CLOCC - Common Lisp Open Code Collection

> * In message <eITNPGHTLjf3MkDgOsgG=yBtNug=@4ax.com>
> * On the subject of "Changing PostScript printer options in gnuplot.lisp (cllib)"
> * Sent on Mon, 29 Apr 2002 19:39:32 +0200
> * Honorable Paolo Amoroso <amoroso@...> writes:
>
> I had to generate color printouts from gnuplot controlled via cllib's
> gnuplot.lisp interface. So I set the CLLIB::TERMINAL-OPTIONS slot of
> CLLIB:+PLOT-TERM-PRINTER+, a CLLIB::PLOT-TERM structure, to an
> appropriate value (i.e. I added "color" to the option string).

(defmethod directive-term (:print)
  (make-plot-term ... ))

make-plot-term is now exported.

-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
<http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
MS Windows vs IBM OS/2: Why marketing matters more than technology...

> * In message <xunlmbou71h.fsf@...>
> * On the subject of "CLOCC and LispWorks"
> * Sent on Mon, 15 Apr 2002 20:54:18 +0200
> * Honorable Martin Thornquist <martint@...> writes:
>
> Error: Defining method #<STANDARD-METHOD PRINT-OBJECT NIL (HASH-TABLE STREAM) 20659E1C> visible from package COMMON-LISP.
> 
> LispWorks has a variable lw:*handle-warn-on-redefinition* to decide
> how to handle such redefinitions, it can be set to :warn to get a
> warning or nil to just do the redefinition silently.

the above is an error, not a warning.

> I'll gladly make/test a patch, but I'm not quite sure what is the best
> way to solve this -- and I don't know CLOCC/CLLIB that well.

--- closio.lisp.~1.16.~	Fri Nov  2 17:31:15 2001
+++ closio.lisp	Tue Apr 16 10:41:20 2002
@@ -83,6 +83,10 @@
 #+allegro
 (eval-when (compile)
   (setf (excl:package-definition-lock (find-package :common-lisp)) nil))
+#+clisp
+(eval-when (compile) (setf (ext:package-lock "CL") nil))
+#+lispworks
+(eval-when (compile) (setf lw:*handle-warn-on-redefinition* nil))
 
 (defmethod print-object ((obj standard-object) (out stream))
   (case *closio-method*
@@ -106,6 +110,10 @@
 #+allegro
 (eval-when (compile)
   (setf (excl:package-definition-lock (find-package :common-lisp)) t))
+#+clisp
+(eval-when (compile) (setf (ext:package-lock "CL") t))
+#+lispworks
+(eval-when (compile) (setf lw:*handle-warn-on-redefinition* :error))
 
 ;;;
 ;;; }}}{{{ macroexpand-r


-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
nobody's life, liberty or property are safe while the legislature is in session

[ Sam Steingold ]

> the above is an error, not a warning.

Yup, as I wrote. However, I see I neglected to mention that
lw:*handle-warn-on-redefinition* can also be :error, sorry about that.
I see you found out by yourself, though.

> <snip patch>

Yeah, that's of course a good way to do it (I didn't see the #+allegro
bit below the method yesterday). Two points, though: this also needs
to be set on loading, and I think it's prudent to keep the original
value of lw:*handle-warn-on-redefinition* in case the user has set it
(this would apply to the allegro and clisp bits too, I guess, but I
don't know those systems). Additionally, the same problem occurs in
miscprint.lisp. The following two patches address these concerns.

--- miscprint.lisp.~1.11.~      Sat Nov  3 12:00:14 2001
+++ miscprint.lisp      Wed Apr 17 10:09:44 2002
@@ -130,12 +130,22 @@
      (alist->hash-table (read st t nil t)))
    rt))
 
+#+lispworks
+(eval-when (:compile-toplevel :load-toplevel)
+  (progn (defvar *lw-saved-handle-warn-on-redefinition*
+          lw:*handle-warn-on-redefinition*)
+        (setq lw:*handle-warn-on-redefinition* nil)))
+
 ;;; beware that some lisps (e.g., CLISP and CMUCL) will not use this
 ;;; method for hash-tables.  it does work with Allegro though.
 (defmethod print-object ((ht hash-table) (out stream))
   (if *print-readably*
       (format out "#h~s" (hash-table->alist ht))
       (call-next-method)))
+
+#+lispworks
+(eval-when (:compile-toplevel :load-toplevel)
+  (setq lw:*handle-warn-on-redefinition* *lw-saved-handle-warn-on-redefinition*))
 
 ;;;
 ;;; property lists


(The following is against the original, without your patch.)

