From haible Mon Mar 8 23:08:07 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA26692; Mon, 8 Mar 93 23:08:07 +0100 Date: Mon, 8 Mar 93 23:08:07 +0100 From: haible (Bruno Haible) Message-Id: <9303082208.AA26692@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: The official CLISP announcement This is CLISP, a Common Lisp implementation. CLISP is mostly CLtL1 compliant. For those who want to use CLOS, I also provide a port of PCL. No other features of CLtL2 or dpANS CL are currently supported. The newest versions will always be available via anonymous ftp from ma2s2.mathematik.uni-karlsruhe.de [129.13.115.2], directory /pub/lisp/clisp/. ------------------------------------------------------------------------------- Common Lisp CLISP Common Lisp is * a convential programming language and an AI language * interactive * a Lisp for professional use Common Lisp programs are * easy to test (interactive) * easy to maintain (depending on programming style) * portable (there is a standard for the language and the library functions) Our Common Lisp CLISP * needs only 1.5 MB of memory * implements 99% of the standard * can call your preferred editor * is freely distributable Common Lisp provides * clear syntax, carefully designed semantics * several data types: numbers, strings, arrays, lists, characters, symbols, structures, streams etc. * runtime typing: the programmer needn't bother about type declarations, but he gets notified on type violations. * many generic functions: 88 arithmetic functions for all kinds of numbers (integers, ratios, floating point numbers, complex numbers), 44 search/filter/sort functions for lists, arrays and strings * automatic memory management (garbage collection) * packaging of programs into modules * macros: every programmer can make his own language extensions Our Common Lisp CLISP provides * an interpreter * a compiler which makes execution of programs 5 times faster * all data types with unlimited size (the size need never be declared, the size of lists and arrays may be changed dynamically) * integers of arbitrary length, unlimited floating point number precision * 594 library functions, 542 of them written in C Get it via anonymous ftp from ma2s2.mathematik.uni-karlsruhe.de [129.13.115.2], directory /pub/lisp/clisp/, or contact Bruno Haible . ------------------------------------------------------------------------------- From @NYX.UNI-KONSTANZ.DE:hoehle@post.inf-wiss.ivp.uni-konstanz.de Tue Mar 9 15:47:17 1993 Return-Path: <@NYX.UNI-KONSTANZ.DE:hoehle@post.inf-wiss.ivp.uni-konstanz.de> Received: from NYX.UNI-KONSTANZ.DE by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA27417; Tue, 9 Mar 93 15:47:17 +0100 Received: from inf-wiss.ivp.uni-konstanz.de by NYX.UNI-KONSTANZ.DE (IBM VM SMTP V2R2) with TCP; Tue, 09 Mar 93 15:47:06 MEZ Received: from stetten.inf-wiss.ivp.uni-konstanz.de by inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA01183; Tue, 9 Mar 93 15:46:31 +0100 Date: Tue, 9 Mar 93 15:46:31 +0100 From: hoehle@post.inf-wiss.ivp.uni-konstanz.de (Joerg-Cyril Hoehle) Message-Id: <9303091446.AA01183@inf-wiss.ivp.uni-konstanz.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Noch ein Test YATS! (yet another test suite) Joerg. hoehle@inf-wiss.ivp.uni-konstanz.de From mycroft@gnu.ai.mit.edu Wed Mar 10 06:18:56 1993 Return-Path: Received: from albert.gnu.ai.mit.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA28978; Wed, 10 Mar 93 06:18:56 +0100 Received: from hal.gnu.ai.mit.edu by albert.gnu.ai.mit.edu (5.65/4.0) with SMTP id ; Wed, 10 Mar 93 00:19:12 -0500 Received: by hal.gnu.ai.mit.edu (AIX 3.2/UCB 5.64/4.0) id ; Wed, 10 Mar 1993 00:18:41 -0500 From: mycroft@gnu.ai.mit.edu Message-Id: <9303100518.AA33611@hal.gnu.ai.mit.edu> Subject: 386BSD port To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Wed, 10 Mar 1993 00:18:40 -0500 (EST) X-Mailer: ELM [version 2.4 PL8] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1138 After running configure, there are several things which need to be changed in unixconf.h: *** unixconf.h.~1~ Tue Mar 9 17:18:16 1993 --- unixconf.h Tue Mar 9 18:21:51 1993 *************** *** 50,50 **** ! #undef STDC_HEADERS --- 50,50 ---- ! #define STDC_HEADERS *************** *** 149,149 **** ! #define RETFREETYPE int --- 149,149 ---- ! #define RETFREETYPE void *************** *** 179,179 **** ! #define RETABORTTYPE int --- 179,179 ---- ! #define RETABORTTYPE void *************** *** 187,187 **** ! #define GETENV_CONST --- 187,187 ---- ! #define GETENV_CONST const *All* of this junk is due to some bogosity in AC_STDC_HEADERS; it explicitly tests whether or not toupper() modifies characters which are not lower case. I can't imagine why it does this, and CLISP doesn't even appear to use toupper(). Now, since STDC_HEADERS is not defined, later checks don't #include , and a bunch of prototypes are not checked. Thus, the other #defines end up being wrong. But we really need to #define STDC_HEADERS for other reasons, and if we don't change the prototypes we lose during compilation. Does that make sense? From loewenst@blaze.rutgers.edu Wed Mar 10 17:38:07 1993 Return-Path: Received: from blaze.rutgers.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03230; Wed, 10 Mar 93 17:38:07 +0100 Received: by blaze.rutgers.edu (5.59/SMI4.0/RU1.5/3.08) id AA10618; Wed, 10 Mar 93 11:38:07 GMT Date: Wed, 10 Mar 93 11:38:07 GMT From: loewenst@blaze.rutgers.edu (David Loewenstern) Message-Id: <9303101638.AA10618@blaze.rutgers.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Error in #\# Reply-To: loewenst@paul.rutgers.edu CLISP appears to have a problem with #\#-dispatching. I have noticed two places where this is an issue: 1. You cannot input an array in the form #(...) to top-level. Arrays do print normally in this format (assuming *print-array* is set correctly, I suppose). 2. I got a copy of Richard Waters' Series macro package, and tried to install it under CLISP. Installation progressed normally, but the package-defined reader-macros #Z and #M can be neither read (from top-level) nor printed. Series installs normally under Allegro CL, and has been stable for at least a year, so I don't believe it's the fault of the package. This is under DOS. These opinions are shareware. If you like the product, please send your $0.02 to David Loewenstern || From haible Wed Mar 10 19:44:04 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03632; Wed, 10 Mar 93 19:44:04 +0100 Date: Wed, 10 Mar 93 19:44:04 +0100 From: haible (Bruno Haible) Message-Id: <9303101844.AA03632@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Error in #\# > CLISP appears to have a problem with #\#-dispatching. > 1. You cannot input an array in the form #(...) to top-level. Arrays > do print normally in this format. There is no problem with #\#. But most of the objects produced by #\# read macros (arrays, pathnames, readtables etc.) are not self-evaluating forms (cf. CLtL section 5.1.1). You cannot use them as forms, you must quote them. Bruno Haible From haible Wed Mar 10 19:49:43 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03677; Wed, 10 Mar 93 19:49:43 +0100 Date: Wed, 10 Mar 93 19:49:43 +0100 From: haible (Bruno Haible) Message-Id: <9303101849.AA03677@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Series package > 2. I got a copy of Richard Waters' Series macro package, and tried to > install it under CLISP. Installation progressed normally, but the > package-defined reader-macros #Z and #M can be neither read (from > top-level) nor printed. The SET-DISPATCH-MACRO-CHARACTER call is buried in the function SERIES::INSTALL, which you must call after loading the series package: clisp (compile-file "series") (load "series") (series::install) (compile-file "seriestest") (load "seriestest") For the Series package to run in CLISP, there are patches needed, however. I have put them in /pub/lisp/clisp/packages on ma2s2.mathematik.uni-karlsruhe.de as well as the whole series package itself. Bruno Haible From hyun@aristotle.ils.nwu.edu Thu Mar 11 22:09:20 1993 Return-Path: Received: from aristotle.ils.nwu.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12192; Thu, 11 Mar 93 22:09:20 +0100 Received: by aristotle.ils.nwu.edu (4.1/SMI-ACNS-1.03) id AA26420; Thu, 11 Mar 93 14:49:09 CST From: hyun@aristotle.ils.nwu.edu (Seungseok Hyun) Message-Id: <9303112049.AA26420@aristotle.ils.nwu.edu> Subject: CLISP with ILISP in Emacs To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Thu, 11 Mar 93 14:49:08 CST In-Reply-To: <9303092210.AA28238@ma2s2.mathematik.uni-karlsruhe.de>; from "Bruno Haible" at Mar 9, 93 11:10 pm X-Mailer: ELM [version 2.3 PL11] Hi! I am trying to figure out how to use CLISP with ILISP from CMU. (ILISP is for interfacing Emacs and a LISP program; I think ILISP offers one of the best LISP environments.) In summary, I should know how to fire "clisp" using "start-process" in Emacs. When I try the following: (start-process "woof" "*scratch*" "/home/hyun/local/bin/clisp" "nil") in a *scratch* buffer, I got an error message something like Process woof exited abnormally with code 1. Since ILISP uses "start-process", ILISP can't fire clisp, either. I initially tried to do this on my 486 Linux and I thought that it might be a Linux problem, so I ftp'ed CLISP for Sparc but got the same result on a Sparc. Other LISP programs like Lucid Lisp, XLisp, Franz Lisp can be fired by the "start-process" in Emacs. What makes CLISP different??? Any idea??? Thanks. Seung -- Seungseok Hyun The Institute for the Learning Sciences (hyun@ils.nwu.edu) Northwestern University From matthias@intellektik.informatik.th-darmstadt.de Fri Mar 12 11:22:38 1993 Return-Path: Received: from rs2.hrz.th-darmstadt.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13131; Fri, 12 Mar 93 11:22:38 +0100 Received: from kiwi.intellektik.informatik.th-darmstadt.de by rs2.hrz.th-darmstadt.de with SMTP id AA16217 (5.65c/IDA-1.4.4 for ); Fri, 12 Mar 1993 11:22:39 +0100 Received: by kiwi.intellektik.informatik.th-darmstadt.de (4.1/Server-1.3/HRZ-THD) id AA11641; Fri, 12 Mar 93 11:22:38 +0100 Date: Fri, 12 Mar 93 11:22:38 +0100 From: Matthias Lindner Message-Id: <9303121022.AA11641@kiwi.intellektik.informatik.th-darmstadt.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CLISP with ILISP in Emacs Hi, I don't know exactly, what your "clisp"-script" looks like, but I guess the last "nil" in the 'start-process' might be the problem. I've just tried '(start-process "woof" "*scratch*" "clisp" "")' and it worked. (Sparc1, emacs 18.56, SunOS 4.1.1) Changing the first line of your "clisp"-script form "#!/bin/sh" to "#!/bin/sh -xv" might give you some hints. I've recently tried to get clisp and ILISP work together, but due to lack of time, I did not succeed. I get an interactive clisp-buffer, but mode reamins :load forever. After M-x panci-lisp, I can reset clisp and type and evaluate expressions in its buffer - but that's all :(. I think there needs some work to be done to patch clisp.lisp, to make all the other ILISP stuff work. If someone out there did it ... I'd be *very* pleased to receive a copy ;) --Matthias ------------------------------------------------------------------------------ Matthias Lindner FG Intellektik, FB Informatik Technische Hochschule Darmstadt Alexanderstr.10, D-6100 Darmstadt TEL: +49 6151 166651 FAX: +49 6151 135326 NET: matthias@intellektik.informatik.th-darmstadt.de ------------------------------------------------------------------------------ From najem@cs.uiuc.edu Fri Mar 12 19:37:28 1993 Return-Path: Received: from herodotus.cs.uiuc.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA14010; Fri, 12 Mar 93 19:37:28 +0100 Received: by herodotus.cs.uiuc.edu (5.65c/IDA-1.2.8) id AA15215; Fri, 12 Mar 1993 12:35:26 -0600 Date: Fri, 12 Mar 1993 12:35:26 -0600 From: "Ziad H. Najem" Message-Id: <199303121835.AA15215@herodotus.cs.uiuc.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: najem@cs.uiuc.edu Subject: a Semi-working CLISP with ILISP in Emacs Hi, I Have hacked an ilisp-mode to get clisp working with ilisp. I'm not very familier with emacs-lisp so the following is actually done by looking at other code and trial-and-error. But it works for the most part. The ilisp package resides in /usr/emacs/ilisp The clisp package resides in /usr/local/lib/lisp The necessery code is spread into two files. One part is in $HOME/.emacs and the other is in /usr/local/lib/lisp/emacs.el. I can start lisp from emacs by executing the 'zlisp' extended command (M-X zlisp) or by issuing the following script from shell #!/bin/bash exec emacs -l /usr/local/lib/lisp/emacs.el -f zlisp Finally, attached is what appears in my $HOME/.emacs and /usr/local/lib/ilisp/emacs.el. Please report any problems to me and I would be interested in seeing any enhancements. Hope that helps, - Ziad najem@cs.uiuc.edu *********************************************************************** Begin Code taken from $HOME/.emacs *********************************************************************** (autoload 'zlisp "/usr/local/lib/lisp/emacs" "Z-Lisp" t) (set-default 'auto-mode-alist (append '(("\\.lsp$" . lisp-mode)) '(("\\.cl$" . lisp-mode)) auto-mode-alist)) (defvar ziad:cl-indent-loaded nil) (setq lisp-mode-hook '(lambda () (require 'zlisp "/usr/local/lib/lisp/emacs") (if ziad:cl-indent-loaded nil (load "cl-indent") (setq ziad:cl-indent-loaded t)) (turn-on-auto-fill) (setq fill-column 75) (setq comment-column 25) (setq completion-ignored-extensions (append '(".lib" ".fas") completion-ignored-extensions)) (setup-gnu-indentation) )) (defun setup-gnu-indentation () (let ((tag 'common-lisp-indent-hook)) (put 'defpackage tag 1) (put 'if tag 2) ;;; For Minerva in Lisp (put 'internalize tag (get 'do tag)) (put 'in-kb tag '(4 2)) (put 'if-prolog tag (get 'if tag)) (put 'if-prolog-let tag '(6 6 4 2)) (put 'when-prolog tag (get 'when tag)) (put 'when-prolog-let tag '(6 4 2)) (put 'unless-prolog tag (get 'unless tag)) (put 'while-prolog tag (get 'when tag)) (put 'while-prolog-let tag (get 'do tag)) (put 'var-let tag (get 'let tag)) (put 'lisp-prover tag (get 'defun tag)) (put 'p-unify tag (get 'defun tag)) (setq lisp-indent-hook 'common-lisp-indent-hook) )) *********************************************************************** End Code taken from $HOME/.emacs *********************************************************************** *********************************************************************** Begin Code taken from /usr/local/lib/lisp/emacs.el *********************************************************************** (setq load-path (cons "/usr/emacs/ilisp" load-path)) (setq ilisp-load-hook '(lambda () (setq ilisp-prefix "\C-c" zlisp-program "/usr/local/lib/lisp/lisp.run -q -M /usr/local/lib/lisp/lispinit.mem -i /usr/local/lib/lisp/cltl2"))) (require 'ilisp) (defun zlisp-check-prompt (old new) "Compare the break level printed at the beginning of the prompt." (let* ((old-level (if (and old (string-match "[0-9]+" old)) (string-to-int (substring old (match-beginning 0) (match-end 0))) 0)) (new-level (if (string-match "[0-9]+" new) (string-to-int (substring new (match-beginning 0) (match-end 0))) 0))) (<= new-level old-level))) (defdialect zlisp "Linux CLisp" clisp (ilisp-load-init 'zlisp "/usr/emacs/ilisp/clisp.lisp") (setq ilisp-load-or-send-command "(and (or (print \"%s\") t) (load \"%s\"))" ilisp-binary-extension "fas" ilisp-motd "Zodiac-Lisp V%s" comint-prompt-regexp "^[^> \t]*>" ilisp-other-prompt "^[ \t]+" comint-fix-error "unwind" comint-continue "continue" comint-interrupt-regexp "\*\* \- Continuable Error" lisp-wait-p t lisp-no-popper nil comint-prompt-status (function (lambda (old line) (comint-prompt-status old line 'zlisp-check-prompt))))) (provide 'zlisp) (require 'cl) (load "fi/utils.el") (load "fi/indent.el") *********************************************************************** End Code taken from /usr/local/lib/lisp/emacs.el *********************************************************************** From haible Sat Mar 20 00:05:12 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA27414; Sat, 20 Mar 93 00:05:12 +0100 Date: Sat, 20 Mar 93 00:05:12 +0100 From: haible (Bruno Haible) Message-Id: <9303192305.AA27414@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: New CLISP version There is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . The list of supported platforms now include Amiga 3000/4000 and Atari TT. The installation procedure on Unix is now simpler. >From the change log: User visible changes -------------------- * Replaced variable *TEMPFILE* by a function EDITOR-TEMPFILE. Portability ----------- * Unix version: Corrected detection of ANSI C header files. * Unix version: Check whether signal handlers are automatically reinstalled when activated. * Made readline library ANSI C compliant and (on Unix) auto-configuring. * In case there is no room for three extra bits in a symbol pointer, allocate three stack words per binding in a VAR frame (instead of two stack words). This completes the Amiga 3000/4000 version. * Completed memory model SPVW_PURE_BLOCKS: There is a large contiguous memory area for each kind of Lisp object. This model is used only if memory can be allocated at prescribed addresses, for example on Unix with mmap(). * More 386BSD support. Other modifications ------------------- * Remove limit on the number of files that may be specified on the command line. Limit was 1000. * Miscellaneous documentation updates. Enjoy! Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From UNM42E@IBM.rhrz.uni-bonn.de Tue Mar 23 18:48:20 1993 Return-Path: Received: from IBM.rhrz.uni-bonn.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07519; Tue, 23 Mar 93 18:48:20 +0100 Message-Id: <9303231748.AA07519@ma2s2.mathematik.uni-karlsruhe.de> Received: from DBNRHRZ1.BITNET by IBM.rhrz.uni-bonn.de (IBM VM SMTP V2R2) with BSMTP id 9209; Tue, 23 Mar 93 19:26:56 MEZ Received: from DBNRHRZ1 (UNM42E) by DBNRHRZ1.BITNET (Mailer R2.08 R208004) with BSMTP id 1138; Tue, 23 Mar 93 19:26:55 MEZ Date: Tue, 23 Mar 93 19:26:19 MEZ From: Michael Stoll Subject: clisp-list Test To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de You may delete this message. It's only a test. - Michael From frankg@halcyon.halcyon.com Mon Mar 29 04:42:20 1993 Return-Path: Received: from halcyon.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19129; Mon, 29 Mar 93 04:42:20 +0200 Received: by halcyon.com id AA21876 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Sun, 28 Mar 1993 18:47:07 -0800 Date: Sun, 28 Mar 1993 18:47:07 -0800 From: Frank Gleason Message-Id: <199303290247.AA21876@halcyon.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: CLX Hi, Sorry if this question has an obvious answer I'm new to lisp. When I try to compile and run CLX I get errors because I don't have a set of functions (module?) that begin with system::. I'm using clisp on Linux, I think the version is 2.6, and pcl+clx.sept92f. Also if I get this working will garnet run under clisp? Thanks in advance Frank Gleason frankg@halcyon.com From nils@guru.ims.uni-stuttgart.de Mon Mar 29 22:24:25 1993 Return-Path: Received: from guru.ims.uni-stuttgart.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA22741; Mon, 29 Mar 93 22:24:25 +0200 Received: by guru.ims.uni-stuttgart.de (Smail3.1.28.1 #2) id m0ndRKX-0001bkC; Mon, 29 Mar 93 22:25 MET Message-Id: Date: Mon, 29 Mar 93 22:25 MET From: nils@guru.ims.uni-stuttgart.de (Cornelis van der Laan) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CLX Frank Gleason wrote: When I try to compile and run CLX I get errors because I don't have a set of functions (module?) that begin with system::. I'm using clisp on Linux, I think the version is 2.6, and pcl+clx.sept92f. Symbols beginning with system:: are located in the system package (and can be variables, functions or just unbound symbols). I can't see any reference to the system package in my sources of CLX that are not conditionalized for lisp dialects that actually have the system package. Could you be more specific on what error message you get ? Are you using the packages specially adapted to clisp (to be found at the usual place in Karlsruhe) or the original sources ? Thanks in advance Frank Gleason frankg@halcyon.com From frankg@halcyon.halcyon.com Mon Mar 29 23:20:11 1993 Return-Path: Received: from halcyon.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA22851; Mon, 29 Mar 93 23:20:11 +0200 Received: by halcyon.com id AA19704 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Mon, 29 Mar 1993 13:20:00 -0800 Date: Mon, 29 Mar 1993 13:20:00 -0800 From: Frank Gleason Message-Id: <199303292120.AA19704@halcyon.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de >>Frank Gleason wrote: >> When I try to compile and run CLX I get errors because I don't >> have a set of functions (module?) that begin with system::. >> I'm using clisp on Linux, I think the version is 2.6, and >> pcl+clx.sept92f. >> >Symbols beginning with system:: are located in the system package >(and can be variables, functions or just unbound symbols). >I can't see any reference to the system package in my sources of >CLX that are not conditionalized for lisp dialects that actually >have the system package. Could you be more specific on what error >message you get ? >Are you using the packages specially adapted to clisp (to be found >at the usual place in Karlsruhe) or the original sources ? I should have done this in the first message you are right. I got the file pcl+clx.sept92f.clisp.tar.Z from gatekeeper.dec.com. Clisp seems to be installed correctly (I installed and ran maxima). I am not at my machine with clisp so I can't copy the message exactly but here is the code segments that are the problem. Grepping for any of these (ie: grep big-endian *) in the clisp source comes up empty. depdefs.lsp line 172 #+CLISP (eval-when (compile eval load) (unless system::*big-endian* (pushnew :clx-little-endian *features*)) ) dependent.lsp line 1430 #+CLISP (defun open-x-stream (host display protocol) protocol ;; unused (let ((stream (system::make-socket-stream host display))) (if (streamp stream) stream (error "Cannot connect to server: ~A:~D" host display)))) dependent.lsp line 1591 #+CLISP (defun buffer-read-default (display vector start end timeout) (declare (type display display) (type buffer-bytes vector) (type array-index start end) (type (or null fixnum) timeout)) #.(declare-buffun) (let ((stream (display-input-stream display))) (cond ((and (eql timeout 0) (not (listen stream))) :timeout) (t (system::read-n-bytes stream vector start (- end start)) nil) ) ) ) dependent.lsp line 1721 #+CLISP (defun buffer-write-default (vector display start end) (declare (type buffer-bytes vector) (type display display) (type array-index start end)) #.(declare-buffun) (system::write-n-bytes (display-output-stream display) vector start (- end start)) nil ) dependent.lsp line 2619 (defun resources-pathname () (when #+(or unix mach) t #-(or unix mach) (search "Unix" (software-type) :test #'char-equal) (or #+(or excl CLISP (and lcl3.0 (not vax-vms)) CMU) (let ((string #-CMU (#+excl sys:getenv #+CLISP system::getenv #+lcl3.0 lcl:environment-variable "XENVIRONMENT") #+CMU (cdr (assoc :xenvironment ext:*environment-list*)))) (when string (pathname string))) (merge-pathnames (user-homedir-pathname) (pathname (concatenate 'simple-string ".Xdefaults-" #+excl (short-site-name) #-excl (machine-instance))))))) From nils@guru.ims.uni-stuttgart.de Tue Mar 30 00:07:47 1993 Return-Path: Received: from guru.ims.uni-stuttgart.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA22969; Tue, 30 Mar 93 00:07:47 +0200 Received: by guru.ims.uni-stuttgart.de (Smail3.1.28.1 #2) id m0ndSwZ-0001alC; Tue, 30 Mar 93 00:08 MET Message-Id: Date: Tue, 30 Mar 93 00:08 MET From: nils@guru.ims.uni-stuttgart.de (Cornelis van der Laan) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Frank Gleason wrote: >>Frank Gleason wrote: >> When I try to compile and run CLX I get errors because I don't >> have a set of functions (module?) that begin with system::. >> I'm using clisp on Linux, I think the version is 2.6, and >> pcl+clx.sept92f. >> >Symbols beginning with system:: are located in the system package >(and can be variables, functions or just unbound symbols). >I can't see any reference to the system package in my sources of >CLX that are not conditionalized for lisp dialects that actually >have the system package. Could you be more specific on what error >message you get ? >Are you using the packages specially adapted to clisp (to be found >at the usual place in Karlsruhe) or the original sources ? I should have done this in the first message you are right. I got the file pcl+clx.sept92f.clisp.tar.Z from gatekeeper.dec.com. Clisp seems to be installed correctly (I installed and ran maxima). I am not at my machine with clisp so I can't copy the message exactly but here is the code segments that are the problem. Grepping for any of these (ie: grep big-endian *) in the clisp source comes up empty. [ code deleted ] You are using and old version of clisp; I've got a rather new one and its source sparkles from functions in the system package. Basically, the system package is for non CL extensions and optimizations, like MAKE-SOCKET-STREAM or specialized i/o routines for faster communication with the Xserver. The best thing would be to get the newest version of clisp; alternatively you could delete the symbol CLISP from *FEATURES* so that the conditionals in the CLX source don't fire. CLX works slower then but works. Use (setq *features* (delete 'CLISP *features*)) to do so. And watch for the endianness of your machine, the i386 is little-endian (long long way for me to find that out...). Nils. ---------------------------------------------------------------- Cornelis van der Laan -- nils@ims.uni-stuttgart.de -- nils@guru.stgt.sub.org # echo Knusper Knusper Knaeuschen > /etc/nologin From haible Tue Mar 30 00:11:09 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23012; Tue, 30 Mar 93 00:11:09 +0200 Date: Tue, 30 Mar 93 00:11:09 +0200 From: haible (Bruno Haible) Message-Id: <9303292211.AA23012@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list, haible@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CLX Frank Gleason observes that the port of CLX to CLISP uses the variable system::*big-endian* and the functions system::getenv, system::make-socket-stream, system::read-n-bytes, system::write-n-bytes. system::getenv and system::make-socket-stream should be there in any case. The other three I have added to CLISP around November 1992, especially with CLX in mind. > I'm using clisp on Linux, I think the version is 2.6, and pcl+clx.sept92f. A version number "2.6" is unknown to me. Try (LISP-IMPLEMENTATION-VERSION). If you get something before January 1993, then upgrade. New Linux binaries are on tsx-11.mit.edu (/pub/linux/packages/lisp/ or something like this) and ma2s2.mathematik.uni-karlsruhe.de /pub/lisp/clisp/linux/. > Also if I get this working will garnet run under clisp? I can't promise anything, as I have only 8 MB RAM and this isn't enough for compiling garnet. Did anyone get through it? Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From @serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de Thu Apr 1 14:22:46 1993 Return-Path: <@serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de> Received: from uni-kl.de (stepsun.uni-kl.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA29758; Thu, 1 Apr 93 14:22:46 +0200 Received: from serv-200.dfki.uni-kl.de by stepsun.uni-kl.de id aa28740; 1 Apr 93 14:21 MET DST Received: from serv-300.dfki.uni-kl.de by serv-200.dfki.uni-kl.de id aa01613; 1 Apr 93 12:22 GMT Date: Thu, 1 Apr 93 12:22:16 GMT From: Bjoern Hoefling To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: hoefling@dfki.uni-kl.de Subject: Q: Clisp on Dos or Linux? Organization: DFKI, University of Kaiserslautern, Germany Message-Id: <9304011222.aa11013@serv-300.dfki.uni-kl.de> Hi, I will begin soon with my master thesis and I have the following problem: I will have to use the hybrid knowledge representation system COLAB which has been developped here (DFKI at the University of Kaiserslautern) on a PC (386 or 486). To run COLAB on a PC forces us to use a Lisp-Dialect available for PC, like CLisp. The decision, whether to use the Clisp version under DOS or under Linux (which has not yet been installed in the Enterprise where I will make the master thesis) depends on the fact how difficult it is to write a nice, inter- active User Interface with it. Therefore I have the following questions: 1. Which user interface toolkits for Lisp-Systems, which are freely available can be used under Clisp with DOS and which under Clisp with Linux (for example: CLX, CLUE, Garnet, XIT, ... or the corresponding tools under DOS/ Windows)? Information about experience with such a toolkit would be also very useful. 2. Are there any publicly available hypertext browser and hypertext editors (an interface to clisp is not strictly required) and/or a cross-referencer under DOS or Linux? I know that Linux seems to be the more promising OS, but is has the disadvantage, that in order to make COLAB run on another PC one has to install first the whole Linux-System (with very much hd-capacity needed). That's why I would like to gather also some information about Clisp under DOS. Thanks in advance for any help. Bjoern Hoefling (hoefling@dfki.uni-kl.de) From schue@ira.uka.de Thu Apr 1 15:11:50 1993 Return-Path: Received: from iraun1.ira.uka.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA29966; Thu, 1 Apr 93 15:11:50 +0200 Received: from ira.uka.de by iraun1.ira.uka.de with SMTP (PP) id <26059-0@iraun1.ira.uka.de>; Thu, 1 Apr 1993 15:05:39 +0200 Received: from deathstar.iaks.ira.uka.de by walhalla.eiss.ira.uka.de (4.1/SMI-4.1) id AA14597; Thu, 1 Apr 93 15:07:10 +0200 Received: from miraculix.iaks.ira.uka.de by deathstar.iaks.ira.uka.de (4.1/SMI-4.1) id AA09157; Thu, 1 Apr 93 15:03:27 +0200 Date: Thu, 1 Apr 93 15:03:27 +0200 From: schue@ira.uka.de (Joachim Schue) Message-Id: <9304011303.AA09157@deathstar.iaks.ira.uka.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Performance of CLISP Are there any results of a comparison between CLISP and Lucid Lisp (Performance, ...) ?. Scince we are thinking about dropping our license of Lucid Lisp I would be very pleased of any estimation of this idea. Thanks in advance, Joachim From nils@guru.ims.uni-stuttgart.de Thu Apr 1 23:05:06 1993 Return-Path: Received: from guru.ims.uni-stuttgart.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00938; Thu, 1 Apr 93 23:05:06 +0200 Received: by guru.ims.uni-stuttgart.de (Smail3.1.28.1 #2) id m0neXNs-0001bPC; Thu, 1 Apr 93 23:05 MET Message-Id: Date: Thu, 1 Apr 93 23:05 MET From: nils@guru.ims.uni-stuttgart.de (Cornelis van der Laan) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Q: Clisp on Dos or Linux? Bjoern Hoeflin wrote: > I will begin soon with my master thesis and I have the following problem: > I will have to use the hybrid knowledge representation system COLAB which has > been developped here (DFKI at the University of Kaiserslautern) on a PC (386 or 486). To run COLAB on a PC forces us to use a Lisp-Dialect available for PC, > like CLisp. The decision, whether to use the Clisp version under DOS or under > Linux (which has not yet been installed in the Enterprise where I will make the master thesis) depends on the fact how difficult it is to write a nice, inter- > active User Interface with it. Therefore I have the following questions: > > 1. Which user interface toolkits for Lisp-Systems, which are freely available > can be used under Clisp with DOS and which under Clisp with Linux > (for example: CLX, CLUE, Garnet, XIT, ... or the corresponding tools under DOS/ > Windows)? Information about experience with such a toolkit would be also very > useful. You can use all CLX based toolkit interfaces in CLISP with an OS that supports socket connections to an Xserver. This need not exclude DOS but (guess what...) it will be painful to get DOS/CLISP to do that. You'd need to look into network drivers like KA9Q, build the proper interface into CLISP, debug it etc etc. And then: still no X server around, since those run best on UNIX boxes and are quite expensive for DOS boxes, although available... > 2. Are there any publicly available hypertext browser and hypertext editors (an interface to clisp is not strictly required) and/or a cross-referencer under DOS or Linux? > > I know that Linux seems to be the more promising OS, but is has the disadvantage, that in order to make COLAB run on another PC one has to install first the > whole Linux-System (with very much hd-capacity needed). That's why I would like > to gather also some information about Clisp under DOS. What is DOS other than big P**? Go Linux or get a decent Unix box. And take all others with you. Another option might be to contact Franz Inc for getting a version of Allegro Common Lisp including CLIM (the Genera like window interface) for PC/Windows. This isn't free and it might be not cheap but does run under DOS and the sources are portable to other platforms that support CLIM like Sparc, RS6000, MacIntosh etc etc. (Email me for more concrete information.) > > Thanks in advance for any help. > > Bjoern Hoefling (hoefling@dfki.uni-kl.de) > > Not much help but then, you didn't pay for it either :,-) Nils PS: I don't reformat those broken lines above, instead, Bjoern: please get used to post/mail with a maximum line width of 70 chars. ---------------------------------------------------------------- Cornelis van der Laan -- nils@ims.uni-stuttgart.de -- nils@guru.stgt.sub.org # echo Knusper Knusper Knaeuschen > /etc/nologin From haible Fri Apr 2 01:02:58 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01062; Fri, 2 Apr 93 01:02:58 +0200 Date: Fri, 2 Apr 93 01:02:58 +0200 From: haible (Bruno Haible) Message-Id: <9304012302.AA01062@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: Performance of CLISP Joachim Schue asks: > Are there any results of a comparison between CLISP and > Lucid Lisp (Performance, ...) ? Results from Timing the Benchmarks in Richard P. Gabriel: Performance and Evaluation of Lisp Systems Systems: 4 = CLISP (Version February 1992, gcc2 -O2) on (SUN Sparcstation 2, 16MB), measured Run Time (< Real Time !) 5 = CLISP (Version July 1992, gcc2 -O) on (486/33MHz, 8MB, Linux), measured Run Time (< Real Time !) 6 = Lucid Common Lisp on SUN-4 (SUN Sparcstation 2, 16MB), measured Total Run Time (< Elapsed Real Time !) 7 = Lucid Common Lisp on SUN-4c (SUN Sparcstation 1 IPC, 8MB), measured Total Run Time (< Elapsed Real Time !) c = compiled m = compiled to machine code Times in seconds 4c 5c 6c 7m TAK 0.83 1.33 0.10 0.09 TAKr 1.60 1.67 0.11 0.11 STAK 1.57 2.77 0.92 0.94 CTAK 3.27 1.99 0.34 0.52 TAKL 3.90 6.47 1.82 0.41 BOYER 17.65 16.14 4.13 7.71 BROWSE 14.18 17.75 5.12 4.85 DESTRU 1.65 3.24 0.48 0.45 TRAVERSinit 13.54 24.08 3.11 2.27 TRAVERSrun 99.26 80.99 10.11 5.43 DERIV 2.61 3.21 0.68 1.95 DDERIV 3.26 3.59 0.79 2.31 DIV2iter 1.18 1.60 0.33 1.49 DIV2recur 2.30 2.13 1.23 3.14 FFT 9.96 18.54 6.73 0.46 PUZZLE 15.31 29.88 5.36 1.16 TRIANGLE 291.84 521.18 42.91 14.74 FPRINT 0.21 0.31 0.14 2.24 FREAD 0.23 0.37 0.18 0.61 TPRINT 2.32 0.53 0.10 [8.71] [] with *PRINT-PRETTY* = T POLY_I_2 0.00 0.00 0.01 0.01 POLY_B_2 0.01 0.01 0.01 0.00 POLY_F_2 0.01 0.01 0.00 0.01 POLY_I_5 0.08 0.08 0.02 0.02 POLY_B_5 0.08 0.10 0.04 1.24 POLY_F_5 0.08 0.09 0.02 0.02 POLY_I_10 0.82 0.97 0.18 0.12 POLY_B_10 1.02 1.15 0.41 0.58 POLY_F_10 0.94 1.01 0.29 1.33 POLY_I_15 5.71 6.79 1.20 0.82 POLY_B_15 8.03 9.10 3.58 5.65 POLY_F_15 6.21 7.19 1.86 2.75 I measured these myself, 4 in April 1992, 5 in July 1992, 6,7 in March 1991. Conclusions: * CLISP is approximately as fast on a Sun Sparc 2 as on a 486 (33 MHz). * Lucid Lisp is always faster than CLISP. This is because Lucid Lisp compiles to native machine code and makes use of the special "tagged" SPARC instructions. Other Notes: * This measured only pure Run Time, not Real Time (also called Elapsed Time). Real Time includes time for swapping and page faults. Since compiled CLISP code tends to be much smaller than compiled Lucid Lisp code this should affect Lucid more than CLISP. * This measured only the speed of compiled code. Development speed also depends on the speed of interpreted code and on the compiler's speed. I recall that Lucid's machine code compiler (7m above) is not very fast. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From yue@maxwell.mmwb.ucsf.EDU Fri Apr 2 09:27:22 1993 Return-Path: Received: from cgl.ucsf.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01374; Fri, 2 Apr 93 09:27:22 +0200 Received: from maxwell.mmwb.ucsf.EDU by cgl.ucsf.EDU (5.65/GSC4.22) id AA18366 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de; Thu, 1 Apr 93 23:26:51 -0800 Received: by maxwell.mmwb.ucsf.EDU (4.1/GSC4.20) id AA00259; Thu, 1 Apr 93 23:26:51 PST Date: Thu, 1 Apr 93 23:26:51 PST From: yue@maxwell.mmwb.ucsf.EDU (Kaizhi Yue) Message-Id: <9304020726.AA00259@maxwell.mmwb.ucsf.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Running CLISP inside Emacs The OS2 port of Clisp makes it possible to run LISP in an Emacs subprocess, which makes it much easier to develop LISP programs. Bruno Haible suggested that I describe some functionalities that I know of. They are mainly the following. M-x run-lisp allows one to start an inferior lisp process in emacs. LISP-SEND-DEFUN, invoked by c-m-x (in the convention of symbolics), sends an lisp expression (usually defun form) inside a LISP buffer to the lisp process to be evaluated and the value will be printed to the inferior lisp buffer. LISP-SEND-COMPILE (c-m-c) allows one to compile a function in a LISP buffer. By defining a proper shell-prompt-pattern (the default is used in the definition of RUN-LISP), the commands SHELL-PREV-COMMAND (alt-p), SHELL-NEXT-COMMAND (alt-n), etc allow one to move the cursor to the previous user inputs. Thus are accomplished functionalities similar to KEYS in OS2, when combined with other emacs shell commands, e.g., SHELL-SEND-INPUT, COPY-LAST-SHELL-INPUT, etc. These shell commands and key bindings are described in the Emacs manual. Most of the functions that are needed such as RUN-LISP, LISP-SEND-DEFUN, SHELL-SEND-INPUT, SHELL-PREV-COMMAND, SHELL-NEXT-COMMAND, etc. already exist in shell.el in EMACS LISP. (The latter two are not found in the OS/2 port of Emacs release but can be found in the Unix port. --- At least I did at our site.) Some slight modifications are needed to make the functions work in OS2. And some are very ad hoc. E.g., when running CLISP in EMACS shell, when the input is more than one line, the read-eval-print loop fails to print the prompt for the next command. So I put in a second (one-line) expression for the lisp process to evaluate so that a prompt will be generated. I do it by evaluating "!" which should be previously defined in CLISP as (setq ! '!). I have included the code for the above functionalities in this post. This includes some LISP code and some Emacs lisp code. They are provided as is (and other standard disclaimers apply). I am sure there are better ways to do these things. But if you would like to take some existing code and play with it, here it is. By the way, Bill Schelter has quite some goodies for using LISP in Emacs shell. E.g., he has a utility for reading documentations of common lisp functions that is similar to C-h d in Emacs. These can be found in his enhancement for AKCL, which is available through FTP at rascal.ics.utexas.edu:pub/akcl-XX.tar.Z. The files are in a directory named something like .../kcl/doc/.... Look for the following files: akcl.el, find-doc.el, etc. ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;;;; Put the following in an init file for ***LISP*** ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; (defmacro compile-if-wellformed (def) `(if (and (eq 'defun (car ,def)) (symbolp (cadr ,def)) (listp (caddr ,def)) (>= (length ,def) 4)) (compile (cadr ,def)(cons 'lambda (cddr ,def))) 'NOT-A-FUNCTION-DEFINITION-FORM-AND-NOT-COMPILED)) (setq ! '!) ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;;; Put the following in an initialization file for Emacs ;;; --- I did not figure out a way of putting them into shell-mode-map. ;;;;;;;;;;;;;;;;;;;;;;; (define-key ext-map "\061" 'shell-next-command) ;;; A-p (define-key ext-map "\031" 'shell-prev-command) ;;; A-n ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;;; Load the following part at startup or through ;;; (autoload 'run-lisp "f:/emacs/custom/inferior-lisp" nil t) ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; (require 'shell) (defun run-lisp (&optional startup-name) "Run an inferior Lisp process, input and output via buffer *lisp*." (interactive nil) ;(interactive "sEnter CL program name: ") (switch-to-buffer (make-shell "lisp" "c:\\os2\\cmd.exe" nil)) (process-send-string "lisp" "f:\\usr\\bin\\lisp.cmd\n") ;; (switch-to-buffer (make-shell "lisp" "c:\\os2\\cmd.exe" "f:\\usr\\bin\\lisp. (inferior-lisp-mode) (make-local-variable 'shell-prompt-pattern) (setq shell-prompt-pattern "^[^#%)>]*[#%)>]+ *")) (defun define-lisp-mode-key (key command) (define-key lisp-mode-map key command) (define-key inferior-lisp-mode-map key command)) ;; %%% copied from /emacs/lisp/lisp-mode.el ;;;; **** to use the kludge in this function, you must have (setq > '>) (defun lisp-send-defun () "Send the current defun to the Lisp process made by run-lisp." (interactive) (save-excursion (end-of-defun) (message "defining common lisp function ....") (let ((end (point))) (beginning-of-defun) ;; Reset so we can grab the value. (setq lisp-last-output-position nil) (process-send-region "lisp" (point) end) (process-send-string "lisp" "!\n") ;;; well, now it works 3/14, while it did (switch-to-buffer (process-buffer(get-process "lisp"))) (em-end-of-buffer) (message "defined")))) ; using the code in shell.el can move the cursor, but the point and mark are w ;;(define-lisp-mode-key "\C-\M-x" 'lisp-send-defun) (define-lisp-mode-key "\e\C-x" 'lisp-send-defun) ;(define-lisp-mode-key "\e\C-y" 'lisp-send-defun) not necessary ;(defun lisp-send-compile () ; ; "Send the current defun to the Lisp process made by run-lisp." ; (interactive) ; (save-excursion ; (end-of-defun) ; (message "Compiling common lisp function ....") ; (let ((end (point))) ; (beginning-of-defun) ; ;; Reset so we can grab the value. ; (setq lisp-last-output-position nil) ; (process-send-string "lisp" "(let ((def ' ") ; (process-send-region "lisp" (point) end) ; (process-send-string "lisp" "))") ; (process-send-string "lisp" ; "(if (and (eq 'defun (car def)) ; (symbolp (cadr def)) ; (listp (caddr def)) ; (every #'atom (caddr def)) ; (>= (length def) 4)) ; (compile (cadr def)(cons 'lambda (cddr def))) ; 'Not-a-function-definition-form))") ; (process-send-string "lisp" "\n")))) ;; this is IN CL (defun lisp-send-compile () "Send the current defun to the run-lisp process for compilation." (interactive) (save-excursion (end-of-defun) (message "Compiling COMMON LISP function ....") (let ((end (point))) (beginning-of-defun) ;; Reset so we can grab the value. (setq lisp-last-output-position nil) (process-send-string "lisp" "(let ((def ' ") (process-send-region "lisp" (point) end) (process-send-string "lisp" "))") ;; closed ...((def...)) (process-send-string "lisp" "(compile-if-wellformed def))") (process-send-string "lisp" ":q\n") ;;; surprise! ))) (define-lisp-mode-key "\e\C-c" 'lisp-send-compile) (define-lisp-mode-key "\e\C-x" 'lisp-send-defun) ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;;; The following is copied from a Unix version of SHELL.EL, ;;; included here for your convenience. They should be loaded ;;; in shell mode or through autoload. ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; (define-key shell-mode-map "\C-i" 'shell-file-name-completion) (define-key shell-mode-map "\C-m" 'shell-send-input) ;;; We have to put these in ext-map. ;;; (define-key shell-mode-map "\061" 'shell-next-command) ;;; (define-key shell-mode-map "\031" 'shell-prev-command) (defun shell-file-name-completion () "Preform file name completion in shell mode" (interactive) (let ((shell-expand-name nil) (shell-expand-dir nil) (shell-expand-file nil) (shell-expand-try nil) (shell-expand-name-begin nil) (shell-expand-name-end nil)) ;; look ahead (re-search-backward shell-pathname-pattern nil t) (forward-char) (setq shell-expand-name-begin (point)) ;; look back (if (re-search-forward shell-pathname-pattern nil 0) (backward-char)) (setq shell-expand-name-end (point)) ;; the name requiring expansion (setq shell-expand-name (buffer-substring shell-expand-name-begin shell-expand-name-end)) ;; directory part of name (setq shell-expand-dir (or (file-name-directory shell-expand-name) default-directory)) ;; file part of name (setq shell-expand-file (file-name-nondirectory shell-expand-name)) ;; do the expansion (setq shell-expand-try (file-name-completion shell-expand-file shell-expand-dir)) ;; display the results (if (eq shell-expand-try t) (message "Sole completion") (if (eq shell-expand-try nil) (message "No match") (if (equal shell-expand-try shell-expand-file) (progn (if shell-completions-window nil (setq shell-completions-window (current-window-configuration))) (message "Making completion list...") (with-output-to-temp-buffer " *Completions*" (display-completion-list (sort (file-name-all-completions shell-expand-try shell-expand-dir) 'string-lessp))) (message "")) (progn ;; put in the expansion (re-search-backward (concat shell-filename-pattern "\\|" shell-pathname-pattern) nil t) (forward-char) (delete-region (point) shell-expand-name-end) (insert shell-expand-try))))))) (defun shell-completion-cleanup () "Clean up windows after shell file name completion. Modified to deal with screens created by show-temp-buffer.el -- MON." (interactive) (if (and (boundp 'read-minibuf-unwind-hook) ;MON (boundp 'epoch-temp-buffer-unwind-screen) ;MON epoch-temp-buffer-unwind-screen) ;MON (funcall read-minibuf-unwind-hook)) ;MON (if shell-completions-window (save-excursion (set-window-configuration shell-completions-window) (setq shell-completions-window nil)))) (defun kill-all-output-from-shell () "Kill shell buffer above current prompt." (interactive) (goto-char (point-max)) (re-search-backward shell-prompt-pattern nil t) (kill-region (point-min) (point)) (goto-char (point-max))) (defun shell-next-command () "Search for the next command in a shell." (interactive) (re-search-forward shell-prompt-pattern nil t)) (defun shell-prev-command () "Search for the previous command in a shell." (interactive) (beginning-of-line) (re-search-backward shell-prompt-pattern nil t) (re-search-forward shell-prompt-pattern nil t)) From yue@maxwell.mmwb.ucsf.EDU Fri Apr 2 09:59:24 1993 Return-Path: Received: from cgl.ucsf.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01427; Fri, 2 Apr 93 09:59:24 +0200 Received: from maxwell.mmwb.ucsf.EDU by cgl.ucsf.EDU (5.65/GSC4.22) id AA18494 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de; Thu, 1 Apr 93 23:58:53 -0800 Received: by maxwell.mmwb.ucsf.EDU (4.1/GSC4.20) id AA00399; Thu, 1 Apr 93 23:58:53 PST Date: Thu, 1 Apr 93 23:58:53 PST From: yue@maxwell.mmwb.ucsf.EDU (Kaizhi Yue) Message-Id: <9304020758.AA00399@maxwell.mmwb.ucsf.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Running CLISP inside Emacs (cont.) To make meta-x run-lisp work for your lisp, it is assumed that you have defined a command file lisp.cmd that is in the path "f:/usr/bin/". For your particular lisp, this line needs to be modified. Alternatively, you may simply call "lisp.exe -M lispinit.mem ..." directly. Also please note that I have assumed that your cmd.exe is in "C:os2" path. Kai From LARSSON@ntcclu.ntc.nokia.com Fri Apr 2 10:03:15 1993 Return-Path: Received: from NTC02.TELE.NOKIA.FI by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01471; Fri, 2 Apr 93 10:03:15 +0200 Date: Fri, 2 Apr 1993 11:04:40 +0400 (EET-DST) From: LARSSON@ntcclu.ntc.nokia.com Message-Id: <930402110440.2220f77a@ntcclu.ntc.nokia.com> Subject: Query: Amount of RAM needed for CLX etc. under Linux To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X-Vmsmail-To: INET::"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Apropos Bjoern Hoeflin's questions concerning user interface tools: I am currently setting up Linux to coexist with OS/2 on an Intel 80486 based PC. How much RAM do I need 1) To be able to run XFree. The Linux docs talk about 8Mb, but will that be sufficient to do anything useful in addition to just running the GUI? 2) To be able to run CLX under CLISP on top of X - and if I add CLUE, (Garnet, XIT), then how much. And when I am ready to implement something useful under those GUIs, then how much RAM should I calculate. In my case 'something useful' most probably has to do with natural language processing. 