*PRINT-PPRINT-DISPATCH* . #<SB-PRETTY:PPRINT-DISPATCH-TABLE
{9647C39}>) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*)
(*PRINT-ARRAY* . T) ...) #<FUNCTION (LAMBDA NIL) {94852A5}>)
  2: (SWANK::DEBUG-IN-EMACS #<TYPE-ERROR {9F05FC1}>)
  3: ((LAMBDA NIL))
  4: (SWANK::CALL-WITH-REDIRECTED-IO #<SWANK::CONNECTION {95FD239}>
#<CLOSURE (LAMBDA NIL) {9F096E5}>)
  5: (SWANK::CALL-WITH-CONNECTION #<SWANK::CONNECTION {95FD239}>
#<CLOSURE (LAMBDA NIL) {9F096D5}>)
  6: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-ACTION-ERROR (ERROR))
#<unavailable argument> #<unavailable argument> #<TYPE-ERROR
{9F05FC1}>)
  7: (SIGNAL #<TYPE-ERROR {9F05FC1}>)
  8: (ERROR TYPE-ERROR)
  9: (SB-KERNEL::OBJECT-NOT-SYMBOL-ERROR-HANDLER #<unavailable
argument> #.(SB-SYS:INT-SAP #XB7BBA4C0)
#<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7BBA1B0 :TYPE (* (STRUCT
SB-VM::OS-CONTEXT-T-STRUCT))> (398))
 10: (SB-KERNEL:INTERNAL-ERROR #.(SB-SYS:INT-SAP #XB7BBA1B0)
#<unavailable argument>)
 11: ("foreign function: call_into_lisp")
 12: ("foreign function: funcall2")
 13: ("foreign function: interrupt_internal_error")
 14: ("foreign function: sigtrap_handler")
 15: (SYMBOL-FUNCTION (SETF IT.BESE.UCW:COMPONENT.PLACE))
 16: ((SB-PCL::FAST-METHOD IT.BESE.ARNESI::EVALUATE/CPS
(IT.BESE.ARNESI::FREE-FUNCTION-OBJECT-FORM T T)) #<unavailable
argument> #<unavailable argument>
#<IT.BESE.ARNESI::FREE-FUNCTION-OBJECT-FORM # {9F00761}> #<unavailable
argument> (IT.BESE.ARNESI::K-FOR-CPS-APPLY
(#<IT.BESE.ARNESI::FREE-FUNCTION-OBJECT-FORM # {9F00761}>
#<IT.BESE.ARNESI::LOCAL-VARIABLE-REFERENCE #1=#:G6097 {9F007F9}>
#<IT.BESE.ARNESI::FREE-APPLICATION-FORM # {9F02369}>) NIL ((:LET #1# .
#<IT.BESE.UCW::PLACE # {9F02221}>) (:LET #:G6098 .
#2=#<IT.BESE.UCW-USER::EXAMPLE-WINDOW {9EF8739}>) . #3=(# # #))
(IT.BESE.ARNESI::K-FOR-EVALUATE/CPS-PROGN (#<IT.BESE.ARNESI::LET*-FORM
# {9F009F1}> #<IT.BESE.ARNESI::LET*-FORM # {9F01131}>) #3#
(IT.BESE.ARNESI::K-FOR-EVALUATE/CPS-PROGN # #3# #))))
 17: (IT.BESE.ARNESI::DRIVE-CPS #<CLOSURE (LAMBDA NIL) {9EF824D}>)
 18: ((LABELS IT.BESE.UCW::CALL-ACTION) #<CLOSURE (LAMBDA NIL) {9326A35}>)
 19: ((SB-PCL::FAST-METHOD IT.BESE.UCW::SERVICE
(IT.BESE.UCW::STANDARD-SESSION-FRAME IT.BESE.UCW::REQUEST-CONTEXT))
#<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW::STANDARD-SESSION-FRAME {9EF7059}>
#<IT.BESE.UCW::COOKIE-SESSION-REQUEST-CONTEXT {9EF6461}>)
 20: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL.
SB-PCL::.ARG0. SB-PCL::.ARG1.)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW::STANDARD-SESSION-FRAME {9EF7059}>
#<IT.BESE.UCW::COOKIE-SESSION-REQUEST-CONTEXT {9EF6461}>)
 21: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL.
SB-PCL::.ARG0. SB-PCL::.ARG1.)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW::STANDARD-SESSION 9
#<IT.BESE.UCW::STANDARD-SESSION-FRAME {9EF7D11}> {92AA189}>
#<IT.BESE.UCW::COOKIE-SESSION-REQUEST-CONTEXT {9EF6461}>)
 22: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL.
SB-PCL::.ARG0. SB-PCL::.ARG1.)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 1>
#<IT.BESE.UCW::COOKIE-SESSION-REQUEST-CONTEXT {9EF6461}>)
 23: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-REQUEST
(IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:REQUEST
IT.BESE.UCW:RESPONSE)) #<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:ARANEIDA-BACKEND 2
{A516781}> #<IT.BESE.UCW::ARANEIDA-REQUEST {9EF6269}>
#<IT.BESE.UCW::ARANEIDA-RESPONSE {9EF6289}>)
 24: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST-RESPONSE
(IT.BESE.UCW::UCW-HANDLER T T)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW::UCW-HANDLER {904C1A1}> #<unavailable
argument> #<ARANEIDA:REQUEST {9EF5BB1}>)
 25: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST (ARANEIDA:HANDLER
T)) #<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW::UCW-HANDLER {904C1A1}> #<ARANEIDA:REQUEST {9EF5BB1}>)
 26: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST (ARANEIDA:HANDLER
T)) #<unavailable argument> #<unavailable argument>
#<ARANEIDA:DISPATCHING-HANDLER {A91D099}> #<ARANEIDA:REQUEST
{9EF5BB1}>)
 27: ((LABELS ARANEIDA::DO-IT) #<ARANEIDA:SERVE-EVENT-HTTP-LISTENER
{A8B0569}> #<SB-SYS:FD-STREAM for "a constant string" {9EF37C1}>)
 28: (ARANEIDA::SERVE-EVENT-HTTP-LISTENER-ACCEPT-ONE-REQUEST
#<ARANEIDA:SERVE-EVENT-HTTP-LISTENER {A8B0569}>)
 29: (SB-IMPL::SUB-SERVE-EVENT NIL 0)
 30: (SB-SYS:WAIT-UNTIL-FD-USABLE 0 :INPUT NIL)
 31: (SB-IMPL::REFILL-BUFFER/FD #<SB-SYS:FD-STREAM for "standard
input" {9001509}>)
 32: (SB-IMPL::INPUT-CHAR/ASCII #<SB-SYS:FD-STREAM for "standard
input" {9001509}> NIL #:EOF-OBJECT)
 33: (READ-CHAR #<SB-SYS:FD-STREAM for "standard input" {9001509}> NIL
#:EOF-OBJECT #<unused argument>)
 34: (READ-CHAR #<SYNONYM-STREAM :SYMBOL SB-SYS:*STDIN* {5012E31}> NIL
#:EOF-OBJECT #<unused argument>)
 35: (READ-CHAR #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL #:EOF-OBJECT #<unused
argument>)
 36: (READ-PRESERVING-WHITESPACE #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL (NIL) T)
 37: (READ-PRESERVING-WHITESPACE #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL (NIL) NIL)
 38: (READ #<SYNONYM-STREAM :SYMBOL SWANK::*CURRENT-STANDARD-INPUT*
{92B4819}> NIL (NIL) NIL)
 39: (SB-IMPL::REPL-READ-FORM-FUN #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-OUTPUT* {92B3371}>)
 40: (SB-IMPL::REPL-FUN NIL)
 41: ((LAMBDA NIL))
 42: ((LAMBDA NIL))
 43: (SB-IMPL::%WITH-REBOUND-IO-SYNTAX #<CLOSURE (LAMBDA NIL) {903A56D}>)
 44: (SB-IMPL::TOPLEVEL-REPL NIL)
 45: (SB-IMPL::TOPLEVEL-INIT)
 46: ((FLET SB-IMPL::RESTART-LISP))


From whalliburton at gmail.com  Sun Aug  7 04:13:46 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Sat, 6 Aug 2005 23:13:46 -0500
Subject: [Bese-devel] Now a problem of a different sort - solved
Message-ID: <4e7bd29e05080621133e8efe67@mail.gmail.com>

Looks like the walker was setting up free-function-object-form name to
be (setf foo) and this was dying on the symbol-function. Replacing
symbol-function with fdefinition fixes the problem.

(defmethod evaluate/cps ((func free-function-object-form) env k)
  (declare (ignore env))
  (if (fboundp (name func))
      (kontinue k (fdefinition (name func)))
    (error "Unbound function ~S." (name func))))

What is the proper way to submit bug fixes? I am new to darcs but have
used cvs for a long time.

Thanks,
Will


From mb at bese.it  Sun Aug  7 06:06:43 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 07 Aug 2005 08:06:43 +0200
Subject: [Bese-devel] Now a problem of a different sort - solved
In-Reply-To: <4e7bd29e05080621133e8efe67@mail.gmail.com> (William
	Halliburton's message of "Sat, 6 Aug 2005 23:13:46 -0500")
References: <4e7bd29e05080621133e8efe67@mail.gmail.com>
Message-ID: <m2y87ee0wc.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> Looks like the walker was setting up free-function-object-form name to
> be (setf foo) and this was dying on the symbol-function. Replacing
> symbol-function with fdefinition fixes the problem.

yeah, ran into that myself yesterday :(

> What is the proper way to submit bug fixes? I am new to darcs but have
> used cvs for a long time.

go to the arnesi directory, do a 'darcs record' it will ask you which
changes you want to include in the patch, when you're done you give
the patch a name. then you need to send off the patch you just
created. do a 'darcs send -o patch' and darcs asks you which patches
(of those available in the repository) you want to include in the new
patch file, in this case you'd just chose the last one. then send an
email and attach the file 'patch'. done.

do this often enough and you get commit privs :)

-- 
-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 Aug  7 06:10:08 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 07 Aug 2005 08:10:08 +0200
Subject: [Bese-devel] Now a problem of a different sort.
In-Reply-To: <4e7bd29e050806174734256863@mail.gmail.com> (William
	Halliburton's message of "Sat, 6 Aug 2005 19:47:47 -0500")
References: <4e7bd29e050806174734256863@mail.gmail.com>
Message-ID: <m2u0i2e0qn.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> Hello.
>
> My last message may not be a problem - it could be caused by using an
> old arnesi without the new cps. I am unable to get ucw_dev to run.

i ran your sample code and got the expected output (not the broken
output you apparently got), so i'm going to ignore the issue unless
you continue to experience problems.

> I noticed that after moving to ucw_dev that the examples did not work
> - KALL not found.

there _should_ be a check in ucw.asd which ensures that you aren't
running a new ucw with an old arnesi, i don't know why that didn't
kick in.

>  Am I using the proper versions? What is the requirements for ucw_dev?

up until a few minutes ago we had the following discrepency:

the ucw repo was expecting the 'old' arnesi cps transformer, but that
didn't exist anywore since i'd merged the patches into the (only)
arnesi repo. so the _only_ way to get things to work was to use the
latest stuff in ucw_dev and arnesi. i've just merged the two branches
(ucw and ucw_dev) and created an arnesi_dev branch which ucw_dev will
follow.

-- 
-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 Aug  7 14:00:06 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 07 Aug 2005 16:00:06 +0200
Subject: [Bese-devel] ucw bug reporting
Message-ID: <m2zmrtc0ex.fsf@soma.local>


hi,

  i just added a new restart to ucw's internal error handling
  mechanism: generate-backtrace-for-emacs. what this restart should do
  is open up a new emacs buffer and fill it with lots and lots of
  details regarding the error. the idea being that you send this to
  the list and I have enough info to at least understand the problem,
  without having to first ask you for a backtrace and let a few days
  go by.

  it's really flaky (both the emacs and lisp sides) and i'd appreciate
  it if people could test it out and make sure it works.

  p.s. - recent ucw breaks on openmcl without the fix provided here: 
  http://article.gmane.org/gmane.lisp.openmcl.devel/883

-- 
-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 Aug  7 14:10:50 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 07 Aug 2005 16:10:50 +0200
Subject: [Bese-devel] ucw bug reporting
In-Reply-To: <m2zmrtc0ex.fsf@soma.local> (Marco Baringer's message of "Sun,
	07 Aug 2005 16:00:06 +0200")
References: <m2zmrtc0ex.fsf@soma.local>
Message-ID: <m2u0i1bzx1.fsf@soma.local>

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

>   i just added a new restart to ucw's internal error handling
>   mechanism: generate-backtrace-for-emacs. what this restart should do
>   is open up a new emacs buffer and fill it with lots and lots of
>   details regarding the error. the idea being that you send this to
>   the list and I have enough info to at least understand the problem,
>   without having to first ask you for a backtrace and let a few days
>   go by.

in case anyone actuallly tries to look at the contents of that buffer
note that it's intended to be lookeed at after "M-x occur RET ^---
RET" at not directly.

-- 
-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 whalliburton at gmail.com  Mon Aug  8 01:32:22 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Sun, 7 Aug 2005 20:32:22 -0500
Subject: [Bese-devel] Backtracking bug still present in my runs.
Message-ID: <4e7bd29e05080718323c851906@mail.gmail.com>

Darcs a fresh arnesi_dev (the website still says that arnesi_cvs is
the development branch). Darcs a fresh ucw_dev. Load into sbcl +
araneida.

The calltest component from the previous email still displays the same behavior.

Will attempt to debug now.

Will


From mb at bese.it  Mon Aug  8 07:01:00 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 08 Aug 2005 09:01:00 +0200
Subject: [Bese-devel] Backtracking bug still present in my runs.
In-Reply-To: <4e7bd29e05080718323c851906@mail.gmail.com> (William
	Halliburton's message of "Sun, 7 Aug 2005 20:32:22 -0500")
References: <4e7bd29e05080718323c851906@mail.gmail.com>
Message-ID: <m2vf2gdiab.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> Darcs a fresh arnesi_dev (the website still says that arnesi_cvs is
> the development branch). 

thanks for pointing that out.

>                           Darcs a fresh ucw_dev. Load into sbcl +
> araneida.
>
> The calltest component from the previous email still displays the
> same behavior.

i get the correct behaviour (openmcl+mod_lisp and sbcl+mod_lisp on
darwin, latest ucw).

here's a short description of the machanics of backtracking which i
hope will help you in your debugging (you may or may not have already
figured this out):

every frame has a list representing the stuff it needs to
backtrack. for every place (which is usally a slot but can be any
setf'able form) have a cons cell composed of value and place, value is
some lisp object and place is an instance of the PLACE class with the
three closures for getting, setting and copying the value.

whenever a reuest comes in we call REINSTATE-BACKTRACK (this, all the
following two methods, are defined in ucw/src/backtrack.lisp) which
simply calls the setter closure and attempts to create the state of
the UI at the time the frame was generated. we now create a new frame
(representing this request/respone cycle) and call CLONE-BACKTRACK to
create the new frame's backtrack list, atm this is just a call to
copy-tree. we then call the action which probably changes the values
of the backtrack'd place and maybe adds new place (by calling the
BACKTRACK method). When the action is done, just _before_ calling the
render-on method, we store the current values of all the places which
need backtracking by calling SAVE-BACKTRACKED. By tracing these 4
methods you can get a very good idea of what's getting saved, what's
getting reloaded and when. 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 Aug  8 07:23:41 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 08 Aug 2005 09:23:41 +0200
Subject: [Bese-devel] openmcl support
Message-ID: <m2ek94dh8i.fsf@soma.local>


UCW's new backtrace stuff triggers a bug in openmcl, you'll need to
add put this to your .openmcl-init or whatever file:

(let ((ccl::*warn-if-redefine-kernel* nil))
  (defun write-a-cons (cons stream level)
    (declare (type cons cons) (type stream stream) (type fixnum level))
    (let ((print-length (get-*print-frob* '*print-length*))
          (level-1 (%i- level 1))
          (head (%car cons))
          (tail (%cdr cons)))
      (declare (type fixnum print-length) (type fixnum level-1))
      (unless (and *print-abbreviate-quote*
                   (write-abbreviate-quote head tail stream level-1))
        (progn
          (pp-start-block stream #\()
          (if (= print-length 0)
              (%write-string "..." stream)
              (progn
                (write-internal stream head level-1 nil)
                (write-internal stream tail level-1
                                (if (atom tail)
                                    print-length
                                    (%i- print-length 1)))))
          (pp-end-block stream #\)))))))

0.14.4 sholud fix 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 whalliburton at gmail.com  Mon Aug  8 16:21:14 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Mon, 8 Aug 2005 11:21:14 -0500
Subject: [Bese-devel] admin page currently broken
Message-ID: <4e7bd29e05080809211e09e4b0@mail.gmail.com>

Attempting to long into admin page with admin/admin. I believe the
error is in the rendering of the admin page.

Tried to trace down the problem, but brain still has not jelled on this code. 

Look like the problem is in that last case of

(defmethod evaluate/cps ((func free-application-form) env k)

The free-application-form is a mapcar over a list with the first
element a cps-closure. Dying on the mapcar because cps-closure is not
a function.

Below is backtrace plus new emacs backtrace (do not know if this is
working as It did not generate much informaton).

Will


Backtrace:
  0: ((SB-PCL::FAST-METHOD
SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT (T)) #<unused argument>
#<unused argument> #<FUNCTION (LAMBDA NIL) {9418345}>)
  1: (SWANK::CALL-WITH-BINDINGS ((*PRINT-PRETTY*) (*PRINT-LEVEL* . 4)
(*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*)
(*PRINT-PPRINT-DISPATCH* . #<SB-PRETTY:PPRINT-DISPATCH-TABLE
{9561021}>) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*)
(*PRINT-ARRAY* . T) ...) #<FUNCTION (LAMBDA NIL) {9418305}>)
  2: (SWANK::DEBUG-IN-EMACS #<TYPE-ERROR {9F2C021}>)
  3: ((LAMBDA NIL))
  4: (SWANK::CALL-WITH-REDIRECTED-IO #<SWANK::CONNECTION {95FD239}>
#<CLOSURE (LAMBDA NIL) {A16BC35}>)
  5: (SWANK::CALL-WITH-CONNECTION #<SWANK::CONNECTION {95FD239}>
#<CLOSURE (LAMBDA NIL) {A16BC25}>)
  6: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-ACTION-ERROR (ERROR T))
#<unavailable argument> #<unavailable argument> #<TYPE-ERROR
{9F2C021}> (#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 4 :DESCRIPTION
"(ERROR TYPE-ERROR)" :LOCALS (# #) :SOURCE-LOCATION (:ERROR "The
source-path (NIL) is not valid.")) #S(IT.BESE.UCW::BACKTRACE-FRAME
:INDEX 5 :DESCRIPTION "(SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER
      #<unavailable argument>
      #.(SB-SYS:INT-SAP #XB7C3C31C)
      #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010 :TYPE (*
                                                              (STRUCT
                                                              
SB-VM::OS-CONTEXT-T-STRUCT))>
      (14 78))" :LOCALS (# # # # #) :SOURCE-LOCATION (:LOCATION # #
#)) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 6 :DESCRIPTION
"(SB-KERNEL:INTERNAL-ERROR
      #.(SB-SYS:INT-SAP #XB7C3C010)
      #<unavailable argument>)" :LOCALS (# # #) :SOURCE-LOCATION
(:LOCATION # # #)) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 7
:DESCRIPTION "(\"foreign function: call_into_lisp\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign
function: call_into_lisp\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
8 :DESCRIPTION "(\"foreign function: funcall2\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign
function: funcall2\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
9 :DESCRIPTION "(\"foreign function: interrupt_internal_error\")"
:LOCALS NIL :SOURCE-LOCATION (:ERROR "The value
  #<SB-DI::BOGUS-DEBUG-FUN \"foreign function: interrupt_internal_error\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
10 :DESCRIPTION "(\"foreign function: sigtrap_handler\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign
function: sigtrap_handler\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
11 :DESCRIPTION "(MAPCAR
      #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
      (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
       #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))"
:LOCALS (# # #) :SOURCE-LOCATION (:LOCATION # # #))
#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 12 :DESCRIPTION "((LAMBDA
(IT.BESE.ARNESI::ARGUMENTS))
      (#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
       (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)))"
:LOCALS (# # # # #) :SOURCE-LOCATION (:LOCATION # # #))
#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 13 :DESCRIPTION
"(IT.BESE.ARNESI::EVALUATE-ARGUMENTS-THEN-APPLY
      #<CLOSURE (LAMBDA #) {9F2785D}>
      NIL
      ((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
       #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
      ((:LET IT.BESE.UCW::SERVER-REPL . #<IT.BESE.UCW::ADMIN-REPL {9F222A1}>)
       (:LET IT.BESE.UCW::CONTROL-PANEL
             . #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>)
       (:BLOCK IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K)
       (:LET IT.BESE.UCW:SELF . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)
       (:LET IT.BESE.UCW::L . #<IT.BESE.UCW::ADMIN-LOGIN
{91C6791}>)))" :LOCALS (# # # # # # # #) :SOURCE-LOCATION (:LOCATION #
# #)) ...))
  7: (SIGNAL #<TYPE-ERROR {9F2C021}>)
  8: (ERROR TYPE-ERROR)
  9: (SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER #<unavailable argument>
#.(SB-SYS:INT-SAP #XB7C3C31C) #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP
#XB7C3C010 :TYPE (* (STRUCT SB-VM::OS-CONTEXT-T-STRUCT))> (14 78))
 10: (SB-KERNEL:INTERNAL-ERROR #.(SB-SYS:INT-SAP #XB7C3C010)
#<unavailable argument>)
 11: ("foreign function: call_into_lisp")
 12: ("foreign function: funcall2")
 13: ("foreign function: interrupt_internal_error")
 14: ("foreign function: sigtrap_handler")
 15: (MAPCAR #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
(#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
 16: ((LAMBDA (IT.BESE.ARNESI::ARGUMENTS))
(#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
(#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)))
      Locals:
        IT.BESE.ARNESI::ARGUMENTS = (#<IT.BESE.ARNESI::CPS-CLOSURE
{9F27B21}> (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
        IT.BESE.ARNESI::FUNC = :<NOT-AVAILABLE>
        #:G763 = :<NOT-AVAILABLE>
        #:G778 = :<NOT-AVAILABLE>
        IT.BESE.ARNESI::K = :<NOT-AVAILABLE>
      Catch-tags:
        #:SB-DEBUG-CATCH-TAG
 17: (IT.BESE.ARNESI::EVALUATE-ARGUMENTS-THEN-APPLY #<CLOSURE (LAMBDA
(IT.BESE.ARNESI::ARGUMENTS)) {9F2785D}> NIL
((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>) ((:LET
IT.BESE.UCW::SERVER-REPL . #<IT.BESE.UCW::ADMIN-REPL {9F222A1}>) (:LET
IT.BESE.UCW::CONTROL-PANEL . #<IT.BESE.UCW::ADMIN-CONTROL-PANEL
{9F20491}>) (:BLOCK IT.BESE.UCW:LOGIN-SUCCESSFUL
IT.BESE.ARNESI::TOPLEVEL-K) (:LET IT.BESE.UCW:SELF .
#1=#<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>) (:LET IT.BESE.UCW::L .
#1#)))
      Locals:
        IT.BESE.ARNESI::ENV = ((:LET IT.BESE.UCW::SERVER-REPL .
#<IT.BESE.UCW::ADMIN-REPL {9F222A1}>) (:LET IT.BESE.UCW::CONTROL-PANEL
. #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>) (:BLOCK
IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K) (:LET
IT.BESE.UCW:SELF . #1=#<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>) (:LET
IT.BESE.UCW::L . #1#))
        IT.BESE.ARNESI::EVALUATED-ARGUMENTS =
((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
        #:G1288 = :<NOT-AVAILABLE>
        #:G1292 = :<NOT-AVAILABLE>
        #:G1296 = :<NOT-AVAILABLE>
        #:G1304 = :<NOT-AVAILABLE>
        IT.BESE.ARNESI::HANDLER = #<CLOSURE (LAMBDA
(IT.BESE.ARNESI::ARGUMENTS)) {9F2785D}>
        IT.BESE.ARNESI::REMAINING-ARGUMENTS = NIL
      Catch-tags:
        #:SB-DEBUG-CATCH-TAG
 18: ((LAMBDA NIL))
 19: (IT.BESE.ARNESI::DRIVE-CPS #<CLOSURE (LAMBDA NIL) {9DC59D5}>)
 20: ((LABELS IT.BESE.UCW::CALL-ACTION) #<CLOSURE (LAMBDA NIL) {9C26ADD}>)
 21: ((SB-PCL::FAST-METHOD IT.BESE.UCW::SERVICE
(IT.BESE.UCW::STANDARD-SESSION-FRAME IT.BESE.UCW::REQUEST-CONTEXT))
#<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW::STANDARD-SESSION-FRAME {A3786D9}>
#<IT.BESE.UCW::STANDARD-REQUEST-CONTEXT {9CD25E9}>)
 22: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL.
SB-PCL::.ARG0. SB-PCL::.ARG1.)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW::STANDARD-SESSION-FRAME {A3786D9}>
#<IT.BESE.UCW::STANDARD-REQUEST-CONTEXT {9CD25E9}>)
 23: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL.
SB-PCL::.ARG0. SB-PCL::.ARG1.)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW::STANDARD-SESSION 3
#<IT.BESE.UCW::STANDARD-SESSION-FRAME {9DC4E49}> {A257291}>
#<IT.BESE.UCW::STANDARD-REQUEST-CONTEXT {9CD25E9}>)
 24: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL.
SB-PCL::.ARG0. SB-PCL::.ARG1.)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW::STANDARD-REQUEST-CONTEXT {9CD25E9}>)
 25: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-REQUEST
(IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:REQUEST
IT.BESE.UCW:RESPONSE)) #<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:ARANEIDA-BACKEND 2
{A5217D1}> #<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}>
#<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}>)
 26: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST-RESPONSE
(IT.BESE.UCW::UCW-HANDLER T T)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW::UCW-HANDLER {9F7F7B1}> #<unavailable
argument> #<ARANEIDA:REQUEST {9CCC129}>)
 27: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST (ARANEIDA:HANDLER
T)) #<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW::UCW-HANDLER {9F7F7B1}> #<ARANEIDA:REQUEST {9CCC129}>)
 28: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST (ARANEIDA:HANDLER
T)) #<unavailable argument> #<unavailable argument>
#<ARANEIDA:DISPATCHING-HANDLER {A91B1B9}> #<ARANEIDA:REQUEST
{9CCC129}>)
 29: ((LABELS ARANEIDA::DO-IT) #<ARANEIDA:SERVE-EVENT-HTTP-LISTENER
{A8A7139}> #<SB-SYS:FD-STREAM for "a constant string" {9CC8EE9}>)
 30: (ARANEIDA::SERVE-EVENT-HTTP-LISTENER-ACCEPT-ONE-REQUEST
#<ARANEIDA:SERVE-EVENT-HTTP-LISTENER {A8A7139}>)
 31: (SB-IMPL::SUB-SERVE-EVENT NIL 0)
 32: (SB-SYS:WAIT-UNTIL-FD-USABLE 0 :INPUT NIL)
 33: (SB-IMPL::REFILL-BUFFER/FD #<SB-SYS:FD-STREAM for "standard
input" {90014F9}>)
 34: (SB-IMPL::INPUT-CHAR/ASCII #<SB-SYS:FD-STREAM for "standard
input" {90014F9}> NIL #:EOF-OBJECT)
 35: (READ-CHAR #<SB-SYS:FD-STREAM for "standard input" {90014F9}> NIL
#:EOF-OBJECT #<unused argument>)
 36: (READ-CHAR #<SYNONYM-STREAM :SYMBOL SB-SYS:*STDIN* {5012E31}> NIL
#:EOF-OBJECT #<unused argument>)
 37: (READ-CHAR #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL #:EOF-OBJECT #<unused
argument>)
 38: (READ-PRESERVING-WHITESPACE #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL (NIL) T)
 39: (READ-PRESERVING-WHITESPACE #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL (NIL) NIL)
 40: (READ #<SYNONYM-STREAM :SYMBOL SWANK::*CURRENT-STANDARD-INPUT*
{92B4819}> NIL (NIL) NIL)
 41: (SB-IMPL::REPL-READ-FORM-FUN #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-OUTPUT* {92B3371}>)
 42: (SB-IMPL::REPL-FUN NIL)
 43: ((LAMBDA NIL))
 44: ((LAMBDA NIL))
 45: (SB-IMPL::%WITH-REBOUND-IO-SYNTAX #<CLOSURE (LAMBDA NIL) {903A56D}>)
 46: (SB-IMPL::TOPLEVEL-REPL NIL)
 47: (SB-IMPL::TOPLEVEL-INIT)
 48: ((FLET SB-IMPL::RESTART-LISP))


Emacs backtrace:

destructure-case failed: (:EVALUATE-IN-EMACS "
 (save-excursion
   (loop
     with buffer-name = \"*UCW Backtrace <%d>*\"
     for id upfrom 0
     for backtrace-buffer = (get-buffer (format buffer-name id))
     while backtrace-buffer
     finally do (switch-to-buffer-other-window (format buffer-name id)))
   (insert \"--- UCW Backtrace
---    Condition: #<TYPE-ERROR {9F2C021}>
#<TYPE-ERROR {9F2C021}> is a TYPE-ERROR. Its slot values are (:DATUM
#1=#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}> :EXPECTED-TYPE #2=(OR
FUNCTION SYMBOL) SB-KERNEL::EXPECTED-TYPE #2# SB-KERNEL::DATUM #1#).
---    Date: 2005-08-08T12:18:29
---    Lisp: \\\"SBCL\\\" \\\"0.9.2\\\"
---    Server: #<IT.BESE.UCW:STANDARD-SERVER
IT.BESE.UCW:ARANEIDA-BACKEND 2 {A5217D1}>
#<IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:ARANEIDA-BACKEND 2
{A5217D1}> is an instance of class #<STANDARD-CLASS
IT.BESE.UCW:STANDARD-SERVER>.
The following slots have :INSTANCE allocation:
 APPLICATIONS         (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/
1> #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
 APPLICATIONS-LOCK    :NULL-LOCK
 STARTED              T
 BACKEND              #<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}>
---    Backend: #<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}>
#<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW:ARANEIDA-BACKEND>.
The following slots have :INSTANCE allocation:
 LISTENER          #<ARANEIDA:SERVE-EVENT-HTTP-LISTENER {A8A7139}>
 DEFAULT-URL       #<ARANEIDA:HTTP-URL \\\"http://127.0.0.1:8080/\\\" {A8838D9}>
 LISTENER-CLASS    ARANEIDA:SERVE-EVENT-HTTP-LISTENER
---    Application: #<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1> is an instance of
class #<STANDARD-CLASS IT.BESE.UCW:STANDARD-APPLICATION>.
The following slots have :INSTANCE allocation:
 URL-PREFIX       \\\"/ucw/admin/\\\"
 SESSION-TYPE     IT.BESE.UCW::STANDARD-SESSION
 TAL-GENERATOR    #<IT.BESE.YACLML:FILE-SYSTEM-GENERATOR {A3F2DC1}>
 WWW-ROOTS       
(#P\\\"/home/conrad/website/ucw/ucw_dev/bin/../wwwroot/./ucw/examples/\\\")
 ENTRY-POINTS     #<HASH-TABLE :TEST EQUAL :COUNT 1 {9D4CBA1}>
 SESSION-TABLE    #<HASH-TABLE :TEST EQUAL :COUNT 1 {9D4CD79}>
 SERVER           #<IT.BESE.UCW:STANDARD-SERVER
IT.BESE.UCW:ARANEIDA-BACKEND 2 {A5217D1}>
---    Request: #<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}>
#<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW::ARANEIDA-REQUEST>.
The following slots have :INSTANCE allocation:
 REQUEST    #<ARANEIDA:REQUEST {9CCC129}>
---    Response: #<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}>
#<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW::ARANEIDA-RESPONSE>.
The following slots have :INSTANCE allocation:
 REQUEST           #<ARANEIDA:REQUEST {9CCC129}>
 HEADERS           ((\\\"Content-Type\\\" . \\\"text/html\\\")
(\\\"Status\\\" . \\\"200\\\"))
 CONTENT-STREAM    #<SB-IMPL::STRING-OUTPUT-STREAM {9CCD8B1}>
--- BACKTRACE
--- FRAME 4
(ERROR TYPE-ERROR)
---   Locals:
SB-DEBUG::ARG-0 ==> 5
SB-DEBUG::ARG-1 ==> TYPE-ERROR
---   Source:
(:ERROR \\\"The source-path (NIL) is not valid.\\\")
--- FRAME 5
(SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER
      #<unavailable argument>
      #.(SB-SYS:INT-SAP #XB7C3C31C)
      #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010 :TYPE (*
                                                              (STRUCT
                                                              
SB-VM::OS-CONTEXT-T-STRUCT))>
      (14 78))
---   Locals:
SB-DEBUG::ARG-0 ==> 4
SB-DEBUG::ARG-1 ==> :<NOT-AVAILABLE>
SB-DEBUG::ARG-2 ==> #.(SB-SYS:INT-SAP #XB7C3C31C)
SB-DEBUG::ARG-3 ==> #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010
:TYPE (* (STRUCT SB-VM::OS-CONTEXT-T-STRUCT))>
SB-DEBUG::ARG-4 ==> (14 78)
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/interr.lisp\\\")
(:POSITION 6404) (:SNIPPET \\\"(deferr object-not-type-error (object
type)
  (error (if (and (typep object 'instance)
		  (layout-invalid (%instance-layout object)))
	     'layout-invalid
	     'type-error)
	 :datum object
	 :expected-type type))

(deferr layout-invalid-error (object la\\\"))
--- FRAME 6
(SB-KERNEL:INTERNAL-ERROR
      #.(SB-SYS:INT-SAP #XB7C3C010)
      #<unavailable argument>)
---   Locals:
SB-DEBUG::ARG-0 ==> 2
SB-DEBUG::ARG-1 ==> #.(SB-SYS:INT-SAP #XB7C3C010)
SB-DEBUG::ARG-2 ==> :<NOT-AVAILABLE>
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/interr.lisp\\\")
(:POSITION 10988) (:SNIPPET \\\"(defun internal-error (context
continuable)
  (declare (type system-area-pointer context))
  (declare (ignore continuable))
  (/show0 \\\\\\\"entering INTERNAL-ERROR, CONTEXT=..\\\\\\\")
  (/hexstr context)
  (infinite-error-protect
   (/show0 \\\\\\\"about to bind ALIEN-CONTE\\\"))
--- FRAME 7
(\\\"foreign function: call_into_lisp\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: call_into_lisp\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 8
(\\\"foreign function: funcall2\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: funcall2\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 9
(\\\"foreign function: interrupt_internal_error\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value
  #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign function:
interrupt_internal_error\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 10
(\\\"foreign function: sigtrap_handler\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: sigtrap_handler\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 11
(MAPCAR
      #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
      (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
       #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
---   Locals:
SB-DEBUG::ARG-0 ==> 2
SB-DEBUG::ARG-1 ==> #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
SB-DEBUG::ARG-2 ==> (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/list.lisp\\\") (:POSITION
41554) (:SNIPPET \\\"(defun mapcar (function list &rest more-lists)
  #!+sb-doc
  \\\\\\\"Apply FUNCTION to successive elements of LIST. Return list of FUNCTION
   return values.\\\\\\\"
  (map1 function (cons list more-lists) :list t))

(defun mapcan (function list &rest more-lists)
  #!+s\\\"))
--- FRAME 12
((LAMBDA (IT.BESE.ARNESI::ARGUMENTS))
      (#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
       (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)))
---   Locals:
IT.BESE.ARNESI::ARGUMENTS ==> (#<IT.BESE.ARNESI::CPS-CLOSURE
{9F27B21}> (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
IT.BESE.ARNESI::FUNC ==> :<NOT-AVAILABLE>
#:G763 ==> :<NOT-AVAILABLE>
#:G778 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::K ==> :<NOT-AVAILABLE>
---   Source:
(:LOCATION (:FILE
\\\"/home/conrad/website/ucw/arnesi_dev/src/cps.lisp\\\") (:POSITION
6830) (:SNIPPET \\\"(apply (fdefinition (operator func)) arguments))))
      (arguments func) '()
      env))))

(defmethod evaluate/cps ((func local-application-form) env k)
  (evaluate/cps-apply (arguments func) (list (list (lookup env :flet
(operator func) :error-p t))) en\\\"))
--- FRAME 13
(IT.BESE.ARNESI::EVALUATE-ARGUMENTS-THEN-APPLY
      #<CLOSURE (LAMBDA #) {9F2785D}>
      NIL
      ((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
       #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
      ((:LET IT.BESE.UCW::SERVER-REPL . #<IT.BESE.UCW::ADMIN-REPL {9F222A1}>)
       (:LET IT.BESE.UCW::CONTROL-PANEL
             . #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>)
       (:BLOCK IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K)
       (:LET IT.BESE.UCW:SELF . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)
       (:LET IT.BESE.UCW::L . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)))
---   Locals:
IT.BESE.ARNESI::ENV ==> ((:LET IT.BESE.UCW::SERVER-REPL .
#<IT.BESE.UCW::ADMIN-REPL {9F222A1}>) (:LET IT.BESE.UCW::CONTROL-PANEL
. #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>) (:BLOCK
IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K) (:LET
IT.BESE.UCW:SELF . #1=#<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>) (:LET
IT.BESE.UCW::L . #1#))
IT.BESE.ARNESI::EVALUATED-ARGUMENTS ==>
((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
#:G1288 ==> :<NOT-AVAILABLE>
#:G1292 ==> :<NOT-AVAILABLE>
#:G1296 ==> :<NOT-AVAILABLE>
#:G1304 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::HANDLER ==> #<CLOSURE (LAMBDA
(IT.BESE.ARNESI::ARGUMENTS)) {9F2785D}>
IT.BESE.ARNESI::REMAINING-ARGUMENTS ==> NIL
---   Source:
(:LOCATION (:FILE
\\\"/home/conrad/website/ucw/arnesi_dev/src/cps.lisp\\\") (:POSITION
12078) (:SNIPPET \\\"(funcall handler (reverse evaluated-arguments))))

;;;; FUNCTION and LAMBDA application

(defmethod evaluate/cps ((lambda lambda-application-form) env k)
  (evaluate/cps-funcall (cons (operator lambda) (arguments lambda)) env k))

;;;; Constants

(defmetho\\\"))
--- FRAME 14
((LAMBDA ()))
---   Locals:
IT.BESE.ARNESI::ENV ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::EVALUATED-ARGUMENTS ==> :<NOT-AVAILABLE>
#:G1277 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::HANDLER ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::REMAINING-ARGUMENTS ==> :<NOT-AVAILABLE> ..)
   [Condition of type SIMPLE-ERROR]

Restarts:
  0: [ABORT-RESPONSE] Abort this response and answer another request
  1: [ABORT] Exit debugger, returning to top level.