--- closio.lisp.bak     Wed Apr 17 09:56:37 2002
+++ closio.lisp Wed Apr 17 10:11:15 2002
@@ -83,6 +83,13 @@
 #+allegro
 (eval-when (compile)
   (setf (excl:package-definition-lock (find-package :common-lisp)) nil))
+#+clisp
+(eval-when (compile) (setf (ext:package-lock "CL") nil))
+#+lispworks
+(eval-when (:compile-toplevel :load-toplevel)
+  (progn (defvar *lw-saved-handle-warn-on-redefinition*
+          lw:*handle-warn-on-redefinition*)
+        (setq lw:*handle-warn-on-redefinition* nil)))
 
 (defmethod print-object ((obj standard-object) (out stream))
   (case *closio-method*
@@ -106,6 +113,11 @@
 #+allegro
 (eval-when (compile)
   (setf (excl:package-definition-lock (find-package :common-lisp)) t))
+#+clisp
+(eval-when (compile) (setf (ext:package-lock "CL") t))
+#+lispworks
+(eval-when (:compile-toplevel :load-toplevel)
+  (setq lw:*handle-warn-on-redefinition* *lw-saved-handle-warn-on-redefinition*))
 
 ;;;
 ;;; }}}{{{ macroexpand-r



Martin
-- 
"An ideal world is left as an exercise to the reader."
                                                 -Paul Graham, On Lisp

> * In message <xun4riaiujn.fsf@...>
> * On the subject of "Re: CLOCC and LispWorks"
> * Sent on Wed, 17 Apr 2002 10:44:12 +0200
> * Honorable Martin Thornquist <martint@...> writes:
>
> [ Sam Steingold ]
> 
> > the above is an error, not a warning.
> 
> Yup, as I wrote. However, I see I neglected to mention that
> lw:*handle-warn-on-redefinition* can also be :error, sorry about that.
> I see you found out by yourself, though.

it was just a guess :-)

> Two points, though: this also needs to be set on loading,

are you sure?  neither CLISP not ACL need it at loading.
please do check.


-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
Small languages require big programs, large languages enable small programs.

[ Sam Steingold ]

>> Two points, though: this also needs to be set on loading,
>
> are you sure?  neither CLISP not ACL need it at loading.
> please do check.

It really does. With just :compile-toplevel compile works fine (of
course), but I get an error afterwards when the system tries to load
the newly compiled file.


Martin
-- 
"An ideal world is left as an exercise to the reader."
                                                 -Paul Graham, On Lisp

> * In message <m3elhlolyo.fsf@...>
> * On the subject of "Re: Basic usage of xml.lisp"
> * Sent on 11 Apr 2002 19:27:11 -0400
> * Honorable Jordan Katz <katz@...> writes:
>
> ;; Loading file /home/katz/.clisprc ...
> ;;  Loading file /usr/local/src/clocc/clocc.fas ...
> *** - This file was produced by another lisp version, must be recompiled.

clocc.lisp is two directory levels above cllib - and it has to be
recompiled too.

> I tried recompiling cllib again with CLISP (when I initially compiled
> I had no errors) but I get the following errors:
> 
> katz@...> make system
> /usr/local/src/clocc/bin/run-lisp -I /usr/local/src/clocc/clocc-top \
>         -x '(funcall (intern "COMPILE-SYSTEM" :mk) "cllib")'
> /usr/local/lib/clisp/base/lisp.run: operating system error during load of initialization file `/usr/local/src/clocc/clocc-top.mem'
>  errno = ENOENT: No such file or directory.
> make: *** [system] Error 1
> 
> How can I fix this?

by reading INSTALL file carefully.
I guess it is poorly written - maybe you can suggest some improvements?