3) Hard disk space required during compilation and setup, assuming some of the sources can then be deleted, if I do not have to rebuild the system frequently. I am also planning to use CLISP plus any feasable combination of the above mentioned GUIs on a HP 9000s700 running HP-UX 8.07 and HP-VUE. Consequently, I would appreciate VERY much to share any experiences of possible quirks associated with these HP systems Thanks in advance, Arne Larsson *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Arne Larsson Nokia Telecommunications Translator Transmission Systems, Customer Services larsson@ntc02.tele.nokia.fi P.O. Box 12, SF-02611 Espoo, Finland larsson@ntcclu.ntc.nokia.com Phone +358 0 5117476, Fax +358 0 51044287 *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* From uho@informatik.uni-kiel.dbp.de Thu Apr 8 13:24:05 1993 Return-Path: Received: from nz11.rz.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13372; Thu, 8 Apr 93 13:24:05 +0200 X400-Received: by mta nz11.rz.uni-karlsruhe.de in /PRMD=uni-karlsruhe/ADMD=dbp/C=de/; Relayed; Thu, 8 Apr 1993 13:24:35 +0200 X400-Received: by /PRMD=UNI-KARLSRUHE/ADMD=DBP/C=DE/; Relayed; Thu, 8 Apr 1993 13:23:13 +0200 X400-Received: by /PRMD=UNI-KIEL/ADMD=DBP/C=DE/; Relayed; Thu, 8 Apr 1993 13:22:23 +0200 Date: Thu, 8 Apr 1993 13:22:23 +0200 X400-Originator: uho@informatik.uni-kiel.dbp.de X400-Recipients: clisp-list@MA2S2.MATHEMATIK.uni-karlsruhe.de X400-Mts-Identifier: [/PRMD=uni-kiel/ADMD=dbp/C=de/;gutemine 005725.0734268143000] X400-Content-Type: P2-1984 (2) From: Ulrich Hoffmann Message-Id: <9304081122.AA20169@minnie.informatik.uni-kiel.dbp.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: IN-PACKAGE changes PACKAGE-USE-LIST inappropriate I think I discovered a problem with IN-PACKAGE. Here is a clisp session protocoll which shows the problem: % clisp ... ------+------ Copyright (c) Bruno Haible, Michael Stoll 1992, 1993 > (make-package 'a) ; first create two new packages # > (make-package 'b) # > (in-package 'user :use '(lisp a)) ; set package-use-list of USER # > (package-use-list 'user) (# #) ; hmm looks ok > (in-package 'user :use '(lisp b)) ; extend use-list with package B # > (package-use-list 'user) ; check if this was succesful (# #) ; oh no! Package A is missing Using USE-PACKAGE instead of the :USE keyword gives the desired result. Any ideas? Greetins, Ulrich I used the SPARC/SUNOS.4.1 Version. Ulrich Hoffmann email: uho@informatik.uni-kiel.dbp.de Institut fuer Informatik, Universitaet Kiel Tel: +49 431 560426 Preusserstr 1-9, D-2300 Kiel 1, Germany Fax: +49 431 566143 From manager@csdeca.cs.missouri.edu Thu Apr 8 20:03:27 1993 Return-Path: Received: from csdeca.cs.missouri.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13708; Thu, 8 Apr 93 20:03:27 +0200 Received: by csdeca.cs.missouri.edu (5.57/Ultrix4.1-C) id AA04118; Thu, 8 Apr 93 13:02:29 -0500 Date: Thu, 8 Apr 93 13:02:29 -0500 From: manager@csdeca.cs.missouri.edu (Benlu Jiang) Message-Id: <9304081802.AA04118@csdeca.cs.missouri.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Clisp I tried to install clisp in DECstation 5000 (RISC, ultrix) But when I build the makefile by makemake mips gcc > makefile I got following error message: makemake: syntax error at line 584: `(' unexpected Could you give me some help? Thanks BJ From haible Sun Apr 11 19:24:51 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA14730; Sun, 11 Apr 93 19:24:51 +0200 Date: Sun, 11 Apr 93 19:24:51 +0200 From: haible (Bruno Haible) Message-Id: <9304111724.AA14730@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: IN-PACKAGE changes PACKAGE-USE-LIST Ulrich Hoffmann has a problem with IN-PACKAGE: > > (package-use-list 'user) > (# # ; hmm looks ok > > (in-package 'user :use '(lisp b)) ; extend use-list with package B > # > > (package-use-list 'user) ; check if this was succesful > (# #) ; oh no! Package A is missing CLtL states that the "existing package is augmented to reflect any new nicknames or new packages in the :USE list". CLISP also removes any packages from the use list that are not specified in the :USE list argument. (Seems more logical to me: The code after the IN-PACKAGE form may be disturbed by an unforeseen package on the use list.) If you have IN-PACKAGE calls for the same package but with different :USE arguments, you should make them equal. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Sun Apr 11 19:35:23 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA14828; Sun, 11 Apr 93 19:35:23 +0200 Date: Sun, 11 Apr 93 19:35:23 +0200 From: haible (Bruno Haible) Message-Id: <9304111735.AA14828@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Clisp on RISC/ultrix > makemake mips gcc > makefile > makemake: syntax error at line 584: `(' unexpected You have a shell /bin/sh that apparently doesn't support function definitions. Get another shell (bash or ksh). Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From jbm@hal.com Fri Apr 16 16:24:08 1993 Return-Path: Received: from hal.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19071; Fri, 16 Apr 93 16:24:08 +0200 Received: from infinity.hal.com by hal.com (4.1/SMI-4.1.1) id AA25825; Fri, 16 Apr 93 07:23:39 PDT Received: by infinity.hal.com (4.1/SMI-4.1.2) id AA09052; Fri, 16 Apr 93 09:23:37 CDT Date: Fri, 16 Apr 93 09:23:37 CDT Message-Id: <9304161423.AA09052@infinity.hal.com> From: jbm@hal.com (Brad Might) Sender: jbm@hal.com To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: bug ;;; dynamic binding of *special* is messed up when ;;; creating structure under (with-open...) (defvar *special* nil) (defstruct base one) (defstruct (derived (:include base) (:print-function print-two)) two) (defun print-two (node stream depth) (princ *special*) (princ " here")) (setq xx (with-open-file (foo "/dev/null") (make-derived))) (setq yy (with-open-file (foo (open "/dev/null"))(make-derived))) ;;; this function should print "T here" but prints "NIL here" (defun funx () (let ((*special* t)) xx)) (defun funy () (let ((*special* t)) yy)) From haible Fri Apr 16 22:47:09 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19726; Fri, 16 Apr 93 22:47:09 +0200 Date: Fri, 16 Apr 93 22:47:09 +0200 From: haible (Bruno Haible) Message-Id: <9304162047.AA19726@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: bug Brad Might thinks: > ;;; this function should print "T here" but prints "NIL here" > (defun funx () (let ((*special* t)) xx)) No, this function shouldn't print anything, and it prints nothing, The value of xx is printed by the read-eval-print loop. When this occurs the function funx has already been left, and the dynamic bindings that were in effect have been undone. If you really want xx to be printed at your program's run time, then do (defun funx () (let ((*special* t)) (print xx))) Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From jbm@hal.com Fri Apr 16 22:55:36 1993 Return-Path: Received: from hal.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19895; Fri, 16 Apr 93 22:55:36 +0200 Received: from infinity.hal.com by hal.com (4.1/SMI-4.1.1) id AA02431; Fri, 16 Apr 93 13:55:02 PDT Received: by infinity.hal.com (4.1/SMI-4.1.2) id AA00234; Fri, 16 Apr 93 15:55:00 CDT Date: Fri, 16 Apr 93 15:55:00 CDT From: jbm@hal.com (Brad Might) Message-Id: <9304162055.AA00234@infinity.hal.com> To: clisp-list Subject: Re: bug In-Reply-To: <9304162047.AA19726@ma2s2.mathematik.uni-karlsruhe.de> References: <9304162047.AA19726@ma2s2.mathematik.uni-karlsruhe.de> Thanks, I see where I went wrong ! Bruno Haible writes: > Brad Might thinks: > > ;;; this function should print "T here" but prints "NIL here" > > (defun funx () (let ((*special* t)) xx)) > > No, this function shouldn't print anything, and it prints nothing, > The value of xx is printed by the read-eval-print loop. When this occurs > the function funx has already been left, and the dynamic bindings that > were in effect have been undone. > > If you really want xx to be printed at your program's run time, then do > > (defun funx () (let ((*special* t)) (print xx))) > > > Bruno Haible > haible@ma2s2.mathematik.uni-karlsruhe.de > From jbm@hal.com Fri Apr 16 22:56:42 1993 Return-Path: Received: from hal.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19908; Fri, 16 Apr 93 22:56:42 +0200 Received: from infinity.hal.com by hal.com (4.1/SMI-4.1.1) id AA02442; Fri, 16 Apr 93 13:56:11 PDT Received: by infinity.hal.com (4.1/SMI-4.1.2) id AA00237; Fri, 16 Apr 93 15:56:09 CDT Date: Fri, 16 Apr 93 15:56:09 CDT From: jbm@hal.com (Brad Might) Message-Id: <9304162056.AA00237@infinity.hal.com> To: clisp-list Subject: Re: bug In-Reply-To: <9304162047.AA19726@ma2s2.mathematik.uni-karlsruhe.de> References: <9304162047.AA19726@ma2s2.mathematik.uni-karlsruhe.de> Bruno Haible writes: > Brad Might thinks: some stuff. ps. I think clisp is very excellent. Is there any internals documentation (in english ?). I am very interested in this stuff. How can I add some functions written in C ? thanks, brad From @cdc2-gw.rrzn.uni-hannover.de,@CDC2.RRZN.UNI-HANNOVER.DE:bergmann@tnt.uni-hannover.de Wed Apr 21 12:29:47 1993 Return-Path: <@cdc2-gw.rrzn.uni-hannover.de,@CDC2.RRZN.UNI-HANNOVER.DE:bergmann@tnt.uni-hannover.de> Received: from cdc2.rrzn.uni-hannover.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05731; Wed, 21 Apr 93 12:29:47 +0200 Received: from DE*DBP*UNI-HANNOVER by cdc2-gw.rrzn.uni-hannover.de via QTFS with X.400; Wed, 21 Apr 93 12:21:16 +0200 Received: from helios.tnt.uni-hannover.de by CDC2.RRZN.UNI-HANNOVER.DE ; 21 Apr 93 12:21:16 Received: from artemis.tnt.uni-hannover.de by helios.tnt.uni-hannover.de (4.1/SMI-4.1) id AA05049; Wed, 21 Apr 93 12:21:32 +0200 Date: Wed, 21 Apr 93 12:21:32 +0200 From: bergmann@tnt.uni-hannover.de (Bernd Bergmann) Message-Id: <9304211021.AA05049@helios.tnt.uni-hannover.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: XIT compilation problems After building PCL, CLX and CLUE i tried to compile XIT from the file ma2s2.mathematik.uni-karlsruhe.de/pub/lisp/clisp/packages/clue+clio+xit.clisp.tar.Z, but the compilation broke with the following error message: ******************************************************************************* .... Compilation of file /new/sources/xit/kernel/clx-auth.lsp is finished. 0 errors, 0 warnings ;; Loading file /new/sources/xit/kernel/clx-auth.fas ... ;; Loading of file /new/sources/xit/kernel/clx-auth.fas is finished. ; COMPILE module MACROS ... Compiling file /new/sources/xit/kernel/macros.lsp ... WARNING in function ADD-OPEN-DISPLAY-HOOK : variable IF-NEEDED is used despite of IGNORE declaration. WARNING in function ADD-CLOSE-DISPLAY-HOOK : variable IF-NEEDED is used despite of IGNORE declaration. WARNING in function ADD-OPEN-TOPLEVEL-HOOK : variable IF-NEEDED is used despite of IGNORE declaration. WARNING in function #:TOP-LEVEL-FORM-38-1-1 : variable ACTIONS is not used. Misspelled or missing IGNORE declaration? WARNING in function #:TOP-LEVEL-FORM-38-2 : variable ACTIONS is not used. Misspelled or missing IGNORE declaration? *** - SYSTEM::%STRUCTURE-REF: NIL is not a structure of type SYSTEM::ANODE 1. Break> ******************************************************************************* ; XIT -- The X User Interface Toolkit ; ; Copyright (c) 1992: Research Group DRUID ; Department of Computer Science ; University of Stuttgart ; Breitwiesenstr. 20-22 ; W-7000 Stuttgart 80 ; Germany ; ; Juergen Herczeg ; Hubertus Hohl ; Matthias Ressel ; Phone: (+49|0) 711 7816 {364,359} ; E-mail: druid@informatik.uni-stuttgart.de My system: Linux pl998 CLISP "March 1993" PCL "July 1b" CLX from pcl+clx.sept92f.clisp.tar.Z Any help will be appreciated. Please email me the answer or suggestion. Thanks in advance, Bernd Bergmann email: bergmann@tnt.uni-hannover.de From @serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de Thu Apr 22 16:14:04 1993 Return-Path: <@serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de> Received: from uni-kl.de (stepsun.uni-kl.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07406; Thu, 22 Apr 93 16:14:04 +0200 Received: from serv-200.dfki.uni-kl.de by stepsun.uni-kl.de id aa08637; 22 Apr 93 16:07 MET DST Received: from serv-300.dfki.uni-kl.de by serv-200.dfki.uni-kl.de id aa13589; 22 Apr 93 14:09 GMT Date: Thu, 22 Apr 93 14:09:43 GMT From: Bjoern Hoefling To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: hoefling@dfki.uni-kl.de Subject: problem with defstruct Organization: DFKI, University of Kaiserslautern, Germany Message-Id: <9304221409.aa11329@serv-300.dfki.uni-kl.de> A friend of me has a problem with clisp and asked me to post it in the clisp mailing list. Here is a short script: ________________________________________________________________________________ > (defstruct (object (:conc-name object-) (:predicate object?) ) (reference 'NUNA) ) OBJECT > (setq a (make-object)) #S(OBJECT :REFERENCE NUNA) > (object-reference a) NUNA > (defstruct (unknown (:predicate unknown?) (:conc-name object-) (:include object) ) (abstract-functions nil :type list) ) UNKNOWN > (object-reference a) *** - SYSTEM::%STRUCTURE-REF: #S(OBJECT :REFERENCE NUNA) ist keine Structure vom Typ UNKNOWN. 1. Break> ________________________________________________________________________________ Thanks for any help Bjoern Hoefling (hoefling@dfki.uni-kl.de) From haible Fri Apr 23 09:54:13 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08040; Fri, 23 Apr 93 09:54:13 +0200 Date: Fri, 23 Apr 93 09:54:13 +0200 From: haible (Bruno Haible) Message-Id: <9304230754.AA08040@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list, haible@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: problem with defstruct Bjoern Hoefling points out a problem that occurs when using DEFSTRUCT with :INCLUDE option and the same :CONC-NAME for both structures: | > (defstruct (object (:conc-name object-) (:predicate object?)) | (reference 'NUNA) | ) | OBJECT | > (setq a (make-object)) | #S(OBJECT :REFERENCE NUNA) | > (object-reference a) | NUNA | > (defstruct (unknown (:predicate unknown?) (:conc-name object-) | (:include object) | ) | (abstract-functions nil :type list) | ) | UNKNOWN | > (object-reference a) | | *** - SYSTEM::%STRUCTURE-REF: #S(OBJECT :REFERENCE NUNA) ist keine | Structure vom Typ UNKNOWN. | 1. Break> By executing the second DEFSTRUCT form, you have in fact redefined the function OBJECT-REFERENCE. The first one accepted all structures of type OBJECT, the second one only those which have the additional slot ABSTRACT-FUNCTIONS. Your code is not covered by CLtL, and it breaks under both CLISP and CMUlisp. Correct it. Either a) Use another CONC-NAME for the including strcutures, or b) Repeat the first DEFSTRUCT form after the second. If you have a tree of including structures, you must repeat all the non-leaves in an leaf-to-root order. For example, change (defstruct a ...) (defstruct (b (:include a)) ...) (defstruct (c (:include b)) ...) (defstruct (d (:include c)) ...) (defstruct (e (:include b)) ...) to (defstruct a ...) (defstruct (b (:include a)) ...) (defstruct (c (:include b)) ...) (defstruct (d (:include c)) ...) (defstruct (e (:include b)) ...) (defstruct (c (:include b)) ...) (defstruct (b (:include a)) ...) (defstruct a ...) Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From pdudey@Willamette.EDU Tue Apr 27 09:28:35 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA11819; Tue, 27 Apr 93 09:28:35 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA13735; Tue, 27 Apr 93 00:23:34 PDT Date: Tue, 27 Apr 1993 00:21:16 -0700 (PDT) From: Peter Dudey Subject: What neat debugging tools don't I know about? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Is there a way to single-step through the evaluation of a form? I've got this problem wherein a variable I try to access inside of a let block (which defines it) claims to "have no value". Any clues? Thanks. # Peter Dudey, 11 kyu, Lisp SubGuru, Order of the Golden Parentheses \FINGER # # The Emergent Church -- "Tenets? You'll have to ask us as a group..." \ME # # If a megacorp added $3 to everybody's bill for 'fees', who'd complain? # # Please mail me plastic spaceships: 900 State St. C-210, Salem, OR 97301 # From haible Tue Apr 27 12:16:09 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12051; Tue, 27 Apr 93 12:16:09 +0200 Date: Tue, 27 Apr 93 12:16:09 +0200 From: haible (Bruno Haible) Message-Id: <9304271016.AA12051@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: What neat debugging tools don't I know about? > Is there a way to single-step through the evaluation of a form? Use (STEP form) and try the Help command/key when at the "Step 1>" prompt. > I've got this problem wherein a variable I try to access inside of a let > block (which defines it) claims to "have no value". Any clues? You can use COMPILE-FILE. The compiler is pretty good in detecting such scoping problems. Are you aware of the difference between LET and LET* ? Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From pdudey@Willamette.EDU Tue Apr 27 18:49:01 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12794; Tue, 27 Apr 93 18:49:01 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA01184; Tue, 27 Apr 93 09:44:02 PDT Date: Tue, 27 Apr 1993 09:43:08 -0700 (PDT) From: Peter Dudey Subject: Re: What neat debugging tools don't I know about? To: clisp-list In-Reply-To: <9304271016.AA12051@ma2s2.mathematik.uni-karlsruhe.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 27 Apr 1993, Bruno Haible wrote: > Are you aware of the difference between LET and LET* ? Just parallel vs. serial assignment, right? # Peter Dudey, 11 kyu, Lisp SubGuru, Order of the Golden Parentheses \FINGER # # The Emergent Church -- "Tenets? You'll have to ask us as a group..." \ME # # If a megacorp added $3 to everybody's bill for 'fees', who'd complain? # # Please mail me plastic spaceships: 900 State St. C-210, Salem, OR 97301 # From pdudey@Willamette.EDU Wed Apr 28 04:05:22 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13271; Wed, 28 Apr 93 04:05:22 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA18226; Tue, 27 Apr 93 19:00:21 PDT Date: Tue, 27 Apr 1993 18:58:07 -0700 (PDT) From: Peter Dudey Subject: What's wrong here? To: clisp-list In-Reply-To: <9304271016.AA12051@ma2s2.mathematik.uni-karlsruhe.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Given these two definitions: (defstruct flugblog (color 'blue) (size 'huge)) (defun foo () ;; ;; Returns the color of a newly-constructed, temporary flugblog ;; (let ((x (make-flugblog))) (eval '(flugblog-color x)))) CLISP chokes on (foo). It complains that x has no value. I've stepped through this problem, and it seems to have something to do with the call to EVAL. Any clues? # Peter Dudey, 11 kyu, Lisp SubGuru, Order of the Golden Parentheses \FINGER # # The Emergent Church -- "Tenets? You'll have to ask us as a group..." \ME # # If a megacorp added $3 to everybody's bill for 'fees', who'd complain? # # Please mail me plastic spaceships: 900 State St. C-210, Salem, OR 97301 # From bkn@ida.liu.se Wed Apr 28 09:15:22 1993 Return-Path: Received: from ida.liu.se (curofix.ida.liu.se) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13455; Wed, 28 Apr 93 09:15:22 +0200 Received: from panix by ida.liu.se (5.65b/ida.minimaster-V1.0b6d5) id AA07806; Wed, 28 Apr 93 09:09:53 +0200 From: Bernt Nilsson Received: from pan16 by panix (5.65b/ida.slave-V1.0b3) id AA05553; Wed, 28 Apr 93 09:09:51 +0200 Received: by pan16 (5.65b/ida.slave-V1.0b3) id AA07623; Wed, 28 Apr 93 09:09:49 +0200 Date: Wed, 28 Apr 93 09:09:49 +0200 Message-Id: <9304280709.AA07623@pan16> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: What's wrong here? > Given these two definitions: > > (defstruct flugblog > (color 'blue) > (size 'huge)) > > (defun foo () > ;; > ;; Returns the color of a newly-constructed, temporary flugblog > ;; > (let ((x (make-flugblog))) > (eval '(flugblog-color x)))) > > CLISP chokes on (foo). It complains that x has no value. I've stepped > through this problem, and it seems to have something to do with the call > to EVAL. Any clues? Yes, "x" is not declared special. Quoting CLtL 2: "eval form [Function] The form is evaluated in the current dynamic environment and a null lexical environment." - You can eigther use a local or a global special declaration: Modify foo; (let ((x (make-flugblog))) (declare (special x)) (eval '(flugblog-color x)))) Or evaluate "(defvar x)" before running (or compiling) foo. -- Bernt ---------------------------------------------------------------------- Bernt Nilsson, Systems Manager | Internet: BKN@IDA.LIU.SE IDA, LiTH (CS Dept. Linkoping U) | uucp: ...!sunic!liuida!bkn S-581 83 Linkoping, Sweden | Bitnet: BKN@SELIUIDA Phone +46 13 281975 From haible Wed Apr 28 12:06:53 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13852; Wed, 28 Apr 93 12:06:53 +0200 Date: Wed, 28 Apr 93 12:06:53 +0200 From: haible (Bruno Haible) Message-Id: <9304281006.AA13852@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: What's wrong here? > (defun foo () > (let ((x (make-flugblog))) > (eval '(flugblog-color x)))) Every time you use EVAL you must be doing something wrong. Simply write (defun foo () (let ((x (make-flugblog))) (flugblog-color x) ) ) Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From syali@cs.Buffalo.EDU Wed Apr 28 17:30:01 1993 Return-Path: Received: from sybil.cs.Buffalo.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA14382; Wed, 28 Apr 93 17:30:01 +0200 Received: by sybil.cs.Buffalo.EDU (4.0/1.01) id AA06976; Wed, 28 Apr 93 11:23:49 EDT Date: Wed, 28 Apr 93 11:23:49 EDT From: syali@cs.Buffalo.EDU (Sy Ali) Message-Id: <9304281523.AA06976@sybil.cs.Buffalo.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: CLISP on a 486DLC-40 (DRDOS6). I've been unable to get CLISP to run on my Cyrix-based machine (the only instance of incompatibility I've yet seen). CLISP load, display the welcome header and prompt, but any keypress results in an immediate exit from the program (if my memory manager is not loaded I get a "Bye.", o/w it hangs with a stack error). I suspect this is an incompatibility in the memory manager. As I very much like CLISP (I trivally ported a 10,000+ line CL program to it, Allegro CL\PC 1.0 was much more painful), I'd like to get a version that works on my machine. Am I outta luck or is there a fix? BTW, this was all under DOS (DRDOS6, but I've duplicated the problem with MSDOS too). My next move is to install Linux :-) Thanks, Sy. From haible Thu Apr 29 00:01:02 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA14673; Thu, 29 Apr 93 00:01:02 +0200 Date: Thu, 29 Apr 93 00:01:02 +0200 From: haible (Bruno Haible) Message-Id: <9304282201.AA14673@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: CLISP on a 486DLC-40 > I've been unable to get CLISP to run on my Cyrix-based machine (the only > instance of incompatibility I've yet seen). The problem you have has already been reported by someone else and analyzed by Eberhard Mattes. When he releases EMX 0.8g, I will build a new DOS release of CLISP, and I hope that it will fix this incompatibility. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From lmulcahy@nyx.cs.du.edu Thu Apr 29 16:07:35 1993 Return-Path: Received: from relay1.UU.NET by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15514; Thu, 29 Apr 93 16:07:35 +0200 Received: from nyx.cs.du.edu by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA12811; Thu, 29 Apr 93 10:01:37 -0400 Received: by nyx.cs.du.edu (4.1/SMI-4.1) id AA20639; Thu, 29 Apr 93 08:01:43 MDT Date: Thu, 29 Apr 93 08:01:43 MDT From: lmulcahy@nyx.cs.du.edu (Larry Mulcahy) Message-Id: <9304291401.AA20639@nyx.cs.du.edu> X-Disclaimer: Nyx is a public access Unix system run by the University of Denver. The University has neither control over nor responsibility for the opinions or correct identity of users. To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: FUNCTION-LAMBDA-EXPRESSION in CLISP? Is there a way to get the function lambda expression? Symbol-function returns a # data type. Is this a defstruct? If so, I'd be satisfied with the slot names. From haible Thu Apr 29 20:02:32 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15755; Thu, 29 Apr 93 20:02:32 +0200 Date: Thu, 29 Apr 93 20:02:32 +0200 From: haible (Bruno Haible) Message-Id: <9304291802.AA15755@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: function-lambda-expression Larry Mulcahy and Joerg Hoehle ask: > Is there a way to get the function lambda expression? > Symbol-function returns a # data type. Is this a defstruct? > If so, I'd be satisfied with the slot names. It is a built-in type. The slots have indices, no names. Here is an implementation of FUNCTION-LAMBDA-EXPRESSION, see CLtL2 p.682. (defun function-lambda-expression (obj) (unless (sys::closurep obj) (error #+DEUTSCH "~: ~ ist keine Funktion." #+ENGLISH "~: ~ is not a function" 'function-lambda-expression obj ) ) (if (not (compiled-function-p obj)) (values (cons 'LAMBDA (sys::%record-ref obj 1)) (vector (sys::%record-ref obj 4) (sys::%record-ref obj 5) (sys::%record-ref obj 6) (sys::%record-ref obj 7) (sys::%record-ref obj 8) ) (sys::%record-ref obj 0) ) (values nil t nil) ) ) Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From pdudey@Willamette.EDU Thu Apr 29 22:25:12 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15883; Thu, 29 Apr 93 22:25:12 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA13661; Thu, 29 Apr 93 13:19:46 PDT Date: Thu, 29 Apr 1993 13:17:30 -0700 (PDT) From: Peter Dudey Subject: A couple of I/O questions To: Clisp mailing list Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII 1) How do I turn the string "13" into the fixnum 13? 2) Is there any way to move the cursor around the screen, so I might build what amounts to an editor? (Actually, I'm writing a program to play the oriental game of Go, and I'd like the user to be able to move the cursor around the board and select a move, rather than typing in coordinates.) Thanks. # Peter Dudey, 11 kyu, Lisp SubGuru, Order of the Golden Parentheses \FINGER # # Church of "Bob's" Sock-Garters / Antartican People's Liberation Front \ME # # "A shadowy flight into the dangerous world of a man who does not exist." # # Please mail me plastic spaceships: 900 State St. C-210, Salem, OR 97301 # From pdudey@Willamette.EDU Fri Apr 30 05:24:40 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16384; Fri, 30 Apr 93 05:24:40 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA10432; Thu, 29 Apr 93 20:19:15 PDT Date: Thu, 29 Apr 1993 20:17:39 -0700 (PDT) From: Peter Dudey Subject: Even yet still more questions To: Clisp mailing list Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Is there a way to "clear" the interpreter, i.e., remove all user definitions and declarations, without exiting and reloading? When I build some horrid, arcane pointer structure, the various items in it are subject to garbage collection when and only when there's nothing pointing to them . . . right? # Peter Dudey, 11 kyu, Lisp SubGuru, Order of the Golden Parentheses \FINGER # # Church of "Bob's" Sock-Garters / Antartican People's Liberation Front \ME # # "A shadowy flight into the dangerous world of a man who does not exist." # # Please mail me plastic spaceships: 900 State St. C-210, Salem, OR 97301 # From haible Fri Apr 30 11:10:44 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16653; Fri, 30 Apr 93 11:10:44 +0200 Date: Fri, 30 Apr 93 11:10:44 +0200 From: haible (Bruno Haible) Message-Id: <9304300910.AA16653@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list, haible@ma2s2.mathematik.uni-karlsruhe.de Subject: A couple of I/O questions Peter Dudey asks: > How do I turn the string "13" into the fixnum 13? Use READ-FROM-STRING or PARSE-INTEGER. > Is there any way to move the cursor around the screen ... ? CLISP contains a fairly portable mini-curses like screen access package. Here is a documentation of it. (It mentions windows, but only one window is possible now: the full (virtual) screen.) 99.2. Random Screen Access -------------------------- (SYSTEM::MAKE-WINDOW) returns a "window stream". As long as this stream is open, the terminal is in cbreak/noecho mode. *TERMINAL-IO* shouldn't be used for input or output during this time. (Use WITH-KEYBOARD and *KEYBOARD-INPUT* instead.) (SYSTEM::WINDOW-SIZE window-stream) returns the window's size, as two values: height (= Ymax+1) and width (= Xmax+1). (SYSTEM::WINDOW-CURSOR-POSITION window-stream) returns the position of the cursor in the window, as two values: line (>=0, <=Ymax, 0 means top), column (>=0, <=Xmax, 0 means left margin). (SYSTEM::SET-WINDOW-CURSOR-POSITION window-stream line column) sets the position of the cursor in the window. (SYSTEM::CLEAR-WINDOW window-stream) clears the window's contents and puts the cursor in the upper left corner. (SYSTEM::CLEAR-WINDOW-TO-EOT window-stream) clears the window's contents from the cursor position to the end of window. (SYSTEM::CLEAR-WINDOW-TO-EOL window-stream) clears the window's contents from the cursor position to the end of line. (SYSTEM::DELETE-WINDOW-LINE window-stream) removes the cursor's line, moves the lines below it up by one line and clears the window's last line. (SYSTEM::INSERT-WINDOW-LINE window-stream) inserts a line at the cursor's line, moving the lines below it down by one line. (SYSTEM::HIGHLIGHT-ON window-stream) switches highlighted output on. (SYSTEM::HIGHLIGHT-OFF window-stream) switches highlighted output off. (SYSTEM::WINDOW-CURSOR-ON window-stream) makes the cursor visible, a cursor block in most implementations. (SYSTEM::WINDOW-CURSOR-OFF window-stream) makes the cursor invisible, in implementations where this is possible. I suggest a macro like (defmacro with-window (&body body) `(LET ((*WINDOW* (SYS::MAKE-WINDOW))) (UNWIND-PROTECT (PROGN ,@body) (CLOSE *WINDOW*)) ) ) to ensure that the window stream is closed in all cases. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Fri Apr 30 11:11:14 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16660; Fri, 30 Apr 93 11:11:14 +0200 Date: Fri, 30 Apr 93 11:11:14 +0200 From: haible (Bruno Haible) Message-Id: <9304300911.AA16660@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list, haible@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: garbage collection Peter Dudey asks: > Is there a way to "clear" the interpreter, i.e., remove all user > definitions and declarations, without exiting and reloading? Assuming you haven't modified system data structures or created new packages, this one might help: (let ((p (find-package "USER"))) (do-symbols (s p) (when (eq (symbol-package s) p) (unintern s p))) ) > When I build some horrid, arcane pointer structure, the various items in > it are subject to garbage collection when and only when there's nothing > pointing to them . . . right? Right. In the simplest case you set a global variable to NIL. It's even simpler if your database is not contained in a global variable at all. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From pdudey@Willamette.EDU Sat May 1 21:33:13 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17698; Sat, 1 May 93 21:33:13 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA00828; Sat, 1 May 93 12:27:29 PDT Date: Sat, 1 May 1993 12:22:48 -0700 (PDT) From: Peter Dudey Subject: Something out for which to watch To: Clisp mailing list Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII If you use a function like this: (defun foo (x) (delete x '(a b c d e f))) the quoted list '(a b c d e f) is NOT immune to the delete! Running the function could actually change it! I'm not sure if this is a bug in CLISP or not--I was under the impression that a quoted, explicit list like that was a constant, and would be "re-created" every time the function was run. Anyway, the problem can of course be solved by using "remove" instead. I would never have been able to find this bug in my program without STEP, which revealed that my list had somehow changed... ! Peter Dudey, 11 kyu, Lisp SubGuru, Order of the Golden Parentheses \FINGER ! ! Reformed Church of James "Eric" the Half-a-Bee, Dipped in Curry \ME ! ! "A shadowy flight into the dangerous world of a man who does not exist." ! ! Please mail me plastic spaceships: 900 State St. C-210, Salem, OR 97301 ! From haible Sun May 2 15:36:54 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18048; Sun, 2 May 93 15:36:54 +0200 Date: Sun, 2 May 93 15:36:54 +0200 From: haible (Bruno Haible) Message-Id: <9305021336.AA18048@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list, haible@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Something out for which to watch [This message is from David Loewenstern .] .. On Sat, 1 May 93 21:33:45 +0200, Peter Dudey said: } If you use a function like this: } (defun foo (x) (delete x '(a b c d e f))) } the quoted list '(a b c d e f) is NOT immune to the delete! Running the } function could actually change it! Yes. I'm fairly certain that CLtL2 makes this trap specific. } I'm not sure if this is a bug in CLISP or not--I was under the impression } that a quoted, explicit list like that was a constant, and would be } "re-created" every time the function was run. It isn't. I know Allegro and Symbolics CL show the same behavior. It would be pretty space-inefficient to have the compiler force every constant list (or array, or structure, I presume?) to be recopied every time it was used. I shudder to think that every construct that looks like (CASE x ((a b c d e f g h i) (foo x)) ((j k l) (bar x))) really disguises a huge amount of consing, since typically CASE is implemented as expanding into MEMBER tests: (LET ((#:G0001 x)) (COND ((MEMBER #:G0001 '(a b c d e f g h i)) (foo x)) ((MEMBER #:G0001 '(j k l))) (bar x))) --- (WITH-STANDARD-DISCLAIMERS (MAKE-INSTANCE 'Signature :NAME "David Loewenstern" :LOCATIONS '((INTERNET "davel@homxc.att.com") (INTERNET "loewenst@paul.rutgers.edu") (AT&T "201-386-6516")) :AFFILIATIONS '("AT&T Bell Labs" "Rutgers, Dept of CS") :SILLY-QUOTE (UNIX:SHELL "/usr/bin/fortune")))) From haible Sun May 2 15:37:37 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18077; Sun, 2 May 93 15:37:37 +0200 Date: Sun, 2 May 93 15:37:37 +0200 From: haible (Bruno Haible) Message-Id: <9305021337.AA18077@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list, haible@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Something out for which to watch Peter Dudey notes: > If you use a function like this: > > (defun foo (x) (delete x '(a b c d e f))) > > the quoted list '(a b c d e f) is NOT immune to the delete! Running the > function could actually change it! Yes, this a trap, even for experienced Lisp programmers. Don't use quoted lists in contexts where you also use destructive operations. Note that using '() for the empty list is always safe. > I'm not sure if this is a bug in CLISP or not--I was under the impression > that a quoted, explicit list like that was a constant, and would be > "re-created" every time the function was run. No, you always refer to the same constant. CLtL2 p. 105 cites X3J13 vote <147>: "EVAL and COMPILE are not permitted either to copy or to coalesce constants appearing in the code they process." David Loewenstern writes: > I'm fairly certain that CLtL2 makes this trap specific. Yes, X3J13 vote <36> (CLtL2 p. 70) states: "It is an error to destructively modify any object that appears as a constant in executable code, whether as a self-evaluating form or within a QUOTE special form." This means that it is the programmer's responsibility. Lisp implementations can detect such situations by putting such objects into read-only memory (on operating systems where this is possible, e.g. Unix). But I didn't implement this because it would disturb beginners when they are playing around with the interpreter: simple things like (NREVERSE '(1 2 3)) would signal an error. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From pdudey@Willamette.EDU Mon May 3 06:30:18 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19103; Mon, 3 May 93 06:30:18 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA11560; Sun, 2 May 93 21:24:20 PDT Date: Sun, 2 May 1993 21:22:47 -0700 (PDT) From: Peter Dudey Subject: The sun4-sunos4 version To: Clisp mailing list Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I've loaded it, and there are two things I don't understand: 1) The README says you no longer need a memory file (?), but you do. 2) Whenever I (exit) from CLISP, it knocks me out of my telnet link. Clues, gurus? # Peter Dudey, 11 kyu, Lisp SubGuru and Genetic Programmer Extraordinaire # # Good stuff to read: "Sunshine, Genius & Rust", in this month's _Analog_ # # "A shadowy flight into the dangerous world of a man who does not exist." # # Please mail me plastic spaceships: 257 NE 13th Street, Salem, OR 97301 # From pdudey@Willamette.EDU Mon May 3 06:38:49 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19147; Mon, 3 May 93 06:38:49 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA12042; Sun, 2 May 93 21:32:57 PDT Date: Sun, 2 May 1993 21:30:05 -0700 (PDT) From: Peter Dudey Subject: Two more things... To: Clisp mailing list Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII 1) Wow, what great compatibility! My code ported over from the DOS version, and I didn't have to change a thing! 2) One difference, though: (read-char) seems to want a return now. Is there any way to read a single keystroke? Since I'm on the topic, does anybody know off the top of their heads if there's a function that will check if a key has been pressed? Thanks, and sorry if I'm cluttering anybody's mailboxes. :) # Peter Dudey, 11 kyu, Lisp SubGuru and Genetic Programmer Extraordinaire # # Good stuff to read: "Sunshine, Genius & Rust", in this month's _Analog_ # # "A shadowy flight into the dangerous world of a man who does not exist." # # Please mail me plastic spaceships: 257 NE 13th Street, Salem, OR 97301 # From mnlrsg@cs.umu.se Mon May 3 10:32:33 1993 Received: from ume.cs.umu.se by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19319; Mon, 3 May 93 10:32:33 +0200 Received: from sandbacka.cs.umu.se by ume.cs.umu.se (5.65+bind 1.7+ida 1.4.2/91-02-01) id AA14472; Mon, 3 May 93 10:25:59 +0200, auth mnlrsg X-Auth-From: mnlrsg Received: by sandbacka.cs.umu.se (5.65+bind 1.7+ida 1.4.2/91-02-01) id AA01081; Mon, 3 May 93 10:25:54 +0200 Return-Path: Date: Mon, 3 May 93 10:25:54 +0200 From: mnlrsg@cs.umu.se Message-Id: <9305030825.AA01081@sandbacka.cs.umu.se> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Hi, folks! First, I have to report 100% (?) portability CLisp PC-version to AKCL for UNIX systems. Second: A friend of mins has some problems getting the minor memory version running on a 386sx with 1Mb RAM. When trying to run a program of more than 40 - 50 lines of code the interpreter responds something like "unable to swap". Is there a way to set the size of the swapfile? Any clues? Rolf A. Sandberg University of Umea, Sweden #No por mucho madrugar amanece mas temprano# From hoehle@post.inf-wiss.ivp.uni-konstanz.de Thu May 6 14:32:16 1993 Return-Path: Received: from inf-wiss.ivp.uni-konstanz.de (post.inf-wiss.ivp.uni-konstanz.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23390; Thu, 6 May 93 14:32:16 +0200 Received: from stetten.inf-wiss.ivp.uni-konstanz.de by inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA23826; Thu, 6 May 93 14:26:13 +0200 Date: Thu, 6 May 93 14:26:13 +0200 From: hoehle@post.inf-wiss.ivp.uni-konstanz.de (Joerg-Cyril Hoehle) Message-Id: <9305061226.AA23826@inf-wiss.ivp.uni-konstanz.de> Received: by stetten.inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA05824; Thu, 6 May 93 14:26:04 +0200 To: clisp-list@[129.13.115.2] Subject: CLISP running as an EMACS subprocess? I've heard that some people were working on some clisp.el package to make CLISP work comfortably as a Lisp-subprocess in EMACS and I'd like to know what the current status of the work is. Is it a comint or an ILISP superset? Joerg. hoehle@inf-wiss.ivp.uni-konstanz.de From hoehle@post.inf-wiss.ivp.uni-konstanz.de Thu May 6 17:55:45 1993 Return-Path: Received: from inf-wiss.ivp.uni-konstanz.de (post.inf-wiss.ivp.uni-konstanz.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23628; Thu, 6 May 93 17:55:45 +0200 Received: from stetten.inf-wiss.ivp.uni-konstanz.de by inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA24518; Thu, 6 May 93 17:49:37 +0200 Date: Thu, 6 May 93 17:49:37 +0200 From: hoehle@post.inf-wiss.ivp.uni-konstanz.de (Joerg-Cyril Hoehle) Message-Id: <9305061549.AA24518@inf-wiss.ivp.uni-konstanz.de> Received: by stetten.inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA05947; Thu, 6 May 93 17:49:28 +0200 To: clisp-list@[129.13.115.2] Cc: Dave Carrico <2fwcsnap@stat1.cc.ukans.edu> Subject: (setf (foo #) value), foo being undefined yet Preliminaries: > (defun fact (n) (setf (foo n) t)) *** - Illegal SETF place: (FOO N) 1. Break> I guess that this is a normal behaviour for CLtLI Lisps. I came accross the following code: * (macroexpand-1 '(defroutine add1-nam ((:nam e) (:nam i) n) (cond ((< i n) (setf e (+ e 1))) (t nil)))) (PROGN (DEFUN ADD1-NAM** (N) (LET ((N (EVAL N))) ; I believe that EVAL to be wrong, ; but that's another story. (COND ((< (I**) N) (SETF (E**) (+ (E**) 1))) (T NIL)) (VALUES))) (DEFMACRO ADD1-NAM (E I N) `(PROGN ,@(DEFTHUNK 'E** E) ,@(DEFTHUNK 'I** I) (MULTIPLE-VALUE-BIND NIL (ADD1-NAM** ,N)) NIL))) T * (defthunk 'e** 'foobar) ((SETF (SYMBOL-FUNCTION 'E**) #'(LAMBDA () FOOBAR)) (SETF (SYMBOL-FUNCTION 'E**-SET) #'(LAMBDA (VAL) (SETF FOOBAR VAL))) (DEFSETF E** E**-SET)) * Loading a file containing the above defroutine leads to the following error: *** - Illegal SETF place: (E**) 1. Break> where EVAL-Frame für Form (LABELS ((ADD1-NAM** (N) (BLOCK ADD1-NAM** (LET ((N (EVAL N))) (COND ((< (I**) N) (SETF (E**) (+ (E**) 1))) (T NIL)) (VALUES))))) #'ADD1-NAM**) 1. Break> The error comes from the (SETF (E**) #) call in ADD1-NAM**, the defsetf for e** would only occur when the add1-nam macro is actually called/used, not in the defroutine call. I wonder whether the above code using a yet undefined setf is correct for CLtLI or CLtLII. In CLtLII, setf, seeing an unknown (setf (f #) #) form, has been expanded to produce a call to the (setf f) function, whereas in CLtLI (setf f) was not considered as being a function name. I'm not sure whether the above code intended to make use of this extended setf behaviour and thus was written for CLtLII. When compiling the above, CMULISP (CLtLII) generates warnings about undefined functions (SETF E**), E** and I**. I'm tempted to believe that the above code is wrong w.r.t. both standards and that it should have been written in another way (perhaps with all needed defsetfs before the defun in the defroutine macro-expansion, even if the used functions are not defined at that time (is this allowed?), and maybe with some preceeding (declare (notinline E** E**-SET))). Thanks for your help, Joerg. hoehle@inf-wiss.ivp.uni-konstanz.de From haible Sun May 9 22:07:36 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA25394; Sun, 9 May 93 22:07:36 +0200 Date: Sun, 9 May 93 22:07:36 +0200 From: haible (Bruno Haible) Message-Id: <9305092007.AA25394@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: CLISP running as an EMACS subprocess? > I've heard that some people were working on some clisp.el package to > make CLISP work comfortably as a Lisp-subprocess in EMACS Ziad Najem's package has circulated in this mailing list and is now contained in ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/mailing-list-archive . Is any other working interface between EMACS and CLISP known? Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Sun May 9 22:08:21 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA25421; Sun, 9 May 93 22:08:21 +0200 Date: Sun, 9 May 93 22:08:21 +0200 From: haible (Bruno Haible) Message-Id: <9305092008.AA25421@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: (setf (foo #) value), foo being undefined yet Joerg Hoehle writes: > I wonder whether the above code using a yet undefined setf is correct > for CLtLI or CLtLII. In CLtLII, setf, seeing an unknown (setf (f #) #) > form, has been expanded to produce a call to the (setf f) function, > whereas in CLtLI (setf f) was not considered as being a function name. True. > I'm not sure whether the above code intended to make use of this > extended setf behaviour and thus was written for CLtLII. Apparently. > I'm tempted to believe that the above code is wrong w.r.t. both > standards and that it should have been written in another way It is wrong only w.r.t. CLTL1, but is will be a good idea to change the macroexpansion of (DEFROUTINE ...) from (PROGN (DEFUN ADD1-NAM** (N) (LET ((N (EVAL N))) (COND ((< (I**) N) (SETF (E**) (+ (E**) 1))) (T NIL)) (VALUES) ) ) (DEFMACRO ADD1-NAM (E I N) `(PROGN ,@(DEFTHUNK 'E** E) ,@(DEFTHUNK 'I** I) (MULTIPLE-VALUE-BIND NIL (ADD1-NAM** ,N)) NIL ) ) ) to something like (PROGN (EVAL-WHEN (COMPILE LOAD EVAL) ,@(DEFTHUNK 'E** E) ,@(DEFTHUNK 'I** I) ) (DEFMACRO ADD1-NAM (E I N) (ONCE-ONLY (E I N) `(PROGN (COND ((< (I**) N) (SETF (E**) (+ (E**) 1))) (T NIL)) (VALUES) ) ) ) ) Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From hoehle@post.inf-wiss.ivp.uni-konstanz.de Mon May 10 12:23:37 1993 Return-Path: Received: from inf-wiss.ivp.uni-konstanz.de (post.inf-wiss.ivp.uni-konstanz.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA29029; Mon, 10 May 93 12:23:37 +0200 Received: from stetten.inf-wiss.ivp.uni-konstanz.de by inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA00695; Mon, 10 May 93 12:15:14 +0200 Date: Mon, 10 May 93 12:15:14 +0200 From: hoehle@post.inf-wiss.ivp.uni-konstanz.de (Joerg-Cyril Hoehle) Message-Id: <9305101015.AA00695@inf-wiss.ivp.uni-konstanz.de> Received: by stetten.inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA06783; Mon, 10 May 93 12:15:05 +0200 To: clisp-list@[129.13.115.2], haible@[129.13.115.2] Subject: Re: (setf (foo #) value), foo being undefined yet In-Reply-To: <9305092008.AA25421@ma2s2.mathematik.uni-karlsruhe.de> References: <9305092008.AA25421@ma2s2.mathematik.uni-karlsruhe.de> I wrote: > I'm not sure whether the above code intended to make use of this > extended setf behaviour and thus was written for CLtLII. Bruno Haible replied: > It is wrong only w.r.t. CLTL1, but is will be a good idea to change > the macroexpansion of (DEFROUTINE ...) from I still think that it's even wrong for CLtLII the way it was done, but anyway I changed it to generate: (PROGN (DEFSETF E** E**-SET) (DEFSETF I** I**-SET) ;; neither E** nor E**-SET are defined now, but they'll be when this ;; function gets called. A (declare (notinline E** E**-SET)) could be ;; in add1-nam** to make things really safe. (DEFUN ADD1-NAM** (N) (LET () (COND ((< (I**) N) (SETF (E**) (+ (E**) 1))) (T NIL)) (VALUES))) (DEFMACRO ADD1-NAM (E I N) `(PROGN ,@(DEFTHUNK 'E** E) ,@(DEFTHUNK 'I** I) (MULTIPLE-VALUE-BIND NIL (ADD1-NAM** ,N)) NIL))) and removed the defsetf from the defthunk generation: > (defthunk 'e** 'bar) ((SETF (SYMBOL-FUNCTION E**) #'(LAMBDA () BAR)) (SETF (SYMBOL-FUNCTION 'E**-SET) #'(LAMBDA (VAL) (SETF BAR VAL)))) making the setf functions known to Lisp before their use. Joerg. hoehle@inf-wiss.ivp.uni-konstanz.de From haible Tue May 11 17:29:17 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00850; Tue, 11 May 93 17:29:17 +0200 Date: Tue, 11 May 93 17:29:17 +0200 From: haible (Bruno Haible) Message-Id: <9305111529.AA00850@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: foreign functions On April 16, 1993, Brad Might asked: > Is there any internals documentation > (in english ?). I am very interested in this stuff. How can I add some > functions written in C ? Such documentation has now been written. It will be contained as src/foreign.txt in the next CLISP source release, which will be announced on this list. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From dsc3pzp@nmrdc1.nmrdc.nnmc.navy.mil Wed May 12 03:39:35 1993 Return-Path: Received: from nmrdc1.nmrdc.nnmc.navy.mil by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02170; Wed, 12 May 93 03:39:35 +0200 Received: by nmrdc1.nmrdc.nnmc.navy.mil id AA02611 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Tue, 11 May 1993 21:33:12 -0400 Date: Tue, 11 May 1993 21:33:12 -0400 From: Philip Perucci Message-Id: <199305120133.AA02611@nmrdc1.nmrdc.nnmc.navy.mil> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: CLOS/CLX on clisp? msg1.rdy From uho@informatik.uni-kiel.dbp.de Wed May 19 17:48:34 1993 Return-Path: Received: from ixgate.gmd.