Backtrace:
  0: ((SB-PCL::FAST-METHOD
SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT (T)) #<unused argument>
#<unused argument> #<FUNCTION (LAMBDA NIL) {9418345}>)
  1: (SWANK::CALL-WITH-BINDINGS ((*PRINT-PRETTY*) (*PRINT-LEVEL* . 4)
(*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*)
(*PRINT-PPRINT-DISPATCH* . #<SB-PRETTY:PPRINT-DISPATCH-TABLE
{9561021}>) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*)
(*PRINT-ARRAY* . T) ...) #<FUNCTION (LAMBDA NIL) {9418305}>)
  2: (SWANK::DEBUG-IN-EMACS #<SIMPLE-ERROR {A29DFB1}>)
  3: ((LAMBDA (CONDITION)) #<SIMPLE-ERROR {A29DFB1}>)
  4: ((LAMBDA (CONDITION)) #<SIMPLE-ERROR {A29DFB1}>)
  5: (SIGNAL #<SIMPLE-ERROR {A29DFB1}>)
  6: (ERROR "destructure-case failed: ~S")
  7: (SWANK::SEND-TO-SOCKET-IO (:EVALUATE-IN-EMACS "
 (save-excursion
   (loop
     with buffer-name = \"*UCW Backtrace <%d>*\"
     for id upfrom 0
     for backtrace-buffer = (get-buffer (format buffer-name id))
     while backtrace-buffer
     finally do (switch-to-buffer-other-window (format buffer-name id)))
   (insert \"--- UCW Backtrace
---    Condition: #<TYPE-ERROR {9F2C021}>
#<TYPE-ERROR {9F2C021}> is a TYPE-ERROR. Its slot values are (:DATUM
#1=#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}> :EXPECTED-TYPE #2=(OR
FUNCTION SYMBOL) SB-KERNEL::EXPECTED-TYPE #2# SB-KERNEL::DATUM #1#).
---    Date: 2005-08-08T12:18:29
---    Lisp: \\\"SBCL\\\" \\\"0.9.2\\\"
---    Server: #<IT.BESE.UCW:STANDARD-SERVER
IT.BESE.UCW:ARANEIDA-BACKEND 2 {A5217D1}>
#<IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:ARANEIDA-BACKEND 2
{A5217D1}> is an instance of class #<STANDARD-CLASS
IT.BESE.UCW:STANDARD-SERVER>.
The following slots have :INSTANCE allocation:
 APPLICATIONS         (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/
1> #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
 APPLICATIONS-LOCK    :NULL-LOCK
 STARTED              T
 BACKEND              #<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}>
---    Backend: #<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}>
#<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW:ARANEIDA-BACKEND>.
The following slots have :INSTANCE allocation:
 LISTENER          #<ARANEIDA:SERVE-EVENT-HTTP-LISTENER {A8A7139}>
 DEFAULT-URL       #<ARANEIDA:HTTP-URL \\\"http://127.0.0.1:8080/\\\" {A8838D9}>
 LISTENER-CLASS    ARANEIDA:SERVE-EVENT-HTTP-LISTENER
---    Application: #<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1> is an instance of
class #<STANDARD-CLASS IT.BESE.UCW:STANDARD-APPLICATION>.
The following slots have :INSTANCE allocation:
 URL-PREFIX       \\\"/ucw/admin/\\\"
 SESSION-TYPE     IT.BESE.UCW::STANDARD-SESSION
 TAL-GENERATOR    #<IT.BESE.YACLML:FILE-SYSTEM-GENERATOR {A3F2DC1}>
 WWW-ROOTS       
(#P\\\"/home/conrad/website/ucw/ucw_dev/bin/../wwwroot/./ucw/examples/\\\")
 ENTRY-POINTS     #<HASH-TABLE :TEST EQUAL :COUNT 1 {9D4CBA1}>
 SESSION-TABLE    #<HASH-TABLE :TEST EQUAL :COUNT 1 {9D4CD79}>
 SERVER           #<IT.BESE.UCW:STANDARD-SERVER
IT.BESE.UCW:ARANEIDA-BACKEND 2 {A5217D1}>
---    Request: #<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}>
#<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW::ARANEIDA-REQUEST>.
The following slots have :INSTANCE allocation:
 REQUEST    #<ARANEIDA:REQUEST {9CCC129}>
---    Response: #<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}>
#<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW::ARANEIDA-RESPONSE>.
The following slots have :INSTANCE allocation:
 REQUEST           #<ARANEIDA:REQUEST {9CCC129}>
 HEADERS           ((\\\"Content-Type\\\" . \\\"text/html\\\")
(\\\"Status\\\" . \\\"200\\\"))
 CONTENT-STREAM    #<SB-IMPL::STRING-OUTPUT-STREAM {9CCD8B1}>
--- BACKTRACE
--- FRAME 4
(ERROR TYPE-ERROR)
---   Locals:
SB-DEBUG::ARG-0 ==> 5
SB-DEBUG::ARG-1 ==> TYPE-ERROR
---   Source:
(:ERROR \\\"The source-path (NIL) is not valid.\\\")
--- FRAME 5
(SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER
      #<unavailable argument>
      #.(SB-SYS:INT-SAP #XB7C3C31C)
      #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010 :TYPE (*
                                                              (STRUCT
                                                              
SB-VM::OS-CONTEXT-T-STRUCT))>
      (14 78))
---   Locals:
SB-DEBUG::ARG-0 ==> 4
SB-DEBUG::ARG-1 ==> :<NOT-AVAILABLE>
SB-DEBUG::ARG-2 ==> #.(SB-SYS:INT-SAP #XB7C3C31C)
SB-DEBUG::ARG-3 ==> #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010
:TYPE (* (STRUCT SB-VM::OS-CONTEXT-T-STRUCT))>
SB-DEBUG::ARG-4 ==> (14 78)
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/interr.lisp\\\")
(:POSITION 6404) (:SNIPPET \\\"(deferr object-not-type-error (object
type)
  (error (if (and (typep object 'instance)
		  (layout-invalid (%instance-layout object)))
	     'layout-invalid
	     'type-error)
	 :datum object
	 :expected-type type))

(deferr layout-invalid-error (object la\\\"))
--- FRAME 6
(SB-KERNEL:INTERNAL-ERROR
      #.(SB-SYS:INT-SAP #XB7C3C010)
      #<unavailable argument>)
---   Locals:
SB-DEBUG::ARG-0 ==> 2
SB-DEBUG::ARG-1 ==> #.(SB-SYS:INT-SAP #XB7C3C010)
SB-DEBUG::ARG-2 ==> :<NOT-AVAILABLE>
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/interr.lisp\\\")
(:POSITION 10988) (:SNIPPET \\\"(defun internal-error (context
continuable)
  (declare (type system-area-pointer context))
  (declare (ignore continuable))
  (/show0 \\\\\\\"entering INTERNAL-ERROR, CONTEXT=..\\\\\\\")
  (/hexstr context)
  (infinite-error-protect
   (/show0 \\\\\\\"about to bind ALIEN-CONTE\\\"))
--- FRAME 7
(\\\"foreign function: call_into_lisp\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: call_into_lisp\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 8
(\\\"foreign function: funcall2\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: funcall2\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 9
(\\\"foreign function: interrupt_internal_error\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value
  #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign function:
interrupt_internal_error\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 10
(\\\"foreign function: sigtrap_handler\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: sigtrap_handler\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 11
(MAPCAR
      #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
      (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
       #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
---   Locals:
SB-DEBUG::ARG-0 ==> 2
SB-DEBUG::ARG-1 ==> #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
SB-DEBUG::ARG-2 ==> (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/list.lisp\\\") (:POSITION
41554) (:SNIPPET \\\"(defun mapcar (function list &rest more-lists)
  #!+sb-doc
  \\\\\\\"Apply FUNCTION to successive elements of LIST. Return list of FUNCTION
   return values.\\\\\\\"
  (map1 function (cons list more-lists) :list t))

(defun mapcan (function list &rest more-lists)
  #!+s\\\"))
--- FRAME 12
((LAMBDA (IT.BESE.ARNESI::ARGUMENTS))
      (#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
       (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)))
---   Locals:
IT.BESE.ARNESI::ARGUMENTS ==> (#<IT.BESE.ARNESI::CPS-CLOSURE
{9F27B21}> (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
IT.BESE.ARNESI::FUNC ==> :<NOT-AVAILABLE>
#:G763 ==> :<NOT-AVAILABLE>
#:G778 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::K ==> :<NOT-AVAILABLE>
---   Source:
(:LOCATION (:FILE
\\\"/home/conrad/website/ucw/arnesi_dev/src/cps.lisp\\\") (:POSITION
6830) (:SNIPPET \\\"(apply (fdefinition (operator func)) arguments))))
      (arguments func) '()
      env))))

(defmethod evaluate/cps ((func local-application-form) env k)
  (evaluate/cps-apply (arguments func) (list (list (lookup env :flet
(operator func) :error-p t))) en\\\"))
--- FRAME 13
(IT.BESE.ARNESI::EVALUATE-ARGUMENTS-THEN-APPLY
      #<CLOSURE (LAMBDA #) {9F2785D}>
      NIL
      ((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
       #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
      ((:LET IT.BESE.UCW::SERVER-REPL . #<IT.BESE.UCW::ADMIN-REPL {9F222A1}>)
       (:LET IT.BESE.UCW::CONTROL-PANEL
             . #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>)
       (:BLOCK IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K)
       (:LET IT.BESE.UCW:SELF . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)
       (:LET IT.BESE.UCW::L . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)))
---   Locals:
IT.BESE.ARNESI::ENV ==> ((:LET IT.BESE.UCW::SERVER-REPL .
#<IT.BESE.UCW::ADMIN-REPL {9F222A1}>) (:LET IT.BESE.UCW::CONTROL-PANEL
. #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>) (:BLOCK
IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K) (:LET
IT.BESE.UCW:SELF . #1=#<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>) (:LET
IT.BESE.UCW::L . #1#))
IT.BESE.ARNESI::EVALUATED-ARGUMENTS ==>
((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
#:G1288 ==> :<NOT-AVAILABLE>
#:G1292 ==> :<NOT-AVAILABLE>
#:G1296 ==> :<NOT-AVAILABLE>
#:G1304 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::HANDLER ==> #<CLOSURE (LAMBDA
(IT.BESE.ARNESI::ARGUMENTS)) {9F2785D}>
IT.BESE.ARNESI::REMAINING-ARGUMENTS ==> NIL
---   Source:
(:LOCATION (:FILE
\\\"/home/conrad/website/ucw/arnesi_dev/src/cps.lisp\\\") (:POSITION
12078) (:SNIPPET \\\"(funcall handler (reverse evaluated-arguments))))

;;;; FUNCTION and LAMBDA application

(defmethod evaluate/cps ((lambda lambda-application-form) env k)
  (evaluate/cps-funcall (cons (operator lambda) (arguments lambda)) env k))

;;;; Constants

(defmetho\\\"))
--- FRAME 14
((LAMBDA ()))
---   Locals:
IT.BESE.ARNESI::ENV ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::EVALUATED-ARGUMENTS ==> :<NOT-AVAILABLE>
#:G1277 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::HANDLER ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::REMAINING-ARGUMENTS ==> :<NOT-AVAILABLE> ..)
  8: (SWANK::EVALUATE-IN-EMACS "
 (save-excursion
   (loop
     with buffer-name = \"*UCW Backtrace <%d>*\"
     for id upfrom 0
     for backtrace-buffer = (get-buffer (format buffer-name id))
     while backtrace-buffer
     finally do (switch-to-buffer-other-window (format buffer-name id)))
   (insert \"--- UCW Backtrace
---    Condition: #<TYPE-ERROR {9F2C021}>
#<TYPE-ERROR {9F2C021}> is a TYPE-ERROR. Its slot values are (:DATUM
#1=#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}> :EXPECTED-TYPE #2=(OR
FUNCTION SYMBOL) SB-KERNEL::EXPECTED-TYPE #2# SB-KERNEL::DATUM #1#).
---    Date: 2005-08-08T12:18:29
---    Lisp: \\\"SBCL\\\" \\\"0.9.2\\\"
---    Server: #<IT.BESE.UCW:STANDARD-SERVER
IT.BESE.UCW:ARANEIDA-BACKEND 2 {A5217D1}>
#<IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:ARANEIDA-BACKEND 2
{A5217D1}> is an instance of class #<STANDARD-CLASS
IT.BESE.UCW:STANDARD-SERVER>.
The following slots have :INSTANCE allocation:
 APPLICATIONS         (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/
1> #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
 APPLICATIONS-LOCK    :NULL-LOCK
 STARTED              T
 BACKEND              #<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}>
---    Backend: #<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}>
#<IT.BESE.UCW:ARANEIDA-BACKEND {A829C89}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW:ARANEIDA-BACKEND>.
The following slots have :INSTANCE allocation:
 LISTENER          #<ARANEIDA:SERVE-EVENT-HTTP-LISTENER {A8A7139}>
 DEFAULT-URL       #<ARANEIDA:HTTP-URL \\\"http://127.0.0.1:8080/\\\" {A8838D9}>
 LISTENER-CLASS    ARANEIDA:SERVE-EVENT-HTTP-LISTENER
---    Application: #<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1> is an instance of
class #<STANDARD-CLASS IT.BESE.UCW:STANDARD-APPLICATION>.
The following slots have :INSTANCE allocation:
 URL-PREFIX       \\\"/ucw/admin/\\\"
 SESSION-TYPE     IT.BESE.UCW::STANDARD-SESSION
 TAL-GENERATOR    #<IT.BESE.YACLML:FILE-SYSTEM-GENERATOR {A3F2DC1}>
 WWW-ROOTS       
(#P\\\"/home/conrad/website/ucw/ucw_dev/bin/../wwwroot/./ucw/examples/\\\")
 ENTRY-POINTS     #<HASH-TABLE :TEST EQUAL :COUNT 1 {9D4CBA1}>
 SESSION-TABLE    #<HASH-TABLE :TEST EQUAL :COUNT 1 {9D4CD79}>
 SERVER           #<IT.BESE.UCW:STANDARD-SERVER
IT.BESE.UCW:ARANEIDA-BACKEND 2 {A5217D1}>
---    Request: #<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}>
#<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW::ARANEIDA-REQUEST>.
The following slots have :INSTANCE allocation:
 REQUEST    #<ARANEIDA:REQUEST {9CCC129}>
---    Response: #<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}>
#<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}> is an instance of class
#<STANDARD-CLASS IT.BESE.UCW::ARANEIDA-RESPONSE>.
The following slots have :INSTANCE allocation:
 REQUEST           #<ARANEIDA:REQUEST {9CCC129}>
 HEADERS           ((\\\"Content-Type\\\" . \\\"text/html\\\")
(\\\"Status\\\" . \\\"200\\\"))
 CONTENT-STREAM    #<SB-IMPL::STRING-OUTPUT-STREAM {9CCD8B1}>
--- BACKTRACE
--- FRAME 4
(ERROR TYPE-ERROR)
---   Locals:
SB-DEBUG::ARG-0 ==> 5
SB-DEBUG::ARG-1 ==> TYPE-ERROR
---   Source:
(:ERROR \\\"The source-path (NIL) is not valid.\\\")
--- FRAME 5
(SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER
      #<unavailable argument>
      #.(SB-SYS:INT-SAP #XB7C3C31C)
      #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010 :TYPE (*
                                                              (STRUCT
                                                              
SB-VM::OS-CONTEXT-T-STRUCT))>
      (14 78))
---   Locals:
SB-DEBUG::ARG-0 ==> 4
SB-DEBUG::ARG-1 ==> :<NOT-AVAILABLE>
SB-DEBUG::ARG-2 ==> #.(SB-SYS:INT-SAP #XB7C3C31C)
SB-DEBUG::ARG-3 ==> #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010
:TYPE (* (STRUCT SB-VM::OS-CONTEXT-T-STRUCT))>
SB-DEBUG::ARG-4 ==> (14 78)
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/interr.lisp\\\")
(:POSITION 6404) (:SNIPPET \\\"(deferr object-not-type-error (object
type)
  (error (if (and (typep object 'instance)
		  (layout-invalid (%instance-layout object)))
	     'layout-invalid
	     'type-error)
	 :datum object
	 :expected-type type))

(deferr layout-invalid-error (object la\\\"))
--- FRAME 6
(SB-KERNEL:INTERNAL-ERROR
      #.(SB-SYS:INT-SAP #XB7C3C010)
      #<unavailable argument>)
---   Locals:
SB-DEBUG::ARG-0 ==> 2
SB-DEBUG::ARG-1 ==> #.(SB-SYS:INT-SAP #XB7C3C010)
SB-DEBUG::ARG-2 ==> :<NOT-AVAILABLE>
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/interr.lisp\\\")
(:POSITION 10988) (:SNIPPET \\\"(defun internal-error (context
continuable)
  (declare (type system-area-pointer context))
  (declare (ignore continuable))
  (/show0 \\\\\\\"entering INTERNAL-ERROR, CONTEXT=..\\\\\\\")
  (/hexstr context)
  (infinite-error-protect
   (/show0 \\\\\\\"about to bind ALIEN-CONTE\\\"))
--- FRAME 7
(\\\"foreign function: call_into_lisp\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: call_into_lisp\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 8
(\\\"foreign function: funcall2\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: funcall2\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 9
(\\\"foreign function: interrupt_internal_error\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value
  #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign function:
interrupt_internal_error\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 10
(\\\"foreign function: sigtrap_handler\\\")
---   Locals:
---   Source:
(:ERROR \\\"The value #<SB-DI::BOGUS-DEBUG-FUN \\\\\\\"foreign
function: sigtrap_handler\\\\\\\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.\\\")
--- FRAME 11
(MAPCAR
      #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
      (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
       #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
---   Locals:
SB-DEBUG::ARG-0 ==> 2
SB-DEBUG::ARG-1 ==> #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
SB-DEBUG::ARG-2 ==> (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
---   Source:
(:LOCATION (:FILE \\\"/usr/lib/sbcl/src/code/list.lisp\\\") (:POSITION
41554) (:SNIPPET \\\"(defun mapcar (function list &rest more-lists)
  #!+sb-doc
  \\\\\\\"Apply FUNCTION to successive elements of LIST. Return list of FUNCTION
   return values.\\\\\\\"
  (map1 function (cons list more-lists) :list t))

(defun mapcan (function list &rest more-lists)
  #!+s\\\"))
--- FRAME 12
((LAMBDA (IT.BESE.ARNESI::ARGUMENTS))
      (#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
       (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)))
---   Locals:
IT.BESE.ARNESI::ARGUMENTS ==> (#<IT.BESE.ARNESI::CPS-CLOSURE
{9F27B21}> (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))
IT.BESE.ARNESI::FUNC ==> :<NOT-AVAILABLE>
#:G763 ==> :<NOT-AVAILABLE>
#:G778 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::K ==> :<NOT-AVAILABLE>
---   Source:
(:LOCATION (:FILE
\\\"/home/conrad/website/ucw/arnesi_dev/src/cps.lisp\\\") (:POSITION
6830) (:SNIPPET \\\"(apply (fdefinition (operator func)) arguments))))
      (arguments func) '()
      env))))

(defmethod evaluate/cps ((func local-application-form) env k)
  (evaluate/cps-apply (arguments func) (list (list (lookup env :flet
(operator func) :error-p t))) en\\\"))
--- FRAME 13
(IT.BESE.ARNESI::EVALUATE-ARGUMENTS-THEN-APPLY
      #<CLOSURE (LAMBDA #) {9F2785D}>
      NIL
      ((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
       #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
      ((:LET IT.BESE.UCW::SERVER-REPL . #<IT.BESE.UCW::ADMIN-REPL {9F222A1}>)
       (:LET IT.BESE.UCW::CONTROL-PANEL
             . #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>)
       (:BLOCK IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K)
       (:LET IT.BESE.UCW:SELF . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)
       (:LET IT.BESE.UCW::L . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)))
---   Locals:
IT.BESE.ARNESI::ENV ==> ((:LET IT.BESE.UCW::SERVER-REPL .
#<IT.BESE.UCW::ADMIN-REPL {9F222A1}>) (:LET IT.BESE.UCW::CONTROL-PANEL
. #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>) (:BLOCK
IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K) (:LET
IT.BESE.UCW:SELF . #1=#<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>) (:LET
IT.BESE.UCW::L . #1#))
IT.BESE.ARNESI::EVALUATED-ARGUMENTS ==>
((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
#<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
#:G1288 ==> :<NOT-AVAILABLE>
#:G1292 ==> :<NOT-AVAILABLE>
#:G1296 ==> :<NOT-AVAILABLE>
#:G1304 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::HANDLER ==> #<CLOSURE (LAMBDA
(IT.BESE.ARNESI::ARGUMENTS)) {9F2785D}>
IT.BESE.ARNESI::REMAINING-ARGUMENTS ==> NIL
---   Source:
(:LOCATION (:FILE
\\\"/home/conrad/website/ucw/arnesi_dev/src/cps.lisp\\\") (:POSITION
12078) (:SNIPPET \\\"(funcall handler (reverse evaluated-arguments))))

;;;; FUNCTION and LAMBDA application

(defmethod evaluate/cps ((lambda lambda-application-form) env k)
  (evaluate/cps-funcall (cons (operator lambda) (arguments lambda)) env k))

;;;; Constants

(defmetho\\\"))
--- FRAME 14
((LAMBDA ()))
---   Locals:
IT.BESE.ARNESI::ENV ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::EVALUATED-ARGUMENTS ==> :<NOT-AVAILABLE>
#:G1277 ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::HANDLER ==> :<NOT-AVAILABLE>
IT.BESE.ARNESI::REMAINING-ARGUMENTS ==> :<NOT-AVAILABLE> ..)
  9: (IT.BESE.UCW::SEND-BACKTRACE-TO-EMACS
#<IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:ARANEIDA-BACKEND 2
{A5217D1}> #<TYPE-ERROR {9F2C021}> (#S(IT.BESE.UCW::BACKTRACE-FRAME
:INDEX 4 :DESCRIPTION "(ERROR TYPE-ERROR)" :LOCALS (# #)
:SOURCE-LOCATION (:ERROR "The source-path (NIL) is not valid."))
#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 5 :DESCRIPTION
"(SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER
      #<unavailable argument>
      #.(SB-SYS:INT-SAP #XB7C3C31C)
      #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XB7C3C010 :TYPE (*
                                                              (STRUCT
                                                              
SB-VM::OS-CONTEXT-T-STRUCT))>
      (14 78))" :LOCALS (# # # # #) :SOURCE-LOCATION (:LOCATION # #
#)) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 6 :DESCRIPTION
"(SB-KERNEL:INTERNAL-ERROR
      #.(SB-SYS:INT-SAP #XB7C3C010)
      #<unavailable argument>)" :LOCALS (# # #) :SOURCE-LOCATION
(:LOCATION # # #)) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 7
:DESCRIPTION "(\"foreign function: call_into_lisp\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign
function: call_into_lisp\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
8 :DESCRIPTION "(\"foreign function: funcall2\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign
function: funcall2\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
9 :DESCRIPTION "(\"foreign function: interrupt_internal_error\")"
:LOCALS NIL :SOURCE-LOCATION (:ERROR "The value
  #<SB-DI::BOGUS-DEBUG-FUN \"foreign function: interrupt_internal_error\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
10 :DESCRIPTION "(\"foreign function: sigtrap_handler\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign
function: sigtrap_handler\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX
11 :DESCRIPTION "(MAPCAR
      #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
      (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
       #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>))"
:LOCALS (# # #) :SOURCE-LOCATION (:LOCATION # # #))
#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 12 :DESCRIPTION "((LAMBDA
(IT.BESE.ARNESI::ARGUMENTS))
      (#<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>
       (#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)))"
:LOCALS (# # # # #) :SOURCE-LOCATION (:LOCATION # # #))
#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 13 :DESCRIPTION
"(IT.BESE.ARNESI::EVALUATE-ARGUMENTS-THEN-APPLY
      #<CLOSURE (LAMBDA #) {9F2785D}>
      NIL
      ((#<IT.BESE.UCW:STANDARD-APPLICATION /ucw/admin/ 1>
        #<IT.BESE.UCW:COOKIE-SESSION-APPLICATION /ucw/examples/ 0>)
       #<IT.BESE.ARNESI::CPS-CLOSURE {9F27B21}>)
      ((:LET IT.BESE.UCW::SERVER-REPL . #<IT.BESE.UCW::ADMIN-REPL {9F222A1}>)
       (:LET IT.BESE.UCW::CONTROL-PANEL
             . #<IT.BESE.UCW::ADMIN-CONTROL-PANEL {9F20491}>)
       (:BLOCK IT.BESE.UCW:LOGIN-SUCCESSFUL IT.BESE.ARNESI::TOPLEVEL-K)
       (:LET IT.BESE.UCW:SELF . #<IT.BESE.UCW::ADMIN-LOGIN {91C6791}>)
       (:LET IT.BESE.UCW::L . #<IT.BESE.UCW::ADMIN-LOGIN
{91C6791}>)))" :LOCALS (# # # # # # # #) :SOURCE-LOCATION (:LOCATION #
# #)) ...))
 10: ((LAMBDA NIL))
 11: ((LAMBDA NIL))
 12: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-REQUEST
(IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:REQUEST
IT.BESE.UCW:RESPONSE)) #<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW:STANDARD-SERVER IT.BESE.UCW:ARANEIDA-BACKEND 2
{A5217D1}> #<IT.BESE.UCW::ARANEIDA-REQUEST {9CCD831}>
#<IT.BESE.UCW::ARANEIDA-RESPONSE {9CCD851}>)
 13: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST-RESPONSE
(IT.BESE.UCW::UCW-HANDLER T T)) #<unavailable argument> #<unavailable
argument> #<IT.BESE.UCW::UCW-HANDLER {9F7F7B1}> #<unavailable
argument> #<ARANEIDA:REQUEST {9CCC129}>)
 14: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST (ARANEIDA:HANDLER
T)) #<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW::UCW-HANDLER {9F7F7B1}> #<ARANEIDA:REQUEST {9CCC129}>)
 15: ((SB-PCL::FAST-METHOD ARANEIDA:HANDLE-REQUEST (ARANEIDA:HANDLER
T)) #<unavailable argument> #<unavailable argument>
#<ARANEIDA:DISPATCHING-HANDLER {A91B1B9}> #<ARANEIDA:REQUEST
{9CCC129}>)
 16: ((LABELS ARANEIDA::DO-IT) #<ARANEIDA:SERVE-EVENT-HTTP-LISTENER
{A8A7139}> #<SB-SYS:FD-STREAM for "a constant string" {9CC8EE9}>)
 17: (ARANEIDA::SERVE-EVENT-HTTP-LISTENER-ACCEPT-ONE-REQUEST
#<ARANEIDA:SERVE-EVENT-HTTP-LISTENER {A8A7139}>)
 18: (SB-IMPL::SUB-SERVE-EVENT NIL 0)
 19: (SB-SYS:WAIT-UNTIL-FD-USABLE 0 :INPUT NIL)
 20: (SB-IMPL::REFILL-BUFFER/FD #<SB-SYS:FD-STREAM for "standard
input" {90014F9}>)
 21: (SB-IMPL::INPUT-CHAR/ASCII #<SB-SYS:FD-STREAM for "standard
input" {90014F9}> NIL #:EOF-OBJECT)
 22: (READ-CHAR #<SB-SYS:FD-STREAM for "standard input" {90014F9}> NIL
#:EOF-OBJECT #<unused argument>)
 23: (READ-CHAR #<SYNONYM-STREAM :SYMBOL SB-SYS:*STDIN* {5012E31}> NIL
#:EOF-OBJECT #<unused argument>)
 24: (READ-CHAR #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL #:EOF-OBJECT #<unused
argument>)
 25: (READ-PRESERVING-WHITESPACE #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL (NIL) T)
 26: (READ-PRESERVING-WHITESPACE #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> NIL (NIL) NIL)
 27: (READ #<SYNONYM-STREAM :SYMBOL SWANK::*CURRENT-STANDARD-INPUT*
{92B4819}> NIL (NIL) NIL)
 28: (SB-IMPL::REPL-READ-FORM-FUN #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-INPUT* {92B4819}> #<SYNONYM-STREAM :SYMBOL
SWANK::*CURRENT-STANDARD-OUTPUT* {92B3371}>)
 29: (SB-IMPL::REPL-FUN NIL)
 30: ((LAMBDA NIL))
 31: ((LAMBDA NIL))
 32: (SB-IMPL::%WITH-REBOUND-IO-SYNTAX #<CLOSURE (LAMBDA NIL) {903A56D}>)
 33: (SB-IMPL::TOPLEVEL-REPL NIL)
 34: (SB-IMPL::TOPLEVEL-INIT)
 35: ((FLET SB-IMPL::RESTART-LISP))


From whalliburton at gmail.com  Tue Aug  9 02:58:06 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Mon, 8 Aug 2005 21:58:06 -0500
Subject: [Bese-devel] Re: admin page currently broken
In-Reply-To: <4e7bd29e05080809211e09e4b0@mail.gmail.com>
References: <4e7bd29e05080809211e09e4b0@mail.gmail.com>
Message-ID: <4e7bd29e0508081958496c64ca@mail.gmail.com>

This can be replicated with

(setf *form* (walk-form '(mapcar (lambda (a) (+ a 1)) '(1 2 3))))
(drive-cps (evaluate/cps *form* nil *toplevel-k*))

Looking at (defmethod evaluate/cps ((func free-application-form) env
k) it looks like a special case needs to be added for mapcar?

Is there a good way to look at the execution of evalutate/cps without tracing?

Will


From whalliburton at gmail.com  Tue Aug  9 04:43:54 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Mon, 8 Aug 2005 23:43:54 -0500
Subject: [Bese-devel] Backtracking bug still present in my runs.
In-Reply-To: <m2vf2gdiab.fsf@soma.local>
References: <4e7bd29e05080718323c851906@mail.gmail.com>
	<m2vf2gdiab.fsf@soma.local>
Message-ID: <4e7bd29e05080821435d6c4f47@mail.gmail.com>

The sample code I submitted also works for me, if I place it into a
containter. The following code should show the bug. Is this invalid
use of render-on? Thanks for the explanation of backtracking.

Will


 (defcomponent calltest-window (simple-window-component)
  ((calltest :component calltest :accessor calltest-window.calltest))
  (:default-initargs :title "calltest")
  (:documentation ""))

(defentry-point "calltest.ucw" (:application *wch-application*) ()
  (call 'calltest-window))

(defmethod render-on ((res response) (app calltest-window))
  (render-on res (calltest-window.calltest app)))


(defcomponent calltest-2 (widget-component)
  ())
   
(defmethod render-on ((res response) (calltest-2 calltest-2))
  (<:as-html "At 2"))

(defcomponent calltest (widget-component)
  ((val :backtrack t :initform 0 :initarg :val :accessor calltest.val)))
   
(defaction calltest-increment ((calltest calltest))
  (incf (calltest.val calltest))
  (if (= (calltest.val calltest) 2) (call 'calltest-2)))

(defmethod render-on ((res response) (calltest calltest))
  (<ucw:a :action (calltest-increment calltest) (<:as-html
(calltest.val calltest))))


From frido at q-software-solutions.de  Tue Aug  9 06:33:46 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 09 Aug 2005 08:33:46 +0200
Subject: [Bese-devel] Errors int the backend araneida.lisp file
Message-ID: <878xzbippx.fsf@flarge.here>

Just a few examples:
(defmethod register-url-handler ((backend araneida-backend) url handler)
  (ucw.backend.dribble "Registering handler ~S at url ~S." handler url backend)
  (araneida:install-handler (araneida:http-listener-handler (listener backend))
			    (make-instance 'ucw-handler :handler handler)
			    (araneida:urlstring (araneida:merge-url (default-url backend) url))
			    nil))

(defmethod unregister-url-handler ((backend araneida-backend) url)
  (ucw.backend.dribble "Unregistreing handler for url ~S." url backend)
  (araneida:uninstall-handler (listener backend)
			      (araneida:merge-url (default-url backend) url)
			      nil))
Well you see there are too many arguments for the format string. So
I'd argue this is wrong.

Regards
Friedrich


From frido at q-software-solutions.de  Tue Aug  9 06:40:04 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 09 Aug 2005 08:40:04 +0200
Subject: [Bese-devel] Bug?
Message-ID: <874q9zipff.fsf@flarge.here>

Well while trying the new ucw stuff I got this message:




The value (SETF COMPONENT.PLACE) is not of type SYMBOL.
   [Condition of type TYPE-ERROR]

Restarts:
  0: [SHOW-BACKTRACE] Send the client a backtrace page.
  1: [RETRY] Clearout the response and retry calling the action.
  2: [SERVER-ERROR] Send the client an internal server error page.
  3: [ABORT-RESPONSE] Abort this response and answer another request
  4: [TERMINATE-THREAD] Terminate this thread (#<THREAD {1000BB08B1}>)

Backtrace:
  0: ((SB-PCL::FAST-METHOD SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT (T)) #<unused argument> #<unused argument> #<FUNCTION (LAMBDA NIL) {1000470029}>)
  1: (SWANK::CALL-WITH-BINDINGS ((*PRINT-PRETTY*) (*PRINT-LEVEL* . 4) (*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*) (*PRINT-PPRINT-DISPATCH* . #<SB-PRETTY:PPRINT-DISPATCH-TABLE {1002D921D1}>) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*) (*PRINT-ARRAY* . T) ...) #<FUNCTION (LAMBDA NIL) {100046FFC9}>)
  2: (SWANK::DEBUG-IN-EMACS #<TYPE-ERROR {1001DE2C71}>)
  3: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-ACTION-ERROR (ERROR)) #<unavailable argument> #<unavailable argument> #<TYPE-ERROR {1001DE2C71}>)


Has anyone used a multithreaded SBCL with UCW?

Regards
Friedrich


From mb at bese.it  Tue Aug  9 06:46:40 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 09 Aug 2005 08:46:40 +0200
Subject: [Bese-devel] Backtracking bug still present in my runs.
In-Reply-To: <4e7bd29e05080821435d6c4f47@mail.gmail.com> (William
	Halliburton's message of "Mon, 8 Aug 2005 23:43:54 -0500")
References: <4e7bd29e05080718323c851906@mail.gmail.com>
	<m2vf2gdiab.fsf@soma.local>
	<4e7bd29e05080821435d6c4f47@mail.gmail.com>
Message-ID: <m27jev4nfz.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> The sample code I submitted also works for me, if I place it into a
> containter. The following code should show the bug. Is this invalid
> use of render-on? Thanks for the explanation of backtracking.

oddly we weren't setting up component slots for backtracking. i had
always thought the default was to backtrack these slots but, looking
through some of my code, i do always add the :backtrack t
option. that's a pretty stupid default value :(

fixed.

-- 
-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 Aug  9 06:50:52 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 09 Aug 2005 08:50:52 +0200
Subject: [Bese-devel] Bug?
In-Reply-To: <874q9zipff.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 09 Aug 2005 08:40:04 +0200")
References: <874q9zipff.fsf@flarge.here>
Message-ID: <m21x534n8z.fsf@soma.local>

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

> Well while trying the new ucw stuff I got this message:

try getting the latest patches from arnesi_dev and ucw_dev:

$ cd /path/to/ucw/
$ darcs pull http://common-lisp.net/project/ucw/repos/ucw_dev
$ cd /path/to/arnesi/
$ darcs pull http://common-lisp.net/project/bese/repos/arnesi_dev

(there has been some movement of the repos lately as i get used to
darcs. those are the official, as-permament-as-can-be, development
repos)

> Has anyone used a multithreaded SBCL with UCW?

not recently as far as i know.

-- 
-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 Aug  9 06:51:41 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 09 Aug 2005 08:51:41 +0200
Subject: [Bese-devel] Bug?
In-Reply-To: <m21x534n8z.fsf@soma.local> (Marco Baringer's message of "Tue,
	09 Aug 2005 08:50:52 +0200")
References: <874q9zipff.fsf@flarge.here> <m21x534n8z.fsf@soma.local>
Message-ID: <m2wtmv38n6.fsf@soma.local>

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

> Friedrich Dominicus <frido at q-software-solutions.de> writes:
>
>> Well while trying the new ucw stuff I got this message:
>
> try getting the latest patches from arnesi_dev and ucw_dev:

and, since those changes affect a bunch of macros, remeber to
recompile all the fasls.

-- 
-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 Aug  9 07:03:31 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 09 Aug 2005 09:03:31 +0200
Subject: [Bese-devel] Re: admin page currently broken
References: <4e7bd29e05080809211e09e4b0@mail.gmail.com>
	<4e7bd29e0508081958496c64ca@mail.gmail.com>
Message-ID: <m2ll3b383g.fsf@soma.local>

William Halliburton <whalliburton at gmail.com> writes:

> This can be replicated with
>
> (setf *form* (walk-form '(mapcar (lambda (a) (+ a 1)) '(1 2 3))))
> (drive-cps (evaluate/cps *form* nil *toplevel-k*))
>
> Looking at (defmethod evaluate/cps ((func free-application-form) env
> k) it looks like a special case needs to be added for mapcar?

we have two options:

1) special case all standard high-order functions

2) use sacla to reimplement them. the plan has always been to use
sacla (a well commented, tested, reimplementation of common-lisp in
common-lisp) for this. try this:

;; define a /cc version of mapcar (can't use defun/cc directly but that's the basic idea)
(setf (get 'mapcar 'defun/cc)
      (make-instance
       'cps-closure
       :code
       (walk-form '(lambda (function list)
                    "Apply FUNCTION to successive elements of lists, return list of results."
                    (do* ((lists (cons list nil))
                          (len (length lists))
                          (args (make-list len) (make-list len))
                          (result (list nil))
                          (splice result))
                         ((do ((l lists (cdr l)) (a args (cdr a)))
                              ((or (null l) (endp (car l))) l)
                            (rplaca a (caar l))
                            (rplaca l (cdar l)))
                          (cdr result))
                      (setq splice
                            (cdr (rplacd
                                  splice
                                  (list (apply function args)))))))
                  nil
                  nil)
       :env
       nil))

(with-call/cc
  (mapcar (lambda (a)
            (+ a 1))
          (list 1 2 3)))
==> (2 3 4)

> Is there a good way to look at the execution of evalutate/cps without tracing?

atm no. however i will be needing to add in a wrapper around each step
of the evaluation of forms, this will make a great point for exactly
that kind of debugging info (and basically step throguh your code for
you).

-- 
-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 Aug  9 12:26:28 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 09 Aug 2005 14:26:28 +0200
Subject: [Bese-devel] What has changed?
Message-ID: <871x53gutn.fsf@flarge.here>

Well I checked the sources with darcs changes, but nowhere is a
comment telling me what goind on with the metaclass option. This code 
;;; Just for testing
(defcomponent ir-test (simple-window-component form-component)
  ((num :accessor num 
        :component (integer-range-field :min-value 0 :max-value 20)))
  (:metaclass standard-component-class))

Has worked, and now I got this error message:



Execution of a form compiled with errors.
Form:
  (DEFCLASS IR-TEST (SIMPLE-WINDOW-COMPONENT FORM-COMPONENT) ((NUM ACCESSOR NUM COMPONENT (INTEGER-RANGE-FIELD MIN-VALUE 0 MAX-VALUE 20))) (METACLASS (METACLASS STANDARD-COMPONENT-CLASS)))
Compile-time-error:
  (in macroexpansion of (DEFCLASS IR-TEST # ...))(hint: For more precise location, try *BREAK-ON-SIGNALS*.)The value of the :metaclass option ((:METACLASS STANDARD-COMPONENT-CLASS)) is not a legal class name.
   [Condition of type SB-INT:COMPILED-PROGRAM-ERROR]

Restarts:
  0: [ABORT] Abort SLIME compilation.
  1: [ABORT] Return to sldb level 1.
  2: [ABORT] Abort SLIME compilation.


I recompiled the whole files and switched back to a single threaded
SBCL to see whether that has broken something or what else is going
on. So what has happened and what am I supposed to do to get that
stuff running again? Dropping the metaclass argument maybe, but if I
had to why didn't I find that mentioned anywhere.

Friedrich


From mb at bese.it  Tue Aug  9 13:04:39 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 09 Aug 2005 15:04:39 +0200
Subject: [Bese-devel] What has changed?
In-Reply-To: <871x53gutn.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 09 Aug 2005 14:26:28 +0200")
References: <871x53gutn.fsf@flarge.here>
Message-ID: <m2hddzmfbs.fsf@soma.local>

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

> Well I checked the sources with darcs changes, but nowhere is a
> comment telling me what goind on with the metaclass option. This code 
> ;;; Just for testing
> (defcomponent ir-test (simple-window-component form-component)
>   ((num :accessor num 
>         :component (integer-range-field :min-value 0 :max-value 20)))
>   (:metaclass standard-component-class))

my bad :(

i recently 'improved' the defcomponent macro giving a bunch of new
options. of course i forgot to actually test the handling of the
:metaclass option. 

grab the latest patch from the ucw_dev repo to fix 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 frido at q-software-solutions.de  Tue Aug  9 13:20:35 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 09 Aug 2005 15:20:35 +0200
Subject: [Bese-devel] general question
Message-ID: <87oe87fdr0.fsf@flarge.here>

is anyone using a recent multithreaded SBCL together with an araneida
backend?

Does server-event-listener  work for you with that SBCL?
Are you able to use  a threaded-http-listener?

I ask because I'm trying to get something like that running for 2 days
now, and the only system I can use at the moment is  a single threaded
SBCL with a server-event-listener. I would be very happy to learn how
one can use a multithreaded SBCL with UCW.

Regards
Friedrich


From rm at seid-online.de  Tue Aug  9 13:37:18 2005
From: rm at seid-online.de (R. Mattes)
Date: Tue, 09 Aug 2005 15:37:18 +0200
Subject: [Bese-devel] general question
In-Reply-To: <87oe87fdr0.fsf@flarge.here>
References: <87oe87fdr0.fsf@flarge.here>
Message-ID: <1123594639.28128.9.camel@hobbes.mh-freiburg.de>

On Tue, 2005-08-09 at 15:20 +0200, Friedrich Dominicus wrote:
> is anyone using a recent multithreaded SBCL together with an araneida
> backend?

As a caveat: my latest tests with Araneida/multithreaded SBCL have been
performed last year. I tried to use the mt Araneida server model for our
application but encounered reproduceable deadlocks/hangs in the
application -- there should be some SBCL mailing list entries about
these in Google's memory. I got the impression that the mt server wasn't
nearly as "maintained" as the serve-event one (and i never had enough
time to dive into the mt server code and fix it).

> Does server-event-listener  work for you with that SBCL?
> Are you able to use  a threaded-http-listener?

We use the serve-event-listener in one of our Applications (even without
a proxy in front of it). But that site has only moderate traffic 
rezensionen.zeit.de). 

> I ask because I'm trying to get something like that running for 2 days
> now, and the only system I can use at the moment is  a single threaded
> SBCL with a server-event-listener. I would be very happy to learn how
> one can use a multithreaded SBCL with UCW.

To me it looks like you'd have to fix two places: araneida's mt server
code as well as SBCL's socket handling in a multithreaded environment.

 HTH Ralf Mattes

> Regards
> Friedrich
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel



From frido at q-software-solutions.de  Tue Aug  9 14:45:37 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 09 Aug 2005 16:45:37 +0200
Subject: [Bese-devel] general question
In-Reply-To: <1123594639.28128.9.camel@hobbes.mh-freiburg.de> (R. Mattes's
	message of "Tue, 09 Aug 2005 15:37:18 +0200")
References: <87oe87fdr0.fsf@flarge.here>
	<1123594639.28128.9.camel@hobbes.mh-freiburg.de>
Message-ID: <877jevf9ta.fsf@flarge.here>

Thanks for taking the time helping me out with that. There are quite
few things I do not understand, e.g. why does serve-event-listener
work for single-threaded SBCL and why that fails on a multi-threaded
SBCL. I checked another backend PortableAserver and it seem that one
can run it with a multi-threaded SBCL, so maybe I should give that a
try. The point is I want to get my applicaton done and not dive into
that stuff yet. I hope this example will be good enough for others to
learn UCW-programming. For me a the moment it is
trial-error-error-error-error trial--- 


However could it be that using the thread stuff from acl-compat
wouldn't be a "good" idea? 