Here, let me help you.
Start with a clean CLOCC, e.g., `rm -rf && cvs up`.
At the very least, remove all the compiled files, like this:

$ rm -f `find . -name "*.fas"`

Open INSTALL in Emacs and read the first part
(building CLOCC with GNU Make):

1. environment:

        Lines 5&6 ==>
        $ export LISPTYPE=clisp

        Line 7-11 ==>
        $ export CLOCC_DUMP=clisp

2. logical hosts: clocc.lisp, luckily, has your setup by default.

3. type "make clocc-top" -- this will compile clocc.lisp and
   defsystem.lisp and create clocc-top.mem (which you were missing)
   
4. type 'cd src/cllib && make system'.

Now, could you please rewrite INSTALL so that it will be more clear?

-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
cogito cogito ergo cogito sum

Sam Steingold <sds@...> writes:

> > * In message <m3elhlolyo.fsf@...>
> > * On the subject of "Re: Basic usage of xml.lisp"
> > * Sent on 11 Apr 2002 19:27:11 -0400
> > * Honorable Jordan Katz <katz@...> writes:
> >
> > ;; Loading file /home/katz/.clisprc ...
> > ;;  Loading file /usr/local/src/clocc/clocc.fas ...
> > *** - This file was produced by another lisp version, must be recompiled.
> 
> clocc.lisp is two directory levels above cllib - and it has to be
> recompiled too.
> 
> > I tried recompiling cllib again with CLISP (when I initially compiled
> > I had no errors) but I get the following errors:
> > 
> > katz@...> make system
> > /usr/local/src/clocc/bin/run-lisp -I /usr/local/src/clocc/clocc-top \
> >         -x '(funcall (intern "COMPILE-SYSTEM" :mk) "cllib")'
> > /usr/local/lib/clisp/base/lisp.run: operating system error during
> > load of initialization file `/usr/local/src/clocc/clocc-top.mem'
> >  errno = ENOENT: No such file or directory.
> > make: *** [system] Error 1
> > 
> > How can I fix this?
> 
> by reading INSTALL file carefully.
> I guess it is poorly written - maybe you can suggest some improvements?
> 
> Here, let me help you.
> Start with a clean CLOCC, e.g., `rm -rf && cvs up`.
> At the very least, remove all the compiled files, like this:
> 
> $ rm -f `find . -name "*.fas"`

I got errors when trying to compile clocc-top after using only rm -f
`find . -name "*.fas"`, but after reinstalling all of CLOCC as you
suggested before make clocc-top worked.  

> 
> Open INSTALL in Emacs and read the first part
> (building CLOCC with GNU Make):
> 
> 1. environment:
> 
>         Lines 5&6 ==>
>         $ export LISPTYPE=clisp
> 
>         Line 7-11 ==>
>         $ export CLOCC_DUMP=clisp

Those variables were set already.  I put them in my ~/.bashrc when I
first installed clocc. 

> 2. logical hosts: clocc.lisp, luckily, has your setup by default.
> 
> 3. type "make clocc-top" -- this will compile clocc.lisp and
>    defsystem.lisp and create clocc-top.mem (which you were missing)

This step went smoothly.

> 4. type 'cd src/cllib && make system'.

I got the following error when I tried this.

Compilation of file
/usr/local/src/clocc/src/tools/metering/metering.lisp is finished.
;; Loading file /usr/local/src/clocc/src/tools/metering/metering.fas ...
*** - symbol NV2 has no value
The following special variables were not defined:
 MONITOR::NV2 MONITOR::OV2
0 errors, 34 warnings

0 errors, 0 warnings
make: *** [system] Error 1

> Now, could you please rewrite INSTALL so that it will be more clear?

