From metawilm at gmail.com  Sun Mar  2 13:08:04 2008
From: metawilm at gmail.com (Willem Broekema)
Date: Sun, 2 Mar 2008 14:08:04 +0100
Subject: [mcclim-devel] Method browser applicable methods improvement
Message-ID: <f6bc9b490803020508k1211e7c8kbf8946e8c5d5c8b0@mail.gmail.com>

In the method browser, the function
COMPUTE-APPLICABLE-METHODS-FROM-SPECIALIZERS determines the applicable
methods by first creating a list of prototype instances for the
specializers. The instances are then fed to
COMPUTE-APPLICABLE-METHODS.

The problem is that fetching prototype instances of built-in classes
is brittle (the function is not part of ANSI), e.g. in Allegro:

CL-USER(78): (mop::class-prototype (find-class 'integer))
Error: The slot EXCL::PROTOTYPE is missing from the object
#<BUILT-IN-CLASS INTEGER> of class #<STANDARD-CLASS BUILT-IN-CLASS>
       during operation SLOT-VALUE
  [condition type: PROGRAM-ERROR]

Restart actions (select using :continue):
 0: Try accessing the slot again
 1: Return a value
 2: Return to Top Level (an "abort" restart).
 3: Abort entirely from this (lisp) process.

See also this discussion:
http://groups.google.nl/group/comp.lang.lisp/browse_frm/thread/a15fd265fa3b540d/f68a63ee410a7f68?lnk=st#f68a63ee410a7f68

The attached patch makes the method browser robust against unavailable
prototype instances, and documents where things still to wrong.

Cheers,
- Willem
-------------- next part --------------
A non-text attachment was scrubbed...
Name: method-browser-prototype.patch
Type: text/x-patch
Size: 2187 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/mcclim-devel/attachments/20080302/620e7ff2/attachment.bin>

From metawilm at gmail.com  Sun Mar  2 15:17:19 2008
From: metawilm at gmail.com (Willem Broekema)
Date: Sun, 2 Mar 2008 16:17:19 +0100
Subject: [mcclim-devel] Drei Lisp syntax fix
Message-ID: <f6bc9b490803020717h3ab95471udadd3129058339c9@mail.gmail.com>

There is a small error in FORM-TO-OBJECT, which causes the method for
incomplete-quote-form to be shadowed by the NO-APPLICABLE-METHOD
default method. The attached patch fixes this.

Cheers,
- Willem
-------------- next part --------------
A non-text attachment was scrubbed...
Name: drei-lisp-syntax.patch
Type: text/x-patch
Size: 779 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/mcclim-devel/attachments/20080302/b4eadbae/attachment.bin>

From athas at sigkill.dk  Sun Mar  2 15:57:27 2008
From: athas at sigkill.dk (Troels Henriksen)
Date: Sun, 02 Mar 2008 16:57:27 +0100
Subject: [mcclim-devel] Drei Lisp syntax fix
In-Reply-To: <f6bc9b490803020717h3ab95471udadd3129058339c9@mail.gmail.com>
	(Willem Broekema's message of "Sun, 2 Mar 2008 16:17:19 +0100")
References: <f6bc9b490803020717h3ab95471udadd3129058339c9@mail.gmail.com>
Message-ID: <87y791i0ew.fsf@lambda.athas.dyndns.dk>

"Willem Broekema" <metawilm at gmail.com> writes:

> There is a small error in FORM-TO-OBJECT, which causes the method for
> incomplete-quote-form to be shadowed by the NO-APPLICABLE-METHOD
> default method. The attached patch fixes this.
>
> Cheers,
> - Willem

Thank you, committed to CVS.

How did you find this bug, by the way? I don't remember when exactly
incomplete-quote-forms are encountered.

