From amoroso@mclink.it Sat Mar 6 04:41:20 1999 Received: from mail1.mclink.it (net128-007.mclink.it [195.110.128.7]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id EAA28664 for ; Sat, 6 Mar 1999 04:41:16 -0800 (PST) Received: from net145-048.mclink.it (net145-048.mclink.it [195.110.145.48]) by mail1.mclink.it (8.9.1/8.9.0) with SMTP id NAA11607 for ; Sat, 6 Mar 1999 13:43:54 +0100 (CET) From: amoroso@mclink.it (Paolo Amoroso) To: clisp-list@clisp.cons.org Subject: CLOS MOP Date: Sat, 06 Mar 1999 12:44:02 GMT Organization: Paolo Amoroso - Milan, ITALY Message-ID: <36e10243.42637@mail.mclink.it> X-Mailer: Forte Agent 1.5/32.451 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Is there any plan to add native--i.e. without PCL--support for the CLOS MOP? I'm just curious, I don't need to use the MOP. Paolo -- Paolo Amoroso From haible@ilog.fr Sun Mar 7 11:03:30 1999 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA12107 for ; Sun, 7 Mar 1999 11:03:25 -0800 (PST) Received: from ilog.ilog.fr ([172.17.4.22]) by sceaux.ilog.fr (8.9.1/8.9.1) with ESMTP id UAA06278 for ; Sun, 7 Mar 1999 20:06:04 +0100 (MET) Received: from jaures.ilog.fr (jaures.ilog.fr [172.16.1.81]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id UAA11169; Sun, 7 Mar 1999 20:05:18 +0100 (MET) From: Bruno Haible Received: (from haible@localhost) by jaures.ilog.fr (8.9.0/8.9.0) id UAA20448; Sun, 7 Mar 1999 20:06:03 +0100 (MET) Date: Sun, 7 Mar 1999 20:06:03 +0100 (MET) Message-Id: <199903071906.UAA20448@jaures.ilog.fr> To: clisp-list@seagull.cons.org Subject: Re: CLOS MOP In-Reply-To: <36e10243.42637@mail.mclink.it> References: <36e10243.42637@mail.mclink.it> Paolo Amoroso asks: > Is there any plan to add native--i.e. without PCL--support for the CLOS > MOP? I'm just curious, I don't need to use the MOP. The meta-object protocol makes it harder for an implementor to optimize things like `make-instance' and generic function call. If someone says which parts of the MOP he needs in order to subclass, say, Gtk classes in Lisp, and it is feasible without penalty on the regular `make-instance' and generic function call, we will do it. Bruno From aler@inf.uc3m.es Sun Mar 14 06:58:35 1999 Received: from elrond.uc3m.es (elrond.uc3m.es [163.117.136.62]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id GAA09147 for ; Sun, 14 Mar 1999 06:58:32 -0800 (PST) Received: from tristan.uc3m.es (tristan.uc3m.es [163.117.129.72]) by elrond.uc3m.es (8.9.3/8.9.1) with SMTP id QAA04027 for ; Sun, 14 Mar 1999 16:12:31 GMT Date: Sun, 14 Mar 1999 16:02:01 +0100 (CET) From: Ricardo Aler Mur X-Sender: aler@tristan.uc3m.es Reply-To: aler@inf.uc3m.es To: clisp-list@seagull.cons.org Subject: CLASP in Clisp In-Reply-To: <199903071906.UAA20448@jaures.ilog.fr> Message-ID: Distribution: world MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, Has anybody tried to use CLASP with Clisp?. CLASP is a statistical package written for Allegro CL. It seems to require CLIM. Any help would be appreciated. Cheers, Ricardo. -------- mailto: "Ricardo Aler Mur" http://grial.uc3m.es/~aler From russell@coulee.tdb.com Thu Mar 18 23:58:15 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id XAA15869 for ; Thu, 18 Mar 1999 23:54:29 -0800 (PST) Received: (qmail 29912 invoked from network); 19 Mar 1999 07:54:29 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 19 Mar 1999 07:54:29 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for clisp-list@seagull.cons.org; Thu, 18 Mar 1999 23:54:29 -0800 (PST) Resent-Message-Id: Resent-To: clisp-list@seagull.cons.org Resent-From: Russell Senior Resent-Date: 18 Mar 1999 23:54:29 -0800 To: clisp-list@seagull.cons.org Subject: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 From: Russell Senior Date: 18 Mar 1999 23:00:57 -0800 Message-ID: <86hfri3qvq.fsf@coulee.tdb.com> X-Mailer: Gnus v5.6.39/Emacs 20.3 Lines: 49 I have tried building from source using both gcc-2.8.1 and egcs-1.1.2. Both (these examples are from the egcs build attempt) fail in the same place during make: /packages/development/egcs/bin/gcc -O -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type -fomit-frame-pointer -Wno-sign-compare -O2 -fexpensive-optimizations -DDYNAMIC_FFI -c foreign.c In file included from foreign.d:5: lispbibl.d:7314: warning: register used for two global register variables foreign.d: In function `foreign_layout': foreign.d:634: `sint64' undeclared (first use in this function) foreign.d:634: (Each undeclared identifier is reported only once foreign.d:634: for each function it appears in.) foreign.d:634: parse error before `sint64' foreign.d:634: warning: no semicolon at end of struct or union foreign.d:635: `uint64' undeclared (first use in this function) foreign.d:645: parse error before `uint64' foreign.d:645: warning: no semicolon at end of struct or union foreign.d: In function `convert_from_foreign': foreign.d:953: `sint64' undeclared (first use in this function) foreign.d:953: parse error before `val' foreign.d:955: `val' undeclared (first use in this function) foreign.d:967: `uint64' undeclared (first use in this function) foreign.d:967: parse error before `val' foreign.d: In function `convert_to_foreign': foreign.d:1663: `sint64' undeclared (first use in this function) foreign.d:1663: parse error before `val' foreign.d:1665: `val' undeclared (first use in this function) foreign.d:1674: `uint64' undeclared (first use in this function) foreign.d:1674: parse error before `val' foreign.d: In function `do_va_start': foreign.d:3099: warning: left-hand operand of comma expression has no effect foreign.d:3101: warning: left-hand operand of comma expression has no effect make: *** [foreign.o] Error 1 russell@dalles[/build/clisp] $ I am also seeing lots of warnings of the form: /packages/development/egcs/bin/gcc -O -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type -fomit-frame-pointer -Wno-sign-compare -O2 -fexpensive-optimizations -DDYNAMIC_FFI -c spvw.c In file included from spvw.d:21: lispbibl.d:7314: warning: register used for two global register variables spvw.d: In function `main': spvw.d:1778: warning: variable `argv_memneed' might be clobbered by `longjmp' or `vfork' Any ideas on what's going wrong here? -- Russell Senior seniorr@teleport.com From sds@eaglets.com Fri Mar 19 06:33:14 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id GAA19367 for ; Fri, 19 Mar 1999 06:33:02 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Fri, 19 Mar 1999 09:30:54 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Fri, 19 Mar 1999 09:30:53 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id JAA11428; Fri, 19 Mar 1999 09:32:20 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: <86hfri3qvq.fsf@coulee.tdb.