All of these steps were crystal clear and I followed all of them when
I had my older version of CLISP.  I guess something must have gone
haywire when I updated CLISP.  However, there's one part of the
INSTALL file that was unclear to me as a Lisp newbie that you didn't
refer to, and that's the last part titled "Lisp only":

;; * compile base files
(in-package :cl-user)
(setq *clocc-root* "c:/gnu/clocc/") ; or whatever ...
(compile-file (concatenate 'string *clocc-root* "clocc"))
(load *)
(compile-file (translate-logical-pathname "clocc:src;defsystem;defsystem"))
(load *)

;; * compile some systems
(setq mk::*central-registry*
      (nconc
       (list
        ;; select the systems you want to use
        (translate-logical-pathname "clocc:src;cllib;")
        (translate-logical-pathname "clocc:src;ext;queues;")
        (translate-logical-pathname "clocc:src;ext;union-find;")
        (translate-logical-pathname "clocc:src;f2cl;packages;")
        (translate-logical-pathname "clocc:src;f2cl;")
        (translate-logical-pathname "clocc:src;port;")
        (translate-logical-pathname "clocc:src;port;configuration;")
        (translate-logical-pathname "clocc:src;port;environment;")
        (translate-logical-pathname "clocc:src;tools;metering;"))
       mk::*central-registry*))
(mk:oos "cllib" :compile)
(mk:oos "f2cl" :compile)
...

;; * add to your init file (~/.clisprc for CLISP, ~/.cmucl-init.lisp for
;;   CMUCL or ~/.clinit.cl for Allegro)
[stuff snipped]

It's clear that the stuff with the comment "add to your init file"
goes in the init file, but I don't understand what the two blocks of
code before that do.  AFAIK, the files refered to in those two blocks
were already compiled with the make's, so I don't know how that part
fits and when I first read the INSTALL file that confused me.
Everything that precedes that part is perfectly clear, though. 

Thanks, I appreciate your patience, 
-- 
Jordan Katz <katz@...>  |  Mind the gap

> * In message <m3zo087cym.fsf@...>
> * On the subject of "Re: Basic usage of xml.lisp"
> * Sent on 12 Apr 2002 18:45:53 -0400
> * Honorable Jordan Katz <katz@...> writes:
>
> ;; Loading file /usr/local/src/clocc/src/tools/metering/metering.fas ...
> *** - symbol NV2 has no value
> The following special variables were not defined:
>  MONITOR::NV2 MONITOR::OV2
> 0 errors, 34 warnings
> 
> 0 errors, 0 warnings
> make: *** [system] Error 1

cvs up again please.


-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
Live free or die.

Sam Steingold <sds@...> writes:

> > * In message <m3zo087cym.fsf@...>
> > * On the subject of "Re: Basic usage of xml.lisp"
> > * Sent on 12 Apr 2002 18:45:53 -0400
> > * Honorable Jordan Katz <katz@...> writes:
> >
> > ;; Loading file /usr/local/src/clocc/src/tools/metering/metering.fas ...
> > *** - symbol NV2 has no value
> > The following special variables were not defined:
> >  MONITOR::NV2 MONITOR::OV2
> > 0 errors, 34 warnings
> > 
> > 0 errors, 0 warnings
> > make: *** [system] Error 1
> 
> cvs up again please.

I did and now 'make system' in src/cllib compiled with 0 errors and 0
warnings.  In my ~/.clisprc I have:

(setq *clocc-root* "/usr/local/src/clocc/") 
(load (concatenate 'string *clocc-root* "clocc"))
(load (translate-logical-pathname "clocc:src;cllib;base"))

I still can't run files that make usage of clocc, though.  I tried
Pearson's slashdot.lisp and got:

*** - A file with name clocc:src;cllib;date does not exist

When I tried gq.lisp from cllib, it returned:

*** - TRANSLATE-LOGICAL-PATHNAME: unknown logical host "CLOCC" in 
#S(LOGICAL-PATHNAME :HOST "CLOCC" :DEVICE NIL
   :DIRECTORY (:ABSOLUTE "SRC" "CLLIB") :NAME "BASE" :TYPE NIL :VERSION NIL)

Sorry to keep bugging but it just won't work for me.  Thanks for the help,
-- 
Jordan Katz <katz@...>  |  Mind the gap