-- 
\  Troels
/\ Henriksen


From metawilm at gmail.com  Sun Mar  2 16:22:06 2008
From: metawilm at gmail.com (Willem Broekema)
Date: Sun, 2 Mar 2008 17:22:06 +0100
Subject: [mcclim-devel] Drei Lisp syntax fix
In-Reply-To: <87y791i0ew.fsf@lambda.athas.dyndns.dk>
References: <f6bc9b490803020717h3ab95471udadd3129058339c9@mail.gmail.com>
	<87y791i0ew.fsf@lambda.athas.dyndns.dk>
Message-ID: <f6bc9b490803020822n7ee64ce8k63f12c01f98b1001@mail.gmail.com>

On Sun, Mar 2, 2008 at 4:57 PM, Troels Henriksen <athas at sigkill.dk> wrote:
>  How did you find this bug, by the way? I don't remember when exactly
>  incomplete-quote-forms are encountered.

I encountered it while entering this form:   (list #1='(

- Willem


From diresu at web.de  Tue Mar  4 06:30:59 2008
From: diresu at web.de (Dietrich Bollmann)
Date: Tue, 04 Mar 2008 15:30:59 +0900
Subject: [mcclim-devel] cpu constantly at 99% with just a simple GUI open
In-Reply-To: <87lk5b2n3m.fsf@lambda.athas.dyndns.dk>
References: <1203340163.21175.39.camel@pippi.pippi>
	<31ffd3c40802180824n900facy282eba82058b8ada@mail.gmail.com>
	<31ffd3c40802180824n5719d310q3d7f4a59d57123a5@mail.gmail.com>
	<1203442440.6318.76.camel@pippi.pippi>
	<1203444234.6318.93.camel@pippi.pippi> <20080219181558.GY16179@radon>
	<1203763407.11026.53.camel@pippi.pippi> <20080223114005.GB16179@radon>
	<1203780425.11026.74.camel@pippi.pippi>
	<87lk5b2n3m.fsf@lambda.athas.dyndns.dk>
Message-ID: <1204612259.13620.47.camel@pippi.pippi>

Hi Troels and David,

Originally I wanted to answer to your emails when everything works ...
but as there are new problems which make it impossible to use sbcl on my
computer I will switch back to cmucl for a while and postpone the
development of the gui until sbcl works again.  As this might take some
time I thought I rather answer to your last emails now.

Following your suggestions I installed sbcl, the lisp userland and
mcclim with clbuild - and beside some minor problems (inbetween them the
crashing of more or less half of the mcclim demodemo examples) things
seemed to work somehow.

As I only need very simple gui functionality, I decided to give it a try
and ported my code from cmucl to sbcl as you told me that this is the
lisp implementation which currently works best with mcclim.

Unfortunately since yesterday there is a new problem with an updated
version of libc6 on debian which renders sbcl unusable on my computer
(See the thread "[Sbcl-devel] SBCL hanging seemingly randomly",
http://sourceforge.net/mailarchive/forum.php?thread_name=1204420182.4466.42.camel%40hake&forum_name=sbcl-devel ) and I will switch back to cmucl and postpone the development of the gui until things work again.

Thanks a lot for your help and the efforts you put into mcclim!
Probably until some time later,

Dietrich

On Sun, 2008-02-24 at 01:51 +0100, Troels Henriksen wrote:
> Dietrich Bollmann <diresu at web.de> writes:
> 
> > As far as I remember it was the calculator.
> >
> > I also tried the drawing tool (first example button) and I think
> > a similar exception message was shown.
> 
> Everything that uses the text-editing substrate will fail. The
> calculator uses it for the display (which is OK), and the drawing
> tool, provided you mean CLIM-FIG, uses it as a status-bar, which is
> really not how it's supposed to be used, but whatever.
> 
> To fix it, you must install the CVS version Flexichain library. This
> is a temporary problem until the next release of Flexichain (after
> which the asdf-installable version will work just fine).
> 



From athas at sigkill.dk  Tue Mar  4 07:54:31 2008
From: athas at sigkill.dk (Troels Henriksen)
Date: Tue, 04 Mar 2008 08:54:31 +0100
Subject: [mcclim-devel] Method browser applicable methods improvement
In-Reply-To: <f6bc9b490803020508k1211e7c8kbf8946e8c5d5c8b0@mail.gmail.com>
	(Willem Broekema's message of "Sun, 2 Mar 2008 14:08:04 +0100")
References: <f6bc9b490803020508k1211e7c8kbf8946e8c5d5c8b0@mail.gmail.com>
Message-ID: <87hcfmc4aw.fsf@lambda.athas.dyndns.dk>

"Willem Broekema" <metawilm at gmail.com> writes:

> The attached patch makes the method browser robust against unavailable
> prototype instances, and documents where things still to wrong.