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15743; Wed, 19 May 93 17:48:34 +0200 X400-Received: by mta ixgate.gmd.de in /PRMD=GmD/ADMD=DbP/C=DE/; Relayed; Wed, 19 May 1993 17:37:29 +0200 X400-Received: by /PRMD=uni-kiel/ADMD=dbp/C=de/; Relayed; Wed, 19 May 1993 17:38:01 +0200 Date: Wed, 19 May 1993 17:38:01 +0200 X400-Originator: uho@informatik.uni-kiel.dbp.de X400-Recipients: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X400-Mts-Identifier: [/PRMD=uni-kiel/ADMD=dbp/C=de/;gutemine 024613.0737825881000] X400-Content-Type: P2-1984 (2) From: Ulrich Hoffmann Message-Id: <9305191537.AA27703@minnie.informatik.uni-kiel.dbp.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Problems with hash-table-setf and system-functions I've discovered a problem with setf on hash-tables with functions. While CLISP compiles the following code without errors, it cannot read back the generated .fas-file: test.lsp------------------------------------------------------------------------ (defvar *v*) (defun f() (setq *v* (make-hash-table)) (setf (gethash 'list) #'list)) -------------------------------------------------------------------------------- minnie% clisp ... | ------+------ Copyright (c) Bruno Haible, Michael Stoll 1992, 1993 > (compile-file "test") Compiling file /tmp_mnt/home/karlo/uho/clicc/src/compiler/test.lsp ... Compilation of file /tmp_mnt/home/karlo/uho/clicc/src/compiler/test.lsp is finished. 0 errors, 0 warnings T > (load "test") ;; Loading file /tmp_mnt/home/karlo/uho/clicc/src/compiler/test.fas ... *** - READ from #: objects printed as #<...> cannot be read back in 1. Break> -------------------------------------------------------------------------------- The problem seems to be, that the (function list) becomes the unprintable object # in the .fas file. Any comments? Ulrich Ulrich Hoffmann email: uho@informatik.uni-kiel.dbp.de Institut fuer Informatik, Universitaet Kiel Tel: +49 431 560426 Preusserstr 1-9, D-2300 Kiel 1, Germany Fax: +49 431 566143 From uho@informatik.uni-kiel.dbp.de Wed May 19 17:51:51 1993 Return-Path: Received: from ixgate.gmd.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15806; Wed, 19 May 93 17:51:51 +0200 X400-Received: by mta ixgate.gmd.de in /PRMD=GmD/ADMD=DbP/C=DE/; Relayed; Wed, 19 May 1993 17:41:53 +0200 X400-Received: by /PRMD=uni-kiel/ADMD=dbp/C=de/; Relayed; Wed, 19 May 1993 17:42:24 +0200 Date: Wed, 19 May 1993 17:42:24 +0200 X400-Originator: uho@informatik.uni-kiel.dbp.de X400-Recipients: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X400-Mts-Identifier: [/PRMD=uni-kiel/ADMD=dbp/C=de/;gutemine 024663.0737826144000] X400-Content-Type: P2-1984 (2) From: Ulrich Hoffmann Message-Id: <9305191541.AA27746@minnie.informatik.uni-kiel.dbp.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Problems with hash-table-setf and system-functions Shame on me - the example should look like: -------------------------------------------------------------------------------- (defvar *v*) (defun f() (setq *v* (make-hash-table)) (setf (gethash 'list *v*) #'list)) -------------------------------------------------------------------------------- Anyway, the effect is still the same. Ulrich Ulrich Hoffmann email: uho@informatik.uni-kiel.dbp.de Institut fuer Informatik, Universitaet Kiel Tel: +49 431 560426 Preusserstr 1-9, D-2300 Kiel 1, Germany Fax: +49 431 566143 From haible Mon May 24 13:17:41 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04323; Mon, 24 May 93 13:17:41 +0200 Date: Mon, 24 May 93 13:17:41 +0200 From: haible (Bruno Haible) Message-Id: <9305241117.AA04323@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: New CLISP version Once more, there is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . The list of supported platforms now include 386BSD 0.1, Coherent 386, Sun3/SunOS4 and DEC5000/Ultrix. Excerpt from the change log: User visible changes -------------------- * New macro DEFPACKAGE. * New function FUNCTION-LAMBDA-EXPRESSION. * New package SCREEN containing some functions for random screen access (Unix, DOS, OS/2, Atari, VMS versions only). * X3J13 vote <5> is implemented: The value of *APPLYHOOK* must now be a function of two arguments. * X3J13 vote <104> is implemented: COMPILE-FILE now binds *READTABLE*. * X3J13 vote <161> is implemented: SHADOW now accepts strings as well as symbols. * Fixed a bug in the compiler which caused the compiler to crash when compiling (CASE key ...) forms with constant key. * Fixed a bug in the compiler which produced unloadable .fas files when compiling forms like (SETF ... (FUNCTION LIST)). * Fixed a bogus error message in the AREF function. * The built-in editor on Unix now handles the Delete key the same way as the Backspace key or Ctrl-H. * Fixed a bug in + that occasionally caused (- x x) to return a long-float zero which was not EQL to the other zeroes of type LONG-FLOAT. * Fixed a bug: Correctly do round-to-even when adding two long-floats. * Fixed a bug: The destructive sequence functions DELETE etc. failed to produce an error when called on a sequence of type LIST and with an :END argument larger than the list's length. * Corrected the branch cuts of the complex ATAN and ATANH functions. Portability ----------- * 68000 and other versions: Fixed a bug in the low-level division algorithm. * Mips and Vax versions: Finished the low-level number crunching routines. * Added Ultrix support. * Added Coherent support. * Finished Sun3 support. Other modifications ------------------- * The file `src/defs2.lsp' contains a list of X3J13 votes and their current status in CLISP. * The file `src/foreign.txt' documents how to do function calls to foreign language routines from within CLISP. * Miscellaneous documentation updates. Enjoy! Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Mon May 24 13:27:45 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05268; Mon, 24 May 93 13:27:45 +0200 Date: Mon, 24 May 93 13:27:45 +0200 From: haible (Bruno Haible) Message-Id: <9305241127.AA05268@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Use September 92 (f) PCL and CLX My port of September 92 PCL to CLISP was incomplete: it lacked support of structure classes. This has now been fixed. Thanks to Cornelis van der Laan who pointed this out to me. I have built a new Sept92f distribution in /pub/lisp/clisp/packages/ on ma2s2.mathematik.uni-karlsruhe.de. July92 PCL or Sept92c PCL are obsolete. Use the Sept92f version. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From garmann@vax1.informatik.fh-regensburg.dbp.de Wed May 26 08:42:22 1993 Return-Path: Received: from ixgate.gmd.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08195; Wed, 26 May 93 08:42:22 +0200 X400-Received: by mta ixgate.gmd.de in /PRMD=GmD/ADMD=DbP/C=DE/; Relayed; Wed, 26 May 1993 08:30:59 +0200 X400-Received: by /PRMD=fh-regensburg/ADMD=dbp/C=de/; Relayed; Wed, 26 May 1993 09:31:02 +0200 Date: Wed, 26 May 1993 09:31:02 +0200 X400-Originator: garmann@vax1.informatik.fh-regensburg.dbp.de X400-Recipients: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X400-Mts-Identifier: [/PRMD=fh-regensburg/ADMD=dbp/C=de/;930526083102] X400-Content-Type: P2-1984 (2) Content-Identifier: 279 From: Udo Garmann Message-Id: <279*/S=garmann/OU=vax1/OU=informatik/PRMD=fh-regensburg/ADMD=dbp/C=de/@MHS> To: Subject: CLISP & ISO5 Hello, A general question: What are the differences between CLISP and the ISO5 documentation in /pub/lisp/islisp on the CLISP-Server? Can it be used as a documentation on CLISP as well? Is the ISO-Standard still under development? Thanx in advance, Udo garmann@vax1.informatik.fh-regensburg.dbp.de From haible Wed May 26 13:48:27 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08712; Wed, 26 May 93 13:48:27 +0200 Date: Wed, 26 May 93 13:48:27 +0200 From: haible (Bruno Haible) Message-Id: <9305261148.AA08712@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CLISP & ISO 8.5 Udo Garmann has discovered the existence of ISLISP 8.5: > What are the differences between CLISP and the ISO5 > documentation in /pub/lisp/islisp on the CLISP-Server? Clisp is an implementation of Common Lisp. ISLisp is a draft for the ISO Standard for the programming language Lisp. It is not Common Lisp, nor Eulisp, nor Scheme. It has some things in common with Common Lisp for historical and political reasons. > Is the ISO-Standard still under development? Yes, the ISO working group for Lisp is still polishing it. The only implementation of it I know of now is an interpreter written by Kent M. Pitman. > Can it be used as a documentation on CLISP as well? No. It happens to be on the same FTP server as CLISP because the ftp.bath.ac.uk server is rather slow when accessed from Germany. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From ae1181t@stnfor.ae.ge.com Fri May 28 02:27:58 1993 Return-Path: Received: from stnfor.ae.ge.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12617; Fri, 28 May 93 02:27:58 +0200 Received: by stnfor.ae.ge.com (5.65/DEC-Ultrix/4.3) id AA05349; Thu, 27 May 1993 20:15:32 -0400 Date: Thu, 27 May 1993 20:15:32 -0400 From: ae1181t@stnfor.ae.ge.com (Osman F Buyukisik) Message-Id: <9305280015.AA05349@stnfor.ae.ge.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: clx with hp710-hpux8.07 hangs Hi, I just built the pcl, clx using clisp. However clx tests do not work on my system. They just sit there and do not even open a window. I have hp710 with hpux8.07. The hp9000s800 stuff seem to work (at least compiles pcl and clx ). Loaded "series" and tests went up the 438 but died there. Has anyone else had similiar problems? fixes? I am trying clisp so that I can use pcl/clx to bring "weyl" algebra system up. AKCL 615 didnt finish compiling pcl! Thanks in advance. Osman From dsc3pzp@nmrdc1.nmrdc.nnmc.navy.mil Sat May 29 21:56:22 1993 Return-Path: Received: from nmrdc1.nmrdc.nnmc.navy.mil by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13979; Sat, 29 May 93 21:56:22 +0200 Received: by nmrdc1.nmrdc.nnmc.navy.mil id AA14896 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Sat, 29 May 1993 15:48:49 -0400 From: Philip Perucci Message-Id: <199305291948.AA14896@nmrdc1.nmrdc.nnmc.navy.mil> Subject: Platform = Linux? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Sat, 29 May 1993 15:48:29 -0400 (EDT) X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 372 Regarding the new version of CLISP... I was under the impression that Linux was a supported platform. Was the ommision of Linux an accident, is it know to compile as one of the other platforms, or is Linux simply not supported? CLISP has been included as "standard" with the Linux SLS distribution (the largest, easiest to install distribution) for many months now... From haible Tue Jun 1 14:57:23 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15398; Tue, 1 Jun 93 14:57:23 +0200 Date: Tue, 1 Jun 93 14:57:23 +0200 From: haible (Bruno Haible) Message-Id: <9306011257.AA15398@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Platform Linux? Philip Perucci wrote: > I was under the impression that Linux was a supported platform. It is. > Was the ommision of Linux an accident...? You are apparently referring to my words: The list of supported platforms now include 386BSD 0.1, Coherent 386, Sun3/SunOS4 and DEC5000/Ultrix. This was the list of new platforms. The full list of platforms where CLISP is known to run on is now: microcomputers: Atari ST Amiga 500-4000 DOS OS/2 Unix workstations: Linux Sun4, SunOS4 Sun386, SunOS4 Sun3, SunOS4 Sun4, SunOS5 386BSD Coherent 386 DEC5000, Ultrix HP9000/800, HP9000/700, HP-UX 8.05 Ports are being done for SGI, Irix VAX, VMS Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From dsc3pzp@nmrdc1.nmrdc.nnmc.navy.mil Tue Jun 1 17:39:38 1993 Return-Path: Received: from nmrdc1.nmrdc.nnmc.navy.mil by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16274; Tue, 1 Jun 93 17:39:38 +0200 Received: by nmrdc1.nmrdc.nnmc.navy.mil id AA04719 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Tue, 1 Jun 1993 11:32:10 -0400 From: Philip Perucci Message-Id: <199306011532.AA04719@nmrdc1.nmrdc.nnmc.navy.mil> Subject: Re: Platform Linux? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Tue, 1 Jun 1993 11:31:57 -0400 (EDT) In-Reply-To: <9306011257.AA15398@ma2s2.mathematik.uni-karlsruhe.de> from "Bruno Haible" at Jun 1, 93 02:59:12 pm X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1260 > > Philip Perucci wrote: > > > I was under the impression that Linux was a supported platform. > > It is. > > > Was the ommision of Linux an accident...? > > You are apparently referring to my words: > > The list of supported platforms now include 386BSD 0.1, Coherent 386, > Sun3/SunOS4 and DEC5000/Ultrix. > > This was the list of new platforms. The full list of platforms where CLISP > is known to run on is now: > > microcomputers: > Atari ST > Amiga 500-4000 > DOS > OS/2 > Unix workstations: > Linux > Sun4, SunOS4 > Sun386, SunOS4 > Sun3, SunOS4 > Sun4, SunOS5 > 386BSD > Coherent 386 > DEC5000, Ultrix > HP9000/800, HP9000/700, HP-UX 8.05 > > Ports are being done for > SGI, Irix > VAX, VMS > > > Bruno Haible > haible@ma2s2.mathematik.uni-karlsruhe.de > Hmmm... I compiled your latest CLISP on my Linux "SLS 1.02", with no crashes (a few warnings, of course). But "make test" died comparing files. Running CLISP gave me the sign-on screen, but running a little 2-page program crashed Linux. FIRST time I ever saw Linux crash! A gentleman from Italy who ported AKCL to Linux says you run AKCL on your Linux box!!! Odd... From baus@rz.uni-sb.de Fri Jun 4 16:54:36 1993 Return-Path: Received: from sbusol.rz.uni-sb.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03473; Fri, 4 Jun 93 16:54:36 +0200 From: "Joerg Baus" Date: Fri, 4 Jun 93 16:42:47 +0200 Message-Id: <9306041442.AA17742@sbusol.rz.uni-sb.de> Received: by sbusol.rz.uni-sb.de (5.65+/v1.0) id AA17742; Fri, 4 Jun 93 16:42:47 +0200 To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: precision of long floats Bruno Haible wrote (in impnotes.txt): "(SETF (LONG-FLOAT-DIGITS) 3322) sets the default precision of long floats to 1000 decimal digits." How does he compute this? What I need is a function (defun precision (n) ... ) to set the precision to n decimal digits. Any ideas? Greetings Joerg Joerg Baus | baus@sbusol.rz.uni-sb.de From hoehle@post.inf-wiss.ivp.uni-konstanz.de Fri Jun 4 17:24:21 1993 Return-Path: Received: from inf-wiss.ivp.uni-konstanz.de (post.inf-wiss.ivp.uni-konstanz.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03535; Fri, 4 Jun 93 17:24:21 +0200 Received: from stetten.inf-wiss.ivp.uni-konstanz.de by inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA01726; Fri, 4 Jun 93 17:10:39 +0200 Date: Fri, 4 Jun 93 17:10:39 +0200 From: hoehle@post.inf-wiss.ivp.uni-konstanz.de (Joerg-Cyril Hoehle) Message-Id: <9306041510.AA01726@inf-wiss.ivp.uni-konstanz.de> Received: by stetten.inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA19903; Fri, 4 Jun 93 17:10:31 +0200 To: Joerg Baus Cc: clisp-list@[129.13.115.2] Subject: precision of long floats In-Reply-To: <9306041442.AA17742@sbusol.rz.uni-sb.de> Joerg Baus writes: > > Bruno Haible wrote (in impnotes.txt): > "(SETF (LONG-FLOAT-DIGITS) 3322) sets the default precision of long > floats to 1000 decimal digits." > > How does he compute this? What I need is a function > > (defun precision (n) > ... ) > > to set the precision to n decimal digits. Any ideas? (log n m) tells you a multiplying factor for the number of digits a representation in base m (for example 2) would take over a representation in base n (for example 10). How about (defun precision (n) "sets default long-float precision to N decimal digits." (setf (long-float-digits) (ceiling (* n (log 10 2)))) (long-float-digits) ; return binary precision actually set ) ? Joerg. hoehle@inf-wiss.ivp.uni-konstanz.de From @ibm3090.rz.uni-karlsruhe.de:ortigosa@INF.UFRGS.BR Thu Jun 10 03:17:41 1993 Return-Path: <@ibm3090.rz.uni-karlsruhe.de:ortigosa@INF.UFRGS.BR> Received: from ibm3090.rz.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA27472; Thu, 10 Jun 93 03:17:41 +0200 Received: from DKAUNI2 by ibm3090.rz.uni-karlsruhe.de (IBM MVS SMTP V2R2.1) with BSMTP id 9048; Thu, 10 Jun 93 03:04:45 MET Received: (from VM.GMD.DE for <@VM.GMD.DE:ortigosa@INF.UFRGS.BR> via BSMTP) Received: (from MAILER@DEARN for MAILER@DKAUNI2 via NJE) (UCLA/Mail V1.500 M-RSCS7079-7079-41); Thu, 10 Jun 93 02:57:53 Received: from brfapesp.bitnet (NJE origin SMTPUSER@BRFAPESP) by VM.GMD.DE (LMail V1.1d/1.7f) with BSMTP id 1900; Thu, 10 Jun 1993 01:12:09 +0200 Received: from vortex.ufrgs.br by brfapesp.bitnet with PMDF#10108; Wed, 9 Jun 1993 12:45 BSC (-0300 C) Received: from inf.ufrgs.br (143.54.2.99) by vortex.ufrgs.br (PMDF V4.2-11 #4493) id <01GZ6AQPX3WW000110@vortex.ufrgs.br>; Wed, 9 Jun 1993 12:35:23 -0300 Received: from jarau.inf.ufrgs.br by inf.ufrgs.br (4.1/SMI-4.1) id AA01357; Wed, 9 Jun 93 12:36:35 EST Received: from itapeva.inf.ufrgs.br by jarau.inf.ufrgs.br (4.1/SMI-4.1) id AA02661; Wed, 9 Jun 93 12:37:53 EST Date: Wed, 09 Jun 1993 12:37:53 -0500 (EST) From: ortigosa@INF.UFRGS.BR(Alvaro Manuel Ortigosa) Subject: clue instalation To: clisp-list@MA2S2.mathematik.uni-karlsruhe.de Message-Id: <9306091537.AA02661@JARAU.INF.UFRGS.BR> X-Envelope-To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Content-Transfer-Encoding: 7BIT I'm trying to install clue + clio packages, and I had several problems; can anyone tell me what I did wrong ? First I took the file pcl+clx.sept92f.clisp.tar.z, and installed pcl and clx, without problems. Then I took the file clue+clio.for-pcl.tar.Z (always from ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/packages). Because I've got problems with the disk space, I deleted the sources of pcl, but I still have sources of clx. When I tried to compile clue, I had a lot of problems (for example, it doesn't compile a "defparameter"). What I want to know is: - Have I got the corrects files ? - Do I need every source file at the time to compile clue? - What another thing could I make wrong ? I hope someone can help me. Thanks in advance. Alvaro Ortigosa From ae1181t@stnfor.ae.ge.com Thu Jun 10 15:12:30 1993 Return-Path: Received: from stnfor.ae.ge.com ([192.35.29.5]) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA27865; Thu, 10 Jun 93 15:12:30 +0200 Received: by stnfor.ae.ge.com (5.65/DEC-Ultrix/4.3) id AA19910; Thu, 10 Jun 1993 08:57:30 -0400 Date: Thu, 10 Jun 1993 08:57:30 -0400 From: ae1181t@stnfor.ae.ge.com (Osman F Buyukisik) Message-Id: <9306101257.AA19910@stnfor.ae.ge.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: <9306091537.AA02661@JARAU.INF.UFRGS.BR> (ortigosa@inf.ufrgs.br.ae.ge.com) Subject: Re: clue instalation I'm trying to install clue + clio packages, and I had several problems; can anyone tell me what I did wrong ? First I took the file pcl+clx.sept92f.clisp.tar.z, and installed pcl and clx, without problems. Then I took the file clue+clio.for-pcl.tar.Z (always from ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/packages). Because I've got problems with the disk space, I deleted the sources of pcl, but I still have sources of clx. When I tried to compile clue, I had a lot of problems (for example, it doesn't compile a "defparameter"). What I want to know is: - Have I got the corrects files ? - Do I need every source file at the time to compile clue? - What another thing could I make wrong ? I think if you look in the READMEs or install files it tells you to get the march version not sept of pcl for clue. Hope this helps. Osman From liotard@mimosa.unice.fr Fri Jun 11 09:40:50 1993 Return-Path: Received: from mimosa.unice.fr by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA28390; Fri, 11 Jun 93 09:40:50 +0200 Received: by mimosa.unice.fr (5.65c/IDA-1.2.8) id AA03959; Fri, 11 Jun 1993 09:24:58 +0100 Date: Fri, 11 Jun 1993 09:24:58 +0100 From: Liotard Jean Message-Id: <199306110824.AA03959@mimosa.unice.fr> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: clue instalation I tried to install clue and clio some time ago and didn't have a better luck. pcl+clx went fine . i could run the demo. (replacing the sept92 pcl version by march92 pcl) for clue+clio , there were some errors (?) in the code. there's a macro ignore-errors expanding into nil, and i had to add an explicit block insert-text-something? because the flet didn't built it. so i managed to run clue : the example ran fine. but all i could do in clio was to open a single window with nothing in it the type of error i got was : add 1 to nil or a bad array access. i got rid of it, but if you manage to get it working (like opening a window with a button in it !) please let me know. jean From dl@hplyot.obspm.circe.fr Tue Jun 15 12:34:53 1993 Return-Path: Received: from hplyot.obspm.circe.fr by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01806; Tue, 15 Jun 93 12:34:53 +0200 Message-Id: <9306151034.AA01806@ma2s2.mathematik.uni-karlsruhe.de> Received: by hplyot.obspm.circe.fr (16.8/16.2) id AA08034; Tue, 15 Jun 93 12:20:35 +0200 Date: Tue, 15 Jun 93 12:20:35 +0200 From: Laurent Demailly To: clisp-list Subject: Re: New CLISP version In-Reply-To: <9305241117.AA04323@ma2s2.mathematik.uni-karlsruhe.de> References: <9305241117.AA04323@ma2s2.mathematik.uni-karlsruhe.de> Hi, I'm trying to build up clisp from sources, on an HP 9000/720 HP-UX 8.07, I got the following error while building with HP CC : everything went fine 'till the 'make lisp.run' part : ------ make.lisp.run.log : ------- cc -c spvw.c unix.d: 425: TCSETATTR redefined lispbibl.d: 2327: syntax error (in preprocessor if) lispbibl.d: 3133: syntax error (in preprocessor if) lispbibl.d: 3181: syntax error (in preprocessor if) lispbibl.d: 3278: syntax error (in preprocessor if) lispbibl.d: 6680: syntax error (in preprocessor if) spvw.d: 2186: syntax error (in preprocessor if) spvw.d: 4623: signal: argument mismatch spvw.d: 4623: signal: argument mismatch spvw.d: 4623: signal: argument mismatch [... 10 pages like that...] spvw.d: 4623: signal: argument mismatch spvw.d: 4623: no space *** Error code 1 Stop. ------------- end of log ------- the first error line is around : # Typinfo: # typecode(object) und mtypecode(object) liefern den Typcode eines # Objektes obj. Bei mtypecode muß er dazu im Speicher liegen. #if !(exact_uint_size_p(oint_type_len) && (tint_type_mask == bit(oint_type_len)-1)) #define typecode(expr) \ ((tint)((oint)(expr) >> oint_type_shift) & (oint_type_mask >> oint_type_shift)) [...] Any hint ? *help* ! ;-) Yours, dl From haible Tue Jun 15 18:50:40 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02400; Tue, 15 Jun 93 18:50:40 +0200 Date: Tue, 15 Jun 93 18:50:40 +0200 From: haible (Bruno Haible) Message-Id: <9306151650.AA02400@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Laurent Demailly is trying to build up clisp from source: > on an HP 9000/720 HP-UX 8.07, I got the following error > while building with HP CC The file unix/PLATFORMS states that HP CC is too buggy for building CLISP. The problem you encountered is only the first of these bugs. Get another C compiler. Or get the binaries of the February version: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/hp9000s800/clisp.tar.z Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From syali@cs.Buffalo.EDU Tue Jun 15 19:22:07 1993 Return-Path: Received: from pegasus.cs.Buffalo.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02486; Tue, 15 Jun 93 19:22:07 +0200 Received: by pegasus.cs.Buffalo.EDU (4.1/1.01) id AA29129; Tue, 15 Jun 93 13:08:32 EDT Date: Tue, 15 Jun 93 13:08:32 EDT From: syali@cs.Buffalo.EDU (Sy Ali) Message-Id: <9306151708.AA29129@pegasus.cs.Buffalo.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: File-write-date Is file-write date supposed to produce an error, as in the below? ---- > (file-write-date "nonexistent.file") *** - FILE-WRITE-DATE: file #"/home/syali/clisp/bin/nonexistent.file" does not exist 1. Break> ---- Steele (pg 655) state "..., or nil if this cannot be determined." The intent appears that a nonexistent file return nil. This is what happens in Lucid, Allegro, and Allegro CL\PC. Sy From haible Mon Jun 21 12:35:31 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06707; Mon, 21 Jun 93 12:35:31 +0200 Date: Mon, 21 Jun 93 12:35:31 +0200 From: haible (Bruno Haible) Message-Id: <9306211035.AA06707@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: File-write-date > Is file-write date supposed to produce an error, as in the below? > Steele (pg 655) state "..., or nil if this cannot be determined." The intent > appears that a nonexistent file return nil. Maybe. The wording is not clear, so you can't rely on your interpretation. PROBE-FILE and some uses of OPEN are to the only pathname functions that are guaranteed not to produce an error even if the file does not exist, so you should use (AND (PROBE-FILE file) (FILE-WRITE-DATE file)) if you want to be careful. Bruno Haible From P2043742@csdvax.csd.unsw.EDU.AU Tue Jun 29 04:52:45 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01125; Tue, 29 Jun 93 04:52:45 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF #2556 ) id <01GZY971YY7A8WZP0M@csdvax.csd.unsw.EDU.AU>; Tue, 29 Jun 1993 12:53:46 +1000 Date: 29 Jun 1993 12:53:46 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Help to build CLUE. To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01GZY971YY7C8WZP0M@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Somebody out there, I am sending this mail in relation to compiling the CLUE system under CLISP. I've downloaded the clue+clio+xit.tar.z from ma2s2.mathematik.uni-karlsruhe.de and I've been trying to compile the system with no success. During buildig CLUE I get the following error message. ; Loading intrinsics.fas *** - The class # was specified as a super-class of the class #; but the meta-classes # and # are incompatible. 1. Break> I need clue for my theses as soon as possible. If somebody out there can help me urgently I would approtiate it. Bela Pecsek. From haible Tue Jun 29 21:52:58 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07106; Tue, 29 Jun 93 21:52:58 +0200 Date: Tue, 29 Jun 93 21:52:58 +0200 From: haible (Bruno Haible) Message-Id: <9306291952.AA07106@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: New CLISP version Once more, there is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . The list of supported platforms now include SGI Irix 4. The port to VMS stagnates, being finished at 85%. Anyone with access to a powerful VMS machine wants to continue? Excerpt from the change log: User visible changes -------------------- * X3J13 vote <89> is implemented: (SETF symbol) is now a valid syntax for a function name in (funname ...), (FUNCTION funname), (DEFUN funname ...), (COMPILE funname ...), (DISASSEMBLE funname), (ED funname), FLET, LABELS etc. New function FDEFINITION. * X3J13 vote <173> is implemented: New special form SYMBOL-MACROLET. The macroexpansion environment passed to macro expanders and to MACROEXPAND-1, MACROEXPAND, GET-SETF-METHOD, GET-SETF-METHOD-MULTIPLE-VALUE now comprises both the variable environment and the function environment. New macro DEFINE-SYMBOL-MACRO. * New macro PRINT-UNREADABLE-OBJECT. * Fixed a bug: The string comparison functions STRING<, STRING>, STRING<=, STRING>=, STRING/=, STRING-LESSP, STRING-GREATERP, STRING-NOT-GREATERP, STRING-NOT-LESSP, STRING-NOT-EQUAL now return the index of the first mismatch instead of T. * Fixed a bug: The functions REMOVE-DUPLICATES and DELETE-DUPLICATES produced an out-of-bounds error when called on a small sequence, with :FROM-END T argument and without :END argument. Portability ----------- * Updated support for DJGPP. Other modifications ------------------- * The macros SETF, PUSH, POP, PUSHNEW have been completely rewritten. * Changed internal representation of special form handlers: They now consume regular memory. This frees one type tag. * Miscellaneous documentation updates. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From pdudey@Willamette.EDU Tue Jun 29 22:22:56 1993 Return-Path: Received: from Willamette.EDU (jupiter.willamette.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07514; Tue, 29 Jun 93 22:22:56 +0200 Received: by Willamette.EDU (4.1/JPC-1.1) id AA03665; Tue, 29 Jun 93 13:22:25 PDT Date: Tue, 29 Jun 1993 13:22:01 -0700 (PDT) From: Peter Dudey Subject: unsubscribe To: Clisp mailing list Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII unsubscribe CLISP-LIST ( Peter Dudey, pdudey@willamette.edu, 257 NE 13th St, Salem, Oregon 97301 ) ( AI researcher / Go player (11 kyu) / Game designer / SF novelist manque ) ( Clinton actually REDUCED the deficit. When did that last happen? ) ( Barney is a velociraptor. "Give me Slack, or give me food, OR KILL ME!" ) From P2043742@csdvax.csd.unsw.EDU.AU Wed Jun 30 06:13:22 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07852; Wed, 30 Jun 93 06:13:22 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF #2556 ) id <01GZZQADY4FK8WZRYU@csdvax.csd.unsw.EDU.AU>; Wed, 30 Jun 1993 14:14:23 +1000 Date: 30 Jun 1993 14:14:23 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Correction [Help to build CLUE] To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01GZZQADY4FM8WZRYU@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Somebody out there, Yesterday I sent a mail in relation to compiling the CLUE system under CLISP. >I've downloaded the clue+clio+xit.tar.z from ma2s2.mathematik.uni-karlsruhe.de >and I've been trying to compile the system with no success. > >During buildig CLUE I get the following error message. > >; Loading intrinsics.fas *** - The class # was >specified as a super-class of the class #; >but the meta-classes # >and # are incompatible. >1. Break> > >I need clue for my theses as soon as possible. > >If you can help me urgently I would approtiate it. Today I recompiled the PCL because the messages were strange and Bruno suspected that something was wrong with the PCL. I get "better" error messages now jet unable to build CLUE. The error message I am geting now is: Compilation of file /tmp_mnt/net/pg1/p2043742/cl2/clisp/clue+clio+xit/clue/intr insics.lsp is finished. 0 errors, 6 warnings The following functions were used but not defined: CONTACT-NAME CONTACT-RESOURCE-PARENT CONTACT-DISPLAY FIND-CONTACT COMPOSITE-CHILDREN CONTACT-PARENT CONTACT-STATE OWNS-FOCUS-P CONTACT-WIDTH CONTACT-HEIGHT CONVERT INITIALIZE-RESOURCE-SLOTS ADD-TO-PARENT ADD-CHILD CHANGE-LAYOUT CONTACT-BACKGROUND CONTACT-DEPTH CONTACT-X CONTACT-Y PCL::|SETF CLUEI CONTACT-STATE| DELETE-CHILD UPDATE-TREE INITIALIZE-GEOMETRY REALIZE INITIAL-STATE-TRANSITION DELETE-TIMER DELETE-MODE DESTROY CONTACT-ROOT CONTACT-BORDER-WIDTH PROCESS-NEXT-EVENT MANAGE-PRIORITY PCL::|SETF CLUEI CONTACT-PRIORITY| CONTACT-EVENT-MASK COMPOSITE-FOCUS ACCEPT-FOCUS-P MANAGE-GEOMETRY MOVE RESIZE ; Loading intrinsics.fas *** - The class # was specified as a super-class of the class #; but the meta-classes # and # are incompatible. touch stamp-fas rm clue.mem rm: clue.mem non-existent *** Error code 2 Stop. Sorry about it. Hopefully now somebody can help me to solve this mith. Yours, Bela From haible Wed Jun 30 12:04:38 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08239; Wed, 30 Jun 93 12:04:38 +0200 Date: Wed, 30 Jun 93 12:04:38 +0200 From: haible (Bruno Haible) Message-Id: <9306301004.AA08239@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: PCL for CLISP For the new version of CLISP announced yesterday I had to patch PCL a bit more. The *only* version of PCL I support from now on is September 1992 (f) PCL. On ma2s2.mathematik.uni-karlsruhe:/pub/lisp/clisp/packages/ the files pcl.sept92f.clisp.tar.z and pcl+clx.sept92f.clisp.tar.z . There was a bug in my earlier ports of September 1992 PCL: structure classes don't work properly. Those who got pcl+clx.sept92f.clisp.tar.z between 20 May 1993 and 11 June 1993 should upgrade to the current one. I wish to thank Scott Musman who pointed me to the problem. Now that structure classes work, there is no more need for July 92 PCL. Neither for March 92 PCL: there have been significant improvements and bugfixes in the meantime. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From P2043742@csdvax.csd.unsw.EDU.AU Thu Jul 1 19:28:29 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA14579; Thu, 1 Jul 93 19:28:29 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF #2556 ) id <01H01K2V2IQO8WZRC4@csdvax.csd.unsw.EDU.AU>; Thu, 1 Jul 1993 21:37:25 +1000 Date: 01 Jul 1993 21:37:25 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Help to build XIT To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H01K2V2SDU8WZRC4@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Clisp-list, I am sending this mail in relation to build XIT. During compilation I get the following error message: . . . ;; Loading of file /tmp_mnt/net/pg1/p2043742/cl2/clisp/clue+clio+xit/kernel/tim ers.fas is finished Compiling file /tmp_mnt/net/pg1/p2043742/cl2/clisp/clue+clio+xit/kernel/interac tion-window.lsp ... *** - Illegal SETF place: (MOUSE-DOCUMENTATION SELF) and the compilation terminates. If somebody was able to build XIT please explain me what's wrong here and how can I overcome this problaim and build XIT successfuly. I look forward to hearing from you. Bela Pecsek From P2043742@csdvax.csd.unsw.EDU.AU Mon Jul 5 05:41:04 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16495; Mon, 5 Jul 93 05:41:04 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #2556) id <01H06OLJXQTS8X05EC@csdvax.csd.unsw.EDU.AU>; Mon, 5 Jul 1993 13:41:16 +1000 Date: Mon, 05 Jul 1993 13:41:16 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Help to build CLUE and XIT please!! To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H06OLJYJR68X05EC@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear CLISP users, I have just built CLUE with CLISP on SGI Indigo (just released PCL and CLX "Sept92f") and everything went just fine but when I try to test it with examples/menu.lsp I get the the following error message: ** - Continuable Error "fixed" cannot be converted to type FONT. If you continue (by typing 'continue'): Retry 1. Break> If somebody knows what the problem is please let me know. Another question: Is there anybody out there who could build XIT with CLISP? If yes please contact me. I look forward to hearing from you, Yours, Bela Pecsek From RLS@waikato.ac.nz Mon Jul 5 07:06:03 1993 Return-Path: Received: from grace.waikato.ac.nz by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16589; Mon, 5 Jul 93 07:06:03 +0200 Message-Id: <9307050506.AA16589@ma2s2.mathematik.uni-karlsruhe.de> Date: Mon, 5 Jul 93 17:05 +1200 From: RLS@waikato.ac.nz Subject: Amiga 600 and Clisp To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" I have just installed CLisp on my Amiga 600 (68000 processor) which has 1MB chip memory plus 4MB fast memory through a PCMCIA expansion. When I run CLisp it uses about 500K of fast memory and then uses up nearly all the chip memory. When I load anything into CLisp it continues to fill up the chip memory, ignoring all the free space on the expansion card. Consequently the chip memory fills up after loading only a few files and CLisp complains that there is no more room for lisp objects. This is silly, the progran should use the fast memory and leave chip memory for the graphics coprocessors, shouldn't it? I may be overlooking something very obvious, so please point it out if you know the solution. No other programs have had this problem with recognizing the presence of the extra memory so I am sure it is a CLisp problem rather than an Amiga problem. If the answer to this problem is already posted somewhere could you please point me to it. P.S. I am using the *-low version as recommended in the documentation that comes with the Amiga version. Thanx in advance, Richard Shepherd. From mcg@waikato.ac.nz Thu Jul 8 00:25:07 1993 Return-Path: Received: from truth.waikato.ac.nz by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03247; Thu, 8 Jul 93 00:25:07 +0200 Received: from fantail.math.waikato.ac.nz by waikato.ac.nz; Thu, 8 Jul 93 10:23 +1200 Received: from pukeko.math.waikato.ac.nz by fantail.math.waikato.ac.nz (4.1/SMI-4.1) id AA02396; Thu, 8 Jul 93 10:27:52 NZS Date: Thu, 8 Jul 93 10:27:52 NZS From: mcg@waikato.ac.nz Subject: CLISP - installation and conversion. To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <9307072227.AA02396@fantail.math.waikato.ac.nz> Hi, I'm an Acorn Archimedes user (standard Ansi C), and I'm wishing to port a version of CLISP to the Archimedes. What I've tried to do so far is obtain the code, unzip it (gzip 1-2-3) and compile it using the installation for the Sun sparc station on a sparc station- this is so that I can get the converted .c files (and to see if the code can be easily compiled o Alas, when I try to run the ./target, it does a 8Meg core dump and when I try to compile the code after makeing the makefile (running ./makemake > makefile) it stuffs up on the spvw compilation - # illegal character ... endif undefined.... I cannot work out (form the .d files) exactly what the .c files should be (comments are in german!)... this leads to my questuions :- a) Does anyone have any suggestions on what my compilation problems are and how to correct it ? b) Does anyone know of anyone with an Archimedes that is also more familiar with CLISP source code, and might like to assist me in trying to convert CLISP? Thanks, Martin. From rainer@wmax60.mathematik.uni-wuerzburg.de Fri Jul 9 12:16:14 1993 Return-Path: Received: from wmax60.mathematik.uni-wuerzburg.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05333; Fri, 9 Jul 93 12:16:14 +0200 Received: from wmax65 by wmax60.mathematik.uni-wuerzburg.de (NX5.67c/NX3.0M) id AA03414; Fri, 9 Jul 93 12:14:40 +0200 From: Rainer Frohnhoefer Message-Id: <9307091014.AA03414@wmax60.mathematik.uni-wuerzburg.de> Received: by wmax65.mathematik.uni-wuerzburg.de (NX5.67c/NX3.0X) id AA00178; Fri, 9 Jul 93 12:14:31 +0200 Date: Fri, 9 Jul 93 12:14:31 +0200 Received: by NeXT.Mailer (1.87.1) Received: by NeXT Mailer (1.87.1) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Clisp on NeXT Now, has anyone build CLISP successfully on a NeXT (Moto CPU) under NeXTStep 3.0? I got the sources, but am no enough a C Freak to figure out all the steps to run through the makefiles. Some system tools seem to be missing ... Thanks , Rainer Frohnhoefer at UNIWUE rainer@wmax60.mathematik.uni-wuerzburg.de ... NeXTMAIL welcome! From rainer@wmax60.mathematik.uni-wuerzburg.de Fri Jul 9 12:31:15 1993 Return-Path: Received: from wmax60.mathematik.uni-wuerzburg.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05401; Fri, 9 Jul 93 12:31:15 +0200 Received: from wmax65 by wmax60.mathematik.uni-wuerzburg.de (NX5.67c/NX3.0M) id AA03442; Fri, 9 Jul 93 12:29:39 +0200 From: Rainer Frohnhoefer Message-Id: <9307091029.AA03442@wmax60.mathematik.uni-wuerzburg.de> Received: by wmax65.mathematik.uni-wuerzburg.de (NX5.67c/NX3.0X) id AA00194; Fri, 9 Jul 93 12:29:37 +0200 Date: Fri, 9 Jul 93 12:29:37 +0200 Received: by NeXT.Mailer (1.87.1) Received: by NeXT Mailer (1.87.1) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de From haible Sat Jul 10 00:38:43 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05999; Sat, 10 Jul 93 00:38:43 +0200 Date: Sat, 10 Jul 93 00:38:43 +0200 From: haible (Bruno Haible) Message-Id: <9307092238.AA05999@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: CLUE installation How to compile CLUE with CLISP: 1. Take clisp-1993-05-24 or older (the clisp-1993-06-29 has two bugs), PCL and CLX (the September 1992 f version), CLUE from the ma2s2:/pub/lisp/clisp/packages/clue+clio+xit.clisp.tar.z . 2. Jean Liotard noted that there is an error in clue/resource.lsp. The definition of IGNORE-ERRORS ought to read #+CLISP (defmacro ignore-errors (&body body) (let ((blockvar (gensym))) `(BLOCK ,blockvar (LET ((*ERROR-HANDLER* #'(LAMBDA (&REST ARGS) (RETURN-FROM ,blockvar (VALUES NIL ARGS)) )) ) ,@body ) ) ) ) (The line ",@body" was missing.) A typical symptom of this problem was observed by Bela Pecsek : > ** - Continuable Error > "fixed" cannot be converted to type FONT. > If you continue (by typing 'continue'): Retry 3. Type "make" in the pcl/ directory, then in the clx/ directory, then move clx.mem to clue+clio+xit/clx/clx.mem, then type "make" in the clue+clio+xit/clue/ directory. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From @vm.urz.Uni-Heidelberg.de:KOCH@novell1.gs.uni-heidelberg.de Mon Jul 12 19:33:06 1993 Return-Path: <@vm.urz.Uni-Heidelberg.de:KOCH@novell1.gs.uni-heidelberg.de> Received: from vm.urz.Uni-Heidelberg.de (vm.hd-net.uni-heidelberg.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12320; Mon, 12 Jul 93 19:33:06 +0200 Received: from pc30.gs.uni-heidelberg.de by vm.urz.Uni-Heidelberg.de (IBM VM SMTP V2R2) with TCP; Mon, 12 Jul 93 19:28:51 CET Received: From NOVELL1_GS/WORKQUEUE by pc30.gs.uni-heidelberg.de via Charon-4.0A-VROOM with IPX id 100.930712192027.352; 12 Jul 93 19:20:45 -0100 Message-Id: To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de From: KOCH@novell1.gs.uni-heidelberg.de Organization: LCL - University of Heidelberg Date: 12 Jul 93 19:20:16 MET Subject: Garnet and CLISP for Linux ? Priority: normal X-Mailer: Pegasus Mail v2.3 (R5). Hello! Did someone try to compile Garnet under CLISP on a Linux-Box? Most of the code seems to compile ok, but loading two modules from the gadget-set produced the following error message: Compiling file /usr/local/garnet/src/gadgets/save-gadget.lsp ...Object SAVE-GADGET Object SAVE-GADGET Compilation of file /usr/local/garnet/src/gadgets/save-gadget.lsp is finished. 0 errors, 0 warnings ;; Loading file /usr/local/garnet/src/gadgets/save-gadget.fas ... *** - SYSTEM::STORE: #\K does not fit into #(1 0 13 0 ... [ total of 8192 numbers, mostly 0's ] ... 0 0), bad type 1. Break> Moreover, the same error occurs when calling the update-window function in Garnet. Since Garnet is running without any problems under CMUCL on the Sun in my office, I assume that this is a CLISP- related problem. I apologize for not having tracked down the problem, but compiling Garnet takes several hours on my linux-box and I do not intend to spend another afternoon trying to hypnotize the activity-LED of my hard disk until I am sure that the problem has not already been solved by someone else. Yours HDK// From P2043742@csdvax.csd.unsw.EDU.AU Tue Jul 13 07:04:36 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12611; Tue, 13 Jul 93 07:04:36 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #2556) id <01H0HXSWVM5I8X0R3K@csdvax.csd.unsw.EDU.AU>; Tue, 13 Jul 1993 15:03:33 +1000 Date: Tue, 13 Jul 1993 15:03:33 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Help to build XIT!! To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H0HXSWVM5K8X0R3K@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear CLISP users, I've been trying to build XIT with the sept92f PCL, CLX and CLUE from clue+clio+xit.clisp.tar.z and I get the following error message: Compilation of file /tmp_mnt/net/pg1/p2043742/cl2/clisp/clue+clio+xit/kernel/vi rtual-dispels.lsp is finished. 0 errors, 15 warnings The following functions were used but not defined: DISPLAY-TEXT DELETE-CHARACTER SET-LINE-POINTS SET-LINE-INNER-POINTS VALIDATE-ENCLOSING-REGION LINE-POINT-COORDINATES SET-LINE-POINT-COORDINATES DISPLAY-LINE DISPLAY-OUTLINE DRAW-OUTLINE ENCLOSING-REGION-OFFSET INVALIDATE-ENCLOSING-REGION FIND-LINE-INNER-POINT DRAG-INNER-POINT-WITH-MOUSE LINE-POINTS ARROW-SIZES DRAW-ARROW DRAW-ARROW-OUTLINE ;; Loading file /tmp_mnt/net/pg1/p2043742/cl2/clisp/clue+clio+xit/kernel/virtua l-dispels.fas ... *** - APPLY: too few arguments given to # If somebody has been able to build XIT please help me. I can't figure out what's wrong here. Allthough I understand the error I can't find the source of it. I need the system as soon as possible for my theses. If somebody can help me I'd approtiate it. Thanks for your help in advance. Yours, Bela. From P2043742@csdvax.csd.unsw.EDU.AU Fri Jul 16 05:16:47 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18519; Fri, 16 Jul 93 05:16:47 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H0M0U3288G8WW2W5@csdvax.csd.unsw.EDU.AU>; Fri, 16 Jul 1993 13:13:06 +1000 Date: Fri, 16 Jul 1993 13:13:06 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Help to build XIT To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H0M0U32RJ68WW2W5@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear CLISP users, I've been trying to build XIT with the sept92f PCL, CLX and CLUE from clue+clio+xit.clisp.tar.z and I get the following error message: Compilation of file /tmp_mnt/net/pg1/p2043742/cl2/clisp/clue+clio+xit/kernel/vi rtual-dispels.lsp is finished. 0 errors, 15 warnings The following functions were used but not defined: DISPLAY-TEXT DELETE-CHARACTER SET-LINE-POINTS SET-LINE-INNER-POINTS VALIDATE-ENCLOSING-REGION LINE-POINT-COORDINATES SET-LINE-POINT-COORDINATES DISPLAY-LINE DISPLAY-OUTLINE DRAW-OUTLINE ENCLOSING-REGION-OFFSET INVALIDATE-ENCLOSING-REGION FIND-LINE-INNER-POINT DRAG-INNER-POINT-WITH-MOUSE LINE-POINTS ARROW-SIZES DRAW-ARROW DRAW-ARROW-OUTLINE ;; Loading file /tmp_mnt/net/pg1/p2043742/cl2/clisp/clue+clio+xit/kernel/virtua l-dispels.fas ... *** - APPLY: too few arguments given to # If somebody has been able to build XIT please help me. I can't figure out what's wrong here. Allthough I understand the error I can't find the source of it. I need the system as soon as possible for my theses. If somebody can help me I'd approciate it. Thanks for your help in advance. Yours, Bela. From P2043742@csdvax.csd.unsw.EDU.AU Sat Jul 17 07:47:36 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU ([149.171.192.3]) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19093; Sat, 17 Jul 93 07:47:36 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H0NKEHKE3K8WW5YL@csdvax.csd.unsw.EDU.AU>; Sat, 17 Jul 1993 15:43:50 +1000 Date: Sat, 17 Jul 1993 15:43:50 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Re: Garnet and CLISP for Linux ? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H0NKEHN2JM8WW5YL@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear "HDK", Sorry about not knowing your name! I am trying to build garnet with CLISP on SGI IRIS INDIGO and I get the same error you mentioned exept I get #\U " *** - SYSTEM::STORE: #\U does not fit into #(1 0 0.....), bad type " instead of #\K you mentioned. If somebody has already tracked down the problem and will or has already sent you a mail how to overcome this problem please forward the mail to me. I'd approcite it. I need garnet for my thesis as soon as possible. Thank you in advance for your help. I look forward to hearing from you, Yours, Bela Pecsek. From P2043742@csdvax.csd.unsw.EDU.AU Sat Jul 17 07:55:01 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU ([149.171.192.3]) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19172; Sat, 17 Jul 93 07:55:01 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H0NKO6KOX88WW5YL@csdvax.csd.unsw.EDU.AU>; Sat, 17 Jul 1993 15:51:38 +1000 Date: Sat, 17 Jul 1993 15:51:38 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Re: Garnet and CLISP for Linux ? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H0NKO6KOXA8WW5YL@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear CLISP users, I have the same problem exept I get the error with #\U exapt of #\K *** - SYSTEM::STORE: #\U does not fit into #(1 0 0 .....), bad type. If somebody has allready tracked down this problem please send me a mail too. I'd approciate it. Yours, Bela Pecsek. From haible Sat Jul 17 17:50:43 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23141; Sat, 17 Jul 93 17:50:43 +0200 Date: Sat, 17 Jul 93 17:50:43 +0200 From: haible (Bruno Haible) Message-Id: <9307171550.AA23141@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: New CLISP version Once more, there is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . The port to VMS still stagnates, being finished at 85%. I haven't heard from anyone with access to a powerful VMS machine. If you know someone who would be willing to help me with the last steps of the port, please tell me. Excerpt from the change log: User visible changes -------------------- * X3J13 vote <150> is mostly implemented: Reading and printing of symbols (and characters with bits or fonts attributes) is now affected by (READTABLE-CASE *READTABLE*), whose value may be one of :UPCASE, :DOWNCASE and :PRESERVE. New place READTABLE-CASE. * Support for immutable objects is implemented. Attempts to modify read-only data will signal an error. Program text and quoted constants loaded from files are considered read-only data. Never again will you be allowed to write programs that destructively modify themselves! This works only for Unix implementations with powerful virtual memory management, currently only Linux and SunOS. To enable it, you have to add -DIMMUTABLE to the CFLAGS in the makefile when building CLISP. * Fixed DEFUN such that (progn (flet ((bar (x) (+ x 10))) (defun bar (y) (if (zerop y) 1 (bar (1- y)))) (bar 1) ) returns 10 instead of 1. * Fixed a bug in the compiler that caused the compiler to crash by endless recursion and stack overflow when compiling assignments to symbol macros. Portability ----------- * Added support for m88000 CPU. * Amiga version: detect 68000 CPU, for better use of present memory. Other modifications ------------------- * Miscellaneous documentation updates. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From reggie@muon.phys.washington.edu Sat Jul 17 21:46:06 1993 Return-Path: Received: from muon.phys.washington.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00328; Sat, 17 Jul 93 21:46:06 +0200 Received: by muon.phys.washington.edu via SMTP (920330.SGI/911001.SGI) for clisp-list@ma2s2.mathematik.uni-karlsruhe.de id AA13635; Sat, 17 Jul 93 12:53:11 -0700 Message-Id: <9307171953.AA13635@muon.phys.washington.edu> To: clisp-list Subject: Re: New CLISP version In-Reply-To: Your message of "Sat, 17 Jul 1993 17:52:52 +0200." <9307171550.AA23141@ma2s2.mathematik.uni-karlsruhe.de> Date: Sat, 17 Jul 1993 12:53:10 -0700 From: Reggie Perry Make a note that on my SGI Indigio running IRIX 4.0.5 I have to change line 56 of unix.d to read: #ifdef HAVE_MMAP #include #include #if defined(UNIX_SYSV) && (defined(sgi) || defined(__sgi)) extern void *mmap (void *addr, int len, int prot, int flags, int fd, off_t off); #else extern caddr_t mmap (caddr_t addr, MMAP_SIZE_T len, int prot, int flags, int fd, off_t off); /* original definition */ #endif ------------------- Reginald S. Perry -- University of Washington, Department of Physics FM-15 Seattle, Washington 98195 reggie@u.washington.edu (General-Mail) reggie@quantum.phys.washington.edu (NeXTMail) From haible Sun Jul 18 17:20:54 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00575; Sun, 18 Jul 93 17:20:54 +0200 Date: Sun, 18 Jul 93 17:20:54 +0200 From: haible (Bruno Haible) Message-Id: <9307181520.AA00575@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Reginald S. Perry writes: > Make a note that on my SGI Indigio running IRIX 4.0.5 I have > to change line 56 of unix.d to read: ... > extern void *mmap (void *addr, int len, int prot, int flags, int fd, off_t off); Thanks. Will be fixed in the next release. Btw, I have built CLISP binaries for Linux with immutable objects support enabled. In /pub/lisp/clisp/linux/ on ma2s2. It requires Linux 0.99.11 and libc-4.4.1. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Tue Jul 20 19:59:08 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02374; Tue, 20 Jul 93 19:59:08 +0200 Date: Tue, 20 Jul 93 19:59:08 +0200 From: haible (Bruno Haible) Message-Id: <9307201759.AA02374@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Garnet and CLISP ? Mr Heinz-Detlev Koch and Mr Bela Pecsek reported problems when building Garnet with CLISP, on Linux resp. SGI/Irix. I can report that I hadn't any problems in doing that on SunOS4. Please take the newest CLISP version, my port of CLX ma2s2:/pub/lisp/clisp/packages/pcl+clx.sept92f.clisp.tar.z and Garnet ma2s2:/pub/lisp/clisp/packages/garnet.clisp.tar.z and tell us whether the problems remain. In case they do, then look at the backtrace. I can't test it on Linux since it won't compile in less than 10 megabytes of RAM, and I have just 8 MB. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From @ibm3090.rz.uni-karlsruhe.de:ACPS7221@RYEVM.RYERSON.CA Tue Jul 20 23:19:20 1993 Return-Path: <@ibm3090.rz.uni-karlsruhe.de:ACPS7221@RYEVM.RYERSON.CA> Received: from ibm3090.rz.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02510; Tue, 20 Jul 93 23:19:20 +0200 Received: from DKAUNI2 by ibm3090.rz.uni-karlsruhe.de (IBM MVS SMTP V2R2.1) with BSMTP id 5816; Tue, 20 Jul 93 23:15:48 MET Received: (from VM.GMD.DE for <@VM.GMD.DE:ACPS7221@RYEVM.RYERSON.CA> via BSMTP) Received: (from MAILER@DEARN for MAILER@DKAUNI2 via NJE) (UCLA/Mail V1.500 M-RSCS2562-2562-24); Tue, 20 Jul 93 23:15:36 Received: from RyeVm.Ryerson.Ca (NJE origin MAILER@RYERSON) by VM.GMD.DE (LMail V1.1d/1.7f) with BSMTP id 5779; Tue, 20 Jul 1993 23:02:51 +0200 Received: from RyeVm.Ryerson.Ca (ACPS7221) by RyeVm.Ryerson.Ca (Mailer R2.08 ptf039) with BSMTP id 1834; Tue, 20 Jul 93 17:00:21 EST Date: Tue, 20 Jul 93 16:57:59 EST From: Michael Caro Subject: Re: Garnet and CLISP ? To: clisp-list Message-Id: <930720.165759.EST.ACPS7221@RyeVm.Ryerson.Ca> In-Reply-To: <9307201759.AA02374@ma2s2.mathematik.uni-karlsruhe.de> Hello Bruno: You are doing a terrific job with CLISP! Keep up the good work. Every thing works perfectly for me. Thanks.:) BTW, why don't you use a swap partition or file, to get the extra few Megs of Ram you need to try and compile Garnet?. Kind Regards, Michael. From lrw@sdsc.edu Wed Jul 21 00:45:15 1993 Return-Path: Received: from mailserver.sdsc.edu (rosebud.sdsc.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02565; Wed, 21 Jul 93 00:45:15 +0200 Received: from zippy.sdsc.edu.sdsc.edu by mailserver.sdsc.edu (4.1/4.7) id AA26520; Tue, 20 Jul 93 15:42:09 PDT Received: by zippy.sdsc.edu.sdsc.edu (5.65/DEC-Ultrix/4.3-sdsc) id AA10772; Tue, 20 Jul 1993 15:41:50 -0700 Date: Tue, 20 Jul 1993 15:41:50 -0700 From: lrw@sdsc.edu Message-Id: <9307202241.AA10772@zippy.sdsc.edu.sdsc.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: can't compile CLISP on a DecStation 5000 I am having the following problem compiling the latest version of CLISP on my DecStation 5000. Does anyone know what is causing the following problems: ------------------------- cut here ---------------------------------- (echo "# 1 \"spvw.i\"" ; cc -E -O -DNO_MULTIMAP_SHM -USINGLE_PASS_COMPILER spvw.c | ./mergestrings) > spvw.i cpp: warning /usr/include/stdlib.h:64: NULL redefined cpp: error lispbibl.d:2326: syntax error cpp: error lispbibl.d:2509: syntax error cpp: error lispbibl.d:2514: syntax error cpp: error lispbibl.d:2519: syntax error cpp: error lispbibl.d:2524: syntax error cpp: error lispbibl.d:2529: syntax error cpp: error lispbibl.d:2534: syntax error cpp: error lispbibl.d:2540: syntax error cpp: error lispbibl.d:3199: syntax error cpp: error lispbibl.d:3247: syntax error cpp: error lispbibl.d:3344: syntax error cpp: error spvw.d:2395: syntax error cpp: error spvw.d:7616: missing endif cc -O -DNO_MULTIMAP_SHM -c spvw.i ccom: Error: lispbibl.d, line 2325: syntax error (((8)==8)||((8)==16)||((8)==32)||((8)==32)) && ( (0xEF000000 >> 24) == (1<<(8))-1)) ----^ ccom: Error: lispbibl.d, line 3159: syntax error object car; } -----------------------------^ ccom: Error: lispbibl.d, line 3161: redeclaration of cons_ typedef cons_ * Cons; --------------^ ccom: Error: lispbibl.d, line 3161: syntax error typedef cons_ * Cons; --------------^ ccom: Error: lispbibl.d, line 3197: syntax error ((24) / (8))== ((24+8) / (8))) -----^ ccom: Error: endif, line 3: syntax error typedef struct { union { object _GCself; uintB flags[sizeof(object)]; } header; ----------------------------------------------------------------------------------------- ------------------------------------------^ ccom: Error: endif, line 9: syntax error } ---------------^ ccom: Error: endif, line 11: redeclaration of symbol_ typedef symbol_ * Symbol; ----------------^ ccom: Error: endif, line 11: syntax error typedef symbol_ * Symbol; ----------------^ ccom: Error: lispbibl.d, line 3343: syntax error (((16)==8)||((16)==16)||((16)==32)||((16)==32)))) -----^ ccom: Error: lispbibl.d, line 3499: syntax error typedef struct { union { object _GCself; uintB flags[sizeof(object)]; } header; ----------------------------------------------------------------------------------------- ------------------------------------------^ ccom: Error: lispbibl.d, line 3502: syntax error } ---------------^ ccom: Error: lispbibl.d, line 3504: redeclaration of bignum_ typedef bignum_ * Bignum; ----------------^ ccom: Error: lispbibl.d, line 3504: syntax error typedef bignum_ * Bignum; ----------------^ ccom: Error: constsym.d, line 32: syntax error symbol_ S_nil; -------------^ (ccom): constsym.d, line 32: cannot recover from earlier errors: goodbye! symbol_ S_nil; -------------------^ *** Error code 1 Stop. ------------------- end of gratuitous error messages ------------------- Len Wanger lrw@sdsc.edu From dudeyp@storm.cs.orst.edu Wed Jul 21 00:57:13 1993 Return-Path: Received: from storm.cs.orst.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02615; Wed, 21 Jul 93 00:57:13 +0200 Received: by storm.cs.orst.edu (5.65/1.34) id AA02231; Tue, 20 Jul 93 15:54:07 -0700 Date: Tue, 20 Jul 1993 15:53:54 -0700 (PDT) From: Peter Dudey Subject: unsubscribe To: clisp-list In-Reply-To: <9307202241.AA10772@zippy.sdsc.edu.sdsc.edu> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII unsubscribe From P2043742@csdvax.csd.unsw.EDU.AU Wed Jul 21 07:29:28 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02711; Wed, 21 Jul 93 07:29:28 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H0T4WQ8LV48WWIBG@csdvax.csd.unsw.EDU.AU>; Wed, 21 Jul 1993 15:25:31 +1000 Date: Wed, 21 Jul 1993 15:25:31 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Problem with Garnet! To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H0T4WQ955E8WWIBG@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear CLISP users, I've just built garnet on SGI IRIS INDIGO (IRIX Release 4.0.5F System V) and everything went just fine. The problem came when I wanted to use it. I wanted to test it with the demos supplied with the system but every time I try to type the commands listed in tourcommands.lsp I get the following error message. *** - UNIX error 9 (EBADF): Bad file number For example: > (create-instance 'mywindow inter:interactor-window (:width 500)) Object MYWINDOW Warning - create-schema is destroying the old #k. #k > (opal:update mywindow) *** - UNIX error 9 (EBADF): Bad file number 1. Break> (create-instance 'myagg opal:aggregate) Object MYAGG #k 1. Break> (s-value mywindow :aggregate myagg) #k 1. Break> (create-instance 'myrect moving-rectangle) Object MYRECT #k 1. Break> (opal:add-component myagg myrect) #k 1. Break> (opal:update mywindow) *** - UNIX error 9 (EBADF): Bad file number 2. Break> And the same with GILT > (gilt::do-go) *** - UNIX error 9 (EBADF): Bad file number 1. Break> Michael Caro has just reported today that for him everything works perfectly. Although I don't know which machine he is using I can assume that something might be wrong with my machine or CLISP but the funny thing that the compilation went perfectly fine. I am using the sept92f (newest) version of clx and pcl so I think CLISP is OK. And anyway the errors are UNIX errors so I assume the errors are UNIX related although I am unfortunately not an UNIX expert. If somebody out there can figure out what's wrong please help me. Thanks for your help in advance. I look forward to hearing from you. Yours, Bela Pecsek From haible Wed Jul 21 23:44:43 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01429; Wed, 21 Jul 93 23:44:43 +0200 Date: Wed, 21 Jul 93 23:44:43 +0200 From: haible (Bruno Haible) Message-Id: <9307212144.AA01429@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: can't compile CLISP on a DecStation 5000 Len Wanger writes: > I am having the following problem compiling the latest version of CLISP > on my DecStation 5000. Does anyone know what is causing the following > problems: > (echo "# 1 \"spvw.i\"" ; cc -E -O -DNO_MULTIMAP_SHM -USINGLE_PASS_COMPILER spvw.c | ./mergestrings) > spvw.i > cpp: warning /usr/include/stdlib.h:64: NULL redefined > cpp: error lispbibl.d:2326: syntax error > cpp: error lispbibl.d:2509: syntax error This is caused by a C preprocessor that does not expand macros with arguments within preprocessor directives. The preprocessor is unusable, you have to use the GNU gcc preprocessor included in the CLISP distribution. In src/makemake.in (around line 394) you can read if [ "$TSYSOS" = coherent ] ; then # If one cannot use the default cc preprocessor: Use GNU cpp instead. You have to replace this by if true; then and restart the "target" command. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Thu Jul 22 00:08:48 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01490; Thu, 22 Jul 93 00:08:48 +0200 Date: Thu, 22 Jul 93 00:08:48 +0200 From: haible (Bruno Haible) Message-Id: <9307212208.AA01490@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Patches to CLISP I just got a good patch for CLISP, implementing X3J13 vote <149>, i.e. fixing the SUBSEQ function. The patch was made by Thomas Kirk and reached me six weeks later, having travelled through two other people's mailbox. This delay is unnecessary. So, please, if you have a useful patch for CLISP, come and share it with us! Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From ambrisko@kpc.com Thu Jul 22 04:02:25 1993 Return-Path: Received: from horn.kpc.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01739; Thu, 22 Jul 93 04:02:25 +0200 Received: by horn.kpc.com (5.65/kpc-930707); id AA03040; Wed, 21 Jul 93 18:59:13 -0700 Received: from pine by kpc.com (5.65/kpc-930604) id AA17582; Wed, 21 Jul 93 19:00:35 -0700 Received: from tasha by mailbox.kpc.com (4.1/kpc-930604) id AA14925; Wed, 21 Jul 93 18:59:22 PDT From: ambrisko@kpc.com (Douglas Ambrisko) Message-Id: <9307220158.AA01565@tasha.kpc.com> Subject: Re: Problem with Garnet! To: P2043742@csdvax.csd.unsw.EDU.AU Date: Wed, 21 Jul 1993 18:58:34 -0800 (PDT) Cc: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: <01H0T4WQ955E8WWIBG@csdvax.csd.unsw.EDU.AU> from "P2043742@csdvax.csd.unsw.EDU.AU" at Jul 21, 93 03:25:31 pm X-Mailer: ELM [version 2.4 PL17] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 978 P2043742@csdvax.csd.unsw.EDU.AU writes: | | Dear CLISP users, | | I've just built garnet on SGI IRIS INDIGO (IRIX Release 4.0.5F System V) | and everything went just fine. The problem came when I wanted to use it. I | wanted to test it with the demos supplied with the system but every time I | try to type the commands listed in tourcommands.lsp I get the following | error message. | | *** - UNIX error 9 (EBADF): Bad file number [stuff deleted] I just grabbed the clisp versin of garnet and built it. I hit the same problem with the "saveinitmem" oni a DEC alpha box . The problem is that garnet assumes you do a load of all compiled modules including opal/defs which calls "initialize-x11-values". So a work-around that I'm doing now is to load the "saveinitmem" version then loading opal/defs. Everything seems to works except for the "subseq" problem where (subseq "hello" 3 nil) bombs. I ended up modifying the clisp source to make it work. Doug A. From schue@ira.uka.de Thu Jul 22 13:47:02 1993 Return-Path: Received: from iraun1.ira.uka.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02902; Thu, 22 Jul 93 13:47:02 +0200 Received: from ira.uka.de by iraun1.ira.uka.de with SMTP (PP) id <28612-0@iraun1.ira.uka.de>; Thu, 22 Jul 1993 13:43:20 +0200 Received: from deathstar.iaks.ira.uka.de by walhalla.eiss.ira.uka.de (4.1/SMI-4.1) id AA20254; Thu, 22 Jul 93 13:45:24 +0200 Received: from majestix.iaks.ira.uka.de by deathstar.iaks.ira.uka.de (4.1/SMI-4.1) id AA20011; Thu, 22 Jul 93 13:39:09 +0200 Date: Thu, 22 Jul 93 13:39:09 +0200 From: schue@ira.uka.de (Joachim Schue) Message-Id: <9307221139.AA20011@deathstar.iaks.ira.uka.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: unsubscribe unsubscribe From P2043742@csdvax.csd.unsw.EDU.AU Thu Jul 22 14:03:51 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02991; Thu, 22 Jul 93 14:03:51 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H0UWXMAZR48WW40G@csdvax.csd.unsw.EDU.AU>; Thu, 22 Jul 1993 21:59:46 +1000 Date: Thu, 22 Jul 1993 21:59:46 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Interesting subseq problem with Garnet. To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H0UWXMBSOY8WW40G@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear CLISP users, Today I tried Garnet again (after loading opal/defs as Doug reported it seemed to work fine at least I didn't get the error *** - UNIX error 9 (EBADF).. as I reported yesterday) but of course I got the error with subseq using 0 in place of the end specifier. *** - SUBSEQ: :END should be an integer >=0, not NIL After this error I grebbed the subseq function from the CMU Common Lisp sources and compiled and loaded it. I tested it first of course and it seems to work just fine. > (subseq "Hello" 3) "lo" > (I am just testing here!!) > (subseq "Hello" 3 4) "l" > (subseq '(a b c d e f) 3 nil) (D E F) > (subseq '#(1 2 3 4 5 6) 3 nil) #(4 5 6) > (subseq "Hello" 3 0) *** - MAKE-STRING: the string length -3 should be nonnegative fixnum 1. Break> And now: > (subseq "Hello" 3 nil) "lo" > Works just fine. But then I wanted to start up Gilt and guess what happened? The Gilt Work Window, Gilt Commands and Gilt Motif Gadgets came up and I was happier than ever and then the shock came. > (gilt:do-go) *** - SUBSEQ: :END should be an integer >=0, not NIL 1. Break> I realy don't know where this error comes from but hopefully somebody out there knows about it. But anyway the sources for subseq are here. ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; (in-package 'lisp) ;;; SUBSEQ cannot default end to the length of sequence since it is not ;;; an error to supply nil for its value. We must test for end being nil ;;; in the body of the function, and this is actually done in the support ;;; routines for other reasons (see above). (defun subseq (sequence start &optional end) "Returns a copy of a subsequence of SEQUENCE starting with element number START and continuing to the end of SEQUENCE or the optional END." (seq-dispatch sequence (list-subseq* sequence start end) (vector-subseq* sequence start end))) (defun vector-subseq* (sequence start &optional end (declare (vector sequence) (fixnum start)) (when (null end) (setf end (length sequence))) (do ((old-index start (1+ old-index)) (new-index 0 (1+ new-index)) (copy (make-sequence-like sequence (- end start))) ((= old-index end) copy) (declare (fixnum old-index new-index)) (setf (aref copy new-index) (aref sequence old-index)))) (defun list-subseq* (sequence start &optional end) (declare (list sequence) (fixnum start)) (if (and end (>= start (the fixnum end))) () (let* ((groveled (nthcdr start sequence)) (result (list (car groveled)))) (if groveled (do ((list (cdr groveled) (cdr list)) (splice result (cdr (rplacd splice (list (car list)))) (index (1+ start) (1+ index))) ((or (atom list) (and end (= index (the fixnum end)))) result) (declare (fixnum index))))))) (defun make-sequence-of-type (type length) "Returns a sequence of the given TYPE and LENGTH." (declare (fixnum length)) (case (type-specifier-atom type) (list (make-list length)) ((bit-vector simple-bit-vector) (make-array length :element-type '(mod 2))) ((string simple-string base-string simple-base-string) (make-string length)) (simple-vector (make-array length)) ((array simple-array vector) (if (listp type) (make-array length :element-type (cadr type)) (make-array length))) (t (make-sequence-of-type (result-type-or-lose type) length)))) (defun result-type-or-lose (type &optional nil-ok) (cond ((subtypep type 'nil) (if nil-ok nil (error "NIL output type invalid for this sequence function."))) ((dolist (seq-type '(list bit-vector string vector) nil) (when (subtypep type seq-type) (return seq-type)))) (t (error "~S is a bad type specifier for sequence functions." type)))) ;;; Seq-Dispatch does an efficient type-dispatch on the given Sequence. (defmacro seq-dispatch (sequence list-form array-form) `(if (listp ,sequence) ,list-form ,array-form)) (defmacro elt-slice (sequences n) "Returns a list of the Nth element of each of the sequences. Used by MAP and friends." `(mapcar #'(lambda (seq) (elt seq ,n)) ,sequences)) (defmacro make-sequence-like (sequence length) "Returns a sequence of the same type as SEQUENCE and the given LENGTH. `(make-sequence-of-type (type-of ,sequence) ,length)) (defmacro type-specifier-atom (type) "Returns the broad class of which TYPE is a specific subclass." `(if (atom ,type) ,type (car ,type))) If somebody knows what's wrong now please tell me. I look forward to hearing from you. Yours, Bela. From eich@crunch.ikp.physik.th-darmstadt.de Thu Jul 22 14:13:46 1993 Return-Path: Received: from rs2.hrz.th-darmstadt.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03047; Thu, 22 Jul 93 14:13:46 +0200 Received: from crunch (crunch.ikp.physik.th-darmstadt.de) by rs2.hrz.th-darmstadt.de with SMTP id AA32133 (5.65c/IDA-1.4.4 for ); Thu, 22 Jul 1993 14:10:17 +0200 Received: by crunch (AIX 3.2/UCB 5.64/Client-1.5/HRZ-THD) id AA17068; Thu, 22 Jul 1993 14:10:17 +0200 Date: Thu, 22 Jul 1993 14:10:17 +0200 From: Egbert Eich Message-Id: <9307221210.AA17068@crunch> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: unsubscribing from this list Dear CLISP users, When I subscribed to this list I received a mail saying (among other things): To unsubscribe from the clisp-list mailing list, send mail to listserv@ma2s2.mathematik.uni-karlsruhe.de including the two words "unsubscribe clisp-list" as message body. Maybe everybody who has lost that mail should keep these lines in a save place. This could avoid flooding peoples mailboxes with useless messages. Egbert. From UNM42E@IBM.rhrz.uni-bonn.de Fri Jul 23 15:58:32 1993 Return-Path: Received: from IBM.rhrz.uni-bonn.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04750; Fri, 23 Jul 93 15:58:32 +0200 Resent-Message-Id: <9307231358.AA04750@ma2s2.mathematik.uni-karlsruhe.de> Received: from DBNRHRZ1.BITNET by IBM.rhrz.uni-bonn.de (IBM VM SMTP V2R2) with BSMTP id 2997; Fri, 23 Jul 93 15:54:10 MEZ Received: from DBNRHRZ1 (UNM42E) by DBNRHRZ1.BITNET (Mailer R2.08 R208004) with BSMTP id 3856; Fri, 23 Jul 93 15:54:09 MEZ Resent-Date: Fri, 23 Jul 93 15:53:09 MEZ Resent-From: Michael Stoll Resent-To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Received: from sfb256.iam.uni-bonn.de by IBM.rhrz.uni-bonn.de (IBM VM SMTP V2R2) with TCP; Fri, 23 Jul 93 15:51:41 MEZ Received: from rhein by sfb256.iam.uni-bonn.de (5.64/0.0) id AA25062; Fri, 23 Jul 93 16:01:08 +0200 Received: by rhein.iam.uni-bonn.de (920330.SGI/911001.SGI) for @sfb256.iam.uni-bonn.de:UNM42E@ibm.rhrz.uni-bonn.de id AA01355; Fri, 23 Jul 93 16:01:04 +0200 Date: Fri, 23 Jul 93 16:01:04 +0200 From: michael@rhein.iam.uni-bonn.de (Michael Stoll) Message-Id: <9307231401.AA01355@rhein.iam.uni-bonn.de> To: UNM42E@IBM.rhrz.uni-bonn.de Subject: Problem with subseq ----------------------------Original message---------------------------- Bela Pecsek writes: [Stuff deleted] > *** - SUBSEQ: :END should be an integer >=0, not NIL > > After this error I grebbed the subseq function from the CMU Common Lisp sources > and compiled and loaded it. I tested it first of course and it seems to work > just fine. [Tests deleted] > But then I wanted to start up Gilt and guess what happened? > The Gilt Work Window, Gilt Commands and Gilt Motif Gadgets came up and I was > happier than ever and then the shock came. > > > (gilt:do-go) > > *** - SUBSEQ: :END should be an integer >=0, not NIL > 1. Break> > > I realy don't know where this error comes from but hopefully somebody out there > knows about it. [Stuff deleted] Propbably the calls to `subseq' have been compiled inline, so that the built-in subseq function is called even if you have redifed subseq later. Try to declare subseq notinline (for example, (proclaim '(notinline subseq)) ) and recompile. Michael From haible Fri Jul 23 16:16:36 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04828; Fri, 23 Jul 93 16:16:36 +0200 Date: Fri, 23 Jul 93 16:16:36 +0200 From: haible (Bruno Haible) Message-Id: <9307231416.AA04828@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Garnet, SUBSEQ Douglas Ambrisko writes: > | *** - UNIX error 9 (EBADF): Bad file number > I just grabbed the clisp versin of garnet and built it. I hit the same > problem with the "saveinitmem" on a DEC alpha box . The problem > is that garnet assumes you do a load of all compiled modules including > opal/defs which calls "initialize-x11-values". Ouch. Open file descriptors don't survive a "saveinitmem", of course. > So a work-around that > I'm doing now is to load the "saveinitmem" version then loading opal/defs. (opal::initialize-x11-values (opal::get-full-display-name)) should work, too. Bela Pecsek writes: > *** - SUBSEQ: :END should be an integer >=0, not NIL As I noted, this will be fixed in the next release. Meanwhile, you can also use the following patch: (in-package "LISP") (proclaim '(notinline subseq)) (setf (symbol-function 'old-subseq) (symbol-function 'subseq)) (defun subseq (seq start &optional (end nil)) (if (null end) (old-subseq seq start) (old-subseq seq start end) ) ) > I was happier than ever and then the shock came. > *** - SUBSEQ: :END should be an integer >=0, not NIL > I really don't know where this error comes from but hopefully somebody > out there knows about it. You have to (proclaim '(notinline subseq)) before compiling Garnet. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de !! To unsubscribe from the clisp-list mailing list, send mail to !! !! listserv@ma2s2.mathematik.uni-karlsruhe.de !! !! including the two words "unsubscribe clisp-list" as message body. !! From P2043742@csdvax.csd.unsw.EDU.AU Tue Jul 27 09:53:50 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08060; Tue, 27 Jul 93 09:53:50 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H11NN0DYCW8WX9AC@csdvax.csd.unsw.EDU.AU>; Tue, 27 Jul 1993 17:48:33 +1000 Date: Tue, 27 Jul 1993 17:48:33 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H11NN0GWGI8WX9AC@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear CLISP users, Thanks for your help to get Garnet up on CLISP for me. After proclaiming subseq notinline, recompiling Garnet and loading opal/defs everything seems to be fine but the speed that is a bit slow on my SGI however in an accaptable range. Thanks again, I look forward to hearing from you. Yours, Bela. From reggie@muon.phys.washington.edu Thu Aug 5 05:19:48 1993 Return-Path: Received: from muon.phys.washington.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17234; Thu, 5 Aug 93 05:19:48 +0200 Received: by muon.phys.washington.edu via SMTP (920330.SGI/911001.SGI) for clisp-list@ma2s2.mathematik.uni-karlsruhe.de id AA00145; Wed, 4 Aug 93 20:13:51 -0700 Message-Id: <9308050313.AA00145@muon.phys.washington.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Clarification Please. Date: Wed, 04 Aug 1993 20:13:51 -0700 From: Reggie Perry I found this while lurking on the net in comp.compilers in the FAQ: language: Common Lisp package: CLISP version: ? parts: bytecode compiler, translator(->C), runtime, library, editor ^^^^^^^^^^^^^^^ author: Bruno Haible , Michael Stoll location: ftp pub/lisp/clisp from ma2s2.mathematik.uni-karlsruhe.de description: CLISP is a Common Lisp (CLtL1) implementation by Bruno Haible of Karlsruhe University and Michael Stoll of Munich University, both in Germany. It needs only 1.5 MB of RAM. German and English versions are available, French coming soon. Packages running in CLISP include PCL and, on Unix machines, CLX. _____________ Is there really a translator to C for clisp? Thanks for the info. Reggie ------------------- Reginald S. Perry -- University of Washington, Department of Physics FM-15 Seattle, Washington 98195 reggie@muon.phys.washington.edu (General-Mail) reggie@u.washington.edu (General-Mail) reggie@quantum.phys.washington.edu (NeXTMail) From haible Thu Aug 5 15:27:18 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17686; Thu, 5 Aug 93 15:27:18 +0200 Date: Thu, 5 Aug 93 15:27:18 +0200 From: haible (Bruno Haible) Message-Id: <9308051327.AA17686@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Clarification Please Reggie Perry writes: > I found this while lurking on the net in comp.compilers in the FAQ: > package: CLISP > parts: bytecode compiler, translator(->C), runtime, library, editor > ^^^^^^^^^^^^^^^ No, CLISP compiles to bytecode only, not to C. > Is there really a translator to C for clisp? No. AKCL is a Common Lisp system that compiles to C, then uses the machine's C compiler to compile to machine code, and AKCL can load such machine code mudules into the running image. CLiCC on the other hand can build stand-alone executables from Common Lisp programs, again using the machine's C compiler. The CLiCC README file: CLiCC is a Common Lisp to C Compiler. It generates C-executables from Common Lisp application programs. CLiCC is not a Common Lisp system. Hence it does not include any program development or debugging support. CLiCC is intended to be used as an addon to existing Common Lisp systems for generating portable applications. CLiCC supports a subset of Common Lisp + CLOS, which we call CL_0 (CommonLisp_0). CL_0 is a strict and very large subset of full Common Lisp + CLOS, without (EVAL ...) and friends. At present, CL_0 is based on CLtL1, but we are working towards CLtL2 and ANSI-CL. The target language is a subset of C. CLiCC is adaptable to gene- rate ANSI-C or K&R-C by using a compiler option. The generated C code is compilable using a conventional C compiler on the target machine, and must be linked with the CLiCC runtime library in order to generate executables. ----------------------------------------------------------------- CLiCC is available via anonymous ftp from ftp.informatik.uni-kiel.de (134.245.15.113) file: kiel/apply/clicc-0.6.1.tar.Z ----------------------------------------------------------------- Wolfgang Goerigk, wg@informatik.uni-kiel.de Ulrich Hoffman, uho@informatik.uni-kiel.de Heinz Knutzen, hk@informatik.uni-kiel.de Christian-Albrechts-Universitaet zu Kiel, Institut fuer Informatik und Praktische Mathematik Preusserstr. 1-9 D-24105 Kiel, Germany ----------------------------------------------------------------- CLiCC runs in CLISP without any modification, but you need at least 8 MB RAM. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Fri Aug 13 16:56:36 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA24897; Fri, 13 Aug 93 16:56:36 +0200 Date: Fri, 13 Aug 93 16:56:36 +0200 From: haible (Bruno Haible) Message-Id: <9308131456.AA24897@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: CLISP on MS Windows Mr Ralf Muschall told me how CLISP can be made to run under Microsoft Windows. From his information I compiled the following instructions. Can someone please verify whether these instructions are correct? Please try it and tell me about your success or failure. Installation under Microsoft Windows: ------------------------------------- CLISP also runs in the DOS box of Microsoft Windows 3.1. To achieve this, the following additional steps are necessary: 1. Get and install ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/dos/clisp-english.zip as described above. Also get ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/dos/delay.exe Then copy delay.exe d:\lib\lisp 2. Get and install ftp.uni-kl.de:/pub/pc/dos/programming/c/dpmigcc1.zip 3. Create a batch file that starts lisp: copy con c:\bat\winclisp.bat d:\lib\lisp\delay.exe 1 c:\rsx\bin\rsx.exe d:\lib\lisp\lisp.exe -M d:\lib\lisp\lispinit.mem %1 %2 %3 %4 %5 %6 %7 %8 %9 [Ctrl-Z] (The paths of course depend on your installation.) 4. Call the PIF editor and input the following: Program file name: c:\bat\winclisp.bat Program title: COMMON LISP Program parameters: Start directory: e:\lisp Screen: Text Memory requirements: requires: 500 maximum: 640 EMS memory: requires: 0 maximum: 0 XMS memory: requires: 1024 maximum: -1 Display: [as you like] Quit_closes_window: [as you like] Execution: [as you like] other_options: [as you like] (You will probably choose the directory which contains your lisp programs as start directory, instead of e:\lisp.) Save it under the name WINCLISP.PIF. 5. In the program manager, in a suitable group: Menu "File" -> "New" -> "Program", window "program properties". Input there: Description: COMMON LISP Command line: winclisp.pif Start directory: e:\lisp Key combination: Ctrl+Alt+Shift+L [as you like] Clicking with the mouse on the such created icon or pressing the key combination given above will now start CLISP. Remarks: * If you want to provide command line parameters for CLISP, set the "Program parameters" field in the PIF editor to "?". You will then always be asked to input arguments for WINCLISP.BAT. * Cut & Paste in DOS boxes (via menu "Edit" -> "Mark" resp. menu "Edit" -> "Insert") inserts an at the end. Therefore one cannot re-edit a pasted line. * But the editing facilities mentioned in CLISP.MAN work. * Bugs in Microsoft Windows and/or RSX can cause system crashes. (That's also the reason for DELAY.EXE.) Good luck! Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From simmons@cs.washington.edu Tue Aug 17 21:30:52 1993 Received: from june.cs.washington.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA26638; Tue, 17 Aug 93 21:30:52 +0200 Received: by june.cs.washington.edu (5.65b/7.1ju) id AA17328; Tue, 17 Aug 93 12:22:35 -0700 Date: Tue, 17 Aug 93 12:22:35 -0700 From: simmons@cs.washington.edu (Dave Simmons) Return-Path: Message-Id: <9308171922.AA17328@june.cs.washington.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Clisp and Microsoft Windows I am part of a group at the University of Washington (Seattle, USA) that is writing image processing software to be used for mathematics education. One of our projects is to set up an image processing environment using Microsoft Windows. We want to have Clisp running in one window, and have it send data and commands to another MS Windows application that manages a set of image windows. To do this, Clisp needs to be compiled as a Windows application so that it can call a C dynamic-link library and perhaps use the Dynamic Data Exchange functions. Has anyone tried making Clisp into a windows application (perhaps using Borland software's "EasyWin" code)? Also, is there any way to make the existing Dos executables run under Windows? Any information or advice would be appreciated. ---Dave David Simmons | CS grad student | U of W Dept. Computer Science From ae1181t@stnfor.ae.ge.com Wed Aug 18 13:53:18 1993 Return-Path: Received: from stnfor.ae.ge.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA26993; Wed, 18 Aug 93 13:53:18 +0200 Received: by stnfor.ae.ge.com (5.65/DEC-Ultrix/4.3) id AA11354; Wed, 18 Aug 1993 07:42:47 -0400 Date: Wed, 18 Aug 1993 07:42:47 -0400 From: ae1181t@stnfor.ae.ge.com (Osman F Buyukisik) Message-Id: <9308181142.AA11354@stnfor.ae.ge.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: <9308171922.AA17328@june.cs.washington.edu> (simmons@cs.washington.edu) Subject: Re: Clisp and Microsoft Windows I think it would take a lot of work. Win3.x is 16 bit and Clisp is based on 32 bits. Since now Clisp can run in a DOS window, may be you can still cut and paste (but probab. just text). Osman From blake@netcom.com Thu Aug 19 06:24:15 1993 Return-Path: Received: from netcom4.netcom.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA27316; Thu, 19 Aug 93 06:24:15 +0200 Received: by netcom4.netcom.com (5.65/SMI-4.1/Netcom) id AA04535; Wed, 18 Aug 93 21:16:33 -0700 From: blake@netcom.com (Blake McBride) Message-Id: <9308190416.AA04535@netcom4.netcom.com> Subject: Re: Clisp and Microsoft Windowsi] To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Wed, 18 Aug 93 21:16:32 PDT In-Reply-To: <9308181142.AA11354@stnfor.ae.ge.com>; from "Osman F Buyukisik" at Aug 18, 93 1:54 pm X-Mailer: ELM [version 2.3 PL11] > > I think it would take a lot of work. Win3.x is 16 bit and Clisp is based on 32 > bits. Since now Clisp can run in a DOS window, may be you can still cut and > paste (but probab. just text). > Osman > > The best way to make CLISP a windows app is by the use of Win32s. Develop a full 32 bit version under Windows NT using Microsofts 32 bit NT C compiler. When linked with the Win32s libary it'll run in true 32 bit mode under Windows 3.1! -- Blake McBride (615) 790-8521 3020 Liberty Hills Drive Franklin, TN 37064 blake@netcom.com U.S.A. From haible Sun Aug 22 19:50:25 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04260; Sun, 22 Aug 93 19:50:25 +0200 Date: Sun, 22 Aug 93 19:50:25 +0200 From: haible (Bruno Haible) Message-Id: <9308221750.AA04260@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: New CLISP version Once more, there is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . Excerpt from the change log: Important note -------------- * Changed bytecode format. All .fas files generated by previous CLISP versions are invalid and must be recompiled. User visible changes -------------------- * X3J13 vote <12> is partially implemented: Native CLOS. New package CLOS containing the macros DEFCLASS, DEFMETHOD, DEFGENERIC, GENERIC-FUNCTION, GENERIC-FLET, GENERIC-LABELS, WITH-SLOTS, WITH-ACCESSORS, the functions FIND-CLASS, (SETF FIND-CLASS), CLASS-OF, CLASS-NAME, (SETF CLASS-NAME), SLOT-VALUE, SLOT-BOUNDP, SLOT-MAKUNBOUND, SLOT-EXISTS-P, CALL-NEXT-METHOD, NEXT-METHOD-P, NO-APPLICABLE-METHOD, NO-PRIMARY-METHOD, NO-NEXT-METHOD, FIND-METHOD, ADD-METHOD, REMOVE-METHOD, COMPUTE-APPLICABLE-METHODS, METHOD-QUALIFIERS, FUNCTION-KEYWORDS, SLOT-MISSING, SLOT-UNBOUND, PRINT-OBJECT, DESCRIBE-OBJECT, MAKE-INSTANCE, INITIALIZE-INSTANCE, REINITIALIZE-INSTANCE, SHARED-INITIALIZE, the classes STANDARD-CLASS, STRUCTURE-CLASS, BUILT-IN-CLASS, STANDARD-OBJECT, STANDARD-GENERIC-FUNCTION, STANDARD-METHOD and all predefined classes, and the method combination STANDARD. See impnotes.txt for details. X3J13 vote <13> is implemented, too. * X3J13 vote <63> is implemented: DESCRIBE receives the output stream as optional argument. New function CLOS:DESCRIBE-OBJECT. * X3J13 vote <149> is implemented: The second argument to SUBSEQ may now be NIL. Thanks to Thomas Kirk. * X3J13 vote <175> is implemented: TAILP now also works on dotted lists. * X3J13 vote <124> is implemented: (DECLARE (OPTIMIZE (DEBUG ...))) is legal. * The type specifier (EQL object), a synonym for (MEMBER object), is now implemented. * New DEFSTRUCT option :INHERIT which is exactly like :INCLUDE except that it doesn't create new accessors for the inherited slots. Use this option to avoid the problems that occur when using the same :CONC-NAME for the new and the inherited structure. * Fixed EXPORT and USE-PACKAGE: Don't signal a name conflict between the symbol being exported and a symbol accessible in the using package if the latter symbol is on the list of shadowing symbols of the using package. * Fixed VECTOR-PUSH-EXTEND. It now works on vectors of dimension 0 and on bit-vectors. * Fixed PRINT-UNREADABLE-OBJECT: Don't output adjacent spaces. * Fixed a bug in the reader that caused an endless loop or endless recursion when resolving #n=/#n# references in objects that already contained cyclic data structures, e.g. when reading (#1=a #.'#2=(b . #2#) #1#). * Fixed a bogus error message in the complex number reader. Portability ----------- * The DOS README explains how to run CLISP in the DOS box of Microsoft Windows. Thanks to Ralf Muschall. * Added Sinix support. Thanks to Manfred Weichel. * Improved installation on OS/2. Other modifications ------------------- * A function proclaimed INLINE will be taken inline in a wider set of circumstances. (Previously INLINE worked only when that function was being used in the same file or in a file that REQUIREd the file defining the function.) * The compiler now produces better code for (APPLY (FUNCTION fun) args). * Changed the default memory size from 2 megabytes to 2.5 megabytes. * Miscellaneous documentation updates. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From reggie@muon.phys.washington.edu Sun Aug 22 23:29:16 1993 Return-Path: Received: from muon.phys.washington.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05606; Sun, 22 Aug 93 23:29:16 +0200 Received: by muon.phys.washington.edu via SMTP (920330.SGI/911001.SGI) for clisp-list@ma2s2.mathematik.uni-karlsruhe.de id AA11238; Sun, 22 Aug 93 14:20:40 -0700 Message-Id: <9308222120.AA11238@muon.phys.washington.edu> To: clisp-list Subject: Re: New CLISP version In-Reply-To: Your message of "Sun, 22 Aug 1993 19:52:15 +0200." <9308221750.AA04260@ma2s2.mathematik.uni-karlsruhe.de> Date: Sun, 22 Aug 1993 14:20:39 -0700 From: Reggie Perry Hi there. I am compiling clisp on an SGI Indigo running IRIX 4.0.5F. I am having problems compiling your latest clisp. here is the problem lisparit0.d:19231: warning: integer overflow in expression lisparit0.d:19231: warning: statement with no effect lisparit0.d:19231: warning: statement with no effect lisparit0.d:19234: warning: address of register variable `ziffern' requested libmld--as0: Error: lisparit0.d, line 20723: cannot write cur table I am using gcc 2.4.5 . Thanks for your help. ------------------- Reginald S. Perry -- University of Washington, Department of Physics FM-15 Seattle, Washington 98195 reggie@muon.phys.washington.edu (General-Mail) reggie@u.washington.edu (General-Mail) reggie@quantum.phys.washington.edu (NeXTMail) From haible Sun Aug 22 23:43:32 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05702; Sun, 22 Aug 93 23:43:32 +0200 Date: Sun, 22 Aug 93 23:43:32 +0200 From: haible (Bruno Haible) Message-Id: <9308222143.AA05702@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Reginald S. Perry writes: > I am having problems compiling your latest clisp. here is the problem > > lisparit0.d:19234: warning: address of register variable `ziffern' requested > libmld--as0: Error: lisparit0.d, line 20723: cannot write cur table The warnings are OK. The error probably comes from the fact that your temporary directory ($TMPDIR or /usr/tmp or /tmp) is full. Temporary files for lisparit0.d may well be one or two megabytes in size. Bruno Haible From matthias@intellektik.informatik.th-darmstadt.de Mon Aug 23 18:08:40 1993 Return-Path: Received: from rs2.hrz.th-darmstadt.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06310; Mon, 23 Aug 93 18:08:40 +0200 Received: from aida.intellektik.informatik.th-darmstadt.de by rs2.hrz.th-darmstadt.de with SMTP id AA25009 (5.65c/IDA-1.4.4 for ); Mon, 23 Aug 1993 17:59:56 +0200 Received: by aida.intellektik.informatik.th-darmstadt.de (4.1/Server-1.3/HRZ-THD) id AA02642; Mon, 23 Aug 93 17:59:51 +0200 Date: Mon, 23 Aug 93 17:59:51 +0200 From: Matthias Lindner Message-Id: <9308231559.AA02642@aida.intellektik.informatik.th-darmstadt.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Clisp and Microsoft Windows > > > > I think it would take a lot of work. Win3.x is 16 bit and Clisp is based on 32 > > bits. Since now Clisp can run in a DOS window, may be you can still cut and > > paste (but probab. just text). > > Osman > > > > > > The best way to make CLISP a windows app is by the use of Win32s. > Develop a full 32 bit version under Windows NT using Microsofts > 32 bit NT C compiler. When linked with the Win32s libary it'll > run in true 32 bit mode under Windows 3.1! > Hi, did someone out there really do this - or is ist just an idea how it might work? If there are versions of CLISP floating around, that run as real win3.x-applications (not in a DOS-window!) please let me know. Thanx --Matthias ------------------------------------------------------------------------------ Matthias Lindner FG Intellektik, FB Informatik Technische Hochschule Darmstadt Alexanderstr.10, D-64283 Darmstadt TEL: +49 6151 166651 FAX: +49 6151 165326 NET: matthias@intellektik.informatik.th-darmstadt.de ------------------------------------------------------------------------------ From matthias@intellektik.informatik.th-darmstadt.de Tue Aug 24 14:49:08 1993 Return-Path: Received: from rs2.hrz.th-darmstadt.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07017; Tue, 24 Aug 93 14:49:08 +0200 Received: from aida.intellektik.informatik.th-darmstadt.de by rs2.hrz.th-darmstadt.de with SMTP id AA38376 (5.65c/IDA-1.4.4 for ); Tue, 24 Aug 1993 14:40:15 +0200 Received: by aida.intellektik.informatik.th-darmstadt.de (4.1/Server-1.3/HRZ-THD) id AA05026; Tue, 24 Aug 93 14:40:10 +0200 Date: Tue, 24 Aug 93 14:40:10 +0200 From: Matthias Lindner Message-Id: <9308241240.AA05026@aida.intellektik.informatik.th-darmstadt.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Hi, yesterday I fetched and compiled the new version of CLISP. Everything worked fine, except for the make-instance function in the new CLOS package. I think the built-in C-function sys::%make-instance does not call initialize-instance on the newly created object. The wrapping lisp-function clos:make-instance does not accept any keyword arguments, that are not defined as :initarg-options to slot definitions in clos:defclass. Is this a bug, or did I misinstall something? --Matthias From haible Tue Aug 24 15:20:35 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07145; Tue, 24 Aug 93 15:20:35 +0200 Date: Tue, 24 Aug 93 15:20:35 +0200 From: haible (Bruno Haible) Message-Id: <9308241320.AA07145@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list, haible@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Matthias Lindner writes > I think the built-in C-function sys::%make-instance does not call > initialize-instance on the newly created object. The wrapping lisp-function > clos:make-instance does not accept any keyword arguments, that are not > defined as :initarg-options to slot definitions in clos:defclass. Yes, this is the current behaviour. I noted in impnotes.txt: MAKE-INSTANCE does not call INITIALIZE-INSTANCE and SHARED-INITIALIZE, but does everything itself. Defining methods on INITIALIZE-INSTANCE or SHARED-INITIALIZE therefore does not modify the behaviour of MAKE-INSTANCE. > Is this a bug, or did I misinstall something? You may call it a bug. I chose this approach to reduce the costs of calling MAKE-INSTANCE. For now, it's 1 generic function call and dispatch and 2 built-in function calls. If MAKE-INSTANCE is required to go through INITIALIZE-INSTANCE and SHARED-INITIALIZE, the costs will be at least 3 GF calls and about 10 other function calls. Unless I do some ugly hacks, modifying #'MAKE-INSTANCE every time a method is added to #'INITIALIZE-INSTANCE or #'SHARED-INITIALIZE. What is common practice in CLOS programs? Are methods for INITIALIZE-INSTANCE frequent or not? What about methods for SHARED-INITIALIZE ? Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From matthias@intellektik.informatik.th-darmstadt.de Tue Aug 24 17:00:32 1993 Return-Path: Received: from rs2.hrz.th-darmstadt.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07620; Tue, 24 Aug 93 17:00:32 +0200 Received: from aida.intellektik.informatik.th-darmstadt.de by rs2.hrz.th-darmstadt.de with SMTP id AA29928 (5.65c/IDA-1.4.4 for ); Tue, 24 Aug 1993 16:51:38 +0200 Received: by aida.intellektik.informatik.th-darmstadt.de (4.1/Server-1.3/HRZ-THD) id AA05498; Tue, 24 Aug 93 16:51:32 +0200 Date: Tue, 24 Aug 93 16:51:32 +0200 From: Matthias Lindner Message-Id: <9308241451.AA05498@aida.intellektik.informatik.th-darmstadt.