> * In message <m38z7r64mf.fsf@...>
> * On the subject of "Re: Basic usage of xml.lisp"
> * Sent on 13 Apr 2002 10:43:36 -0400
> * Honorable Jordan Katz <katz@...> writes:
>
> (setq *clocc-root* "/usr/local/src/clocc/") 
> (load (concatenate 'string *clocc-root* "clocc"))
> (load (translate-logical-pathname "clocc:src;cllib;base"))

load autoloads too:
(load (translate-logical-pathname "cllib:auto"))

> I still can't run files that make usage of clocc, though.  I tried
> Pearson's slashdot.lisp and got:
> 
> *** - A file with name clocc:src;cllib;date does not exist

please tell him to replace
(load "clocc:src;cllib;date")
with
(load (translate-logical-pathname "clocc:src;cllib;date"))

(alternatively, setq custom:*parse-namestring-ansi*  to t in ~/.clisprc
or run clisp with -ansi option)

> When I tried gq.lisp from cllib, it returned:
> 
> *** - TRANSLATE-LOGICAL-PATHNAME: unknown logical host "CLOCC" in 
> #S(LOGICAL-PATHNAME :HOST "CLOCC" :DEVICE NIL
>    :DIRECTORY (:ABSOLUTE "SRC" "CLLIB") :NAME "BASE" :TYPE NIL :VERSION NIL)

what exactly did you do?

-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
Who is General Failure and why is he reading my hard disk?

* Sam Steingold <sds@...> [2002-04-13 15:08:24 -0400]:

> > I still can't run files that make usage of clocc, though.  I tried
> > Pearson's slashdot.lisp and got:
> > 
> > *** - A file with name clocc:src;cllib;date does not exist
> 
> please tell him to replace
> (load "clocc:src;cllib;date")

What makes you think I do that? I don't. What I do do is:

,----[ Excerpt from <URL:http://www.davep.org/lisp/slashdot.lisp> ]
| ;; Uses clocc <URL:http://clocc.sourceforge.net/>.
| (eval-when (compile load eval)
|   (require :cllib-date #p"clocc:src;cllib;date")
|   (require :cllib-url  #p"clocc:src;cllib;url")
|   (require :cllib-html #p"clocc:src;cllib;html")
|   (require :cllib-xml  #p"clocc:src;cllib;xml"))
`----

which works perfectly.

> with
> (load (translate-logical-pathname "clocc:src;cllib;date"))

How does `translate-logical-pathname' differ from `#p'? How does it differ
in the case of this poster's problem?

-- 
Dave Pearson
http://www.davep.org/

I am replying through the list.
please do not contact me directly about CLOCC, CLLIB or CLISP.

> * In message <m3bscn2de9.fsf@...>
> * On the subject of "Re: Basic usage of xml.lisp"
> * Sent on 13 Apr 2002 22:59:58 -0400
> * Honorable Jordan Katz <katz@...> writes:
>
> > > When I tried gq.lisp from cllib, it returned:
> > > 
> > > *** - TRANSLATE-LOGICAL-PATHNAME: unknown logical host "CLOCC" in 
> > > #S(LOGICAL-PATHNAME :HOST "CLOCC" :DEVICE NIL
> > >    :DIRECTORY (:ABSOLUTE "SRC" "CLLIB") :NAME "BASE" :TYPE NIL :VERSION NIL)
> 
> I just ran the file under clisp..
> 
> $ clisp /usr/local/src/clocc/src/cllib/gq.lisp
> 
> *** - TRANSLATE-LOGICAL-PATHNAME: unknown logical host "CLOCC" in 
> #S(LOGICAL-PATHNAME :HOST "CLOCC" :DEVICE NIL
>    :DIRECTORY (:ABSOLUTE "SRC" "CLLIB") :NAME "BASE" :TYPE NIL
> :VERSION NIL)

try `clisp -i /usr/local/src/clocc/src/cllib/gq.lisp`
(without -i, you ignore ~/.clisprc and you do not get the prompt!)

-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
Are you smart enough to use Lisp?