Thank you, patch applied.

-- 
\  Troels
/\ Henriksen


From metawilm at gmail.com  Sun Mar 23 00:14:56 2008
From: metawilm at gmail.com (Willem Broekema)
Date: Sun, 23 Mar 2008 01:14:56 +0100
Subject: [mcclim-devel] Drei issues
Message-ID: <f6bc9b490803221714y30edecavceec6b4a5d303ab9@mail.gmail.com>

Hi,
The following two issues in Drei came up in Allegro 8.1:

1. In Drei/syntax.lisp in update-lex there is a call to undefined
function low-mark.

2. In the listener, I try to show a picture from a file. Typing in the
accept box for the filename gives the following error (normally caught
in av-do-accept, but I disabled the catch-all handler):

Error: missing value for argument DREI::RELATIVE-CHUNK-END-OFFSET
  [condition type: SIMPLE-ERROR]

Restart actions (select using :continue):
 0: Return to application command loop
 1: Return to Top Level (an "abort" restart).
 2: Abort entirely from this (lisp) process.
[1] CLIM-USER(63): :zo
Evaluation stack:

   (ERROR SIMPLE-ERROR :FORMAT-CONTROL ...)
 ->(DREI::FETCH-CHUNK #<DREI:BUFFER-LINE @ #x117fa84a> 1)
   (DREI:BUFFER-VIEW-STROKE-PUMP #<DREI:TEXTUAL-DREI-SYNTAX-VIEW name:
*scratch* 1 @ #x1113c3b2>
                                 #S(DREI::DISPLAYED-STROKE
:START-OFFSET NIL ...) ...)
   ((METHOD DREI:STROKE-PUMP-WITH-SYNTAX
     (DREI:TEXTUAL-DREI-SYNTAX-VIEW
DREI-FUNDAMENTAL-SYNTAX:FUNDAMENTAL-SYNTAX T T))
      #<DREI:TEXTUAL-DREI-SYNTAX-VIEW name: *scratch* 1 @ #x1113c3b2>
      #<DREI-FUNDAMENTAL-SYNTAX:FUNDAMENTAL-SYNTAX @ #x1113c3c2> ...)
   [... DREI:STROKE-PUMP-WITH-SYNTAX ]
   ((METHOD DREI:STROKE-PUMP (DREI:DREI-SYNTAX-VIEW T T))
      #<DREI:TEXTUAL-DREI-SYNTAX-VIEW name: *scratch* 1 @ #x1113c3b2>
      #S(DREI::DISPLAYED-STROKE :START-OFFSET NIL ...) ...)
   ((METHOD DREI:STROKE-PUMP :AROUND (DREI:DREI-BUFFER-VIEW
DREI::DISPLAYED-STROKE T))
      #<DREI:TEXTUAL-DREI-SYNTAX-VIEW name: *scratch* 1 @ #x1113c3b2>
      #S(DREI::DISPLAYED-STROKE :START-OFFSET NIL ...) ...)
   ((:INTERNAL (:EFFECTIVE-METHOD 3 NIL NIL T T) 0)
      #<DREI:TEXTUAL-DREI-SYNTAX-VIEW name: *scratch* 1 @ #x1113c3b2>
      #S(DREI::DISPLAYED-STROKE :START-OFFSET NIL ...) ...)
   (DREI::PUT-STROKE #<DREI:TEXTUAL-DREI-SYNTAX-VIEW name: *scratch* 1
@ #x1113c3b2>
                     #S(DREI::DISPLAYED-LINE :START-OFFSET 0 ...) ...)
   (DREI::DRAW-LINE-STROKES
      #<CLIM-LISTENER::LISTENER-INTERACTOR-PANE
CLIM-LISTENER::INTERACTOR @ #x1109c912>
      #<DREI:TEXTUAL-DREI-SYNTAX-VIEW name: *scratch* 1 @ #x1113c3b2> ...)