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Russell Senior's message of "Fri, 19 Mar 1999 00:03:44 -0800 (PST)" Date: 19 Mar 1999 09:32:20 -0500 Message-ID: Lines: 60 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <86hfri3qvq.fsf@coulee.tdb.com> >>>> On the subject of "Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1" >>>> Sent on Fri, 19 Mar 1999 00:03:44 -0800 (PST) >>>> Honorable Russell Senior writes: >> I have tried building from source using both gcc-2.8.1 and egcs-1.1.2. >> >> Both (these examples are from the egcs build attempt) fail in the same >> place during make: >> >> /packages/development/egcs/bin/gcc -O -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type -fomit-frame-pointer -Wno-sign-compare -O2 -fexpensive-optimizations -DDYNAMIC_FFI -c foreign.c >> In file included from foreign.d:5: >> lispbibl.d:7314: warning: register used for two global register variables >> foreign.d: In function `foreign_layout': >> foreign.d:634: `sint64' undeclared (first use in this function) >> foreign.d:634: (Each undeclared identifier is reported only once >> foreign.d:634: for each function it appears in.) >> foreign.d:634: parse error before `sint64' >> foreign.d:634: warning: no semicolon at end of struct or union >> foreign.d:635: `uint64' undeclared (first use in this function) >> foreign.d:645: parse error before `uint64' >> foreign.d:645: warning: no semicolon at end of struct or union >> foreign.d: In function `convert_from_foreign': >> foreign.d:953: `sint64' undeclared (first use in this function) >> foreign.d:953: parse error before `val' >> foreign.d:955: `val' undeclared (first use in this function) >> foreign.d:967: `uint64' undeclared (first use in this function) >> foreign.d:967: parse error before `val' >> foreign.d: In function `convert_to_foreign': >> foreign.d:1663: `sint64' undeclared (first use in this function) >> foreign.d:1663: parse error before `val' >> foreign.d:1665: `val' undeclared (first use in this function) >> foreign.d:1674: `uint64' undeclared (first use in this function) >> foreign.d:1674: parse error before `val' >> foreign.d: In function `do_va_start': >> foreign.d:3099: warning: left-hand operand of comma expression has no effect >> foreign.d:3101: warning: left-hand operand of comma expression has no effect >> make: *** [foreign.o] Error 1 >> russell@dalles[/build/clisp] $ I've never seen this. Please indicate what sources you are using (release, snapshot), as well as the configure command you issued. >> I am also seeing lots of warnings of the form: >> >> /packages/development/egcs/bin/gcc -O -W -Wswitch -Wcomment -Wpointer-arith -Wimplicit -Wreturn-type -fomit-frame-pointer -Wno-sign-compare -O2 -fexpensive-optimizations -DDYNAMIC_FFI -c spvw.c >> In file included from spvw.d:21: >> lispbibl.d:7314: warning: register used for two global register variables >> spvw.d: In function `main': >> spvw.d:1778: warning: variable `argv_memneed' might be clobbered by `longjmp' or `vfork' >> >> Any ideas on what's going wrong here? this is okay. -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Don't use force -- get a bigger hammer. From russell@coulee.tdb.com Fri Mar 19 07:42:18 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id HAA20140 for ; Fri, 19 Mar 1999 07:42:05 -0800 (PST) Received: (qmail 32319 invoked from network); 19 Mar 1999 15:42:05 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 19 Mar 1999 15:42:05 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Fri, 19 Mar 1999 07:42:04 -0800 (PST) To: clisp-list@seagull.cons.org Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: From: Russell Senior Date: 19 Mar 1999 07:42:03 -0800 In-Reply-To: Sam Steingold's message of "Fri, 19 Mar 1999 06:37:56 -0800 (PST)" Message-ID: <86pv657agk.fsf@coulee.tdb.com> Lines: 17 X-Mailer: Gnus v5.6.39/Emacs 20.3 >>>>> "Sam" == Sam Steingold writes: Sam> I've never seen this. Please indicate what sources you are using Sam> (release, snapshot), as well as the configure command you issued. Sorry. It is the clisp-1999-01-08 release. I configured from a build directory with a pretty plain-jane: /src/clisp-1999-01-08/configure --srcdir=/src/clisp-1999-01-08 --prefix=/packages/development/lisp/clisp then followed the supplied directions on makemake and config.lsp. The error occurred during the `make' step. -- Russell Senior seniorr@teleport.com From sds@eaglets.com Fri Mar 19 08:27:01 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA20721 for ; Fri, 19 Mar 1999 08:25:52 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Fri, 19 Mar 1999 11:21:28 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Fri, 19 Mar 1999 11:21:28 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id LAA11594; Fri, 19 Mar 1999 11:21:29 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: <86pv657agk.fsf@coulee.tdb.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Russell Senior's message of "Fri, 19 Mar 1999 07:45:42 -0800 (PST)" Date: 19 Mar 1999 11:21:29 -0500 Message-ID: Lines: 33 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <86pv657agk.fsf@coulee.tdb.com> >>>> On the subject of "Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1" >>>> Sent on Fri, 19 Mar 1999 07:45:42 -0800 (PST) >>>> Honorable Russell Senior writes: >> >>>>> "Sam" == Sam Steingold writes: >> >> Sam> I've never seen this. Please indicate what sources you are using >> Sam> (release, snapshot), as well as the configure command you issued. >> >> Sorry. It is the clisp-1999-01-08 release. >> >> I configured from a build directory with a pretty plain-jane: >> >> /src/clisp-1999-01-08/configure --srcdir=/src/clisp-1999-01-08 --prefix=/packages/development/lisp/clisp >> >> then followed the supplied directions on makemake and config.lsp. >> The error occurred during the `make' step. this is fairly strange. I have a similar system (egcs 1.1.1, glibc 2.1) and I do not see this. where did you get the source tree? (cellar.goems.com? cons.org?) Not that it should matter... :-) If you do not insist on compiling yourself, you can find glib2.1 binary RPMs on ftp://cellar.goems.com/pub/clisp. If you do insist on compiling yourself, you might try to edit foreign.d and replacing the offending symbols which whatever is defined on your system. -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. The world is coming to an end. Please log off. From russell@coulee.tdb.com Fri Mar 19 09:55:29 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id JAA21763 for ; Fri, 19 Mar 1999 09:55:18 -0800 (PST) Received: (qmail 623 invoked from network); 19 Mar 1999 17:55:17 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 19 Mar 1999 17:55:17 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Fri, 19 Mar 1999 09:55:16 -0800 (PST) To: clisp-list@seagull.cons.org Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: From: Russell Senior Date: 19 Mar 1999 09:55:16 -0800 In-Reply-To: Sam Steingold's message of "Fri, 19 Mar 1999 08:32:09 -0800 (PST)" Message-ID: <86yakt5pq3.fsf@coulee.tdb.com> Lines: 27 X-Mailer: Gnus v5.6.39/Emacs 20.3 >>>>> "Sam" == Sam Steingold writes: Sam> this is fairly strange. I have a similar system (egcs 1.1.1, Sam> glibc 2.1) and I do not see this. where did you get the source Sam> tree? (cellar.goems.com? cons.org?) Not that it should Sam> matter... :-) I got it (1999-01-08) originally from cons.org back in early February. $ md5sum clispsrc.tar.gz 2faa2d691742f2ab4d259c51fe1afaf5 clispsrc.tar.gz Sam> [...] If you do not insist on compiling yourself, you can find glib2.1 Sam> binary RPMs on ftp://cellar.goems.com/pub/clisp. If you do Sam> insist on compiling yourself, you might try to edit foreign.d and Sam> replacing the offending symbols which whatever is defined on your Sam> system. I prefer building from source. I'll investigate. I am also going to try the 1999-03-19 version from cellar.goems.com and see how that goes. Thanks. -- Russell Senior seniorr@teleport.com From sds@eaglets.com Fri Mar 19 10:44:50 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id KAA22438 for ; Fri, 19 Mar 1999 10:44:46 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Fri, 19 Mar 1999 13:42:12 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Fri, 19 Mar 1999 13:42:11 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id NAA24748; Fri, 19 Mar 1999 13:43:37 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: <86yakt5pq3.fsf@coulee.tdb.