> * In message <20020415111832.R4159@...>
> * On the subject of "Re: Basic usage of xml.lisp"
> * Sent on Mon, 15 Apr 2002 11:18:32 +0100
> * Honorable Dave Pearson <davep@...> writes:
>
> * Sam Steingold <sds@...> [2002-04-13 15:08:24 -0400]:
> 
> > > I still can't run files that make usage of clocc, though.  I tried
> > > Pearson's slashdot.lisp and got:
> > > 
> > > *** - A file with name clocc:src;cllib;date does not exist
> > 
> > please tell him to replace
> > (load "clocc:src;cllib;date")
> 
> What makes you think I do that? I don't. What I do do is:
> 
> ,----[ Excerpt from <URL:http://www.davep.org/lisp/slashdot.lisp> ]
> | ;; Uses clocc <URL:http://clocc.sourceforge.net/>.
> | (eval-when (compile load eval)
> |   (require :cllib-date #p"clocc:src;cllib;date")
> |   (require :cllib-url  #p"clocc:src;cllib;url")
> |   (require :cllib-html #p"clocc:src;cllib;html")
> |   (require :cllib-xml  #p"clocc:src;cllib;xml"))
> `----
> 
> which works perfectly.
> 
> > with
> > (load (translate-logical-pathname "clocc:src;cllib;date"))
> 
> How does `translate-logical-pathname' differ from `#p'? How does it differ
> in the case of this poster's problem?

when CLISP has custom:*parse-namestring-ansi* nil, #p"" does not
recognise logical pathnames.

-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
Lisp is a language for doing what you've been told is impossible. - Kent Pitman

* Sam Steingold <sds@...> [2002-04-16 10:35:50 -0400]:

> > How does `translate-logical-pathname' differ from `#p'? How does it
> > differ in the case of this poster's problem?
> 
> when CLISP has custom:*parse-namestring-ansi* nil, #p"" does not recognise
> logical pathnames.

Would I be right in thinking that, in such a circumstance, my code wouldn't
be running in an ANSI CL?

-- 
Dave Pearson
http://www.davep.org/

> * In message <20020416161715.S4159@...>
> * On the subject of "Re: Basic usage of xml.lisp"
> * Sent on Tue, 16 Apr 2002 16:17:15 +0100
> * Honorable Dave Pearson <davep@...> writes:
>
> * Sam Steingold <sds@...> [2002-04-16 10:35:50 -0400]:
> 
> > > How does `translate-logical-pathname' differ from `#p'? How does it
> > > differ in the case of this poster's problem?
> > 
> > when CLISP has custom:*parse-namestring-ansi* nil, #p"" does not recognise
> > logical pathnames.
> 
> Would I be right in thinking that, in such a circumstance, my code
> wouldn't be running in an ANSI CL?

lisps have their quirks.
adding a TRANSLATE-LOGICAL-PATHNAME call is a small price to pay for
both of us to avoid having to deal with these kinds of questions.

e.g., I had to modify all calls to FIND-PACKAGE in CLOCC/PORT to
accommodate the Allegro's idea of case-sensitivity.

life is tough.

-- 
Sam Steingold (http://www.podval.org/~sds) running RedHat7.2 GNU/Linux
Read, think and remember! <http://www.iris.org.il> <http://www.memri.org/>
<http://www.palestine-central.com/> <http://www.mideasttruth.com/>
Old Age Comes at a Bad Time.

* Sam Steingold <sds@...> [2002-04-16 12:28:01 -0400]:

> > Would I be right in thinking that, in such a circumstance, my code
> > wouldn't be running in an ANSI CL?
> 
> lisps have their quirks.
> adding a TRANSLATE-LOGICAL-PATHNAME call is a small price to pay for
> both of us to avoid having to deal with these kinds of questions.
> 
> e.g., I had to modify all calls to FIND-PACKAGE in CLOCC/PORT to
> accommodate the Allegro's idea of case-sensitivity.
> 
> life is tough.

So that's a "yes" then?

-- 
Dave Pearson
http://www.davep.org/