... more older frames ...
[1] CLIM-USER(64):

That error is thrown in fetch-chunk, which does:
  (destructuring-bind (relative-chunk-end-offset . objectp)
      (aref (chunks line) chunk-index)
 and at that point, (chunks line) = #((2)) while chunk-index = 1.


Cheers,
- Willem


From athas at sigkill.dk  Sun Mar 23 08:58:34 2008
From: athas at sigkill.dk (Troels Henriksen)
Date: Sun, 23 Mar 2008 09:58:34 +0100
Subject: [mcclim-devel] Drei issues
In-Reply-To: <f6bc9b490803221714y30edecavceec6b4a5d303ab9@mail.gmail.com>
	(Willem Broekema's message of "Sun, 23 Mar 2008 01:14:56 +0100")
References: <f6bc9b490803221714y30edecavceec6b4a5d303ab9@mail.gmail.com>
Message-ID: <87hcexzuit.fsf@lambda.athas.dyndns.dk>

"Willem Broekema" <metawilm at gmail.com> writes:

> Hi,
> The following two issues in Drei came up in Allegro 8.1:
>
> 1. In Drei/syntax.lisp in update-lex there is a call to undefined
> function low-mark.

This is not a very high-priority problem, since the problematic code
in syntax.lisp is not used anywhere anymore, and should probably just
be removed. (It's part of an older parser generator that turned out to
be too slow).

> 2. In the listener, I try to show a picture from a file. Typing in the
> accept box for the filename gives the following error (normally caught
> in av-do-accept, but I disabled the catch-all handler):

Do you have the most recent CVS version? I fixed a bug that could
cause this error a week or so ago.

> That error is thrown in fetch-chunk, which does:
>   (destructuring-bind (relative-chunk-end-offset . objectp)
>       (aref (chunks line) chunk-index)
>  and at that point, (chunks line) = #((2)) while chunk-index = 1.

This used to happen when redisplay was performed at a time where the
view of the buffer had not been updated to account for buffer
changes. The dialog implementation used to be a major culprit at
triggering this. Could you please provide me with the exact steps
needed to reproduce this?

-- 
\  Troels
/\ Henriksen


From metawilm at gmail.com  Sun Mar 23 10:13:24 2008
From: metawilm at gmail.com (Willem Broekema)
Date: Sun, 23 Mar 2008 11:13:24 +0100
Subject: [mcclim-devel] Drei issues
In-Reply-To: <87hcexzuit.fsf@lambda.athas.dyndns.dk>
References: <f6bc9b490803221714y30edecavceec6b4a5d303ab9@mail.gmail.com>
	<87hcexzuit.fsf@lambda.athas.dyndns.dk>
Message-ID: <f6bc9b490803230313r41c6f8f1y2401e48b1325dd38@mail.gmail.com>

On Sun, Mar 23, 2008 at 9:58 AM, Troels Henriksen <athas at sigkill.dk> wrote:
>  > 2. In the listener, I try to show a picture from a file. Typing in the
>  > accept box for the filename gives the following error (normally caught
>  > in av-do-accept, but I disabled the catch-all handler):
>
>  Do you have the most recent CVS version? I fixed a bug that could
>  cause this error a week or so ago.

Indeed, after updating the display of an xpm image works. Apologies
for the false alarm.