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Russell Senior's message of "Fri, 19 Mar 1999 10:00:13 -0800 (PST)" Date: 19 Mar 1999 13:43:36 -0500 Message-ID: Lines: 17 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <86yakt5pq3.fsf@coulee.tdb.com> >>>> On the subject of "Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1" >>>> Sent on Fri, 19 Mar 1999 10:00:13 -0800 (PST) >>>> Honorable Russell Senior writes: >> >> I prefer building from source. I'll investigate. I am also going >> to try the 1999-03-19 version from cellar.goems.com and see how that >> goes. please beware - this is a snapshot: it might not compile either! [it includes UNICODE support though] -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Those who can laugh at themselves will never cease to be amused. From sds@eaglets.com Mon Mar 22 10:11:08 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id KAA26053 for ; Mon, 22 Mar 1999 10:10:35 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Mon, 22 Mar 1999 13:09:29 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Mon, 22 Mar 1999 13:09:28 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id NAA11665; Mon, 22 Mar 1999 13:10:24 -0500 Sender: "Shteingold, Sam" To: of list Subject: Re: merge-pathnames bug? References: <199902151353.OAA13028@zeus.gmd.de> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Joerg Hoehle's message of "Mon, 15 Feb 1999 06:05:06 -0800 (PST)" Date: 22 Mar 1999 13:10:24 -0500 Message-ID: Lines: 20 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <199902151353.OAA13028@zeus.gmd.de> >>>> On the subject of "Re: merge-pathnames bug?" >>>> Sent on Mon, 15 Feb 1999 06:05:06 -0800 (PST) >>>> Honorable Joerg Hoehle writes: >> >> This doesn't seem to be documented in _impnotes.txt. It is now. see (tomorrow) ftp://cellar.goems.com/pub/clisp/impnotes.html#pathmerge >> I'd vote for the change too. I implemented the change under the `-a' (more ANSI CL compliance) flag. It will be in the tomorrows snapshot. -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Those who don't know lisp are destined to reinvent it, poorly. From russell@coulee.tdb.com Tue Mar 23 23:41:26 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id XAA16668 for ; Tue, 23 Mar 1999 23:41:17 -0800 (PST) Received: (qmail 24760 invoked from network); 24 Mar 1999 07:41:36 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 24 Mar 1999 07:41:36 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Tue, 23 Mar 1999 23:41:35 -0800 (PST) To: clisp-list@seagull.cons.org Cc: Multiple recipients of list Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: From: Russell Senior Date: 23 Mar 1999 23:41:34 -0800 In-Reply-To: Sam Steingold's message of "Fri, 19 Mar 1999 10:52:41 -0800 (PST)" Message-ID: <86d81zbahd.fsf@coulee.tdb.com> Lines: 36 X-Mailer: Gnus v5.6.39/Emacs 20.3 Russell> I prefer building from source. I'll investigate. Under linux-2.2.4 using both gcc-2.8.1 and egcs-1.1.2 and glibc-2.1, building clisp-1999-01-08 from source succeeded after making the following change: diff -r -u clisp-1999-01-08/src/unix.d clisp-1999-01-08-munged/src/unix.d --- clisp-1999-01-08/src/unix.d Thu Jan 7 08:33:51 1999 +++ clisp-1999-01-08-munged/src/unix.d Tue Mar 23 21:22:15 1999 @@ -845,7 +845,7 @@ extern_C void* dlopen (const char * library, int flag); extern_C void* dlsym (void* handle, DLSYM_CONST char * symbol); extern_C int dlclose (void* handle); - extern_C DLERROR_CONST char * dlerror (void); +# extern_C DLERROR_CONST char * dlerror (void); #define HAVE_DYNLOAD #endif I observed some mystery segfaults during the builds which seemed to go away when I reran make, giving the appearance of a possible intermittent hardware (thermal?) problem with my box. This might explain some of the problems I saw earlier, so I can't say with any confidence that the new linux kernel had any influence on this evenings success. Also, in my local version I have modified src/lisparit.d to use 'DOUBLE-FLOAT instead of 'SINGLE-FLOAT for *DEFAULT-FLOAT-FORMAT* and *READ-DEFAULT-FLOAT-FORMAT*. This causes the `make check' to fail several tests, apparently innocously (`make check' succeeded before this change). Perhaps if the tests rely on these values being 'SINGLE-FLOAT they should be set explicitly in tests.lsp? -- Russell Senior seniorr@teleport.com From russell@coulee.tdb.com Wed Mar 24 02:11:53 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id CAA18059 for ; Wed, 24 Mar 1999 02:11:39 -0800 (PST) Received: (qmail 25394 invoked from network); 24 Mar 1999 10:12:16 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 24 Mar 1999 10:12:16 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Wed, 24 Mar 1999 02:12:16 -0800 (PST) To: clisp-list@seagull.cons.org Subject: clisp-1999-01-08 install buglet References: <86d81zbahd.fsf@coulee.tdb.com> From: Russell Senior Date: 24 Mar 1999 02:12:16 -0800 In-Reply-To: Russell Senior's message of "Tue, 23 Mar 1999 23:45:42 -0800 (PST)" Message-ID: <864snbrybj.fsf@coulee.tdb.com> Lines: 10 X-Mailer: Gnus v5.6.39/Emacs 20.3 Linux-2.2.4, gcc-2.8.1, glibc-2.1. When configuring with a --prefix=foo option, the specified directory `foo' is not automatically created with `make install'. I don't know precisely how this is dealt with in other packages. -- Russell Senior seniorr@teleport.com From sds@eaglets.com Wed Mar 24 06:10:25 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id GAA20037 for ; Wed, 24 Mar 1999 06:10:08 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Wed, 24 Mar 1999 09:09:40 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Wed, 24 Mar 1999 09:09:40 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id JAA25471; Wed, 24 Mar 1999 09:10:21 -0500 Sender: "Shteingold, Sam" To: list Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: <86d81zbahd.fsf@coulee.tdb.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Russell Senior's message of "Tue, 23 Mar 1999 23:47:21 -0800 (PST)" Date: 24 Mar 1999 09:10:21 -0500 Message-ID: Lines: 54 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <86d81zbahd.fsf@coulee.tdb.com> >>>> On the subject of "Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1" >>>> Sent on Tue, 23 Mar 1999 23:47:21 -0800 (PST) >>>> Honorable Russell Senior writes: >> Russell> I prefer building from source. I'll investigate. >> >> Under linux-2.2.4 using both gcc-2.8.1 and egcs-1.1.2 and glibc-2.1, >> building clisp-1999-01-08 from source succeeded after making the >> following change: >> >> diff -r -u clisp-1999-01-08/src/unix.d clisp-1999-01-08-munged/src/unix.d >> --- clisp-1999-01-08/src/unix.d Thu Jan 7 08:33:51 1999 >> +++ clisp-1999-01-08-munged/src/unix.d Tue Mar 23 21:22:15 1999 >> @@ -845,7 +845,7 @@ >> extern_C void* dlopen (const char * library, int flag); >> extern_C void* dlsym (void* handle, DLSYM_CONST char * symbol); >> extern_C int dlclose (void* handle); >> - extern_C DLERROR_CONST char * dlerror (void); >> +# extern_C DLERROR_CONST char * dlerror (void); >> #define HAVE_DYNLOAD >> #endif yep. I am getting this with the current CVS tree. I hope this will be fixed in the next release. >> I observed some mystery segfaults during the builds which seemed to >> go away when I reran make, giving the appearance of a possible >> intermittent hardware (thermal?) problem with my box. This might >> explain some of the problems I saw earlier, so I can't say with any >> confidence that the new linux kernel had any influence on this >> evenings success. yep. I am getting these too ("... indicative of gcc bug &c") they go away on a second or third attempt. >> Also, in my local version I have modified src/lisparit.d to use >> 'DOUBLE-FLOAT instead of 'SINGLE-FLOAT for *DEFAULT-FLOAT-FORMAT* >> and *READ-DEFAULT-FLOAT-FORMAT*. This causes the `make check' to >> fail several tests, apparently innocously (`make check' succeeded >> before this change). Perhaps if the tests rely on these values >> being 'SINGLE-FLOAT they should be set explicitly in tests.lsp? no-no-no!!! HyperSpec/Body/var_stread-de_oat-formatst.html If you want a different value for these variables, set them in ~/.clisrc or dump you own memory image (like I do). -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. A man paints with his brains and not with his hands. From sds@eaglets.com Wed Mar 24 06:21:15 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id GAA20226 for ; Wed, 24 Mar 1999 06:21:14 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Wed, 24 Mar 1999 09:20:30 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Wed, 24 Mar 1999 09:20:29 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id JAA25486; Wed, 24 Mar 1999 09:21:09 -0500 Sender: "Shteingold, Sam" To: of list Subject: Re: clisp-1999-01-08 install buglet References: <864snbrybj.fsf@coulee.tdb.