Regards
Friedrich


From frido at q-software-solutions.de  Tue Aug  9 15:39:28 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 09 Aug 2005 17:39:28 +0200
Subject: [Bese-devel] Changes II
Message-ID: <873bpjf7bj.fsf@flarge.here>

Well as written before I'm trying to get ucw working for me. I was a
bit successfull with that around a week or so ago. But now all the
things start tumbling down. So what has changes in the action
handling?

this code has worked for quite a time but now I got an error message:
             (<ucw:form 
               :action (user-details prods)
               (<:p
                (<:input :type "submit" :value "Yes"
                         :name "am-customer")
                (<:input :type "submit" :value "No"
                         :name ""))))))))


with this user-details action
(defaction user-details ((transaction trans-info))
  ;; (inspect (ucw::all-params (ucw::context.request *context*)))
  (let ((products (products-in-basket transaction))
        (sum (trans-price transaction)))
    (let ((trans-data (make-instance 'transaction))
          (p-ids (map 'list #'(lambda (item)
                                (list (product-id item)
                                      (num-copies item)
                                      (price item)))
                                products)))
      (db-update-tid-to-pid-mapping (trans-obj trans-data) p-ids sum)
  (let ((btn-text (ucw::get-parameter (ucw::context.request *context*) "am-customer")))
    (inspect btn-text)
    (if btn-text
      (call 'shop-login :message "You customer data please"
            :transaction trans-data)
      (call 'info-message :message "Not yet a customer."
            :transaction trans-data))))))


The error message is:
The value #<IT.BESE.ARNESI::CPS-CLOSURE {1000085D01}>
is not of type
  (OR FUNCTION SYMBOL).
   [Condition of type TYPE-ERROR]

Restarts:
  0: [SHOW-BACKTRACE] Send the client a backtrace page.
  1: [RETRY] Clearout the response and retry calling the action.
  2: [SERVER-ERROR] Send the client an internal server error page.
  3: [GENERATE-BACKTRACE-FOR-EMACS] Generate a bug report in Emacs.
  4: [ABORT-RESPONSE] Abort this response and answer another request
  5: [ABORT] Exit debugger, returning to top level.

Backtrace:
  0: ((SB-PCL::FAST-METHOD SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT (T)) #<unused argument> #<unused argument> #<FUNCTION (LAMBDA NIL) {1000DFE5D9}>)
  1: (SWANK::CALL-WITH-BINDINGS ((*PRINT-PRETTY*) (*PRINT-LEVEL* . 4) (*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*) (*PRINT-PPRINT-DISPATCH* . #<SB-PRETTY:PPRINT-DISPATCH-TABLE {1001A80C71}>) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*) (*PRINT-ARRAY* . T) ...) #<FUNCTION (LAMBDA NIL) {1000DFE579}>)
  2: (SWANK::DEBUG-IN-EMACS #<TYPE-ERROR {100008FC11}>)
  3: ((LAMBDA NIL))
  4: (SWANK::CALL-WITH-REDIRECTED-IO #<SWANK::CONNECTION {1001237481}> #<CLOSURE (LAMBDA NIL) {10004918C9}>)
  5: (SWANK::CALL-WITH-CONNECTION #<SWANK::CONNECTION {1001237481}> #<CLOSURE (LAMBDA NIL) {10004918A9}>)
  6: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-ACTION-ERROR (ERROR T)) #<unavailable argument> #<unavailable argument> #<TYPE-ERROR {100008FC11}> (#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 4 :DESCRIPTION "(ERROR TYPE-ERROR)" :LOCALS (# #) :SOURCE-LOCATION (:ERROR "Source filename not recorded for #<FUNCTION ERROR>")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 5 :DESCRIPTION "(SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER

so what do I have to change to get this action done?

Friedrich


From mb at bese.it  Tue Aug  9 16:00:42 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 09 Aug 2005 18:00:42 +0200
Subject: [Bese-devel] Changes II
In-Reply-To: <873bpjf7bj.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 09 Aug 2005 17:39:28 +0200")
References: <873bpjf7bj.fsf@flarge.here>
Message-ID: <m2k6ivje1h.fsf@soma.local>

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

> Well as written before I'm trying to get ucw working for me. I was a
> bit successfull with that around a week or so ago. But now all the
> things start tumbling down. So what has changes in the action
> handling?

pardon me for breaking your app :(

what has happened is that we've moved from a macro transformer to an
interpreter and the work still isn't quite finished. 

atm actions have the following, temporary, limitation: you can't pass
a function generated in the action (a form like #'(lambda ...)) to a
function which wasn't generated by defun/cc. basically this won't
work:

(mapcar (lambda ...) ...)

nor anything similar.

the reason is that the interpreter, when it sees a lambda form,
doesn't generate a function object but generates a cps-closure
object. this cps-closure object can contain call/cc (unlike the old
cps-transformer) but you need to have a definition of mapcar (or just
map in your case) which knows how to deal with cps-closure objects.

i'm currently (as we speak) cut 'n pasting code from sacla to
implement most of the common higher order common-lisp function
(mapXYZ, removeXYZ, etc.) in order to fix your code you will to
refactor the action so that the (lambda ...) form is in a second
function defined outside of the action and you only pass non-function
objects to that second function.

-- 
-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 Aug  9 16:06:21 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 09 Aug 2005 18:06:21 +0200
Subject: [Bese-devel] Changes II
In-Reply-To: <873bpjf7bj.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 09 Aug 2005 17:39:28 +0200")
References: <873bpjf7bj.fsf@flarge.here>
Message-ID: <87pssndria.fsf@flarge.here>

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

> Well as written before I'm trying to get ucw working for me. I was a
> bit successfull with that around a week or so ago. But now all the
> things start tumbling down. So what has changes in the action
> handling?
>
> this code has worked for quite a time but now I got an error message:
>              (<ucw:form 
>                :action (user-details prods)
>                (<:p
>                 (<:input :type "submit" :value "Yes"
>                          :name "am-customer")
>                 (<:input :type "submit" :value "No"
>                          :name ""))))))))
>
>
> with this user-details action
> (defaction user-details ((transaction trans-info))
>   ;; (inspect (ucw::all-params (ucw::context.request *context*)))
>   (let ((products (products-in-basket transaction))
>         (sum (trans-price transaction)))
>     (let ((trans-data (make-instance 'transaction))
>           (p-ids (map 'list #'(lambda (item)
>                                 (list (product-id item)
>                                       (num-copies item)
>                                       (price item)))
>                                 products)))

Ok I figured out the following the above line can not be used I have
to write an own function for that, now maybe someone knows why I have
to put it outside this action. 

So the following rewrite 

(defun gen-pids (products)
  (map 'list #'(lambda (item)
                 ;; (inspect item)
                 (list (product-id item)
                       (num-copies item)
                       (price item)))
       products))
  


(defaction user-details ((transaction trans-info))
  ;; (inspect (ucw::all-params (ucw::context.request *context*)))
  (let ((products (products-in-basket transaction))
        (sum (trans-price transaction)))

    (let ((trans-data (make-instance 'transaction))
          (p-ids (gen-pids products)))
;;       (inspect p-ids)
      (db-update-tid-to-pid-mapping (trans-obj trans-data) p-ids sum)
  (let ((btn-text (ucw::get-parameter (ucw::context.request *context*) "am-customer")))
    ;; (inspect btn-text)
    (if btn-text
      (call 'shop-login :message "You customer data please"
            :transaction trans-data)
      (call 'info-message :message "Not yet a customer."
            :transaction trans-data))))))


make the whole stuff work again. If that is on purpose could someone
give me an explanation please?

Regards
Friedrich


From smoof-ra at elder-gods.org  Tue Aug  9 18:45:25 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Tue, 9 Aug 2005 14:45:25 -0400
Subject: [Bese-devel] funcallable instances
Message-ID: <20050809184525.GA16399@melkor.elder-gods.org>

I don't know if this patch is actually a good idea to commit, because
it will only work on very recent sbcl, no cmucl or openmcl, but it's
kinda cool :-)

	--larry 
-------------- next part --------------

New patches:

[make closure/cc a funcallable instance (in sbcl)
smoof-ra at elder-gods.org**20050809182900
 
 funcalling a closure/cc will just call it with a toplevel continuation
 that way cc closures can be called totally transparently by non-cc code
 as long as the cc-closure doesn't call/cc.
 
] {
hunk ./src/cc-interpreter.lisp 144
-   (env :accessor env :initarg :env)))
+   (env :accessor env :initarg :env))
+  #+sbcl (:metaclass mopp:funcallable-standard-class))
+
+#+sbcl
+(defmethod initialize-instance :after ((fun closure/cc) &rest initargs)
+  (declare (ignore initargs))
+  (mopp:set-funcallable-instance-function 
+   fun 
+   #'(lambda (&rest args)
+       (drive-interpreter/cc 
+	(apply-lambda/cc fun
+			 args
+			 *toplevel-k*)))))
+
}

Context:

[Remove occurences of 'cps' in the api. We don't actually cps transforme anymore so this is misleading.
Marco Baringer <mb at bese.it>**20050809125933] 
[Remove 'cps' from test suite, replace it with 'call/cc' 
Marco Baringer <mb at bese.it>**20050809104818] 
[Rename cps.lisp to cc-interpreter.lisp
Marco Baringer <mb at bese.it>**20050809104737] 
[Fix typo in fold-strings' docstring
Marco Baringer <mb at bese.it>**20050809083006] 
[Trivial change to the name of the gensym generated by DOLIST*
Marco Baringer <mb at bese.it>**20050809061102] 
[Fix lexical-variables and lexical-functions on clisp
Marco Baringer <mb at bese.it>**20050807223244] 
[Use FDEFINITION instead of SYMBOL-FUNCTION to get a function from a function name.
Marco Baringer <mb at bese.it>**20050807222932] 
[Make sure we only pass symbols to functions like GET and MACRO-FUNCTION
Marco Baringer <mb at bese.it>**20050807222905] 
[Implement lexical-functions for CLISP
Marco Baringer <mb at bese.it>**20050807204738] 
[Implement lexical-variables and lexical-functions for NIL environments
Marco Baringer <mb at bese.it>**20050807204711] 
[Remove arnesi.el from ssytem def
Marco Baringer <mb at bese.it>**20050807204654] 
[Fix evaluation of #'(foo bar) in cps interpreter
Marco Baringer <mb at bese.it>**20050806182653] 
[Delete arnesi.el. SLIME is perfectly able to figure out the indententation by itself.
Marco Baringer <mb at bese.it>**20050807075500] 
[aparently global variables can be found in sbcl lexical environments
smoof-ra at elder-gods.org**20050804164859] 
[Implement environment-p and lexical-variables for CLISP
Marco Baringer <mb at bese.it>**20050804165821] 
[Make the lexenv stuff use generic-functions and methods
Marco Baringer <mb at bese.it>**20050804161857] 
[Fixup lexical-variables and lexical-functions for OpenMCL
Marco Baringer <mb at bese.it>**20050804152727
 
 This patch causes lexical-variables to no longer return ignored
 variables and symbol-macrolets. We've also implemented
 lexical-functions (though we do some hackery to convert functions
 names to something "normal" (ie SETF::|FOO::BAR| ==> (SETF FOO::BAR))
 
] 
[Typo in lexical-variables for sbcl (we were accessing lexenv-funs instead of lexenv-vars)
Marco Baringer <mb at bese.it>**20050804152051] 
[Change lexical-variables for sbcl so that it doesn't return ignored variables
Marco Baringer <mb at bese.it>**20050804150841] 
[Fix lexical-variables for cmucl to not return ignored variables
Marco Baringer <mb at bese.it>**20050804150256] 
[Typo in previous patch
Marco Baringer <mb at bese.it>**20050804150242] 
[Implement lexical-functions for cmucl
Marco Baringer <mb at bese.it>**20050804143350] 
[recognise flets from the lexical environment (on sbcl)
smoof-ra at elder-gods.org**20050803222732] 
[Rewrite multiple-value-setf so that my simple mind can understand it.
Marco Baringer <mb at bese.it>**20050803104652] 
[Added cps evaluation of THE forms
Marco Baringer <mb at bese.it>**20050803092059] 
[minor comment fixup
Marco Baringer <mb at bese.it>**20050803085322] 
[Moved defclass progv-form to keep the walker classes defined in alphabetical order 
Marco Baringer <mb at bese.it>**20050803085254] 
[Added walker class for THE forms
Marco Baringer <mb at bese.it>**20050803085210] 
[allow new special forms to be added to the walker by shadowing *walker-handlers*
smoof-ra at elder-gods.org**20050802165355] 
[Minor spacing fixs to the previous patch
Marco Baringer <mb at bese.it>**20050802152421] 
[progv
smoof-ra at elder-gods.org**20050802150342] 
[labels can have declarations inside the body
smoof-ra at elder-gods.org**20050801193433] 
[declares needs to be copied in the labels handler just like the other lambda-function-form slots
smoof-ra at elder-gods.org**20050801193107] 
[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] 
[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] 
[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
ddb2f89ea7fc94e072aa677fbd9138407abe37c2

From smoof-ra at elder-gods.org  Wed Aug 10 04:07:38 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 10 Aug 2005 00:07:38 -0400
Subject: [Bese-devel] method combination
Message-ID: <20050810040738.GA21777@melkor.elder-gods.org>

here's standard method combination for the interpreter.  Also adds &rest support.

       --larry 
-------------- next part --------------


(in-package :arnesi)

(fmakunbound 'supercool)

(defgeneric supercool (x)
  (:method-combination cc-standard))

(defmethod/cc supercool ((x integer))
  (if (next-method-p)
      (call-next-method x))
  (print (+ 2000 x)))

(defmethod/cc supercool :before (x)
  (print 'before))

(defmethod/cc supercool :after (x)
  (print 'after))

(with-call/cc 
  (supercool 7))

;BEFORE 
;2007 
;AFTER 

(defmethod/cc supercool (x)
  (print (+ 1000 x)))

(with-call/cc 
  (supercool 7))

;BEFORE 
;1007
;2007 
;AFTER 

(defmethod/cc supercool :around (x)
  (print 'pre)
  (call-next-method x)
  (print 'post))

(defmethod/cc supercool :around ((x integer))
  (print 'ipre)
  (call-next-method x)
  (print 'ipost))

(with-call/cc 
  (supercool 7))

;IPRE 
;PRE 
;BEFORE 
;1007 
;2007 
;AFTER 
;POST 
;IPOST

(defparameter *c* nil)

(defmethod/cc supercool :before (x)
  (print 'before)
  (let/cc k (setq *c* k)))

(with-call/cc 
  (supercool 7))

;IPRE 
;PRE 
;BEFORE 

(kall *c*)

;1007 
;2007 
;AFTER 
;POST 
;IPOST 


-------------- next part --------------

New patches:

[ method combination 
smoof-ra at elder-gods.org**20050810035825] {
hunk ./src/cc-interpreter.lisp 170
+
+    ((and (eql 'call-next-method (operator func))
+	  (second (multiple-value-list (lookup env :next-method t))))
+     (aif (lookup env :next-method t)
+	  (evaluate-arguments-then-apply
+	   (lambda (arguments)
+	     (apply-lambda/cc it arguments k))
+	   (arguments func) '() env)
+	  (error "no next method")))
+
+    ((and (eql 'next-method-p (operator func))
+	  (second (multiple-value-list (lookup env :next-method t))))
+     (kontinue k (lookup env :next-method t)))
hunk ./src/cc-interpreter.lisp 264
-        (optional-function-argument-form
+        (rest-function-argument-form
+	 (setf env (register env :let (name parameter) remaining-arguments))
+	 (setf remaining-arguments nil))
+	(optional-function-argument-form
hunk ./src/cc-interpreter.lisp 591
-(defmacro defmethod/cc (name arguments &body body)
-  `(progn
-     (setf (get ',name 'defmethod/cc) t)
-     (defmethod ,name ,arguments
-       (declare (ignorable ,@(extract-argument-names arguments :allow-specializers t)))
-       (make-instance 'closure/cc
-                      :code (walk-form '(lambda ,(extract-argument-names arguments :allow-specializers t
-                                                                         :keep-lambda-keywords t)
-                                         , at body)
-                                       nil nil)
-                      :env nil))))
+; for emacs:  (setf (get 'defmethod/cc 'common-lisp-indent-function) 'lisp-indent-defmethod)
+
+(defmacro defmethod/cc (name &rest args)
+  (let ((qlist (if (symbolp (car args))
+		   (prog1 
+		       (list (car args))
+		     (setf args (cdr args))))))
+    (destructuring-bind (arguments &body body) args
+      `(progn
+	 (setf (get ',name 'defmethod/cc) t)
+	 (defmethod ,name , at qlist ,arguments
+	   (declare (ignorable ,@(extract-argument-names arguments :allow-specializers t)))
+	   (make-instance 'closure/cc
+			  :code (walk-form '(lambda ,(extract-argument-names arguments :allow-specializers t
+									     :keep-lambda-keywords t)
+					     , at body)
+					   nil nil)
+			  :env nil))))))
+
hunk ./src/cc-interpreter.lisp 614
+
+
+(defun closure-with-nextmethod (closure next)
+  (make-instance 'closure/cc 
+		 :code (code closure)
+		 :env (register (env closure) :next-method t next)))
+
+(defun closure-with-befores (closure befores)
+  (make-instance 'closure/cc 
+		 :code (walk-form `(lambda (&rest args)
+				     ,@(loop 
+					  for before in befores
+					  collect `(apply ,before args))
+				     (apply ,closure args)))
+		 :env nil))
+
+(defun closure-with-afters (closure afters)
+  (make-instance 'closure/cc 
+		 :code (walk-form `(lambda (&rest args)
+				     (prog1 
+					 (apply ,closure args)
+				       ,@(loop 
+					    for after in afters
+					    collect `(apply ,after args)))))
+		 :env nil))
+
+(define-method-combination cc-standard
+    (&key (around-order :most-specific-first)
+          (before-order :most-specific-first)
+          (primary-order :most-specific-first)
+          (after-order :most-specific-last))
+  ((around (:around))
+   (before (:before))
+   (primary () :required t)
+   (after (:after)))
+  
+  (labels ((effective-order (methods order)
+             (ecase order
+               (:most-specific-first methods)
+               (:most-specific-last (reverse methods))))
+	   (primary-wrap (methods &optional nextmethod)
+	     (case (length methods)
+	       (1 `(closure-with-nextmethod 
+		    (call-method ,(first methods))
+		    ,nextmethod))
+	       (t `(closure-with-nextmethod 
+		    (call-method ,(first methods))
+		    ,(primary-wrap (cdr methods) nextmethod)))))
+	   (call-methods (methods)
+	     `(list ,@(loop 
+			 for m in methods
+			 collect `(call-method ,m)))))
+    (let* (;; reorder the methods based on the -order arguments
+           (around  (effective-order around around-order))
+           (before  (effective-order before before-order))
+           (primary (effective-order primary primary-order))
+           (after   (effective-order after after-order))
+           (form    (primary-wrap primary)))
+      (when after 
+	(setf form `(closure-with-afters ,form ,(call-methods after))))
+      (when before 
+	(setf form `(closure-with-befores ,form ,(call-methods before))))
+      (when around
+	(setf form (primary-wrap around form)))
+      form)))
+
+
+
}

Context:

[Added copyright notice to cc-interpreter.lisp
Marco Baringer <mb at bese.it>**20050809151738] 
[Remove occurences of 'cps' in the api. We don't actually cps transforme anymore so this is misleading.
Marco Baringer <mb at bese.it>**20050809125933] 
[Remove 'cps' from test suite, replace it with 'call/cc' 
Marco Baringer <mb at bese.it>**20050809104818] 
[Rename cps.lisp to cc-interpreter.lisp
Marco Baringer <mb at bese.it>**20050809104737] 
[Fix typo in fold-strings' docstring
Marco Baringer <mb at bese.it>**20050809083006] 
[Trivial change to the name of the gensym generated by DOLIST*
Marco Baringer <mb at bese.it>**20050809061102] 
[Fix lexical-variables and lexical-functions on clisp
Marco Baringer <mb at bese.it>**20050807223244] 
[Use FDEFINITION instead of SYMBOL-FUNCTION to get a function from a function name.
Marco Baringer <mb at bese.it>**20050807222932] 
[Make sure we only pass symbols to functions like GET and MACRO-FUNCTION
Marco Baringer <mb at bese.it>**20050807222905] 
[Implement lexical-functions for CLISP
Marco Baringer <mb at bese.it>**20050807204738] 
[Implement lexical-variables and lexical-functions for NIL environments
Marco Baringer <mb at bese.it>**20050807204711] 
[Remove arnesi.el from ssytem def
Marco Baringer <mb at bese.it>**20050807204654] 
[Fix evaluation of #'(foo bar) in cps interpreter
Marco Baringer <mb at bese.it>**20050806182653] 
[Delete arnesi.el. SLIME is perfectly able to figure out the indententation by itself.
Marco Baringer <mb at bese.it>**20050807075500] 
[aparently global variables can be found in sbcl lexical environments
smoof-ra at elder-gods.org**20050804164859] 
[Implement environment-p and lexical-variables for CLISP
Marco Baringer <mb at bese.it>**20050804165821] 
[Make the lexenv stuff use generic-functions and methods
Marco Baringer <mb at bese.it>**20050804161857] 
[Fixup lexical-variables and lexical-functions for OpenMCL
Marco Baringer <mb at bese.it>**20050804152727
 
 This patch causes lexical-variables to no longer return ignored
 variables and symbol-macrolets. We've also implemented
 lexical-functions (though we do some hackery to convert functions
 names to something "normal" (ie SETF::|FOO::BAR| ==> (SETF FOO::BAR))
 
] 
[Typo in lexical-variables for sbcl (we were accessing lexenv-funs instead of lexenv-vars)
Marco Baringer <mb at bese.it>**20050804152051] 
[Change lexical-variables for sbcl so that it doesn't return ignored variables
Marco Baringer <mb at bese.it>**20050804150841] 
[Fix lexical-variables for cmucl to not return ignored variables
Marco Baringer <mb at bese.it>**20050804150256] 
[Typo in previous patch
Marco Baringer <mb at bese.it>**20050804150242] 
[Implement lexical-functions for cmucl
Marco Baringer <mb at bese.it>**20050804143350] 
[recognise flets from the lexical environment (on sbcl)
smoof-ra at elder-gods.org**20050803222732] 
[Rewrite multiple-value-setf so that my simple mind can understand it.
Marco Baringer <mb at bese.it>**20050803104652] 
[Added cps evaluation of THE forms
Marco Baringer <mb at bese.it>**20050803092059] 
[minor comment fixup
Marco Baringer <mb at bese.it>**20050803085322] 
[Moved defclass progv-form to keep the walker classes defined in alphabetical order 
Marco Baringer <mb at bese.it>**20050803085254] 
[Added walker class for THE forms
Marco Baringer <mb at bese.it>**20050803085210] 
[allow new special forms to be added to the walker by shadowing *walker-handlers*
smoof-ra at elder-gods.org**20050802165355] 
[Minor spacing fixs to the previous patch
Marco Baringer <mb at bese.it>**20050802152421] 
[progv
smoof-ra at elder-gods.org**20050802150342] 
[labels can have declarations inside the body
smoof-ra at elder-gods.org**20050801193433] 
[declares needs to be copied in the labels handler just like the other lambda-function-form slots
smoof-ra at elder-gods.org**20050801193107] 
[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] 
[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] 
[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
1047890ba1af393a7dd7bc77565211c5d3a5ff11

From mb at bese.it  Wed Aug 10 05:32:03 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 10 Aug 2005 07:32:03 +0200
Subject: [Bese-devel] method combination
In-Reply-To: <20050810040738.GA21777@melkor.elder-gods.org> (Larry D'Anna's
	message of "Wed, 10 Aug 2005 00:07:38 -0400")
References: <20050810040738.GA21777@melkor.elder-gods.org>
Message-ID: <m23bpie4rw.fsf@soma.local>

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

> here's standard method combination for the interpreter.  Also adds
> &rest support.

oh my. that is really really cool.

-- 
-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  Wed Aug 10 07:38:44 2005
From: marco_simon at web.de (Marco Simon)
Date: Wed, 10 Aug 2005 09:38:44 +0200
Subject: [Bese-devel] Documentation
Message-ID: <42F9AF04.6070008@web.de>

Hi list,

once again some questions from a lisp and ucw-beginner:
Marco Baringer mentioned in his great "hello-world" video
that one could use the (ok x y) function for catching the field-
values from a html-form for further processing.

The ok-functions seem only to return one single field y out of
the component x. So I can process a single html-form-field. But
how can I process an html-form with several fields ? I just tried
something like (<ucw:form :action (list (ok myform field1) (ok myform 
field2)))
in order to get the fields' value returned as a list, but that didn't 
seem to work.

What other actions beside (ok x y) could I have used ? I guess it isn't the
only one for processing html-formular-data. And much more important:
How could I answer that question by myself ? At this point I am used to
pic some function-reference and have a look for the fitting function.

A futher problem, which perhaps is related to my developement-environment
(xemacs via ssh, cmucl, slime/swank, ucw) is that my emacs gets stuck as 
soon
as any error appears while my ucw-application is running. The the 
browser starts
loading the next "page" endlessly, my emacs/slime shows me the the 
debugger, but
I'm not able do anything futher in emacs because it doesn't react on any 
further input.
Any Ideas what the problem might be ? Compile-time-errors can be 
debugged without
any problems and without getting stuck.


Thanks for ideas, hints and anwers,
 best regards
 Marco Simon





From mb at bese.it  Wed Aug 10 08:40:59 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 10 Aug 2005 10:40:59 +0200
Subject: [Bese-devel] Documentation
In-Reply-To: <42F9AF04.6070008@web.de> (Marco Simon's message of "Wed,
	10 Aug 2005 09:38:44 +0200")
References: <42F9AF04.6070008@web.de>
Message-ID: <m23bpiqj50.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> Hi list,
>
> once again some questions from a lisp and ucw-beginner:
> Marco Baringer mentioned in his great "hello-world" video
> that one could use the (ok x y) function for catching the field-
> values from a html-form for further processing.

ok is an action which takes two arguments: 1) the component 2) the
value. what it does is to cause the call form which created the
component (the first argument to OK) to return the second argument
pased to ok.

(with the new cc-interpreter i am seriously considering redefining OK
like this:

(defaction ok ((c component) &rest values)
  (values-list values))

)

> The ok-functions seem only to return one single field y out of the
> component x. So I can process a single html-form-field. But how can
> I process an html-form with several fields ? I just tried something
> like (<ucw:form :action (list (ok myform field1) (ok myform
> field2))) in order to get the fields' value returned as a list, but
> that didn't seem to work.

nope. there are basically two (or three) options for collecting form
info.

1:

(let ((field1 ...) (field2 ...) (field3 ...))
  (<ucw:form :action (ok component (list field1 field2 field3))
    (<ucw:text :accessor field1)
    (<ucw:select :accessor field2
      (<ucw:option :value (something))
      (<ucw:option :value (something-else)))
    (<ucw:input :type "checkbox" :accessor field3)))

in this case you'd use the component like this:

(defaction whatever ((c component))
  (destructuring-bind (f1 f2 f3)
      (call 'my-form-component)
    (do-stuff f1 f2 ...)))

2:

(<ucw:form :action (ok component)
  (<ucw:text :accessor (field1 component))
  (<ucw:select :accessor (field2 component)
    (<ucw:option :value (something))
    (<ucw:option :value (something-else)))
  (<ucw:input :type "checkbox" :accessor (field3 component)))

and now you'd use the component like this:

(defaction whatever ((c component))
  (let ((form-data (call 'my-form-component)))
    (do-stuff (field1 form-data) (field2 form-data) ...)))

or 3:

(<ucw:form :action (ok component)
  (<:text :name "f1")
  (<:select :name "f2"
    (<:option :value "a string")
    (<:option :value "another string"))
  (<ucw:input :type "checkbox" :name "f3"))

and you'd use it like this:

(defaction whatever ((c component))
  (let* ((form-data (call 'my-form-component))
         (f1 (ucw::get-parameter "f1" (ucw::context.request *context*)))
          ...)
    (do-stuff f1 f2 ...)))


i've used both #1 and #2 depending on what felt easier at the
time. i've never used #3 but i have seen it "in the wild" and will
gladly accept a patch which makes it slightly easier to type. #3 has
the draw back that the values passed around must all be strings, but
has the advantage that you can use it in public urls (iow you don't
need an already existing session object).

> Wha other actions beside (ok x y) could I have used ? I guess it isn't the
> only one for processing html-formular-data. And much more important:

There are no 'form processing' actions. All actions are 'form
processing' actions. does this make any sense? an action is just a
function which takes a component as the first parameter and is defined
via defaction. you can write an action which takes as may other args
as you want and you can get the parameters to an action from just
about anywhere.

> How could I answer that question by myself ? At this point I am used to
> pic some function-reference and have a look for the fitting function.

basically anything defined via defaction.

with a very very recent ucw/arnesi you can inspect the variable
arnesi::*cc-functions* and find out all the functions defined via
defmethod/cc (the macro underlying defaction).

> A futher problem, which perhaps is related to my
> developement-environment (xemacs via ssh, cmucl, slime/swank, ucw)
> is that my emacs gets stuck as soon as any error appears while my
> ucw-application is running. The the browser starts loading the next
> "page" endlessly, my emacs/slime shows me the the debugger, but I'm
> not able do anything futher in emacs because it doesn't react on any
> further input.  Any Ideas what the problem might be ?
> Compile-time-errors can be debugged without any problems and without
> getting stuck.

this is really really bad :( this basically makes an instantaneous
edit/compile/debug cycle (the core of lisp development) impossible.

slime has a know bug where, when an sldb buffer first pops up, the
first key you hit (whatever it is) generates a <buffer is read-only>
error, but subsequent key presses should work.

first off a few questions: what version of cmucl? is the entire setup
running remotely or is xemacs local and everything else is remote?
what's in the *slime-events* buffer? have you run
mp::startup-idle-and-toplevel-loops? (i believe slime runs this
automatically on startup, but adding it to .cmuclrc can't hurt)

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  Wed Aug 10 14:16:10 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 10 Aug 2005 16:16:10 +0200
Subject: [Bese-devel] 
	&key, &allow-other-keys in with-call/cc [more breakage]
Message-ID: <m27jetri6t.fsf@soma.local>


hi,

  I just commited a patch to arnesi_dev which allows for &key in
  lambdas lists (and therefore in actions as well). this is an
  essential step on the way to integrating sacla. i've probably broken
  things along the 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 mb at bese.it  Wed Aug 10 14:22:10 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 10 Aug 2005 16:22:10 +0200
Subject: [Bese-devel] method combination
In-Reply-To: <m23bpie4rw.fsf@soma.local> (Marco Baringer's message of "Wed,
	10 Aug 2005 07:32:03 +0200")
References: <20050810040738.GA21777@melkor.elder-gods.org>
	<m23bpie4rw.fsf@soma.local>
Message-ID: <m2ll39q3cd.fsf@soma.local>

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

> Larry D'Anna <smoof-ra at elder-gods.org> writes:
>
>> here's standard method combination for the interpreter.  Also adds
>> &rest support.

  unless you're carefull you may not realize exactly for cool larry's
  patch is.

  let's pretend (i actually have this setup) that you have three or
  four task components which represent the central workflows of your
  app, and they all need to update userinfo after a certain amount of
  time (so they all subclass the my-generic-task component):

  (defaction start :before ((task my-generic-task))
    (when (user-data-too-old (call 'my-login))
      (call 'update-user-info)))

  come on, how fucking cool is that? instant "web aspect oriented
  programming". if you make the my-login-component use ajax (aka
  javascript) you get 'ajax aspect oriented programming' for more
  buzzwordness. of course, since this app runs on expensive servers
  and i wear a coat and tie when i meet the client it's actually
  "enterprise ajax aspect oriented programming objects".

p.s. - i've looked through (and even commited!) larry's patch but i
       haven't actually tested it yet...

p.p.s. - this is begging for a video :)

-- 
-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 whalliburton at gmail.com  Wed Aug 10 16:51:02 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Wed, 10 Aug 2005 11:51:02 -0500
Subject: [Bese-devel] Incorrect behavior with continuations.
Message-ID: <4e7bd29e05081009515b8d20f1@mail.gmail.com>

Using the latest arnesi. (Same behavior with arnesi darcs from Aug 7).

Embedding a call/cc withing a with-call/cc does not seem to complete
the execution.

(with-call/cc 
 (call/cc (lambda (cc) ()))
 3)

returns nill but should return 3

Will


From smoof-ra at elder-gods.org  Wed Aug 10 18:08:26 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 10 Aug 2005 14:08:26 -0400
Subject: [Bese-devel] Incorrect behavior with continuations.
In-Reply-To: <4e7bd29e05081009515b8d20f1@mail.gmail.com>
References: <4e7bd29e05081009515b8d20f1@mail.gmail.com>
Message-ID: <20050810180826.GA17829@melkor.elder-gods.org>

* William Halliburton (whalliburton at gmail.com) [050810 12:51]:
> Using the latest arnesi. (Same behavior with arnesi darcs from Aug 7).
> 
> Embedding a call/cc withing a with-call/cc does not seem to complete
> the execution.
> 
> (with-call/cc 
>  (call/cc (lambda (cc) ()))
>  3)
> 
> returns nill but should return 3


http://common-lisp.net/pipermail/bese-devel/2005-June/000546.html

	--larry 



From joseph at mammalia.net  Wed Aug 10 19:56:54 2005
From: joseph at mammalia.net (R. Joseph Wright)
Date: Wed, 10 Aug 2005 12:56:54 -0700
Subject: [Bese-devel] method combination
In-Reply-To: <20050810040738.GA21777@melkor.elder-gods.org>
References: <20050810040738.GA21777@melkor.elder-gods.org>
Message-ID: <24672315-29A1-4836-8A1A-11F0AAEB3C6A@mammalia.net>


On Aug 9, 2005, at 9:07 PM, Larry D'Anna wrote:

> here's standard method combination for the interpreter.  Also adds  
> &rest support.

Can you explain this part?

 >(defparameter *c* nil)
 >
 >defmethod/cc supercool :before (x)
 >  (print 'before)
 >  (let/cc k (setq *c* k)))
 >
 >(with-call/cc
 >  (supercool 7))
 >
 >;IPRE
 >;PRE
 >;BEFORE
 >
 >(kall *c*)
 >
 >;1007
 >;2007
 >;AFTER
 >;POST
 >;IPOST

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



From smoof-ra at elder-gods.org  Wed Aug 10 20:08:35 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Wed, 10 Aug 2005 16:08:35 -0400
Subject: [Bese-devel] method combination
In-Reply-To: <24672315-29A1-4836-8A1A-11F0AAEB3C6A@mammalia.net>
References: <20050810040738.GA21777@melkor.elder-gods.org>
	<24672315-29A1-4836-8A1A-11F0AAEB3C6A@mammalia.net>
Message-ID: <20050810200835.GA25692@melkor.elder-gods.org>

* R. Joseph Wright (joseph at mammalia.net) [050810 15:57]:
> 
> On Aug 9, 2005, at 9:07 PM, Larry D'Anna wrote:
> 
> >here's standard method combination for the interpreter.  Also adds  
> >&rest support.
> 
> Can you explain this part?
> 
> >(defparameter *c* nil)
> >
> >defmethod/cc supercool :before (x)
> >  (print 'before)
> >  (let/cc k (setq *c* k)))
> >
> >(with-call/cc
> >  (supercool 7))
> >
> >;IPRE
> >;PRE
> >;BEFORE
> >
> >(kall *c*)
> >
> >;1007
> >;2007
> >;AFTER
> >;POST
> >;IPOST

It's the same as the part before except this time we save the
continuation and quit in the :before method, and then resume
the continuation on a separate line.  The point of the patch is
to make sure that the logic that runs the various methods of a
generic function occurs in the cc-interpreter, so call/cc can be 
used in any of the methods.

     --larry


From whalliburton at gmail.com  Thu Aug 11 04:05:03 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Wed, 10 Aug 2005 23:05:03 -0500
Subject: [Bese-devel] Incorrect behavior with continuations.
In-Reply-To: <20050810180826.GA17829@melkor.elder-gods.org>
References: <4e7bd29e05081009515b8d20f1@mail.gmail.com>
	<20050810180826.GA17829@melkor.elder-gods.org>
Message-ID: <4e7bd29e050810210573e4214b@mail.gmail.com>

Thank you for the pointer. Setting *call/cc-returns* gives me the
behavior I wanted.

I did notice that your examples do not work with the current arnesi.
You have to currently call kall on the continuation and they also give
a different behavior.

(with-call/cc 
  (print (list 'foo
	       (let/cc k
		 (kall k 'bar) 
		 'baz))))

Prints (FOO BAR) and returns (FOO BAR) regardless of *call/cc-returns*.

Will


On 8/10/05, Larry D'Anna <smoof-ra at elder-gods.org> wrote:
> * William Halliburton (whalliburton at gmail.com) [050810 12:51]:
> > Using the latest arnesi. (Same behavior with arnesi darcs from Aug 7).
> >
> > Embedding a call/cc withing a with-call/cc does not seem to complete
> > the execution.
> >
> > (with-call/cc
> >  (call/cc (lambda (cc) ()))
> >  3)
> >
> > returns nill but should return 3
> 
> 
> http://common-lisp.net/pipermail/bese-devel/2005-June/000546.html
> 
>         --larry
> 
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel
>


From Alan-Shields at omrf.ouhsc.edu  Thu Aug 11 21:23:23 2005
From: Alan-Shields at omrf.ouhsc.edu (Alan Shields)
Date: Thu, 11 Aug 2005 16:23:23 -0500
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
Message-ID: <E1E3KWJ-0007RT-00@inquisitor.omrf.org>

Thu Aug 11 16:20:32 CDT 2005  Alan Shields <Alan-Shields at omrf.ouhsc.edu>
  * Add :fiveam to *features*
  I've found that I prefer to add testing code in my other code, but I dislike
  adding :fiveam to my package requirements list.
  
  This allows users of fiveam to delineate testing code with #+fiveam() - whereupon
  it will be totally ignored by those who do not have fiveam.
  
  Happiness for all and love ensue.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/x-darcs-patch
Size: 1186 bytes
Desc: A darcs patch for your repository!
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050811/a213d3d6/attachment.bin>

From mb at bese.it  Fri Aug 12 12:32:41 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 12 Aug 2005 14:32:41 +0200
Subject: [Bese-devel] httpd backend
Message-ID: <m2u0hvpc7q.fsf@soma.local>


i taught the httpd backend how to serve up files (so that the example
app in its default configuration gets stylesheets and javascript files).

this would be really cool, if it weren't for the fact that it doesn't
work :(. if i use curl or if i manually put the urls of stylesheet.css
into my browser the files come back normally. i've compared the data
the httpd backend sends to what apache sends (with mod_lisp everything
works fine) and it can't seem to figure out what the difference
is. anybody have any ideas?

p.s. - it works with safari but not with explorer, camino or firefox.

-- 
-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 Aug 12 14:34:08 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 12 Aug 2005 16:34:08 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> (Alan Shields's message
	of "Thu, 11 Aug 2005 16:23:23 -0500")
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org>
Message-ID: <m2ek8zns0v.fsf@soma.local>

Alan Shields <Alan-Shields at omrf.ouhsc.edu> writes:

> Thu Aug 11 16:20:32 CDT 2005  Alan Shields <Alan-Shields at omrf.ouhsc.edu>
>   * Add :fiveam to *features*
>   I've found that I prefer to add testing code in my other code, but I dislike
>   adding :fiveam to my package requirements list.

i am hesitant to modify *features*, it's not really supposed to be
used for this (afaict). you could do this:

(defun :fiveam ()
  (if (find-package :it.bese.fiveam)
      '(and)
      '(or)))

...

#+#.(fiveam)
(test
  ...)

if you want to avoid giving a keyword a function value you'll have to
fully qualify the function name:

#+#.(my-package:fiveam)

is this Good Enough(TM)? (since you want the code to run without
fiveam loaded you will need to define the fiveam function for every
package which uses fiveam (or any other test suite for that
matter)). if this becomes common i could distribute a 'boot strap'
file for fiveam which contains:

(in-package :common-lisp-user)

(defpackage :it.bese.fiveam
  (:nicknames :5am)
  (:export :?))

(defun 5am:? ()
  (if (find-package :it.bese.fiveam)
      '(and)
      '(or)))

and you'd litter your code with:

#.#+(5am:?)
(test ...)

you'd then include this file in all your projects. loading fiveam will
replace the package definition but until then the reader will be
happy. on second thought, i think this idea sucks.