- Willem


From metawilm at gmail.com  Sun Mar 23 10:26:12 2008
From: metawilm at gmail.com (Willem Broekema)
Date: Sun, 23 Mar 2008 11:26:12 +0100
Subject: [mcclim-devel] "Show file" gives blanc window
Message-ID: <f6bc9b490803230326g23022802qbee3f603cb252a1f@mail.gmail.com>

On Allegro 8.1 I have an issue with showing a text file contents, via
the "Show file" menu option. Sometimes the contents is shown, but
about 80% of the time the window popping up is blanc. If the text file
contains several lines then the lines quickly scroll by when the
window is opened, but the result is still a blanc window. Resizing the
window does not trigger drawing either.

Regards,
- Willem


From idurand at labri.fr  Fri Mar 28 15:16:31 2008
From: idurand at labri.fr (Irene DURAND)
Date: Fri, 28 Mar 2008 16:16:31 +0100
Subject: [mcclim-devel] pb with latest version
Message-ID: <47ED0BCF.5010102@labri.fr>

Hello,

I did a cvs update in the wrong window and got my working mcclim updated 
without wanting it.
I did remove all object files and recompile everything.
But none of my clim applications run. So I tried with the clim-demo 
calculator which fails
as well (although with a different error but the error is always related 
to DREI).

Does everybody have this problem or just me? Should I try to go back to 
an earlier
version?

Thanks in advance for your help.

Ir?ne
PS: I am on linux with SBCL 1.0.9

CLIM-DEMO> (calculator)
---------------------------------------------------------------------------------------------------------------------------------------------------------------
There is no applicable method for the generic function
  #<STANDARD-GENERIC-FUNCTION DREI-BUFFER:BUFFER (8)>
when called with arguments
  (#<CLIM-INTERNALS::DREI-TEXT-FIELD-SUBSTRATE "(Unnamed Pane)" {AD180A9}>).
   [Condition of type SIMPLE-ERROR]

Restarts:
  0: [ABORT-REQUEST] Abort handling SLIME request.
  1: [TERMINATE-THREAD] Terminate this thread (#<THREAD "repl-thread" 
{CA7BA99}>)

Backtrace:
  0: ((SB-PCL::FAST-METHOD NO-APPLICABLE-METHOD (T)) #<unavailable 
argument> #<unavailable argument> #<STANDARD-GENERIC-FUNCTION 
DREI-BUFFER:BUFFER (8)>)
  1: ((SB-PCL::FAST-METHOD GADGET-VALUE (DREI:DREI-GADGET-PANE)) 
#<unavailable argument> #<unavailable argument> 
#<CLIM-INTERNALS::DREI-TEXT-FIELD-SUBSTRATE "(Unnamed Pane)" {AD180A9}>)
  2: ((LABELS #:G125) #<CLIM-CLX::CLX-MEDIUM {AD46CF1}>)



From idurand at labri.fr  Fri Mar 28 15:41:17 2008
From: idurand at labri.fr (Irene DURAND)
Date: Fri, 28 Mar 2008 16:41:17 +0100
Subject: [mcclim-devel] Re: pb with latest version
In-Reply-To: <47ED0BCF.5010102@labri.fr>
References: <47ED0BCF.5010102@labri.fr>
Message-ID: <47ED119D.2060305@labri.fr>

After recompiling also subsystems and updating clx the calculator works.
But my SLS application which used to work fails with the following error:
The function DREI-BUFFER::DELETE-ELEMENTS* is undefined.
Again something related to DREI.
Has anybody an idea of the problem?
Ir?ne

The function DREI-BUFFER::DELETE-ELEMENTS* is undefined.
   [Condition of type UNDEFINED-FUNCTION]

Restarts:
  0: [ABORT] Return to application command loop
  1: [ABORT-REQUEST] Abort handling SLIME request.
  2: [TERMINATE-THREAD] Terminate this thread (#<THREAD "repl-thread" 
{ADDE181}>)

Backtrace:
  0: ("bogus stack frame")
  1: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL. 
SB-PCL::.ARG0. SB-PCL::.ARG1. SB-PCL::.ARG2.)) #<unavailable argument> 
#<unavailable argument> #<DREI::EXTENDED-STANDARD-BUFFER size:4 
{BE88399}> 0 4)
  2: ((LAMBDA (SB-PCL::.PV-CELL. SB-PCL::.NEXT-METHOD-CALL. 
SB-PCL::.ARG0. SB-PCL::.ARG1. SB-PCL::.ARG2.)) #<unavailable argument> 
#<unavailable argument> #<DREI:DREI-BUFFER size:4 {BE88359}> 0 4)