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Russell Senior's message of "Wed, 24 Mar 1999 02:15:53 -0800 (PST)" Date: 24 Mar 1999 09:21:09 -0500 Message-ID: Lines: 22 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <864snbrybj.fsf@coulee.tdb.com> >>>> On the subject of "clisp-1999-01-08 install buglet" >>>> Sent on Wed, 24 Mar 1999 02:15:53 -0800 (PST) >>>> Honorable Russell Senior writes: >> Linux-2.2.4, gcc-2.8.1, glibc-2.1. >> >> When configuring with a --prefix=foo option, the specified directory >> `foo' is not automatically created with `make install'. I don't know >> precisely how this is dealt with in other packages. first, you shouldn't do something like this. second, it's not hard to fix it: just call mkdir -p instead of just mkdir. third, I am not sure that all mkdir implementations support -p, so I will wait for Bruno to come back from his vacation before committing this. -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. .sigs are like your face - rarely seen by you and uglier than you think From sparrow@sparrow.diamant.ru Thu Mar 25 09:11:29 1999 Received: from sparrow.diamant.ru (qmailr@sparrow.diamant.ru [194.186.43.21]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id JAA06553 for ; Thu, 25 Mar 1999 09:11:24 -0800 (PST) Received: (qmail 5410 invoked by uid 50002); 25 Mar 1999 17:15:03 -0000 To: clisp-list@clisp.cons.org Subject: communication via unix pipes MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii From: Alexander Vorobiev Date: 25 Mar 1999 20:15:02 +0300 Message-ID: Lines: 50 User-Agent: Gnus/5.07008 (Pterodactyl Gnus v0.80) XEmacs/20.4 (Emerald) Hello, everybody! I am trying to write a program (using clisp on linux) which is supposed to run as a subprocess and communicate with its parent (big C program) through 2 unidirectional pipes. File descriptors of the pipes are given to my program in command line. Information is transmitted as 32-bit packets (unsigned long in C terminology). The first thing i tried was to use /dev/fd/* files (as I know appropriate file descriptors): #!/usr/local/bin/clisp -q -C (defvar *receiver-pathname* (make-pathname :directory '(:absolute "dev" "fd") :name (car lisp:*args*))) (defvar *sender-pathname* (make-pathname :directory '(:absolute "dev" "fd") :name (cadr lisp:*args*))) (defconstant +header-size+ 4) ; I want to read a header of some big packet (defvar *header* (make-array +header-size+ :element-type '(unsigned-byte 32) :initial-element 0)) (defun read-header () (with-open-file (sender *sender-pathname* :direction :input :element-type '(unsigned-byte 32)) (read-sequence *header* sender))) (read-header) Unfortunately, this didn't work: *** - file #P"/proc/5395/fd/pipe:[1339598]" does not exist Is it possible to make clisp deal with such "special" files properly? Or maybe it is possible to avoid using /dev/fd/* files and to connect streams directly to open file descriptors? And the last question about C to lisp communication: should I use some ffi tools to convert data for transmitting to/from parent process written in C? Thanks in advance, Alexander From russell@coulee.tdb.com Thu Mar 25 10:58:33 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10] (may be forged)) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id KAA07954 for ; Thu, 25 Mar 1999 10:58:26 -0800 (PST) Received: (qmail 2185 invoked from network); 25 Mar 1999 18:58:07 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 25 Mar 1999 18:58:07 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Thu, 25 Mar 1999 10:58:06 -0800 (PST) To: clisp-list@seagull.cons.org Cc: Multiple recipients of list Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: From: Russell Senior Date: 25 Mar 1999 10:58:06 -0800 In-Reply-To: Sam Steingold's message of "Wed, 24 Mar 1999 06:14:26 -0800 (PST)" Message-ID: <86d81xz9a9.fsf@coulee.tdb.com> Lines: 23 X-Mailer: Gnus v5.6.39/Emacs 20.3 >>>>> "Sam" == Sam Steingold writes: Russell> I observed some mystery segfaults during the builds which Russell> seemed to go away when I reran make, giving the appearance of Russell> a possible intermittent hardware (thermal?) problem with my Russell> box. This might explain some of the problems I saw earlier, Russell> so I can't say with any confidence that the new linux kernel Russell> had any influence on this evenings success. Sam> yep. I am getting these too ("... indicative of gcc bug &c") Sam> they go away on a second or third attempt. New information implies that it might be a linux kernel problem. The errors I was seeing were signal 9 (KILL) processes deaths apparently resulting from memory management issues in the kernel. With linux-2.2.4, I am now seeing these problems primarily during the configuration stage, and not during the compile. I suspect this is what caused my earlier problems with the long long: confusion from abnormally terminated autoconf tests. -- Russell Senior seniorr@teleport.com From russell@coulee.tdb.com Thu Mar 25 11:15:38 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id LAA08275 for ; Thu, 25 Mar 1999 11:15:22 -0800 (PST) Received: (qmail 2310 invoked from network); 25 Mar 1999 19:16:15 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 25 Mar 1999 19:16:15 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Thu, 25 Mar 1999 11:16:15 -0800 (PST) To: clisp-list@seagull.cons.org Cc: Multiple recipients of list Subject: Re: clisp-1999-01-08 install buglet References: From: Russell Senior Date: 25 Mar 1999 11:16:15 -0800 In-Reply-To: Sam Steingold's message of "Wed, 24 Mar 1999 06:27:33 -0800 (PST)" Message-ID: <86aex1z8g0.fsf@coulee.tdb.com> Lines: 29 X-Mailer: Gnus v5.6.39/Emacs 20.3 >>>>> "Sam" == Sam Steingold writes: Russell> Linux-2.2.4, gcc-2.8.1, glibc-2.1. Russell> When configuring with a --prefix=foo option, the specified Russell> directory `foo' is not automatically created with `make Russell> install'. I don't know precisely how this is dealt with in Russell> other packages. Sam> first, you shouldn't do something like this. Why not? I find it quite useful. Sam> second, it's not hard to fix it: just call mkdir -p instead of Sam> just mkdir. Sam> third, I am not sure that all mkdir implementations support -p, Sam> so I will wait for Bruno to come back from his vacation before Sam> committing this. Automake seems to provide a script called mkinstalldirs for automake'd/autoconf'd packages, which I expect resolves many portability issues. I believe I have seen the mkdir -p option used elsewhere. -- Russell Senior seniorr@teleport.com From sds@eaglets.com Thu Mar 25 11:19:33 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA08376 for ; Thu, 25 Mar 1999 11:19:15 