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Bruno Haible wrote: > > You may call it a bug. I chose this approach to reduce the costs of calling > MAKE-INSTANCE. For now, it's 1 generic function call and dispatch and 2 > built-in function calls. If MAKE-INSTANCE is required to go through > INITIALIZE-INSTANCE and SHARED-INITIALIZE, the costs will be at least 3 GF > calls and about 10 other function calls. Unless I do some ugly hacks, > modifying #'MAKE-INSTANCE every time a method is added to > #'INITIALIZE-INSTANCE or #'SHARED-INITIALIZE. > > What is common practice in CLOS programs? Are methods for INITIALIZE-INSTANCE > frequent or not? What about methods for SHARED-INITIALIZE ? > > My experience in CLOS-programming is not very extensive, however many programs, I saw used INITIALIZE-INSTANCE to do most of the work. I use CLOS (and used PCL before) to build a simple window system based on the StdWin interface of CLISP (which I've finished). I think it is very desirable, to be able to do something like (make-instance 'menu-window :title "Lisps" :items (list "CLISP" "LUCID" "ALLEGRO")) to create a pop-up-menu. Of course it is possible, to write a wrapping function MAKE-MENU-WINDOW, that explicitly calls initialize-instance, but why should this be necessary? CLtL2 (sorry I don't have CLtL1) says on page 847: "The generic function initialize-instance is called by make-instance to initialize a newly created instance." The same can be found in the PCL documentation "Functions in the Programmer Interface" page 2-57. I think ist's not a good idea to pay with incompatibilty for sparing a few function calls. However - that's *my* opinion - of course. Anyway - after changing MAKE-INSTANCE - everything worked fine. The performance of the new CLOS is very impressing - compared to PCL. Pretty good work - thanks! --Matthias ------------------------------------------------------------------------------ Matthias Lindner FG Intellektik, FB Informatik Technische Hochschule Darmstadt Alexanderstr.10, D-64283 Darmstadt TEL: +49 6151 166651 FAX: +49 6151 165326 NET: matthias@intellektik.informatik.th-darmstadt.de ------------------------------------------------------------------------------ From egdorf@zaphod.lanl.gov Tue Aug 24 18:37:03 1993 Return-Path: Received: from mailhost.lanl.gov by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07737; Tue, 24 Aug 93 18:37:03 +0200 Received: from zaphod.lanl.gov by mailhost.lanl.gov (5.65/1.14) id AA02142; Tue, 24 Aug 93 10:27:13 -0600 Received: by zaphod.lanl.gov.a7-mimsy (4.1/SMI-4.1) id AA01529; Tue, 24 Aug 93 10:28:11 MDT Date: Tue, 24 Aug 93 10:28:11 MDT From: egdorf@zaphod.lanl.gov (Skip Egdorf) Message-Id: <9308241628.AA01529@zaphod.lanl.gov.a7-mimsy> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: Bruno Haible's message of Tue, 24 Aug 93 15:21:18 +0200 <9308241320.AA07145@ma2s2.mathematik.uni-karlsruhe.de> Subject: RE: New CLISP version (and make-instance) You may call it a bug. I chose this approach to reduce the costs of calling MAKE-INSTANCE. For now, it's 1 generic function call and dispatch and 2 built-in function calls. If MAKE-INSTANCE is required to go through INITIALIZE-INSTANCE and SHARED-INITIALIZE, the costs will be at least 3 GF calls and about 10 other function calls. Unless I do some ugly hacks, modifying #'MAKE-INSTANCE every time a method is added to #'INITIALIZE-INSTANCE or #'SHARED-INITIALIZE. What is common practice in CLOS programs? Are methods for INITIALIZE-INSTANCE frequent or not? What about methods for SHARED-INITIALIZE ? One data point: Yes, frequent to both! If make-instance does not call initialize-instance as described in the spec, it will not work for ANY CLOS programs in my current suite. Modifying the behaviour of initialize-instance and shared-initialize is a very basic mechanism used throughout every CLOS program of which I am aware. Sorry Bruno, that is probably not the answer you wanted to hear... Skip Egdorf hwe@lanl.gov From lrw@sdsc.edu Tue Aug 24 23:52:16 1993 Return-Path: Received: from mailserver.sdsc.edu (rosebud.sdsc.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07880; Tue, 24 Aug 93 23:52:16 +0200 Received: from toastman.sdsc.edu by mailserver.sdsc.edu (4.1/4.8) id AA23109; Tue, 24 Aug 93 14:43:45 PDT Received: by toastman.sdsc.edu (5.65/DEC-OSF1/4.3-sdsc) id AA03779; Tue, 24 Aug 1993 14:43:13 -0700 Message-Id: <9308242143.AA03779@toastman.sdsc.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de (clisp-list) Subject: Re: New CLISP version (and make-instance) In-Reply-To: Your message of "Tue, 24 Aug 93 18:38:39 +0200." <9308241628.AA01529@zaphod.lanl.gov.a7-mimsy> Date: Tue, 24 Aug 93 14:43:13 +28416 From: lrw@sdsc.edu X-Mts: smtp > What is common practice in CLOS programs? Are methods for >INITIALIZE-INSTANCE frequent or not? What about methods for SHARED-INITIALIZE ? I am just a CLOS beginner but I have used one or both of those methods in every CLOS class I have written. Len Wanger San Diego Supercomputer Center. From loewenst@paul.rutgers.edu Wed Aug 25 00:16:28 1993 Return-Path: Received: from att.att.com (att-out.att.com) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07938; Wed, 25 Aug 93 00:16:28 +0200 From: David.Loewenstern@att.com (David Loewenstern, Bell Labs, 201-386-6516) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Received: by cbfsb.cb.att.com (4.1/EMS main.cf 1.23 3/29/93 (SMI-4.1)) id AA04197; Tue, 24 Aug 93 10:54:40 EDT Received: from hogpa.ho.att.com by cbfsb (4.1/EMS main.cf 1.23 3/29/93 (SMI-4.1)) id AA04180; Tue, 24 Aug 93 10:54:01 EDT Received: from hogpb.ho.att.com by hogpa.ho.att.com (4.1/SMI-4.0) id AA28790; Tue, 24 Aug 93 10:53:53 EDT Received: by hogpb.ho.att.com using UUCP (4.1/6.0c-FWP); id AA16498; Tue, 24 Aug 93 10:53:52 EDT Received: from paul.rutgers.edu by att.att.com; Tue, 24 Aug 93 16:42 EDT Date: Tue, 24 Aug 93 10:53:52 EDT Content-Length: 876 Content-Type: text Message-Id: <9308241453.AA16498@hogpb.ho.att.com> Original-From: att.com!David.Loewenstern (David Loewenstern, Bell Labs, 201-386-6516) Original-To: ma2s2.mathematik.uni-karlsruhe.de!clisp-list In-Reply-To: Bruno Haible's message of Tue, 24 Aug 93 15:21:18 +0200 <9308241320.AA07145@ma2s2.mathematik.uni-karlsruhe.de> Subject: Re: New CLISP version Reply-To: loewenst@paul.rutgers.edu (David Loewenstern) ... On Tue, 24 Aug 93 15:21:18 +0200, haible@ma2s2.mathematik.uni-karlsruhe.de (Bruno Haible) said: } What is common practice in CLOS programs? Are methods for INITIALIZE-INSTANCE } frequent or not? What about methods for SHARED-INITIALIZE ? I've written methods for both INITIALIZE-INSTANCE and SHARED-INITIALIZE, to derive default slot values from other slot values. I haven't had a situation in which I couldn't have easily switched from INITIALIZE-INSTANCE to SHARED-INITIALIZE or vice versa. Perhaps you could make the compliant version standard, and provide the non-compliant .fas file for people who needed the speed? (Or you could get really fancy and add a non-compliant-class metaclass.) These opinions are shareware. If you like the product, please send your $0.02 to David Loewenstern From michael@ifki50.informatik.fh-muenchen.de Wed Aug 25 08:18:58 1993 Return-Path: Received: from ifki50.informatik.fh-muenchen.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08386; Wed, 25 Aug 93 08:18:58 +0200 Received: by ifki50.informatik.fh-muenchen.de (AIX 3.2/UCB 5.64/4.03) id AA22978; Wed, 25 Aug 1993 08:09:48 +0200 Date: Wed, 25 Aug 1993 08:09:48 +0200 From: michael@ifki50.informatik.fh-muenchen.de Message-Id: <9308250609.AA22978@ifki50.informatik.fh-muenchen.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Port fuer RS6000 ? Ich habe versuht Clisp auf die RS6000 zu portieren. Ich komme aber nicht weiter als bis zum Interpreter. Der Compiler läuft nicht, sonern produziert nur ser viele Meldungen. Mache ich Arbeit doppelt? Giebt es eine Quellcodedukomentation Vielen Dank Michael Kirstein From haible Thu Aug 26 12:23:51 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA09901; Thu, 26 Aug 93 12:23:51 +0200 Date: Thu, 26 Aug 93 12:23:51 +0200 From: haible (Bruno Haible) Message-Id: <9308261023.AA09901@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version (and make-instance) OK, enough feedback. I thank you all! I asked ! What is common practice in CLOS programs? Are methods for INITIALIZE-INSTANCE ! frequent or not? What about methods for SHARED-INITIALIZE ? Matthias Lindner and Joerg Hoehle pointed out that INITIALIZE-INSTANCE methods are very frequent. Skip Egdorf, Len Wanger and David Loewenstern mentioned SHARED-INITIALIZE as well. So, I have rewritten MAKE-INSTANCE so that it behaves as if it were calling INITIALIZE-INSTANCE, and I'll probably do the same for SHARED-INITIALIZE. David Loewenstern: > Perhaps you could make the compliant version standard, and provide > the non-compliant .fas file for people who needed the speed? No, the new MAKE-INSTANCE is even faster than the old one. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From deleeuw@stat.ucla.edu Sat Aug 28 10:39:02 1993 Received: from julia.math.ucla.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA11327; Sat, 28 Aug 93 10:39:02 +0200 Received: from laplace.stat.ucla.edu by julia.math.ucla.edu via SMTP (Sendmail 5.61/1.07) id AA26180; Sat, 28 Aug 93 01:30:32 -0700 Return-Path: Received: by laplace.stat.ucla.edu (Sendmail 4.1/1.08) id AA00210; Sat, 28 Aug 93 01:23:57 PDT Date: Sat, 28 Aug 93 01:23:57 PDT From: Jan Deleeuw Message-Id: <9308280823.AA00210@laplace.stat.ucla.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: clue problem Hello, I setup clisp, clx, pcl, garnet, maxima without problems. They all seem to function well. But clue gets me into trouble. I have directories foo/clx foo/pcl foo/clue and foo/clio, and both clx.mem and pcl.mem are sitting confortable in their directories. I then say make -f Makefile.clisp in the clue directory, and things start to compile and load, but then screech to a halt with *** - FIND-CLASS: WINDOW does not name a class when compiling intrinsics.lsp. My system is SunOS 4.1.3 on a Sun 470 server, but I guess that is sort of irrelevant since I only use what comes with clisp. It is a known problem, with a known solution, or should I become more specific. --- Jan From haible Sat Aug 28 15:45:43 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA11510; Sat, 28 Aug 93 15:45:43 +0200 Date: Sat, 28 Aug 93 15:45:43 +0200 From: haible (Bruno Haible) Message-Id: <9308281345.AA11510@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: clue problem Jan Deleeuw has a problem compiling CLUE: > *** - FIND-CLASS: WINDOW does not name a class > > when compiling intrinsics.lsp. CLX can define its types using DEFSTRUCT or DEFCLASS. You won't notice a difference when running only CLX. But CLUE requires that CLX has used DEFCLASS. In the pcl+clx.sept92f.clisp distribution, CLX uses DEFCLASS only if PCL is present. So you need to rebuild clx.mem, starting from pcl.mem. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From deleeuw@stat.ucla.edu Tue Aug 31 05:48:28 1993 Received: from julia.math.ucla.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16988; Tue, 31 Aug 93 05:48:28 +0200 Received: from laplace.stat.ucla.edu by julia.math.ucla.edu via SMTP (Sendmail 5.61/1.07) id AA05336; Mon, 30 Aug 93 20:39:34 -0700 Return-Path: Received: by laplace.stat.ucla.edu (Sendmail 4.1/1.08) id AA04493; Mon, 30 Aug 93 20:32:45 PDT From: Jan Deleeuw Message-Id: <9308310332.AA04493@laplace.stat.ucla.edu> Subject: batch mode To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Mon, 30 Aug 1993 20:32:44 -0700 (PDT) X-Mailer: ELM [version 2.4 PL20] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 683 + Jan Deleeuw has a problem compiling CLUE: + + > *** - FIND-CLASS: WINDOW does not name a class + > + > when compiling intrinsics.lsp. + + CLX can define its types using DEFSTRUCT or DEFCLASS. You won't notice a + difference when running only CLX. But CLUE requires that CLX has used + DEFCLASS. In the pcl+clx.sept92f.clisp distribution, CLX uses DEFCLASS + only if PCL is present. So you need to rebuild clx.mem, starting from + pcl.mem. Sorry to be so dense, but I do build clx.mem with the Makefile.clisp that comes with the clx distribution, and that uses clisp -M ../pcl/pcl.mem when doing compile-clx. Isn't this "starting from pcl.mem" ? --- Jan From P2043742@csdvax.csd.unsw.EDU.AU Wed Sep 1 05:58:53 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17753; Wed, 1 Sep 93 05:58:53 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H2FQ4DG6CW91WY1G@csdvax.csd.unsw.EDU.AU>; Wed, 1 Sep 1993 13:58:32 +1000 Date: Wed, 01 Sep 1993 13:58:32 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Clisp and Garnet-2.2 To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H2FQ4DJE3M91WY1G@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Clisp Users, I am using Garnet-2.1 on SGI and I have just realised that Garnet-2.2 is available. Before spending time to bild it myself (I have restricted time and disk space!) I would like to ask some question about it. Has anybody been able to bild it with Clisp? Is it really 2-3 times faster then Garnet-2.1? Thanks for your repply in advanced. Yours, Bela Pecsek From rbfeens@cs.vu.nl Wed Sep 1 10:34:01 1993 Return-Path: Received: from top.cs.vu.nl by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18029; Wed, 1 Sep 93 10:34:01 +0200 Received: from fluit.cs.vu.nl by top.cs.vu.nl id aa07399; 1 Sep 93 10:33 MET DST Date: Wed, 1 Sep 93 10:33:26 MET DST From: RB Feenstra To: clisp-list Subject: Re: Clisp and Garnet-2.2 Message-Id: <9309011033.aa12364@fluit.cs.vu.nl> Dear CLISP users, Today I tried to compile Garnet version 2.2 beta, which, according to its documentation, has CLISP support. Unfortunately, when following the instructions in src/README.CLISP, the compilation halts with a message indicating a compiler bug: > (load "garnet-compiler.lsp") ;; Loading file garnet-compiler.lsp ... %%%%%%%%%%%%%% Compiling Utils %%%%%%%%%%%%%%% ;; Loading file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/utils/utils-compiler.lsp ... Compiling file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/utils/general.lsp ... WARNING in function PROBE-DIRECTORY-1 : variable ARGS is not used. Misspelled or missing IGNORE declaration? Compilation of file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/utils/general.lsp is finished. 0 errors, 1 warning ;; Loading file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/utils/general.fas ... ;; Loading of file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/utils/general.fas is finished. ;; Loading of file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/utils/utils-compiler.lsp is finished. %%%%%%%%%%%%%% Compiling KR %%%%%%%%%%%%%%% ;; Loading file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/kr/kr-compiler.lsp ... Compiling file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/kr/kr-macros.lsp ... Compilation of file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/kr/kr-macros.lsp is finished. 0 errors, 0 warnings The following functions were used but not defined: PRINT-THE-SCHEMA PRINT-THE-SLOT TYPE-TO-FN SCHEMA-P VALUE-FN S-VALUE-FN PROCESS-SLOTS CREATION-MESSAGE ;; Loading file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/kr/kr-macros.fas ... ;; Loading of file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/kr/kr-macros.fas is finished. Compiling file /tmp_mnt/net/tmp/tmp/rbfeens/garnet/src/kr/kr.lsp ... *** - Compiler bug!! Occurred in OPTIMIZE-LABEL. 1. Break> where EVAL frame for form (COMPILE-FILE (MERGE-PATHNAMES USER::FILE USER::GARNET-KR-SRC)) 1. Break> I'm running a CLISP with (machine-version) equal to "SUN4C", (lisp-implementation-version) is "June 1993". I didn't have encounter this problem with garnet 2.1. Any suggestions of what may be wrong? Remco Feenstra (rbfeens@cs.vu.nl) From haible Wed Sep 1 16:05:32 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA22528; Wed, 1 Sep 93 16:05:32 +0200 Date: Wed, 1 Sep 93 16:05:32 +0200 From: haible (Bruno Haible) Message-Id: <9309011405.AA22528@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: New CLISP version There is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . It fixes the shortcomings/bugs regarding the CLOS of the previous version. The compiler bug uncovered by Garnet 2.2 isn't fixed yet, I haven't had time to look at it. >From now on, CLISP will also be available via anonymous ftp from Mark Kantrowitz' Lisp repository, host ftp.cs.cmu.edu, directory user/ai/lang/lisp/impl/clisp/. Please save FTP bandwidth by using this ftp server if you are in USA. Excerpt from the change log: Important note -------------- * Changed bytecode format. All .fas files generated by previous CLISP versions are invalid and must be recompiled. User visible changes -------------------- * Unix version only: The command line option -M now also accepts memory images that were compressed using GNU gzip. * Fixed a CLOS bug: MAKE-INSTANCE now behaves as if it were calling INITIALIZE-INSTANCE, and INITIALIZE-INSTANCE and REINITIALIZE-INSTANCE behave as if they were calling SHARED-INITIALIZE. * Fixed a CLOS bug: NO-APPLICABLE-METHOD failed to be called in some cases when no method was applicable. * Relating to X3J13 vote <89>: The first argument to the functions FUNCALL and APPLY, denoting a function, may now also be a list (SETF symbol). * Fixed a bogus error message that occurred when storing an out-of-range integer into an array of element-type (UNSIGNED-BYTE 32). Portability ----------- * Added NetBSD support. Thanks to Charles Hannum. Other modifications ------------------- * Added Jeff Dalton's "brief CLOS guide". * Miscellaneous documentation updates. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From P2043742@csdvax.csd.unsw.EDU.AU Fri Sep 3 06:16:40 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA25815; Fri, 3 Sep 93 06:16:40 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H2IJACZOU891X4LN@csdvax.csd.unsw.EDU.AU>; Fri, 3 Sep 1993 14:16:01 +1000 Date: Fri, 03 Sep 1993 14:16:01 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: String to List coerceing To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H2IJAD0HS291X4LN@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Clisp Users, I have a seemingly trivial question I do not know the answere for. I'd like to read a line from a file and coerce it to list. Say, I have a file with a line: 100 1.3 4.2 5.8 I'd like the line to be converted to (100 1.3 4.2 5.8) when I read it in. What is the simplest way to do it? Thanks for your help in addvance. Yours, Bela Pecsek From loewenst@paul.rutgers.edu Fri Sep 3 07:56:40 1993 Return-Path: Received: from paul.rutgers.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA25906; Fri, 3 Sep 93 07:56:40 +0200 Received: by paul.rutgers.edu (5.59/SMI4.0/RU1.5/3.08) id AA27062; Fri, 3 Sep 93 01:56:03 EDT Date: Fri, 3 Sep 93 01:56:03 EDT From: loewenst@paul.rutgers.edu (David Loewenstern) Message-Id: <9309030556.AA27062@paul.rutgers.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: P2043742@csdvax.csd.unsw.edu.au In-Reply-To: <01H2IJAD0HS291X4LN@csdvax.csd.unsw.EDU.AU> "P2043742@csdvax.csd.unsw.edu.au" Subject: Re: String to List coerceing Reply-To: loewenst@paul.rutgers.edu (David Loewenstern) ... On 3 Sep 93, P2043742 (whoever that is) said: } I have a seemingly trivial question I do not know the answere for. } I'd like to read a line from a file and coerce it to list. } Say, I have a file with a line: } 100 1.3 4.2 5.8 } I'd like the line to be converted to (100 1.3 4.2 5.8) when I read it in. } What is the simplest way to do it? } Thanks for your help in addvance. ;;;I'll ignore the question of detecting the end-of-file. ;;;If you know how many numbers are on each line: (DEFUN Read-Into-List (filestream length) (LOOP repeat length collect (READ filestream))) ;;; ... it's pretty easy to do this without the LOOP macro, too. ;;;If you don't: (DEFUN Readline-Into-List (filestream) (READ-FROM-STRING (CONCATENATE 'STRING "(" (READ-LINE filestream) ")"))) These opinions are shareware. If you like the product, please send your $0.02 to David Loewenstern || From deleeuw@stat.ucla.edu Fri Sep 3 09:00:10 1993 Received: from julia.math.ucla.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA25982; Fri, 3 Sep 93 09:00:10 +0200 Received: from laplace.stat.ucla.edu by julia.math.ucla.edu (Sendmail 4.1/1.07) id AA28028; Fri, 3 Sep 93 00:00:30 PDT Return-Path: Received: by laplace.stat.ucla.edu (Sendmail 4.1/1.08) id AA13134; Thu, 2 Sep 93 23:53:26 PDT From: Jan Deleeuw Message-Id: <9309030653.AA13134@laplace.stat.ucla.edu> Subject: batch mode To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Thu, 2 Sep 1993 23:53:25 -0700 (PDT) X-Mailer: ELM [version 2.4 PL20] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 586 I have the following problem (Sun Sparc Server 470, SunOS 4.1.3). I have compiled pcl and clx from the sept92 clisp version, which means that the clisp used to compile clx loads pcl.mem. In the dumped clx, I can do > pcl::*defclass-times* (COMPILE LOAD EVAL) > xlib::*def-clx-class-use-defclass* (XLIB:WINDOW XLIB:DRAWABLE XLIB:PIXMAP) Now I want to use the dumped clx, which contains pcl, to compile clue. This fails ; Compiling intrinsics.lsp Compiling file /m1/lisp/clisp/packages/xit/clue/intrinsics.lsp ... *** - FIND-CLASS: WINDOW does not name a class Any ideas ? --- Jan From pet@prip.tuwien.ac.at Sun Sep 5 15:36:06 1993 Return-Path: Received: from email.tuwien.ac.at by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00503; Sun, 5 Sep 93 15:36:06 +0200 Received: from prip5.prip.tuwien.ac.at by email.tuwien.ac.at with SMTP (PP) id <15953-0@email.tuwien.ac.at>; Sun, 5 Sep 1993 15:34:56 +0200 Received: from prip9.prip.tuwien.ac.at.tuwien.ac.at by prip.tuwien.ac.at (4.1/SMI-4.1) id AA00193; Sun, 5 Sep 93 15:35:09 +0200 Date: Sun, 5 Sep 93 15:35:09 +0200 From: pet@prip.tuwien.ac.at (Johann Petrak) Message-Id: <9309051335.AA00193@prip.tuwien.ac.at> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Has anyone compiled the new CLISP version for OS/2 ???????? I am unable to do that myself, I don't have the disk space... From haible Tue Sep 7 13:13:08 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02147; Tue, 7 Sep 93 13:13:08 +0200 Date: Tue, 7 Sep 93 13:13:08 +0200 From: haible (Bruno Haible) Message-Id: <9309071113.AA02147@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: Re: Clisp and Garnet-2.2 Remco Feenstra observed a CLISP compiler bug when trying to compile Garnet-2.2. Here is a patch for it. After applying it, recompile compiler.lsp and rebuild lispinit.mem as described in the README file. *** clisp-1993-09-01/src/compiler.lsp Sun Aug 29 22:00:23 1993 --- clisp/src/compiler.lsp Tue Sep 7 00:56:57 1993 *************** *** 9952,9958 **** (optimize-part code) ; weitere mögliche Optimierung ; weitere mögliche Optimierungen: (optimize-label false-label) ; wegen verringerter Referenzen ! (optimize-short index code) ; wegen obigem optimize-part ) (FALSE ; Referenz auf true-label streichen --- 9952,9958 ---- (optimize-part code) ; weitere mögliche Optimierung ; weitere mögliche Optimierungen: (optimize-label false-label) ; wegen verringerter Referenzen ! (optimize-short index) ; wegen obigem optimize-part ) (FALSE ; Referenz auf true-label streichen *************** *** 9967,9973 **** (optimize-part code) ; weitere mögliche Optimierung ; weitere mögliche Optimierungen: (optimize-label true-label) ; wegen verringerter Referenzen ! (optimize-short index code) ; wegen obigem optimize-part ) ) ) ) (JMP (let ((label (second item))) --- 9967,9973 ---- (optimize-part code) ; weitere mögliche Optimierung ; weitere mögliche Optimierungen: (optimize-label true-label) ; wegen verringerter Referenzen ! (optimize-short index) ; wegen obigem optimize-part ) ) ) ) (JMP (let ((label (second item))) Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From ckclark@MIT.EDU Tue Sep 7 21:12:30 1993 Return-Path: Received: from MIT.EDU (MIT.MIT.EDU) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03275; Tue, 7 Sep 93 21:12:30 +0200 Received: from M16-034-18.MIT.EDU by MIT.EDU with SMTP id AA27012; Tue, 7 Sep 93 15:09:59 EDT Date: Tue, 7 Sep 93 15:09:59 EDT Message-Id: <9309071909.AA27012@MIT.EDU> From: Calvin Clark To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: What does "No more room for LISP objects" mean? This is probably a stupid question, but bear with me; I'm new to Common Lisp. Sometimes, when compiling large packages like garnet, I get the error: No more room for LISP objects and I'm not sure what this means. What boundary am I hitting? This occurs even when the clisp process is not taking up a lot of memory compared to the real memory of the system. If this error indicates that the process is hitting a system resource limit, which one is it? If not, is there some way I can modify the source so that this won't happen? (BTW, I've only seen this error on DECstation/ULTRIX, not on Sparc/Solaris.) A (possibly) related question: can someone explain the output of the clisp ``room'' function, which seems to be still CLtL1, since it takes no arguments? -Calvin From haible Tue Sep 7 22:07:01 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03337; Tue, 7 Sep 93 22:07:01 +0200 Date: Tue, 7 Sep 93 22:07:01 +0200 From: haible (Bruno Haible) Message-Id: <9309072007.AA03337@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: What does "No more room for LISP objects" mean? Calvin Clark asks: > No more room for LISP objects > What boundary am I hitting? Some versions of clisp allocate their memory at startup time. The amount of memory can be specified by the -m command line option. (See the manual page clisp.man.) When this memory is filled up, you will get the error you described. Other versions of clisp allocate memory dynamically. Until the OS refuses to grant more memory to clisp. (The drawback of this strategy is that some Unices crash when their RAM or swap space is filled up. The clisp process will be aborted without any opportunity to save the computation's results.) > I've only seen this error on DECstation/ULTRIX, not on Sparc/Solaris. SunOS has a working mmap() system call, which clisp uses for its memory allocation. This results in the dynamic strategy. On Ultrix, clisp apparently chooses shared memory (SystemV IPC) for its memory allocation, and there are so many limits on shared memory that a fixed memory size is inevitable. > can someone explain the output of the clisp ``room'' function The first value is the number of bytes currently occupied by Lisp objects. The second value is the number of bytes that can be allocated before the next garbage collection occurs. Well, this could have been explained in impnotes.txt, but I hate to bother users with such low-level details. > ``room'' function, which seems to be still CLtL1, since it takes no > arguments? Would you really prefer some more verbose ROOM function? If you want to reduce your program's memory consumption, you should use the TIME macro. Perhaps some more sophisticated metering facility should be invented? - I think that the global values output by ROOM cannot give useful hints. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From @serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de Wed Sep 8 10:57:09 1993 Return-Path: <@serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de> Received: from uni-kl.de (stepsun.uni-kl.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03751; Wed, 8 Sep 93 10:57:09 +0200 Received: from serv-200.dfki.uni-kl.de by stepsun.uni-kl.de id aa26288; 8 Sep 93 10:55 MET DST Received: from serv-300.dfki.uni-kl.de by serv-200.dfki.uni-kl.de id aa06439; 8 Sep 93 8:55 GMT Date: Wed, 8 Sep 93 8:55:41 GMT From: Bjoern Hoefling To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: hoefling@dfki.uni-kl.de Subject: connection to tcl/tk Organization: DFKI, University of Kaiserslautern, Germany Message-Id: <9309080855.aa14487@serv-300.dfki.uni-kl.de> Hello everybody! I have the following problem concerning the connection between clisp and the tcl/tk toolkit. Tcl/tk is a GUI which I want to use to build a graphical user interface for a program written in clisp. Until now I have discovered 2 possibi- lities to realize a connection between a CommonLisp and tcl/tk. The first is to create a bidirectional stream between Lisp and the tcl/tk shell wish. For example in Lucid Common Lisp this can be realized in the following way: (defun init-interface () (setq *lisp-to-wish-stream* (run-program "wish" :input :stream :output :stream :wait nil))) After that Lisp can send messages to wish by the created stream and the output of wish is sent to Lisp. Unfortunately run-program does not exist in clisp and until now I discovered only the possibility to realize a communication via 2 files, one for the messages from lisp to tcl and one for the way back and Lisp and wish must explicitly read from/write on that files. Does anybody know a solution for clisp which is more elegant and more like the Lucid solution? If this problem can only be solved on the operating system level, I am using both clisp under Linux on a PC (486) and under Unix on Sun. Thanks in advance for any help. ********************************************************************* Bjoern HOEFLING, DFKI, Postfach 2080, W-67608 Kaiserslautern, Germany hoefling@dfki.uni-kl.de Phone: +49-631-205-3487 Fax: -3210 ********************************************************************* From haible Fri Sep 10 12:52:18 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06445; Fri, 10 Sep 93 12:52:18 +0200 Date: Fri, 10 Sep 93 12:52:18 +0200 From: haible (Bruno Haible) Message-Id: <9309101052.AA06445@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: connection to tcl/tk Bjoern Hoefling asks how to realize a connection between a CommonLisp and tcl/tk. > create a bidirectional stream between Lisp and the tcl/tk shell wish. CLISP supports unidirectional streams to other programs through the functions MAKE-PIPE-INPUT-STREAM and MAKE-PIPE-OUTPUT-STREAM. I haven't implemented bidirectional streams because that could lead to deadlocks. See impnotes.txt for details. > I discovered only the possibility to realize a communication via 2 files, > one for the messages from lisp to tcl and one for the way back and Lisp > and wish must explicitly read from/write on that files. You may simplify this by using FIFOs (named pipes) instead of ordinary files. Try something like that: (shell "mkfifo /tmp/wish-in") (shell "mkfifo /tmp/wish-out") (setq *lisp-to-wish-stream* (make-two-way-stream (open "/tmp/wish-out" :direction :input) (open "/tmp/wish-in" :direction :output) ) ) (shell "wish < /tmp/wish-in > /tmp/wish-out") (delete-file "/tmp/wish-in") (delete-file "/tmp/wish-out") > I am using both clisp under Linux on a PC (486) and under Unix on Sun. mkfifo is contained in the GNU file utilities. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From brent@amber.ssd.csd.harris.com Fri Sep 10 14:04:06 1993 Return-Path: Received: from travis.csd.harris.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06835; Fri, 10 Sep 93 14:04:06 +0200 Received: from amber.ssd.csd.harris.com by travis.csd.harris.com (5.61/harris-5.1) id AA24425; Fri, 10 Sep 93 08:01:52 -0400 Received: by amber (5.61/CX/UX-5.0) id AA23702; Fri, 10 Sep 93 08:01:51 -0400 Date: Fri, 10 Sep 93 08:01:51 -0400 From: brent@amber.ssd.csd.harris.com (Brent Benson) Message-Id: <9309101201.AA23702@amber> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: <9309101052.AA06445@ma2s2.mathematik.uni-karlsruhe.de> (haible@ma2s2.mathematik.uni-karlsruhe.de) Subject: Re: connection to tcl/tk # From: haible@ma2s2.mathematik.uni-karlsruhe.de (Bruno Haible) # # CLISP supports unidirectional streams to other programs through the # functions MAKE-PIPE-INPUT-STREAM and MAKE-PIPE-OUTPUT-STREAM. I haven't # implemented bidirectional streams because that could lead to deadlocks. # See impnotes.txt for details. ... # You may simplify this by using FIFOs (named pipes) instead of ordinary files. # Try something like that: # # (shell "mkfifo /tmp/wish-in") # (shell "mkfifo /tmp/wish-out") # (setq *lisp-to-wish-stream* # (make-two-way-stream (open "/tmp/wish-out" :direction :input) # (open "/tmp/wish-in" :direction :output) # ) ) # (shell "wish < /tmp/wish-in > /tmp/wish-out") # (delete-file "/tmp/wish-in") # (delete-file "/tmp/wish-out") # Why not implement MAKE-PIPE-IO-STREAM in terms of this solution and make it part of CLISP? From haible Fri Sep 10 16:46:50 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00531; Fri, 10 Sep 93 16:46:50 +0200 Date: Fri, 10 Sep 93 16:46:50 +0200 From: haible (Bruno Haible) Message-Id: <9309101446.AA00531@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2 index From haible Fri Sep 10 19:58:00 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01098; Fri, 10 Sep 93 19:58:00 +0200 Date: Fri, 10 Sep 93 19:58:00 +0200 From: haible (Bruno Haible) Message-Id: <9309101758.AA01098@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: connection to tcl/tk > Why not implement MAKE-PIPE-IO-STREAM in terms of this solution and > make it part of CLISP? Because it can lead to deadlocks. When process A waits for process B and process B waits for process A to put something onto the pipe, both processes will block. Besides, such cycles in the data flow graph between processes can normally be avoided. If the contents of either pipe can be determined a priori, then use a file. Furthermore, I can't implement it for the OS/2 version of CLISP, only for Unix. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de PS: I apologize for my last `message'. The archive of this mailing list can be retrieved via anonymous FTP from ma2s2.mathematik.uni-karlsruhe.de, file /pub/lisp/clisp/mailing-list-archive, and now also by sending mail to listserv@ma2s2.mathematik.uni-karlsruhe.de including the three words "get clisp-list mailing-list-archive" as message body. From haible Mon Sep 13 10:38:11 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00502; Mon, 13 Sep 93 10:38:11 +0200 Date: Mon, 13 Sep 93 10:38:11 +0200 From: haible (Bruno Haible) Message-Id: <9309130838.AA00502@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: New CLISP version Some time ago, Johann Petrak asked: > Has anyone compiled the new CLISP version for OS/2 ???????? > I am unable to do that myself, I don't have the disk space... Yes. It's on ma2s2:/pub/lisp/clisp/binaries/os2/clisp.zip . Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From P2043742@csdvax.csd.unsw.EDU.AU Thu Sep 16 05:50:54 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03668; Thu, 16 Sep 93 05:50:54 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H30O51BHGW91Z394@csdvax.csd.unsw.EDU.AU>; Thu, 16 Sep 1993 13:48:24 +1000 Date: Thu, 16 Sep 1993 13:48:24 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Problems with Garnet-2.2 To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H30O51CAEQ91Z394@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Clisp Users, I have just built Garnet2.2 on SGI Indigo and it runs 2-3 times faster then version 2.1. First I tested it with my application I am developing and everything seemed to be just fine. After this I saved the memory image. When I start up Garnet using this memory image, sometimes I get the following continuable error: ** - Continuable Error X-Error: Asynchronous FONT-ERROR in request 2395 (last request was 2535) Code 5 6.0 [ChangeGC] ID #x2800001 If you continue (by typing 'continue'): Ignore 1. Break> I got this error when I am using the saved memory image "only".(as far as I can tell now without extensive testing) When I start up the system loading the modules everything seemes to be OK. The other problem is with agate. When I trying to start it up I get this: > (agate:do-go) *** - index too large 1. Break> Has anybody encountered this problaims or something is wrong with my system. I am using Clisp that is available from the pub/lisp/clisp/binaries/sgi-irix4 and the jun 7 pcl+clx.sept92f.clisp.tar.z. Has anyone built clisp using the newest sources for SGI Indigo if yes please make the image available for me somehow because I haven't got gcc on my machine and Clisp as far as I know can't be built using the Irix native cc compiler. If someone built it allready please place it into ma2s2.mathematik.uni-karlsruhe.de where I can download it from. I'd approciate it very mutch. Yours, Bela Pecsek. From rbfeens@cs.vu.nl Fri Sep 17 16:55:50 1993 Return-Path: Received: from top.cs.vu.nl by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02151; Fri, 17 Sep 93 16:55:50 +0200 Received: from fluit.cs.vu.nl by top.cs.vu.nl id aa26107; 17 Sep 93 16:52 MET DST Date: Fri, 17 Sep 93 16:51:59 MET DST From: RB Feenstra To: clisp-list Subject: Re: Problems with Garnet-2.2 Message-Id: <9309171651.aa05995@fluit.cs.vu.nl> [Bela Pecsek describes problems (X FONT-ERROR) with garnet2.2beta & CLISP, that occur only if a saved lisp image is restarted.] I have encountered the same problems you described on a sun4. I suppose the cause is that you did a plain (saveinitmem) to save your Lisp image instead of using the special garnet functions opal:disconnect-garnet before saving the image, and using opal::initialize-x11-values after startup instead of the special opal:reconnect-garnet call, which was designed for this purpose. The Garnet 2.2 beta changes document also mentions a new call opal:Make-Image (see section 5.12) which does all this, but it seems that the current version doesn't support CLISP. Adding a (saveinitmem) at the right spot in utils.lsp won't be difficult though, but I haven't had the time to try this. If this solves your problems, it would be convenient to merge this fix to the main garnet source and adapt garnets README.CLISP to suggest using this new function. Remco Feenstra From syali@cs.Buffalo.EDU Mon Sep 20 01:28:30 1993 Return-Path: Received: from hydra.cs.Buffalo.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03581; Mon, 20 Sep 93 01:28:30 +0200 Received: by hydra.cs.Buffalo.EDU (4.1/1.01) id AA04374; Sun, 19 Sep 93 19:24:34 EDT Date: Sun, 19 Sep 93 19:24:34 EDT From: syali@cs.Buffalo.EDU (Sy Ali) Message-Id: <9309192324.AA04374@hydra.cs.Buffalo.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: CHANGE-CLASS not implemented? Is this not implemented in the latest version of CLISP? I can't seem to find it, and before I go writing it want to make sure its not there. Thanks, Sy From P2043742@csdvax.csd.unsw.EDU.AU Mon Sep 20 04:34:01 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03717; Mon, 20 Sep 93 04:34:01 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H366LWMYW09202JC@csdvax.csd.unsw.EDU.AU>; Mon, 20 Sep 1993 12:31:02 +1000 Date: Mon, 20 Sep 1993 12:31:02 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Re: Problems with Garnet-2.2 To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H366LWOB429202JC@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Remco, Thanks for your fast reply to my mail. I tried out the suggested opal:Make-Image to save the image and it works just fine. I guess I should have tried it out before asking questions but I thought saveinitmem should work too. (It was working with Garnet-2.1) This function is implemented in Garnet-2.2 for Clisp too allready thanks to Bruno. Allthough the problem with Agate (*** - index too large) is still remained. Thanks for your help again. Yours, Bela From haible Mon Sep 20 16:27:24 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04684; Mon, 20 Sep 93 16:27:24 +0200 Date: Mon, 20 Sep 93 16:27:24 +0200 From: haible (Bruno Haible) Message-Id: <9309201427.AA04684@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CHANGE-CLASS not implemented? Sy Ali asks: > Is this not implemented in the latest version of CLISP? No, I didn't implement CHANGE-CLASS. It is hardly ever used, and it violates a fundamental property of Lisp: an object's type must be constant as time passes. (Would you be happy if a string could be "changed" to become a number?) If you really need CHANGE-CLASS, you may use PCL. It's bigger and slower than CLISP's native CLOS, but it has CHANGE-CLASS. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From syali@cs.Buffalo.EDU Mon Sep 20 18:08:27 1993 Return-Path: Received: from hydra.cs.Buffalo.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04977; Mon, 20 Sep 93 18:08:27 +0200 Received: by hydra.cs.Buffalo.EDU (4.1/1.01) id AA11354; Mon, 20 Sep 93 12:04:24 EDT Date: Mon, 20 Sep 93 12:04:24 EDT From: syali@cs.Buffalo.EDU (Sy Ali) Message-Id: <9309201604.AA11354@hydra.cs.Buffalo.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CHANGE-CLASS not implemented? From clisp-list@ma2s2.mathematik.uni-karlsruhe.de Mon Sep 20 10:30:32 1993 Date: Mon, 20 Sep 93 16:27:35 +0200 Errors-To: haible@ma2s2.mathematik.uni-karlsruhe.de Originator: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Errors-To: haible@ma2s2.mathematik.uni-karlsruhe.de Reply-To: clisp-list Sender: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Version: 5.5 -- Copyright (c) 1991/92, Anastasios Kotsikonas From: haible@ma2s2.mathematik.uni-karlsruhe.de (Bruno Haible) To: Multiple recipients of list Subject: Re: CHANGE-CLASS not implemented? Sy Ali asks: > Is this not implemented in the latest version of CLISP? No, I didn't implement CHANGE-CLASS. It is hardly ever used, and it violates a fundamental property of Lisp: an object's type must be constant as time passes. (Would you be happy if a string could be "changed" to become a number?) If you really need CHANGE-CLASS, you may use PCL. It's bigger and slower than CLISP's native CLOS, but it has CHANGE-CLASS. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de --- I'm familiar with all the above arguments. I'm also capable of writing a CHANGE-CLASS of my own. I was just wondering if this was an oversight on your part, since it is part of the specification. It's your lisp (which I'm happy for) and you can decide if you want to be conformant or not. Thanks, Sy From P2043742@csdvax.csd.unsw.EDU.AU Tue Sep 21 09:52:28 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05825; Tue, 21 Sep 93 09:52:28 +0200 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H37S3K97409207LB@csdvax.csd.unsw.EDU.AU>; Tue, 21 Sep 1993 15:57:20 +1000 Date: Tue, 21 Sep 1993 15:57:20 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Problem with Garnet2.2 again. To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H37S3KASZ69207LB@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Clisp users, I am using Garnet2.2 and when I want to change a slot in a property window I get the following error: *** - FUNCALL: the function GARNET-GADGETS::CONDITION is undefined 1. Break> I think this error comes from the Careful-Eval macro where It seemes to me that the condition system is used and Clisp does not now how to handle handler-case and (error (condition) .... Or am I missing something? If this is the case has somebody changed this macro jet? If yes please tell me how to do it. Another problem: I can not find the CLISP.README file for Garnet2.2!! Where is it? Yours, Bela. From haible Tue Sep 21 14:52:08 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06440; Tue, 21 Sep 93 14:52:08 +0200 Date: Tue, 21 Sep 93 14:52:08 +0200 From: haible (Bruno Haible) Message-Id: <9309211252.AA06440@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Problem with Garnet2.2 again. Bela Pecsek tests Garnet 2.2: > *** - FUNCALL: the function GARNET-GADGETS::CONDITION is undefined > 1. Break> > > I think this error comes from the Careful-Eval macro where It seemes to me > that the condition system is used and Clisp does not now how to handle > handler-case Yes. This is the problem. I am sending patches for this and some other problems to the Garnet developers. Anyone wants to implement the condition system (HANDLER-CASE etc.) for clisp? > I can not find the CLISP.README file for Garnet2.2!! Where is it? The Garnet developers are trying to make the README.CLISP obsolete. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From jasont@cs.tamu.edu Fri Sep 24 00:14:25 1993 Return-Path: Received: from clavin (clavin.cs.tamu.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08899; Fri, 24 Sep 93 00:14:25 +0200 Received: from cs.tamu.edu (slx5.cs.tamu.edu) by clavin (AA03912); Thu, 23 Sep 93 17:07:35 CDT Date: Thu, 23 Sep 93 17:07:35 CDT From: jasont@cs.tamu.edu (Jason T Tobias) Message-Id: <9309232207.AA03912@clavin> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: documentation? X-Sun-Charset: US-ASCII Sorry if this is an FAQ, but is there any documentation of general commands for CLISP? I have the PC load of CLISP. Thanks -Jason From reggie@muon.phys.washington.edu Mon Sep 27 00:56:11 1993 Return-Path: Received: from muon.phys.washington.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16824; Mon, 27 Sep 93 00:56:11 +0100 Received: by muon.phys.washington.edu via SMTP (920330.SGI/911001.SGI) for clisp-list@ma2s2.mathematik.uni-karlsruhe.de id AA18784; Sun, 26 Sep 93 16:52:38 -0700 Message-Id: <9309262352.AA18784@muon.phys.washington.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Problems compiling clisp under A/UX. Date: Sun, 26 Sep 1993 16:52:38 -0700 From: Reggie Perry Hi there. I am trying to compile clisp on a Mac Workgroup Server 95 running A/UX 3.0.1. I get this error that baffles me. I will provide the transcript. Any help would be grately appreciated. Thanks. Reggie gcc -O -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type -fomit-frame-pointer -O2 -DNO_MULTIMAP_SHM -c spvw.c spvw.d: In function `gc_compact_from_varobject_page': spvw.d:3116: warning: `p2' may be used uninitialized in this function spvw.d:3117: warning: `l2' may be used uninitialized in this function spvw.d: In function `gc_compact_from_cons_page': spvw.d:3170: warning: `p2' may be used uninitialized in this function spvw.d:3171: warning: `l2' may be used uninitialized in this function spvw.d: In function `gc_compact_heap': spvw.d:3231: warning: address of register variable `pages_sorted' requested spvw.d: In function `sigwinch_handler': spvw.d:5278: warning: passing arg 2 of `signal' from incompatible pointer type spvw.d: In function `interrupt_handler': spvw.d:5293: warning: passing arg 2 of `signal' from incompatible pointer type spvw.d: In function `alarm_handler': spvw.d:5313: warning: passing arg 2 of `signal' from incompatible pointer type spvw.d: In function `main': spvw.d:6389: warning: integer overflow in expression spvw.d:6868: warning: passing arg 2 of `signal' from incompatible pointer type spvw.d:6911: warning: passing arg 2 of `signal' from incompatible pointer type spvw.d:6913: warning: passing arg 2 of `signal' from incompatible pointer type spvw.d:6293: warning: variable `argv_memneed' may be clobbered by `longjmp' or `vfork' spvw.c:3589:"Malformed operand" -- Statement 'movel #symbol_tab_data+536875808,a4@+' ignored spvw.c:3613:"Malformed operand" -- Statement 'movel #symbol_tab_data+536875808,a4@+' ignored spvw.c:3731:"Malformed operand" -- Statement 'movel %sp@(188),a4@+' ignored spvw.c:3761:"Malformed operand" -- Statement 'movel a4@-,%a0@(16)' ignored spvw.c:4252:"Malformed operand" -- Statement 'movel %sp@(196),a4@+' ignored spvw.c:4253:"Malformed operand" -- Statement 'movel %sp@(192),a4@+' ignored spvw.c:4287:"Malformed operand" -- Statement 'movel a4@-,%a0@' ignored spvw.c:4288:"Malformed operand" -- Statement 'movel a4@-,%a0@(4)' ignored spvw.c:4307:"Malformed operand" -- Statement 'movel %sp@(192),a4@+' ignored spvw.c:4308:"Malformed operand" -- Statement 'movel %sp@(188),a4@+' ignored spvw.c:4338:"Malformed operand" -- Statement 'movel a4@-,%a0@' ignored spvw.c:4339:"Malformed operand" -- Statement 'movel a4@-,%a0@(4)' ignored spvw.c:4394:"Malformed operand" -- Statement 'movel a4@-,%d0' ignored spvw.c:4647:"Malformed operand" -- Statement 'movel %d0,a4@+' ignored spvw.c:5327:"Malformed operand" -- Statement 'movel %d0,a4@+' ignored spvw.c:5333:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored spvw.c:10221:"Malformed operand" -- Statement 'movel %a0@(4),a4@+' ignored spvw.c:10249:"Malformed operand" -- Statement 'movel %a6,a4@+' ignored spvw.c:10264:"Malformed operand" -- Statement 'movel a4@-,%d0' ignored spvw.c:11594:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored spvw.c:11614:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored spvw.c:11620:"Malformed operand" -- Statement 'movel mv_space,a4@+' ignored spvw.c:12115:"Malformed operand" -- Statement 'movel #0,a4@+' ignored spvw.c:12116:"Malformed operand" -- Statement 'movel #0,a4@+' ignored spvw.c:12194:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored spvw.c:12221:"Malformed operand" -- Statement 'movel mv_space,a4@+' ignored spvw.c:12227:"Malformed operand" -- Statement 'movel a4@-,%d0' ignored spvw.c:12245:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored spvw.c:12267:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored spvw.c:12268:"Malformed operand" -- Statement 'movel #symbol_tab_data+536894816,a4@+' ignored spvw.c:12276:"Malformed operand" -- Statement 'movel %d0,a4@+' ignored spvw.c:12296:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored spvw.c:12372:"Malformed operand" -- Statement 'movel object_tab+1028,a4@+' ignored spvw.c:12400:"Malformed operand" -- Statement 'movel %d1,a4@+' ignored Make: *** Error: Update of `spvw.o' terminated with exit code 1 ------------------- Reginald S. Perry -- University of Washington, Department of Physics FM-15 Seattle, Washington 98195 reggie@muon.phys.washington.edu (General-Mail) reggie@u.washington.edu (General-Mail) reggie@quantum.phys.washington.edu (NeXTMail) From haible Mon Sep 27 14:27:20 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17211; Mon, 27 Sep 93 14:27:20 +0100 Date: Mon, 27 Sep 93 14:27:20 +0100 From: haible (Bruno Haible) Message-Id: <9309271327.AA17211@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: New CLISP version Once more, there is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . Excerpt from the change log: User visible changes -------------------- * COMPILE-FILE now handles top-level forms as described in CLtL2 p. 90: The macroexpansion of a top-level form is considered a top-level form. When a top-level form is a PROGN, LOCALLY, COMPILER-LET, SYMBOL-MACROLET or MACROLET form, the subforms of its body are considered top-level forms too and are compiled separately. This allows things like (PROGN (DEFCONSTANT X 3141) (DEFUN GET-X () X)) or (PROGN (DEFCLASS SHIP () ()) (DEFUN SHIP-P (X) (TYPEP X 'SHIP))) to work. Warning: If you want to inhibit this splitting into separate forms (for example, because several of the forms refer to the same uninterned symbol), you have to use (LET () (PROGN ...)) instead of (PROGN ...). * X3J13 vote <144> is implemented: New macro DECLAIM. * X3J13 vote <113> is implemented: The macro LOCALLY is now a special form. * Fixed a serious bug in the compiler: When compiling functions inline whose parameters are special variables, incorrect code could be generated. * Fixed a bug in the compiler which caused the compiler to crash in some rare cases, notably when compiling Garnet-2.2. * Fixed a bug that prevented the debugger commands DOWN and BOTTOM from working on Linux and SunOS. * Fixed a CLOS bug: Specifying &KEY without &REST in generic function lambda lists didn't have the desired effect. * DEFCLASS supports both :DEFAULT-INITARGS syntaxes. * DEFGENERIC is more tolerant about invalid declarations. Portability ----------- * Added Dynix/ptx Sequent support. Thanks to Esther Davis. * Added support for RS/6000 CPU and AIX. Thanks to Gabor Herr. Other modifications ------------------- * Made the memory allocation and (ROOM), (TIME ...) more virtual memory efficient (less paging on Unix and DOS). Thanks to Don Cohen. * Added Geoffrey Gordon's "LISP tutorial". * Renamed stdwin.lsp to stdwin2.lsp. * Miscellaneous documentation updates. Binaries have been rebuilt for the following architectures: from clisp-1993-09-27: Linux 928808 Sep 26 00:37 linux/clisp-english.tar.z from clisp-1993-09-01: Sun4, SunOS 4 1015709 Sep 24 22:57 sun4-sunos4/clisp-english.tar.z Atari ST 916677 Sep 17 17:01 atari/clisp.tar.z OS/2 1109019 Sep 13 11:08 os2/clisp.zip Sun4, Solaris 2.1 1165140 Sep 7 22:30 sun4-solaris2/clisp.tar.z Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Mon Sep 27 15:28:46 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17372; Mon, 27 Sep 93 15:28:46 +0100 Date: Mon, 27 Sep 93 15:28:46 +0100 From: haible (Bruno Haible) Message-Id: <9309271428.AA17372@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Problems compiling clisp under A/UX. Reggie Perry has trouble compiling clisp on Apple's A/UX 3.0.1: > I get this error that baffles me. > spvw.c:12245:"Malformed operand" -- Statement 'movel %a0,a4@+' ignored My speedups for clisp on 680x0 processors require GNU C and are written in the MIT assembly language syntax of the 680x0. Apple's assembler apparently uses another syntax (maybe the Motorola syntax). You may - either rewrite the 680x0 __asm__ stuff to use the other syntax, or - pretend to clisp the CPU were something like an 680x0, but not an 680x0 proper (modify lispbibl.d for that purpose), or - pretend to clisp the compiler were not GNU C (use "gcc -U__GNUC__" as compiler instead of "gcc"). Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From reggie@muon.phys.washington.edu Mon Sep 27 19:33:15 1993 Return-Path: Received: from muon.phys.washington.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17924; Mon, 27 Sep 93 19:33:15 +0100 Received: by muon.phys.washington.edu via SMTP (920330.SGI/911001.SGI) for clisp-list@ma2s2.mathematik.uni-karlsruhe.de id AA21839; Mon, 27 Sep 93 11:29:27 -0700 Message-Id: <9309271829.AA21839@muon.phys.washington.edu> To: clisp-list Subject: Re: Problems compiling clisp under A/UX. In-Reply-To: Your message of "Mon, 27 Sep 1993 15:30:35 BST." <9309271428.AA17372@ma2s2.mathematik.uni-karlsruhe.de> Date: Mon, 27 Sep 1993 11:29:26 -0700 From: Reggie Perry Thanks for the fix. Now for the next problem :-) I had this happen: gcc -O -U__GNUC__ -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type -fomit-frame-pointer -O2 -c lisparit0.c lisparit0.d:1663: ari68020.c: No such file or directory Make: *** Error: Update of `lisparit0.o' terminated with exit code 1 Make: [line 789 in /root/src/clisp-1993-09-27/aux/makefile] I did this: ./comment5 ../src/ari68020.mot.d | ./ansidecl > ari68020.d and then this happened: gcc -O -U__GNUC__ -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type -fomit-frame-pointer -O2 -L. spvw.o spvwtabf.o spvwtabs.o spvwtabo.o eval.o control.o pathname.o stream.o socket.o io.o array.o hashtabl.o list.o package.o record.o sequence.o charstrg.o debug.o misc.o predtype.o symbol.o lisparit0.o stdwin.o graph.o unixaux.o -lreadline -ltermcap -L/usr/lib -lX11 -o lisp.run ld warning: Multiply defined symbol PC, in /lib/libtermcap.a, has more than onesize undefined first referenced symbol in file pttob spvw.o tcgetattr stream.o mulu32_ array.o copy_loop_up lisparit0.o copy_loop_down lisparit0.o fill_loop_down lisparit0.o or_loop_up lisparit0.o xor_loop_up lisparit0.o and_loop_up lisparit0.o eqv_loop_up lisparit0.o nand_loop_up lisparit0.o nor_loop_up lisparit0.o andc2_loop_up lisparit0.o orc2_loop_up lisparit0.o not_loop_up lisparit0.o and_test_loop_up lisparit0.o clear_loop_down lisparit0.o shift1left_loop_down lisparit0.o shiftleft_loop_down lisparit0.o shift1right_loop_up lisparit0.o shiftrightsigned_loo lisparit0.o test_loop_up lisparit0.o addto_loop_down lisparit0.o neg_loop_down lisparit0.o subfrom_loop_down lisparit0.o compare_loop_up lisparit0.o shiftrightcopy_loop_ lisparit0.o clear_loop_up lisparit0.o mulu_loop_down lisparit0.o muluadd_loop_down lisparit0.o add_loop_down lisparit0.o dec_loop_down lisparit0.o inc_loop_down lisparit0.o divucopy_loop_up lisparit0.o shiftleftcopy_loop_d lisparit0.o divu_6432_3232_ lisparit0.o mulusub_loop_down lisparit0.o shiftright_loop_up lisparit0.o sub_loop_down lisparit0.o divu_loop_up lisparit0.o mulusmall_loop_down lisparit0.o subx_loop_down lisparit0.o fill_loop_up lisparit0.o tcsetattr unixaux.o tcdrain unixaux.o tcflush unixaux.o ld fatal: Symbol referencing errors. No output written to lisp.run collect: ld returned 13 exit status What little thing am I missing? Reggie ------------------- Reginald S. Perry -- University of Washington, Department of Physics FM-15 Seattle, Washington 98195 reggie@muon.phys.washington.edu (General-Mail) reggie@u.washington.edu (General-Mail) reggie@quantum.phys.washington.edu (NeXTMail) From johnm@netcom.com Tue Oct 5 23:53:12 1993 Return-Path: Received: from netcom.netcom.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA29583; Tue, 5 Oct 93 23:53:12 +0100 Received: by netcom.netcom.com (5.65/SMI-4.1/Netcom) id AA14616; Tue, 5 Oct 93 15:53:17 -0700 Date: Tue, 5 Oct 93 15:53:17 -0700 From: johnm@netcom.com (J_Gerard Malecki) Message-Id: <9310052253.AA14616@netcom.netcom.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de get CLISP-LIST mailing-list-archive From puolamak@cc.helsinki.fi Wed Oct 6 13:23:59 1993 Return-Path: Received: from kruuna.helsinki.fi by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00125; Wed, 6 Oct 93 13:23:59 +0100 Received: from klaava.Helsinki.FI by kruuna.helsinki.fi with SMTP id AA27923 (5.65c8/IDA-1.4.4 for ); Wed, 6 Oct 1993 14:22:54 +0200 Received: by klaava.Helsinki.FI (4.1/SMI-4.1) id AA07423; Wed, 6 Oct 93 14:02:22 +0200 From: puolamak@cc.helsinki.fi (Kai Rainer Puolamaki) Message-Id: <9310061202.AA07423@klaava.Helsinki.FI> Subject: Re: your mail To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Wed, 6 Oct 1993 14:02:21 +0200 (EET) In-Reply-To: <9310052253.AA14616@netcom.netcom.com> from "J_Gerard Malecki" at Oct 5, 93 11:55:37 pm Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 38 get CLISP-LIST mailing-list-archive From puolamak@cc.helsinki.fi Tue Oct 12 09:36:36 1993 Return-Path: Received: from kruuna.helsinki.fi by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06844; Tue, 12 Oct 93 09:36:36 +0100 Received: from klaava.Helsinki.FI by kruuna.helsinki.fi with SMTP id AA01870 (5.65c8/IDA-1.4.4 for ); Tue, 12 Oct 1993 10:34:38 +0200 Received: by klaava.Helsinki.FI (4.1/SMI-4.1) id AA23881; Tue, 12 Oct 93 10:34:37 +0200 From: puolamak@cc.helsinki.fi (Kai Rainer Puolamaki) Message-Id: <9310120834.AA23881@klaava.Helsinki.FI> Subject: get clisp-list archieve To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Tue, 12 Oct 1993 10:34:36 +0200 (EET) In-Reply-To: <9310052253.AA14616@netcom.netcom.com> from "J_Gerard Malecki" at Oct 5, 93 11:55:37 pm Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 37 get CLISP-LIST mailing-list-archive From @serv-200.dfki.uni-kl.de:elsbernd@dfki.uni-kl.de Tue Oct 12 10:23:09 1993 Return-Path: <@serv-200.dfki.uni-kl.de:elsbernd@dfki.uni-kl.de> Received: from uni-kl.de (stepsun.uni-kl.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07028; Tue, 12 Oct 93 10:23:09 +0100 Received: from serv-200.dfki.uni-kl.de by stepsun.uni-kl.de id aa17500; 12 Oct 93 10:21 MET Received: from isg-200.dfki.uni-kl.de by serv-200.dfki.uni-kl.de id aa15070; 12 Oct 93 9:21 GMT Date: Tue, 12 Oct 93 9:21:43 GMT From: Klaus Elsbernd To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: get CLISP-LIST mailing-list-archive Organization: German Research Center for AI (DFKI) Message-Id: <9310120921.aa20878@isg-200.dfki.uni-kl.de> From tom@jsk.t.u-tokyo.ac.jp Tue Oct 12 11:50:30 1993 Received: from monoceros (jsk.t.u-tokyo.ac.jp) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07278; Tue, 12 Oct 93 11:50:30 +0100 Received: from aquila.jsk.t.u-tokyo.ac.jp by monoceros (4.1/6.4J.6-MX) id AA06214; Tue, 12 Oct 93 19:48:56 JST Received: by aquila.jsk.t.u-tokyo.ac.jp (4.1/6.4J.6-MX) id AA00294; Tue, 12 Oct 93 19:51:06 JST Date: Tue, 12 Oct 93 19:51:06 JST From: tom@jsk.t.u-tokyo.ac.jp (Tomohiro Shibata) Return-Path: Message-Id: <9310121051.AA00294@aquila.jsk.t.u-tokyo.ac.jp> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Q:Xit and usage of make-socket-stream Hello everyone! I am new to this mailing list and now trying to install clue+clio+xit on clisp into SparcIPX. It seems to have succeeded making clx because demo worked correctly. But below message logged while makeing clue. ----------- begin ------------- clisp -m 5MB -M ../clx/clx.mem -q -i defsystem -x '(compile-clue)' ;; Loading file /home/monoceros/tshibata/Softs/clisp/packages/clue+clio+xit/clue/defsystem.fas ... ;; Loading of file /home/monoceros/tshibata/Softs/clisp/packages/clue+clio+xit/clue/defsystem.fas is finished. ; Loading clue.fas ; Loading clx-patch.fas ; Loading window-doc.fas ; Loading event-parse.fas ; Loading defcontact.fas ; Compiling intrinsics.lsp Compiling file /home/monoceros/tshibata/Softs/clisp/packages/clue+clio+xit/clue/intrinsics.lsp ... *** - FIND-CLASS: WINDOW does not name a class touch stamp-fas rm clue.mem rm: override protection 644 for clue.mem? y clisp -m 5MB -M ../clx/clx.mem -q -i defsystem -x '(load-clue) (saveinitmem)' ;; Loading file /home/monoceros/tshibata/Softs/clisp/packages/clue+clio+xit/clue/defsystem.fas ... ;; Loading of file /home/monoceros/tshibata/Softs/clisp/packages/clue+clio+xit/clue/defsystem.fas is finished. ; Loading clue.fas ; Loading clx-patch.fas ; Loading window-doc.fas ; Loading event-parse.fas ; Loading defcontact.fas ; Loading intrinsics.fas *** - UNIX error 13 (EACCES): Permission denied 2721264 ; 1360632 mv lispinit.mem clue.mem ---------end------------------ I don't know what above error is. Produced binary is 2.8MB which is almost same as clx. Please tell me what I should do. There is another question. I found SYSTEM function MAKE-SOCKET-STREM, but couldn't find its usage. Please tell me any information, or does anyone has programmed a function which return two-way-stream between clisp and forked executable program? Thanks in advance. p.s. I receive mail which contents is "get CLISP-LISP mailing-list-archive" many times. I'm sure to have get it by ftp. Why those mail come frequently? -------------------------------- Tomohiro Shibata (email: tom@jsk.t.u-tokyo.ac.jp) Inoue-Inaba Laboratory, Department of Mechano-informatics, University of Tokyo; Japan From jbishop@babel.ling.nwu.edu Tue Oct 12 13:53:10 1993 Received: from relay.acns.nwu.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07524; Tue, 12 Oct 93 13:53:10 +0100 Received: from babel.ling.nwu.edu by relay.acns.nwu.edu (4.1/SMI-ACNS-relay-1.7) id AA15126; Tue, 12 Oct 93 07:51:45 CDT Return-Path: Received: by babel.ling.nwu.edu (4.1/SMI-4.0) id AA10140; Tue, 12 Oct 93 07:51:23 CDT Date: Tue, 12 Oct 1993 07:50:55 -0500 (CDT) From: Jeff Bishop Sender: Jeff Bishop Reply-To: Jeff Bishop Subject: "get" messages To: clisp-list In-Reply-To: <9310121051.AA00294@aquila.jsk.t.u-tokyo.ac.jp> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII I have also received numerous "get ..." messages. Apparently, some users are mistaking the clisp-list address for a file server. I don't know if there is an email address which supports "get" but everything can be received by anonymous ftp from what I can tell. Jeff Bishop jbishop@nwu.edu From haible Tue Oct 12 16:10:43 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07821; Tue, 12 Oct 93 16:10:43 +0100 Date: Tue, 12 Oct 93 16:10:43 +0100 From: haible (Bruno Haible) Message-Id: <9310121510.AA07821@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Q:Xit and usage of make-socket-stream Mr Tomohiro Shibata has problems when compiling CLUE: > ; Compiling intrinsics.lsp > Compiling file /home/monoceros/tshibata/Softs/clisp/packages/clue+clio+xit/clue/intrinsics.lsp ... > *** - FIND-CLASS: WINDOW does not name a class Mr Jan Deleeuw had the same problem. It may come from the buggy top-level form handling in CLISP versions older than 1993-09-27. Please try getting that newest CLISP version. > Produced binary is 2.8MB which is almost same as clx. Note also that you should be able to use CLISP's native CLOS instead of PCL when building CLX for CLUE. The resulting memory image will be much smaller. > I found SYSTEM function MAKE-SOCKET-STREAM, but couldn't find its usage. It is the low-level function for building the connection to the X server. See clx/dependent.lsp for its usage. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de PS: Send administrative request such as "get clisp-list mailing-list-archive" or "unsubscribe clisp-list" to listserv@ma2s2.mathematik.uni-karlsruhe.de ^^^^^^^^ From deleeuw@math.ucla.edu Tue Oct 12 19:52:41 1993 Received: from julia.math.ucla.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA08430; Tue, 12 Oct 93 19:52:41 +0100 Received: from sonia.math.ucla.edu by julia.math.ucla.edu (Sendmail 4.1/1.07) id AA23757; Tue, 12 Oct 93 11:52:34 PDT Return-Path: Received: by sonia.math.ucla.edu (Sendmail 4.1/1.07) id AA00313; Tue, 12 Oct 93 11:52:33 PDT From: Jan Deleeuw Message-Id: <9310121852.AA00313@sonia.math.ucla.edu> Subject: clue To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Tue, 12 Oct 1993 11:52:33 -0700 (PDT) X-Mailer: ELM [version 2.4 PL22] Content-Type: text Content-Length: 736 The intrinsics.lsp compiling error in building CLUE is not due to the version of clisp. It also happens with the newest version. I shall try to go through the native CLOS next. ...... I have tried. Better, but no cigar. I now get ; Compiling ./intrinsics.lsp Compiling file /m1/lisp/clisp/packages/clue/intrinsics.lsp ... *** - DEFCLASS BASIC-CONTACT: superclass # should belong to class STANDARD-CLASS This seems to be the same problem in a slightly different form. --- Jan -- Jan de Leeuw; UCLA Statistics; phone (310)-825-9550; fax (310)-206-6673 mail: 405 Hilgard Ave, Los Angeles, CA 90024-1555 home: 130 Vista Place, Venice, CA 90291-3224; phone (310)-396-3171 INTERNET: deleeuw@stat.ucla.edu From tom@jsk.t.u-tokyo.ac.jp Wed Oct 13 13:50:51 1993 Received: from monoceros (jsk.t.u-tokyo.ac.jp) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA09323; Wed, 13 Oct 93 13:50:51 +0100 Received: from delphinus (delphinus.jsk.t.u-tokyo.ac.jp) by monoceros (4.1/6.4J.6-MX) id AA19155; Wed, 13 Oct 93 21:49:13 JST Received: by delphinus (4.1/6.4J.6-MX) id AA03198; Wed, 13 Oct 93 21:49:14 JST Date: Wed, 13 Oct 93 21:49:14 JST From: tom@jsk.t.u-tokyo.ac.jp (Tomohiro Shibata) Return-Path: Message-Id: <9310131249.AA03198@delphinus> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: Bruno Haible's message of Tue, 12 Oct 93 16:13:19 +0100 <9310121510.AA07821@ma2s2.mathematik.uni-karlsruhe.de> Subject: Re: Q:Xit and usage of make-socket-stream >>> ; Compiling intrinsics.lsp >>> Compiling file /home/monoceros/tshibata/Softs/clisp/packages/clue\ >>>clio+xit/clue/intrinsics.lsp ... >>> *** - FIND-CLASS: WINDOW does not name a class >> >>Mr Jan Deleeuw had the same problem. It may come from the buggy top-level >>form handling in CLISP versions older than 1993-09-27. Please try getting >>that newest CLISP version. OK, I got newest CLISP version by ftp and made it. Then, I decoded 'pcl+clx.sept92f.clisp.tar.z' to make pcl and clx. It seemed to be succeed making them. But after all the same error occured when making CLUE which source is clisp/packages/clue.clisp.tar.z or clisp/packages/clue+clio+xit.clisp.tar.z Is there anyone out who could succeed making clue+clio+xit on Sun4? >>> I found SYSTEM function MAKE-SOCKET-STREAM, but couldn't find its usage. >>It is the low-level function for building the connection to the X server. >>See clx/dependent.lsp for its usage. I believe provided MAKE-SOCKET-STREAM is only for connetion with X. I want a function which is : >>>but couldn't find its usage. Please tell me any information, or does >>>anyone has programmed a function which return two-way-stream between >>>clisp and forked executable program? Should I implement it by remake? p.s. Should I mail it only to Mr. Haible like this case(issues of making binaries)? -------------------------------- Tomohiro Shibata (email: tom@jsk.t.u-tokyo.ac.jp) Inoue-Inaba Laboratory, Department of Mechano-informatics, University of Tokyo; Japan From vargasje@charlie.ece.scarolina.edu Wed Oct 13 15:18:25 1993 Return-Path: Received: from woodstock.ece.scarolina.edu ([129.252.22.23]) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA09466; Wed, 13 Oct 93 15:18:25 +0100 Received: by woodstock.ece.scarolina.edu (5.65/DEC-Ultrix/4.3) id AA06420; Wed, 13 Oct 1993 10:16:46 -0400 Date: Wed, 13 Oct 1993 10:16:46 -0400 From: vargasje@charlie.ece.scarolina.edu (Dr J. E. Vargas) Message-Id: <9310131416.AA06420@woodstock.ece.scarolina.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Q:Xit and usage of make-socket-stream I am trying to build CLX on top of CLISP, on a DECstation 3100. First I tried the make clx, but CLISP is invoked and nothing happens. Then I started clisp, loaded defsystem.lsp, and then give the command (compile-clx), which so far runs fine, but since this is not what the README file says I am supposed to do, I wonder if, at the end, I will get some error. So the question is: Is the Makefile in the .../clx directory (from clx-for-clisp.tar) wrong? I am doing the right thing bypassing it? vargasje@ece.scarolina.edu From haible Wed Oct 13 18:45:07 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13883; Wed, 13 Oct 93 18:45:07 +0100 Date: Wed, 13 Oct 93 18:45:07 +0100 From: haible (Bruno Haible) Message-Id: <9310131745.AA13883@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Building CLX > Is the Makefile in the .../clx directory (from clx-for-clisp.tar) > wrong? I am doing the right thing bypassing it? Instead of using the CLX makefile you can as well do each of these steps: clisp -q -c defsystem clisp -m 4MB -q -i defsystem > (compile-clx) clisp -m 4MB -q -i defsystem > (load-clx "" :macrosp t) > (saveinitmem) mv lispinit.mem clx.mem Works as well. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From tom@jsk.t.u-tokyo.ac.jp Thu Oct 14 14:16:27 1993 Received: from monoceros (jsk.t.u-tokyo.ac.jp) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA14883; Thu, 14 Oct 93 14:16:27 +0100 Received: from aquila.jsk.t.u-tokyo.ac.jp by monoceros (4.1/6.4J.6-MX) id AA28750; Thu, 14 Oct 93 22:14:32 JST Received: by aquila.jsk.t.u-tokyo.ac.jp (4.1/6.4J.6-MX) id AA00869; Thu, 14 Oct 93 22:16:47 JST Date: Thu, 14 Oct 93 22:16:47 JST From: tom@jsk.t.u-tokyo.ac.jp (Tomohiro Shibata) Return-Path: Message-Id: <9310141316.AA00869@aquila.jsk.t.u-tokyo.ac.jp> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: Bruno Haible's message of Fri, 10 Sep 93 12:52:18 +0200 <9309101052.AA06445@ma2s2.mathematik.uni-karlsruhe.de> Subject: mkfifo Hello. There has been some discuttion on the mailing list on how to make bidirectional communication between clisp and other programs using Unix. below is one solutoin provided by Mr. Haible. >> (shell "mkfifo /tmp/wish-in") >> (shell "mkfifo /tmp/wish-out") >> (setq *lisp-to-wish-stream* >> (make-two-way-stream (open "/tmp/wish-out" :direction :input) >> (open "/tmp/wish-in" :direction :output) >> ) ) >> (shell "wish < /tmp/wish-in > /tmp/wish-out") >> (delete-file "/tmp/wish-in") >> (delete-file "/tmp/wish-out") As for my check, it doesn't work as they are seen above. when (open pipename-out :direction :input) is entered, prompt never comes. But (open pipename-in :direction :output) is entered, I can get #. I think only it can work in case of `:direction :io '. And I tried: >> (shell "command < pipename-in > pipename-out") , but I couln't succeed. Has Mr. Hoefling already succeeded to connect with WISH? Any suggestion and information are welcome. -------------------------------- Tomohiro Shibata (email: tom@jsk.t.u-tokyo.ac.jp) Inoue-Inaba Laboratory, Department of Mechano-informatics, University of Tokyo; Japan From matthias@intellektik.informatik.th-darmstadt.de Thu Oct 14 14:43:10 1993 Return-Path: Received: from rs2.hrz.th-darmstadt.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15011; Thu, 14 Oct 93 14:43:10 +0100 Received: from aida.intellektik.informatik.th-darmstadt.de by rs2.hrz.th-darmstadt.de with SMTP id AA16216 (5.65c/IDA-1.4.4 for ); Thu, 14 Oct 1993 14:41:26 +0100 Received: by aida.intellektik.informatik.th-darmstadt.de (4.1/Server-1.3/HRZ-THD) id AA20042; Thu, 14 Oct 93 14:41:21 +0100 Date: Thu, 14 Oct 93 14:41:21 +0100 From: Matthias Lindner Message-Id: <9310141341.AA20042@aida.intellektik.informatik.th-darmstadt.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: mkfifo hi, I don't know, if my solution to the problem ist the most elegant one, but anyway - it works (at least for me ;): (defun RUN-PROGRAM (prog &key arguments input output (if-output-exists :overwrite) (wait t)) (let* ((outf (cond ((stringp output) output) ((pathnamep output) (namestring output)) ((eq :stream output) (do ((pn (symbol-name (gensym "/tmp/clrpO")) (symbol-name (gensym "/tmp/clrpO")))) ((null (probe-file pn)) (unless (zerop (shell (format NIL "~A ~A" *mkfifo-prog* pn))) (error "Can't create fifo - sorry!")) pn))) (T NIL))) (inf (cond ((or (stringp input) (pathnamep input)) input) ((eq :stream input) (do ((pn (symbol-name (gensym "/tmp/clrpI")) (symbol-name (gensym "/tmp/clrpI")))) ((null (probe-file pn)) (unless (zerop (shell (format NIL "~A ~A" *mkfifo-prog* pn))) (error "Can't create fifo - sorry!")) (open pn :direction :output :if-exists :overwrite)))) (T NIL))) (cmd (format NIL "~A~{ ~A~}~@[ < ~A~]~@[ ~A~] 2>&1 ~@[ &~]" prog arguments (and inf (namestring inf)) (and outf (format NIL "~A ~A" (case if-output-exists (:append ">> ") (:overwrite "> ") (:error (when (probe-file outf) (error "Outfile ~A exists!" outf)) "> ") (otherwise (error "Strange if-output-exists spec ~A" if-output-exists))) outf)) (or (not wait) (eq :stream input) (eq :stream output)))) (res (shell (format NIL "exec /bin/sh -c ~S" cmd)))) (cond ((and (eq :stream input) (eq :stream output)) (let ((outs (open outf :direction :input))) (values (make-two-way-stream outs inf) outs inf))) ((eq :stream input) inf) ((eq :stream output) (open outf :direction :input)) (T res)))) (defun START-WISH (&key (wish *wish-prog*) (arguments NIL)) (multiple-value-setq (*wish* *wish-i-stream* *wish-o-stream*) (run-program wish :arguments arguments :input :stream :output :stream :wait NIL))) START-WISH returns a two way stream, that may be used to read from and write to the wish process. However, you should never try to use "read" or "read-line" on this stream, but use your own reader, that is implemented in terms of read-char-no-hang and/or listen. Also remember to use "flush output" on the wish-side, after every output-operation, or you will not see anything on the lisp-stream (until the io-buffer gets filled). Have fun --Matthias ------------------------------------------------------------------------------ Matthias Lindner FG Intellektik, FB Informatik Technische Hochschule Darmstadt Alexanderstr.10, D-64283 Darmstadt TEL: +49 6151 166651 FAX: +49 6151 165326 NET: matthias@intellektik.informatik.th-darmstadt.de ------------------------------------------------------------------------------ From @serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de Thu Oct 14 16:48:55 1993 Return-Path: <@serv-200.dfki.uni-kl.de:hoefling@dfki.uni-kl.de> Received: from uni-kl.de (stepsun.uni-kl.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15319; Thu, 14 Oct 93 16:48:55 +0100 Received: from serv-200.dfki.uni-kl.de by stepsun.uni-kl.de id aa28395; 14 Oct 93 16:47 MET Received: from serv-300.dfki.uni-kl.de by serv-200.dfki.uni-kl.de id aa01663; 14 Oct 93 15:47 GMT Received: from ws-302.dfki.uni-kl.de by serv-300.dfki.uni-kl.de id aa16165; 14 Oct 93 15:47 GMT To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: hoefling@dfki.uni-kl.de Subject: Re: mkfifo Date: Thu, 14 Oct 1993 16:46:59 +0100 From: Bjoern Hoefling Message-Id: <9310141547.aa16165@serv-300.dfki.uni-kl.de> Tomohiro Shibata wrote: > There has been some discuttion on the mailing list on how to make > bidirectional communication between clisp and other programs using Unix. > below is one solutoin provided by Mr. Haible. >> (shell "mkfifo /tmp/wish-in") >> (shell "mkfifo /tmp/wish-out") >> (setq *lisp-to-wish-stream* >> (make-two-way-stream (open "/tmp/wish-out" :direction :input) >> (open "/tmp/wish-in" :direction :output) >> ) ) >> (shell "wish < /tmp/wish-in > /tmp/wish-out") >> (delete-file "/tmp/wish-in") >> (delete-file "/tmp/wish-out") > > As for my check, it doesn't work as they are seen above. > when (open pipename-out :direction :input) is entered, prompt never comes. > But (open pipename-in :direction :output) is entered, I can get > #. > I think only it can work in case of `:direction :io '. And I tried: >> (shell "command < pipename-in > pipename-out") , > but I couln't succeed. Has Mr. Hoefling already succeeded to connect > with WISH? > Any suggestion and information are welcome. As I am directly adressed, here my response. In my masters thesis I am using clisp under linux and UNIX (Sun). The graphical user interface for my program is written in tcl/tk by another student and he has tried to explain his solution for the problem for you. His solution is rather special and is realized mostly in tcl and not in clisp, but it could be done in a similar way in C with another UNIX-process. Here is his solution: ________________________________________________________________________________ My solution works completely the other way round: I don't call wish from within lisp, but start the lisp interpreter from within the wish. I think this is more useful because normally i want to evaluate lisp functions in response to tk events. I need the "addinput patch" which is available at harbor.ecn.purdue.edu. The following code works with tcl7.0/tk3.3 and a widget shell with the appropriate patch applied. # the proc ChildProcess forks a new process from exec file cmd # and sets it's stdio to inPipe/outPipe proc ChildProcess {cmd inPipe outPipe } { if {[set childPid [fork]] == 0} { # overwrite stdin with inPipe dup $inPipe stdin # close the inPipe at the lisp side close $inPipe # do the same with stdout/stderr dup $outPipe stdout dup $outPipe stderr close $outPipe # disconnect process from terminal (useful for signal handling) id process group set # run cmd (lisp,shell, or whatever) execl $cmd # will never make it here... } return $childPid } # create fifo (named pipe) # set your path to the mknod binary here! proc mkfifo name { # set mknod /bin/mknod # linux set mknod /usr/etc/mknod # sun catch { exec $mknod $name p } msg if { "$msg" != "" } { puts stderr $msg } } # send lisp command to lisp interp. proc to-lisp {cmd} { global tclOutPipe puts $tclOutPipe $cmd flush $tclOutPipe } # signal handler proc quitCmd { } { global tclOutPipe tclInPipe tclDataInPipe catch { to-lisp "(quit)" removeinput $tclInPipe removeinput $tclDataInPipe close $tclOutPipe close $tclInPipe close $tclDataInPipe unlink [format "/tmp/wishin-%d" [pid] ] } destroy . } # handlers for receiving input from one of the pipes # the catch is needed for suns where i get some strange error message proc process-toplevel {event fileid} { catch { if { [gets $fileid line] < 0 } { quitCmd } else { echo toplevel: $line } } } proc process-data {event fileid} { catch { if { [lgets $fileid line] < 0 } { quitCmd } else { echo data: $line } } msg if { "$msg" != "" } { puts stderr $msg } } # main program: # first i create two pipes for communication to/from lisp *terminal-io*. pipe lispInPipe tclOutPipe pipe tclInPipe lispOutPipe # next i create a named pipe for data communication lisp -> wish (i want to read lisp data # from within tcl) set pipename [format "/tmp/wishin-%d" [pid] ] mkfifo $pipename # the named pipe MUST be opened non-blocking, because otherwise it will hang # until the other side is opened (see unix manual) set tclDataInPipe [open $pipename {RDONLY NONBLOCK} ] # sorry, but i don't know if this is really necessary - at least it doesn't cause harm .... fcntl $tclInPipe NONBLOCK 1 fcntl $tclDataInPipe NONBLOCK 0 # install addinput handlers addinput -read $tclInPipe "process-toplevel %E %F" addinput -read $tclDataInPipe "process-data %E %F" # fork off clisp ChildProcess clisp $lispInPipe $lispOutPipe # set signal handler so that clisp gets terminated when you kill wish signal trap { SIGTERM SIGQUIT SIGINT } quitCmd ________________________________________________________________________________ I hope this answer helps at least those who use the combination wish/clisp. ********************************************************************* Bjoern HOEFLING, DFKI, Postfach 2080, W-67608 Kaiserslautern, Germany hoefling@dfki.uni-kl.de Phone: +49-631-205-3487 Fax: -3210 ********************************************************************* From tom@jsk.t.u-tokyo.ac.jp Fri Oct 15 03:19:07 1993 Received: from monoceros (jsk.t.u-tokyo.ac.jp) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16027; Fri, 15 Oct 93 03:19:07 +0100 Received: from lupus.jsk.t.u-tokyo.ac.jp by monoceros (4.1/6.4J.6-MX) id AA01115; Fri, 15 Oct 93 11:17:06 JST Received: by lupus.jsk.t.u-tokyo.ac.jp (4.1/6.4J.6-MX) id AA00353; Fri, 15 Oct 93 11:19:21 JST Date: Fri, 15 Oct 93 11:19:21 JST From: tom@jsk.t.u-tokyo.ac.jp (Tomohiro Shibata) Return-Path: Message-Id: <9310150219.AA00353@lupus.jsk.t.u-tokyo.ac.jp> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: Matthias Lindner's message of Thu, 14 Oct 93 14:45:52 +0100 <9310141341.AA20042@aida.intellektik.informatik.th-darmstadt.de> Subject: Re: mkfifo Mr. Matthias wrote: >>I don't know, if my solution to the problem ist the most elegant one, but >>anyway - it works (at least for me ;): >>START-WISH returns a two way stream, that may be used to read from >>and write to >>the wish process. However, you should never try to use "read" or >>"read-line" >>on this stream, but use your own reader, that is implemented in terms of >>read-char-no-hang and/or listen. Also remember to use "flush output" on >>the wish-side, after every output-operation, or you will not see >>anything on the lisp-stream (until the io-buffer gets filled). >> >>Have fun >>--Matthias Thank you for your answer and your careful notes! At last I could succeed inter proccess communicating(of course I don't mention about IPC :-). I also thank you Mr. Haible and Mr. Hoefling for your quick reply. Now I will try to make a communication between Clisp and Transputer both of under Unix. I also have a dream to buy 486-machine and use Linux.:-) With BEST regards, -------------------------------- Tomohiro Shibata (email: tom@jsk.t.u-tokyo.ac.jp) Inoue-Inaba Laboratory, Department of Mechano-informatics, University of Tokyo; Japan From P2043742@csdvax.csd.unsw.EDU.AU Sat Oct 16 03:06:31 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17006; Sat, 16 Oct 93 03:06:31 +0100 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H46HANCMV48WZF36@csdvax.csd.unsw.EDU.AU>; Sat, 16 Oct 1993 12:06:18 +1000 Date: Sat, 16 Oct 1993 12:06:18 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: I need the newest version of Clisp for SGI!! To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H46HAND65E8WZF36@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Clisp Users, I am useing an old version of Clisp on a SGI Iris Indigo that is available from ma2s2.mathematik.uni-karsruhe.de because due to space limitation I am not allowed to have GCC on my machine and with the Irix native CC compiler Clisp can not be built. If somebody has allready built the newest version of clisp for SGI please post it to the above mentioned depository to make me able to download and use it on my machine. I'd approciate it. Yours, Bela Pecsek. From reggie@muon.phys.washington.edu Sat Oct 16 03:41:22 1993 Return-Path: Received: from muon.phys.washington.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17079; Sat, 16 Oct 93 03:41:22 +0100 Received: by muon.phys.washington.edu via SMTP (920330.SGI/911001.SGI) for clisp-list@ma2s2.mathematik.uni-karlsruhe.de id AA09503; Fri, 15 Oct 93 19:40:20 -0700 Message-Id: <9310160240.AA09503@muon.phys.washington.edu> To: clisp-list Subject: Re: I need the newest version of Clisp for SGI!! In-Reply-To: Your message of "Sat, 16 Oct 1993 03:08:17 BST." <01H46HAND65E8WZF36@csdvax.csd.unsw.EDU.AU> Date: Fri, 15 Oct 1993 19:40:19 -0700 From: Reggie Perry If Bruno tells me how he wants it set up, I would be glad to upload the system for you. -Reggie From tom@jsk.t.u-tokyo.ac.jp Sat Oct 16 14:53:55 1993 Received: from monoceros.jsk.t.u-tokyo.ac.jp (jsk.t.u-tokyo.ac.jp) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17284; Sat, 16 Oct 93 14:53:55 +0100 Received: by monoceros.jsk.t.u-tokyo.ac.jp (4.1/6.4J.6-MX) id AA02684; Sat, 16 Oct 93 22:51:36 JST Date: Sat, 16 Oct 93 22:51:36 JST From: tom@jsk.t.u-tokyo.ac.jp (Tomohiro Shibata) Return-Path: Message-Id: <9310161351.AA02684@monoceros.jsk.t.u-tokyo.ac.jp> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de In-Reply-To: Bruno Haible's message of Fri, 10 Sep 93 12:52:18 +0200 <9309101052.AA06445@ma2s2.mathematik.uni-karlsruhe.de> Subject: pipe communication Here is a summary to realize making CLISP communicate other process using named pipe. Mr. Haible wrote: >> (shell "mkfifo /tmp/wish-in") >> (shell "mkfifo /tmp/wish-out") >> (setq *lisp-to-wish-stream* >> (make-two-way-stream (open "/tmp/wish-out" :direction :input) >> (open "/tmp/wish-in" :direction :output) >> ) ) >> (shell "wish < /tmp/wish-in > /tmp/wish-out") >> (delete-file "/tmp/wish-in") >> (delete-file "/tmp/wish-out") But it doesn't work at the (open .... :input). I got a Mr.Matthias's program which works correctly by this mailing-list, and found differece of two programs, that is: 1. opening named pipe for input can succeed if it has already connected to output of any program. 2. Order to open two streams must be that fist output and second input. (defun open-test nil (setq cmd (format NIL "program < /tmp/in > /tmp/out 2>&1 &")) (shell (format NIL "exec /bin/sh -c ~S" cmd)) (setq outf (open "/tmp/in" :direction :output)) (setq inf (open "/tmp/out" :direction :input))) So, simple example is a above function which can succeed. Sincerely. -------------------------------- Tomohiro Shibata (email: tom@jsk.t.u-tokyo.ac.jp) Inoue-Inaba Laboratory, Department of Mechano-informatics, University of Tokyo; Japan From marcusli@io.org Sun Oct 17 00:29:24 1993 Return-Path: Received: from io.org by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17770; Sun, 17 Oct 93 00:29:24 +0100 Received: by io.org (Smail3.1.28.1 #4) id m0ooL0M-0004cVC; Sat, 16 Oct 93 19:25 EDT Message-Id: From: marcusli@io.org (Marcus Li) Subject: clisp for MacOS? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Sat, 16 Oct 1993 19:25:46 -0400 (EDT) X-Mailer: ELM [version 2.4 PL22] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 127 Hello, I am looking for a MacOS version of clisp. Please email info if there is such a thing. Thanks in advance. Marcus Li From jbishop@babel.ling.nwu.edu Sun Oct 17 17:38:26 1993 Received: from relay.acns.nwu.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18064; Sun, 17 Oct 93 17:38:26 +0100 Received: from babel.ling.nwu.edu by relay.acns.nwu.edu (4.1/SMI-ACNS-relay-1.7) id AA04070; Sun, 17 Oct 93 11:36:05 CDT Return-Path: Received: by babel.ling.nwu.edu (4.1/SMI-4.0) id AA21478; Sun, 17 Oct 93 11:35:36 CDT Date: Sun, 17 Oct 1993 11:32:54 -0500 (CDT) From: Jeff Bishop Subject: Illegal characters/ .LSP files To: Clisp List Cc: Bruno Haible Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I have found that whenever I edit any Lisp files on the OS/2 system editor (or any of several other text editors which run on the same platform) I get an error message when I attempt to load them. It says ;; Loading file C:\Common Lisp\chapter5.lsp ... *** - READ from #: ill egal character #\ Jeff Bishop jbishop@nwu.edu From jbishop@babel.ling.nwu.edu Sun Oct 17 17:42:29 1993 Received: from relay.acns.nwu.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18118; Sun, 17 Oct 93 17:42:29 +0100 Received: from babel.ling.nwu.edu by relay.acns.nwu.edu (4.1/SMI-ACNS-relay-1.7) id AA04433; Sun, 17 Oct 93 11:40:11 CDT Return-Path: Received: by babel.ling.nwu.edu (4.1/SMI-4.0) id AA21484; Sun, 17 Oct 93 11:39:42 CDT Date: Sun, 17 Oct 1993 11:37:10 -0500 (CDT) From: Jeff Bishop Subject: Illegal characters/ .LSP files To: Clisp List Cc: Bruno Haible Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I have found that whenever I edit any Lisp files on the OS/2 system editor (or any of several other text editors which run on the same platform) I get an error message when I attempt to load them. It says ;; Loading file C:\Common Lisp\chapter5.lsp ... *** - READ from #: ill egal character #\ I *suspect* it might have something to do with ^M at the end of a line, since I have experienced no such problems with the Unix version. Does anyone know how to edit the files to make them readable in a DOS, Windows, or OS/2 environment? Jeff Bishop jbishop@nwu.edu From LARSSON@ntcclu.ntc.nokia.com Mon Oct 18 09:47:42 1993 Return-Path: Received: from NTC02.TELE.NOKIA.FI by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18517; Mon, 18 Oct 93 09:47:42 +0100 Date: Mon, 18 Oct 1993 10:46:26 +0300 (EET) From: LARSSON@ntcclu.ntc.nokia.com Message-Id: <931018104626.208065b8@ntcclu.ntc.nokia.com> Subject: RE: Illegal characters/ .LSP files To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X-Vmsmail-To: SMTP%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Jeff Bishop asked: I have found that whenever I edit any Lisp files on the OS/2 system editor (or any of several other text editors which run on the same platform) I get an error message when I attempt to load them. It says ;; Loading file C:\Common Lisp\chapter5.lsp ... *** - READ from #: ill egal character #\ I *suspect* it might have something to do with ^M at the end of a line, since I have experienced no such problems with the Unix version. Does anyone know how to edit the files to make them readable in a DOS, Windows, or OS/2 environment? This problem is caused by the End Of File character (Ctrl-Z in DOS and OS/2). You can avoid it by using the TE/2 editor (download from ftp-os2.cdrom.com or nic.funet.fi), which does NOT write Ctrl-Z at the end of files. I think GNU Emacs will do the same service, but I haven't tested it yet. Arne *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Arne Larsson Nokia Telecommunications Translator Transmission Systems, Customer Services larsson@ntc02.tele.nokia.fi P.O. Box 12, SF-02611 Espoo, Finland larsson@ntcclu.ntc.nokia.com Phone +358 0 5117476, Fax +358 0 51044287 Arne.Larsson@ntc.nokia.com *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* From johann@mail4.ai.univie.ac.at Mon Oct 18 13:38:23 1993 Return-Path: Received: from milano.ai.univie.ac.at by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18836; Mon, 18 Oct 93 13:38:23 +0100 Received: by milano.ai.univie.ac.at id AA08273 (5.65c/IDA-1.4.4 for clisp-list ); Mon, 18 Oct 1993 13:35:27 +0100 Date: Mon, 18 Oct 1993 13:35:27 +0100 From: Johann Petrak Message-Id: <199310181235.AA08273@milano.ai.univie.ac.at> To: clisp-list Subject: RE: Illegal characters/ .LSP files In-Reply-To: <931018104626.208065b8@ntcclu.ntc.nokia.com> References: <931018104626.208065b8@ntcclu.ntc.nokia.com> LARSSON@ntcclu.ntc.nokia.com writes: > Jeff Bishop asked: > > I have found that whenever I edit any Lisp files on the OS/2 system > editor (or any of several other text editors which run on the same > platform) I get an error message when I attempt to load them. It says > ;; Loading file C:\Common Lisp\chapter5.lsp ... > *** - READ from # Lisp\\chapter5.lsp">: ill > egal character #\ > > I *suspect* it might have something to do with ^M at the end of a line, > since I have experienced no such problems with the Unix version. Does > anyone know how to edit the files to make them readable in a DOS, > Windows, or OS/2 environment? > > This problem is caused by the End Of File character (Ctrl-Z > in DOS and OS/2). You can avoid it by using the TE/2 editor > (download from ftp-os2.cdrom.com or nic.funet.fi), which does > NOT write Ctrl-Z at the end of files. > > I think GNU Emacs will do the same service, but I haven't > tested it yet. Another way is to tell the LISP reader to treat Ctrl-Z as whitespace: (set-syntax-from-char #\Code26 #\Space) Johann Petrak, Email: johann@ai.univie.ac.at Phone: +43/222/533-61-12 Austrian Research Institute for Schottengasse 3 Artificial Intelligence A-1010 Vienna, Austria From laytonj@sun.soe.clarkson.edu Tue Oct 19 00:23:47 1993 Return-Path: Received: from omnigate.clarkson.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20226; Tue, 19 Oct 93 00:23:47 +0100 Received: from sun.soe.clarkson.edu by omnigate.clarkson.edu with SMTP (PP) id <06454-0@omnigate.clarkson.edu>; Mon, 18 Oct 1993 19:23:05 -0400 Received: from eagle1.mie.clarkson.edu.noname by sun.soe.clarkson.edu (4.1/SMI-4.1) id AA02827; Mon, 18 Oct 93 19:20:51 EDT Date: Mon, 18 Oct 93 19:20:51 EDT From: laytonj@sun.soe.clarkson.edu (jeffrey b.layton) Message-Id: <9310182320.AA02827@sun.soe.clarkson.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Real Newbie Question Sorry to bother everyone, I just joined the list (actually several days) and didn't see any postings, so I thought I would ask a question. I just got the clisp sources and tried to compile them on my Sparc10 under SunOS 4.1.3. I followed the instructions in the INSTALL file in the unix subdirectory. However, when I went the directory where I'm going to install clisp, I couldn't makemake. I've searched in the src files and I can't find makemake (only makemake.in). I'd really like to compile this on my machine (rather than downloadthe binaries). Any ideas what I'm doing wrong? Thanks very much, Jeff Layton (Internet: laytonj@sun.soe.clarkson.edu) From RLS@waikato.ac.nz Tue Oct 19 08:44:00 1993 Return-Path: Received: from grace.waikato.ac.nz by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20424; Tue, 19 Oct 93 08:44:00 +0100 Message-Id: <9310190744.AA20424@ma2s2.mathematik.uni-karlsruhe.de> Date: Tue, 19 Oct 93 20:26 +1300 From: RLS@waikato.ac.nz Subject: RE: Illegal characters/ .LSP files To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" In reply to Mr. Bishops problems with loading files, I too have found the same problem when loading dos format text files into CLisp on Unix. If you delete the ^M's and ^Z at the end of the file then Unix doesn't mind them any more---actually it's only the ^Z, the ^M's aren't a problem. This is because Unix takes ^D as eof while dos takes ^Z (I think!). However OS/2 CLisp should take files created with an OS/2 editor---otherwise how could an OS/2-ite survive? Richard Shepherd. From haible Tue Oct 19 17:13:03 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21372; Tue, 19 Oct 93 17:13:03 +0100 Date: Tue, 19 Oct 93 17:13:03 +0100 From: haible (Bruno Haible) Message-Id: <9310191613.AA21372@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Installation Question Jeff Layton has a problem when installing clisp: > However, when I went the directory > where I'm going to install clisp, I couldn't makemake. I've searched in the > src files and I can't find makemake (only makemake.in). The makemake file is created by the "target" command in the directory you specify as argument to "target". Did you interrupt the "target" command, or did it fail in some other way? Retry the "target ..." command and observe if you see any error message. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Tue Oct 19 17:50:50 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21509; Tue, 19 Oct 93 17:50:50 +0100 Date: Tue, 19 Oct 93 17:50:50 +0100 From: haible (Bruno Haible) Message-Id: <9310191650.AA21509@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Illegal characters/ .LSP files Some DOS and OS/2 users (Mr. Bishop being one of them) have been affected by an editor which appends Ctrl-Z to text files, the OS/2 "enhanced" editor for example. Richard Shepherd proposes: > However OS/2 CLisp should take files created with an OS/2 editor--- > otherwise how could an OS/2-ite survive? Ctrl-Z was used as EOF character 10 years ago. I think it dates back to operating systems like CP/M which didn't store a file's length in bytes, only its size in disk blocks. It is of no use at all nowadays. > This is because Unix takes ^D as eof while dos takes ^Z (I think!). Yes, with the distinction that Unix doesn't save the Ctrl-D in the file. My advice is to use an editor which does not append this prehistoric character to your files. Arne Larsson mentioned one, and there are many more good editors around. Only if I can load the clisp executable into an editor, save it, and it comes out unmodified, only then I consider that editor usable for day to day work. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From dliu@faraday-gw.njit.edu Tue Oct 19 19:52:19 1993 Return-Path: Received: from njitgw.njit.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21818; Tue, 19 Oct 93 19:52:19 +0100 Received: by njitgw.njit.edu (5.57/Ultrix3.0-C) id AA20948; Tue, 19 Oct 93 14:49:31 -0400 Received: from ark.njit.edu by faraday-gw.njit.edu (4.1/SMI-4.1) id AA29100; Tue, 19 Oct 93 14:49:22 EDT Date: Tue, 19 Oct 93 14:49:22 EDT From: dliu@faraday-gw.njit.edu (Dong Liu) Received: by ark.njit.edu (4.1/SMI-4.1) id AA13036; Tue, 19 Oct 93 14:49:22 EDT To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: compiler error when compiling garnet 2.2 Mailer: VM 5.32 (beta) for GNU Emacs 18.59.5 Message-Id: I installed clisp-1993-09-01 on both a Sun Sparc2 and a Linux machines, when compiling the new garnet version 2.2, clisp stops at compiling kr/kr.lsp, the error message is *** - Compiler bug!! Occurred in OPTIMIZE-LABEL. Dose anyone have a solution? Thanks -dong From haible Tue Oct 19 20:03:45 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21890; Tue, 19 Oct 93 20:03:45 +0100 Date: Tue, 19 Oct 93 20:03:45 +0100 From: haible (Bruno Haible) Message-Id: <9310191903.AA21890@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: compiler error when compiling garnet 2.2 This has already been discussed on this mailing list: > clisp stops at compiling kr/kr.lsp, the error message is > *** - Compiler bug!! Occurred in OPTIMIZE-LABEL. > Does anyone have a solution? Get clisp-1993-09-27. Same place as usual. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From dxs@evolving.com Thu Oct 21 06:18:50 1993 Return-Path: Received: from csn.org by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA24305; Thu, 21 Oct 93 06:18:50 +0100 Received: by csn.org with UUCP id AA00801 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Wed, 20 Oct 1993 23:15:51 -0600 Received: by kafka (AIX 3.2/UCB 5.64/4.03) id AA23474; Mon, 18 Oct 1993 12:50:29 -0600 Date: Mon, 18 Oct 1993 12:50:29 -0600 From: dxs@evolving.com (Dan Stanger) Message-Id: <9310181850.AA23474@kafka> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: clisp source file expansion i downloaded the clisp source from ftp.cs.cmu.edu. the files have a extension .z. does anyone have source to the program to expand them or could they tell me where to find the source in the usa? thanks, dan stanger From RLS@waikato.ac.nz Thu Oct 21 09:08:25 1993 Return-Path: Received: from truth.waikato.ac.nz by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA24472; Thu, 21 Oct 93 09:08:25 +0100 Message-Id: <9310210808.AA24472@ma2s2.mathematik.uni-karlsruhe.de> Date: Thu, 21 Oct 93 21:05 +1300 From: RLS@waikato.ac.nz Subject: RE: clisp source file expansion To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" My gues would be gzip, becoming standard on unix. If you have access to a unix machine it may already be there. The syntax is gzip -d file where file.z is your compressed stuff. gzip can also be obtained for DOS, the syntax to use it exactly the same as above. If you don't have gzip I believe it is available from many ftp sites (though I can't name one for you!). Richard Shepherd. From dxs@evolving.com Sat Oct 23 13:03:37 1993 Return-Path: Received: from nz11 by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AB26130; Sat, 23 Oct 93 13:03:37 +0100 Received: from co.edu by nz11.rz.uni-karlsruhe.de with SMTP (PP) id <02856-0@nz11.rz.uni-karlsruhe.de>; Fri, 22 Oct 1993 22:14:06 +0100 Received: by mail-2.csn.net with UUCP id AA13733 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Fri, 22 Oct 1993 15:12:42 -0600 Received: from tunabnc by kafka (AIX 3.2/UCB 5.64/4.03) id AA24129; Fri, 22 Oct 1993 14:41:13 -0600 Received: by tunabnc (AIX 3.2/UCB 5.64/4.03) id AA41735; Fri, 22 Oct 1993 14:40:29 -0600 Date: Fri, 22 Oct 1993 14:40:29 -0600 From: dxs@evolving.com (Dan Stanger) Message-Id: <9310222040.AA41735@tunabnc> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: building clisp-1993-08-22 i am having a problem building. there is a problem in the configure in src where the code is as follows. # A program that outputs its argument count: cat > conftest.c < #ifdef __STDC__ main (int argc, char** argv) #else main () int argc; char** argv; #endif { printf("%d\n",argc); exit(0); } EOF eval $compile if i dont use the ansi c compiler, then the code fails because the line containing main should be main(argc, argv) int argc; char** argv; is this a error in configure? if i try to use the ansi c compiler, i get errors in machine.h i am trying to port this to a rs6000. From haible Sat Oct 23 14:36:42 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA26378; Sat, 23 Oct 93 14:36:42 +0100 Date: Sat, 23 Oct 93 14:36:42 +0100 From: haible (Bruno Haible) Message-Id: <9310231336.AA26378@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: building clisp-1993-08-22 on RS6000 Dan Stanger notes: > there is a problem in the configure > the line containing main should be > main(argc, argv) int argc; char** argv; You are right. Mr Wolfgang Golubski told me about it, and it has been fixed in clisp-1993-09-27. > i am trying to port this to a rs6000. There are RS/6000 binaries in ma2s2:/pub/lisp/clisp/binaries/rs6000/. They have been provided by Gabor Herr . Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From rv-server@cs.brown.edu Sat Oct 23 23:57:08 1993 Return-Path: Received: from cs.brown.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16488; Sat, 23 Oct 93 23:57:08 +0100 Received: from vegas.cs.brown.edu by cs.brown.edu (5.64+/Bullwinkle-1.7) id AA25688; Sat, 23 Oct 93 18:53:41 -0400 Date: Sat, 23 Oct 93 18:53:41 -0400 From: rv-server@cs.brown.edu Message-Id: <9310232253.AA25688@cs.brown.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: building clisp on Linux box I tried but failed building clisp on a linux box. The configure, though lengthy, completed successfully, but compilation exhausted virtual memory. My 386-64k 40MHz 8M ram 8M swap linux box choked when compiling the file spvwtabs.c any ideas? rodrigo vanegas rv@cs.brown.edu From haible Sun Oct 24 00:05:55 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA16554; Sun, 24 Oct 93 00:05:55 +0100 Date: Sun, 24 Oct 93 00:05:55 +0100 From: haible (Bruno Haible) Message-Id: <9310232305.AA16554@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: building clisp on Linux box Rodrigo Vanegas observes: > My 386-64k 40MHz 8M ram 8M swap linux box choked when compiling the > file spvwtabs.c Yes, when using gcc-2.4.5 you need 8 MB RAM and 10 MB swap space. You can get away with much less memory if you install gcc-2.5.0. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From ian@calvin.ds.boeing.com Wed Oct 27 00:00:54 1993 Return-Path: Received: from atc.boeing.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20523; Wed, 27 Oct 93 00:00:54 +0100 Received: by atc.boeing.com (5.57) id AA26303; Tue, 26 Oct 93 15:58:13 -0700 Received: from calvin.ds.boeing.com (calvin.ds.boeing.com) by mercury.boeing.com (4.1/SMI-4.1) id AA28874; Tue, 26 Oct 93 16:05:32 PDT Received: by calvin.ds.boeing.com (5.57/Ultrix3.0-C) id AA21535; Tue, 26 Oct 93 15:57:11 -0700 Date: Tue, 26 Oct 93 15:57:11 -0700 From: ian@calvin.ds.boeing.com (Ian Searle) Message-Id: <9310262257.AA21535@ calvin.ds.boeing.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Need Help with SVr4 Installation. Hello, I have successfully installed CLISP on Sun and DEC workstations, but I am having a very difficult time with CLISP on a 486 running SVr4.0. I get assembler errors using gcc. Using cc I can get it to compile, but lisp.run won't do anything because it thinks there is only 66k of memory (this is not correct). I am adding -DNO_MULTIMAP_SHM to the CFLAGS def in the makefile. I saw the note in there that says: (*) The lisp.run you get is a program that reliably crashes your machine. But I was hoping that this could be resolved. Can anybody out there help? Thanks, Ian Searle From timcenko@cs.columbia.edu Wed Oct 27 16:47:50 1993 Return-Path: Received: from cs.columbia.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21369; Wed, 27 Oct 93 16:47:50 +0100 Received: from ground.cs.columbia.edu by cs.columbia.edu (5.65c/0.6/jba+ad+jtt for clisp-list@ma2s2.mathematik.uni-karlsruhe.de) with SMTP id AA08033; Wed, 27 Oct 1993 11:43:41 -0400 Received: by ground.cs.columbia.edu id AA18693 (5.65c/IDA-1.4.4.5/jba/jtt/ad for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Wed, 27 Oct 1993 11:43:40 -0400 Date: Wed, 27 Oct 93 11:43:40 EDT From: Aleksandar Timcenko To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Question: how to increase the stack Message-Id: I am running clisp DOS/windows on a 486 with 8M RAM + 16M swap. The application that I am trying to port from SUN compiles smoothly but when it runs I get the following stack too small! esp A1A1C top B2000 size 10000 Program terminate by signal 9 , SIGKILL Is there any way around this? I'd appreciate any comments/suggestions/condolences. Thanks, Alex From snfeldma%teebox.franken.de@castle.franken.de Thu Oct 28 05:01:05 1993 Return-Path: Received: from faui45.informatik.uni-erlangen.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21846; Thu, 28 Oct 93 05:01:05 +0100 Received: from castle.franken.de by uni-erlangen.de with UUCP; id AA17361 (5.65c-5/7.3v-FAU); Thu, 28 Oct 1993 04:56:54 +0100 Received: by castle.franken.de (V1.16/wusel) id <7tcw@castle.franken.de>; Thu, 28 Oct 93 01:13:49 MEZ Received: by teeBox.franken.de (Smail3.1.28.1 #6) id m0osL2H-000DCoC; Thu, 28 Oct 93 00:16 GMT Message-Id: Date: Thu, 28 Oct 1993 00:16:16 +0000 (GMT) In-Reply-To: <9310262257.AA21535@ calvin.ds.boeing.com> from "Ian Searle" at Oct 27, 93 00:03:37 am X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2839 From: snfeldma@teebox.franken.de (Sebastian Feldmann) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Need Help with SVr4 Installation. > > Hello, I have successfully installed CLISP on Sun and DEC > workstations, but I am having a very difficult time with CLISP on a > 486 running SVr4.0. I get assembler errors using gcc. Using cc I can > get it to compile, but lisp.run won't do anything because it thinks > there is only 66k of memory (this is not correct). I am adding > -DNO_MULTIMAP_SHM to the CFLAGS def in the makefile. I saw the note in > there that says: > > (*) The lisp.run you get is a program that reliably crashes your machine. > [...] I am using a 486 SVR4.2 and have the same problem. During compilation I get *many* warnings of the kind: " "lispbibl.d", line 6764: warning: trigraph sequence replaced "spvw.d", line 1784: warning: integer overflow detected: op "<<" "spvw.d", line 1797: warning: integer overflow detected: op "<<" "spvw.d", line 1798: warning: integer overflow detected: op "<<" "spvw.d", line 1817: warning: integer overflow detected: op "<<" "spvw.d", line 5270: warning: integer overflow detected: op "<<" ... "spvw.d", line 6027: warning: improper pointer/integer combination: op "=" "spvw.d", line 6029: warning: improper pointer/integer combination: op "=" "spvw.d", line 6135: warning: argument is incompatible with prototype: arg #2 "spvw.d", line 6178: warning: argument is incompatible with prototype: arg #2 "spvw.d", line 6180: warning: argument is incompatible with prototype: arg #2 "spvw.d", line 6733: warning: assignment type mismatch ... "eval.d", line 2149: warning: integer overflow detected: op "<<" "eval.d", line 2706: warning: integer overflow detected: op "<<" "eval.d", line 2777: warning: integer overflow detected: op "<<" "lisparit0.d", line 2119: warning: integer overflow detected: op "<<" "lisparit0.d", line 2150: warning: integer overflow detected: op "<<" "lisparit0.d", line 2150: warning: integer overflow detected: op "U-" "lisparit0.d", line 2203: warning: integer overflow detected: op "<<" "lisparit0.d", line 2203: warning: integer overflow detected: op "-" "lisparit0.d", line 3839: warning: integer overflow detected: op "<<" " a first look for the reason directs me to " oint_addr_shift-addr_shift " so it seems to be a wrong definition of oint_addr_shift and/or addr_shift or some kind of macro-trouble in macros using these I looked at clisp for the first time a few days ago. So chances are not very high I'll solve that problem soon. Any help welcome ... so long Sebastian +-----------------------------------------------------------------------------+ | Sebastian Feldmann snfeldma@teebox.franken.de Fax/Data:(49) 9122-61696 | | do Smalltalk or Scheme to avoid brain damage ogin: nuucp word: nuucp | | InfoFiles als: ~/info or ~/index | +-----------------------------------------------------------------------------+ From ian@calvin.ds.boeing.com Thu Oct 28 16:08:22 1993 Return-Path: Received: from atc.boeing.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23514; Thu, 28 Oct 93 16:08:22 +0100 Received: by atc.boeing.com (5.57) id AA25720; Thu, 28 Oct 93 08:05:22 -0700 Received: from calvin.ds.boeing.com (calvin.ds.boeing.com) by mercury.boeing.com (4.1/SMI-4.1) id AA04640; Thu, 28 Oct 93 08:12:39 PDT Received: by calvin.ds.boeing.com (5.57/Ultrix3.0-C) id AA13351; Thu, 28 Oct 93 08:04:19 -0700 Date: Thu, 28 Oct 93 08:04:19 -0700 From: ian@calvin.ds.boeing.com (Ian Searle) Message-Id: <9310281504.AA13351@ calvin.ds.boeing.com> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: In-Reply-To: (snfeldma@teebox.franken.de) Subject: Re: Need Help with SVr4 Installation. >"lisparit0.d", line 2203: warning: integer overflow detected: op "<<" >"lisparit0.d", line 2203: warning: integer overflow detected: op "-" >"lisparit0.d", line 3839: warning: integer overflow detected: op "<<" > >" > >a first look for the reason directs me to " oint_addr_shift-addr_shift " >so it seems to be a wrong definition of oint_addr_shift and/or addr_shift >or some kind of macro-trouble in macros using these > >I looked at clisp for the first time a few days ago. So chances are not very >high I'll solve that problem soon. > >Any help welcome ... > >so long > Sebastian I have decided to ignore these compiler warnings for the moment. I got the same kind of warnings when I compiled CLISP on a Sun-4 and a DEC 5000, and CLISP seems to run OK (at least well enough to support MAXIMA). -Ian Searle From haible Thu Oct 28 16:29:22 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23635; Thu, 28 Oct 93 16:29:22 +0100 Date: Thu, 28 Oct 93 16:29:22 +0100 From: haible (Bruno Haible) Message-Id: <9310281529.AA23635@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Need Help with SVr4 Installation. > During compilation I get *many* warnings of the kind: > "lispbibl.d", line 6764: warning: trigraph sequence replaced > "spvw.d", line 1784: warning: integer overflow detected: op "<<" > "lisparit0.d", line 2119: warning: integer overflow detected: op "<<" > ... All these warnings are harmless. I have seen them all and checked the clisp source. On the other hand, if the CLISP compiler emits warnings about your Lisp code, you should better take a look at them. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From xjam@ginkgo.CS.Berkeley.EDU Fri Oct 29 00:02:36 1993 Return-Path: Received: from ginkgo.CS.Berkeley.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23912; Fri, 29 Oct 93 00:02:36 +0100 Received: from localhost (xjam@localhost) by ginkgo.CS.Berkeley.EDU (8.6.2/8.5) id PAA14388; Thu, 28 Oct 1993 15:59:07 -0700 Message-Id: <199310282259.PAA14388@ginkgo.CS.Berkeley.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: call-next-method in CLOS Reply-To: xjam@cork.CS.Berkeley.EDU Date: Thu, 28 Oct 1993 15:59:06 -0700 From: (The Crossjammer) While compiling clue/shells.lsp I ran across a call to call-next-method which actually passes arguments. The builtin CLOS of CLISP seems to disallow this but my CLTL2 copy says it's applicable. Any quick fixes and/or clarifications? xjam@cork.Berkeley.EDU Fiiiive Thousand Boomin Watts...... -The JBeez They can't come on and play me in prime time, | Free Millipede coin op Cause I know the time, cause I'm gettin' mine. | and House music at my I get on the mix late in the night... -Public Enemy | place! Drop by sometime. From haible Fri Oct 29 19:43:57 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA25178; Fri, 29 Oct 93 19:43:57 +0100 Date: Fri, 29 Oct 93 19:43:57 +0100 From: haible (Bruno Haible) Message-Id: <9310291843.AA25178@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: call-next-method in CLOS Brian Dennis writes: > While compiling clue/shells.lsp I ran across a call to call-next-method > which actually passes arguments. The builtin CLOS of CLISP seems to > disallow this but my CLTL2 copy says it's applicable. The builtin CLOS of CLISP doesn't support CALL-NEXT-METHOD with arguments. > Any quick fixes and/or clarifications? Try this one: *** clue/shells.lsp% Fri Oct 29 19:33:11 1993 --- clue/shells.lsp Fri Oct 29 19:33:11 1993 *************** *** 127,135 **** (defmethod (setf contact-state) :around (new-state (shell override-shell)) ;; :iconic is equivalent to :withdrawn for override-shell ! (call-next-method ! (if (eq :iconic new-state) :withdrawn new-state) ! shell)) --- 127,135 ---- (defmethod (setf contact-state) :around (new-state (shell override-shell)) ;; :iconic is equivalent to :withdrawn for override-shell ! (if (eq :iconic new-state) ! ((setf contact-state) :withdrawn shell) ! (call-next-method))) One such a usage of CALL-NEXT-METHOD also appeared in the Apply2C compiler, and it could be patched away like this one. As long as no package really needs CALL-NEXT-METHOD with arguments, I won't implement it. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From pao@ai.mit.edu Fri Oct 29 20:21:27 1993 Return-Path: Received: from life.ai.mit.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA25320; Fri, 29 Oct 93 20:21:27 +0100 Received: from SLUGGO.AI.MIT.EDU by life.ai.mit.edu (4.1/AI-4.10) for clisp-list@ma2s2.mathematik.uni-karlsruhe.de id AA22023; Fri, 29 Oct 93 15:16:42 EDT Date: Fri, 29 Oct 1993 15:16-0400 From: Patrick A. O'Donnell Subject: Re: call-next-method in CLOS To: clisp-list In-Reply-To: <9310291843.AA25178@ma2s2.mathematik.uni-karlsruhe.de> Message-Id: <19931029191620.0.PAO@SLUGGO.AI.MIT.EDU> Date: Fri, 29 Oct 1993 14:45 EDT From: haible@ma2s2.mathematik.uni-karlsruhe.de (Bruno Haible) Brian Dennis writes: > Any quick fixes and/or clarifications? Try this one: *** clue/shells.lsp% Fri Oct 29 19:33:11 1993 --- clue/shells.lsp Fri Oct 29 19:33:11 1993 *************** *** 127,135 **** (defmethod (setf contact-state) :around (new-state (shell override-shell)) ;; :iconic is equivalent to :withdrawn for override-shell ! (call-next-method ! (if (eq :iconic new-state) :withdrawn new-state) ! shell)) --- 127,135 ---- (defmethod (setf contact-state) :around (new-state (shell override-shell)) ;; :iconic is equivalent to :withdrawn for override-shell ! (if (eq :iconic new-state) ! ((setf contact-state) :withdrawn shell) ! (call-next-method))) One such a usage of CALL-NEXT-METHOD also appeared in the Apply2C compiler, and it could be patched away like this one. As long as no package really needs CALL-NEXT-METHOD with arguments, I won't implement it. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de To be fair, you should point out that this changes the operation of (setf contact-state) for override-shell, specifically for any subclass of override-shell with daemon methods on (setf contact-state). - Pat From snfeldma%teebox.franken.de@castle.franken.de Sat Oct 30 22:12:26 1993 Return-Path: Received: from faui45.informatik.uni-erlangen.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00588; Sat, 30 Oct 93 22:12:26 +0100 Received: from castle.franken.de by uni-erlangen.de with UUCP; id AA11426 (5.65c-5/7.3v-FAU); Sat, 30 Oct 1993 22:07:46 +0100 Received: by castle.franken.de (V1.16/wusel) id <7x09@castle.franken.de>; Sat, 30 Oct 93 20:15:53 MEZ Received: by teeBox.franken.de (Smail3.1.28.1 #6) id m0otLqR-000DEZC; Sat, 30 Oct 93 19:20 GMT Message-Id: Date: Sat, 30 Oct 1993 19:20:14 +0000 (GMT) X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 777 From: snfeldma@teebox.franken.de (Sebastian Feldmann) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: HEUREKA 486/SVR4.2 Salve, just fetched a newer version of clisp ... and all my trouble with that "66k of memory is not enough for me" has gone away :-) my configuration now is: 486/50 16MB - SVR4.2 (Onsite) clisp-1993-09-27 add -DNO_MULTIMAP_SHM -DSVR4 -DUHC (as described in unix/PLATFORMS) add -lsocket -lnsl (in the makefile lisp.run: ....) and all went fine so long Sebastian +-----------------------------------------------------------------------------+ | Sebastian Feldmann snfeldma@teebox.franken.de Data: 09122-61696 | | do Smalltalk or Scheme to avoid brain damage ogin: nuucp word: nuucp | | +clisp InfoFiles als: ~/info or ~/index | +-----------------------------------------------------------------------------+ From ROUSSEL@hep.Physics.McGill.CA Fri Nov 5 04:15:11 1993 Return-Path: Received: from hepvs6.Physics.McGill.CA by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12194; Fri, 5 Nov 93 04:15:11 +0100 Date: Thu, 4 Nov 1993 22:09:32 -0500 (EST) From: ROUSSEL@hep.Physics.McGill.CA Message-Id: <931104220932.22001936@hep.Physics.McGill.CA> Subject: Problems with compilation of maxima under clisp(os/2 version) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X-Vmsmail-To: SMTP%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Hello everybody. I'm trying to compile maxima under the os/2 version of clisp but I'm having a problem. At some point in the compilation of maxima there is a file, mload.fas, which won't load because "the function user-homedir-pathname is undefined". user-homedir-pathname appears twice in the source code of mload.lsp but indeed, it is not defined. The only solution that I've found to this problem was to delete this name. I would like to know if this is going to give me problems when running maxima, and if there's another (better) solution to this problem. Thanks for any help. Harold Roussel From haible Mon Nov 8 18:37:03 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA17856; Mon, 8 Nov 93 18:37:03 +0100 Date: Mon, 8 Nov 93 18:37:03 +0100 From: haible (Bruno Haible) Message-Id: <9311081737.AA17856@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2 Subject: New CLISP version Once more, there is a new version of CLISP at the usual place: ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp/source . Excerpt from the change log: User visible changes -------------------- * The STDWIN library can now be made part of CLISP by calling makemake with argument --with-stdwin. * X3J13 vote <108> is implemented: New packages COMMON-LISP and COMMON-LISP-USER. * X3J13 vote <112> is implemented: LOAD binds the variables *LOAD-PATHNAME* and *LOAD-TRUENAME*, COMPILE-FILE binds the variables *COMPILE-FILE-PATHNAME* and *COMPILE-FILE-TRUENAME*. New variables *LOAD-PATHNAME*, *LOAD-TRUENAME*, *COMPILE-FILE-PATHNAME*, *COMPILE-FILE-TRUENAME*. * X3J13 vote <26> is implemented: COMPILE-FILE has a keyword argument :PRINT. New variables *COMPILE-WARNINGS*, *COMPILE-VERBOSE*, *COMPILE-PRINT*. * Fixed two bugs in DEFPACKAGE: *PACKAGE* is not modified by DEFPACKAGE any more, and the :INTERN option now works. * Fixed FUNCTION-LAMBDA-EXPRESSION such that it works on built-in functions. Thanks to France Dacar. * Fixed a serious bug in GCD on 16-bit processors: (GCD x y) with 2^k - 2^(k-16) <= |x|,|y| < 2^k could cause a division by zero. Portability ----------- * Added support for DEC Alpha AXP running OSF/1 Unix. * Added support for Apple Mac II running A/UX. * Added support for Atari ST/TT running MiNT. Thanks to Scott Kolodzieski. * Support for SunOS4 cc now works. * The makefile now also works with "parallel" make utilities (e.g. GNU make with option -j). Thanks to Rick Sladkey. * Easier porting: The most dangerous optimizations (inline asm() statements, global register allocations, ...) are inhibited if CLISP is compiled with option SAFETY=3. * Some modifications for the sake of gcc-2.5. Other modifications ------------------- * Speed up generic function dispatch. * Some kinds of objects take less space (all CPUs except 680x0). * The list of X3J13 votes and their current status in CLISP has been moved from file `src/defs2.lsp' to file `src/cltl2.txt'. * The code for setting up a connection to an X server (file `src/socket.d') has been completely rewritten. * Miscellaneous documentation updates. New binaries have been built for: DEC Alpha AXP, OSF 1.3 1375711 Oct 31 02:10 decalpha-osf/clisp.tar.z IBM RS/6000, AIX 3.2 1357509 Oct 15 09:40 rs6000/clisp.tar.z Binaries have been rebuilt for the following architectures: Sun3, SunOS 4 1055703 Oct 13 19:08 sun3-sunos4/clisp.tar.z HP 9000/800, HP-UX 8 1115787 Oct 23 23:44 hp9000s800/clisp.tar.z Sun4, SunOS 4 1044615 Nov 2 19:11 sun4-sunos4/clisp-english.tar.z Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From xjam@ginkgo.CS.Berkeley.EDU Wed Nov 10 08:02:28 1993 Return-Path: Received: from ginkgo.CS.Berkeley.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20136; Wed, 10 Nov 93 08:02:28 +0100 Received: from localhost (xjam@localhost) by ginkgo.CS.Berkeley.EDU (8.6.4/8.5) id WAA06569; Tue, 9 Nov 1993 22:56:19 -0800 Message-Id: <199311100656.WAA06569@ginkgo.CS.Berkeley.EDU> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Clisp, ilisp and completion Reply-To: xjam@cork.CS.Berkeley.EDU Date: Tue, 09 Nov 1993 22:56:18 -0800 From: (The Crossjammer) Does the command line completion in clisp mess up anyone elses' use of clisp with ilisp? I'm not using the clisp specific stuff that's in the archive so that may be the solution. Alternatively, is there any means to disable the readline library without rebuilding clisp? --Bri From matthias@intellektik.informatik.th-darmstadt.de Wed Nov 10 11:51:33 1993 Return-Path: Received: from rs2.hrz.th-darmstadt.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20362; Wed, 10 Nov 93 11:51:33 +0100 Received: from aida.intellektik.informatik.th-darmstadt.de by rs2.hrz.th-darmstadt.de with SMTP id AA12788 (5.65c/IDA-1.4.4 for ); Wed, 10 Nov 1993 11:45:02 +0100 Received: by aida.intellektik.informatik.th-darmstadt.de (4.1/Server-1.3/HRZ-THD) id AA05066; Wed, 10 Nov 93 11:44:55 +0100 Date: Wed, 10 Nov 93 11:44:55 +0100 From: Matthias Lindner Message-Id: <9311101044.AA05066@aida.intellektik.informatik.th-darmstadt.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Clisp, ilisp and completion --------- Received message begins Here --------- > > Does the command line completion in clisp mess up anyone elses' use of > clisp with ilisp? I'm not using the clisp specific stuff that's in the > archive so that may be the solution. I have some trouble, with ILISP and CLISP (which means that I've to type several 'Ctrl-D's after some ilisp calls e.g. load-file-lisp), but this did *NOT* go away, when I relinked CLISP without the readline lib. BTW what's "the clisp specific stuff that's in the archive"? > > Alternatively, is there any means to disable the readline library without > rebuilding clisp? No - a '--no-readline' command line switch was nice (Hi Bruno :) )but it does not exist. --Matthias ------------------------------------------------------------------------------ Matthias Lindner FG Intellektik, FB Informatik Technische Hochschule Darmstadt Alexanderstr.10, D-64283 Darmstadt TEL: +49 6151 166651 FAX: +49 6151 165326 NET: matthias@intellektik.informatik.th-darmstadt.de ------------------------------------------------------------------------------ From xjam@ginkgo.CS.Berkeley.EDU Wed Nov 10 18:31:17 1993 Return-Path: Received: from ginkgo.CS.Berkeley.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20776; Wed, 10 Nov 93 18:31:17 +0100 Received: from localhost (xjam@localhost) by ginkgo.CS.Berkeley.EDU (8.6.4/8.5) id JAA06894; Wed, 10 Nov 1993 09:25:14 -0800 Message-Id: <199311101725.JAA06894@ginkgo.CS.Berkeley.EDU> To: clisp-list Subject: Re: Clisp, ilisp and completion In-Reply-To: Your message of "Wed, 10 Nov 1993 11:54:46 +0100." <9311101044.AA05066@aida.intellektik.informatik.th-darmstadt.de> Date: Wed, 10 Nov 1993 09:25:13 -0800 From: (The Crossjammer) >>>>> In article <9311101044.AA05066@aida.intellektik.informatik.th-darmstadt.de>, Matthias Lindner writes: Matthias> I have some trouble, with ILISP and CLISP (which means that I've to type Matthias> several 'Ctrl-D's after some ilisp calls e.g. load-file-lisp), but this Matthias> did *NOT* go away, when I relinked CLISP without the readline lib. Matthias> BTW what's "the clisp specific stuff that's in the archive"? I guess I should be more specific. My problem is that if there are tabs in a string that commands like "eval-prev-defun" send to clisp, clisp tries to do expansion and weird things result. I managed to get around it once by building clisp without readline, but this is a pain if you want to keep up with latest versions by just getting the binaries. Meanwhile, in the archives of the clisp list there is some elisp code that you're supposed to be able to plug into ILISP that'll create a clisp dialect. I haven't tried it though. xjam@cork.Berkeley.EDU Fiiiive Thousand Boomin Watts...... -The JBeez They can't come on and play me in prime time, | Free Millipede coin op Cause I know the time, cause I'm gettin' mine. | and House music at my I get on the mix late in the night... -Public Enemy | place! Drop by sometime. From haible Thu Nov 11 22:10:57 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA22824; Thu, 11 Nov 93 22:10:57 +0100 Date: Thu, 11 Nov 93 22:10:57 +0100 From: haible (Bruno Haible) Message-Id: <9311112110.AA22824@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Clisp, ilisp and completion > Alternatively, is there any means to disable the readline library without > rebuilding clisp? On Unix: Send the output to a pipe, for example clisp | tee For DOS, I don't know. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From cole@elvis.concert.net Fri Nov 12 14:39:32 1993 Return-Path: Received: from elvis.concert.net by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23488; Fri, 12 Nov 93 14:39:32 +0100 Received: by elvis.concert.net (5.65/tas-concert/aug93) id AA01422; Fri, 12 Nov 93 08:32:37 -0500 From: Derrick Cole Message-Id: <9311121332.AA01422@elvis.concert.net> Subject: first message to the list, have a newbie question To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Fri, 12 Nov 1993 08:32:37 -0500 (EST) X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2451 Greetings! I've been exposed to c-lisp for a grand total of < 12 hours, so please pardon my newbie-esque approach. I just pulled down the 93-11-08 code and attempted to compile. I'm running SunOS 4.1.3 on a Sparc10 and compiling with gcc-2.5.2. First pass, compiling with -DIMMUTABLE failed, but completed without it. I looked at spvw.d, made a change, and now -DIMMUTABLE (at least) compiles: *** spvw.d.orig Thu Nov 11 22:28:44 1993 --- spvw.d Thu Nov 11 22:28:18 1993 *************** *** 1139,1146 **** }); } #define exitmap() \ ! if (!(bigblock_ptr == &bigblock[0])) ! close_mapid(bigblock[0].mm_mapid); #endif #endif # MULTIMAP_MEMORY_VIA_FILE --- 1139,1147 ---- }); } #define exitmap() \ ! { if (!(bigblock_ptr == &bigblock[0])) \ ! close_mapid(bigblock[0].mm_mapid); \ ! } #endif #endif # MULTIMAP_MEMORY_VIA_FILE "make test" passes, but "make testsuite" fails with the following alltest.erg file: Form: (PRIN1-TO-STRING MOST-POSITIVE-DOUBLE-FLOAT) SOLL: "8.988465674311579d307" CLISP: "1.7976931348623157d308" Form: (PRIN1-TO-STRING LEAST-POSITIVE-DOUBLE-FLOAT) SOLL: "5.562684646268004d-309" CLISP: "2.2250738585072014d-308" Form: (PRIN1-TO-STRING LEAST-NEGATIVE-DOUBLE-FLOAT) SOLL: "-5.562684646268004d-309" CLISP: "-2.2250738585072014d-308" Form: (PRIN1-TO-STRING MOST-NEGATIVE-DOUBLE-FLOAT) SOLL: "-8.988465674311579d307" CLISP: "-1.7976931348623157d308" However, if I invoke lisp.run and execute the forms by hand, I get: > (PRIN1-TO-STRING MOST-POSITIVE-DOUBLE-FLOAT) ".11111111111111111111111111111111111111111111111111111d1024" > (PRIN1-TO-STRING LEAST-POSITIVE-DOUBLE-FLOAT) ".10000000000000000000000000000000000000000000000000000d-1021" > (PRIN1-TO-STRING LEAST-NEGATIVE-DOUBLE-FLOAT) "-.10000000000000000000000000000000000000000000000000000d-1021" > (PRIN1-TO-STRING MOST-NEGATIVE-DOUBLE-FLOAT) "-.11111111111111111111111111111111111111111111111111111d1024" > I'm afraid to susbstitute these values, as I'm unsure as to whether or not I've got a compilation problem. Any and all help would certainly be much appreciated! Thanks, Derrick -- | / _____/ | / _____/ Derrick Cole, KC4WEJ / __/ / / / | / / MCNC Information Technologies / / / / | / / Research Triangle Park, NC __/ __/ _______/ __/ __/ _______/ cole@concert.net From haible Fri Nov 12 15:26:32 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23606; Fri, 12 Nov 93 15:26:32 +0100 Date: Fri, 12 Nov 93 15:26:32 +0100 From: haible (Bruno Haible) Message-Id: <9311121426.AA23606@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: first message to the list, have a newbie question Derrick Cole was "exposed to c-lisp for a grand total of < 12 hours" and already found and fixed a bug. Congratulations! > "make test" passes, but "make testsuite" fails with the following alltest.erg > file: > > Form: (PRIN1-TO-STRING MOST-POSITIVE-DOUBLE-FLOAT) > SOLL: "8.988465674311579d307" > CLISP: "1.7976931348623157d308" This is due to the fact that on Sparc, CLISP uses the floating format prescribed by the hardware for DOUBLE-FLOATs. It differs from CLISP's soft floats. Consider the testsuite to have succeeded. > However, if I invoke lisp.run and execute the forms by hand, I get: > > > (PRIN1-TO-STRING MOST-POSITIVE-DOUBLE-FLOAT) > ".11111111111111111111111111111111111111111111111111111d1024" Don't invoke lisp.run without arguments (except on Atari) - it contains only one half of clisp. Not even DEFUN or SETF. And floating point numbers are printed in binary because floatpri.lsp has not been loaded... Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From dasch@darkwing.uoregon.edu Sat Nov 13 00:12:29 1993 Return-Path: Received: from darkwing.uoregon.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA24313; Sat, 13 Nov 93 00:12:29 +0100 Received: from localhost (dasch@localhost) by darkwing.uoregon.edu (8.6.4/8.6) id PAA24332; Fri, 12 Nov 1993 15:05:28 -0800 Date: Fri, 12 Nov 1993 15:05:28 -0800 From: John Daschbach Message-Id: <199311122305.PAA24332@darkwing.uoregon.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: maxima,clisp,os2, load fails I have been trying to run maxima with clisp under os/2. I have the maxima package from the clisp tree at cmu, so I think this is the correct version to try. Maxima works with the default files, but I have been unable to 'load()' any files. I always get a EONENT error, "file not found" for any load command. The varible file_search contains the correct pathnames. I have tried the trace facility in maxima, but it only traces up to the point where it runs FILE_SEARCH and then returns the same error. Calling FILE_SEARCH yields the same error for any file. Clisp loads files ok, but the maxima code does not. It looks like there is no debugger present in maxima as far as I can tell. Has anybody been able to load files from maxima running in clisp under os2? Any ideas on how to proceed to debug this? -John From jbishop@babel.ling.nwu.edu Sun Nov 14 22:47:36 1993 Received: from relay.acns.nwu.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA00558; Sun, 14 Nov 93 22:47:36 +0100 Received: from babel.ling.nwu.edu by relay.acns.nwu.edu (4.1/SMI-ACNS-relay-1.7) id AA23245; Sun, 14 Nov 93 15:40:14 CST Return-Path: Received: by babel.ling.nwu.edu (4.1/SMI-4.0) id AA27497; Sun, 14 Nov 93 15:40:10 CST Date: Sun, 14 Nov 1993 15:39:53 -0600 (CST) From: Jeff Bishop Sender: Jeff Bishop Reply-To: Jeff Bishop Subject: Loop macro To: Clisp List Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII I am having some trouble with the loop macro on CLISP. When I try the loop-for-var ... constructions, it complains that var is undefined rather than defining it as a local variable. Any suggestions for making LOOP work according to Common Lisp conventions? __________________________________________________________________________ | | | | Jeff Bishop | | | | "Real programmers don't document. If it | | Department of Linguistics | was hard to write, it should be hard to | | Northwestern University | understand." | |______________________________|__________________________________________| From haible Mon Nov 15 18:51:53 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01658; Mon, 15 Nov 93 18:51:53 +0100 Date: Mon, 15 Nov 93 18:51:53 +0100 From: haible (Bruno Haible) Message-Id: <9311151751.AA01658@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Loop macro Jeff Bishop writes: > I am having some trouble with the loop macro on CLISP. When I try the > loop-for-var ... constructions, it complains that var is undefined rather > than defining it as a local variable. Any suggestions for making LOOP > work according to Common Lisp conventions? CLISP's LOOP macro is a CLtL1 loop, a simple loop. The (LOOP FOR var ...) constructs use the CLtL2 loop macro. I have an implementation of it, but I don't distribute it with clisp because I discourage the use of LOOP as long as the ANSI CL draft has inconsistencies in its specification. It is on ma2s2.mathematik.uni-karlsruhe.de: /pub/lisp/clisp/packages/UNSUPPORTED/loop.lsp.z. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From philp@universe.digex.net Tue Nov 16 16:30:50 1993 Return-Path: Received: from universe.digex.net by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03153; Tue, 16 Nov 93 16:30:50 +0100 Received: by universe.digex.net id AA02897 (5.67a8/IDA-1.5 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Tue, 16 Nov 1993 10:23:11 -0500 Date: Tue, 16 Nov 1993 10:17:28 -0500 (EST) From: Phil Perucci Subject: Compile 11/08 on Linux? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Anyone out there working on porting the latest clisp (11/08) to the current Linux (Slackware 1.1.0)? Using the previous Slackware 1.0.5 and previous clisp source, compile was fine. But both the 11/08 and previous clisp source version fail under Slackware 1.1.0. I could post the output of make, if someone could give me the syntax for saving EVERYTHING to a file, for example "make 2>&1 myfile". Can't remember the syntax... ============================================================================== Phil Perucci | "All postings are my own opinion - all comments Systems Programmer | are intended for research/educational purposes" ============================================================================== From MORRIS1@DUCVAX.AUBURN.EDU Thu Nov 18 06:27:39 1993 Return-Path: Received: from DUCVAX.AUBURN.EDU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04808; Thu, 18 Nov 93 06:27:39 +0100 Received: from DUCVAX.AUBURN.EDU by DUCVAX.AUBURN.EDU (PMDF #12431) id <01H5FU61E6CG0005CA@DUCVAX.AUBURN.EDU>; Wed, 17 Nov 1993 23:19 CST Date: Wed, 17 Nov 1993 23:19 CST From: MORRIS1@DUCVAX.AUBURN.EDU Subject: CLISP Compiler To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H5FU61E6CG0005CA@DUCVAX.AUBURN.EDU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" This may be a really dumb question, but I'm very new to CLISP and LISP in general. I'm using CLISP on an IBM-compatible 386 computer using MSDOS. Is there a way to compile my programs into an executable format (.exe or .com)? Thanks for the help! Steve From philp@universe.digex.net Thu Nov 18 13:40:13 1993 Return-Path: Received: from universe.digex.net by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05290; Thu, 18 Nov 93 13:40:13 +0100 Received: by universe.digex.net id AA23420 (5.67a8/IDA-1.5 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Thu, 18 Nov 1993 07:32:13 -0500 Date: Thu, 18 Nov 1993 07:25:58 -0500 (EST) From: Phil Perucci Subject: edit/test cycle? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Simple question... With CLISP, what is the best edit/test/code/edit, etc cycle (under X)? One xterm for editing, another for "clisp -c"? About all I have to go on is the CLISP man page and "LISP 3rd Edition" by Winston/Horn. BTW- the environment is Linux (Slackware 1.1.0), XFree86 2.0. ============================================================================== Phil Perucci | "All postings are my own opinion - all comments Systems Programmer | are intended for research/educational purposes" ============================================================================== From RLS@waikato.ac.nz Thu Nov 18 23:38:48 1993 Return-Path: Received: from truth.waikato.ac.nz by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA05989; Thu, 18 Nov 93 23:38:48 +0100 Message-Id: <9311182238.AA05989@ma2s2.mathematik.uni-karlsruhe.de> Date: Fri, 19 Nov 93 11:30 +1300 From: RLS@waikato.ac.nz Subject: RE: edit/test cycle? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" I've just started using Linux and openwin, this makes the environment almost identical to the familiar sun/sparc environment. I find a nice way to use CLisp is to have one xterm running CLisp and a textedit window with my source in it. One can then just copy and paste between them, especially if you've got the 3 button mouse going with the middle button pasting into the xterm. Another xterm is handy to have around from which to start other textedit's as needed. Hope this helps, Richard Shepherd. From gt8134b@prism.gatech.edu Fri Nov 19 03:42:01 1993 Return-Path: Received: from hydra.gatech.edu (hydra-rich.gatech.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06167; Fri, 19 Nov 93 03:42:01 +0100 Received: by hydra.gatech.edu (5.67a/3.1) id AA27619; Thu, 18 Nov 1993 21:33:55 -0500 From: gt8134b@prism.gatech.edu (Robert Sanders) Message-Id: <199311190233.AA27619@prism.gatech.edu> Subject: Version 1993-11-08 under Linux To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de (CLISP mailing list) Date: Thu, 18 Nov 1993 21:33:55 -0500 (EST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 779 I've compiled version 1993-11-08 for Linux and found a minor bug: in unixconf.h, SHMAT_CONST should be defined to nothing. Also, with this configuration: Linux 0.99.13 ALPHAr libc 4.4.4 gcc 2.4.3 I got a SIGSEGV after the "Bye" message when it quit. I traced this to somewhere at the end of main(). Not having the time to recompile everything -g, I just recompiled spvw.c with -g. (I had been compiling with -O2 -m486). This version of the lisp.run did not cause the segfault. I'm assuming it's an optimization bug in gcc 2.4.3; spvw.c is the only file I had to compile with optimization off. I will investigate further. I'm installing gcc 2.5.4 tonight. I just got my copy of CLtL2 and having CLISP to use it with is heavenly. Kudos to Bruno and Michael! From gt8134b@prism.gatech.edu Fri Nov 19 03:50:22 1993 Return-Path: Received: from hydra.gatech.edu (hydra-rich.gatech.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06255; Fri, 19 Nov 93 03:50:22 +0100 Received: by hydra.gatech.edu (5.67a/3.1) id AA29466; Thu, 18 Nov 1993 21:42:16 -0500 From: gt8134b@prism.gatech.edu (Robert Sanders) Message-Id: <199311190242.AA29466@prism.gatech.edu> Subject: Compiling 1993-11-08 for HP-UX 8.07 To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de (CLISP mailing list) Date: Thu, 18 Nov 1993 21:42:15 -0500 (EST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 576 I tried to compile CLISP for the HP9000s700 at work today. Most things went fine except the arihppa file; it seems to be in HP assembler syntax, and I'm using GAS with GCC. So I changed the makefile line from ... "|| as" to "|| /bin/as". The CLISP that resulted from this worked well except for one thing: if I pressed CTRL-C it died from a SIGALRM. I briefly looked at unix.c, unixaux.c, and spvw.c and the alarm_handler() and interrupt_handler() is somewhat complex so I didn't have time to explore more. I'll investigate further tomorrow but any hints are welcome. From haible Sun Nov 21 22:17:54 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01786; Sun, 21 Nov 93 22:17:54 +0100 Date: Sun, 21 Nov 93 22:17:54 +0100 From: haible (Bruno Haible) Message-Id: <9311212117.AA01786@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Compile 11/08 on Linux? Phil Perucci was wondering why CLISP 1993-11-08 didn't compile on Linux Slackware 1.1.0. The reason is that this Slackware release uses gcc-2.5 in combination with a set of system's include files (inc-4.4.4) which had not been updated for gcc-2.5. The CLISP source has been updated for gcc-2.5. > previous clisp source version fail under Slackware 1.1.0. Don't use previous CLISP sources with gcc-2.5. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Mon Nov 22 00:19:53 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02064; Mon, 22 Nov 93 00:19:53 +0100 Date: Mon, 22 Nov 93 00:19:53 +0100 From: haible (Bruno Haible) Message-Id: <9311212319.AA02064@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CLISP Compiler Steven A. Morris asks: > I'm using CLISP on an IBM-compatible 386 computer using MSDOS. > Is there a way to compile my programs into an executable format (.exe or > .com)? No. The CLISP compiler doesn't produce machine code. You can use the (undocumented) SAVEMEM function to dump memory images that can be given as argument to the -M option of clisp. For example: clisp > (load "rubik.fas") > (savemem "rubik.mem") > (exit) Then call clisp -M rubik.mem Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Mon Nov 22 01:02:06 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02240; Mon, 22 Nov 93 01:02:06 +0100 Date: Mon, 22 Nov 93 01:02:06 +0100 From: haible (Bruno Haible) Message-Id: <9311220002.AA02240@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: edit/test cycle? Phil Perucci has an interesting question: > With CLISP, what is the best edit/test/code/edit, etc cycle (under X)? > One xterm for editing, another for "clisp -c"? The way I use to work on Linux with X11 is this: - editor: aXe-5.1 (has multiple windows, pull-down menus, parentheses matching) - clisp running in one xterm, sometimes with Capture/Logging enabled. - developing cycle: edit & save, switch to clisp xterm, (load "..."), test several functions individually, maybe using the stepper, then switch back to the editor window. - when I think everything works, use clisp -c and eliminate the errors and warnings. Bruno Haible From haible Mon Nov 22 01:21:23 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02371; Mon, 22 Nov 93 01:21:23 +0100 Date: Mon, 22 Nov 93 01:21:23 +0100 From: haible (Bruno Haible) Message-Id: <9311220021.AA02371@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Version 1993-11-08 under Linux Robert Sanders found a problem when compiling and running clisp on Linux: > libc 4.4.4 gcc 2.4.3 > I got a SIGSEGV after the "Bye" message when it quit. This is a problem in the Linux header file /usr/include/setjmp.h that goes away if you upgrade to gcc-2.5.3 and libc-4.4.4b. We have to thank Richard Stallman: he modified gcc to fix this problem. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From gt8134b@prism.gatech.edu Mon Nov 22 01:57:15 1993 Return-Path: Received: from hydra.gatech.edu (hydra-rich.gatech.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02460; Mon, 22 Nov 93 01:57:15 +0100 Received: from prism (acme.gatech.edu) by hydra.gatech.edu (5.67a/3.1) id AA27877; Sun, 21 Nov 1993 19:48:37 -0500 Received: by prism.gatech.edu (5.67a/1.0) id AA04811; Sun, 21 Nov 1993 19:48:33 -0500 From: gt8134b@prism.gatech.edu Message-Id: <199311220048.AA04811@prism> Subject: Re: Compiling 1993-11-08 for HP-UX 8.07 To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de (CLISP mailing list) Date: Sun, 21 Nov 1993 19:48:33 -0500 (EST) X-Mailer: ELM [version 2.4 PL22] Content-Type: text Content-Length: 698 [Bruno, I hope you don't mine me forwarding this to the CLISP list.] >From haible@ma2s2.mathematik.uni-karlsruhe.de Sun Nov 21 19:40:59 1993 > > The CLISP that resulted from this worked well except for one thing: > > if I pressed CTRL-C it died from a SIGALRM. > > the alarm_handler() > > and interrupt_handler() is somewhat complex so I didn't have time > > to explore more. > > The problem isn't there: If I try to interrupt running programs using Well, I found a workaround that seems to work, although it results in less responsive behavior due to alarm()'s resolution: I undefined HAVE_ITIMER in unixconf.h. Until the problem in readline is found, I suggest that this be the default. From gt8134b@prism.gatech.edu Mon Nov 22 02:06:08 1993 Return-Path: Received: from hydra.gatech.edu (hydra-rich.gatech.