Irene DURAND a ?crit :
> Hello,
>
> I did a cvs update in the wrong window and got my working mcclim 
> updated without wanting it.
> I did remove all object files and recompile everything.
> But none of my clim applications run. So I tried with the clim-demo 
> calculator which fails
> as well (although with a different error but the error is always 
> related to DREI).
>
> Does everybody have this problem or just me? Should I try to go back 
> to an earlier
> version?
>
> Thanks in advance for your help.
>
> Ir?ne
> PS: I am on linux with SBCL 1.0.9
>



From athas at sigkill.dk  Fri Mar 28 18:31:00 2008
From: athas at sigkill.dk (Troels Henriksen)
Date: Fri, 28 Mar 2008 19:31:00 +0100
Subject: [mcclim-devel] Re: pb with latest version
In-Reply-To: <47ED119D.2060305@labri.fr> (Irene DURAND's message of "Fri, 28
	Mar 2008 16:41:17 +0100")
References: <47ED0BCF.5010102@labri.fr> <47ED119D.2060305@labri.fr>
Message-ID: <87wsnmwviz.fsf@lambda.athas.dyndns.dk>

Irene DURAND <idurand at labri.fr> writes:

> After recompiling also subsystems and updating clx the calculator works.
> But my SLS application which used to work fails with the following error:
> The function DREI-BUFFER::DELETE-ELEMENTS* is undefined.
> Again something related to DREI.
> Has anybody an idea of the problem?

CVS McCLIM requires Flexichain 1.5.1. This is the version that is
installed via ASDF-Install and also the version accessible via
Flexichain CVS.

-- 
\  Troels
/\ Henriksen


From rpgoldman at sift.info  Fri Mar 28 19:08:48 2008
From: rpgoldman at sift.info (Robert P. Goldman)
Date: Fri, 28 Mar 2008 14:08:48 -0500
Subject: [mcclim-devel] Re: pb with latest version
In-Reply-To: <87wsnmwviz.fsf@lambda.athas.dyndns.dk>
References: <47ED0BCF.5010102@labri.fr> <47ED119D.2060305@labri.fr>
	<87wsnmwviz.fsf@lambda.athas.dyndns.dk>
Message-ID: <47ED4240.3070408@sift.info>

Troels Henriksen wrote:
> Irene DURAND <idurand at labri.fr> writes:
>
>> After recompiling also subsystems and updating clx the calculator works.
>> But my SLS application which used to work fails with the following error:
>> The function DREI-BUFFER::DELETE-ELEMENTS* is undefined.
>> Again something related to DREI.
>> Has anybody an idea of the problem?
>
> CVS McCLIM requires Flexichain 1.5.1. This is the version that is
> installed via ASDF-Install and also the version accessible via
> Flexichain CVS.
>

There is a bug in mcclim.asd, then:

[rpg at cryptonomicon mcclim]$ fgrep -i flexichain *.asd
mcclim.asd:    :depends-on (:clim-lisp :spatial-trees (:version
"flexichain" "1.4.1"))
mcclim.asd:  :depends-on (:flexichain :esa-mcclim :clim-core
#+#.(mcclim.system::dep-on-swank) :swank)


(:version "flexichain" "1.5.1") should be substituted in both places
where flexichain appears as a :depends-on.

I could do this, but my build environment is not in a good state, so it
might be better if someone else, who could test this, would make the patch.

If we make this patch, it will make it obvious to the hapless McCLIM
loader when he or she has a bad dependency.

Probably we should require a specific version of :spatial-trees, as well.