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Thu, 25 Mar 1999 14:17:59 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Thu, 25 Mar 1999 14:17:59 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id OAA30983; Thu, 25 Mar 1999 14:19:29 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: communication via unix pipes References: Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Alexander Vorobiev's message of "Thu, 25 Mar 1999 09:19:09 -0800 (PST)" Date: 25 Mar 1999 14:19:29 -0500 Message-ID: Lines: 24 X-Mailer: Gnus v5.7/Emacs 20.3 Privet! First of all, I don't think I can help you with the particular problem you are having. I have no idea what can go wrong. Nevertheless, I'd think that the method you are using is not optimal. Generally speaking, file descriptors are so low-level things that I would be reluctant to use them for communication between programs. Also, by using /dev/fd/, you are limiting portability of your program to modern unixes (basically, to Linux and Solaris :-). >> I am trying to write a program (using clisp on linux) which is >> supposed to run as a subprocess and communicate with its parent (big >> C program) through 2 unidirectional pipes. File descriptors of the why use 2 pipes when you can use one socket? you can even make the two program communicate via TCP/IP (and have them running on different computers). -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Never trust a man who can count to 1024 on his fingers. From sds@eaglets.com Thu Mar 25 11:44:57 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA08986 for ; Thu, 25 Mar 1999 11:44:19 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Thu, 25 Mar 1999 14:43:08 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Thu, 25 Mar 1999 14:43:08 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id OAA31181; Thu, 25 Mar 1999 14:44:39 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: <86d81xz9a9.fsf@coulee.tdb.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Russell Senior's message of "Thu, 25 Mar 1999 11:02:54 -0800 (PST)" Date: 25 Mar 1999 14:44:38 -0500 Message-ID: Lines: 35 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <86d81xz9a9.fsf@coulee.tdb.com> >>>> On the subject of "Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1" >>>> Sent on Thu, 25 Mar 1999 11:02:54 -0800 (PST) >>>> Honorable Russell Senior writes: >> >>>>> "Sam" == Sam Steingold writes: >> >> Russell> I observed some mystery segfaults during the builds which >> Russell> seemed to go away when I reran make, giving the appearance of >> Russell> a possible intermittent hardware (thermal?) problem with my >> Russell> box. This might explain some of the problems I saw earlier, >> Russell> so I can't say with any confidence that the new linux kernel >> Russell> had any influence on this evenings success. >> >> Sam> yep. I am getting these too ("... indicative of gcc bug &c") >> Sam> they go away on a second or third attempt. >> >> New information implies that it might be a linux kernel problem. >> The errors I was seeing were signal 9 (KILL) processes deaths yeah, I seem to recall something like this... >> apparently resulting from memory management issues in the kernel. >> With linux-2.2.4, I am now seeing these problems primarily during >> the configuration stage, and not during the compile. I suspect this >> is what caused my earlier problems with the long long: confusion >> from abnormally terminated autoconf tests. hmmm - can you send a meaningful bug report to Linus? (I cannot!) -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. You think Oedipus had a problem -- Adam was Eve's mother. From sds@eaglets.com Thu Mar 25 12:14:16 1999 Received: from relay7.UU.NET (relay7.UU.NET [192.48.96.17]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id MAA09754 for ; Thu, 25 Mar 1999 12:14:04 -0800 (PST) Received: from mail.eaglets.com by relay7.UU.NET with ESMTP (peer crosschecked as: [208.235.77.228]) id QQgicm22941 for ; Thu, 25 Mar 1999 15:13:29 -0500 (EST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Thu, 25 Mar 1999 15:10:50 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Thu, 25 Mar 1999 15:10:50 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id PAA31326; Thu, 25 Mar 1999 15:10:50 -0500 Sender: "Shteingold, Sam" To: of list Subject: Re: clisp-1999-01-08 install buglet References: <86aex1z8g0.fsf@coulee.tdb.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Russell Senior's message of "Thu, 25 Mar 1999 11:19:06 -0800 (PST)" Date: 25 Mar 1999 15:10:50 -0500 Message-ID: Lines: 40 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <86aex1z8g0.fsf@coulee.tdb.com> >>>> On the subject of "Re: clisp-1999-01-08 install buglet" >>>> Sent on Thu, 25 Mar 1999 11:19:06 -0800 (PST) >>>> Honorable Russell Senior writes: >> >>>>> "Sam" == Sam Steingold writes: >> >> Russell> Linux-2.2.4, gcc-2.8.1, glibc-2.1. >> >> Russell> When configuring with a --prefix=foo option, the specified >> Russell> directory `foo' is not automatically created with `make >> Russell> install'. I don't know precisely how this is dealt with in >> Russell> other packages. >> >> Sam> first, you shouldn't do something like this. >> >> Why not? I find it quite useful. how useful can is be - creating a top level directory just for one package? if you do that, why install at all? clisp can run in-place just fine. >> Sam> second, it's not hard to fix it: just call mkdir -p instead of >> Sam> just mkdir. >> >> Sam> third, I am not sure that all mkdir implementations support -p, >> Sam> so I will wait for Bruno to come back from his vacation before >> Sam> committing this. >> >> Automake seems to provide a script called mkinstalldirs for >> automake'd/autoconf'd packages, which I expect resolves many >> portability issues. oh come on - you cannot expect *everyone* to install automake, can you? or do you want mkinstalldirs to be distributed with clisp? -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. When we write programs that "learn", it turns out we do and they don't. From russell@coulee.tdb.com Thu Mar 25 12:35:26 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id MAA10197 for ; Thu, 25 Mar 1999 12:35:15 -0800 (PST) Received: (qmail 2758 invoked from network); 25 Mar 1999 20:36:09 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 25 Mar 1999 20:36:09 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Thu, 25 Mar 1999 12:36:08 -0800 (PST) To: clisp-list@seagull.cons.org Subject: Re: Problems building clisp-1999-01-08 on linux-2.2.1, glibc-2.1 References: From: Russell Senior Date: 25 Mar 1999 12:36:08 -0800 In-Reply-To: Sam Steingold's message of "Thu, 25 Mar 1999 11:49:50 -0800 (PST)" Message-ID: <86r9qdxq6f.fsf@coulee.tdb.com> Lines: 15 X-Mailer: Gnus v5.6.39/Emacs 20.3 >>>>> "Sam" == Sam Steingold writes: Sam> hmmm - can you send a meaningful bug report to Linus? I have been testing a few things for him and supplying my results. I haven't heard anything definitive back yet. Curiously, I am only seeing the `Out of memory for ' (where is usually cc1, but sometimes sed, sh, etc) syslog messages during a CLISP configure. I configured/compiled gcc and the kernel several times yesterday and didn't see any there. Apparently, the CLISP configure is doing some magical spock-nerve-pinch on the kernel. -- Russell Senior seniorr@teleport.com From russell@coulee.tdb.com Thu Mar 25 13:02:54 1999 Received: from bonneville.tdb.com (bonneville.tdb.com [198.107.132.10] (may be forged)) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id NAA10820 for ; Thu, 25 Mar 1999 13:02:53 -0800 (PST) Received: (qmail 2914 invoked from network); 25 Mar 