-- 
-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 Aug 12 14:39:43 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 12 Aug 2005 16:39:43 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <m2ek8zns0v.fsf@soma.local> (Marco Baringer's message of "Fri,
	12 Aug 2005 16:34:08 +0200")
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> <m2ek8zns0v.fsf@soma.local>
Message-ID: <m27jernrrk.fsf@soma.local>

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

> (in-package :common-lisp-user)
>
> (defpackage :it.bese.fiveam

of course, this should :it.bese.fiveam.bootstrap or something if i
want it to actually work :)

-- 
-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 asf at boinkor.net  Fri Aug 12 15:23:14 2005
From: asf at boinkor.net (Andreas Fuchs)
Date: Fri, 12 Aug 2005 17:23:14 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <m2ek8zns0v.fsf@soma.local>
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> <m2ek8zns0v.fsf@soma.local>
Message-ID: <871x4z5gd9.wl%asf@boinkor.net>

Today, Marco Baringer <mb at bese.it> wrote:
> Alan Shields <Alan-Shields at omrf.ouhsc.edu> writes:
>
>> Thu Aug 11 16:20:32 CDT 2005 Alan Shields
>> <Alan-Shields at omrf.ouhsc.edu>
>> * Add :fiveam to *features*
>> I've found that I prefer to add testing code in my other code, but I dislike
>> adding :fiveam to my package requirements list.
>
> i am hesitant to modify *features*, it's not really supposed to be
> used for this (afaict). you could do this:
>
> (defun :fiveam ()
>   (if (find-package :it.bese.fiveam)
>       '(and)
>       '(or)))

Careful, there!

<antifuchs> suite:testp should return '(or) and '(and), of course (:
<Xof> n-uhuh
<Xof> '(:or) '(:and)
<antifuchs> ah, right.
<antifuchs> I was in the keyword package (:
<Xof> lots of people get that wrong, including clrfi authors

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


From mb at bese.it  Fri Aug 12 15:37:22 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 12 Aug 2005 17:37:22 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <871x4z5gd9.wl%asf@boinkor.net> (Andreas Fuchs's message of "Fri,
	12 Aug 2005 17:23:14 +0200")
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org>
	<m2ek8zns0v.fsf@soma.local> <871x4z5gd9.wl%asf@boinkor.net>
Message-ID: <m2irybmaj1.fsf@soma.local>

Andreas Fuchs <asf at boinkor.net> writes:

> Today, Marco Baringer <mb at bese.it> wrote:
>> Alan Shields <Alan-Shields at omrf.ouhsc.edu> writes:
>>
>>> Thu Aug 11 16:20:32 CDT 2005 Alan Shields
>>> <Alan-Shields at omrf.ouhsc.edu>
>>> * Add :fiveam to *features*
>>> I've found that I prefer to add testing code in my other code, but I dislike
>>> adding :fiveam to my package requirements list.
>>
>> i am hesitant to modify *features*, it's not really supposed to be
>> used for this (afaict). you could do this:
>>
>> (defun :fiveam ()
>>   (if (find-package :it.bese.fiveam)
>>       '(and)
>>       '(or)))
>
> Careful, there!

um, the function definition is not read in the keyword package. notice
that the function is named :FIVEAM so you can use it like this:

#+#.(fiveam) since the string "fiveam" _will_ be read in the keyword
package.

-- 
-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 asf at boinkor.net  Fri Aug 12 15:40:37 2005
From: asf at boinkor.net (Andreas Fuchs)
Date: Fri, 12 Aug 2005 17:40:37 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <m2irybmaj1.fsf@soma.local>
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> <m2ek8zns0v.fsf@soma.local>
	<871x4z5gd9.wl%asf@boinkor.net> <m2irybmaj1.fsf@soma.local>
Message-ID: <87zmrn40zu.wl%asf@boinkor.net>

Today, Marco Baringer <mb at bese.it> wrote:
> Andreas Fuchs <asf at boinkor.net> writes:
>> Today, Marco Baringer <mb at bese.it> wrote:
>>> (defun :fiveam ()
>>>   (if (find-package :it.bese.fiveam)
>>>       '(and)
>>>       '(or)))
>>
>> Careful, there!
>
> um, the function definition is not read in the keyword
> package.

Right, so symbols it returns should be in the keyword package. It
should be '(:or) and '(:and), not '(or) and '(and).

> notice that the function is named :FIVEAM so you can use it like
> this: #+#.(fiveam) since the string "fiveam" _will_ be read in the
> keyword package.

Correct.

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


From mb at bese.it  Fri Aug 12 15:55:01 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 12 Aug 2005 17:55:01 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <87zmrn40zu.wl%asf@boinkor.net> (Andreas Fuchs's message of "Fri,
	12 Aug 2005 17:40:37 +0200")
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org>
	<m2ek8zns0v.fsf@soma.local> <871x4z5gd9.wl%asf@boinkor.net>
	<m2irybmaj1.fsf@soma.local> <87zmrn40zu.wl%asf@boinkor.net>
Message-ID: <m2d5ojm9pm.fsf@soma.local>

Andreas Fuchs <asf at boinkor.net> writes:

> Today, Marco Baringer <mb at bese.it> wrote:
>> Andreas Fuchs <asf at boinkor.net> writes:
>>> Today, Marco Baringer <mb at bese.it> wrote:
>>>> (defun :fiveam ()
>>>>   (if (find-package :it.bese.fiveam)
>>>>       '(and)
>>>>       '(or)))
>>>
>>> Careful, there!
>>
>> um, the function definition is not read in the keyword
>> package.
>
> Right, so symbols it returns should be in the keyword package. It
> should be '(:or) and '(:and), not '(or) and '(and).

good catch. how can something so simple be so damn confusing?

-- 
-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 pjb at informatimago.com  Fri Aug 12 15:59:42 2005
From: pjb at informatimago.com (Pascal Bourguignon)
Date: Fri, 12 Aug 2005 17:59:42 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <m2irybmaj1.fsf@soma.local>
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> <m2ek8zns0v.fsf@soma.local>
	<871x4z5gd9.wl%asf@boinkor.net> <m2irybmaj1.fsf@soma.local>
Message-ID: <17148.51054.734431.64434@thalassa.informatimago.com>

Marco Baringer writes:
> Andreas Fuchs <asf at boinkor.net> writes:
> 
> > Today, Marco Baringer <mb at bese.it> wrote:
> >> Alan Shields <Alan-Shields at omrf.ouhsc.edu> writes:
> >>
> >>> Thu Aug 11 16:20:32 CDT 2005 Alan Shields
> >>> <Alan-Shields at omrf.ouhsc.edu>
> >>> * Add :fiveam to *features*
> >>> I've found that I prefer to add testing code in my other code, but I dislike
> >>> adding :fiveam to my package requirements list.
> >>
> >> i am hesitant to modify *features*, it's not really supposed to be
> >> used for this (afaict). you could do this:
> >>
> >> (defun :fiveam ()
> >>   (if (find-package :it.bese.fiveam)
> >>       '(and)
> >>       '(or)))
> >
> > Careful, there!
> 
> um, the function definition is not read in the keyword package. notice
> that the function is named :FIVEAM so you can use it like this:
> 
> #+#.(fiveam) since the string "fiveam" _will_ be read in the keyword
> package.

Why should it be read in the keyword package?

[39]> #.(zaphod-beeblebrox)


*** - EVAL: undefined function ZAPHOD-BEEBLEBROX
The following restarts are available:
USE-VALUE      :R1      You may input a value to be used instead of (FDEFINITION 'ZAPHOD-BEEBLEBROX).
RETRY          :R2      Retry
STORE-VALUE    :R3      You may input a new value for (FDEFINITION 'ZAPHOD-BEEBLEBROX).
ABORT          :R4      ABORT
Break 1 [40]> *package*
#<PACKAGE COMMON-LISP-USER>
Break 1 [40]> :a
[41]> (find-symbol "ZAPHOD-BEEBLEBROX" "KEYWORD")
NIL ;
NIL
[42]> (find-symbol "ZAPHOD-BEEBLEBROX" "COMMON-LISP-USER")
ZAPHOD-BEEBLEBROX ;
:INTERNAL


You should write:   #+#.(:fiveam)



-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
Until real software engineering is developed, the next best practice
is to develop with a dynamic system that has extreme late binding in
all aspects. The first system to really do this in an important way
is Lisp. -- Alan Kay


From asf at boinkor.net  Fri Aug 12 16:15:38 2005
From: asf at boinkor.net (Andreas Fuchs)
Date: Fri, 12 Aug 2005 18:15:38 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <17148.51054.734431.64434@thalassa.informatimago.com>
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> <m2ek8zns0v.fsf@soma.local>
	<871x4z5gd9.wl%asf@boinkor.net> <m2irybmaj1.fsf@soma.local>
	<17148.51054.734431.64434@thalassa.informatimago.com>
Message-ID: <87y8773zdh.wl%asf@boinkor.net>

Today, Pascal Bourguignon <pjb at informatimago.com> wrote:
>> #+#.(fiveam) since the string "fiveam" _will_ be read in the
>> #keyword
>> package.
>
> Why should it be read in the keyword package?

Because the spec says so. See 2.4.8.17.

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


From carlos.ungil at bluewin.ch  Fri Aug 12 18:56:56 2005
From: carlos.ungil at bluewin.ch (Carlos Ungil)
Date: Fri, 12 Aug 2005 20:56:56 +0200
Subject: [Bese-devel] yaclml test suite failures
Message-ID: <8ee8d4efa73b44cd3ce8570ed47d50d0@bluewin.ch>

Hello,

I had a problems with ucw on linux (only part of the html output was 
visible, and the pages couldn't be rendered) and I thought it was 
related to the indentation inserted inside tags. Finally it looks like 
the origin of my problems was elsewhere (I don't understand why, but 
everything works now).

However, I noticed that this indentation doesn't pass the tests  (later 
I've seen in some old mails that this is indeed a feature) . There are 
as well errors in the test suite (the output for sbcl 0.9.3 on linux 
FC3 is included at the end of this message).

Maybe if the test suite is not valid it can be removed, or at least 
this fact could be mentioned in the README).

Cheers,

Carlos

.................................
Did 39 checks.
    Pass: 31 (79%)
    Skip: 0 ( 0%)
    Fail: 8 (20%)

Failure Details:
NON-TAL []:
     "<pre
   >
   one

two
</pre
 >" was not STRING= to "<pre>
   one

two
</pre>".
NON-TAL []:
     "<a
   ><span
   /></a
 >" was not STRING= to "<a><span/></a>".
NON-TAL []:
     "<a
/>" was not STRING= to "<a/>".
TAL-WHEN []:
     Unexpected Error: #<TYPE-ERROR {9E0FBA9}>..
TAL-REPLACE []:
     Unexpected Error: #<TYPE-ERROR {9E30FE1}>..
TAL-CONTENT []:
     Unexpected Error: #<TYPE-ERROR {9E4FD89}>..
TAL-INCLUDE []:
     Unexpected Error: #<SIMPLE-ERROR {9E7CEF9}>..
TAL-ENV []:
     Unexpected Error: #<UNDEFINED-FUNCTION MAKE-TAL-ENVIRONMENT 
{9E83501}>..



From mb at bese.it  Fri Aug 12 22:31:16 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 13 Aug 2005 00:31:16 +0200
Subject: [Bese-devel] yaclml test suite failures
In-Reply-To: <8ee8d4efa73b44cd3ce8570ed47d50d0@bluewin.ch> (Carlos Ungil's
	message of "Fri, 12 Aug 2005 20:56:56 +0200")
References: <8ee8d4efa73b44cd3ce8570ed47d50d0@bluewin.ch>
Message-ID: <m264ua9497.fsf@soma.local>

Carlos Ungil <carlos.ungil at bluewin.ch> writes:

> However, I noticed that this indentation doesn't pass the tests
> (later I've seen in some old mails that this is indeed a feature) .

'feature' is a strong word. however the funny indentation is the only
way i can think of to indent html code without changing how browsers
interpret the code and without having to know wether a block should be
display: block or display: inline (and since this is impossible to
know with divs and spans we don't have much choice).

> There are as well errors in the test suite (the output for sbcl 0.9.3
> on linux FC3 is included at the end of this message).

thanks for pointing this out. (i think the last time i ran yaclml's
test suite was in mid 2004...)

> Maybe if the test suite is not valid it can be removed, or at least
> this fact could be mentioned in the README).

it should be updated, it will probably be removed (ucw is a far better
test suite than anything i could come up with by hand).

-- 
-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 Aug 13 16:11:07 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 13 Aug 2005 18:11:07 +0200
Subject: [Bese-devel] Changes II
In-Reply-To: <87pssndria.fsf@flarge.here> (Friedrich Dominicus's message of
	"Tue, 09 Aug 2005 18:06:21 +0200")
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
Message-ID: <87zmrl95r8.fsf@flarge.here>

May I ask if anyone has encountered the same problem than I did? Is it
on purpose and why does it happen. I'd appriciate to know what this
extraction of the lambda was necessary.

Regards
Friedrich


From mb at bese.it  Sun Aug 14 11:24:42 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 14 Aug 2005 13:24:42 +0200
Subject: [Bese-devel] Changes II
In-Reply-To: <87zmrl95r8.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sat, 13 Aug 2005 18:11:07 +0200")
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
	<87zmrl95r8.fsf@flarge.here>
Message-ID: <m23bpc3gn9.fsf@soma.local>

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

> May I ask if anyone has encountered the same problem than I did? Is it
> on purpose and why does it happen. I'd appriciate to know what this
> extraction of the lambda was necessary.

did you not recieve this email?

http://common-lisp.net/pipermail/bese-devel/2005-August/000756.html

or was i simply not clear enough?

-- 
-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  Sun Aug 14 12:36:29 2005
From: marco_simon at web.de (Marco Simon)
Date: Sun, 14 Aug 2005 14:36:29 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <m23bpc3gn9.fsf@soma.local>
References: <873bpjf7bj.fsf@flarge.here>
	<87pssndria.fsf@flarge.here>	<87zmrl95r8.fsf@flarge.here>
	<m23bpc3gn9.fsf@soma.local>
Message-ID: <42FF3ACD.6020602@web.de>

Hello list,

could somebody please give me short "hello world" on how to
use the tal-template engine ?
I thought it would be as easy as declaring an component with
the super-class "template-component"  and a template-name, but
that seemed not to be enough.
I even had a look at Marco Baringers's "counter" example which uses the
tal-template mechanism - but didn't manage it.
So could you please give me some example-lines which show how
to implement *.tal - files ?

Beside that (but this hasnt to do with ucw) I'm looking for a lisp-function
which escapes all " within a string to \" but didn't find any predifined 
cl-function
so far. Any hints ?

Thanks a lot
 best regards
  Marco




From whalliburton at gmail.com  Sun Aug 14 14:39:49 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Sun, 14 Aug 2005 09:39:49 -0500
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <42FF3ACD.6020602@web.de>
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
	<87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de>
Message-ID: <4e7bd29e05081407395c10d780@mail.gmail.com>

Ok. Here is an answer the second part. 

(defun escape-char (char str)
  "Return string with CHAR escaped in string STR with #\\"
  (with-output-to-string (s)
      (loop
       for c across str
       do (cond
          ((char= c char)
	   (write-char #\\ s)
	   (write-char c s))
	  (t (write-char c s))))))

On 8/14/05, Marco Simon <marco_simon at web.de> wrote:
> Hello list,
> 
> could somebody please give me short "hello world" on how to
> use the tal-template engine ?
> I thought it would be as easy as declaring an component with
> the super-class "template-component"  and a template-name, but
> that seemed not to be enough.
> I even had a look at Marco Baringers's "counter" example which uses the
> tal-template mechanism - but didn't manage it.
> So could you please give me some example-lines which show how
> to implement *.tal - files ?
> 
> Beside that (but this hasnt to do with ucw) I'm looking for a lisp-function
> which escapes all " within a string to \" but didn't find any predifined
> cl-function
> so far. Any hints ?
> 
> Thanks a lot
>  best regards
>   Marco
> 
> 
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel
>


From mlist at rogers.com  Sun Aug 14 23:39:11 2005
From: mlist at rogers.com (vedm)
Date: 14 Aug 2005 19:39:11 -0400
Subject: [Bese-devel] UCW "Hello World" error 
Message-ID: <86ll34ks0w.fsf@localhost.localdomain>


I tried UCW with CMUCL and SBCL (Debian unstable). In both of them when
I try to load http://127.0.0.1:8080/ucw/examples/index.ucw in the
browser, I get the following error in the SLIME debugger:

Type-error in KERNEL::OBJECT-NOT-SYMBOL-ERROR-HANDLER:
   (SETF IT.BESE.UCW:COMPONENT.PLACE) is not of type SYMBOL
   [Condition of type TYPE-ERROR]

Any idea how to fix this?

-- 
vedm


From whalliburton at gmail.com  Mon Aug 15 01:15:19 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Sun, 14 Aug 2005 20:15:19 -0500
Subject: [Bese-devel] UCW "Hello World" error
In-Reply-To: <86ll34ks0w.fsf@localhost.localdomain>
References: <86ll34ks0w.fsf@localhost.localdomain>
Message-ID: <4e7bd29e05081418151f0fab42@mail.gmail.com>

Hello,

I believe the answer you are looking for is at

http://common-lisp.net/pipermail/bese-devel/2005-August/000732.html
http://common-lisp.net/pipermail/bese-devel/2005-August/000733.html

This is fixed in latest ucw_dev + arnesi_dev.

Will


On 14 Aug 2005 19:39:11 -0400, vedm <mlist at rogers.com> wrote:
> 
> I tried UCW with CMUCL and SBCL (Debian unstable). In both of them when
> I try to load http://127.0.0.1:8080/ucw/examples/index.ucw in the
> browser, I get the following error in the SLIME debugger:
> 
> Type-error in KERNEL::OBJECT-NOT-SYMBOL-ERROR-HANDLER:
>    (SETF IT.BESE.UCW:COMPONENT.PLACE) is not of type SYMBOL
>    [Condition of type TYPE-ERROR]
> 
> Any idea how to fix this?
> 
> --
> vedm
> _______________________________________________
> 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  Mon Aug 15 05:46:45 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 07:46:45 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <42FF3ACD.6020602@web.de> (Marco Simon's message of "Sun,
	14 Aug 2005 14:36:29 +0200")
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
	<87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de>
Message-ID: <m23bpbycoq.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> could somebody please give me short "hello world" on how to use the
> tal-template engine?
> I thought it would be as easy as declaring an component with the
> super-class "template-component" and a template-name, but that
> seemed not to be enough.
> I even had a look at Marco Baringers's "counter" example which uses
> the tal-template mechanism - but didn't manage it.  So could you
> please give me some example-lines which show how to implement *.tal
> - files ?

using a tal template requires that 1) we create the .tal file and 2)
we tell ucw haw to find it.

#1 is, i hope, obvious, #2 requires that we pass :template-name as an
initarg when creating the component (most of the time, but not always,
you'll use the class option :default-initargs) and setting up the
proper tal-generator. a tal-generator is an object which, given a
template filename, returns a function which, given a tal evironment,
actually prints the page to *yaclml-stream*. at some point i planned
on having multiple types of tal-generators, for now all we have is the
file-system-generator, and i don't think we need more. every
application has its own generator (which is _not_ tied to the single
component), this lets use the same component with different templates
in different apps (though the same component with differnt templates
in the same app will require multiple component classes).

(defvar *example-application*
  (make-instance 'cookie-session-application
                 :url-prefix "/ucw/examples/"
                 :tal-generator (make-instance 'yaclml:file-system-generator
                                               :cachep t
                                               :root-directories (list *ucw-tal-root*))
                 :www-roots (list (merge-pathnames "./ucw/examples/" *ucw-tal-root*))))

that's the form whcih creates the example application object, notice
the :tal-generator keyword. file-system-generator has a list of
directories, in this case just *ucw-tal-root*, which it uses to search
for the actually template file. if our file-system-generator has
"/foo/" on its list of root-directories and we're looking for a
template named "here/there/whatever.tal" it will load the file at
"/foo/here/there/whatever.tal". 

in the case of the counter example the template is named
"ucw/examples/counter.tal" and the root-directories of the example
app's generator is just (list "/path/to/ucw/wwwroot/") hence
counter.tal needs to live at
"/path/to/ucw/wwwroot/ucw/examples/counter.tal". 

does this make sense?

once we've found the template we need to actually use it. all tal
templates are eventually converted to lisp code and use yaclml macros
underneath, with some extensions. when you use an anchor tag in a tal
file (if you've setup the xmlns properly) you'll eventually "run" the
<:A macro, when you see <ucw:option value="foobar">blah
blah</ucw:option> that is exactly the same as (<ucw:option :value
"foobar" "blah blah blah") in regular lisp code. the tal attribute
extensions are (or should be) documentated somewhere, here's a quick
summary:

Attributes:

tla:content="(something)" - (SOMETHING) is evaluated to produce a
string and that is used as the body of the tag.

tal:replace="(something)" - the entire tag (including all attributes)
is replaced with the string returned by (SOMETHING).

tal:dolist="(something)" - (SOMETHING) returns a list of tal
environment objects (which are themselves lists of alists, clos
objects or hash tables). the body of the tag is executed once for each
environment object returned by (SOMETHING), the current env is
extended with that environment.

tal:in-package="package-name" - The package ti use for evaluating
code, interneing names, etc.

Tags:

<tal:tal> - does nothing in and of itself, but it provides a place to
put tal:attributes when you don't want to add another tag. <div> or
<span> should work just as well but browsers aren't perfect.

<tal:include> - include another tal template.

  tal:name="whateverl.tal" - the name of the of the file to include (not
  evaluated). the file is found using the same mechanism as described
  above.

  param:xyz="(something)" - something is evaluated and its return
  value (a list object) is put into included file's environment under
  the name xyz. (xyz is converted into a symbol in the current tal
  package).

  <param:xyz> ... </param:xyz> - xyz is bound to the _string_ which
  the body of the tag produces.

Other things:

tal attributes are normally evaluated. there is a read macro bound to
#\$ which will lookup in the current env the value associated with the
symbol. So $whatever in a tal attribute correspands to the value of
the lisp object bonud to the name whatever (whatever is a symbol in
the current tal package (see tal:in-package above)).

non tal attributes are generally strings which are left alone. the
exception is if the contain ${ ... } or @{ ... }. in this case the
body of ... is evaluated it it's return value (a string in the case of
$, a list of strings in the case of @) is substituted for the text
occupied by the ${ ... } code. example:

<table tal:in-package=":it.bese.ucw-user">
  <!-- all symbols, etc will be interned into the :it.bese.ucw-user
       package -->
  <tr tal:dolist="$rows"> 
    <!-- loop over the tal env bound to ROWS
    <td style='background: #${ (if $even-row "000000" "333333") };"
        tal:content="$row-value">
      <!-- even rows get a white background, odd rows get a light gray
           background. this text inside the ttag will be replace
           with the value associated with $body -->
      dummy value goes here
    </td>
  </tr>
</table>

hth.

p.s. - this is the way things _should_ be. i haven't done a lot of tal
coding lately so there may be deviations (which are bugs and sholud be
reported as such).

-- 
-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 Aug 15 06:29:46 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 08:29:46 +0200
Subject: [Bese-devel] ucw and parenscript
Message-ID: <m27jenww4l.fsf@soma.local>


as of festerday now ucw depends on parenscript. You can get
parenscript from the darcs repo here:

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

-- 
-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 Aug 15 08:23:41 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 10:23:41 +0200
Subject: [Bese-devel] ucw and parenscript
In-Reply-To: <7c23adaa05081501203302f321@mail.gmail.com> (Ivan Toshkov's
	message of "Mon, 15 Aug 2005 11:20:09 +0300")
References: <m27jenww4l.fsf@soma.local>
	<7c23adaa05081501203302f321@mail.gmail.com>
Message-ID: <m2zmrjr4ky.fsf@soma.local>

Ivan Toshkov <itoshkov at gmail.com> writes:

> I've modified it your version of parenscript and send it here. 
> Incidentally, `dolist' has the same problem as `dotimes'.  This patch
> fixes that, too.

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  Mon Aug 15 08:25:04 2005
From: marco_simon at web.de (Marco Simon)
Date: Mon, 15 Aug 2005 10:25:04 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <4e7bd29e05081407395c10d780@mail.gmail.com>
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>	
	<87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>	
	<42FF3ACD.6020602@web.de>
	<4e7bd29e05081407395c10d780@mail.gmail.com>
Message-ID: <43005160.2070601@web.de>

Hi William,

I fear I didn't explain correctly what's my need.
Perhaps I should tell what I wanna do:

I want to embed some html or js snippets directly into
my ucw-code. (something like
(<:as-is code_snippet)

An easy example would be:
(<:as-is "<script type=\"text/javascript\"> alert(\"Hello 
World\")</script)")

As you can see I had to escape several double-quotation-marks in order to
make lisp handle this as one Sting. This "escaping by hand" might be ok for
short and handwritten snippets, but if I've got some longer js-code with 
lots
of " in it, it will become some work to escape all all of them " by hand 
(if even possible).

This is why I'am looking for some function which escapes all " within 
that snipped.
In php I'd easyly write someting like this:
 <? echo addslashes($snipped) ?> or
<? $mysnipped = '<script type="text/javascript"> alert ("Hello World") 
</script>' ?>

I'm absolutely sure that something similar is even possible with lisp, 
but I didn't
realize how to do so far.

Thanks for your hints,
 best regards
 Marco







William Halliburton schrieb:

>Ok. Here is an answer the second part. 
>
>(defun escape-char (char str)
>  "Return string with CHAR escaped in string STR with #\\"
>  (with-output-to-string (s)
>      (loop
>       for c across str
>       do (cond
>          ((char= c char)
>	   (write-char #\\ s)
>	   (write-char c s))
>	  (t (write-char c s))))))
>
>On 8/14/05, Marco Simon <marco_simon at web.de> wrote:
>  
>
>>Hello list,
>>
>>could somebody please give me short "hello world" on how to
>>use the tal-template engine ?
>>I thought it would be as easy as declaring an component with
>>the super-class "template-component"  and a template-name, but
>>that seemed not to be enough.
>>I even had a look at Marco Baringers's "counter" example which uses the
>>tal-template mechanism - but didn't manage it.
>>So could you please give me some example-lines which show how
>>to implement *.tal - files ?
>>
>>Beside that (but this hasnt to do with ucw) I'm looking for a lisp-function
>>which escapes all " within a string to \" but didn't find any predifined
>>cl-function
>>so far. Any hints ?
>>
>>Thanks a lot
>> best regards
>>  Marco
>>
>>
>>_______________________________________________
>>bese-devel mailing list
>>bese-devel at common-lisp.net
>>http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel
>>
>>    
>>
>
>
>  
>




From itoshkov at gmail.com  Mon Aug 15 08:20:09 2005
From: itoshkov at gmail.com (Ivan Toshkov)
Date: Mon, 15 Aug 2005 11:20:09 +0300
Subject: [Bese-devel] ucw and parenscript
In-Reply-To: <m27jenww4l.fsf@soma.local>
References: <m27jenww4l.fsf@soma.local>
Message-ID: <7c23adaa05081501203302f321@mail.gmail.com>

On 8/15/05, Marco Baringer <mb at bese.it> wrote:
> 
> as of festerday now ucw depends on parenscript. You can get
> parenscript from the darcs repo here:
> 
> http://common-lisp.net/project/ucw/repos/parenscript
> 

Some time ago I found the following problem with parenscipt's dotimes:

>  (js:js (dotimes (i args.length) (print i)))
"for (var i = 0; i != args.length; i = i++) {
  print(i);
}"

As you can see this is an infinite loop unless `args.length' is zero. 
Unfortunately, I was unable to reach parenscript's author and left the
patch in the cliki page.

I've modified it your version of parenscript and send it here. 
Incidentally, `dolist' has the same problem as `dotimes'.  This patch
fixes that, too.

Cheers,
Ivan

-- 
All languages are equal,
but Lisp is more equal than others.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patch
Type: application/octet-stream
Size: 1985 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050815/fa5c1778/attachment.obj>

From mb at bese.it  Mon Aug 15 08:34:09 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 10:34:09 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <43005160.2070601@web.de> (Marco Simon's message of "Mon,
	15 Aug 2005 10:25:04 +0200")
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
	<87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de> <4e7bd29e05081407395c10d780@mail.gmail.com>
	<43005160.2070601@web.de>
Message-ID: <m2hddrr43i.fsf@soma.local>

Marco Simon <marco_simon at web.de> writes:

> Hi William,
>
> I fear I didn't explain correctly what's my need.
> Perhaps I should tell what I wanna do:
>
> I want to embed some html or js snippets directly into
> my ucw-code. (something like
> (<:as-is code_snippet)
>
> An easy example would be:
> (<:as-is "<script type=\"text/javascript\"> alert(\"Hello
> World\")</script)")

grab a very recent ucw and parenscript (see other email to
bese-devel):

(<ucw:script
  (alert "Hello World"))

that is sooooo much better.

having said that, what you probably want is just an easy way to read
in a string and setup the delimiting chars based on the content of the
string:

CL-USER> (defun |#/ reader| (stream sub-char prefix)
           (declare (ignore sub-char prefix))
           (with-output-to-string (string)
             (loop
                with delimiter = (read-char stream t nil t)
                for char = (read-char stream t nil t)
                until (char= delimiter char)
                do (write-char char string))))
                
|#/ reader|
CL-USER> (set-dispatch-macro-character #\# #\/ #'|#/ reader|)
T
CL-USER> #/"foo"
"foo"
CL-USER> #/!foo!
"foo"
CL-USER> #/'foo'
"foo"
CL-USER> 

You could just as well make it always use #\/ and avoid the extra
chars.

> As you can see I had to escape several double-quotation-marks in
> order to make lisp handle this as one Sting. This "escaping by hand"
> might be ok for short and handwritten snippets, but if I've got some
> longer js-code with lots of " in it, it will become some work to
> escape all all of them " by hand (if even possible).

you could also just put everything in an external js file and pass the
:javascript arg to simple-window-component.

> This is why I'am looking for some function which escapes all " within
> that snipped.
> In php I'd easyly write someting like this:
>  <? echo addslashes($snipped) ?> or
> <? $mysnipped = '<script type="text/javascript"> alert ("Hello World")
> </script>' ?>
>
> I'm absolutely sure that something similar is even possible with
> lisp, but I didn't realize how to do so far.

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 Aug 15 09:36:40 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 11:36:40 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <20050815091319.GA58357@pvv.ntnu.no> (astor@pvv.ntnu.no's message
	of "Mon, 15 Aug 2005 11:13:19 +0200")
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
	<87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de> <4e7bd29e05081407395c10d780@mail.gmail.com>
	<43005160.2070601@web.de> <m2hddrr43i.fsf@soma.local>
	<20050815091319.GA58357@pvv.ntnu.no>
Message-ID: <m28xz3pmmv.fsf@soma.local>

astor at pvv.ntnu.no writes:

> On Mon, Aug 15, 2005 at 10:34:09AM +0200, Marco Baringer wrote:
>> 
>> grab a very recent ucw and parenscript (see other email to
>> bese-devel):
>> 
>> (<ucw:script
>>   (alert "Hello World"))
>> 
>> that is sooooo much better.
>> 
>
> (<ucw:header-script ...) or something similart would also be nice.
> Pages are supposed to be faster when scripts are put in the HEADER
> section since this section can be processed before/while the page is
> being laid out.

sorry, i don't follow, why not just put it in the header then?

it's not available quite yet the :inline-javascript
simple-window-component initarg will soon take a list of parenscript
code and not a string as it does now.

-- 
-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 Aug 15 10:15:55 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 15 Aug 2005 12:15:55 +0200
Subject: [Bese-devel] Got in troubles again
Message-ID: <87pssfh5es.fsf@flarge.here>

I do not understand what's wrong with the following code:

;;; Just for testing
(defcomponent ir-test (simple-window-component form-component)
  ((num :accessor num 
        :component (integer-range-field :min-value 0 :max-value 20))
   (foo-num :accessor foo-num
            :initarg :foo-num 
            :component (integer-field))))



(defmethod render-on ((res response) (tf ir-test))
  (render-on res (num tf))
  (render-on res (foo-num tf)))



(defentry-point "ti.ucw" (:application *shop*) ()
  (call 'ir-test :foo-num 10))


Error message I got is:
There is no applicable method for the generic function
  #<STANDARD-GENERIC-FUNCTION (SETF IT.BESE.UCW:COMPONENT.PLACE) (1)>
when called with arguments
  (#<IT.BESE.UCW::PLACE (SLOT-VALUE IT.BESE.UCW::COMP
                                    IT.BESE.UCW::SLOT-NAME) {100116B301}>
   10).
   [Condition of type SIMPLE-ERROR]

Restarts:
  0: [SHOW-BACKTRACE] Send the client a backtrace page.
  1: [RETRY] Clearout the response and retry calling the action.
  2: [SERVER-ERROR] Send the client an internal server error page.
  3: [GENERATE-BACKTRACE-FOR-EMACS] Generate a bug report in Emacs.
  4: [ABORT-RESPONSE] Abort this response and answer another request
  5: [ABORT] Exit debugger, returning to top level.


It was my impresson that I can put in the same argument to call as I
can for make-intance So what does this ..... error message mean? What
do I have to to get that running? 

Regards
Friedrich


From mb at bese.it  Mon Aug 15 10:28:08 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 12:28:08 +0200
Subject: [Bese-devel] Got in troubles again
In-Reply-To: <87pssfh5es.fsf@flarge.here> (Friedrich Dominicus's message of
	"Mon, 15 Aug 2005 12:15:55 +0200")
References: <87pssfh5es.fsf@flarge.here>
Message-ID: <m2ll33o5on.fsf@soma.local>

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

> I do not understand what's wrong with the following code:
>
> ;;; Just for testing
> (defcomponent ir-test (simple-window-component form-component)
>   ((num :accessor num 
>         :component (integer-range-field :min-value 0 :max-value 20))
>    (foo-num :accessor foo-num
>             :initarg :foo-num 
>             :component (integer-field))))
>
>
>
> (defmethod render-on ((res response) (tf ir-test))
>   (render-on res (num tf))
>   (render-on res (foo-num tf)))
>
>
>
> (defentry-point "ti.ucw" (:application *shop*) ()
>   (call 'ir-test :foo-num 10))

the foo-num slot is assumed to be component, you've set it to 10,
which isn't a component. ucw barfs when it tries to set the place of
the number 10.

what you want is this:

(defmethod shared-initialize :after ((c ir-test) slot-names &key foo-num &allow-other-keys)
  (setf (lisp-value (foo-num c)) foo-num))

iow you want the value passed as :foo-num to end up in a slot of the
underlying component. 

the fact that this shared-initialize is neccessary is pretty ugly, but
i have yet to figure out a good way to structure it, suggestions
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 mb at bese.it  Mon Aug 15 10:37:16 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 12:37:16 +0200
Subject: [Bese-devel] Got in troubles again
In-Reply-To: <m2ll33o5on.fsf@soma.local> (Marco Baringer's message of "Mon,
	15 Aug 2005 12:28:08 +0200")
References: <87pssfh5es.fsf@flarge.here> <m2ll33o5on.fsf@soma.local>
Message-ID: <m2ek8vo59f.fsf@soma.local>

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

>
> what you want is this:
>
> (defmethod shared-initialize :after ((c ir-test) slot-names &key foo-num &allow-other-keys)
>   (setf (lisp-value (foo-num c)) foo-num))
>
> iow you want the value passed as :foo-num to end up in a slot of the
> underlying component. 
>
> the fact that this shared-initialize is neccessary is pretty ugly, but
> i have yet to figure out a good way to structure it, suggestions
> welcome.

this could alse be nice:

(defentry-point "ti.ucw" (:application *shop*) ()
  (call 'ir-test :foo-num `(:lisp-value 10)))

basically we take anything passed as an initarg to a component slot
and pass that as initargs when creating the underlying component. this
would also allow you to do stuff like:

(call 'ir-test :foo-num `(:lisp-value ,*default-value*
                          :client-value ""))

good enough?

-- 
-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 astor at pvv.ntnu.no  Mon Aug 15 11:50:48 2005
From: astor at pvv.ntnu.no (astor at pvv.ntnu.no)
Date: Mon, 15 Aug 2005 13:50:48 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <m28xz3pmmv.fsf@soma.local>
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
	<87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de>
	<4e7bd29e05081407395c10d780@mail.gmail.com>
	<43005160.2070601@web.de> <m2hddrr43i.fsf@soma.local>
	<20050815091319.GA58357@pvv.ntnu.no> <m28xz3pmmv.fsf@soma.local>
Message-ID: <20050815115048.GD58357@pvv.ntnu.no>

On Mon, Aug 15, 2005 at 11:36:40AM +0200, Marco Baringer wrote:
> astor at pvv.ntnu.no writes:
> > Pages are supposed to be faster when scripts are put in the HEADER
> > section since this section can be processed before/while the page is
> > being laid out.
> 
> sorry, i don't follow, why not just put it in the header then?
> 
> it's not available quite yet the :inline-javascript
> simple-window-component initarg will soon take a list of parenscript
> code and not a string as it does now.
> 

Typically you want to keep the javascript and the HTML close to each
other for a component.  So you might want an easy way to add to the
"header" section in the way that render-on adds to the "body" section.
Adding to an inline-javascript slot or something like that would do
the trick I guess.

astor


From mb at bese.it  Mon Aug 15 13:25:38 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 15:25:38 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <20050815115048.GD58357@pvv.ntnu.no> (astor@pvv.ntnu.no's message
	of "Mon, 15 Aug 2005 13:50:48 +0200")
References: <873bpjf7bj.fsf@flarge.here> <87pssndria.fsf@flarge.here>
	<87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de> <4e7bd29e05081407395c10d780@mail.gmail.com>
	<43005160.2070601@web.de> <m2hddrr43i.fsf@soma.local>
	<20050815091319.GA58357@pvv.ntnu.no> <m28xz3pmmv.fsf@soma.local>
	<20050815115048.GD58357@pvv.ntnu.no>
Message-ID: <m2slxbmiwd.fsf@soma.local>

astor at pvv.ntnu.no writes:

> On Mon, Aug 15, 2005 at 11:36:40AM +0200, Marco Baringer wrote:
>> astor at pvv.ntnu.no writes:
>> > Pages are supposed to be faster when scripts are put in the HEADER
>> > section since this section can be processed before/while the page is
>> > being laid out.
>> 
>> sorry, i don't follow, why not just put it in the header then?
>> 
>> it's not available quite yet the :inline-javascript
>> simple-window-component initarg will soon take a list of parenscript
>> code and not a string as it does now.
>> 
>
> Typically you want to keep the javascript and the HTML close to each
> other for a component.  So you might want an easy way to add to the
> "header" section in the way that render-on adds to the "body" section.
> Adding to an inline-javascript slot or something like that would do
> the trick I guess.

currenlty we don't have a way for a nested componnet to affect the
tags generated by the window component, nor do we have an already
existing mechanism which would allow this.

would it be worth the effort to make simple-window-component (or
something) walk the component tree before the render-on phase and
collect extra head headers (like script or even meta or link)?

-- 
-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 Aug 15 14:19:01 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 15 Aug 2005 16:19:01 +0200
Subject: [Bese-devel] Got in troubles again
In-Reply-To: <m2ll33o5on.fsf@soma.local> (Marco Baringer's message of "Mon,
	15 Aug 2005 12:28:08 +0200")
References: <87pssfh5es.fsf@flarge.here> <m2ll33o5on.fsf@soma.local>
Message-ID: <87fytbgu5m.fsf@flarge.here>

Thanks I found out half an hour after my posting how this works and
was "hacking" something similiar but it's really ugly. Assume you have
an object with 10 slots and 10 initargs, there must be  a better
way, but I have no idea yet. It's a pitty.

Friedrich
              


From astor at pvv.ntnu.no  Mon Aug 15 15:34:45 2005
From: astor at pvv.ntnu.no (astor at pvv.ntnu.no)
Date: Mon, 15 Aug 2005 17:34:45 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <m2slxbmiwd.fsf@soma.local>
References: <87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de>
	<4e7bd29e05081407395c10d780@mail.gmail.com>
	<43005160.2070601@web.de> <m2hddrr43i.fsf@soma.local>
	<20050815091319.GA58357@pvv.ntnu.no> <m28xz3pmmv.fsf@soma.local>
	<20050815115048.GD58357@pvv.ntnu.no> <m2slxbmiwd.fsf@soma.local>
Message-ID: <20050815153445.GF58357@pvv.ntnu.no>

On Mon, Aug 15, 2005 at 03:25:38PM +0200, Marco Baringer wrote:
> 
> currenlty we don't have a way for a nested componnet to affect the
> tags generated by the window component, nor do we have an already
> existing mechanism which would allow this.
> 
> would it be worth the effort to make simple-window-component (or
> something) walk the component tree before the render-on phase and
> collect extra head headers (like script or even meta or link)?
> 

I think so.  But just to complete the picture, I can think of the
following use-cases:

o Sometimes you want to place supporting javascript in HEAD.  This is
  the proper way to ensure that they are available.  Putting the script
  in BODY could often be buggy since the code might not be available
  when an event fires in the javascript world.

o Sometimes a component might want to place supporting javascript in
  HEAD _once_ even if the component is rendered multiple times.
  (maybe a property-like interface would be good for this stuff).

o Maybe a component would like to produce an "external"
  .js-file for that particular component and refer to that .js-file
  _once_ in HEAD if the component is being used.  This could be a good
  solution for complex components (AJAX maybe).

o Sometimes you want to put javascript in BODY, but AFAIK, that will
  typically not be code that simply defines functions, but rather code
  that does stuff, like your alert example.  But this is supported
  rather well already I think.