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02530; Mon, 22 Nov 93 02:06:08 +0100 Received: from prism (acme.gatech.edu) by hydra.gatech.edu (5.67a/3.1) id AA00469; Sun, 21 Nov 1993 19:57:31 -0500 Received: by prism.gatech.edu (5.67a/1.0) id AA05130; Sun, 21 Nov 1993 19:57:27 -0500 From: gt8134b@prism.gatech.edu Message-Id: <199311220057.AA05130@prism> Subject: Re: Version 1993-11-08 under Linux To: haible@ma2s2.mathematik.uni-karlsruhe.de (Bruno Haible) Date: Sun, 21 Nov 1993 19:57:27 -0500 (EST) Cc: clisp-list@ma2s2.mathematik.uni-karlsruhe.de (CLISP mailing list) In-Reply-To: <9311220031.AA02429@ma2s2.mathematik.uni-karlsruhe.de> from "Bruno Haible" at Nov 22, 93 01:31:15 am X-Mailer: ELM [version 2.4 PL22] Content-Type: text Content-Length: 2049 > > in unixconf.h, SHMAT_CONST should be defined to nothing. > Oops. I should have noticed this. What does the declaration of shmat() > look like in your /usr/include/sys/shm.h ? extern char *shmat __P((int __shmid, char *__shmaddr, int __shmflg)); > I had the same problem some weeks ago, and I thought it was a problem in gcc. > I sent a bug report to the gcc people, and Stallman patched gcc. The problem > was that gcc must not treat "setjmp" like any other C function, but on Linux > with a new libc setjmp is really called __setjmp - and gcc didn't know about > it. Hmm. Is it that setjmp() should be declared __attribute__ ((noreturn)), or something other that gcc needs to know? > > I will investigate further. I'm installing gcc 2.5.4 tonight. > Recall that gcc-2.5 requires inc-4.4.4b or newer. Hmm. I compiled gcc 2.5.4 myself with library 4.4.4 (non-beta). Maybe I should upgrade my libc and include files. I found a couple of new problems; the first is that kernel 0.99.13 ALPHAr (and a few earlier releases, too) have changed kmalloc() so that it can allocate a maximum of 4072 bytes or so. THe upshot of that is that SHMMAX needs to be reduced because fewer page pointers can fit on one kmalloc()ed page. I think I used 0x3fa000. Another problem was that I couldn't compile maxima. It compiled all the files fine, then started loading, got to mopers.fas, and sat there spinning. I left it for a couple of hours with no results. Then I did an strace on it, and it was continually readdir()ing the root directory and /bin! Do you have any idea what this is caused by? And starting clisp with 4 megs of memory caused it to eventually fail with "No room for Lisp Objects" when attempting to compile maxima. This could just be a side-effect of the above problem. If I send sigquit to clisp, it doesn't get to remove the main shared memory segment(s). Though they are attached 42 times, they aren't destroyed. Could they be deleted so that when clisp terminates they are automaticaly disposed of? thanks, From gt8134b@prism.gatech.edu Mon Nov 22 02:16:33 1993 Return-Path: Received: from hydra.gatech.edu (hydra-rich.gatech.edu) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA02598; Mon, 22 Nov 93 02:16:33 +0100 Received: from prism (acme.gatech.edu) by hydra.gatech.edu (5.67a/3.1) id AA03518; Sun, 21 Nov 1993 20:07:53 -0500 Received: by prism.gatech.edu (5.67a/1.0) id AA05764; Sun, 21 Nov 1993 20:07:44 -0500 From: gt8134b@prism.gatech.edu Message-Id: <199311220107.AA05764@prism> Subject: Re: edit/test cycle? To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Sun, 21 Nov 1993 20:07:43 -0500 (EST) In-Reply-To: <9311220002.AA02240@ma2s2.mathematik.uni-karlsruhe.de> from "Bruno Haible" at Nov 22, 93 01:05:48 am X-Mailer: ELM [version 2.4 PL22] Content-Type: text Content-Length: 3505 > Phil Perucci has an interesting question: > > With CLISP, what is the best edit/test/code/edit, etc cycle (under X)? > - editor: aXe-5.1 (has multiple windows, pull-down menus, parentheses matching) Emacs has all those, too. I don't use pull-down menus much. If I can dgo a day without touching my mouse I'm happy :-) Emacs has exceptional Lisp editing facilities (of course). It does paren matching, pretty indenting via TAB, and the inferior-lisp functionality that saves me loads of time. Basically, you start CLISP in an Emacs buffer and communicate via that buffer. However, you can also use Emacs commands to send code to the inferior (no offense, Bruno :-) Lisp process without having to go through all the cut-and-paste or save-and-load drudgery. It won't work if you're doing screen-oriented work, but otherwise it's heaven. Emacs 19 under X has things like syntactical highlighting/colorizing of code and multiple screens (X windows). I heartily recommend Emacs for anyone, but especially for Lisp users. I recommend 8 MB or more for Lisp development under X; I have 8 and it works pretty well. A (huge) side bonus is that Emacs can be customized by Lisp code. I couldn't live without it. Here are some of the Emacs commands for dealing with other Lisp processes. You would, of course, bind them to handy keys. lisp-arglist-command Variable: Command to query inferior Lisp for a function's arglist. lisp-compile-defun Function: Compile the current defun in the inferior Lisp process. lisp-compile-defun-and-go Function: Compile the current defun in the inferior Lisp, and switch to its buffer. lisp-compile-file Function: Compile a Lisp file in the inferior Lisp process. lisp-compile-region Function: Compile the current region in the inferior Lisp process. lisp-compile-region-and-go Function: Compile the current region in the inferior Lisp, and switch to its buffer. lisp-complete-symbol Function: Perform completion on Lisp symbol preceding point. That symbol is lisp-describe-sym Function: Send a command to the inferior Lisp to describe symbol SYM. lisp-describe-sym-command Variable: Command to query inferior Lisp for a variable's documentation. lisp-eval-defun Function: Send the current defun to the inferior Lisp process. lisp-eval-defun-and-go Function: Send the current defun to the inferior Lisp, and switch to its buffer. lisp-eval-last-sexp Function: Send the previous sexp to the inferior Lisp process. lisp-eval-region Function: Send the current region to the inferior Lisp process. lisp-eval-region-and-go Function: Send the current region to the inferior Lisp, and switch to its buffer. lisp-function-doc-command Variable: Command to query inferior Lisp for a function's documentation. lisp-load-file Function: Load a Lisp file into the inferior Lisp process. lisp-send-defun Function: Send the current defun to the Lisp process made by M-x run-lisp. lisp-show-arglist Function: Send a query to the inferior Lisp for the arglist for function FN. lisp-show-function-documentation Function: Send a command to the inferior Lisp to give documentation for function FN. lisp-show-variable-documentation Function: Send a command to the inferior Lisp to give documentation for function FN. lisp-var-doc-command Variable: Command to query inferior Lisp for a variable's documentation. switch-to-lisp Function: Switch to the inferior Lisp process buffer. From P2043742@csdvax.csd.unsw.EDU.AU Tue Nov 23 06:55:55 1993 Return-Path: Received: from csdvax.csd.unsw.EDU.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04360; Tue, 23 Nov 93 06:55:55 +0100 Received: from csdvax.csd.unsw.EDU.AU by csdvax.csd.unsw.EDU.AU (PMDF V4.2-13 #4042) id <01H5NUAQDDI88X6Z9Y@csdvax.csd.unsw.EDU.AU>; Tue, 23 Nov 1993 16:50:19 +1000 Date: Tue, 23 Nov 1993 16:50:19 +1000 From: P2043742@csdvax.csd.unsw.EDU.AU Subject: Problem with load gadget in Garnet2.2. To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Message-Id: <01H5NUAQDN5E8X6Z9Y@csdvax.csd.unsw.EDU.AU> X-Vms-To: IN%"clisp-list@ma2s2.mathematik.uni-karlsruhe.de" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Garnet and Clisp users. I have some problems with the load gadget of Garnet2.2. My problems are: 1. When the window pops up the files that are in the directory I specified in the :initial-directory slot of the gadget don't get listed in the Load Window. 2. When I type the name of the file in the load window that are in the directory I specified Garnet does not load it instead it gives a beep. The value that is returned by the gadget is not the same I typed in. My load gadget was: (create-instance 'Load-Gadget gg:motif-load-gadget (:window-top 200) (:window-left 400) (:initial-directory "/tmp_mnt/net/pg1/p2043742/clisp/garnet2.2/turnint/") (:foreground-color opal:motif-green) (:selection-function #'turn::Load-Function)) Originaly the file I wanted to read was in that directory. When I typed in the filename the system beeped as if the file was not in the directory or didn't exist at all. Then I changed the :initial-directory to "/tmp_mnt/net/pg1/p2043742/clisp/garnet2.2/" and now if I type ./turnint/filename the value that is returned by the gadget is #"../filename". I copyed the file into directory /tmp_mnt/net/pg1/p2043742/clisp/garnet2.2/ and now the system loads the file. Could someone explain what is happening hire. I am using CLISP on SGI Indigo. Yours, Bela Pecsek. From ckclark@MIT.EDU Sun Nov 28 07:09:23 1993 Return-Path: Received: from MIT.EDU (ATHENA-AS-WELL.MIT.EDU) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA10521; Sun, 28 Nov 93 07:09:23 +0100 Received: from HODGE.MIT.EDU by MIT.EDU with SMTP id AA26538; Sun, 28 Nov 93 01:08:49 EST Received: by hodge.MIT.EDU (5.57/4.7) id AA21043; Sun, 28 Nov 93 01:08:34 -0500 Date: Sun, 28 Nov 93 01:08:34 -0500 Message-Id: <9311280608.AA21043@hodge.MIT.EDU> From: Calvin Clark To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: *error-handler* The variable *error-handler* is not documented in impnotes.txt. Is it likely to be preserved in future versions of CLISP? I would like to be able to use it in macros like this: (defmacro with-handled-error (&body body) (let ((blockvar (gensym))) `(block ,blockvar (let ((*error-handler* #'(lambda (arg1 &rest args) (apply 'format (cons *error-output* args)) (return-from ,blockvar nil)))) ,@body)))) The macro works like this: > (with-handled-error (pathname "~nouser")) UNIX error 2 (ENOENT): No such file or directory NIL > (with-handled-error (open "/dev/rrz0a" :direction :input)) UNIX error 13 (EACCES): Permission denied NIL > What are the potential caveats in using *error-handler* this way? Are there better ways in CLISP to deal with system errors in code? -Calvin From hjstein@MATH.HUJI.AC.IL Sun Nov 28 12:41:38 1993 Return-Path: Received: from sunset.ma.huji.ac.il by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA10661; Sun, 28 Nov 93 12:41:38 +0100 Received: by sunset.ma.huji.ac.il id AA13281 (5.65c/HUJI 4.152 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Sun, 28 Nov 1993 13:43:06 +0200 Date: Sun, 28 Nov 1993 13:43:06 +0200 From: "Harvey J. Stein" Message-Id: <199311281143.AA13281@sunset.ma.huji.ac.il> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Maxima problem Cc: hjstein@MATH.HUJI.AC.IL I'm sorry to bother From hjstein@MATH.HUJI.AC.IL Sun Nov 28 12:59:48 1993 Return-Path: Received: from sunset.ma.huji.ac.il by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA10758; Sun, 28 Nov 93 12:59:48 +0100 Received: by sunset.ma.huji.ac.il id AA13324 (5.65c/HUJI 4.152 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Sun, 28 Nov 1993 14:01:16 +0200 Date: Sun, 28 Nov 1993 14:01:16 +0200 From: "Harvey J. Stein" Message-Id: <199311281201.AA13324@sunset.ma.huji.ac.il> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Maxima problem Cc: hjstein@MATH.HUJI.AC.IL (By the way, please excuse the previous message. I meant to cancel the message but accidentally sent it instead.) I just tried to compile & run maxima under clisp. I got: *** - EVAL: undefined function SYSTEM::MAKE-WINDOW Maxima 4.153 Sun Aug 19 12:02:38 CDT 1990 (with enhancements by W. Schelter). when I tried to run maxima. It started up, but when trying to do anything, I get the message: *** - symbol *WINDOW* has no value I remember something on this list about this problem, and I recall that a newer version of CLISP fixes it (I'm running version "September 1993" as reported by (lisp-implementation-version)). So, I connected to ma2s2.mathematik.uni-karlsruhe.de to down-load the latest version of CLISP, and to try again. However, when I checked the README file, it said that the newest version of CLISP requires a Linux kernel of at least .99pl11. I'm running .99pl10. Do I really need to upgrade the kernel before I can run this binary? Can I compile a working version of CLISP under .99pl10? I figured I'd ask before doing the download and enduring the pain of compilation. Thanks, Harvey Stein Department of Mathematics Hebrew University hjstein@math.huji.ac.il From haible Mon Nov 29 21:17:23 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12416; Mon, 29 Nov 93 21:17:23 +0100 Date: Mon, 29 Nov 93 21:17:23 +0100 From: haible (Bruno Haible) Message-Id: <9311292017.AA12416@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Maxima problem Harvey Stein too has been bitten by the following problem: > ... run maxima under clisp. I got: > *** - EVAL: undefined function SYSTEM::MAKE-WINDOW > Maxima 4.153 Sun Aug 19 12:02:38 CDT 1990 (with enhancements by W. Schelter). > *** - symbol *WINDOW* has no value Please get and install /pub/lisp/clisp/packages/maxima.clisp.patch1 from ma2s2.mathematik.uni-karlsruhe.de. The function MAKE-WINDOW sits in # since the CLISP version of May 1993. He asks whether CLISP requires a certain Linux kernel version: > However, when I checked the > README file, it said that the newest version of CLISP requires a Linux > kernel of at least .99pl11. I'm running .99pl10. Do I really need to > upgrade the kernel before I can run this binary? Can I compile a > working version of CLISP under .99pl10? CLISP can be compiled on any Linux version newer than 0.95. The binary distribution for Linux is linked against libc-4.4.1 which requires Linux kernel 0.99.11 or newer. If you don't want to upgrade your Linux, you will have to recompile CLISP from source. Just start it, then sit and wait. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Mon Nov 29 21:28:55 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12490; Mon, 29 Nov 93 21:28:55 +0100 Date: Mon, 29 Nov 93 21:28:55 +0100 From: haible (Bruno Haible) Message-Id: <9311292028.AA12490@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: *error-handler* Calvin Clark defines WITH-HANDLED-ERROR and asks: > What are the potential caveats in using *error-handler* this way? None. Your macro is perfectly fine, except that a call (fresh-line *error-output*) before (apply #'format *error-output* args) wouldn't hurt. > Are there better ways in CLISP to deal with system errors in code? Not yet. > The variable *error-handler* is not documented in impnotes.txt. Is it > likely to be preserved in future versions of CLISP? Yes. So many ports of packages depend on it. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From hjstein@MATH.HUJI.AC.IL Tue Nov 30 06:55:12 1993 Return-Path: Received: from sunset.ma.huji.ac.il by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13030; Tue, 30 Nov 93 06:55:12 +0100 Received: by sunset.ma.huji.ac.il id AA23680 (5.65c/HUJI 4.152 for clisp-list ); Tue, 30 Nov 1993 07:56:21 +0200 Date: Tue, 30 Nov 1993 07:56:21 +0200 From: "Harvey J. Stein" Message-Id: <199311300556.AA23680@sunset.ma.huji.ac.il> To: clisp-list In-Reply-To: <9311292017.AA12416@ma2s2.mathematik.uni-karlsruhe.de> Subject: Re: Maxima problem Bruno Haible writes: > CLISP can be compiled on any Linux version newer than 0.95. The > binary distribution for Linux is linked against libc-4.4.1 which > requires Linux kernel 0.99.11 or newer. Are you sure about this? I've been running libc-4.4.1 under kernel 0.99.10 for about two months, and I installed libc-4.4.4 about a week ago. I haven't noticed any problems... Harvey Stein From d93perwi@und.ida.liu.se Wed Dec 1 20:57:30 1993 Return-Path: Received: from ida.liu.se (curofix.ida.liu.se) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA19569; Wed, 1 Dec 93 20:57:30 +0100 Received: from und.ida.liu.se (astmatix.ida.liu.se) by ida.liu.se (5.65b/ida.minimaster-V1.0b6d5) id AA17364; Wed, 1 Dec 93 20:57:01 +0100 Received: from odalix by und.ida.liu.se (5.65b/ida.minimaster-V1.0b6d2) id AA10283; Wed, 1 Dec 93 20:56:58 +0100 From: Per Widerlund Received: from odal5 by odalix (5.65b/ida.slave-V1.0b3) id AA27572; Wed, 1 Dec 93 20:56:56 +0100 Received: by odal5 (5.65b/ida.slave-V1.0b3) id AA20958; Wed, 1 Dec 93 20:56:54 +0100 Date: Wed, 1 Dec 93 20:56:54 +0100 Message-Id: <9312011956.AA20958@odal5> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Questions... Ok, I have some general and some Amiga-specific questions... Can I get the - function to accept LONG-FLOAT in any way? I get 0.0 from (- 3.00000000001 3.0) instead of 0.00000000001. Where can I find an editor for the Amiga who can handle all the parentheses? (Something like Emacs would be great but GNU-Emacs is to LARGE for my system.) Where can I find Amiga-specific CLisp source? (Graphic packages and such...) /Per Widerlund From hoehle@post.inf-wiss.ivp.uni-konstanz.de Thu Dec 2 14:32:00 1993 Return-Path: Received: from inf-wiss.ivp.uni-konstanz.de (post.inf-wiss.ivp.uni-konstanz.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20340; Thu, 2 Dec 93 14:32:00 +0100 Received: from stetten.inf-wiss.ivp.uni-konstanz.de by inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA11079; Thu, 2 Dec 93 14:31:23 +0100 Date: Thu, 2 Dec 93 14:31:23 +0100 Message-Id: <9312021331.AA11079@inf-wiss.ivp.uni-konstanz.de> From: hoehle@post.inf-wiss.ivp.uni-konstanz.de (Joerg-Cyril Hoehle) To: clisp-list@[129.13.115.2] Cc: Per Widerlund Subject: Questions... In-Reply-To: <9312011956.AA20958@odal5> References: <9312011956.AA20958@odal5> Per Widerlund writes: > Can I get the - function to accept LONG-FLOAT in any way? > I get 0.0 from (- 3.00000000001 3.0) instead of 0.00000000001. Have a look at *DEFAULT-FLOAT-FORMAT* and *READ-DEFAULT-FLOAT-FORMAT* or use the 3.0001l0 (see CLtLII p.18ff) syntax. If you need more precision than the default for long-floats, have a look at (LONG-FLOAT-DIGITS) and (SETF (LONG-FLOAT-DIGITS) nn). > Where can I find an editor for the Amiga who can handle all the parentheses? > (Something like Emacs would be great but GNU-Emacs is to LARGE for my system.) Before I got enough memory to run GNU-Emacs, I used MG (aka MicroGNUEmacs) which comes pretty close. To be found on Fish 352. You can get a modified version that displays ISO chars from me. > Where can I find Amiga-specific CLisp source? > (Graphic packages and such...) The only Amiga-specific source so far is rexx.lsp. Plans are to port stdwin to the Amiga, but that's more of C than LISP hacking. Other plans are the creation of a good (portable, not Amiga-only) alien package in order to be able to call all .library functions, but no work has been done on that for some time now. Other plans are to improve the ARexx interface a little bit. So far the way to do graphics is to use some ARexx support package that allows access to any library functions. The next best thing in the future would be the use of the portable STDWIN library. Joerg Hoehle. hoehle@inf-wiss.uni-konstanz.de From marcus@ee.pdx.edu Thu Dec 2 18:38:42 1993 Return-Path: Received: from ursula.ee.pdx.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA20937; Thu, 2 Dec 93 18:38:42 +0100 Received: by ursula.ee.pdx.edu (4.1/SMI-4.1) id AA23597; Thu, 2 Dec 93 09:39:20 PST Date: Thu, 2 Dec 93 09:39:20 PST From: marcus@ee.pdx.edu (Marcus Daniels) Message-Id: <9312021739.AA23597@ursula.ee.pdx.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: trouble porting clisp I'm trying to bring up clisp-1993-11-08 on m68k NEXSTEP 3.1. I've configured it so it compiles with few complaints. I can compile clisp with the `strict bsd' and posix configurations. To compile several files, it was necessary to use SAFETY=3 (with -g, the compiler (gcc 2.2.2) generated local symbols that weren't defined, which caused the assembler to fail). Also, for some reason, NeXT's gcc wouldn't grok the register ... __asm__("register_name") directives in eval.d... I've tried the various memory-management options, but none work `out of the box'. NeXT doesn't document the mmap functions, and several of the relevant constants aren't defined. So the configuration that I've started working with is the SPVW_MIXED_BLOCKS -- neither the HAVE_MMAP_ANON or HAVE_MMAP_DEVZERO are defined. Here is what happens when lisp.run is invoked initially: *** - The value of *TERMINAL-IO* was not a stream: #. It has been changed to #. *** - The value of *DEBUG-IO* was not a stream: #. It has been changed to #. *** - No more room for LISP objects *** - No more room for LISP objects *** - No more room for LISP objects *** - No more room for LISP objects *** - No more room for LISP objects *** - No more room for LISP objects *** - No more room for LISP objects *** - No more room for LISP objects *** - No more room for LISP objects: RESET *** - The value of *STANDARD-OUTPUT* was not a stream: #. It has been changed to #. *** - No more room for LISP objects: RESET *** - The value of *STANDARD-INPUT* was not a stream: #. It has been changed to #. *** - No more room for LISP objects: RESET The garbage collection that triggers the `no room' messages is triggered by an attempt to display the `breakprompt_string', which apparently isn't initialized yet (the length field in the sstring structure is 537564620, and the characters arent `Break>'. ). I'm guessing there is a obvious oversight I'm making that explains this behavior. (BTW, the NeXT installation note isn't valid for 3.0). Any suggestions? From marcus@ee.pdx.edu Fri Dec 3 14:35:54 1993 Return-Path: Received: from ursula.ee.pdx.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21859; Fri, 3 Dec 93 14:35:54 +0100 Received: by ursula.ee.pdx.edu (4.1/SMI-4.1) id AA17035; Fri, 3 Dec 93 05:36:23 PST Date: Fri, 3 Dec 93 05:36:23 PST From: marcus@ee.pdx.edu (Marcus Daniels) Message-Id: <9312031336.AA17035@ursula.ee.pdx.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: CLCS for CLISP Has anyone ported CLCS to CLISP? If so, does it work with CLISP's CLOS, or do you need to use PCL? From hoehle@post.inf-wiss.ivp.uni-konstanz.de Fri Dec 3 14:48:40 1993 Return-Path: Received: from inf-wiss.ivp.uni-konstanz.de (post.inf-wiss.ivp.uni-konstanz.de) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA21935; Fri, 3 Dec 93 14:48:40 +0100 Received: from stetten.inf-wiss.ivp.uni-konstanz.de by inf-wiss.ivp.uni-konstanz.de (4.1/SMI-4.1) id AA14175; Fri, 3 Dec 93 14:47:57 +0100 Date: Fri, 3 Dec 93 14:47:57 +0100 Message-Id: <9312031347.AA14175@inf-wiss.ivp.uni-konstanz.de> From: hoehle@post.inf-wiss.ivp.uni-konstanz.de (Joerg-Cyril Hoehle) To: clisp-list@[129.13.115.2] Subject: CLCS for CLISP In-Reply-To: <9312031336.AA17035@ursula.ee.pdx.edu> References: <9312031336.AA17035@ursula.ee.pdx.edu> Marcus Daniels writes: > Has anyone ported CLCS to CLISP? If so, does it work with CLISP's CLOS, or > do you need to use PCL? I don't think anyone has ported CLCS to CLISP, because that would require lots of changes to any function generating an error in CLISP, affecting the LISP and C source files. Lots of work. However it's still possible to make your own CLCS than can only handle errors generated by your application. But it won't be able to intercept CLISP generated ones. Joerg Hoehle. hoehle@inf-wiss.uni-konstanz.de From haible Fri Dec 3 20:26:28 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA27054; Fri, 3 Dec 93 20:26:28 +0100 Date: Fri, 3 Dec 93 20:26:28 +0100 From: haible (Bruno Haible) Message-Id: <9312031926.AA27054@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CLCS for CLISP Marcus Daniels asks: > Has anyone ported CLCS to CLISP? Assuming you are reffering to the condition system contained in the clcs/ subdirectory of the PCL and CLX distribution: I didn't port it because a. its implementation of HANDLER-BIND conses too much, b. as Joerg Hoehle pointed out, it is not of much use unless CLISP's run-time system generates conditions. > If so, does it work with CLISP's CLOS, or do you need to use PCL? CLCS doesn't need much more than DEFSTRUCT. It will certainly work with CLISP's CLOS. To avoid redundant work, I suggest you wait until the condition system David Gadbois has written for CLISP has been integrated. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From marcus@ee.pdx.edu Tue Dec 7 14:14:54 1993 Return-Path: Received: from ursula.ee.pdx.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA01518; Tue, 7 Dec 93 14:14:54 +0100 Received: by ursula.ee.pdx.edu (4.1/SMI-4.1) id AA10668; Tue, 7 Dec 93 05:14:50 PST Date: Tue, 7 Dec 93 05:14:50 PST From: marcus@ee.pdx.edu (Marcus Daniels) Message-Id: <9312071314.AA10668@ursula.ee.pdx.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: poor make-pipe-stream performance I'm archiving big records (up to 1.6 meg) as structures, using the #S() syntax. It is generated by a simple database interface program that communicates through standard I/O. The records consists primarily of double-vectors (6 sets, of 300-30000 elements). My plan was to use make-pipe-input-stream to `read' from this program. I've found that pipe I/O performance is not as good as AKCL. Reading directly from a file yields equivalent performance to AKCL, so I don't believe it is a O.S. thing (and there is plenty of resident memory free in all cases). CLISP and AKCL are both compiled with gcc-2.5.6 -O6 on Linux. CLISP doesn't use shared memory due to the 4MB limitation (in the following example the memory use is around 7.5 MB). Here were the results for the largest record (27951 * 6 * double-vector + 3 identifier fields). Real-time `Run-time' CLISP Straight (local) file: 37.6 35.1 CLISP MKFIFO file: 258.3 151.04 CLISP MAKE-PIPE-INPUT-STREAM 232.9 145.27 AKCL Straight (local) file: 33.0 N/A AKCL RUN-PROCESS 58.0 N/A The CLISP MKFIFO/MAKE-PIPE exec'ed way the shell... So the fact that AKCL uses execvp with no shell wouldn't seem to matter. Also, I played with using dd to reblock stuff for CLISP, but that didn't seem to accomplish anything.. I'd be willing to pay the 25 second penalty for such records, if I could use pipe, but not a 3 minute penalty. Are there any buffering parameters I might play with? What else could account for this discrepancy? The subshell approach was intended to be tolerable, hopefully temporary, substitute for a socket based or Postgres backend (perhaps something like Picasso or CLING). Has anyone ever worked out a simple socket oriented (i.e. database) interface for CLISP? I noticed that the only socket support CLISP includes is apparently for the sake of CLX -- not for general use. Is there any (technical) reason not to provide such an interface (other than it could be difficult or impossible to provide GPL'ed on all platforms). But a DJGPP network capable CLISP would come in _very_ handy for all the DOS bound PC crud around here! regards, Marcus Daniels From michaelw@karl.cs.su.oz.au Thu Dec 9 07:19:44 1993 Return-Path: Received: from joyce.cs.su.OZ.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA03719; Thu, 9 Dec 93 07:19:44 +0100 Message-Id: <9312090619.AA03719@ma2s2.mathematik.uni-karlsruhe.de> Received: from karl.cs.su.oz.au by joyce.cs.su.OZ.AU (mail from michaelw for clisp-list@ma2s2.mathematik.uni-karlsruhe.de) with MHSnet; Thu, 09 Dec 1993 17:17:57 +1100 Date: Thu, 9 Dec 1993 17:09:38 +1100 From: michaelw@karl.cs.su.OZ.AU (Michael Wise) Subject: CLISP to replace XLISP To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de A short while ago I placed a request in news for information about reasonably sized LISP implementations for use in teaching. For some time I've been using XLISP, and in fact have found it to be very satisfactory. The problem with it has been, however, the lack of a CLOS implementation. A number of people responded to that posting recommending CLISP, so here I am. Dumb question time. The doc that comes with the system basically says that CLISP implements CLtL1 plus parts of CLtL2, and parts of CLOS. As I have now obtained and am currently installing the latst version (clisp-1993-11-08), do I still need PCL if I want the complete CLOS? Secondly, is there anywhere a list of exactly which functions (and options to those functions) are supported. Thanks MichaelW Michael J. Wise Basser Department of Computer Science, F09 Sydney University, N.S.W. 2006 Australia Telephone: +61 2 692 4156 Messages: +61 2 692 3423 Facsimile: +61 2 692 3838 Internet/AARnet: michaelw@cs.su.oz.au UUCP.net: {mcvax,nttlab,ukc,uunet}!munnari!karl.cs.su.oz.au!michaelw -- "We have met the enemy, and they is us" - Pogo (Walt Kelly), 1970 From haible Thu Dec 9 13:39:41 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04143; Thu, 9 Dec 93 13:39:41 +0100 Date: Thu, 9 Dec 93 13:39:41 +0100 From: haible (Bruno Haible) Message-Id: <9312091239.AA04143@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: CLISP to replace XLISP Michael Wise asks: > Dumb question time. The doc that comes with the system basically > says that CLISP implements CLtL1 plus parts of CLtL2, and parts of > CLOS. That's true. > is there anywhere a list of exactly which functions > (and options to those functions) are supported. See impnotes.txt. Or evaluate (do-external-symbols (s "CLOS") (print s)) > As I have now obtained and am currently installing the latst version > (clisp-1993-11-08), do I still need PCL if I want the complete CLOS? You won't need PCL for most CLOS programs. Note that PCL isn't a "complete CLOS" either. For example, it doesn't support the :METHOD option in DEFGENERIC. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Thu Dec 9 17:26:41 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA04807; Thu, 9 Dec 93 17:26:41 +0100 Date: Thu, 9 Dec 93 17:26:41 +0100 From: haible (Bruno Haible) Message-Id: <9312091626.AA04807@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: poor make-pipe-stream performance Marcus Daniels compares CLISP pipes to AKCL pipes: > Real-time `Run-time' > CLISP Straight (local) file: 37.6 35.1 > CLISP MKFIFO file: 258.3 151.04 > CLISP MAKE-PIPE-INPUT-STREAM 232.9 145.27 > AKCL Straight (local) file: 33.0 N/A > AKCL RUN-PROCESS 58.0 N/A CLISP buffers regular files, because I can assume that the contents of a regular file won't change. When CLISP OPENs anything else (socket, fifo, pipe to another program) I didn't dare to make these streams buffered since reading too much or writing too late may have undesired effects (e.g. if one end of such a stream is interactive, or require FINISH-OUTPUT after every write operation, or even dead-locks). Thus when you READ from a pipe, READ-CHAR calls the operating system's read() function once for every character. It's not fast, but it is safe. You may use temporary files instead of pipes. Which to use depends on the amount of data you wish to transport, and on the operating system of course. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From dxs@evolving.com Fri Dec 10 22:57:10 1993 Return-Path: Received: from csn.org by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06317; Fri, 10 Dec 93 22:57:10 +0100 Received: by csn.org with UUCP id AA27984 (5.65c/IDA-1.4.4 for clisp-list@ma2s2.mathematik.uni-karlsruhe.de); Fri, 10 Dec 1993 14:55:07 -0700 Received: by kafka (AIX 3.2/UCB 5.64/4.03) id AA11773; Fri, 10 Dec 1993 14:31:15 -0700 Date: Fri, 10 Dec 1993 14:31:15 -0700 From: dxs@evolving.com (Dan Stanger) Message-Id: <9312102131.AA11773@kafka> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: tex conversion to postscript does anyone have postscrip versions of the cl-protos documentation or would anybody be willing to run tex on the files that come with cl-protos for me? i dont have access to tex on any machine here. thanks, dan stanger dxs@evolving.com From marcus@ee.pdx.edu Sat Dec 11 11:01:44 1993 Return-Path: Received: from ursula.ee.pdx.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA06805; Sat, 11 Dec 93 11:01:44 +0100 Received: by ursula.ee.pdx.edu (4.1/SMI-4.1) id AA22542; Sat, 11 Dec 93 02:00:58 PST Date: Sat, 11 Dec 93 02:00:58 PST From: marcus@ee.pdx.edu (Marcus Daniels) Message-Id: <9312111000.AA22542@ursula.ee.pdx.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: CLOS questions.. 1. The _impnotes mention that CALL-NEXT-METHOD doesn't take arguments. Is there a stylistic (etc.) reason one should be not be changing arguments on such calls? For me, it is a fairly common thing to do. 2. No MAKE-LOAD-FORM/SAVING-SLOTS? From marcus@ee.pdx.edu Sun Dec 12 00:25:38 1993 Return-Path: Received: from ursula.ee.pdx.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA07077; Sun, 12 Dec 93 00:25:38 +0100 Received: by ursula.ee.pdx.edu (4.1/SMI-4.1) id AA25936; Sat, 11 Dec 93 15:24:37 PST Date: Sat, 11 Dec 93 15:24:37 PST From: marcus@ee.pdx.edu (Marcus Daniels) Message-Id: <9312112324.AA25936@ursula.ee.pdx.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: hash-table puzzlement Could someone be so kinda as to tell me why this program reports different values after compilation? (eval-when (load compile eval) (defvar base) (defvar copies) (defstruct base (table (make-hash-table :test #'equal) :type hash-table)) (defstruct copies (table1 (make-hash-table :test #'equal) :type hash-table) (table2 (make-hash-table :test #'equal) :type hash-table)) (defun make-and-store-values-in-table () (setq base (make-base)) (setf (gethash "STR1" (base-table base)) 1) (setf (gethash "STR2" (base-table base)) 2) (setf (gethash "STR3" (base-table base)) 3) base ) (defun make-and-store-values-in-two-tables () (setq copies (make-copies)) (maphash #'(lambda (key value) (setf (gethash key (copies-table1 copies)) value) (setf (gethash key (copies-table2 copies)) value)) (base-table base)) copies ) ) (defconstant base-stored `#.(make-and-store-values-in-table)) (defconstant copies-stored `#.(make-and-store-values-in-two-tables)) (terpri) (princ (gethash "STR1" (copies-table1 copies-stored))) (terpri) (princ (gethash "STR1" (copies-table2 copies-stored))) (terpri) From miki@soum.co.jp Mon Dec 13 15:08:42 1993 Return-Path: Received: from soumgw.soum.co.jp by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA09913; Mon, 13 Dec 93 15:08:42 +0100 Received: from force.soum.co.jp (root@force.soum.co.jp [133.112.48.2]) by soumgw.soum.co.jp (8.5+2.0Wb2/3.019Wb-SOUM-GW) with SMTP id XAA01822; Mon, 13 Dec 1993 23:06:03 +0900 Received: by force.soum.co.jp (5.67+1.6W/2.8Wb-SOUM-MASTER) with SMTP id AA12386; Mon, 13 Dec 93 23:06:00 JST Message-Id: <9312131406.AA12386@force.soum.co.jp> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Cc: miki@soum.co.jp Subject: Compiling 1993-11-08 version on DEC Alpha OSF/1 V1.3 Date: Mon, 13 Dec 1993 23:05:59 +0900 From: Susumu Miki As CLISP is provided as a binary-kit for DEC Alpha running OSF/1, I've tried to compile it from source files, version 1993-11-08. I've installed gcc-2.5.5, and done all steps just like described as unix/INSTALL and unix/PLATFORMS as follows: 1. Perform "./target alpha-dec-osf1-with-gcc" in the top directory of the source tree. 2. Change directory as "cd alpha-dec-osf1-with-gcc" and type "./makemake > makefile" 3. Edit makefile for adding "-DNO_SINGLEMAP -DNO_MULTIMAP_SHM -DWIDE" to CFLAGS. 4. Type "make config.lsp" and edit it for site specific informations. 5. Type "make" Then I get clisp.run, that dumps core in the last step: ---- ./lisp.run -x "(load \"init.lsp\") (saveinitmem) (exit)" [2] + 20679 exit 1 make $* > make.log 2>&1 sh: 22894 Memory fault - core dumped *** Exit 139 Stop. ---- Would someone tell me how can I get an working executable file from the source files distributed. Thanks in advance. Susumu Miki miki@soum.co.jp Research and Development Dept. SOUM Corporation VOICE: 03-5453-1251 From haible Mon Dec 13 22:50:17 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA10448; Mon, 13 Dec 93 22:50:17 +0100 Date: Mon, 13 Dec 93 22:50:17 +0100 From: haible (Bruno Haible) Message-Id: <9312132150.AA10448@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Compiling 1993-11-08 version on DEC Alpha OSF/1 V1.3 > As CLISP is provided as a binary-kit for DEC Alpha running OSF/1, > I've tried to compile it from source files, version 1993-11-08. Support for DEC Alpha wasn't finished in that release. Since you are so desperately trying to compile it yourself, I made the necessary diffs (relative to clisp version 1993-11-08) available on ma2s2: /pub/lisp/clisp/binaries/decalpha-osf/decalpha-diffs.z . Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Mon Dec 13 23:02:12 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA10557; Mon, 13 Dec 93 23:02:12 +0100 Date: Mon, 13 Dec 93 23:02:12 +0100 From: haible (Bruno Haible) Message-Id: <9312132202.AA10557@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: hash-table puzzlement Marcus Daniels asks: > Could someone be so kinda as to tell me why this program reports > different values after compilation? > ... > (defconstant copies-stored `#.(make-and-store-values-in-two-tables)) This is a bug in the READ function: When it reads a hash-table and resolves #n= / #n# references it forgets to recompute the hash codes and to reorganize the hash table. This bug will be fixed in the next release. Thank you for reporting it. Use (gc) as a temporary workaround. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From miki@soum.co.jp Tue Dec 14 02:10:31 1993 Return-Path: Received: from soumgw.soum.co.jp by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA10676; Tue, 14 Dec 93 02:10:31 +0100 Received: from force.soum.co.jp (root@force.soum.co.jp [133.112.48.2]) by soumgw.soum.co.jp (8.5+2.0Wb2/3.019Wb-SOUM-GW) with SMTP id KAA15052; Tue, 14 Dec 1993 10:07:50 +0900 Received: by force.soum.co.jp (5.67+1.6W/2.8Wb-SOUM-MASTER) with SMTP id AA14899; Tue, 14 Dec 93 10:07:47 JST Message-Id: <9312140107.AA14899@force.soum.co.jp> To: clisp-list Subject: Re: Compiling 1993-11-08 version on DEC Alpha OSF/1 V1.3 In-Reply-To: Your message of "Mon, 13 Dec 1993 22:53:42 +0100." <9312132150.AA10448@ma2s2.mathematik.uni-karlsruhe.de> Date: Tue, 14 Dec 1993 10:07:46 +0900 From: Susumu Miki In article <9312132150.AA10448@ma2s2.mathematik.uni-karlsruhe.de>, haible@ma2s2.mathematik.uni-karlsruhe.de (Bruno Haible) writes: Bruno> Errors-To: haible@ma2s2.mathematik.uni-karlsruhe.de Bruno> Originator: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Bruno> Errors-To: haible@ma2s2.mathematik.uni-karlsruhe.de Bruno> Version: 5.5 -- Copyright (c) 1991/92, Anastasios Kotsikonas > Support for DEC Alpha wasn't finished in that release. Since you are so > desperately trying to compile it yourself, I made the necessary diffs > (relative to clisp version 1993-11-08) available on ma2s2: > /pub/lisp/clisp/binaries/decalpha-osf/decalpha-diffs.z . I've taken this via ftp, and will try again on it. Thank you very much for your efforts. Susumu Miki miki@soum.co.jp Research and Development Dept. SOUM Corporation VOICE: 03-5453-1251 From michaelw@karl.cs.su.oz.au Tue Dec 14 02:32:50 1993 Return-Path: Received: from joyce.cs.su.OZ.AU by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA10732; Tue, 14 Dec 93 02:32:50 +0100 Message-Id: <9312140132.AA10732@ma2s2.mathematik.uni-karlsruhe.de> Received: from karl.cs.su.oz.au by joyce.cs.su.OZ.AU (mail from michaelw for clisp-list@ma2s2.mathematik.uni-karlsruhe.de) with MHSnet; Tue, 14 Dec 1993 12:30:13 +1100 Date: Tue, 14 Dec 1993 12:24:27 +1100 From: michaelw@karl.cs.su.OZ.AU (Michael Wise) Subject: porting CLISP to MIPS R4000 boxes running RISC/os 5.01 To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de I'm not having an easy time at all porting CLISP to our MIPS R4000 boxes running RISC/os 5.01. Problem is that RISC/os 5.01 tries to cover three different flavours of UNIX: svr3, svr4 and bsd43, which is a nice idea in principle. However, what happens in practice is that confusion arises among the various include files which have a nasty habit of laughing at each other. (Been there, done that!) Has anyone gotten to the other side of this exercise; I'd very much like to get on with looking at CLISP itself! Cheers MichaelW From miki@soum.co.jp Wed Dec 15 05:58:49 1993 Return-Path: Received: from soumgw.soum.co.jp by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA12077; Wed, 15 Dec 93 05:58:49 +0100 Received: from force.soum.co.jp (root@force.soum.co.jp [133.112.48.2]) by soumgw.soum.co.jp (8.5+2.0Wb2/3.019Wb-SOUM-GW) with SMTP id NAA26117; Wed, 15 Dec 1993 13:55:52 +0900 Received: by force.soum.co.jp (5.67+1.6W/2.8Wb-SOUM-MASTER) with SMTP id AA12944; Wed, 15 Dec 93 13:55:46 JST Message-Id: <9312150455.AA12944@force.soum.co.jp> To: clisp-list Cc: miki@soum.co.jp Subject: Re: Compiling 1993-11-08 version on DEC Alpha OSF/1 V1.3 In-Reply-To: Your message of "Mon, 13 Dec 1993 22:53:42 +0100." <9312132150.AA10448@ma2s2.mathematik.uni-karlsruhe.de> Date: Wed, 15 Dec 1993 13:55:45 +0900 From: Susumu Miki In article <9312132150.AA10448@ma2s2.mathematik.uni-karlsruhe.de> haible@ma2s2.mathematik.uni-karlsruhe.de (Bruno Haible) writes: > Support for DEC Alpha wasn't finished in that release. Since you are so > desperately trying to compile it yourself, I made the necessary diffs > (relative to clisp version 1993-11-08) available on ma2s2: > /pub/lisp/clisp/binaries/decalpha-osf/decalpha-diffs.z . I applied this patch, and get executable "lisp.run" which does not dumps core, now. It passes "make test", but failes "make testsuite". It tells me as follows: ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ EQL-OK: T (SUBTYPEP 'BIT-VECTOR 'VECTOR) EQL-OK: T (SUBTYPEP 'SIMPLE-BIT-VECTOR 'SIMPLE-VECTOR) EQL-OK: NIL T Bye. ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Then, I tried on the same version for "sun4m" machines and get the same result.... Is this a bug in the current version of CLisp??? Susumu Miki Research and Development Dept. SOUM Corporation VOICE: 03-5453-1251 From haible Wed Dec 15 20:50:04 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA13018; Wed, 15 Dec 93 20:50:04 +0100 Date: Wed, 15 Dec 93 20:50:04 +0100 From: haible (Bruno Haible) Message-Id: <9312151950.AA13018@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: Compiling 1993-11-08 version on DEC Alpha OSF/1 V1.3 Susumu Miki : > It passes "make test", but failes "make testsuite". > Is this a bug in the current version of CLisp??? If everything else works and only the testsuite fails, then it is a bug in the testsuite :-) The suite/*.erg files contain the tests that failed. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Fri Dec 17 19:31:00 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA15871; Fri, 17 Dec 93 19:31:00 +0100 Date: Fri, 17 Dec 93 19:31:00 +0100 From: haible (Bruno Haible) Message-Id: <9312171831.AA15871@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list Subject: distributing CLISP You all know that CLISP is free software. One can download it via FTP. What do those do who don't have FTP access? They try to get it on floppy disks. I am not willing to distribute CLISP on disks to normal users (not co-developers) myself. If someone is willing to do that, he can. By GPL he can take any fee the customer is willing to pay. If you want to send disks with CLISP on them to people all over the world, please let me know. I'll forward you the addresses of people who contact me. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From marcus@ee.pdx.edu Sat Dec 18 21:17:15 1993 Return-Path: Received: from ursula.ee.pdx.edu by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA18059; Sat, 18 Dec 93 21:17:15 +0100 Received: by ursula.ee.pdx.edu (4.1/SMI-4.1) id AA03259; Sat, 18 Dec 93 12:14:55 PST Date: Sat, 18 Dec 93 12:14:55 PST From: marcus@ee.pdx.edu (Marcus Daniels) Message-Id: <9312182014.AA03259@ursula.ee.pdx.edu> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: single floats, NaNs I inconsistently get NaNs sometimes when reading negative single-floats of the form ``-629.25'' ``-800.1'' , etc.. When I read them as -629.25d0, it works. Why for is this thing? From kaehms@sedbsvr.se.ssd.lmsc.lockheed.com Wed Dec 22 23:29:06 1993 Return-Path: Received: from eagle.is.lmsc.lockheed.com by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA23326; Wed, 22 Dec 93 23:29:06 +0100 Received: from sedbsvr.ssd.lmsc.lockheed.com by eagle.is.lmsc.lockheed.com (5.65/Ultrix4.3-C) id AA03442; Wed, 22 Dec 1993 14:24:14 -0800 Message-Id: <9312222224.AA03442@eagle.is.lmsc.lockheed.com> Received: by sedbsvr.se.ssd.lmsc.lockheed.com (1.36.108.7/16.2) id AA22343; Wed, 22 Dec 1993 14:24:54 -0800 From: Bob Kaehms Subject: Building CLisp on an HP-700 8.07?? (help) To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Wed, 22 Dec 93 14:24:54 PST Cc: haible@ma2s2.mathematik.uni-karlsruhe.de Mailer: Elm [revision: 70.30] Hi, I've just started to try and build Clisp (clisp-1993-11-08) on an HP 700 running 8.07. I'm running into problems. After running the all-to-ascii program (it looked like a couple of the files had illegal characters, or unrecognizable ones in the filename and thus were not ptocessed by the program), the target program, configuring the config.lsp file, and running make, I get a few warnings about pointers and a funny one in various files s.a.: debug.d: 1125: warning- Illegal ^A, ^B, ^C, or ^D in source. Changed to space. 1125: warning- Illegal ^A, ^B, ^C, or ^D in source. ... ...it finally chokes with the following few messages: cc -Aa -z -D_HPUX_SOURCE -c misc.c unix.d: 445: warning- Redefinition of macro TCSETATTR. cc -Aa -z -D_HPUX_SOURCE -c predtype.c unix.d: 445: warning- Redefinition of macro TCSETATTR. cc -Aa -z -D_HPUX_SOURCE -c symbol.c unix.d: 445: warning- Redefinition of macro TCSETATTR. cc -Aa -z -D_HPUX_SOURCE -c lisparit0.c unix.d: 445: warning- Redefinition of macro TCSETATTR. lisparit0.d: 14584: Overflowed replacement buffer. *** Error code 1 I've tried defining CC both as cc -Aa -z -D_HPUX_SOURCE as well as c89 -z -D_HPUX_SOURCE ...Any hints would be appreciated! From michaelw@karl.cs.su.oz.au Mon Dec 27 09:26:51 1993 Return-Path: Received: from joyce.cs.su.OZ.AU ([129.78.8.208]) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA26535; Mon, 27 Dec 93 09:26:51 +0100 Message-Id: <9312270826.AA26535@ma2s2.mathematik.uni-karlsruhe.de> Received: from karl.cs.su.oz.au by joyce.cs.su.OZ.AU (mail from michaelw for clisp-list@ma2s2.mathematik.uni-karlsruhe.de) with MHSnet; Mon, 27 Dec 1993 19:21:48 +1100 Date: Mon, 27 Dec 1993 19:10:16 +1100 From: michaelw@karl.cs.su.OZ.AU (Michael Wise) Subject: the joys (?!!) of porting CLISP To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de So far I have tried porting it to R4000 RISCos 5.01, got almost no way down the track before it was total confused and hope less (due to the multiple universes being supported - best with the bsd flavour); R3000 with RISCos 4.52 - slightly better, but still got bogged. I then decided to move to the new Sparc Sun (running Solaris 2.XX) because we're about the move from the MIPS's in any case to a Sun multiprocessor. This attempt has been more successful. Main discovery was to define: LD_LIBRARY_PATH=/usr/ccs/lib:/usr/lib:/usr/openwin/lib and for good measure, change -Y P,/usr/ccs/lib:/usr/lib to -Y P,/usr/ccs/lib:/usr/lib:/usr/openwin/lib in the definition for lisp.run . I've now gotten stuck when trying to make interpreted.mem . After getting started ok and loading defseq.lsp, backquot.lsp, defmacro.lsp, macros1.lsp, macros2.lsp ... the following happens: ;; Loading file defs1.lsp ...*** Error code 139 make: Fatal error: Command failed for target `interpreted.mem' Any bright ideas? BTW, any way that the header can be suppressed at run time. The problem is that when the system will be run in batch mode (when student assignments are being marked), the extra output generated will have to be filtered or it will confuse the marking program. It's there- fore far easier to suppress it in the first place. Cheers MichaelW From dogan@srdc.metu.edu.tr Mon Dec 27 16:51:55 1993 Return-Path: Received: from srdc.metu.edu.tr ([144.122.230.11]) by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA26781; Mon, 27 Dec 93 16:51:55 +0100 Received: from karakoy.metu.edu.tr (karakoy.srdc.metu.edu.tr) by srdc.metu.edu.tr (4.1/SMI-4.1) id AA22710; Mon, 27 Dec 93 17:47:51 +0200 From: dogan@srdc.metu.edu.tr (Hakan DOGAN) Message-Id: <9312271447.AA22710@srdc.metu.edu.tr> Subject: compiler of clisp To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Date: Mon, 27 Dec 1993 17:47:58 +0300 (GMT+3:00) In-Reply-To: <9312222224.AA03442@eagle.is.lmsc.lockheed.com> from "Bob Kaehms" at Dec 22, 93 11:32:35 pm Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 160 hi . I want to learn if there's a compiler for clisp to run on unix workstations (sun or dec alpha) ? Hakan DOGAN dogan@srdc.metu.edu.tr From haible Mon Dec 27 21:45:28 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA28465; Mon, 27 Dec 93 21:45:28 +0100 Date: Mon, 27 Dec 93 21:45:28 +0100 From: haible (Bruno Haible) Message-Id: <9312272045.AA28465@ma2s2.mathematik.uni-karlsruhe.de> To: , clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: compiler of clisp Hakan Dogan asks: > there's a compiler for clisp to run on unix workstations (sun or dec alpha) ? CLISP includes a compiler, and there are binaries for Suns and DEC Alphas on ma2s2.mathematik.uni-karlsruhe.de. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Mon Dec 27 21:54:54 1993 Return-Path: Received: by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA28596; Mon, 27 Dec 93 21:54:54 +0100 Date: Mon, 27 Dec 93 21:54:54 +0100 From: haible (Bruno Haible) Message-Id: <9312272054.AA28596@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2.mathematik.uni-karlsruhe.de Subject: Re: the joys (?!!) of porting CLISP > BTW, any way that the header can be suppressed at run time. "man clisp" tells you about the command line option -q which stands for "quiet". Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de From haible Tue Dec 28 18:39:06 1993 Return-Path: Received: from ma2s3.mathematik.uni-karlsruhe.de by ma2s2.mathematik.uni-karlsruhe.de (4.1/SMI-4.0) id AA29955; Tue, 28 Dec 93 18:39:06 +0100 Date: Tue, 28 Dec 93 18:39:06 +0100 From: haible (Bruno Haible) Message-Id: <9312281739.AA29955@ma2s2.mathematik.uni-karlsruhe.de> To: clisp-list@ma2s2 Subject: Re: the joys (?!!) of porting CLISP Michael Wise experienced problems when building clisp on Solaris 2.3: > ;; Loading file defs1.lsp ...*** Error code 139 > make: Fatal error: Command failed for target `interpreted.mem' He had an unusable C compiler and should upgrade to the newest gcc. Bruno Haible haible@ma2s2.mathematik.uni-karlsruhe.de