1999 21:02:40 -0000 Received: from coulee.tdb.com (root@198.107.133.11) by mail.tdb.com with SMTP; 25 Mar 1999 21:02:40 -0000 Received: by coulee.tdb.com via sendmail from stdin id (Debian Smail3.2.0.101) for russell@tdb.com; Thu, 25 Mar 1999 13:02:40 -0800 (PST) To: clisp-list@seagull.cons.org Subject: Re: clisp-1999-01-08 install buglet References: From: Russell Senior Date: 25 Mar 1999 13:02:39 -0800 In-Reply-To: Sam Steingold's message of "Thu, 25 Mar 1999 12:21:02 -0800 (PST)" Message-ID: <86lnglxoy8.fsf@coulee.tdb.com> Lines: 31 X-Mailer: Gnus v5.6.39/Emacs 20.3 >>>>> "Sam" == Sam Steingold writes: Sam> first, you shouldn't do something like this. Russell> Why not? I find it quite useful. Sam> how useful can is be - creating a top level directory just for Sam> one package? if you do that, why install at all? clisp can run Sam> in-place just fine. It is useful for installing multiple versions, and for later uninstalling (rm -rf foo). In contrast, running in place retains the original source and build by-products, which otherwise could be easily removed. Russell> Automake seems to provide a script called mkinstalldirs for Russell> automake'd/autoconf'd packages, which I expect resolves many Russell> portability issues. Sam> oh come on - you cannot expect *everyone* to install automake, Sam> can you? or do you want mkinstalldirs to be distributed with Sam> clisp? Automake (I think it is) provides a `make dist' target which builds a distribution tarball. The script is included in the tarball. In CLISP's case, the mkinstalldirs script could perhaps be distributed with the package and the Makefiles adapted to use it. -- Russell Senior seniorr@teleport.com From sparrow@sparrow.diamant.ru Fri Mar 26 05:52:51 1999 Received: from sparrow.diamant.ru (qmailr@sparrow.diamant.ru [194.186.43.21]) by seagull.cdrom.com (8.8.8/8.6.6) with SMTP id FAA21217 for ; Fri, 26 Mar 1999 05:52:05 -0800 (PST) Received: (qmail 6471 invoked by uid 50002); 26 Mar 1999 13:55:42 -0000 To: clisp-list@seagull.cons.org Subject: Re: communication via unix pipes References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii From: Alexander Vorobiev Date: 26 Mar 1999 16:55:41 +0300 In-Reply-To: Sam Steingold's message of "Thu, 25 Mar 1999 11:38:34 -0800 (PST)" Message-ID: Lines: 40 User-Agent: Gnus/5.07008 (Pterodactyl Gnus v0.80) XEmacs/20.4 (Emerald) Hi! Sam Steingold writes: > First of all, I don't think I can help you with the particular problem > you are having. I have no idea what can go wrong. Nevertheless, I'd btw acl5/linux trial edition has no problems with opening /dev/fd/* files - the resulting stream is an instance of a subclass of terminal-stream. Does clisp have something like that? > think that the method you are using is not optimal. Generally speaking, > file descriptors are so low-level things that I would be reluctant to > use them for communication between programs. Unfortunately I cannot change parent program - I just have the communication protocol specs I have to obey > Also, by using /dev/fd/, you are limiting portability of your program to > modern unixes (basically, to Linux and Solaris :-). It's ok ;) Wouldn't it be a good idea to have some low-level tools to facilitate systems programming? I think that unlike some scheme implementations (guile), this tools should be integrated into high-level lisp concepts (class/conditions hierarchies etc). For example in acl it is possible to specify unix file descriptors as arguments to 'open' > why use 2 pipes when you can use one socket? > you can even make the two program communicate via TCP/IP (and have them > running on different computers). If I could choose means of communication I would definitely use corba (ilu) - it's amazing! pity that clisp does not have corba bindings yet. Alexander -- It's not just a language, it's an adventure. Common Lisp. From dtillman@palantir.cannonexpress.com Fri Mar 26 10:19:32 1999 Received: from palantir.cannonexpress.com (dtillman@palantir.cannonexpress.com [208.136.255.65]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id KAA24148 for ; Fri, 26 Mar 1999 10:19:10 -0800 (PST) Received: (from dtillman@localhost) by palantir.cannonexpress.com (8.9.2/8.9.2) id MAA11921; Fri, 26 Mar 1999 12:20:47 -0600 (CST) Date: Fri, 26 Mar 1999 12:20:47 -0600 (CST) Message-Id: <199903261820.MAA11921@palantir.cannonexpress.com> X-Authentication-Warning: palantir.cannonexpress.com: dtillman set sender to dtillman@palantir.cannonexpress.com using -f From: To: clisp-list@seagull.cons.org Subject: Require and CLOS Hello all, I am playing with CLISP on SCO and attempting to use CLOS. When I first start CLISP, CLOS is not loaded so I have a (require 'clos) at the top of my code. Is this not the proper way to do this? I get: ;; Loading file ele.lsp ... ;; Loading file /usr/local/lib/lisp/src/clos.lsp ... ** - Continuable Error Redefining the COMMON LISP macro WITH-SLOTS If you continue (by typing 'continue'): The old definition will be lost 1. Break> Thank you for your thoughts on this. -Dave From sds@eaglets.com Fri Mar 26 11:15:34 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA24936 for ; Fri, 26 Mar 1999 11:14:48 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Fri, 26 Mar 1999 14:13:45 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Fri, 26 Mar 1999 14:13:44 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id OAA02936; Fri, 26 Mar 1999 14:15:06 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: Require and CLOS References: <199903261820.MAA11921@palantir.cannonexpress.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: 's message of "Fri, 26 Mar 1999 10:24:20 -0800 (PST)" Date: 26 Mar 1999 14:15:06 -0500 Message-ID: Lines: 42 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <199903261820.MAA11921@palantir.cannonexpress.com> >>>> On the subject of "Require and CLOS" >>>> Sent on Fri, 26 Mar 1999 10:24:20 -0800 (PST) >>>> Honorable writes: >> Hello all, I am playing with CLISP on SCO and attempting >> to use CLOS. When I first start CLISP, CLOS is not loaded >> so I have a (require 'clos) at the top of my code. Is this >> not the proper way to do this? I get: >> >> >> >> ;; Loading file ele.lsp ... >> ;; Loading file /usr/local/lib/lisp/src/clos.lsp ... >> ** - Continuable Error >> Redefining the COMMON LISP macro WITH-SLOTS >> If you continue (by typing 'continue'): The old definition will be lost >> 1. Break> where did you get clisp? what is the version (clisp --version)? the current version of clisp has clos in it by default: $ clisp -q [1]> (defclass z () (a b c)) # [2]> (describe (make-instance 'z)) Description of # This is an instance of the CLOS class #. Slots: A unbound B unbound C unbound -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. An elephant is a mouse with an operating system. From dtillman@palantir.cannonexpress.com Fri Mar 26 11:47:33 1999 Received: from palantir.cannonexpress.com (dtillman@palantir.cannonexpress.com [208.136.255.65]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA25446 for ; Fri, 26 Mar 1999 11:47:22 -0800 (PST) Received: (from dtillman@localhost) by palantir.cannonexpress.com (8.9.2/8.9.2) id NAA13400; Fri, 26 Mar 1999 13:48:54 -0600 (CST) Date: Fri, 26 Mar 1999 13:48:54 -0600 (CST) Message-Id: <199903261948.NAA13400@palantir.cannonexpress.com> X-Authentication-Warning: palantir.cannonexpress.com: dtillman set sender to dtillman@palantir.cannonexpress.com using -f From: To: Sam Steingold CC: clisp-list@clisp.cons.org In-reply-to: (message from Sam Steingold on Fri, 26 Mar 1999 11:20:30 -0800 (PST)) Subject: Re: Require and CLOS References: >>>> Sam Steingold writes: > where did you get clisp? Hmm. Good question. Is has been some months since I picked up this version (for SCO) and I don't recall where from. > what is the version (clisp --version)? CLISP 1995-04-04 (April 1995) A bit old... > the current version of clisp has clos in it by default: > > $ clisp -q > > [1]> (defclass z () (a b c)) > # And I get: $ clisp -q > (defclass z () (a b c)) *** - EVAL: undefined function DEFCLASS 1. Break> *features* (CLOS LOOP COMPILER CLISP CLTL1 COMMON-LISP INTERPRETER CLISP3 LOGICAL-PATHNAMES UNIX) 1. Break> *modules* NIL Perhaps this package is badly damaged and I should dump it. Anyone have the current version of CLISP compiled for SCO. -Dave From sds@eaglets.com Fri Mar 26 11:54:53 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA25572 for ; Fri, 26 Mar 1999 11:54:14 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Fri, 26 Mar 1999 14:53:06 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Fri, 26 Mar 1999 14:53:06 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id OAA03073; Fri, 26 Mar 1999 14:54:27 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: communication via unix pipes References: Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Alexander Vorobiev's message of "Fri, 26 Mar 1999 05:56:54 -0800 (PST)" Date: 26 Mar 1999 14:54:27 -0500 Message-ID: Lines: 25 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message >>>> On the subject of "Re: communication via unix pipes" >>>> Sent on Fri, 26 Mar 1999 05:56:54 -0800 (PST) >>>> Honorable Alexander Vorobiev writes: >> >> For example in acl it is possible to specify unix file descriptors as >> arguments to 'open' I can do that - but I am afraid that Bruno will (rightfully!) kill me for that when he comes back from his vacation. >> > why use 2 pipes when you can use one socket? >> > you can even make the two program communicate via TCP/IP (and have them >> > running on different computers). >> >> If I could choose means of communication I would definitely use corba (ilu) >> - it's amazing! pity that clisp does not have corba bindings yet. you are welcome to write it. -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Don't hit a man when he's down -- kick him; it's easier. From sds@eaglets.com Fri Mar 26 11:57:36 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA25650 for ; Fri, 26 Mar 1999 11:57:30 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Fri, 26 Mar 1999 14:56:33 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Fri, 26 Mar 1999 14:56:32 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id OAA03082; Fri, 26 Mar 1999 14:57:54 -0500 Sender: "Shteingold, Sam" To: clisp-list@clisp.cons.org Subject: Re: Require and CLOS References: <199903261948.NAA13400@palantir.cannonexpress.com> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: 's message of "Fri, 26 Mar 1999 13:48:54 -0600 (CST)" Date: 26 Mar 1999 14:57:53 -0500 Message-ID: Lines: 19 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <199903261948.NAA13400@palantir.cannonexpress.com> >>>> On the subject of "Re: Require and CLOS" >>>> Sent on Fri, 26 Mar 1999 13:48:54 -0600 (CST) >>>> Honorable writes: >> >>>> Sam Steingold writes: >> > what is the version (clisp --version)? >> >> CLISP 1995-04-04 (April 1995) almost 4 years old. please look at ftp://clisp.cons.org/pub/lisp/clisp -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Our business is run on trust. We trust you will pay in advance. From rmz@dunk.follo.net Sat Mar 27 08:58:37 1999 Received: from ns1.yes.no (ns1.yes.no [195.204.136.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA08567 for ; Sat, 27 Mar 1999 08:58:11 -0800 (PST) Received: from dunk.follo.net (dunk.follo.net [195.204.143.225]) by ns1.yes.no (8.9.1a/8.9.1) with ESMTP id RAA08318 for ; Sat, 27 Mar 1999 17:59:01 +0100 (CET) Received: (from rmz@localhost) by dunk.follo.net (8.8.8/8.8.6) id RAA04685 for clisp-list@seagull.cons.org; Sat, 27 Mar 1999 17:59:01 +0100 (CET) Date: Sat, 27 Mar 1999 17:59:00 +0100 From: Bjorn Remseth To: clisp-list@seagull.cons.org Subject: Re: Require and CLOS Message-ID: <19990327175900.A4618@dunk.follo.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.3i In-Reply-To: ; from Sam Steingold on Fri, Mar 26, 1999 at 12:20:18PM -0800 Organization: Yes Interactive AS On Fri, Mar 26, 1999 at 12:20:18PM -0800, Sam Steingold wrote: > please look at ftp://clisp.cons.org/pub/lisp/clisp The most current file is always named ".../source/clispsrc.tar.gz". This is a really bad idea, since it makes it difficult to refer to a specific uniquely to a specific snapshot. When making a FreeBSD port (or any other installation system system using a specific release, fetched from an URL), it may break whenever a new release is made. Not good. If instead the release snapshot was given an unique name, for instance "clispsrc-1999-09-22.tar.gz", this problem should disappear. The "clispsrc.tar.gz" target may still exist, but should be updated to point to the most current release. Bruno, is this something which is simple to do? I would start using it almost immediately in our internal FreeBSD port (which then may be published in the FreeBSD ports collection, something we can't do now since the port is using our our internal (but uniquely named :) snapshot copy of clisp as an authorative source.) (Rmz) From amoroso@mclink.it Sat Mar 27 11:45:20 1999 Received: from mail1.mclink.it (net128-007.mclink.it [195.110.128.7]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA10296 for ; Sat, 27 Mar 1999 11:45:03 -0800 (PST) Received: from net145-008.mclink.it (net145-008.mclink.it [195.110.145.8]) by mail1.mclink.it (8.9.1/8.9.0) with SMTP id UAA05482 for ; Sat, 27 Mar 1999 20:45:02 +0100 (CET) From: amoroso@mclink.it (Paolo Amoroso) To: clisp-list@clisp.cons.org Subject: Bug in one of the demos of NCLX Date: Sat, 27 Mar 1999 19:44:49 GMT Organization: Paolo Amoroso - Milan, ITALY Message-ID: <36fd315d.38577@mail.mclink.it> X-Mailer: Forte Agent 1.5/32.451 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I've found a minor bug in one of the demos included with the NCLX package bundled with the RPM source distribution of CLISP 1999.01.08 for Linux. The bug is in file clisp-1999.01.08/modules/clx/new-clx/demos/qix.lsp and it's caused by a typo in the last form: (format T "~& CLX-DEMOS:QIX :host :width :height :delay~ ~% The famous swirling vectors.") ~% Call (clx-demos:qix) or (clx-demos:qix :delay 0).") The fix is obvious: Just remove the ") characters from the next to last line. Paolo P.S. There are problems also with sokoban.lsp. -- Paolo Amoroso From amoroso@mclink.it Sun Mar 28 06:59:50 1999 Received: from mail1.mclink.it (net128-007.mclink.it [195.110.128.7]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id GAA18670 for ; Sun, 28 Mar 1999 06:59:17 -0800 (PST) Received: from net145-035.mclink.it (net145-035.mclink.it [195.110.145.35]) by mail1.mclink.it (8.9.1/8.9.0) with SMTP id QAA11057 for ; Sun, 28 Mar 1999 16:59:42 +0200 (CEST) From: amoroso@mclink.it (Paolo Amoroso) To: clisp-list@clisp.cons.org Subject: -a option for ANSI CL compliance Date: Sun, 28 Mar 1999 14:59:32 GMT Organization: Paolo Amoroso - Milan, ITALY Message-ID: <36fdfe23.39578@mail.mclink.it> X-Mailer: Forte Agent 1.5/32.451 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Concerning the -a option for improving ANSI CL compliance, the CLISP manual states: -a ANSI CL compliant: Comply with the ANSI CL specifi- cation even on those issues where ANSI CL is bro- ken. This option is provided for maximum portabil- ity of Lisp programs. It is not useful for actual everyday work. What are those issues where ANSI