So to me it seems like a component might want to produce:
o Global code (external .js-file)
o Per request, "per instance" code in HEAD.
o Per request, "per instance" code in BODY.
o Per request, "per class" code in HEAD.

astor


From mb at bese.it  Mon Aug 15 15:53:55 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 15 Aug 2005 17:53:55 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <20050815153445.GF58357@pvv.ntnu.no> (astor@pvv.ntnu.no's message
	of "Mon, 15 Aug 2005 17:34:45 +0200")
References: <87zmrl95r8.fsf@flarge.here> <m23bpc3gn9.fsf@soma.local>
	<42FF3ACD.6020602@web.de> <4e7bd29e05081407395c10d780@mail.gmail.com>
	<43005160.2070601@web.de> <m2hddrr43i.fsf@soma.local>
	<20050815091319.GA58357@pvv.ntnu.no> <m28xz3pmmv.fsf@soma.local>
	<20050815115048.GD58357@pvv.ntnu.no> <m2slxbmiwd.fsf@soma.local>
	<20050815153445.GF58357@pvv.ntnu.no>
Message-ID: <m2vf27qjqk.fsf@soma.local>

astor at pvv.ntnu.no writes:

> So to me it seems like a component might want to produce:
> o Global code (external .js-file)
> o Per request, "per class" code in HEAD.

these two can be done with the same mechanism? if we had a method
add-to-head-once-only:

(defmethod add-to-head-once-only ((c my-component))
  (<:script :src "foo.js")
  (<:meta ...)
  (<:link ...))

> o Per request, "per instance" code in HEAD.

how does the name add-to-head sound?

> o Per request, "per instance" code in BODY.

as you said i'm fairly pleased with what we have now.

-- 
-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 Alan-Shields at omrf.ouhsc.edu  Mon Aug 15 18:31:21 2005
From: Alan-Shields at omrf.ouhsc.edu (Alan Shields)
Date: Mon, 15 Aug 2005 13:31:21 -0500
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <m2ek8zns0v.fsf@soma.local>
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> <m2ek8zns0v.fsf@soma.local>
Message-ID: <20050815183121.GA18785@inquisitor.omrf.org>

On Fri, Aug 12, 2005 at 04:34:08PM +0200, Marco Baringer wrote:
> Alan Shields <Alan-Shields at omrf.ouhsc.edu> writes:
> 
> > Thu Aug 11 16:20:32 CDT 2005  Alan Shields <Alan-Shields at omrf.ouhsc.edu>
> >   * Add :fiveam to *features*
> >   I've found that I prefer to add testing code in my other code, but I dislike
> >   adding :fiveam to my package requirements list.
> 
> i am hesitant to modify *features*, it's not really supposed to be
> used for this (afaict).

Several packages modify *features* (such as CL-PPCRE, and
SPLIT-SEQUENCE) to include their package name. ASDF as well.

While CLTL specifies that *features* contains tokens "provided by the
implementation", CLHS specifies that the tokens "correspond to some
aspect of the implementation or environment." This difference would seem
to be deliberate.

I am no common lisp expert, but it would seem to me that *features*
exists to solve exactly the problem that we're facing here: conditional
inclusion of code based on the environment in which the code is loaded.

As you say yourself further in the thread: the #. solution leads to
uglyness.

Alan


From smoof-ra at elder-gods.org  Mon Aug 15 20:31:06 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Mon, 15 Aug 2005 16:31:06 -0400
Subject: [Bese-devel] declarations
Message-ID: <20050815203105.GA843@melkor.elder-gods.org>


Here's another declarations patch.  Now they have parent pointers and
are parsed into useful classes.

    --larry 
-------------- next part --------------

New patches:

[more usefull declaration objects
smoof-ra at elder-gods.org**20050815202603] {
hunk ./src/walk.lisp 141
-(defun split-body (body env &key (docstring t) (declare t))
+(defun split-body (body env &key parent (docstring t) (declare t))
hunk ./src/walk.lisp 143
-	(decl nil)
+	(newdecls nil)
hunk ./src/walk.lisp 155
-                            (multiple-value-setf (env decl) (parse-declaration dec env))
-			    (when decl
-			      (push decl decls))))
+                            (multiple-value-setf (env newdecls) (parse-declaration dec env parent))
+			    (setf decls (append newdecls decls))))
hunk ./src/walk.lisp 180
-(defun parse-declaration (declaration environment)
-  (macrolet ((extend-env ((var list) &rest datum)
-               `(dolist (,var ,list)
-                  (setf environment (register environment :declare , at datum)))))
-    (destructuring-bind (type &rest arguments)
-        declaration
-      (case type
-        (dynamic-extent
-         (extend-env (var arguments)
-                     var `(dynamic-extent)))        
-        (ftype
-         (extend-env (function-name (cdr arguments))
-                     function-name `(ftype ,(first arguments))))
-        (ignorable
-         (extend-env (var arguments)
-                     var `(ignorable)))
-        (ignore
-         (extend-env (var arguments)
-                     var `(ignorable)))
-        (inline
-          (extend-env (function arguments) function `(ignorable)))
-        (notinline
-         (extend-env (function arguments) function `(notinline)))
-        (optimize
-         (extend-env (optimize-spec arguments) 'optimize optimize-spec))
-        (special
-         (extend-env (var arguments) var `(special)))
-        (type
-         (extend-env (var (rest arguments)) var `(type ,(first arguments))))
-        (t
-         (extend-env (var arguments) var `(type ,type))))))
-  (values environment (make-instance 'declaration-form :source declaration)))
+(defclass optimize-declaration-form (declaration-form)
+  ((optimize-spec :accessor optimize-spec :initarg :optimize-spec)))
+
+(defclass variable-declaration-form (declaration-form)
+  ((name :accessor name :initarg :name)))
+
+(defclass function-declaration-form (declaration-form)
+  ((name :accessor name :initarg :name)))
+
+(defclass dynamic-extent-declaration-form (variable-declaration-form)
+  ())
+
+(defclass ignorable-declaration-form-mixin (declaration-form)
+  ())
+
+(defclass variable-ignorable-declaration-form (variable-declaration-form ignorable-declaration-form-mixin)
+  ())
+
+(defclass function-ignorable-declaration-form (function-declaration-form ignorable-declaration-form-mixin)
+  ())
+
+(defclass special-declaration-form (variable-declaration-form)
+  ())
+
+(defclass type-declaration-form (variable-declaration-form)
+  ((type-form :accessor type-form :initarg :type-form)))
+
+(defclass ftype-declaration-form (function-declaration-form)
+  ((type-form :accessor type-form :initarg :type-form)))
+
+(defclass notinline-declaration-form (function-declaration-form)
+  ())
+
+
+(defun parse-declaration (declaration environment parent)
+  (let ((declares nil))
+    (flet ((funname (form)
+	     (if (and (consp form) (eql (car form) 'function))
+		 (cadr form)
+		 nil)))
+      (macrolet ((mkdecl (varname formclass &rest rest)
+		   `(make-instance ,formclass :parent parent :source (list type ,varname) , at rest))		 
+		 (extend-env ((var list) newdeclare &rest datum)
+		   `(dolist (,var ,list)
+		      (when ,newdeclare (push ,newdeclare declares))
+		      (setf environment (register environment :declare , at datum)))))
+	(destructuring-bind (type &rest arguments)
+	    declaration
+	  (case type
+	    (dynamic-extent
+	     (extend-env (var arguments)
+			 (mkdecl var 'variable-ignorable-declaration-form) 
+			 var `(dynamic-extent)))        
+	    (ftype
+	     (extend-env (function-name (cdr arguments))
+			 (make-instance 'ftype-declaration-form 
+					:parent parent
+					:source `(ftype ,(first arguments) function-name)
+					:name function-name
+					:type-form (first arguments))
+			 function-name `(ftype ,(first arguments))))
+	    ((ignore ignorable)
+	     (extend-env (var arguments)
+			 (aif (funname var)
+			      (mkdecl var 'function-ignorable-declaration-form :name it)
+			      (mkdecl var 'variable-ignorable-declaration-form :name var))
+			 var `(ignorable)))
+	    (inline
+	      (extend-env (function arguments) 
+			  (mkdecl function 'function-ignorable-declaration-form :name function)
+			  function `(ignorable)))
+	    (notinline
+	     (extend-env (function arguments)
+			 (mkdecl function 'notinline-declaration-form :name function)
+			 function `(notinline)))
+	    (optimize
+	     (extend-env (optimize-spec arguments) 
+			 (mkdecl optimize-spec 'optimize-declaration-form :optimize-spec optimize-spec)
+			 'optimize optimize-spec))
+	    (special
+	     (extend-env (var arguments) 
+			 (mkdecl var 'special-declaration-form :name var)
+			 var `(special)))
+	    (type
+	     (extend-env (var (rest arguments))
+			 (make-instance 'type-declaration-form 
+					:parent parent
+					:source `(type ,(first arguments) var)
+					:name var
+					:type-form (first arguments))
+			 var `(type ,(first arguments))))
+	    (t
+	     (extend-env (var arguments)
+			 (make-instance 'type-declaration-form 
+					:parent parent
+					:source `(,type var)
+					:name var
+					:type-form type)
+			 var `(type ,type)))))))
+    (when (null declares)
+      (setq declares (list (make-instance 'declaration-form :parent parent :source declaration))))
+    (values environment declares)))
hunk ./src/walk.lisp 285
-      (split-body forms env :docstring docstring :declare declare)
+      (split-body forms env :parent parent :docstring docstring :declare declare)
}

Context:

[Change the name of the walker handler for atom so that it doesn't used to walk forms whose car is CL:ATOM.
Marco Baringer <mb at bese.it>**20050812114613] 
[Added required specifier :primary to cc-standard method combination
Marco Baringer <mb at bese.it>**20050812114343
 
 This prevents people from accidentaly using defmethod on a cc generic
 function. You must now either use defmethod/cc (which retains the same
 API as before), or pass the :primary method qualifier (and if you do
 we'll assume you know what you're doing).
 
] 
[Update the list of exported symbols
Marco Baringer <mb at bese.it>**20050812102709] 
[Two minor indentation fixups
Marco Baringer <mb at bese.it>**20050812102429] 
[Remove type declarations form join-string (ucw passes in strings which violate some of the type declarations)
Marco Baringer <mb at bese.it>**20050812102333] 
[JOIN-STRINGS was not returning the result :(
Marco Baringer <mb at bese.it>**20050812070730] 
[Added more tests for cc-standard method combination
Marco Baringer <mb at bese.it>**20050811152516] 
[Make the arguments to a continuation all option (the equivalent of evaluating (values))
Marco Baringer <mb at bese.it>**20050811152449] 
[Added cc definintions of common standard functions whcih take functions as parameters (assoc and mapXYZ)
Marco Baringer <mb at bese.it>**20050811152400] 
[Added missing cleane-argument-list function and tests
Marco Baringer <mb at bese.it>**20050811130722] 
[Fix generic/method/function lambda list handling and manipulation in walker and cc-interpreter
Marco Baringer <mb at bese.it>**20050811130002] 
[Added support for &key, &allow-other-keys in lambda/cc (and therefore defun/cc and defmethod/cc)
Marco Baringer <mb at bese.it>**20050810140725] 
[keyword-function-argument-form is not a subclass of optional-argument-form, even though they are superficially similar
Marco Baringer <mb at bese.it>**20050810140659] 
[Move debugging code into DEFK where have access to the name of the continuation we're calling (as apposed to KLAMBDA)
Marco Baringer <mb at bese.it>**20050810121305] 
[Refactor cc-interpreter into a few smaller files in the call-cc directory
Marco Baringer <mb at bese.it>**20050810104819] 
[Refactor FOLD-STRING (Patch by: Janis Dzerins <jdz at dir.lv>)
Marco Baringer <mb at bese.it>**20050810103438] 
[the fdefinition/cc table must remeber, other than the def, whether it was a function or a method (since we treat those two differently)
Marco Baringer <mb at bese.it>**20050810101541] 
[Added :method-combination option to the expaniosn of the defgeneric/cc macro
Marco Baringer <mb at bese.it>**20050810101446] 
[Added *debug-evaluate/cc* and related functions
Marco Baringer <mb at bese.it>**20050810063228] 
[Use fedinition/cc, not fdefinition, for defun/cc
Marco Baringer <mb at bese.it>**20050810055126] 
[Typo in name of k-for-apply/cc/optional-argument-default-value continuation
Marco Baringer <mb at bese.it>**20050810055105] 
[Don't use GET so that we don't break package locks. Add now function fdefinition/cc
Marco Baringer <mb at bese.it>**20050810054327
 
 Basically it's a bad idea to use #'get and #'(setf get) on standard
 symbols, however this is something we're going to want to do. We know
 have our own hash-table (*cc-functions*) which maps symbols to closure
 objects.
 
] 
[ method combination 
smoof-ra at elder-gods.org**20050810035825] 
[make closure/cc a funcallable instance (in sbcl)
smoof-ra at elder-gods.org**20050809182900
 
 funcalling a closure/cc will just call it with a toplevel continuation
 that way cc closures can be called totally transparently by non-cc code
 as long as the cc-closure doesn't call/cc.
 
] 
[Added copyright notice to cc-interpreter.lisp
Marco Baringer <mb at bese.it>**20050809151738] 
[Remove occurences of 'cps' in the api. We don't actually cps transforme anymore so this is misleading.
Marco Baringer <mb at bese.it>**20050809125933] 
[Remove 'cps' from test suite, replace it with 'call/cc' 
Marco Baringer <mb at bese.it>**20050809104818] 
[Rename cps.lisp to cc-interpreter.lisp
Marco Baringer <mb at bese.it>**20050809104737] 
[Fix typo in fold-strings' docstring
Marco Baringer <mb at bese.it>**20050809083006] 
[Trivial change to the name of the gensym generated by DOLIST*
Marco Baringer <mb at bese.it>**20050809061102] 
[Fix lexical-variables and lexical-functions on clisp
Marco Baringer <mb at bese.it>**20050807223244] 
[Use FDEFINITION instead of SYMBOL-FUNCTION to get a function from a function name.
Marco Baringer <mb at bese.it>**20050807222932] 
[Make sure we only pass symbols to functions like GET and MACRO-FUNCTION
Marco Baringer <mb at bese.it>**20050807222905] 
[Implement lexical-functions for CLISP
Marco Baringer <mb at bese.it>**20050807204738] 
[Implement lexical-variables and lexical-functions for NIL environments
Marco Baringer <mb at bese.it>**20050807204711] 
[Remove arnesi.el from ssytem def
Marco Baringer <mb at bese.it>**20050807204654] 
[Fix evaluation of #'(foo bar) in cps interpreter
Marco Baringer <mb at bese.it>**20050806182653] 
[Delete arnesi.el. SLIME is perfectly able to figure out the indententation by itself.
Marco Baringer <mb at bese.it>**20050807075500] 
[aparently global variables can be found in sbcl lexical environments
smoof-ra at elder-gods.org**20050804164859] 
[Implement environment-p and lexical-variables for CLISP
Marco Baringer <mb at bese.it>**20050804165821] 
[Make the lexenv stuff use generic-functions and methods
Marco Baringer <mb at bese.it>**20050804161857] 
[Fixup lexical-variables and lexical-functions for OpenMCL
Marco Baringer <mb at bese.it>**20050804152727
 
 This patch causes lexical-variables to no longer return ignored
 variables and symbol-macrolets. We've also implemented
 lexical-functions (though we do some hackery to convert functions
 names to something "normal" (ie SETF::|FOO::BAR| ==> (SETF FOO::BAR))
 
] 
[Typo in lexical-variables for sbcl (we were accessing lexenv-funs instead of lexenv-vars)
Marco Baringer <mb at bese.it>**20050804152051] 
[Change lexical-variables for sbcl so that it doesn't return ignored variables
Marco Baringer <mb at bese.it>**20050804150841] 
[Fix lexical-variables for cmucl to not return ignored variables
Marco Baringer <mb at bese.it>**20050804150256] 
[Typo in previous patch
Marco Baringer <mb at bese.it>**20050804150242] 
[Implement lexical-functions for cmucl
Marco Baringer <mb at bese.it>**20050804143350] 
[recognise flets from the lexical environment (on sbcl)
smoof-ra at elder-gods.org**20050803222732] 
[Rewrite multiple-value-setf so that my simple mind can understand it.
Marco Baringer <mb at bese.it>**20050803104652] 
[Added cps evaluation of THE forms
Marco Baringer <mb at bese.it>**20050803092059] 
[minor comment fixup
Marco Baringer <mb at bese.it>**20050803085322] 
[Moved defclass progv-form to keep the walker classes defined in alphabetical order 
Marco Baringer <mb at bese.it>**20050803085254] 
[Added walker class for THE forms
Marco Baringer <mb at bese.it>**20050803085210] 
[allow new special forms to be added to the walker by shadowing *walker-handlers*
smoof-ra at elder-gods.org**20050802165355] 
[Minor spacing fixs to the previous patch
Marco Baringer <mb at bese.it>**20050802152421] 
[progv
smoof-ra at elder-gods.org**20050802150342] 
[labels can have declarations inside the body
smoof-ra at elder-gods.org**20050801193433] 
[declares needs to be copied in the labels handler just like the other lambda-function-form slots
smoof-ra at elder-gods.org**20050801193107] 
[oops i forgot to actually make the declaration-form instances
smoof-ra at elder-gods.org**20050801185641] 
[initial support for remembering declarations
smoof-ra at elder-gods.org**20050801184329
 
 this patch adds a mixin called implicit-progn-with-declare-mixin and 
 uses it instead of implicit-progn-mixin in all the places that allow declares.
 It has slot which should contain a list of the declares at the top of the 
 implicit-progn.  This patch doesn't do anything clever with the declares, 
 it just creates declaration-form objects and points their source slots 
 at the original declares, however it would be easy to modify parse-declaration
 to generate more usefull declaration objects.
 
] 
[Call the property :FEATURES, not FEATURES
Marco Baringer <mb at bese.it>**20050729103229] 
[Rename asdf property version to features, add "cps-interpreter"
Marco Baringer <mb at bese.it>**20050728120238] 
[dont need to register allow-other-keys because its not a binding
smoof-ra at elder-gods.org**20050727153603] 
[fixed type name mismatch for allow-other-keys-function-arguement-form
smoof-ra at elder-gods.org**20050727152456] 
[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:
5f28ccaa2a4bdc7b3974b3e4ad0762bdf6f6e9b6

From mb at bese.it  Tue Aug 16 05:43:06 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 16 Aug 2005 07:43:06 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
References: <E1E3KWJ-0007RT-00@inquisitor.omrf.org> <m2ek8zns0v.fsf@soma.local>
	<20050815183121.GA18785@inquisitor.omrf.org>
Message-ID: <m2y872l9n9.fsf@soma.local>

Alan Shields <Alan-Shields at omrf.ouhsc.edu> writes:

> Several packages modify *features* (such as CL-PPCRE, and
> SPLIT-SEQUENCE) to include their package name. ASDF as well.

well, if split-sequence does it than why can't i? :)

> While CLTL specifies that *features* contains tokens "provided by the
> implementation", CLHS specifies that the tokens "correspond to some
> aspect of the implementation or environment." This difference would seem
> to be deliberate.
>
> I am no common lisp expert, but it would seem to me that *features*
> exists to solve exactly the problem that we're facing here: conditional
> inclusion of code based on the environment in which the code is loaded.

the issue with using *features* is mainly one of compatability. people
have often used on feature (like #+sbcl) to actually imply something
else (for example
#+sbcl-has-a-weird-internal-function-named-foo). this has lead to a
push away from the use of *features*.

> As you say yourself further in the thread: the #. solution leads to
> uglyness.

consider me convinced. i'm going to do this, but i'm going to put
:5am, not :fiveam (that logic being that i've already used up that
name for the package so no harm can come from using it for *features*
as well).

-- 
-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 Aug 16 07:34:01 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Tue, 16 Aug 2005 09:34:01 +0200
Subject: [Bese-devel] Got in troubles again
In-Reply-To: <87fytbgu5m.fsf@flarge.here> (Friedrich Dominicus's message of
	"Mon, 15 Aug 2005 16:19:01 +0200")
References: <87pssfh5es.fsf@flarge.here> <m2ll33o5on.fsf@soma.local>
	<87fytbgu5m.fsf@flarge.here>
Message-ID: <8764u6qqs6.fsf@flarge.here>

I though about it quite some time now what about a thing like the
following. As I understand a integer-field just can have one value,
the initarg name is unimporant then, because he value should end in
lisp-value. Shouldn't it be possible to write something along this
lines then?

(defclass t1 ()
  ((lisp-value :accessor lisp-value)))

(defmethod shared-initialize :after ((obj t1) slot-names &rest
                                     initargs &key &allow-other-keys)
  (setf (lisp-value obj) (cadr initargs))
  obj)

(defvar *t1-obj* nil)
(setf *t1-obj* (make-instance 't1 :num 10))

I guess one better check that initargs really consist of one element,
but I hope I could state the "base idea". 

However, I'm sure I'm missing something even in this context and
definitly in the UCW context.

Regards
Friedrich





From mb at bese.it  Tue Aug 16 08:52:07 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 16 Aug 2005 10:52:07 +0200
Subject: [Bese-devel] Got in troubles again
References: <87pssfh5es.fsf@flarge.here> <m2ll33o5on.fsf@soma.local>
	<87fytbgu5m.fsf@flarge.here> <8764u6qqs6.fsf@flarge.here>
Message-ID: <m2oe7ynu14.fsf@soma.local>

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

> I though about it quite some time now what about a thing like the
> following. As I understand a integer-field just can have one value,
> the initarg name is unimporant then, because he value should end in
> lisp-value. Shouldn't it be possible to write something along this
> lines then?
>
> (defclass t1 ()
>   ((lisp-value :accessor lisp-value)))
>
> (defmethod shared-initialize :after ((obj t1) slot-names &rest
>                                      initargs &key &allow-other-keys)
>   (setf (lisp-value obj) (cadr initargs))
>   obj)

sorry, why do you use &rest + #'cadr instead of &key?

> (defvar *t1-obj* nil)
> (setf *t1-obj* (make-instance 't1 :num 10))
>
> I guess one better check that initargs really consist of one element,
> but I hope I could state the "base idea". 

the base idea is correct, but there are situations where the
lisp-value and the client-value will be different (though this is
admittedly rare for an integer field), for example select input fields
may want an initial client-value of "---" to correspond to some
default lisp value.

> However, I'm sure I'm missing something even in this context and
> definitly in the UCW context.

no, i don't think so. i think it's just, as it currently stands, ugly.


-- 
-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 Aug 16 11:38:08 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 16 Aug 2005 13:38:08 +0200
Subject: [Bese-devel] lookingup symbols in parenscript code
Message-ID: <m2d5oem7rz.fsf@soma.local>


question: seeing as how javascript doesn't have namespaces, why does
parenscript? consider this code (assuming we've not imported the js
package):

(js:js* `(new (*abstract.*event-observer ,(lookup-js-id field))))

since NEW is not eql to JS:NEW that will get compiled into:

"new(AbstractEventObserver('foobar'))"

instead of what i wanted:

"new AbstractEventObserver('foobar')"

macros aside (since those are actually lisp functions) is there any
point to this behavior? does this code make any sense:

(js:js
  (defun ucw::foo () t)
  (defun cl-user::foo () nil))

currently it doesn't even work (both UCW::FOO and CL-USER::FOO are
converted to "foo") but it could be made to work, if someone could
convince me that it's worth it (i really don't think it is).

-- 
-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 Aug 16 11:41:55 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 16 Aug 2005 13:41:55 +0200
Subject: [Bese-devel] lookingup symbols in parenscript code
In-Reply-To: <m2d5oem7rz.fsf@soma.local> (Marco Baringer's message of "Tue,
	16 Aug 2005 13:38:08 +0200")
References: <m2d5oem7rz.fsf@soma.local>
Message-ID: <m264u6m7lo.fsf@soma.local>

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

> question: seeing as how javascript doesn't have namespaces, why does
> parenscript? consider this code (assuming we've not imported the js
> package):

sorry, that's makes no sense at all.

what i meant to say was:

question: seeing as how javascript doesn't have namespaces, why does
parenscript consider symbols in different packages to be different?
take for example this code (assuming we've not imported the js
package):

-- 
-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 alexander.kjeldaas at gmail.com  Tue Aug 16 13:11:20 2005
From: alexander.kjeldaas at gmail.com (Alexander Kjeldaas)
Date: Tue, 16 Aug 2005 15:11:20 +0200
Subject: [Bese-devel] How to use tal-templates
In-Reply-To: <m2vf27qjqk.fsf@soma.local>
References: <87zmrl95r8.fsf@flarge.here>
	<4e7bd29e05081407395c10d780@mail.gmail.com> <43005160.2070601@web.de>
	<m2hddrr43i.fsf@soma.local> <20050815091319.GA58357@pvv.ntnu.no>
	<m28xz3pmmv.fsf@soma.local> <20050815115048.GD58357@pvv.ntnu.no>
	<m2slxbmiwd.fsf@soma.local> <20050815153445.GF58357@pvv.ntnu.no>
	<m2vf27qjqk.fsf@soma.local>
Message-ID: <b2ec2dee050816061133af4cd4@mail.gmail.com>

On 8/15/05, Marco Baringer <mb at bese.it> wrote:
> astor at pvv.ntnu.no writes:
> 
> > So to me it seems like a component might want to produce:
> > o Global code (external .js-file)
> > o Per request, "per class" code in HEAD.
> 
> these two can be done with the same mechanism? if we had a method
> add-to-head-once-only:
> 
> (defmethod add-to-head-once-only ((c my-component))
>   (<:script :src "foo.js")
>   (<:meta ...)
>   (<:link ...))
> 
> > o Per request, "per instance" code in HEAD.
> 
> how does the name add-to-head sound?
>

Good.

astor


From Alan-Shields at omrf.ouhsc.edu  Tue Aug 16 18:45:23 2005
From: Alan-Shields at omrf.ouhsc.edu (Alan Shields)
Date: Tue, 16 Aug 2005 13:45:23 -0500
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
Message-ID: <E1E56R9-0002lE-00@inquisitor.omrf.org>

Tue Aug 16 13:44:54 CDT 2005  Alan Shields <Alan-Shields at omrf.ouhsc.edu>
  * Add :fiveam to *features*
  I've found that I prefer to add testing code in my other code, but I dislike
  adding :fiveam to my package requirements list.
  
  This allows users of fiveam to delineate testing code with #+fiveam() - whereupon
  it will be totally ignored by those who do not have fiveam.
  
  Happiness for all and love ensue.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/x-darcs-patch
Size: 1144 bytes
Desc: A darcs patch for your repository!
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050816/953e4ddb/attachment.bin>

From mb at bese.it  Tue Aug 16 19:00:30 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 16 Aug 2005 21:00:30 +0200
Subject: [Bese-devel] darcs patch: Add :fiveam to *features*
In-Reply-To: <E1E56R9-0002lE-00@inquisitor.omrf.org> (Alan Shields's message
	of "Tue, 16 Aug 2005 13:45:23 -0500")
References: <E1E56R9-0002lE-00@inquisitor.omrf.org>
Message-ID: <m2slx9u2pd.fsf@soma.local>

Alan Shields <Alan-Shields at omrf.ouhsc.edu> writes:

> Tue Aug 16 13:44:54 CDT 2005  Alan Shields <Alan-Shields at omrf.ouhsc.edu>
>   * Add :fiveam to *features*
>   I've found that I prefer to add testing code in my other code, but I dislike
>   adding :fiveam to my package requirements list.
>   
>   This allows users of fiveam to delineate testing code with #+fiveam() - whereupon
>   it will be totally ignored by those who do not have fiveam.
>   
>   Happiness for all and love ensue.

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 tim at fractaldragon.net  Tue Aug 16 19:14:19 2005
From: tim at fractaldragon.net (Tim Lavoie)
Date: Tue, 16 Aug 2005 14:14:19 -0500
Subject: [Bese-devel] YACML needs join-strings-return-value
Message-ID: <20050816191419.GA19136@fractaldragon.net>

Hi all,

I've just updated my local darcs repositories for YACML, arnesi and
UCW, and am now running into the following problem.

  YACLML requires the join-strings-return-value feature of system :ARNESI.
:ARNESI currently only provides ("v1.4.0" "v1.4.1" "cps-interpreter").
Try pull'ing the latest arnesi or send an email to bese-devel at common-lisp.net

Am I just a little out of sync on these, or have I mangled something else?

	Thanks,
	Tim

-- 
"Some people's idea of free speech is that they are free to say
anything they like, but if anyone says anything back, that is an
outrage."
    -- Winston Churchill


From alex.mizrahi at gmail.com  Tue Aug 16 17:21:05 2005
From: alex.mizrahi at gmail.com (Alex Mizrahi)
Date: Tue, 16 Aug 2005 20:21:05 +0300
Subject: [Bese-devel] arnesi version mismatch?
Message-ID: <007001c5a286$e430d4d0$0e3705c3@ALEXCOMP>

hello

i was going to try latest version of uncommon web, but it failed to load:

 YACLML requires the join-strings-return-value feature of system :ARNESI.
:ARNESI currently only provides ("v1.4.0" "v1.4.1" "cps-interpreter").
Try pull'ing the latest arnesi or send an email to
bese-devel at common-lisp.net

with latest libs from darcs repository.
are there some problems on my system or it's inconsistent state of
repositories?

with best regards, Alex 'killer_storm' Mizrahi.



From bobstopper at bobturf.org  Tue Aug 16 23:28:48 2005
From: bobstopper at bobturf.org (Robert Marlow)
Date: Wed, 17 Aug 2005 07:28:48 +0800
Subject: [Bese-devel] arnesi version mismatch?
In-Reply-To: <007001c5a286$e430d4d0$0e3705c3@ALEXCOMP>
References: <007001c5a286$e430d4d0$0e3705c3@ALEXCOMP>
Message-ID: <87k6iljwb3.wl%bobstopper@bobturf.org>

I got this too. Try pulling from
http://common-lisp.net/project/ucw/repos/ucw_dev/
and
http://common-lisp.net/project/bese/repos/arnesi_dev/
instead. That's how I got around it.


At Tue, 16 Aug 2005 20:21:05 +0300,
Alex Mizrahi wrote:
> 
> hello
> 
> i was going to try latest version of uncommon web, but it failed to load:
> 
>  YACLML requires the join-strings-return-value feature of system :ARNESI.
> :ARNESI currently only provides ("v1.4.0" "v1.4.1" "cps-interpreter").
> Try pull'ing the latest arnesi or send an email to
> bese-devel at common-lisp.net
> 
> with latest libs from darcs repository.
> are there some problems on my system or it's inconsistent state of
> repositories?
> 
> with best regards, Alex 'killer_storm' Mizrahi.
> 
> _______________________________________________
> 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 Aug 17 05:24:45 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 17 Aug 2005 07:24:45 +0200
Subject: [Bese-devel] YACML needs join-strings-return-value
In-Reply-To: <20050816191419.GA19136@fractaldragon.net> (Tim Lavoie's message
	of "Tue, 16 Aug 2005 14:14:19 -0500")
References: <20050816191419.GA19136@fractaldragon.net>
Message-ID: <m21x4tqgo2.fsf@soma.local>

Tim Lavoie <tim at fractaldragon.net> writes:

> Hi all,
>
> I've just updated my local darcs repositories for YACML, arnesi and
> UCW, and am now running into the following problem.
>
>   YACLML requires the join-strings-return-value feature of system :ARNESI.
> :ARNESI currently only provides ("v1.4.0" "v1.4.1" "cps-interpreter").
> Try pull'ing the latest arnesi or send an email to bese-devel at common-lisp.net

i think you're just pulling from arnesi and not arnesi_dev. i plan on
moving the _dev branches (which work) over to the stable branches
(which are broken) Real Soon Now(TM). in the mean time make sure
you're pulling from:

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

-- 
-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 bobstopper at bobturf.org  Wed Aug 17 05:44:39 2005
From: bobstopper at bobturf.org (bobstopper at bobturf.org)
Date: Wed, 17 Aug 2005 13:44:39 +0800
Subject: [Bese-devel] darcs patch: removall of *s from tal-environment
Message-ID: <20050817054439.50C31CB9ED@spengler.bobturf.org>

Wed Aug 17 13:43:29 WST 2005  bobstopper at bobturf.org
  * removall of *s from tal-environment
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/x-darcs-patch
Size: 2088 bytes
Desc: A darcs patch for your repository!
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050817/73b9be0d/attachment.bin>

From mb at bese.it  Wed Aug 17 10:37:45 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 17 Aug 2005 12:37:45 +0200
Subject: [Bese-devel] ucw coolness story
Message-ID: <m2psscj1c6.fsf@soma.local>


hi,

  i'm sure you all know this already, but it really made me smile:

  i'm currently working on an app for managing the patients in a
hospital. one of the (many, many, many) things it needs to do is
generate statistics, one of those statistics requires that you chose a
private clinic from a list or type the name into a text box. i
initially had these two actions:

  (defaction dati-fatturazione ((c component) (istituto-name string))
    (dati-fatturazione c (find-istituto istituto-name)))

  (defaction dati-fatturazione ((c component) (istituto istituto))
    ...)

  (sorry for the weird mix of english and italian. 'istituto' is what
the app calls the private clinics and dati-fatturazione basically
means billing info.)

  one of the things that kept coming up was that, when using the 'type
the name' form people would misspell the name of the clinic, which
would trigger an error page since find-istituto returned NIL. so i
changed the first method to this:

  (defaction dati-fatturazione ((c component) (istituto-name string))
    (aif (find-istituto istituto-name)
         (dati-fatturazione c it)
         (call 'info-message :message "No clinic with that name.")))

  then we decided i'd be nice if the app tried to guess what clinic
they wanted, so the action became:

  (defaction dati-fatturazione ((c component) (istituto-name string))
    (aif (find-istituto istituto-name)
         (dati-fatturazione c it)
         (aif (possible-istituto istituto-name)
              (dati-fatturazione c it)
              (call 'info-message :message "No clinic with that name."))))

  then we decided that, since some ambiguity can result, it'd be nice
to show them the list of possible clinics and let them chose:

  (defaction dati-fatturazione ((c component) (istituto-name string))
    (aif (find-istituto istituto-name)
         (dati-fatturazione c it)
         (let ((possibilities (possible-istituti istituto-name)))
           (case (length possibilities)
             (0 (call 'info-message :message "Nessun istituto con quel nome."))
             (1 (dati-fatturazione c (first possibilities)))
             (t (dati-fatturazione
                 c
                 (call 'option-dialog
                       :message (format nil "Nessun istituto col nome ~S"
                                            istituto-name)
                       :options (mapcar (lambda (istituto)
                                          (cons istituto
                                                (format nil "Dati per l'istituto ~A."
                                                            (nome istituto))))
                                        possibilities))))))))

  notice that in the n possibilities case we still call
dati-fatturazione but we get the istituto from an option-dialog
component.  

  anyway, my point: sure i could have done that in any web framework
but i don't think i would have if it hadn't been so easy. some things
need to be _easy_, and not just possible, in order to take advantage
of them.

p.s. - among other things i've been doing some java work and trust me,
nothing hurts as much as xml configuration files and
edit-wait-compile-wait-debug-wait when you've tasted lisp and ucw (and
slime).

p.p.s. - yes, i'm bragging about something i wrote. sue me.

-- 
-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  Wed Aug 17 12:56:28 2005
From: der_julian at web.de (Julian Stecklina)
Date: Wed, 17 Aug 2005 14:56:28 +0200
Subject: [Bese-devel] ucw coolness story
In-Reply-To: <m2psscj1c6.fsf@soma.local>
References: <m2psscj1c6.fsf@soma.local>
Message-ID: <20050817145628.069eb7ff@localhost>

On Wed, 17 Aug 2005 12:37:45 +0200
"Marco Baringer" <mb at bese.it> wrote:

> p.p.s. - yes, i'm bragging about something i wrote. sue me.

But you surely have a point. :)

Regards,
-- 
Julian Stecklina

(Of course SML does have its weaknesses, but by comparison, a
 discussion of C++'s strengths and flaws always sounds like an
 argument about whether one should face north or east when one
 is sacrificing one's goat to the rain god.)         -- Thant Tessman


From pjb at informatimago.com  Fri Aug 19 01:47:58 2005
From: pjb at informatimago.com (Pascal Bourguignon)
Date: Fri, 19 Aug 2005 03:47:58 +0200 (CEST)
Subject: [Bese-devel] enctype multipart/form-data hangs
Message-ID: <20050819014758.E26FB10FBF3@thalassa.informatimago.com>

Hello,

When I specify:
 (<ucw:form :accept-charset "utf-8" :enctype "multipart/form-data" ...)
the server hangs in read-request:


2005-08-04T02:56.10 +WARN+ UCW-LOGGER: Attempting to publish "../wwwroot/html/" at "/app/bellerophon/" but mod_lisp backend does not support publish-directory.
2005-08-04T02:56.10 +INFO+ UCW-LOGGER: Starting up standard server #<STANDARD-SERVER MOD-LISP-BACKEND 3 #x20553DFE>.
1. Trace: 
(IT.BESE.UCW::READ-REQUEST '#<MOD-LISP-BACKEND #x20553DC6>
 '#<IO UNBUFFERED SOCKET-STREAM CHARACTER 0.0.0.0:3001>)
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: Reading request from #<IO UNBUFFERED SOCKET-STREAM CHARACTER 0.0.0.0:3001>
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "content-length"="689"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "content-type"="multipart/form-data; boundary=---------------------------90641591112385285421222996919"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "remote-ip-addr"="62.93.174.79"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "remote-ip-port"="49423"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "script-filename"="/local/html/local/app"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "server-ip-addr"="62.93.174.79"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "server-ip-port"="80"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "server-protocol"="HTTP/1.1"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "method"="POST"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "url"="/app/bellerophon/"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "server-id"="bellerophon"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "server-baseversion"="Apache/1.3.27"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "modlisp-version"="2.42"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Accept"="text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Accept-Charset"="ISO-8859-1,utf-8;q=0.7,*;q=0.7"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Accept-Encoding"="gzip,deflate"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Accept-Language"="en-us,en;q=0.9,fr;q=0.7,es;q=0.6,pt;q=0.4,ca;q=0.3,de;q=0.1"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Connection"="keep-alive"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Content-Length"="689"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Content-Type"="multipart/form-data; boundary=---------------------------90641591112385285421222996919"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Cookie"="ucw-session-id=chsuiNZkEyjfeyOyOTZexwfMEwvrlLCJBaFpldek"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Host"="thalassa.informatimago.com"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Keep-Alive"="300"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "Referer"="http://thalassa.informatimago.com/app/bellerophon/"
2005-08-04T02:56.19 +DRIBBLE+ UCW-LOGGER: "User-Agent"="Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7) Gecko/20040713"
;; hangs here.
C-c C-c



But when I remove :enctype: (<ucw:form :accept-charset "utf-8" ...)
then it works:

1. Trace: 
(IT.BESE.UCW::READ-REQUEST '#<MOD-LISP-BACKEND #x205510F6>
 '#<IO UNBUFFERED SOCKET-STREAM CHARACTER 0.0.0.0:3001>)
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: Reading request from #<IO UNBUFFERED SOCKET-STREAM CHARACTER 0.0.0.0:3001>
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "content-length"="78"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "content-type"="application/x-www-form-urlencoded"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "remote-ip-addr"="192.168.0.64"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "remote-ip-port"="49550"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "script-filename"="/local/html/local/app"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "server-ip-addr"="62.93.174.79"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "server-ip-port"="80"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "server-protocol"="HTTP/1.1"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "method"="POST"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "url"="/app/bellerophon/"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "server-id"="bellerophon"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "server-baseversion"="Apache/1.3.27"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "modlisp-version"="2.42"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Accept"="*/*"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Accept-Encoding"="gzip, deflate"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Accept-Language"="en"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Connection"="close"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Content-Length"="78"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Content-Type"="application/x-www-form-urlencoded"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Cookie"="ucw-session-id=wKpectxuxvEFJVNNeKKbqhNzBIceMmbVhXLwAcck"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Host"="thalassa.informatimago.com"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "Referer"="http://thalassa.informatimago.com/app/bellerophon/"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: "User-Agent"="Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en) AppleWebKit/312.1.1 (KHTML, like Gecko) Safari/312"
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: Body: s=wKpectxuxvEFJVNNeKKbqhNzBIceMmbVhXLwAcck&f=YqmRmtUMXbqakxkoYYbM&a=rVJKgVJpjM
1. Trace: IT.BESE.UCW::READ-REQUEST ==> #<IT.BESE.UCW::MOD-LISP-REQUEST #x20861F46>
2005-08-04T03:25.54 +INFO+ UCW-LOGGER: Serving action UCW::SWITCH-COMPONENT in session "wKpectxuxvEFJVNNeKKbqhNzBIceMmbVhXLwAcck".
#### #<MENU-COMPONENT #x2079F8BE>
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: Sending mod-lisp response.
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: Sending ----
2005-08-04T03:25.54 +DRIBBLE+ UCW-LOGGER: 
<div style="border: 2px solid #000000;"
  ><!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
      >Utilisateur BELLEROPHON</title
    ><link href="stylesheet.css" rel="stylesheet" type="text/css"
/></head
  >


I hoped to be able to use "multipart/form-data" to get better utf-8
encoding support...

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/

The world will now reboot.  don't bother saving your artefacts.