From rpgoldman at sift.info  Fri Mar 28 19:52:46 2008
From: rpgoldman at sift.info (Robert P. Goldman)
Date: Fri, 28 Mar 2008 14:52:46 -0500
Subject: [mcclim-devel] Re: pb with latest version
In-Reply-To: <47ED4240.3070408@sift.info>
References: <47ED0BCF.5010102@labri.fr> <47ED119D.2060305@labri.fr>
	<87wsnmwviz.fsf@lambda.athas.dyndns.dk>
	<47ED4240.3070408@sift.info>
Message-ID: <47ED4C8E.4030608@sift.info>

Robert P. Goldman wrote:
> Troels Henriksen wrote:
>> Irene DURAND <idurand at labri.fr> writes:
>>
>>> After recompiling also subsystems and updating clx the calculator works.
>>> But my SLS application which used to work fails with the following error:
>>> The function DREI-BUFFER::DELETE-ELEMENTS* is undefined.
>>> Again something related to DREI.
>>> Has anybody an idea of the problem?
>> CVS McCLIM requires Flexichain 1.5.1. This is the version that is
>> installed via ASDF-Install and also the version accessible via
>> Flexichain CVS.
>>
> 
> There is a bug in mcclim.asd, then:
> 
> [rpg at cryptonomicon mcclim]$ fgrep -i flexichain *.asd
> mcclim.asd:    :depends-on (:clim-lisp :spatial-trees (:version
> "flexichain" "1.4.1"))
> mcclim.asd:  :depends-on (:flexichain :esa-mcclim :clim-core
> #+#.(mcclim.system::dep-on-swank) :swank)
> 
> 
> (:version "flexichain" "1.5.1") should be substituted in both places
> where flexichain appears as a :depends-on.
> 
> I could do this, but my build environment is not in a good state, so it
> might be better if someone else, who could test this, would make the patch.
> 
> If we make this patch, it will make it obvious to the hapless McCLIM
> loader when he or she has a bad dependency.
> 
> Probably we should require a specific version of :spatial-trees, as well.
> 

OK, I made the patch, and it passed my rigorous testing procedure, viz.
 mcclim loading with asdf initially crashed because I had the old
flexichain, and then succeeded after I did a cvs up on flexichain (well,
except for my CLX is messed up...).

Will commit soon...



From idurand at labri.fr  Fri Mar 28 20:03:14 2008
From: idurand at labri.fr (idurand at labri.fr)
Date: Fri, 28 Mar 2008 21:03:14 +0100
Subject: [mcclim-devel] Re: pb with latest version
Message-ID: <20080328210314.bwrtbe8ajkk40gws@webmail.labri.fr>

Thanks for all you answers. I will try with Flexichain 1.5.1 on monday
and keep you informed.
Ir?ne

Troels Henriksen <athas at sigkill.dk> a ?crit :

> Irene DURAND <idurand at labri.fr> writes:
>
>> After recompiling also subsystems and updating clx the calculator works.
>> But my SLS application which used to work fails with the following error:
>> The function DREI-BUFFER::DELETE-ELEMENTS* is undefined.
>> Again something related to DREI.
>> Has anybody an idea of the problem?
>
> CVS McCLIM requires Flexichain 1.5.1. This is the version that is
> installed via ASDF-Install and also the version accessible via
> Flexichain CVS.
>
> --
> \  Troels
> /\ Henriksen
>




From idurand at labri.fr  Mon Mar 31 07:12:37 2008
From: idurand at labri.fr (Irene DURAND)
Date: Mon, 31 Mar 2008 09:12:37 +0200
Subject: [mcclim-devel] Re: mcclim-devel Digest, Vol 39, Issue 4
In-Reply-To: <20080328170245.D8CDA4406D@common-lisp.net>
References: <20080328170245.D8CDA4406D@common-lisp.net>
Message-ID: <47F08EE5.5060704@labri.fr>

Hello,

All my problems are gone with the latest version of flexichain 1.5.1.
Thanks for you help.
Ir?ne

>
> Today's Topics:
>
>    1. pb with latest version (Irene DURAND)
>    2. Re: pb with latest version (Irene DURAND)
>
>
> ----------------------------------------------------------------------
>