From mb at bese.it  Fri Aug 19 11:09:45 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 19 Aug 2005 13:09:45 +0200
Subject: [Bese-devel] enctype multipart/form-data hangs
References: <20050819014758.E26FB10FBF3@thalassa.informatimago.com>
Message-ID: <m2iry2p4hy.fsf@soma.local>

Pascal Bourguignon <pjb at informatimago.com> writes:

> I hoped to be able to use "multipart/form-data" to get better utf-8
> encoding support...

i do not believe this a utf-8 issue. i took the form example and
switched it over to multipart/form-data, while i don't get the hanging
problems you have all of the form data is getting lost somewhere down
in rfc2388 :(

-- 
-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  Fri Aug 19 11:36:04 2005
From: jdz at dir.lv (Janis Dzerins)
Date: Fri, 19 Aug 2005 14:36:04 +0300
Subject: [Bese-devel] enctype multipart/form-data hangs
In-Reply-To: <m2iry2p4hy.fsf@soma.local>
References: <20050819014758.E26FB10FBF3@thalassa.informatimago.com>
	<m2iry2p4hy.fsf@soma.local>
Message-ID: <4305C424.3030109@dir.lv>

Marco Baringer wrote:
> Pascal Bourguignon <pjb at informatimago.com> writes:
> 
> 
>>I hoped to be able to use "multipart/form-data" to get better utf-8
>>encoding support...
> 
> 
> i do not believe this a utf-8 issue. i took the form example and
> switched it over to multipart/form-data, while i don't get the hanging
> problems you have all of the form data is getting lost somewhere down
> in rfc2388 :(
> 

First, be sure to update to the latest release of rfc2388.

If it the problem persists, I'd be glad to test / correct it -- I would 
need a test case (self-contained preferrably) that would exhibit this 
problem.

--
Janis Dzerins


From rusabd at gmail.com  Fri Aug 19 07:56:06 2005
From: rusabd at gmail.com (Ruslan Abdulkhalikov)
Date: Fri, 19 Aug 2005 13:56:06 +0600
Subject: [Bese-devel] port #1
Message-ID: <2f11360405081900562e9c0195@mail.gmail.com>

Hello ladies and gentlemen.

I've been using UCW for few weeks and found that patches are applied
on _dev repository frequently...   So, could you suggest me some CL
implementation which is checked first after OpenMCL? CMUCL  SBCL,
CLISP, Allegro? Actually SBCL works almost fine for my very simple
examples but I failed to make work UCW against CLISP on Windows...

Thanks.

Best Regards,
Ruslan


From frido at q-software-solutions.de  Sat Aug 20 07:08:51 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 09:08:51 +0200
Subject: [Bese-devel] another question about the ucw way
Message-ID: <87y86xrsos.fsf@flarge.here>

Ok, I'm back again ;-(, let me introduce a simple example
assume you'd ahve the class
(defclass foo ()
          ((num-1 ...)
           (num-2 ...)
           ... some other slots))


Now you like to present it so you write a render-on method.
(defmethod render-on ((res response) (foo foo))
.... 
all you need to render that class.

Now you want to be able to edit the slots of the class foo. So you'd
need another render-on method but you still have the first one, so
what are you supposed to do? 

it may be that my understanding of ucw is so sparse (despite the fact
that I spend now at least 3 weeks on it), that I can't see the obvious
but for me there is no way to write different render-on
methods for one class. On the other hand even if there would I see
that  lot of code would have to be duplicated, but the difference is
just:
one time you like to have an editable view and the other time you just
want a "read-only" view. 

The clumsy workaround I think I have to take is
declare diverse classes
(defclass foo ()
....

(defclass editable-foo()....

and the like. However I would need to copy the slots over to the new
class just for the filling the slots and write another method. So I
had to write nearly identical classes over and over again and copy
values from one class to another. This can't be the "right" solution. 

A solution might be adding another parameter to the render-on method
which is named view. Of course one would have to write specialiced
methods for the different views but one would not have to invent
things like editable-foo. 

The render-on method would then look like
(defgeneric render-on (response class view))
(defmethod render-on ((res response) (foo foo) (editable-view
editable-view))

What is your opinion about that? 
Have I missed the obvious ucw way?


Regards
Friedrich


From whalliburton at gmail.com  Thu Aug 18 02:47:33 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Wed, 17 Aug 2005 21:47:33 -0500
Subject: [Bese-devel] Sbcl disassembly pretty printer
Message-ID: <4e7bd29e05081719475eb0d370@mail.gmail.com>

Hello. You wanted a copy of the sbcl disassembly pretty printer. I
have attached the files.

The disassembler will stay permanently at
http://licentiae.com/diss.ucw from now out.  I hope to put annotation
based on pattern matching.

I used displaced arrays all over the place in order to not copy the
text data, is this practice ok or do you know of possible problems? Do
you know of a more succinct way to map-over-lines?

 Will
-------------- next part --------------
A non-text attachment was scrubbed...
Name: disassemble-component.lisp
Type: application/octet-stream
Size: 7082 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050817/420e77ec/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nasm_doc.lisp
Type: application/octet-stream
Size: 25309 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050817/420e77ec/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: diss.css
Type: text/css
Size: 874 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050817/420e77ec/attachment.css>

From mb at bese.it  Sat Aug 20 10:47:40 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 20 Aug 2005 12:47:40 +0200
Subject: [Bese-devel] another question about the ucw way
In-Reply-To: <87y86xrsos.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sat, 20 Aug 2005 09:08:51 +0200")
References: <87y86xrsos.fsf@flarge.here>
Message-ID: <m2acjcyjeb.fsf@soma.local>

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

> Now you want to be able to edit the slots of the class foo. So you'd
> need another render-on method but you still have the first one, so
> what are you supposed to do? 

ucw does not, directly, deal with this issue. the only thing we have
so far are the presentations (which works but i don't particularly
like even though i've been used them to develop some very complex
apps) and drew's lisp-on-lines (darcs get
http://tech.coop/darcs/lisp-on-lines).

if you do come up with something you like i'd love to hear about 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 Aug 20 10:58:19 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 12:58:19 +0200
Subject: [Bese-devel] renamed parenscript?
Message-ID: <87wtmgj2no.fsf@flarge.here>

Well if you check out the latest ucw_dev stuff you'll find a
parenscript in the modules to have, it seem this has to be renamed to 
klammerscript. 

Regards
Friedrich


From mb at bese.it  Sat Aug 20 11:02:57 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 20 Aug 2005 13:02:57 +0200
Subject: [Bese-devel] port #1
In-Reply-To: <2f11360405081900562e9c0195@mail.gmail.com> (Ruslan
	Abdulkhalikov's message of "Fri, 19 Aug 2005 13:56:06 +0600")
References: <2f11360405081900562e9c0195@mail.gmail.com>
Message-ID: <m264u0yiou.fsf@soma.local>

Ruslan Abdulkhalikov <rusabd at gmail.com> writes:

> Hello ladies and gentlemen.
>
> I've been using UCW for few weeks and found that patches are applied
> on _dev repository frequently...   So, could you suggest me some CL
> implementation which is checked first after OpenMCL? CMUCL  SBCL,
> CLISP, Allegro? Actually SBCL works almost fine for my very simple
> examples but I failed to make work UCW against CLISP on Windows...

quite a few people use _dev with sbcl. i generally test on sbcl then
cmucl then clisp, though lately i've been testing on openmcl and clisp
only.

i do not have easy access to a windows machine, what issues are you
running into?

-- 
-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 Aug 20 11:37:14 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 13:37:14 +0200
Subject: [Bese-devel] small patch to ucw-tags
Message-ID: <87pss8j0ut.fsf@flarge.here>

I found the following typo in the mentioned fiel here's a patch. 

Regards
Friedrich

--- ucw-tags.lisp.old	2005-08-20 13:35:15.000000000 +0200
+++ ucw-tags.lisp	2005-08-20 13:33:17.000000000 +0200
@@ -231,7 +231,7 @@
 (deftag-macro <ucw:script (&body body)
   `(<:script :type "text/javascript"
              (<:as-is ~% "// <![CDATA[" ~%
-                      (js:js* , at body)
+                      (js:js , at body)
                       ~% "// ]]>" ~%)))
 
 ;; Copyright (c) 2003-2005 Edward Marco Baringer


From frido at q-software-solutions.de  Sat Aug 20 12:14:57 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 14:14:57 +0200
Subject: [Bese-devel] another patch
Message-ID: <87ll2wiz3y.fsf@flarge.here>

for error.lisp

Regards
Friedrich

--- error.lisp.2005-08-20	2005-08-20 13:58:00.000000000 +0200
+++ error.lisp	2005-08-20 13:58:07.000000000 +0200
@@ -52,7 +52,7 @@
      for index upfrom 0
      for div-id = (concatenate 'string "frame-" (princ-to-string index) "-details")
      do (<:div
-	  (<:p (<:button :onclick (js:js-inline* `(toggle-display ,div-id)) "[ + ]")
+	  (<:p (<:button :onclick (js:js-inline `(toggle-display ,div-id)) "[ + ]")
 	       " "
 	       (<:as-html
 		 (let ((desc (backtrace-frame-description frame))



From mb at bese.it  Sat Aug 20 12:45:40 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 20 Aug 2005 14:45:40 +0200
Subject: [Bese-devel] small patch to ucw-tags
In-Reply-To: <87pss8j0ut.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sat, 20 Aug 2005 13:37:14 +0200")
References: <87pss8j0ut.fsf@flarge.here>
Message-ID: <m2u0hkwzd7.fsf@soma.local>

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

> --- ucw-tags.lisp.old	2005-08-20 13:35:15.000000000 +0200
> +++ ucw-tags.lisp	2005-08-20 13:33:17.000000000 +0200
> @@ -231,7 +231,7 @@
>  (deftag-macro <ucw:script (&body body)
>    `(<:script :type "text/javascript"
>               (<:as-is ~% "// <![CDATA[" ~%
> -                      (js:js* , at body)
> +                      (js:js , at body)
>                        ~% "// ]]>" ~%)))

this isn't a feature, not a bug :)

if you use js:js there's no way to inject data into the generated
javascript like this:

(with-unique-js-name (my-func)
  (<ucw:script
    `(defun ,my-func (x)
       (do-stuff-with-x)))
  ...
  (<:input :on-change (js:js-inline* `(,my-func this.value))))

though maybe it would be better to follow the js naming convention and
have <ucw:script and <ucw:script*.

-- 
-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 Aug 20 12:46:48 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 20 Aug 2005 14:46:48 +0200
Subject: [Bese-devel] small patch to ucw-tags
In-Reply-To: <m2u0hkwzd7.fsf@soma.local> (Marco Baringer's message of "Sat,
	20 Aug 2005 14:45:40 +0200")
References: <87pss8j0ut.fsf@flarge.here> <m2u0hkwzd7.fsf@soma.local>
Message-ID: <m2pss8wzbb.fsf@soma.local>

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

> this isn't a feature, not a bug :)

this _is_ a feature, not a 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  Sat Aug 20 12:48:15 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 20 Aug 2005 14:48:15 +0200
Subject: [Bese-devel] renamed parenscript?
In-Reply-To: <87wtmgj2no.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sat, 20 Aug 2005 12:58:19 +0200")
References: <87wtmgj2no.fsf@flarge.here>
Message-ID: <m2ll2wwz8w.fsf@soma.local>

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

> Well if you check out the latest ucw_dev stuff you'll find a
> parenscript in the modules to have, it seem this has to be renamed to 
> klammerscript. 

sorry, i don't understand.

-- 
-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  Sat Aug 20 13:43:00 2005
From: der_julian at web.de (Julian Stecklina)
Date: Sat, 20 Aug 2005 15:43:00 +0200
Subject: [Bese-devel] Sbcl disassembly pretty printer
In-Reply-To: <4e7bd29e05081719475eb0d370@mail.gmail.com>
References: <4e7bd29e05081719475eb0d370@mail.gmail.com>
Message-ID: <20050820154300.10bf21ec@localhost>

On Wed, 17 Aug 2005 21:47:33 -0500
William Halliburton <whalliburton at gmail.com> wrote:

> The disassembler will stay permanently at
> http://licentiae.com/diss.ucw from now out.  I hope to put annotation
> based on pattern matching.

Bad Gateway

The proxy server received an invalid response from an upstream server.

??

Regards,
-- 
Julian Stecklina

(Of course SML does have its weaknesses, but by comparison, a
 discussion of C++'s strengths and flaws always sounds like an
 argument about whether one should face north or east when one
 is sacrificing one's goat to the rain god.)         -- Thant Tessman


From frido at q-software-solutions.de  Sat Aug 20 14:39:27 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 16:39:27 +0200
Subject: [Bese-devel] renamed parenscript?
In-Reply-To: <m2ll2wwz8w.fsf@soma.local> (Marco Baringer's message of "Sat,
	20 Aug 2005 14:48:15 +0200")
References: <87wtmgj2no.fsf@flarge.here> <m2ll2wwz8w.fsf@soma.local>
Message-ID: <87hddkisf4.fsf@flarge.here>

sorry for beeing unclear the ucw.asd files says:

  :depends-on (:arnesi :yaclml :swank :iterate :rfc2388 :klammerscript))
(in fact it says :parenscript at the end but after install
parenscript via asdf-install you read in the asd file

(defpackage :klammerscript.system
  (:use :cl :asdf))

(in-package :klammerscript.system)

(defsystem :klammerscript


...
 

so the name seems to be klammerscript, at least with the changes in
ucw I got this compiled.

Regards
Friedrich


From frido at q-software-solutions.de  Sat Aug 20 14:45:15 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 16:45:15 +0200
Subject: [Bese-devel] small patch to ucw-tags
In-Reply-To: <m2u0hkwzd7.fsf@soma.local> (Marco Baringer's message of "Sat,
	20 Aug 2005 14:45:40 +0200")
References: <87pss8j0ut.fsf@flarge.here> <m2u0hkwzd7.fsf@soma.local>
Message-ID: <87d5o8is5g.fsf@flarge.here>

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

>   (<:input :on-change (js:js-inline* `(,my-func this.value))))
>
> though maybe it would be better to follow the js naming convention and
> have <ucw:script and <ucw:script*.

could you please tell me where it is defined then? 


grep -n 'js\*' parenscript/css.lisp parenscript/introduction.lisp parenscript/js-html.lisp parenscript/js.lisp parenscript/macros.lisp parenscript/package.lisp parenscript/reference.lisp parenscript/tutorial.lisp parenscript/utils.lisp yaclml/**/*.lisp arnesi/**/*.lisp

no output.

the same is true here for the other script is there another
parenscript? 

Regards
Friedrich




From mb at bese.it  Sat Aug 20 15:22:44 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 20 Aug 2005 17:22:44 +0200
Subject: [Bese-devel] small patch to ucw-tags
In-Reply-To: <87d5o8is5g.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sat, 20 Aug 2005 16:45:15 +0200")
References: <87pss8j0ut.fsf@flarge.here> <m2u0hkwzd7.fsf@soma.local>
	<87d5o8is5g.fsf@flarge.here>
Message-ID: <m2acjcws3f.fsf@soma.local>

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

> "Marco Baringer" <mb at bese.it> writes:
>
>>   (<:input :on-change (js:js-inline* `(,my-func this.value))))
> >
>> though maybe it would be better to follow the js naming convention and
>> have <ucw:script and <ucw:script*.
>
> could you please tell me where it is defined then? 

i believe you're using manuel's 0.1.0 release. you'll need to use my
branch:

darcs get http://common-lisp.net/project/ucw/repos/parenscript/

this contains the rename from klammerscript to parenscript, the js*
macros, and a few other minor changes.

-- 
-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 Aug 20 17:02:18 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 19:02:18 +0200
Subject: [Bese-devel] small patch to ucw-tags
In-Reply-To: <m2acjcws3f.fsf@soma.local> (Marco Baringer's message of "Sat,
	20 Aug 2005 17:22:44 +0200")
References: <87pss8j0ut.fsf@flarge.here> <m2u0hkwzd7.fsf@soma.local>
	<87d5o8is5g.fsf@flarge.here> <m2acjcws3f.fsf@soma.local>
Message-ID: <873bp4ilt1.fsf@flarge.here>

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

>
> darcs get http://common-lisp.net/project/ucw/repos/parenscript/
>
> this contains the rename from klammerscript to parenscript, the js*
> macros, and a few other minor changes.
You're right about that, but how should I know?

Regards
Friedrich


From mb at bese.it  Sat Aug 20 17:10:21 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 20 Aug 2005 19:10:21 +0200
Subject: [Bese-devel] small patch to ucw-tags
In-Reply-To: <873bp4ilt1.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sat, 20 Aug 2005 19:02:18 +0200")
References: <87pss8j0ut.fsf@flarge.here> <m2u0hkwzd7.fsf@soma.local>
	<87d5o8is5g.fsf@flarge.here> <m2acjcws3f.fsf@soma.local>
	<873bp4ilt1.fsf@flarge.here>
Message-ID: <m24q9kwn42.fsf@soma.local>

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

> "Marco Baringer" <mb at bese.it> writes:
>
>>
>> darcs get http://common-lisp.net/project/ucw/repos/parenscript/
>>
>> this contains the rename from klammerscript to parenscript, the js*
>> macros, and a few other minor changes.
> You're right about that, but how should I know?

i sent an email to the list:

http://common-lisp.net/pipermail/bese-devel/2005-August/000789.html

and i added a note to the README. Since manuel has given me his 'do
what with it' approval i will update the cliki page as well to point
to this new (and maintained) 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 frido at q-software-solutions.de  Sat Aug 20 17:17:37 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sat, 20 Aug 2005 19:17:37 +0200
Subject: [Bese-devel] small patch to ucw-tags
In-Reply-To: <m24q9kwn42.fsf@soma.local> (Marco Baringer's message of "Sat,
	20 Aug 2005 19:10:21 +0200")
References: <87pss8j0ut.fsf@flarge.here> <m2u0hkwzd7.fsf@soma.local>
	<87d5o8is5g.fsf@flarge.here> <m2acjcws3f.fsf@soma.local>
	<873bp4ilt1.fsf@flarge.here> <m24q9kwn42.fsf@soma.local>
Message-ID: <87vf20h6j2.fsf@flarge.here>

Ok, I did not have seen that mail. Me bad. Will try to make it better
in the future.

Regards
Friedrich


From whalliburton at gmail.com  Sat Aug 20 20:46:48 2005
From: whalliburton at gmail.com (William Halliburton)
Date: Sat, 20 Aug 2005 15:46:48 -0500
Subject: [Bese-devel] Sbcl disassembly pretty printer
In-Reply-To: <20050820154300.10bf21ec@localhost>
References: <4e7bd29e05081719475eb0d370@mail.gmail.com>
	<20050820154300.10bf21ec@localhost>
Message-ID: <4e7bd29e05082013464feb274@mail.gmail.com>

Server was rebooted by provider for Xen upgrade. Sorry. Have not made
sbcl start on boot. Should be up now. Will add it to boot sequence.

Are there any tricks to adding sbcl/ucw to linux boot sequence? Just
an open ended question, as I do not think I will have problems.

I have added regular expressions for the instructions. (see blue
comment line at wch.web::foo). If anyone has any similar
hacks/knowledge, please send them my way.

Will

On 8/20/05, Julian Stecklina <der_julian at web.de> wrote:
> On Wed, 17 Aug 2005 21:47:33 -0500
> William Halliburton <whalliburton at gmail.com> wrote:
> 
> > The disassembler will stay permanently at
> > http://licentiae.com/diss.ucw from now out.  I hope to put annotation
> > based on pattern matching.
> 
> Bad Gateway
> 
> The proxy server received an invalid response from an upstream server.
> 
> ??
> 
> Regards,
> --
> Julian Stecklina
> 
> (Of course SML does have its weaknesses, but by comparison, a
>  discussion of C++'s strengths and flaws always sounds like an
>  argument about whether one should face north or east when one
>  is sacrificing one's goat to the rain god.)         -- Thant Tessman
> _______________________________________________
> bese-devel mailing list
> bese-devel at common-lisp.net
> http://common-lisp.net/cgi-bin/mailman/listinfo/bese-devel
>


From frido at q-software-solutions.de  Sun Aug 21 03:03:24 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sun, 21 Aug 2005 05:03:24 +0200
Subject: [Bese-devel] general question of understanding
Message-ID: <87br3s6lfn.fsf@flarge.here>

Well ucw is way too much a mystery too me, as you've seen but giving
up is not what I like to do. So I have a general question
Do I understand it correctly that creating of a component can not be
done with make-instance?

Regards
Friedrich


From frido at q-software-solutions.de  Sun Aug 21 03:28:44 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Sun, 21 Aug 2005 05:28:44 +0200
Subject: [Bese-devel] what's wrong with the following code?
Message-ID: <874q9k6k9f.fsf@flarge.here>

(defaction shop-login ((login shop-login) username password)
  (setf (login.username login) username
        (login.password login) password)
  (let* ((user-data (check-credentials login))
         (shipping-adr (make-instance 'shipping-adr :adr (car (qss.db::shipping-address user-data)))))
        ;; car is needed because a join slot always returns a list!     
    (inspect login)
    (inspect shipping-adr)
  (if shipping-adr
      (call 'customer-input-page
            :customer-data user-data
            :address shipping-adr
            :payment-details nil)
      (call 'info-message 
            :message "Login failed"))))

The shipping-adr class is defined this way:
(defcomponent address-mixin ()
  ((adr :initarg :adr :accessor adr
        :backtrack t
        :initform (make-instance 'qss.db::shipping-address)))
  (:metaclass standard-component-class))

;; (forget-class :shipping-adr)
(defclass shipping-adr (address-mixin)
  ()
  (:metaclass standard-component-class))

If I inspect login I see the following:
0. CONTINUATION: (IT.BESE.ARNESI::TOPLEVEL-K)
1. CALLING-COMPONENT: #<TRANS-INFO {1001F24851}>
2. PLACE: #<IT.BESE.UCW::PLACE (CONTEXT.WINDOW-COMPONENT
                                *CONTEXT*) {1001391471}>

that's ok, the shipping-adr has the following
0. CONTINUATION: "unbound"
1. CALLING-COMPONENT: "unbound"
2. PLACE: "unbound"

which seems itself reasonable, but then I do the following
(defmethod render-on ((res response) (customer-page customer-input-page))
  (<:h1 "Your adress data: ")
  (render-on res (address customer-page)))

which then calls this:
(defmethod render-on ((res response) (adr shipping-adr))
  (let ((address (adr adr)))
    (inspect adr)
    (<ucw:form :action (edit-address adr)
               (<:table 
                (dolist (item '(street zip city country email))
                  (present-a-slot-as-table-row item (slot-value address item))))
               (<:p
                (<ucw:submit :action (edit-address adr)
                             :value "Edit")))))

This is rendered fine.

But then I push the button and got:
The slot IT.BESE.UCW::PLACE is unbound in the object #<SHIPPING-ADR {10017C2171}>.
   [Condition of type UNBOUND-SLOT]

Restarts:
  0: [SHOW-BACKTRACE] Send the client a backtrace page.
  1: [RETRY] Clearout the response and retry calling the action.
  2: [SERVER-ERROR] Send the client an internal server error page.
  3: [GENERATE-BACKTRACE-FOR-EMACS] Generate a bug report in Emacs.
  

That's of course true, but not what I like to see of course I'd like
to call the next action and go on from there. 

So how am I supposed to build such a action-render-render-action 
chain?

what i want is:
checking the login credentials
if the the customer can be found then show me his/her address
if the data have changed allow to change them 
change the data 
display the changed data
go on wiht the state as if he/she had not changed the address
display e.g the payment terms the customer has taken.
allow changes as needed 
etc.

Help would be very appreciated.

Regards
Friedrich


From rusabd at gmail.com  Mon Aug 22 07:12:08 2005
From: rusabd at gmail.com (Ruslan Abdulkhalikov)
Date: Mon, 22 Aug 2005 13:12:08 +0600
Subject: [Bese-devel] port #1
In-Reply-To: <m264u0yiou.fsf@soma.local>
References: <2f11360405081900562e9c0195@mail.gmail.com>
	<m264u0yiou.fsf@soma.local>
Message-ID: <2f11360405082200127a543c1f@mail.gmail.com>

On 8/20/05, Marco Baringer <mb at bese.it> wrote:
> Ruslan Abdulkhalikov <rusabd at gmail.com> writes:
> 
> > Hello ladies and gentlemen.
> >
> > I've been using UCW for few weeks and found that patches are applied
> > on _dev repository frequently...   So, could you suggest me some CL
> > implementation which is checked first after OpenMCL? CMUCL  SBCL,
> > CLISP, Allegro? Actually SBCL works almost fine for my very simple
> > examples but I failed to make work UCW against CLISP on Windows...
> 
> quite a few people use _dev with sbcl. i generally test on sbcl then
> cmucl then clisp, though lately i've been testing on openmcl and clisp
> only.
> 
> i do not have easy access to a windows machine, what issues are you
> running into?

I tried it a week ago - it just hangs on initial examples page
(127.0.0.1:8080/ucw/examples/index.ucw) - did not replied with any
result... But it can be some local firewall issue, I do not know...

Win32 is not absolutly necessary for me,  actually SBCL and clisp on
linux is ok :) But I thought about creating local web applications
running on the most popular client OS (win). and UCW is great choice
for web applications running on client's box since perfomance is not
issue in this case...


> --
> -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

-- 
Ruslan Abdulkhalikov
-- Quantum Mechanics is God's version of "Trust me."


From mb at bese.it  Tue Aug 23 12:35:31 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 23 Aug 2005 14:35:31 +0200
Subject: [Bese-devel] general question of understanding
In-Reply-To: <87br3s6lfn.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sun, 21 Aug 2005 05:03:24 +0200")
References: <87br3s6lfn.fsf@flarge.here>
Message-ID: <m2br3ozv8s.fsf@soma.local>

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

> Well ucw is way too much a mystery too me, as you've seen but giving
> up is not what I like to do. So I have a general question
> Do I understand it correctly that creating of a component can not be
> done with make-instance?

it can be done (call macro expands into a make-instance form). however
the call-component method does some neccassary bookkeeping work (like
setting up the continuation, the place, the parent, etc.) which you'll
need to repeat if you just use make-instance. of course, this
bookkeeping stuff requires that there be a valid *context* object in
the dynamic environment.

basically it can be done, but it's a goodly amoun of work. why are you
trying to do 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 a_bakic at yahoo.com  Tue Aug 23 16:31:04 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Tue, 23 Aug 2005 09:31:04 -0700 (PDT)
Subject: [Bese-devel] portable aserve/sbcl/utf-8 (cl-ppcre?)
Message-ID: <20050823163105.57395.qmail@web40614.mail.yahoo.com>

Hi,

I am using the aserve backend with sbcl, and am trying to pass utf-8 requests
and responses through. I modified the definition of
*default-aserve-external-format* in aserve/main.cl to contain #+sbcl :utf-8
(and also added "charset=utf-8" to content-type), but am still getting errors
like this:

18:00:46 - While reading http request from 127.0.0.1: invalid array index 197
for
#*0000000000000000000000000000000010111110000110010000000000111111100000000000000000000000000111101000000000000000000000000001110
(should be nonnegative and <127)

Looks like it's coming from cl-ppcre. Is anyone familiar with this stuff? (I
recently added cl-automaton, a regexp-engine, to climacs to get around
cl-ppcre's inability to handle Unicode.)

Thanks,
Alex


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


From a_bakic at yahoo.com  Tue Aug 23 19:02:50 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Tue, 23 Aug 2005 12:02:50 -0700 (PDT)
Subject: [Bese-devel] portable aserve/sbcl/utf-8 (cl-ppcre?)
In-Reply-To: <20050823163105.57395.qmail@web40614.mail.yahoo.com>
Message-ID: <20050823190250.23025.qmail@web40621.mail.yahoo.com>

> I am using the aserve backend with sbcl, and am trying to pass utf-8 requests
> and responses through.

I just found out that Araneida seems to work fine with sbcl and utf-8. (BTW,
one should probably use #+(or unicode sb-unicode) instead of #+sbcl in
aserve/main.cl.)

Alex


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


From frido at q-software-solutions.de  Wed Aug 24 07:35:02 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Wed, 24 Aug 2005 09:35:02 +0200
Subject: [Bese-devel] general question of understanding
In-Reply-To: <m2br3ozv8s.fsf@soma.local> (Marco Baringer's message of "Tue,
	23 Aug 2005 14:35:31 +0200")
References: <87br3s6lfn.fsf@flarge.here> <m2br3ozv8s.fsf@soma.local>
Message-ID: <87mzn7vlcp.fsf@flarge.here>

You've to see it togeher with the other posting. I try to get the
latter running and am failing...
I have not idea what I can make about the posted render construct. It
seems I had to fill in all the missing slots, but that can't be the
solution really.

Regards
Friedrich





From frido at q-software-solutions.de  Wed Aug 24 08:09:47 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Wed, 24 Aug 2005 10:09:47 +0200
Subject: [Bese-devel] what's wrong with the following code?
In-Reply-To: <874q9k6k9f.fsf@flarge.here> (Friedrich Dominicus's message of
	"Sun, 21 Aug 2005 05:28:44 +0200")
References: <874q9k6k9f.fsf@flarge.here>
Message-ID: <87irxvvjqs.fsf@flarge.here>

could that be a bug? the declaration of the shipping-adr class is:

(defcomponent address-mixin ()
  ((adr :initarg :adr :accessor adr
        :backtrack t
        :initform (make-instance 'qss.db::shipping-address)))
  (:metaclass standard-component-class))

;; (forget-class :shipping-adr)
(defclass shipping-adr (address-mixin)
  ()
  (:metaclass standard-component-class))

So it should be a standard component class, looking through the
sources I found:
(defun initialize-component-class-after (class)
  ":after initialization function for standard-component-class objects.
...

shouldn't that fill the missing slots?

Regards
Friedrich


From joe13x at hotmail.com  Wed Aug 24 09:30:15 2005
From: joe13x at hotmail.com (P. Joe Bootvong)
Date: Wed, 24 Aug 2005 16:30:15 +0700
Subject: [Bese-devel] Bug in UCW Examples: Add some numbers?
Message-ID: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>

Hi,

I think this is a bug, or may be I don't understand how it is supposed to 
work.

Step to reproduce:

* Start UCW Examples Goto "Add some numbers" example.
* Enter 2 for number of number to read.
* Enter 10 for first number that was asked, then OK
* Push the browser's back button.
* Enter 5 for the a number, then OK.

Result: The sum is: 5

Expected Result:
There are two sensible behavior I can think of:

1.) UCW think I go back to edit the last value I have entered. So It should 
prompt me for another value. In addition to enter 10, then back to edit to 
5, I should be asked for one more number.
Or
2.) UCW think there is nothing in the page indicating that the value I went 
back and submit is first value. So it took the new value as the second 
number and add both result in, And return the result of 15.

The current behavior is neither of this.




From digash at gmail.com  Wed Aug 24 14:05:08 2005
From: digash at gmail.com (Dimitry Gashinsky)
Date: Wed, 24 Aug 2005 10:05:08 -0400
Subject: [Bese-devel] Bug in UCW Examples: Add some numbers?
In-Reply-To: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>
References: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>
Message-ID: <e2c76ba505082407051f8be4ea@mail.gmail.com>

I get a different behavior. It asks me for another number and then
adds it to 5 following your example. Which makes perfect sense to me.

You can try it here https://dam.digash.com/ucw/examples/index.ucw

Regards,
DiG

On 8/24/05, P. Joe Bootvong <joe13x at hotmail.com> wrote:
> Hi,
> 
> I think this is a bug, or may be I don't understand how it is supposed to
> work.
> 
> Step to reproduce:
> 
> * Start UCW Examples Goto "Add some numbers" example.
> * Enter 2 for number of number to read.
> * Enter 10 for first number that was asked, then OK
> * Push the browser's back button.
> * Enter 5 for the a number, then OK.
> 
> Result: The sum is: 5
> 
> Expected Result:
> There are two sensible behavior I can think of:
> 
> 1.) UCW think I go back to edit the last value I have entered. So It should
> prompt me for another value. In addition to enter 10, then back to edit to
> 5, I should be asked for one more number.
> Or
> 2.) UCW think there is nothing in the page indicating that the value I went
> back and submit is first value. So it took the new value as the second
> number and add both result in, And return the result of 15.
> 
> The current behavior is neither of this.
> 
> 
> _______________________________________________
> 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 mb at bese.it  Wed Aug 24 17:01:48 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 24 Aug 2005 19:01:48 +0200
Subject: [Bese-devel] Bug in UCW Examples: Add some numbers?
In-Reply-To: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl> (P. Joe Bootvong's
	message of "Wed, 24 Aug 2005 16:30:15 +0700")
References: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>
Message-ID: <m23bozw9oj.fsf@soma.local>

"P. Joe Bootvong" <joe13x at hotmail.com> writes:

> Expected Result:
> There are two sensible behavior I can think of:
>
> 1.) UCW think I go back to edit the last value I have entered. So It
>   should prompt me for another value. In addition to enter 10, then
>   back to edit to 5, I should be asked for one more number.

this is what should happen.

the problem is this (in ucw/examples/sum.lisp):

 (loop
   repeat how-many
   sum (call 'read-a-number) into total
   finally (call 'info-message :message (format nil "The sum is: ~D." total)))

the repeat clause of loop does not play well with
continuations. basically repeat setups a local counter and medifies
that value each time through the loop. however since this is a
modification and not a fresh binding calling the contiunation does not
undo the modification (this is the way things should be (though i
damit that it looks wrong in this particular case)).

the only way to fix is to use recursion (which gets 'undone' when
stepping back to a previous continuation):

(defaction start ((s sum))
  (let ((how-many (call 'read-a-number
                        :label "How many numbers should we read?")))
    (labels ((sum-numbers (how-many sum)
                (if (zerop how-many)
                    ;; all done
                    (call 'info-message
                          :message (format nil "The sum is: ~D." sum))
                    ;; more to go
                    (sum-numbers (1- how-many)
                                 (+ sum (call 'read-a-number))))))
      (sum-numbers how-many 0))))

NB: while writing this fix i discovered a bug in cc interpreter, local
function application was broken. UPDATE ARNESI.

> 2.) UCW think there is nothing in the page indicating that the value I
>   went back and submit is first value. So it took the new value as the
>   second number and add both result in, And return the result of 15.
>
> The current behavior is neither of this.

nope :(

-- 
-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  Thu Aug 25 12:56:53 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Thu, 25 Aug 2005 14:56:53 +0200
Subject: [Bese-devel] a thing I can't believe
Message-ID: <87pss2899m.fsf@flarge.here>

Amd I now going nuts. How can I prefill text input field in a form?

(defmethod render-on ((res response) (field text-field))
  (<ucw:input :type "text" :class "ucw-text-field" :accessor (client-value field)
              :size (size field)))

No way to give an initial value? 

This can't be 

Friedrich


From frido at q-software-solutions.de  Thu Aug 25 13:14:18 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Thu, 25 Aug 2005 15:14:18 +0200
Subject: [Bese-devel] Add to my last mail
Message-ID: <87ll2q88gl.fsf@flarge.here>

No it can't be and it isn't at least that has turned out to work one
has to use something along this lines:
(<ucw:input :type "text" :accessor (lisp-value (street adr))))))

Friedrich


From mb at bese.it  Fri Aug 26 08:12:04 2005
From: mb at bese.it (Marco Baringer)
Date: Fri, 26 Aug 2005 10:12:04 +0200
Subject: [Bese-devel] Bug in UCW Examples: Add some numbers?
In-Reply-To: <m23bozw9oj.fsf@soma.local> (Marco Baringer's message of "Wed,
	24 Aug 2005 19:01:48 +0200")
References: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>
	<m23bozw9oj.fsf@soma.local>
Message-ID: <m24q9dce23.fsf@soma.local>

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

> the only way to fix is to use recursion (which gets 'undone' when
> stepping back to a previous continuation):
>
> (defaction start ((s sum))
>   (let ((how-many (call 'read-a-number
>                         :label "How many numbers should we read?")))
>     (labels ((sum-numbers (how-many sum)
>                 (if (zerop how-many)
>                     ;; all done
>                     (call 'info-message
>                           :message (format nil "The sum is: ~D." sum))
>                     ;; more to go
>                     (sum-numbers (1- how-many)
>                                  (+ sum (call 'read-a-number))))))
>       (sum-numbers how-many 0))))

this is another way to fix the problem. we're using ucw's backtracking
mechanism to work around the fact that continuations don't undo side
effects.

(defaction start ((s sum))
  (loop
     for how-many = (call 'read-a-number :label "How many numbers should we read?")
     initially (backtrack (ucw::context.current-frame *context*) (make-place how-many))
     do (loop
           for count below how-many
           initially (backtrack (ucw::context.current-frame *context*)
                                (make-place count))
           sum (call 'read-a-number) into total
           finally (call 'info-message
                         :message (format nil "The sum is: ~D." total)))))

now that i think about it may be worth the effort, since we've already
got the neccessary infrastructure, to have an action automatically
backtrack all local variables. though this will help in a lot of casse
it will still leave weird edge cases sucha s this one:

(let ((a (cons 1 2)))
  (setf (car a) 'foo)
  ;; the modification to the car of a is not backtracked
  )

the only way to fix this would be to use deep copying of all
backtracked places, this sounds like a _horrible_ idea.

-- 
-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 Aug 26 10:04:52 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Fri, 26 Aug 2005 12:04:52 +0200
Subject: [Bese-devel] unclear usage again
Message-ID: <87mzn53tff.fsf@flarge.here>

Well there is the class form-element which has two slots client-value
and lisp-value, as I understand you can use the client-value field to
initialize the content of  form field. After it has been displayd to
the user and the user has changed  it one can use read-client-value
which however fills lisp-value with the content of the client-value
field. 

So I'd like to know how lisp-value and client-value play together and
what purpose they have. 

Regards
Friedrich


From frido at q-software-solutions.de  Fri Aug 26 10:21:28 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Fri, 26 Aug 2005 12:21:28 +0200
Subject: [Bese-devel] yet another question (suggestion?)
Message-ID: <87irxt3snr.fsf@flarge.here>

Well for the moment I have find at least a way to proceed. But there
are a still a lot of open questions. Does anyone here have used UCW
together with CLSQL? for the time beeing it seems on needs quite few
extra classed to get them going together. e.g you like to read an
object from a database and "display" it. That's the "easy" part, but
then you migh want to change the display so you have to put e.g an
text-field out for a field you like to change. to initialize the view
it seems on has to write:

  (let ((address (adr adr)))
    ;; (inspect adr)
    (when address 
      (setf (ucw::client-value (zip adr)) (zip address)
            (ucw::client-value (street adr)) (street address)
            (ucw::client-value (city adr)) (city address)
            (ucw::client-value (country adr)) (country address)
            (ucw::client-value (email adr)) (email address)))
...

so me seems this isn't an unusual request you want to initialze the
few from a database objects,but one has to use a "private" feature of
ucw to achieve that. I think client-value should be an exported
function.

Ok now let's see further. You got your view, now you change the value
now you have to propagate the changes in the view to the database so
you have to use code like this IIUC.

(defun update-db-address-from-view (slot-name address)
  (setf (slot-value  (adr address) slot-name) 
        (read-client-value (slot-value address slot-name))))

Of course one can generalize it but me things at least this back/forth
between has to be code explicitly. Am I missing something or is that a
correct impression? 


As I understand lisp-on-lines, the author there has worked for
autmating this task. Is anyone using lisp-on-lines and can confirm
that?

Thanks
Friedrich


From smoof-ra at elder-gods.org  Fri Aug 26 13:34:41 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Fri, 26 Aug 2005 09:34:41 -0400
Subject: [Bese-devel] Bug in UCW Examples: Add some numbers?
In-Reply-To: <m24q9dce23.fsf@soma.local>
References: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>
	<m23bozw9oj.fsf@soma.local> <m24q9dce23.fsf@soma.local>
Message-ID: <20050826133441.GA4420@melkor.elder-gods.org>

* Marco Baringer (mb at bese.it) [050826 04:13]:
> this is another way to fix the problem. we're using ucw's backtracking
> mechanism to work around the fact that continuations don't undo side
> effects.
> 
> (defaction start ((s sum))
>   (loop
>      for how-many = (call 'read-a-number :label "How many numbers should we read?")
>      initially (backtrack (ucw::context.current-frame *context*) (make-place how-many))
>      do (loop
>            for count below how-many
>            initially (backtrack (ucw::context.current-frame *context*)
>                                 (make-place count))
>            sum (call 'read-a-number) into total
>            finally (call 'info-message
>                          :message (format nil "The sum is: ~D." total)))))
> 
> now that i think about it may be worth the effort, since we've already
> got the neccessary infrastructure, to have an action automatically
> backtrack all local variables. though this will help in a lot of casse
> it will still leave weird edge cases sucha s this one:
> 
> (let ((a (cons 1 2)))
>   (setf (car a) 'foo)
>   ;; the modification to the car of a is not backtracked
>   )
> 
> the only way to fix this would be to use deep copying of all
> backtracked places, this sounds like a _horrible_ idea.

Maybe we should have looping constructs that create new bindings at
each iteration, and encourage people to use those.  I don't think the
behavior ucw is exhibiting here is a bug at all.  The bug is only in
the example code that doesn't take into account the fact that it might 
be continued at the same point more than once.

      --larry 


From chalub at gmail.com  Sat Aug 27 02:10:11 2005
From: chalub at gmail.com (Fabricio Chalub)
Date: 26 Aug 2005 23:10:11 -0300
Subject: [Bese-devel] Two problems with CMUCL HEAD 2005-08-26 (19b) in UCW
	and ARNESI (dev)
Message-ID: <87pss0870c.fsf@eq.local>

Hello!

I'm trying UCW (latest dev repositories, etc.) on CMUCL HEAD
(2005-08-26) (19b) and am having two different problems; the first is
with the actual compilation of UCW per se and an attempt in fixing it is
provided, while the second is in one of the examples.

The problem is an error raised by ARNESI when CMUCL loads/compiles
control-flow.lisp in rerl/standard-component, as follows (note also the
minor typo `lexcial'):

   Error in function (METHOD IT.BESE.ARNESI::LEXICAL-FUNCTIONS NIL
   (C::LEXENV)):

   Sorry, don't know how to handle the lexcial function spec 
   (IT.BESE.UCW:ANSWER . #<C::FUNCTIONAL #x5853F82D NAME= NIL>).
   [Condition of type SIMPLE-ERROR]

Which is raised by LEXICAL-FUNCTIONS in LEXENV.LISP.  It appears that
the function spec (ANSWER . #<C::FUNCTIONAL) fell through the cases
handled by the #cmu version of that function.

An ugly attempt at covering this case follows, but it surely does not
solve the problem satisfactorily as I have no experience in CMUCL's
internals.  My solution is to add another ELSE IF before the ERROR call,
considering the 'C::FUNCTIONAL case as follows:

--- lexenv.lisp.~1~     2005-08-25 23:38:43.000000000 -0300
+++ lexenv.lisp 2005-08-26 22:32:07.531838568 -0300
@@ -126,10 +126,13 @@
                   (eql 'system:macro (second func-spec)))
        ;; except that we don't return macros for now
        do (progn)
-     ;; if we get here we're confused :(
-     else
-       do (error "Sorry, don't know how to handle the lexcial function spec ~S."
-                 func-spec)))
+       else if (typep (cdr func-spec) 'C::FUNCTIONAL)
+             collect (car func-spec)
+            ;; handle the case  (NAME . #<C::FUNCTIONAL>)
+            ;; if we get here we're confused :(
+            else
+            do (error "Sorry, don't know how to handle the lexcial function spec ~S."
+                      func-spec)))

With the above patch the LEXICAL-FUNCTION apparently works, except now
that the PRESENTATIONS example of UCW fails with the following error.

  Generic function CREATE-PERSON: Invalid generic function parameter name
  ((LAST-NAME 
  LAST-NAME))

Which completely baffles me as I cannot find any problems in
CREATE-PERSON on a superficial look. I don't discard this being a
fallout from my "patch". 

Anybody else have similar problems with CMUCL?

(Ignoring the PRESENTATIONS example, the rest works perfectly well,
thanks for the magnificent work.)

Best regards,
Fabricio



From mb at bese.it  Sat Aug 27 10:36:55 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 12:36:55 +0200
Subject: [Bese-devel] a thing I can't believe
In-Reply-To: <87pss2899m.fsf@flarge.here> (Friedrich Dominicus's message of
	"Thu, 25 Aug 2005 14:56:53 +0200")
References: <87pss2899m.fsf@flarge.here>
Message-ID: <m2slwveke0.fsf@soma.local>

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

> Amd I now going nuts. How can I prefill text input field in a form?
>
> (defmethod render-on ((res response) (field text-field))
>   (<ucw:input :type "text" :class "ucw-text-field" :accessor (client-value field)
>               :size (size field)))
>
> No way to give an initial value? 

two options:

1) use the :value attribute.

2) since text-field your own component use :initform on the
   client-value slot (or initialize the client-value in a
   shared-initialize method or whatever).

-- 
-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 Aug 27 10:40:27 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 12:40:27 +0200
Subject: [Bese-devel] unclear usage again
In-Reply-To: <87mzn53tff.fsf@flarge.here> (Friedrich Dominicus's message of
	"Fri, 26 Aug 2005 12:04:52 +0200")
References: <87mzn53tff.fsf@flarge.here>
Message-ID: <m2ll2nek84.fsf@soma.local>

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

> So I'd like to know how lisp-value and client-value play together and
> what purpose they have. 

for a text input the relation between the two is trivial (they're both
strings). for a number input the relation is a bit more interesting so
i'll use that.

the first thing to realize is that the client-value will _always_ be a
string, while the lisp-value may be a string or any lisp object (like
a number). when we call the submit action on a form element it calls
the read-client-value method which converts (and validates) the string
the client sent into a lisp object and fills the lisp-value slot
appropiatly.

in code terms:

(when (validate-client-value form-element)
  (setf (lisp-value form-element) 
        (read-client-value form-element)))

-- 
-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 Aug 27 10:46:57 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 12:46:57 +0200
Subject: [Bese-devel] yet another question (suggestion?)
In-Reply-To: <87irxt3snr.fsf@flarge.here> (Friedrich Dominicus's message of
	"Fri, 26 Aug 2005 12:21:28 +0200")
References: <87irxt3snr.fsf@flarge.here>
Message-ID: <m2d5nzejxa.fsf@soma.local>

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

> Of course one can generalize it but me things at least this back/forth
> between has to be code explicitly. Am I missing something or is that a
> correct impression? 

your impression is correct. ucw does _not_ provide support for mapping
between db abjects and web forms, this is an intentional design
chocie. there are over 18 gazillion ways of doing this mapping, each
with its own pros and cons. i'm not about to chose one and use only
that. i have so far used various db backends (clsql, elephant,
prevalence, ccl:save-application or sb-ext:save-lisp-and-die) and for
each one i've needed a fundamentally different way of mapping the db
objects to web forms.

the presentation facility is my initial (not very successfull) attempt
to automate this job. the way i use the presentations is to a have a
macro, in each application, which creates the clos class, the
presentations, and any db specific infrastructure.

> as I understand lisp-on-lines, the author there has worked for
> autmating this task. Is anyone using lisp-on-lines and can confirm
> that?

i am not activly using lisp-on-lines, though i have looked at it and
believe that your time would be better spent learning drew's code than
the presentation/form stuff i've written.

-- 
-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 emailmac at gmail.com  Sat Aug 27 07:11:27 2005
From: emailmac at gmail.com (Mac Chan)
Date: Sat, 27 Aug 2005 00:11:27 -0700
Subject: [Bese-devel] Two problems with CMUCL HEAD 2005-08-26 (19b) in UCW
	and ARNESI (dev)
In-Reply-To: <E1E8r1C-0008Tv-Rf@e4ward.com>
References: <E1E8r1C-0008Tv-Rf@e4ward.com>
Message-ID: <4877ae64050827001172bdd5f3@mail.gmail.com>

On 26 Aug 2005 23:10:11 -0300, Fabricio Chalub wrote:
> I'm trying UCW (latest dev repositories, etc.) on CMUCL HEAD
> (2005-08-26) (19b) and am having two different problems; the first is
> with the actual compilation of UCW per se and an attempt in fixing it is
> provided, while the second is in one of the examples.

Marco was kind enough to walk me thru these two problems on irc yesterday.
I'm not sure if this is a complete fix but here you go:


arnesi_dev diff

Index: src/call-cc/generic-functions.lisp
===================================================================
--- src/call-cc/generic-functions.lisp  (revision 203)
+++ src/call-cc/generic-functions.lisp  (working copy)
@@ -127,11 +127,11 @@
           ((or required-function-argument-form
                specialized-function-argument-form)
            (push (name arg) generic-lambda-list))
-          (keyword-function-argument-form
-           (pushnew '&key generic-lambda-list)
-           (push (list (list (keyword-name arg)
-                             (name arg)))
-                 generic-lambda-list))
+          
+                 (keyword-function-argument-form 
+                  (pushnew '&key generic-lambda-list) 
+                  (push (list (keyword-name arg)) generic-lambda-list))
+
           (rest-function-argument-form
            (push '&rest generic-lambda-list)
            (push (name arg) generic-lambda-list))
Index: src/lexenv.lisp
===================================================================
--- src/lexenv.lisp     (revision 203)
+++ src/lexenv.lisp     (working copy)
@@ -126,6 +126,9 @@
                   (eql 'system:macro (second func-spec)))
        ;; except that we don't return macros for now
        do (progn)
+       ;; sometimes, oddly, we get (NAME DEF) 
+       else if (symbolp (first func-spec)) 
+               collect (first func-spec)
      ;; if we get here we're confused :(
      else
        do (error "Sorry, don't know how to handle the lexcial
function spec ~S."




Also, here are some minor fixes for cmucl 

ucw_dev diff

Index: src/control.lisp
===================================================================
--- src/control.lisp    (revision 203)
+++ src/control.lisp    (working copy)
@@ -1,4 +1,4 @@
-(in-package :ucw)
+(in-package :it.bese.ucw)
 
 (defun threaded-lisp-p ()
   #+(and sbcl sb-thread) t
Index: bin/start.lisp
===================================================================
--- bin/start.lisp      (revision 203)
+++ bin/start.lisp      (working copy)
@@ -2,6 +2,19 @@
 
 (in-package :common-lisp-user)
 
+#+cmu
+(defun init-cmu-mp ()
+  ;; this isn't strictly necessary, but scheduling feels very coarse
+  ;; without startup-idle-and-top-level-loops, leading to answer delays
+  ;; of about 1s per request.
+  (unless (find-if
+           #'(lambda (proc) (string= (mp:process-name proc) "Top Level Loop"))
+           (mp:all-processes))
+    (mp::startup-idle-and-top-level-loops)))
+
+#+cmu
+(init-cmu-mp)
+
 ;;;; * UCW server initialization "script" 
 
@@ -66,6 +79,9 @@
 
 (register-application *default-server* ucw::*admin-application*)
 
+#+cmu
+(defparameter swank:*communication-style* :spawn)
+


From mb at bese.it  Sat Aug 27 10:58:39 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 12:58:39 +0200
Subject: [Bese-devel] Bug in UCW Examples: Add some numbers?
In-Reply-To: <20050826133441.GA4420@melkor.elder-gods.org> (Larry D'Anna's
	message of "Fri, 26 Aug 2005 09:34:41 -0400")
References: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>
	<m23bozw9oj.fsf@soma.local> <m24q9dce23.fsf@soma.local>
	<20050826133441.GA4420@melkor.elder-gods.org>
Message-ID: <m2y86nd4tc.fsf@soma.local>

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

> Maybe we should have looping constructs that create new bindings at
> each iteration, and encourage people to use those.  I don't think the
> behavior ucw is exhibiting here is a bug at all.  The bug is only in
> the example code that doesn't take into account the fact that it might 
> be continued at the same point more than once.

there are a few different issues:

1)

(dolist (i my-list)
  (<:li (<ucw:a :action (do-stuff i))))

here we don't know whether dolist will create a new binding or not and
so we don't know if the links will each have their own binding or if
the same binding is modified. fortunetly this case (which is the more
common one) is easily solvable:

(dolist (i my-list)
  (let ((i i))
    (<ucw:a :action (do-stuff i))))

or use dolist* which, other than allowing destructuring of the
elements, also guarantees a fresh binding each time:

(dolist* (i my-list)
  (<ucw:a :action (do-stuff i)))

1a)

(loop 
  for i in my-list
  do (call 'foo-bar))

here we'd need a loop* macro which, like dolist vs. dolist* above,
guaranteed a fresh binding on each iteration. the solutions here are:
a) re implement loop (i'll accept a patch but i'm not about to do it);
b) use iterate, this would require first fixing iterate's use
macroexpand; c) don't use loop or wrap the (call 'foo-bar) form with
(let ((i i)).

for now (and probably for a long time) (c) is our best option.

2)

(loop
  repeat 5
  do (call 'foo-bar))

this is more problematic. we do _not_ want to create a new binding
here, if we don't modify the same biding then our loop won't
loop. however we want that modification to get magically undone when
we use the back button. fixing this would require a ucw:loop macro or
something which setup the internal counter for backtracking.

at least, i think these are the issues and the associated solutions,
i'd love to be proven wrong.

-- 
-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 Aug 27 11:21:04 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 13:21:04 +0200
Subject: [Bese-devel] Two problems with CMUCL HEAD 2005-08-26 (19b) in UCW
	and ARNESI (dev)
In-Reply-To: <87pss0870c.fsf@eq.local> (Fabricio Chalub's message of "26 Aug
	2005 23:10:11 -0300")
References: <87pss0870c.fsf@eq.local>
Message-ID: <m2irxrd3rz.fsf@soma.local>

Fabricio Chalub <chalub at gmail.com> writes:

> An ugly attempt at covering this case follows, but it surely does not
> solve the problem satisfactorily as I have no experience in CMUCL's
> internals.  My solution is to add another ELSE IF before the ERROR call,
> considering the 'C::FUNCTIONAL case as follows:

what we're doing in the lexical-xyz functions isn't kosher, there is
no 'clean' solution (unless all implementations started providing
environment access like allegro is doing).

> --- lexenv.lisp.~1~     2005-08-25 23:38:43.000000000 -0300
> +++ lexenv.lisp 2005-08-26 22:32:07.531838568 -0300
> @@ -126,10 +126,13 @@
>                    (eql 'system:macro (second func-spec)))
>         ;; except that we don't return macros for now
>         do (progn)
> -     ;; if we get here we're confused :(
> -     else
> -       do (error "Sorry, don't know how to handle the lexcial function spec ~S."
> -                 func-spec)))
> +       else if (typep (cdr func-spec) 'C::FUNCTIONAL)
> +             collect (car func-spec)
> +            ;; handle the case  (NAME . #<C::FUNCTIONAL>)
> +            ;; if we get here we're confused :(
> +            else
> +            do (error "Sorry, don't know how to handle the lexcial function spec ~S."
> +                      func-spec)))

applied, thanks.

> With the above patch the LEXICAL-FUNCTION apparently works, except now
> that the PRESENTATIONS example of UCW fails with the following error.
>
>   Generic function CREATE-PERSON: Invalid generic function parameter name
>   ((LAST-NAME 
>   LAST-NAME))
>
> Which completely baffles me as I cannot find any problems in
> CREATE-PERSON on a superficial look. I don't discard this being a
> fallout from my "patch". 

the problem is that cumcl does not support generic-function lambda
lists like this:

(defgeneric foobar (&key ((:keyword variable))))

this is, according to my reading of the spec and #lisp's denziens, a
legal defgeneric form. this looks like a bug in cmucl. however, i just
commited a patch which only passes the keyword if the user actually
supplied one differenet from the argument. so basically this:

(defmethod/cc foo (&key bar))

will now work, but this:

(defmethod/cc foo (&key ((:goo bar))))

still won't.

-- 
-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 Aug 27 11:29:57 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 13:29:57 +0200
Subject: [Bese-devel] Two problems with CMUCL HEAD 2005-08-26 (19b) in UCW
	and ARNESI (dev)
In-Reply-To: <4877ae64050827001172bdd5f3@mail.gmail.com> (Mac Chan's message
	of "Sat, 27 Aug 2005 00:11:27 -0700")
References: <E1E8r1C-0008Tv-Rf@e4ward.com>
	<4877ae64050827001172bdd5f3@mail.gmail.com>
Message-ID: <m2br3jd3d6.fsf@soma.local>

Mac Chan <emailmac at gmail.com> writes:

> Also, here are some minor fixes for cmucl 

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 binarin at agava.com  Sat Aug 27 14:51:13 2005
From: binarin at agava.com (Alexey Lebedeff)
Date: Sat, 27 Aug 2005 18:51:13 +0400
Subject: [Bese-devel] UTF-8
Message-ID: <871x4fcu1q.fsf@binarin.domain>


I'm using ucw with mod_lisp backend and sbcl.

To make it work with utf-8 I made few small changes to arnesi & ucw.

Changes to arnesi:

- When utf-8 is used, encoding of characters > 127 with '&xXXXX'; is not
  the right thing. So i dissallowed such encoding when sb-unicode feature
  is present. This is not quite correct - not all users of
  unicode-enabled sbcl wants this behavior(that's for sure) - but my
  superficial knowledge of arnesi doesn't allow me to make this change in
  better way.

- nunescape-as-uri is now accepting external-format argument, same as
  unescape-as-uri.

Changes to ucw:

- Allow to specify various external-formats for different purposes:
  unescaping of urls, communication with slime and http.

Accompanied by setting slime-net-coding-system on emacs side, it works
pretty well.


And also unrelated to external formats change: 
There was missing argument in call to ucw.backend.dribble. So, when I was
starting ucw, sbcl complained to me about "format" arguments mismatch,
resulting in impossibilty to start ucw without user intervention.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: arnesi.CHANGESET
Type: text/x-patch
Size: 14060 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050827/48400d40/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ucw.CHANGESET
Type: text/x-patch
Size: 15907 bytes
Desc: inline
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050827/48400d40/attachment-0001.bin>
-------------- next part --------------

-- 
[1180'99:11-6a] [BMSTU-AK:5-111]

From mb at bese.it  Sat Aug 27 19:15:40 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 21:15:40 +0200
Subject: [Bese-devel] UTF-8
In-Reply-To: <871x4fcu1q.fsf@binarin.domain> (Alexey Lebedeff's message of
	"Sat, 27 Aug 2005 18:51:13 +0400")
References: <871x4fcu1q.fsf@binarin.domain>
Message-ID: <m2br3jb38j.fsf@soma.local>

Alexey Lebedeff <binarin at agava.com> writes:

> I'm using ucw with mod_lisp backend and sbcl.
>
> To make it work with utf-8 I made few small changes to arnesi & ucw.
>
> Changes to arnesi:
>
> - When utf-8 is used, encoding of characters > 127 with '&xXXXX'; is not
>   the right thing. So i dissallowed such encoding when sb-unicode feature
>   is present. This is not quite correct - not all users of
>   unicode-enabled sbcl wants this behavior(that's for sure) - but my
>   superficial knowledge of arnesi doesn't allow me to make this change in
>   better way.
>
> - nunescape-as-uri is now accepting external-format argument, same as
>   unescape-as-uri.

applied. thanks!

> Changes to ucw:
>
> - Allow to specify various external-formats for different purposes:
>   unescaping of urls, communication with slime and http.
>
> Accompanied by setting slime-net-coding-system on emacs side, it works
> pretty well.

when trying to apply this i get this message:

soma:~/lisp/ucw mb$ darcs apply patch 
darcs: bug in get_extra.
Most likely this is caused by a bug that existed in darcs prior
to version 1.0.1.  Details for dealing with this issue can be found
at http://darcs.net/DarcsWiki/Issues1.0.1
soma:~/lisp/ucw mb$ cd ../arnesi/

the solution suggested by that page is to use an older version of
darcs, which i don't have. could you try upgrading darcs and
reverting/rerecording the pages?

otherwise i can apply the differences and just create a new patch,
though you'll then have to unrecord your patch and apply mine.

suggestions?

> And also unrelated to external formats change: 
> There was missing argument in call to ucw.backend.dribble. So, when I was
> starting ucw, sbcl complained to me about "format" arguments mismatch,
> resulting in impossibilty to start ucw without user intervention.

thanks anyway.

-- 
-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  Sat Aug 27 19:29:11 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sat, 27 Aug 2005 12:29:11 -0700 (PDT)
Subject: [Bese-devel] a minor typo fix
Message-ID: <20050827192912.36032.qmail@web40610.mail.yahoo.com>

Hi,

diff -rN old-ucw_dev/src/yaclml/ucw-tags.lisp
new-ucw_dev/src/yaclml/ucw-tags.lisp
62c62
<           `(defun set-action-paramater (value)
---
>           `(defun set-action-parameter (value)

will make the presentation demo work better.

Alex


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


From mb at bese.it  Sat Aug 27 19:43:35 2005
From: mb at bese.it (Marco Baringer)
Date: Sat, 27 Aug 2005 21:43:35 +0200
Subject: [Bese-devel] a minor typo fix
In-Reply-To: <20050827192912.36032.qmail@web40610.mail.yahoo.com> (Aleksandar
	Bakic's message of "Sat, 27 Aug 2005 12:29:11 -0700 (PDT)")
References: <20050827192912.36032.qmail@web40610.mail.yahoo.com>
Message-ID: <m23bovb1y0.fsf@soma.local>

Aleksandar Bakic <a_bakic at yahoo.com> writes:

> Hi,
>
> diff -rN old-ucw_dev/src/yaclml/ucw-tags.lisp
> new-ucw_dev/src/yaclml/ucw-tags.lisp
> 62c62
> <           `(defun set-action-paramater (value)
> ---
>>           `(defun set-action-parameter (value)
>
> will make the presentation demo work better.

that explains a lot of things :)

thanks a bunch.

-- 
-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  Sat Aug 27 19:55:05 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sat, 27 Aug 2005 12:55:05 -0700 (PDT)
Subject: [Bese-devel] a minor typo fix
In-Reply-To: <m23bovb1y0.fsf@soma.local>
Message-ID: <20050827195506.87119.qmail@web40604.mail.yahoo.com>

> > will make the presentation demo work better.
> 
> that explains a lot of things :)

Well, just a little bit better. I am now getting the following type error in
the Javascript console:

UCW Examples
http://localhost:8080/ucw/examples/index.ucw?s=YBreWHzydwOKZzlqLFGDItCPxMPCLtYUYJrSnxXE&f=EmrdzaAflPPTnivSpBky&a=BIvWFqxFkJ
Event thread: click
Error:
name: TypeError
message: Statement on line 4: Could not convert undefined or null to object
Backtrace:
  Line 4 of inline#1 script in
http://localhost:8080/ucw/examples/index.ucw?s=YBreWHzydwOKZzlqLFGDItCPxMPCLtYUYJrSnxXE&f=EmrdzaAflPPTnivSpBky&a=BIvWFqxFkJ
    this.form["a"].value = value;
  Line 1 of  script 
    setActionParameter("vjDxpiFLbt");
return true;
  At unknown location
    [statement source code not available]

Alex


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


From binarin at agava.com  Sat Aug 27 20:12:05 2005
From: binarin at agava.com (Alexey Lebedeff)
Date: Sun, 28 Aug 2005 00:12:05 +0400
Subject: [Bese-devel] UTF-8
In-Reply-To: <m2br3jb38j.fsf@soma.local> (Marco Baringer's message of "Sat,
	27 Aug 2005 21:15:40 +0200")
References: <871x4fcu1q.fsf@binarin.domain> <m2br3jb38j.fsf@soma.local>
Message-ID: <87slwvw356.fsf@binarin.domain>

## On Sat, 27 Aug 2005 21:15:40 +0200
## you wrote:

 >> Changes to ucw:
 >>
 >> - Allow to specify various external-formats for different purposes:
 >>   unescaping of urls, communication with slime and http.
 >>
 >> Accompanied by setting slime-net-coding-system on emacs side, it works
 >> pretty well.

 MB> when trying to apply this i get this message:

 MB> soma:~/lisp/ucw mb$ darcs apply patch 
 MB> darcs: bug in get_extra.
 MB> Most likely this is caused by a bug that existed in darcs prior
 MB> to version 1.0.1.  Details for dealing with this issue can be found
 MB> at http://darcs.net/DarcsWiki/Issues1.0.1
 MB> soma:~/lisp/ucw mb$ cd ../arnesi/

 MB> the solution suggested by that page is to use an older version of
 MB> darcs, which i don't have. could you try upgrading darcs and
 MB> reverting/rerecording the pages?
I have darcs-1.0.2 from debian stable (and 1.0.3 is available from
testing/unstable). So i think this is not about 1.0.1 bug. The working
copy against which i maded patches was fetched only today.

 MB> otherwise i can apply the differences and just create a new patch,
 MB> though you'll then have to unrecord your patch and apply mine.
That will be ok.

-- 
[1180'99:11-6a] [BMSTU-AK:5-111]


From a_bakic at yahoo.com  Sat Aug 27 22:05:34 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sat, 27 Aug 2005 15:05:34 -0700 (PDT)
Subject: [Bese-devel] a minor typo fix
In-Reply-To: <20050827195506.87119.qmail@web40604.mail.yahoo.com>
Message-ID: <20050827220534.57237.qmail@web40622.mail.yahoo.com>

BTW, I changed ucw-tags.lisp to get "this.form.a.value=...", but this resulted
in the same error. On the other hand, Firefox says "Error: this.form has no
properties".

Alex


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


From a_bakic at yahoo.com  Sat Aug 27 22:42:27 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sat, 27 Aug 2005 15:42:27 -0700 (PDT)
Subject: [Bese-devel] a minor typo fix
In-Reply-To: <20050827220534.57237.qmail@web40622.mail.yahoo.com>
Message-ID: <20050827224227.82832.qmail@web40628.mail.yahoo.com>

This seems to be one correct way:

diff -rN old-ucw_dev/src/yaclml/ucw-tags.lisp
new-ucw_dev/src/yaclml/ucw-tags.lisp
41c41
<                           (set-action-parameter
---
>                           (set-action-parameter this.form
62,63c62,63
<           `(defun set-action-paramater (value)
<              (setf (slot-value (aref this.form ,',+action-parameter-name+)
'value)
---
>           `(defun set-action-parameter (form value)
>              (setf (slot-value (slot-value form ,(intern
+action-parameter-name+)) 'value)
102c102,103
<                                        (set-action-parameter
,(register-action---
>                                        (set-action-parameter this.form
>                                                            ,(register-action
166c167
<                            (set-action-parameter ,(make-new-action
(context.current-frame *context*)
---
>                            (set-action-parameter this.form ,(make-new-action
(context.current-frame *context*)

Alex


		
__________________________________ 
Yahoo! Mail 
Stay connected, organized, and protected. Take the tour: 
http://tour.mail.yahoo.com/mailtour.html 



From drewc at tech.coop  Sun Aug 28 14:22:58 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Sun, 28 Aug 2005 07:22:58 -0700
Subject: [Bese-devel] yet another  minor typo fix
In-Reply-To: <20050827224227.82832.qmail@web40628.mail.yahoo.com>
References: <20050827224227.82832.qmail@web40628.mail.yahoo.com>
Message-ID: <4311C8C2.8010307@tech.coop>

it's "text/css", not "test/css" :).

drewc
-------------- next part --------------
A non-text attachment was scrubbed...
Name: css.patch
Type: text/x-patch
Size: 14467 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050828/68296736/attachment.bin>

From drewc at tech.coop  Sun Aug 28 14:27:53 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Sun, 28 Aug 2005 07:27:53 -0700
Subject: [Bese-devel] patch: add :target attribute to yaclml form
In-Reply-To: <4311C8C2.8010307@tech.coop>
References: <20050827224227.82832.qmail@web40628.mail.yahoo.com>
	<4311C8C2.8010307@tech.coop>
Message-ID: <4311C9E9.3050208@tech.coop>

A non-text attachment was scrubbed...
Name: target.patch
Type: text/x-patch
Size: 1456 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050828/8e6dca21/attachment.bin>

From drewc at tech.coop  Sun Aug 28 14:44:25 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Sun, 28 Aug 2005 07:44:25 -0700
Subject: [Bese-devel] arnesi patch: enhanced defmethod/cc to work in the
 case of a null argument list and missing body.]
Message-ID: <4311CDC9.1000702@tech.coop>


-------------- next part --------------
A non-text attachment was scrubbed...
Name: method.patch
Type: text/x-patch
Size: 14912 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050828/6cd628b6/attachment.bin>

From a_bakic at yahoo.com  Sun Aug 28 00:32:02 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sat, 27 Aug 2005 17:32:02 -0700 (PDT)
Subject: [Bese-devel] another presentation issue
Message-ID: <20050828003202.14830.qmail@web40601.mail.yahoo.com>

Hi,

I am able to edit a person's data, but when I click OK, I get an error:

Backtrace:
  0: ((SB-PCL::FAST-METHOD SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT (T))
#<unused argument> #<unused argument> #<FUNCTION (LAMBDA NIL) {242D435}>)
  1: (SWANK::CALL-WITH-BINDINGS ((*PRINT-PRETTY*) (*PRINT-LEVEL* . 4)
(*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*)
(*PRINT-PPRINT-DISPATCH* . #<SB-PRETTY:PPRINT-DISPATCH-TABLE {5D1F7A1}>)
(*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*) (*PRINT-ARRAY* . T)
...) #<FUNCTION (LAMBDA NIL) {242D3F5}>)
  2: (SWANK::DEBUG-IN-EMACS #<TYPE-ERROR {97FEC89}>)
  3: ((LAMBDA NIL))
  4: (SWANK::CALL-WITH-REDIRECTED-IO #<SWANK::CONNECTION {9237991}> #<CLOSURE
(LAMBDA NIL) {9E0936D}>)
  5: (SWANK::CALL-WITH-CONNECTION #<SWANK::CONNECTION {9237991}> #<CLOSURE
(LAMBDA NIL) {9E0935D}>)
  6: ((SB-PCL::FAST-METHOD IT.BESE.UCW::HANDLE-ACTION-ERROR (ERROR T))
#<unavailable argument> #<unavailable argument> #<TYPE-ERROR {97FEC89}>
(#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 4 :DESCRIPTION "(ERROR TYPE-ERROR)"
:LOCALS (# #) :SOURCE-LOCATION (:ERROR "The source-path (NIL) is not valid."))
#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 5 :DESCRIPTION
"(SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER
      #<unavailable argument>
      #.(SB-SYS:INT-SAP #XF7B85478)
      #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XF7B85168 :TYPE (*
                                                              (SB-ALIEN:STRUCT
                                                              
SB-VM::OS-CONTEXT-T-STRUCT))>
      (14 78))" :LOCALS (# # # # #) :SOURCE-LOCATION (:LOCATION # # #))
#S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 6 :DESCRIPTION
"(SB-KERNEL:INTERNAL-ERROR
      #.(SB-SYS:INT-SAP #XF7B85168)
      #<unavailable argument>)" :LOCALS (# # #) :SOURCE-LOCATION (:LOCATION # #
#)) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 7 :DESCRIPTION "(\"foreign function:
call_into_lisp\")" :LOCALS NIL :SOURCE-LOCATION (:ERROR "The value
#<SB-DI::BOGUS-DEBUG-FUN \"foreign function: call_into_lisp\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 8
:DESCRIPTION "(\"foreign function: funcall2\")" :LOCALS NIL :SOURCE-LOCATION
(:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign function: funcall2\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 9
:DESCRIPTION "(\"foreign function: interrupt_internal_error\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value
  #<SB-DI::BOGUS-DEBUG-FUN \"foreign function: interrupt_internal_error\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 10
:DESCRIPTION "(\"foreign function: sigtrap_handler\")" :LOCALS NIL
:SOURCE-LOCATION (:ERROR "The value #<SB-DI::BOGUS-DEBUG-FUN \"foreign
function: sigtrap_handler\">
is not of type
  SB-DI::COMPILED-DEBUG-FUN.")) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 11
:DESCRIPTION "(FDEFINITION #<PERSON {5DA9001}>)" :LOCALS (# #) :SOURCE-LOCATION
(:ERROR "The source-path (NIL) is not valid.")) #S(IT.BESE.UCW::BACKTRACE-FRAME
:INDEX 12 :DESCRIPTION "((LAMBDA (IT.BESE.ARNESI::ARGUMENTS))
      (#<PERSON {5E717B1}> #<PERSON {5E717F9}>))" :LOCALS (#) :SOURCE-LOCATION
(:LOCATION # # #)) #S(IT.BESE.UCW::BACKTRACE-FRAME :INDEX 13 :DESCRIPTION
"(IT.BESE.ARNESI::DRIVE-INTERPRETER/CC #<CLOSURE (LAMBDA #) {97F8FAD}>)"
:LOCALS (#) :SOURCE-LOCATION (:LOCATION # # #)) ...))
  7: (SIGNAL #<TYPE-ERROR {97FEC89}>)
  8: (ERROR TYPE-ERROR)
  9: (SB-KERNEL::OBJECT-NOT-TYPE-ERROR-HANDLER #<unavailable argument>
#.(SB-SYS:INT-SAP #XF7B85478) #<SB-ALIEN-INTERNALS:ALIEN-VALUE :SAP #XF7B85168
:TYPE (* (SB-ALIEN:STRUCT SB-VM::OS-CONTEXT-T-STRUCT))> (14 78))
 10: (SB-KERNEL:INTERNAL-ERROR #.(SB-SYS:INT-SAP #XF7B85168) #<unavailable
argument>)
 11: ("foreign function: call_into_lisp")
 12: ("foreign function: funcall2")
 13: ("foreign function: interrupt_internal_error")
 14: ("foreign function: sigtrap_handler")
 15: (FDEFINITION #<PERSON {5DA9001}>)
 16: ((LAMBDA (IT.BESE.ARNESI::ARGUMENTS)) (#<PERSON {5E717B1}> #<PERSON
{5E717F9}>))
 17: (IT.BESE.ARNESI::DRIVE-INTERPRETER/CC #<CLOSURE (LAMBDA NIL) {97F8FAD}>)
 18: ((LAMBDA (IT.BESE.ARNESI::ARGUMENTS)) (#<PERSON-EDITOR {9E82281}> #<PERSON
{5DA9001}>))
 19: (IT.BESE.ARNESI::DRIVE-INTERPRETER/CC #<CLOSURE (LAMBDA NIL) {9769D35}>)
 20: ((LABELS IT.BESE.UCW::CALL-ACTION) #<CLOSURE (LAMBDA NIL) {9722B8D}>)
 21: ((SB-PCL::FAST-METHOD IT.BESE.UCW::SERVICE
(IT.BESE.UCW::STANDARD-SESSION-FRAME IT.BESE.UCW::REQUEST-CONTEXT))
#<unavailable argument> #<unavailable argument>
#<IT.BESE.UCW::STANDARD-SESSION-FRAME {976BA31}>
#<IT.BESE.UCW::COOKIE-SESSION-REQUEST-CONTEXT {974C031}>)

Alex


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


From a_bakic at yahoo.com  Sun Aug 28 00:35:51 2005
From: a_bakic at yahoo.com (Aleksandar Bakic)
Date: Sat, 27 Aug 2005 17:35:51 -0700 (PDT)
Subject: [Bese-devel] another presentation issue
In-Reply-To: <20050828003202.14830.qmail@web40601.mail.yahoo.com>
Message-ID: <20050828003551.75745.qmail@web40610.mail.yahoo.com>

Sorry, I forgot a piece:

The value #<PERSON {5DA9001}> is not of type (OR SYMBOL CONS).
   [Condition of type TYPE-ERROR]

(Wild guess: (OR SYMBOL CONS) comes from function name.)

Alex


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


From smoof-ra at elder-gods.org  Sun Aug 28 04:00:09 2005
From: smoof-ra at elder-gods.org (Larry D'Anna)
Date: Sun, 28 Aug 2005 00:00:09 -0400
Subject: [Bese-devel] Bug in UCW Examples: Add some numbers?
In-Reply-To: <m2y86nd4tc.fsf@soma.local>
References: <BAY103-F18EC5F4A80CCD8C385CE8ED5A80@phx.gbl>
	<m23bozw9oj.fsf@soma.local> <m24q9dce23.fsf@soma.local>
	<20050826133441.GA4420@melkor.elder-gods.org>
	<m2y86nd4tc.fsf@soma.local>
Message-ID: <20050828040009.GA24774@melkor.elder-gods.org>

* Marco Baringer (mb at bese.it) [050827 06:59]:
> (loop
>   repeat 5
>   do (call 'foo-bar))

If we had ucw:loop that translated this into 

(labels ((theloop (n)
                  (when (> n 0)
                    (call 'foo-bar)
                    (theloop (1- n)))))
  (theloop 5)))

everything would work great.


          --larry 


From mb at bese.it  Sun Aug 28 08:21:01 2005
From: mb at bese.it (Marco Baringer)
Date: Sun, 28 Aug 2005 10:21:01 +0200
Subject: [Bese-devel] yet another  minor typo fix
In-Reply-To: <4311C8C2.8010307@tech.coop> (Drew Crampsie's message of "Sun,
	28 Aug 2005 07:22:58 -0700")
References: <20050827224227.82832.qmail@web40628.mail.yahoo.com>
	<4311C8C2.8010307@tech.coop>
Message-ID: <m2wtm6a2vm.fsf@soma.local>

Drew Crampsie <drewc at tech.coop> writes:

> it's "text/css", not "test/css" :).

now i'm worried. i get the 'bug in get_extra' with this patch and the
:target patch you sent. :(

-- 
-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 Aug 28 16:59:13 2005
From: der_julian at web.de (Julian Stecklina)
Date: Sun, 28 Aug 2005 18:59:13 +0200
Subject: [Bese-devel] Compilation error with ACL 7
Message-ID: <20050828185913.3e86a073@localhost>

Hello,

I just tried to compile a current UCW with Allegro CL 7.0 on FreeBSD
and got:

;;; Compiling file /home/blitz/src/ucw/src/rerl/backtracking.lisp
;;; Writing fasl file /home/blitz/src/ucw/src/rerl/backtracking.fasl
Error: Class #<STANDARD-CLASS IT.BESE.ARNESI::APPLICATION-FORM> is not
yet finalized.
  [condition type: PROGRAM-ERROR]

Regards,
-- 
Julian Stecklina

(Of course SML does have its weaknesses, but by comparison, a
 discussion of C++'s strengths and flaws always sounds like an
 argument about whether one should face north or east when one
 is sacrificing one's goat to the rain god.)         -- Thant Tessman


From chalub at gmail.com  Sun Aug 28 17:06:10 2005
From: chalub at gmail.com (Fabricio Chalub)
Date: 28 Aug 2005 14:06:10 -0300
Subject: [Bese-devel] Problems with MP::STARTUP-IDLE-AND-TOP-LEVEL-LOOPS in
	start.lisp
Message-ID: <87slwuout9.fsf@eq.local>

Hi,

a side effect of adding STARTUP-IDLE-AND-TOP-LEVEL-LOOPS in start.lisp
is that CMUCL stops loading start.lisp at that point, since that
function creates a new REPL.  A search on Google shows that people
usually recommends that function as an addition to .CMUCL-INIT.LISP.

It seems that some further magic is needed for the correct MP
initialization.  A possible source of inspiration comes from
PORTEABLEASERVE's INSTALL.lisp that has been bit by this problem in the
past ('DOUBLE-KLUDGE', below).  Unfortunately, I can't retrieve the last
version from CVS as Sourceforge appears to be down for maintenance now.

---- begin of INSTALL.LISP cut
  ;; Startup multiprocessing.
  ;;
  ;; this isn't strictly necessary, but scheduling feels very coarse
  ;; before the evaluation of (startup-idle-and-top-level-loops) --
  ;; answer delays of about 1s per http request.
  ;;
  ;; KLUDGE: startup-idle-and-top-level-loops can only be evaluated
  ;; once, so we look for something resembling an existing idle loop
  ;; before invoking it.
  #||
  #+mp
  (unless (find-if
           #'(lambda (proc) (string= (mp:process-name proc) "Idle Loop"))
           (mp:all-processes))
    (mp::startup-idle-and-top-level-loops))
  ||#
  ;; DOUBLE KLUDGE: The preceding (commented-out) form caused the
  ;; loading of INSTALL.lisp to abort silently (!), so we do the
  ;; following, pilfered from eclipse the window manager:
  #+mp
  (setf mp::*idle-process* mp::*initial-process*)
---- end of INSTALL.LISP cut

So, apparently UCW should use

   #+(and cmu mp)
   (setf mp::*idle-process* mp::*initial-process*)

instead of the MP::S-I-A-T-L-L + FIND-IF call.

Cheers,
Fabricio



From drewc at tech.coop  Mon Aug 29 23:52:24 2005
From: drewc at tech.coop (Drew Crampsie)
Date: Mon, 29 Aug 2005 16:52:24 -0700
Subject: Quick Lisp-on-Lines tutorial (was Re: [Bese-devel] yet another
	question (suggestion?))
In-Reply-To: <87ek8fsuy9.fsf@flarge.here>
References: <87irxt3snr.fsf@flarge.here> <43106736.3090005@tech.coop>
	<87ek8fsuy9.fsf@flarge.here>
Message-ID: <43139FB8.6060300@tech.coop>

Friedrich Dominicus wrote:
> Drew Crampsie <drewc at tech.coop> writes:
> 
> 
>>Friedrich Dominicus wrote:

> 
> Here's the requested stuff. 
> 
> I have however quite few questions:
> I have a clsq database model. Is it possible to just exchange
> def-view-class with the here mentioend macro?

Might work, unless your model happens to use inheritance, in which case 
not without adding support for inheritance to LoL.

> 
> I tried to get hairierpub running, but without tweaking the files it
> failed here:

The version of hairiepub that is in that repo probably does not build. 
The whole thing is in a big state of flux right now, and has never been 
released anyway.

> I'm (more and more) sceptical about the usefullness of ucw and sorry
> hairiepub does not help to improve on my doubts.

Well, pulling pre-alpha software and expecting it to work out-of-the-box 
  is a little crazy.

[snipped whining]

> See my message about the read-only before a modifiable version. Nobody
> has explained it to me and I have worked three days on it to figure
> out what's going on without success. 

I don't know what post you are refering to, but the likely reason nobody 
has answered it is because it is not a "smart question". See 
http://www.catb.org/~esr/faqs/smart-questions.html for details about 
what i mean.

Also, when someone gives you a good detailed answer, responding as you 
did to this one will probably piss them off enough to not care about 
your problems anymore.

you're welcome.

drewc



From jan at rychter.com  Mon Aug 29 09:46:47 2005
From: jan at rychter.com (Jan Rychter)
Date: Mon, 29 Aug 2005 11:46:47 +0200
Subject: [Bese-devel] Minor fixes
Message-ID: <m2ll2l9it4.fsf@tnuctip.rychter.com>

Tiny annoyances, but one was causing an error when trying to use the
httpd backend.

--J.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: jwr-ucw-fixes.patch
Type: text/x-patch
Size: 14589 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050829/de522074/attachment.bin>

From frido at q-software-solutions.de  Mon Aug 29 10:07:55 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 29 Aug 2005 12:07:55 +0200
Subject: Quick Lisp-on-Lines tutorial (was Re: [Bese-devel] yet another
	question (suggestion?))
In-Reply-To: <43139FB8.6060300@tech.coop> (Drew Crampsie's message of "Mon,
	29 Aug 2005 16:52:24 -0700")
References: <87irxt3snr.fsf@flarge.here> <43106736.3090005@tech.coop>
	<87ek8fsuy9.fsf@flarge.here> <43139FB8.6060300@tech.coop>
Message-ID: <871x4d6op0.fsf@flarge.here>

Drew Crampsie <drewc at tech.coop> writes:

> Friedrich Dominicus wrote:
>> Drew Crampsie <drewc at tech.coop> writes:
>>
>>>Friedrich Dominicus wrote:
>
>> Here's the requested stuff. I have however quite few questions:
>> I have a clsq database model. Is it possible to just exchange
>> def-view-class with the here mentioend macro?
>
> Might work, unless your model happens to use inheritance, in which
> case not without adding support for inheritance to LoL.
>
>> I tried to get hairierpub running, but without tweaking the files it
>> failed here:
>
> The version of hairiepub that is in that repo probably does not
> build. The whole thing is in a big state of flux right now, and has
> never been released anyway.
Well I feel that is unfair, the README contains the following:
> HairiePub requires LISP-ON-LINES, cl-l10n and POSTGRESQL
> 
> Add the following lines to your /etc/postgres/pg_hba.conf
> 
> ## HairiePub User
> local   all         hairiepub                                             password
> host    all         hairiepub         127.0.0.1         255.255.255.255   password
> 
> And then create a user and a database in postgres :
> 
> template1=# CREATE USER hairiepub PASSWORD 'hp01';
>  CREATE USER
> template1=# CREATE DATABASE hairiepub OWNER hairiepub;
>  CREATE DATABASE
> 
> Once you've done that you are ready to create that database. There is a file 'database/sql-schema.lisp'. Load it into a running lisp (i use C-c C-k in slime) to create the database.
> 
> Make sure ASDF can see the system :
> 
> drewc at osiris:~/.sbcl/systems $ ln -s /home/drewc/src/hairiepub/hairiepub.asd /home/drewc/.sbcl/systems/
> 
> 
> once you've done that, it should be a simple matter of running './bin/ucwctl start && ./bin/ucwctl attach'
> 
> 
So I can't see that it's not supposed to work.
>
>> I'm (more and more) sceptical about the usefullness of ucw and sorry
>> hairiepub does not help to improve on my doubts.
>
> Well, pulling pre-alpha software and expecting it to work
> out-of-the-box is a little crazy.
Well then you should at least mention it. 

>
> [snipped whining]
Well it was an exact description on all the things going wrong. So why
do you start blaming me of whining? I send it to you because I though
you might be interested, sorry for misjudging you.



> I don't know what post you are refering to, but the likely reason
> nobody has answered it is because it is not a "smart question". See
> http://www.catb.org/~esr/faqs/smart-questions.html for details about
> what i mean.
Well I repeat this question that you got an idea. I don'f feel that
the question was too dump. 
(defaction shop-login ((login shop-login) username password)
  (setf (login.username login) username
        (login.password login) password)
  (let* ((user-data (check-credentials login))
         (shipping-adr (make-instance 'shipping-adr :adr (car (qss.db::shipping-address user-data)))))
        ;; car is needed because a join slot always returns a list!     
    (inspect login)
    (inspect shipping-adr)
  (if shipping-adr
      (call 'customer-input-page
            :customer-data user-data
            :address shipping-adr
            :payment-details nil)
      (call 'info-message 
            :message "Login failed"))))

The shipping-adr class is defined this way:
(defcomponent address-mixin ()
  ((adr :initarg :adr :accessor adr
        :backtrack t
        :initform (make-instance 'qss.db::shipping-address)))
  (:metaclass standard-component-class))

;; (forget-class :shipping-adr)
(defclass shipping-adr (address-mixin)
  ()
  (:metaclass standard-component-class))

If I inspect login I see the following:
0. CONTINUATION: (IT.BESE.ARNESI::TOPLEVEL-K)
1. CALLING-COMPONENT: #<TRANS-INFO {1001F24851}>
2. PLACE: #<IT.BESE.UCW::PLACE (CONTEXT.WINDOW-COMPONENT
                                *CONTEXT*) {1001391471}>

that's ok, the shipping-adr has the following
0. CONTINUATION: "unbound"
1. CALLING-COMPONENT: "unbound"
2. PLACE: "unbound"

which seems itself reasonable, but then I do the following
(defmethod render-on ((res response) (customer-page customer-input-page))
  (<:h1 "Your adress data: ")
  (render-on res (address customer-page)))

which then calls this:
(defmethod render-on ((res response) (adr shipping-adr))
  (let ((address (adr adr)))
    (inspect adr)
    (<ucw:form :action (edit-address adr)
               (<:table 
                (dolist (item '(street zip city country email))
                  (present-a-slot-as-table-row item (slot-value address item))))
               (<:p
                (<ucw:submit :action (edit-address adr)
                             :value "Edit")))))

This is rendered fine.

But then I push the button and got:
The slot IT.BESE.UCW::PLACE is unbound in the object #<SHIPPING-ADR {10017C2171}>.
   [Condition of type UNBOUND-SLOT]

Restarts:
  0: [SHOW-BACKTRACE] Send the client a backtrace page.
  1: [RETRY] Clearout the response and retry calling the action.
  2: [SERVER-ERROR] Send the client an internal server error page.
  3: [GENERATE-BACKTRACE-FOR-EMACS] Generate a bug report in Emacs.
  

That's of course true, but not what I like to see of course I'd like
to call the next action and go on from there. 

So how am I supposed to build such a action-render-render-action 
chain?

what i want is:
checking the login credentials
if the the customer can be found then show me his/her address
if the data have changed allow to change them 
change the data 
display the changed data
go on wiht the state as if he/she had not changed the address
display e.g the payment terms the customer has taken.
allow changes as needed 
etc.

Help would be very appreciated.





>
> Also, when someone gives you a good detailed answer, responding as you
> did to this one will probably piss them off enough to not care about
> your problems anymore.
Well I can not see in any message that I'd made someone down. Oh, well
I asked dump question also but if I thought I had to ask them one can
assume that I had my reasons. Now I'm getting a bit angry about your
wording. I did spend a lot of time getting things going. I wrote > 30
k code to implement a shop solution. It took me a lot of work and I run
into all kind of troubles. What I especially dislike about your
wording is that you insnuate that I did not try. And what I dislike
also is that it seems one is not allowed to criticize things. You made
your stuff available as example I suppose. I tried it. I followed you
instructions but it turned out that it won't work the intended way, I
send you a long reply about the toubles, but instead of saying. "yeah,
there are problems" are you starting flaming around. If you feel that
is ok, well go for it. I won't bother you again.

Friedrich



From mb at bese.it  Mon Aug 29 11:55:27 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 29 Aug 2005 13:55:27 +0200
Subject: [Bese-devel] UTF-8
In-Reply-To: <871x4fcu1q.fsf@binarin.domain> (Alexey Lebedeff's message of
	"Sat, 27 Aug 2005 18:51:13 +0400")
References: <871x4fcu1q.fsf@binarin.domain>
Message-ID: <m2ll2lkle8.fsf@soma.local>

Alexey Lebedeff <binarin at agava.com> writes:

> Changes to ucw:
>
> - Allow to specify various external-formats for different purposes:
>   unescaping of urls, communication with slime and http.

it seems that there's an encoding issue in my mailer (or
someplace). would you mind gzip'ing this (so that mailers will leave
the contents alone) and resending it as an attachement?

-- 
-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 Aug 29 11:56:11 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 29 Aug 2005 13:56:11 +0200
Subject: [Bese-devel] yet another  minor typo fix
In-Reply-To: <4311C8C2.8010307@tech.coop> (Drew Crampsie's message of "Sun,
	28 Aug 2005 07:22:58 -0700")
References: <20050827224227.82832.qmail@web40628.mail.yahoo.com>
	<4311C8C2.8010307@tech.coop>
Message-ID: <m2hdd9kld0.fsf@soma.local>

Drew Crampsie <drewc at tech.coop> writes:

> it's "text/css", not "test/css" :).

would you mind reseding the patch as an attachement (so that hopefully
the various mailers involved won't screw up the contents of the
patch)?

-- 
-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 binarin at agava.com  Mon Aug 29 12:00:23 2005
From: binarin at agava.com (Alexey Lebedeff)
Date: Mon, 29 Aug 2005 16:00:23 +0400
Subject: [Bese-devel] UTF-8
In-Reply-To: <m2ll2lkle8.fsf@soma.local> (Marco Baringer's message of "Mon,
	29 Aug 2005 13:55:27 +0200")
References: <871x4fcu1q.fsf@binarin.domain> <m2ll2lkle8.fsf@soma.local>
Message-ID: <87y86lq7fs.fsf@binarin.domain>

## On Mon, 29 Aug 2005 13:55:27 +0200
## you wrote:

 >> Changes to ucw:
 >>
 >> - Allow to specify various external-formats for different purposes:
 >>   unescaping of urls, communication with slime and http.

 MB> it seems that there's an encoding issue in my mailer (or
 MB> someplace). would you mind gzip'ing this (so that mailers will leave
 MB> the contents alone) and resending it as an attachement?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ucw.CHANGESET.gz
Type: application/octet-stream
Size: 5411 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050829/aae942a4/attachment.obj>
-------------- next part --------------



-- 
[1180'99:11-6?] [BMSTU_-_fuck'??????_??:5-111]

From mb at bese.it  Mon Aug 29 12:05:53 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 29 Aug 2005 14:05:53 +0200
Subject: [Bese-devel] UTF-8
In-Reply-To: <87y86lq7fs.fsf@binarin.domain> (Alexey Lebedeff's message of
	"Mon, 29 Aug 2005 16:00:23 +0400")
References: <871x4fcu1q.fsf@binarin.domain> <m2ll2lkle8.fsf@soma.local>
	<87y86lq7fs.fsf@binarin.domain>
Message-ID: <m24q99kkwu.fsf@soma.local>


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 slawek.zak at gmail.com  Mon Aug 29 12:33:09 2005
From: slawek.zak at gmail.com (Slawek Zak)
Date: Mon, 29 Aug 2005 14:33:09 +0200
Subject: [Bese-devel] Compilation error with ACL 7
In-Reply-To: <20050828185913.3e86a073@localhost>
References: <20050828185913.3e86a073@localhost>
Message-ID: <787bbe1c050829053341ca2c75@mail.gmail.com>

On 8/28/05, Julian Stecklina <der_julian at web.de> wrote:
> Hello,
> 
> I just tried to compile a current UCW with Allegro CL 7.0 on FreeBSD
> and got:
> 
> ;;; Compiling file /home/blitz/src/ucw/src/rerl/backtracking.lisp
> ;;; Writing fasl file /home/blitz/src/ucw/src/rerl/backtracking.fasl
> Error: Class #<STANDARD-CLASS IT.BESE.ARNESI::APPLICATION-FORM> is not
> yet finalized.
>   [condition type: PROGRAM-ERROR]

It's a known problem. I've reported it to Marco. You can get around it
by finalizing the classes manually during compilation with
finalize-inheritance.

/S


From carlos.ungil at bluewin.ch  Mon Aug 29 15:35:12 2005
From: carlos.ungil at bluewin.ch (Carlos Ungil)
Date: Mon, 29 Aug 2005 17:35:12 +0200
Subject: Quick Lisp-on-Lines tutorial (was Re: [Bese-devel] yet another
	question (suggestion?))
In-Reply-To: <871x4d6op0.fsf@flarge.here>
References: <87irxt3snr.fsf@flarge.here> <43106736.3090005@tech.coop>
	<87ek8fsuy9.fsf@flarge.here> <43139FB8.6060300@tech.coop>
	<871x4d6op0.fsf@flarge.here>
Message-ID: <1f84f5672c39cf8e710095dd07a21bfe@bluewin.ch>

Hello,

On Aug 29, 2005, at 12:07 PM, Friedrich Dominicus wrote:
> Well I feel that is unfair, the README contains the following:
>> HairiePub requires LISP-ON-LINES, cl-l10n and POSTGRESQL
>> ....
> So I can't see that it's not supposed to work.
>> Well, pulling pre-alpha software and expecting it to work
>> out-of-the-box is a little crazy.
> Well then you should at least mention it.

Given that hairypub is based on lisp-on-lines, and lisp-on-lines' 
README file states that
 > This is a very early release, and there are more bugs [than] docs. 
HERE BE DRAGONS!
you could have imagined that it wouldn't be terrible reliable.

Cheers,

Carlos

PD: I hope Drew and Marco won't stop publishing their code for fear of 
complaints about incomplete or inaccurate documentation.




From frido at q-software-solutions.de  Mon Aug 29 16:28:14 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Mon, 29 Aug 2005 18:28:14 +0200
Subject: Quick Lisp-on-Lines tutorial (was Re: [Bese-devel] yet another
	question (suggestion?))
In-Reply-To: <1f84f5672c39cf8e710095dd07a21bfe@bluewin.ch> (Carlos Ungil's
	message of "Mon, 29 Aug 2005 17:35:12 +0200")
References: <87irxt3snr.fsf@flarge.here> <43106736.3090005@tech.coop>
	<87ek8fsuy9.fsf@flarge.here> <43139FB8.6060300@tech.coop>
	<871x4d6op0.fsf@flarge.here>
	<1f84f5672c39cf8e710095dd07a21bfe@bluewin.ch>
Message-ID: <87mzn06735.fsf@flarge.here>

Yes if you see it that way, you're right. So I better stay away from
both I guess.

Regards
Friedrich


From mb at bese.it  Mon Aug 29 17:18:41 2005
From: mb at bese.it (Marco Baringer)
Date: Mon, 29 Aug 2005 19:18:41 +0200
Subject: [Bese-devel] Minor fixes
In-Reply-To: <m2ll2l9it4.fsf@tnuctip.rychter.com> (Jan Rychter's message of
	"Mon, 29 Aug 2005 11:46:47 +0200")
References: <m2ll2l9it4.fsf@tnuctip.rychter.com>
Message-ID: <m2zmr04q6m.fsf@soma.local>

Jan Rychter <jan at rychter.com> writes:

> New patches:
>
> [Add a missing argument.
> Jan Rychter <jan at rychter.com>**20050829094338] {
> hunk ./src/backend/httpd.lisp 500
> -                                                  headers body)
> +                                                  (cdr it) headers body)
> }

Aleksandar Bakic's previous patch had already fixed this (the patch is
infact exactly the same). what's the best thing (as far as using darcs
is concrened) to do here, apply yours and the record a trivial
merge-conflict patch or just not apply yours?

-- 
-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 jan at rychter.com  Mon Aug 29 19:44:14 2005
From: jan at rychter.com (Jan Rychter)
Date: Mon, 29 Aug 2005 21:44:14 +0200
Subject: [Bese-devel] Minor fixes
In-Reply-To: <m2zmr04q6m.fsf@soma.local> (Marco Baringer's message of "Mon, 29
	Aug 2005 19:18:41 +0200")
References: <m2ll2l9it4.fsf@tnuctip.rychter.com> <m2zmr04q6m.fsf@soma.local>
Message-ID: <m2y86k7ckx.fsf@tnuctip.rychter.com>

Marco Baringer wrote:
> Jan Rychter <jan at rychter.com> writes:
> > New patches:
> >
> > [Add a missing argument.
> > Jan Rychter <jan at rychter.com>**20050829094338] {
> > hunk ./src/backend/httpd.lisp 500
> > -                                                  headers body)
> > +                                                  (cdr it) headers body)
> > }
> 
> Aleksandar Bakic's previous patch had already fixed this (the patch is
> infact exactly the same). what's the best thing (as far as using darcs
> is concrened) to do here, apply yours and the record a trivial
> merge-conflict patch or just not apply yours?

No idea about darcs, but if I were you I'd just drop my patch.

--J.


From tim at fractaldragon.net  Tue Aug 30 05:59:06 2005
From: tim at fractaldragon.net (Tim Lavoie)
Date: Tue, 30 Aug 2005 00:59:06 -0500
Subject: [Bese-devel] Getting at higher-level state in contained components
Message-ID: <20050830055905.GA15494@fractaldragon.net>

Hi all,

I've got a window component defined, where there are pieces which I
would like to use to control other components, or the top-level
window.

It starts out basically like this:

(defcomponent foo-window (template-component simple-window-component)
  ((nav1 :component side-nav1 :accessor foo-window.nav1)
   (body :component simple-container :accessor foo-window.body)
   (profile :component profile :accessor ww-profile))

  ........

The body is the main content area, nav1 would be something like the
typical side-bar navigation area, and profile is a user profile which
I've changed from defaults to results from an SQL query once they've
logged in.

What I would like to do ideally is get a reference in one component to
another, without manually hard-coding the path to use the appropriate
number of parent links. This does work, but is fragile in the face of
changing page structure. This allows me already to have navigation
links do something to the main content area as an example, or get
something from the profile.

Is there a nice way to define this sort of find-in-parents behaviour
already? It seems obvious, but I didn't find anything of the sort
already. I think it would work well and allow arbitrary nesting
without changing template code, which would be quite nice. Any
suggestions?

	Thanks,
	Tim



From pjb at informatimago.com  Tue Aug 30 07:34:55 2005
From: pjb at informatimago.com (Pascal Bourguignon)
Date: Tue, 30 Aug 2005 09:34:55 +0200 (CEST)
Subject: [Bese-devel] can we do coroutines with cps?
Message-ID: <20050830073455.A741510F9F6@thalassa.informatimago.com>


With cps from ucw-0.3.9, I'm trying to implement coroutines suspend/resume:

(defparameter *coroutine* (cons t t))

(defun/cc resume ()
  (print `(:resume-1 *coroutine* = ,*coroutine*))
  (setf (cdr *Coroutine*) t)
  (call/cc (car *coroutine*))
  (print `(:resume-2 *coroutine* = ,*coroutine*)))

(defun/cc suspend (escape)
  (print `(:suspend-1 *coroutine* = ,*coroutine*))
  (setf (car *Coroutine*) (progn (setf (cdr *Coroutine*) nil)
                                 (call/cc (function identity))))
  (print `(:suspend-2 *coroutine* = ,*coroutine*))
  (unless (cdr *coroutine*)
    (funcall escape nil))
  (print `(:suspend-3 *coroutine* = ,*coroutine*)))

(defun/cc routine (escape)
  (let ((i 0))
    (loop
       (print `(:routine *coroutine* = ,*coroutine*))
       (print (incf i))
       (suspend escape))))

(defun/cc f (dummy)
  (print `(:f *coroutine* = ,*coroutine*))
  (call/cc (function routine)))

(with-call/cc
    (progn   (setf (car *Coroutine*) (function f))
             (loop repeat 4 do (print :before) (resume) (print :after))))

Unfortunately this doesn't work: routine is not resumed.  It seems the
(call/cc (function identity)) doens't return the current continuation,
but jumps to the end of with-call/cc:

:BEFORE 
(:RESUME-1 *COROUTINE* =
 (#<FUNCTION F (DUMMY) (DECLARE (SYSTEM::IN-DEFUN F))
    (BLOCK F
     (LET ((#:K-12203 (OR IT.BESE.ARNESI::*K* #'IT.BESE.ARNESI::TOPLEVEL-K)))
      (DECLARE (IGNORABLE #:K-12203))
      (IT.BESE.ARNESI::DRIVE-CPS
       (PROGN (PRINT (LIST ':F '*COROUTINE* '= *COROUTINE*))
        (LAMBDA NIL
         (FUNCALL #'IT.BESE.ARNESI::TOPLEVEL-K
          (FUNCALL #'ROUTINE
           (IT.BESE.ARNESI::MAKE-CALL/CC-K #:K-12203))))))))>)) 
(:F *COROUTINE* =
 (#<FUNCTION F (DUMMY) (DECLARE (SYSTEM::IN-DEFUN F))
    (BLOCK F
     (LET ((#:K-12203 (OR IT.BESE.ARNESI::*K* #'IT.BESE.ARNESI::TOPLEVEL-K)))
      (DECLARE (IGNORABLE #:K-12203))
      (IT.BESE.ARNESI::DRIVE-CPS
       (PROGN (PRINT (LIST ':F '*COROUTINE* '= *COROUTINE*))
        (LAMBDA NIL
         (FUNCALL #'IT.BESE.ARNESI::TOPLEVEL-K
          (FUNCALL #'ROUTINE (IT.BESE.ARNESI::MAKE-CALL/CC-K #:K-12203))))))))>
  . T)) 
(:ROUTINE *COROUTINE* =
 (#<FUNCTION F (DUMMY) (DECLARE (SYSTEM::IN-DEFUN F))
    (BLOCK F
     (LET ((#:K-12203 (OR IT.BESE.ARNESI::*K* #'IT.BESE.ARNESI::TOPLEVEL-K)))
      (DECLARE (IGNORABLE #:K-12203))
      (IT.BESE.ARNESI::DRIVE-CPS
       (PROGN (PRINT (LIST ':F '*COROUTINE* '= *COROUTINE*))
        (LAMBDA NIL
         (FUNCALL #'IT.BESE.ARNESI::TOPLEVEL-K
          (FUNCALL #'ROUTINE (IT.BESE.ARNESI::MAKE-CALL/CC-K #:K-12203))))))))>
  . T)) 
1 
(:SUSPEND-1 *COROUTINE* =
 (#<FUNCTION F (DUMMY) (DECLARE (SYSTEM::IN-DEFUN F))
    (BLOCK F
     (LET ((#:K-12203 (OR IT.BESE.ARNESI::*K* #'IT.BESE.ARNESI::TOPLEVEL-K)))
      (DECLARE (IGNORABLE #:K-12203))
      (IT.BESE.ARNESI::DRIVE-CPS
       (PROGN (PRINT (LIST ':F '*COROUTINE* '= *COROUTINE*))
        (LAMBDA NIL
         (FUNCALL #'IT.BESE.ARNESI::TOPLEVEL-K
          (FUNCALL #'ROUTINE (IT.BESE.ARNESI::MAKE-CALL/CC-K #:K-12203))))))))>
  . T)) 
#<COMPILED-FUNCTION IT.BESE.ARNESI::MAKE-CALL/CC-K-1>


Is it possible to do co-routines with arnesi cps?

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
Kitty like plastic.
Confuses for litter box.
Don't leave tarp around.


From mb at bese.it  Tue Aug 30 13:13:36 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 30 Aug 2005 15:13:36 +0200
Subject: [Bese-devel] can we do coroutines with cps?
In-Reply-To: <20050830073455.A741510F9F6@thalassa.informatimago.com> (Pascal
	Bourguignon's message of "Tue, 30 Aug 2005 09:34:55 +0200 (CEST)")
References: <20050830073455.A741510F9F6@thalassa.informatimago.com>
Message-ID: <m2wtm336v3.fsf@soma.local>

Pascal Bourguignon <pjb at informatimago.com> writes:

> Unfortunately this doesn't work: routine is not resumed.  It seems the
> (call/cc (function identity)) doens't return the current continuation,
> but jumps to the end of with-call/cc:

there's a variable, *call/cc-returns* which controls this
behaviour. try setting it t T and (after a recompile since it has
effect at macro expansion time) see what happens.

this should work (i consider this email a bug report), i'm now trying
to figure out why the interpeter breaks on 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 jan at rychter.com  Tue Aug 30 13:33:37 2005
From: jan at rychter.com (Jan Rychter)
Date: Tue, 30 Aug 2005 15:33:37 +0200
Subject: [Bese-devel] UCW guidance
Message-ID: <m2r7cb5z2m.fsf@tnuctip.rychter.com>

I need some guidance on what is the best way to design UCW applications:

1. I needed to enclose some components within a TAL template page. There
   is a menu/navigation bar and the main application area. I ended up
   defining a window object subclassing window-component and
   template-component. So far so good. But now I have two
   simple-container slots, one for the menu and one for the body. Each
   container slot in turn contains the menu component and the component
   that is the current body. This extra level of indirection (the
   simple-containers) seems unnecessary, but without it I couldn't
   figure out how to switch the body component. Any advice?

2. I'm using (defmethod shared-initialize :after...) to add my
   subcomponents to the main window-component (copied from the
   examples). Surely there has to be a better way?

3. When you have a three-step-process (say, a shopping cart checkout),
   would you rather implement three separate components that call each
   other or one component with internal (backtracked) state, going
   through the three states?

4. I believe this question is similar to the one posted earlier by Tim
   Lavoie. I ended up writing the following piece of code:

(defaction switch-to ((component main-menu) new-label)
  (setf (selected-item component) new-label)
  (switch-component (main.body (ucw::parent (ucw::parent component))) new-label))

Needless to say, I find this to be rather ugly. There are two
ucw::parent calls because of the extra level of indirection that I had
to design in (see question 1), but even if it wasn't for that, I'd need
at least one ucw::parent call. There surely is a better way?

Also, I've just been bitten by the cps transformer -- I was trying to
generate an <ucw:action inside a loop clause. Took me a while to find
out that using dolist instead of loop solves my problem... (I guess this
is to be expected in the ucw_dev branch and will change with the cps
interpreter?)

Any advice would be much appreciated!

thanks,
--J.


From mb at bese.it  Tue Aug 30 15:14:53 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 30 Aug 2005 17:14:53 +0200
Subject: [Bese-devel] Getting at higher-level state in contained components
In-Reply-To: <20050830055905.GA15494@fractaldragon.net> (Tim Lavoie's message
	of "Tue, 30 Aug 2005 00:59:06 -0500")
References: <20050830055905.GA15494@fractaldragon.net>
Message-ID: <m2acizzcb6.fsf@soma.local>

Tim Lavoie <tim at fractaldragon.net> writes:

> Hi all,
>
> I've got a window component defined, where there are pieces which I
> would like to use to control other components, or the top-level
> window.
>
> It starts out basically like this:
>
> (defcomponent foo-window (template-component simple-window-component)
>   ((nav1 :component side-nav1 :accessor foo-window.nav1)
>    (body :component simple-container :accessor foo-window.body)
>    (profile :component profile :accessor ww-profile))
>
>   ........
>
> The body is the main content area, nav1 would be something like the
> typical side-bar navigation area, and profile is a user profile which
> I've changed from defaults to results from an SQL query once they've
> logged in.
>
> What I would like to do ideally is get a reference in one component to
> another, without manually hard-coding the path to use the appropriate
> number of parent links. This does work, but is fragile in the face of
> changing page structure. This allows me already to have navigation
> links do something to the main content area as an example, or get
> something from the profile.

at some point you're going to be forced to tell one component about
the other component, and this is going to require, on some level, a
hard coded path. that said it'd be easier for the nav component if you
informed it about the body component, instead of having the nav bar's
code littered with calls to (body (parent self)).

(defcomponent navbar ()
  ((nav-tree)
   (current-node)
   (body)))

(defmethod render-on ((res response) (navbar navbar))
  (dolist* (node nav-tree)
    (<ucw:a :action (switch-component node (body navbar))
            (<:as-html (title node)))))

and then in the window's shard-initialize tell the navbar what
component it should control:

(defmethod shard-initialize :after ((foo foo-window) slot-names &rest ignore)
  (setf (body (navbar foo)) (body foo)))

could that work? (i'm not really sure what answer you want here (and
there are 1e20 possibilities) so i'm just guessing).

> Is there a nice way to define this sort of find-in-parents behaviour
> already? It seems obvious, but I didn't find anything of the sort
> already. I think it would work well and allow arbitrary nesting
> without changing template code, which would be quite nice. Any
> suggestions?

could you be a bit more precise about what you want? we already have
the parent accessor (every component knows what component its nested
in) but something a well done navbar component would be a good idea as
well. of course, it'd be built on top of a well done tree-view
component...

-- 
-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 Aug 30 15:35:07 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 30 Aug 2005 17:35:07 +0200
Subject: [Bese-devel] UCW guidance
In-Reply-To: <m2r7cb5z2m.fsf@tnuctip.rychter.com> (Jan Rychter's message of
	"Tue, 30 Aug 2005 15:33:37 +0200")
References: <m2r7cb5z2m.fsf@tnuctip.rychter.com>
Message-ID: <m2d5nvxwt0.fsf@soma.local>

Jan Rychter <jan at rychter.com> writes:

> I need some guidance on what is the best way to design UCW applications:
>
> 1. I needed to enclose some components within a TAL template page. There
>    is a menu/navigation bar and the main application area. I ended up
>    defining a window object subclassing window-component and
>    template-component. So far so good. But now I have two
>    simple-container slots, one for the menu and one for the body. Each
>    container slot in turn contains the menu component and the component
>    that is the current body. This extra level of indirection (the
>    simple-containers) seems unnecessary, but without it I couldn't
>    figure out how to switch the body component. Any advice?

why make the navbar a container? could something like the component in
the other mail (the one i just sent to Tim Lavoie) work? the body
component you pass to navbar need not even be a contianer, all it
needs is a switch-component method.

> 2. I'm using (defmethod shared-initialize :after...) to add my
>    subcomponents to the main window-component (copied from the
>    examples). Surely there has to be a better way?

the example-window should be written like this:

(defcomponent example-window (simple-window-component)
  ((body :component
         (simple-container
          :current-component-name 'example-welcome
          :contents
          `((example-welcome .     ,(make-instance 'example-welcome))
            (counter .             ,(make-instance 'counter))
            (transaction-example . ,(make-instance 'transaction-example))
            (example-form .        ,(make-instance 'example-form))
            (sum .                 ,(make-instance 'sum))
            (presentations-example . ,(make-instance 'presentations-example))
            (file-upload-example . ,(make-instance 'file-upload-example))))
         :accessor example-window.body)))

no, i don't know why it's not already done that way and yes, this
could be improved on.

> 3. When you have a three-step-process (say, a shopping cart checkout),
>    would you rather implement three separate components that call each
>    other or one component with internal (backtracked) state, going
>    through the three states?

personally i prefer the former (three components) especially since
people (like those pesky clients) keep changing the order of things
and it's easier for me to move three seperate component around in a
single action than change the state handling in a single component.

> 4. I believe this question is similar to the one posted earlier by Tim
>    Lavoie. I ended up writing the following piece of code:
>
> (defaction switch-to ((component main-menu) new-label)
>   (setf (selected-item component) new-label)
>   (switch-component (main.body (ucw::parent (ucw::parent component))) new-label))
>
> Needless to say, I find this to be rather ugly. There are two
> ucw::parent calls because of the extra level of indirection that I had
> to design in (see question 1), but even if it wasn't for that, I'd need
> at least one ucw::parent call. There surely is a better way?

i'd do it like so:

(defaction switch-to ((menu main-menu) new-label)
  (setf (selected-item menu) new-label
        (container.current-component-name (controlled-component menu)) label))

setting up the controlled-component slot should be done in the
shared-initialize method of the component which contains the navbar
(and presumably the 'content' component).

if this seems like a good idea (at the very least better than the
nothing we have now) i'll implement it in the example app.

> Also, I've just been bitten by the cps transformer -- I was trying to
> generate an <ucw:action inside a loop clause. Took me a while to find
> out that using dolist instead of loop solves my problem... (I guess this
> is to be expected in the ucw_dev branch and will change with the cps
> interpreter?)

no, the 'fresh-binding' vs. 'one-binding+modification' is a buch
deeper issue i can't really "fix." as odd as it may seem the current
effect is what is supposed to happen.

> Any advice would be much appreciated!

i hope this helps. the more info (doubts, questions, other
possiblities) the better i can help and the more ideas we can bounce
around.

i think i've said this in the past but i'll repeat it here just in
case. component based continuation enabled web frameworks (libraries,
platform whatever the hell they're called today) are young (ucw
especially so). i do not know what the 'right' answer to these
questions could be: never ever believe that i know what i'm talking
about.

-- 
-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 Aug 30 15:52:31 2005
From: mb at bese.it (Marco Baringer)
Date: Tue, 30 Aug 2005 17:52:31 +0200
Subject: [Bese-devel] patches
Message-ID: <m2psrvwhfk.fsf@soma.local>


i just push'd all the patches i've recieved in the past few days
(arnesi, ucw and yaclml stuff). i made a bit a mess of the emails and
the patches and i'm pretty sure i forgot something, not that i can
figure out what... if there's a patch you sent which i haven't applied
would you mind reminding me about it?

p.s. - how cool is it to have this 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 asbjxrn at bjxrnstad.net  Tue Aug 30 16:28:03 2005
From: asbjxrn at bjxrnstad.net (=?ISO-8859-1?Q?Asbj=F8rn_Bj=F8rnstad?=)
Date: Wed, 31 Aug 2005 00:28:03 +0800
Subject: [Bese-devel] UCW guidance
In-Reply-To: <m2d5nvxwt0.fsf@soma.local>
References: <m2r7cb5z2m.fsf@tnuctip.rychter.com> <m2d5nvxwt0.fsf@soma.local>
Message-ID: <531E4BD0-FA00-4535-A685-6F0637C5AE18@bjxrnstad.net>


On 30 Aug 2005, at 11:35 PM, Marco Baringer wrote:

> Jan Rychter <jan at rychter.com> writes:

>> 4. I believe this question is similar to the one posted earlier by  
>> Tim
>>    Lavoie. I ended up writing the following piece of code:
>>
>> (defaction switch-to ((component main-menu) new-label)
>>   (setf (selected-item component) new-label)
>>   (switch-component (main.body (ucw::parent (ucw::parent  
>> component))) new-label))
>>
>> Needless to say, I find this to be rather ugly. There are two
>> ucw::parent calls because of the extra level of indirection that I  
>> had
>> to design in (see question 1), but even if it wasn't for that, I'd  
>> need
>> at least one ucw::parent call. There surely is a better way?
>>
>
> i'd do it like so:
>
> (defaction switch-to ((menu main-menu) new-label)
>   (setf (selected-item menu) new-label
>         (container.current-component-name (controlled-component  
> menu)) label))
>
> setting up the controlled-component slot should be done in the
> shared-initialize method of the component which contains the navbar
> (and presumably the 'content' component).
>
> if this seems like a good idea (at the very least better than the
> nothing we have now) i'll implement it in the example app.


The way I've done it is to have a slot "top-pane" in all components  
and work
my way down the tree. That way I can access different containers  
depending
on need. For example switching the content-pane becomes something like:

(defmacro bodyswitch-link (label component)
   `(<ucw:a :action (switch-component (content-pane (top-pane  
page)) ,component)
                     ,label))

(defmethod render-on ((res response) (page guest-navigation))
    (<:ul :id "contextnav"
       (<:li (bodyswitch-link "login" 'login))
       (<:li (bodyswitch-link "register" 'register-user)))

(Refactored from my real code, and not tested, but you get the Idea.)

I also switch menus and other components that are not in the body pane,
it becomes more tricky if the tree of components/containers becomes more
complex, but so far it's working.
-- 
   -asbjxrn



From tim at fractaldragon.net  Tue Aug 30 16:40:30 2005
From: tim at fractaldragon.net (Tim Lavoie)
Date: Tue, 30 Aug 2005 11:40:30 -0500
Subject: [Bese-devel] Getting at higher-level state in contained components
In-Reply-To: <m2acizzcb6.fsf@soma.local>
References: <20050830055905.GA15494@fractaldragon.net>
	<m2acizzcb6.fsf@soma.local>
Message-ID: <20050830164030.GA30979@fractaldragon.net>

On Tue, Aug 30, 2005 at 05:14:53PM +0200, Marco Baringer wrote:
> >
> > What I would like to do ideally is get a reference in one component to
> > another, without manually hard-coding the path to use the appropriate
> > number of parent links. This does work, but is fragile in the face of
> > changing page structure. This allows me already to have navigation
> > links do something to the main content area as an example, or get
> > something from the profile.
> 
> at some point you're going to be forced to tell one component about
> the other component, and this is going to require, on some level, a
> hard coded path. that said it'd be easier for the nav component if you
> informed it about the body component, instead of having the nav bar's
> code littered with calls to (body (parent self)).

Yeah, that's why I asked. I had tested out the idea with some (body
(parent self)) -type code in the TAL template for the nav component,
but wasn't thrilled with it.

> 
> (defcomponent navbar ()
>   ((nav-tree)
>    (current-node)
>    (body)))
> 
> (defmethod render-on ((res response) (navbar navbar))
>   (dolist* (node nav-tree)
>     (<ucw:a :action (switch-component node (body navbar))
>             (<:as-html (title node)))))
> 
> and then in the window's shard-initialize tell the navbar what
> component it should control:
> 
> (defmethod shard-initialize :after ((foo foo-window) slot-names &rest ignore)
>   (setf (body (navbar foo)) (body foo)))
> 
> could that work? (i'm not really sure what answer you want here (and
> there are 1e20 possibilities) so i'm just guessing).


Yes, I think that should work fine, and is certainly more direct than
what I had been thinking; rather than simply coding in the reference
as you showed here, I started playing with the idea of a method to
follow the chain of (parent self) calls until either the top is
reached, or the desired component found.

I suspect I won't be doing the sort of nesting to make this approach
really needed, so I'll try your way first.


> > Is there a nice way to define this sort of find-in-parents behaviour
> > already? It seems obvious, but I didn't find anything of the sort
> > already. I think it would work well and allow arbitrary nesting
> > without changing template code, which would be quite nice. Any
> > suggestions?
> 
> could you be a bit more precise about what you want? we already have
> the parent accessor (every component knows what component its nested
> in) but something a well done navbar component would be a good idea as
> well. of course, it'd be built on top of a well done tree-view
> component...

Heh. I'll try the simple and expedient method for the time being, but
you are right, there is room for a couple of helpful widgets.

	Thanks,	
	Tim

-- 
I've finally learned what "upward compatible" means.  It means we get to
keep all our old mistakes.
                -- Dennie van Tassel


From jan at rychter.com  Wed Aug 31 07:30:25 2005
From: jan at rychter.com (Jan Rychter)
Date: Wed, 31 Aug 2005 09:30:25 +0200
Subject: [Bese-devel] UCW file upload example
In-Reply-To: <m21x5si5fy.fsf@soma.local> (Marco Baringer's message of "Thu, 21
	Jul 2005 16:45:05 +0200")
References: <E1Dspfs-0002Qm-EN@e4ward.com> <m21x5si5fy.fsf@soma.local>
Message-ID: <m2slwq36ni.fsf@tnuctip.rychter.com>

[replying to a fairly old thread...]

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

 Marco> 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.

 Marco> no, it definetly can't. but even if it could wouldn't it be
 Marco> better to use something like ftp (or any protocol which allows
 Marco> 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.

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

Has anybody worked on this? I've just hit the same problem.

On a gigabit or fast ethernet LAN uploading files 1GB in size really
isn't an issue -- and users really want it to be simple (meaning: no
FTP, just in-browser buttons).

--J.


From mb at bese.it  Wed Aug 31 11:49:56 2005
From: mb at bese.it (Marco Baringer)
Date: Wed, 31 Aug 2005 13:49:56 +0200
Subject: [Bese-devel] UCW file upload example
In-Reply-To: <m2slwq36ni.fsf@tnuctip.rychter.com> (Jan Rychter's message of
	"Wed, 31 Aug 2005 09:30:25 +0200")
References: <E1Dspfs-0002Qm-EN@e4ward.com> <m21x5si5fy.fsf@soma.local>
	<m2slwq36ni.fsf@tnuctip.rychter.com>
Message-ID: <m2r7camil7.fsf@soma.local>

Jan Rychter <jan at rychter.com> writes:

> Has anybody worked on this? I've just hit the same problem.

i will need it soon-ish (mid september maybe) so it'll happen around
then, but i can't dedicate any time to it now.

> On a gigabit or fast ethernet LAN uploading files 1GB in size really
> isn't an issue -- and users really want it to be simple (meaning: no
> FTP, just in-browser buttons).

too true.

-- 
-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 Aug 31 14:58:34 2005
From: frido at q-software-solutions.de (Friedrich Dominicus)
Date: Wed, 31 Aug 2005 16:58:34 +0200
Subject: [Bese-devel] small patch suggestion
Message-ID: <87psrurw4l.fsf@flarge.here>

After Drew has put me in the ground (to some extend he's right) here's
a small addition to form.lisp. I'm not sure if it's good idea, or if
it would be better to derive another component, but I feel if things
are available in standard HTML they should be easy accessible to
Common Lisp also

I guess it's clear what the patch is supposed to do. It should limit
the entry field length, I replaces the nil :initform to size because I
do not think that size could be anything else but a number. I tried
this patch here and it "seems" to be ok. So feel free to ignore or
apply this patch Marco. The numbers are arbitrary, I do not know if
there is a useful default for it.

BTW. I'm quite happy that UCW exists, but I feld and sometimes feel
lost in it's usage. 

Regards
Friedrich


--- form.lisp.2005-08-31	2005-08-20 13:45:59.000000000 +0200
+++ form.lisp	2005-08-31 16:47:24.000000000 +0200
@@ -52,12 +52,14 @@
 ;;;; Generic Text 
 
 (defclass text-field (form-element)
-  ((size :accessor size :initarg :size :initform nil))
+  ((size :accessor size :initarg :size :initform 0) 
+   (maxlength :accessor maxlength :initarg :maxlength :initform 20))
   (:metaclass standard-component-class))
 
 (defmethod render-on ((res response) (field text-field))
   (<ucw:input :type "text" :class "ucw-text-field" :accessor (client-value field)
-              :size (size field)))
+              :size (size field)
+              :maxlength (maxlength field)))
 
 (defvar *completing-text-field-prefix-counter* 0)