CL is broken? Paolo -- Paolo Amoroso From unk6@rz.uni-karlsruhe.de Sun Mar 28 08:14:54 1999 Received: from mailgate.rz.uni-karlsruhe.de (nz40.rz.uni-karlsruhe.de [129.13.197.4]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA19408 for ; Sun, 28 Mar 1999 08:14:29 -0800 (PST) Received: from rz114s0.uni-karlsruhe.de (gilbert@isdn216-171.rz.uni-karlsruhe.de [129.13.216.171]) by mailgate.rz.uni-karlsruhe.de with smtp (Exim 2.04 #3) id 10RIDO-0000Tk-00; Sun, 28 Mar 1999 18:15:11 +0200 To: clisp-list@seagull.cons.org Subject: Re: Bug in one of the demos of NCLX References: <36fd315d.38577@mail.mclink.it> From: Gilbert Baumann Date: 28 Mar 1999 18:10:39 +0200 In-Reply-To: amoroso@mclink.it's message of Sat, 27 Mar 1999 11:45:30 -0800 (PST) Lines: 21 X-Mailer: Gnus v5.2.25/XEmacs 19.14 Message-Id: amoroso@mclink.it (Paolo Amoroso) writes: > (format T "~& CLX-DEMOS:QIX :host :width :height :delay~ > ~% The famous swirling vectors.") > ~% Call (clx-demos:qix) or (clx-demos:qix :delay 0).") > > The fix is obvious: Just remove the ") characters from the next to last > line. Uff?! How come, that it took so long, that this was noticed? It seems like this quirk is in since the very early days of new-clx. Thanks for reporting that. > P.S. > There are problems also with sokoban.lsp. Answered by private email. Gilbert. -- ;;; You know you have hacked Lisp too much, when you m-c-x in a C buffer. From sds@eaglets.com Sun Mar 28 10:10:19 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id KAA20770 for ; Sun, 28 Mar 1999 10:08:38 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Sun, 28 Mar 1999 13:08:15 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Sun, 28 Mar 1999 13:08:15 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id NAA12513; Sun, 28 Mar 1999 13:09:18 -0500 Sender: "Shteingold, Sam" To: recipients of list Subject: Re: -a option for ANSI CL compliance References: <36fdfe23.39578@mail.mclink.it> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: amoroso@mclink.it's message of "Sun, 28 Mar 1999 07:06:28 -0800 (PST)" Date: 28 Mar 1999 13:09:18 -0500 Message-ID: Lines: 24 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <36fdfe23.39578@mail.mclink.it> >>>> On the subject of "-a option for ANSI CL compliance" >>>> Sent on Sun, 28 Mar 1999 07:06:28 -0800 (PST) >>>> Honorable amoroso@mclink.it (Paolo Amoroso) writes: >> Concerning the -a option for improving ANSI CL compliance, the CLISP manual >> states: >> >> -a ANSI CL compliant: Comply with the ANSI CL specifi- >> cation even on those issues where ANSI CL is bro- >> ken. This option is provided for maximum portabil- >> ity of Lisp programs. It is not useful for actual >> everyday work. >> >> What are those issues where ANSI CL is broken? please see "impnotes.html#ansi" (http://www.goems.com/~sds/data/impnotes.html#ansi or ftp://cellar.goems.com/pub/clisp/impnotes.html#ansi) -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. The world will end in 5 minutes. Please log out. From mjkoskin@sci.fi Mon Mar 29 05:27:01 1999 Received: from ds9.sci.fi (ds9.sci.fi [195.74.0.54]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id FAA01201 for ; Mon, 29 Mar 1999 05:26:25 -0800 (PST) Received: from sci.fi (MCDX.dyn.saunalahti.fi [195.197.4.210]) by ds9.sci.fi (8.9.1/8.9.1) with ESMTP id QAA08845 for ; Mon, 29 Mar 1999 16:27:36 +0300 (EET DST) Sender: root@ds9.sci.fi Message-ID: <36FF7F4A.200D0BB1@sci.fi> Date: Mon, 29 Mar 1999 16:25:30 +0300 From: Matti Koskinen X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.35 i586) MIME-Version: 1.0 To: Multiple recipients of list Subject: Newest clisp broken? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit hello all i've tried to build common music -package with the latest clisp (this year's version) but compilation terminates due to errors. I built clisp from sources using linux and rh5.1. then on my cds of red hat there's an old version (May 1997) of clisp and with that build goes ok. common music can be found from ftp://ccrma-ftp.stanford.edu/pub/Lisp/cm/sources/cm.tar.gz if someone wants to look this. in my system compilation aborts in midi.lisp. thanks -matti mjkoskin@sci.fi From sds@eaglets.com Mon Mar 29 07:34:30 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id HAA02573 for ; Mon, 29 Mar 1999 07:34:13 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Mon, 29 Mar 1999 10:33:25 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Mon, 29 Mar 1999 10:33:24 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id KAA22814; Mon, 29 Mar 1999 10:34:19 -0500 Sender: "Shteingold, Sam" To: clisp-list@seagull.cons.org Subject: Re: Newest clisp broken? References: <36FF7F4A.200D0BB1@sci.fi> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Matti Koskinen's message of "Mon, 29 Mar 1999 05:32:07 -0800 (PST)" Date: 29 Mar 1999 10:34:19 -0500 Message-ID: Lines: 24 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <36FF7F4A.200D0BB1@sci.fi> >>>> On the subject of "Newest clisp broken?" >>>> Sent on Mon, 29 Mar 1999 05:32:07 -0800 (PST) >>>> Honorable Matti Koskinen writes: >> hello all >> >> i've tried to build common music -package with the >> latest clisp (this year's version) but compilation >> terminates due to errors. I built clisp from sources >> using linux and rh5.1. then on my cds of red hat >> there's an old version (May 1997) of clisp and with >> that build goes ok. >> common music can be found from >> ftp://ccrma-ftp.stanford.edu/pub/Lisp/cm/sources/cm.tar.gz if someone >> wants to look this. in my system compilation aborts >> in midi.lisp. could you please post the error message? -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. main(a){a="main(a){a=%c%s%c;printf(a,34,a,34);}";printf(a,34,a,34);} From mjkoskin@sci.fi Mon Mar 29 10:57:13 1999 Received: from ds9.sci.fi (ds9.sci.fi [195.74.0.54]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id KAA04784 for ; Mon, 29 Mar 1999 10:56:07 -0800 (PST) Received: from sci.fi (MCCCLXXXVII.dyn.saunalahti.fi [195.197.4.187]) by ds9.sci.fi (8.9.1/8.9.1) with ESMTP id VAA23109 for ; Mon, 29 Mar 1999 21:56:16 +0300 (EET DST) Sender: root@ds9.sci.fi Message-ID: <36FFCC4B.42E5ECD5@sci.fi> Date: Mon, 29 Mar 1999 21:54:03 +0300 From: Matti Koskinen X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.35 i586) MIME-Version: 1.0 To: clisp-list@seagull.cons.org Subject: Re: Newest clisp broken? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sam Steingold wrote: > > could you please post the error message? > > -- > Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux > Micros**t is not the answer. Micros**t is a question, and the answer is Linux, > (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. > main(a){a="main(a){a=%c%s%c;printf(a,34,a,34);}";printf(a,34,a,34);} this is the tail of dribble: The following special variables were not defined: CHANNEL-MESSAGE-CHANNEL MESSAGE CHANNEL-MESSAGE-OPCODE CHANNEL-MESSAGE-DATA1 CHANNEL-MESSAGE-DATA2 NOTE-OFF-CHANNEL NOTE-OFF-KEY NOTE-OFF-VELOCITY NOTE-ON-CHANNEL NOTE-ON-KEY NOTE-ON-VELOCITY KEY-PRESSURE-CHANNEL KEY-PRESSURE-KEY KEY-PRESSURE-PRESSURE CONTROL-CHANGE-CHANNEL CONTROL-CHANGE-CONTROLLER CONTROL-CHANGE-VALUE PROGRAM-CHANGE-CHANNEL PROGRAM-CHANGE-PROGRAM CHANNEL-PRESSURE-CHANNEL CHANNEL-PRESSURE-PRESSURE PITCH-BEND-CHANNEL PITCH-BEND-LSB PITCH-BEND-MSB SYSTEM-MESSAGE-ROUTE SYSTEM-MESSAGE-STATUS SYSTEM-MESSAGE-DATA1 SYSTEM-MESSAGE-DATA2 SYSEX-ROUTE MTC-QUARTER-FRAME-ROUTE SONG-POSITION-ROUTE SONG-POSITION-LSB SONG-POSITION-MSB SONG-SELECT-ROUTE SONG-SELECT-SONG CABLE-SELECT-ROUTE CABLE-SELECT-CABLE TUNE-REQUEST-ROUTE EOX-ROUTE TIMING-CLOCK-ROUTE TIMING-TICK-ROUTE START-ROUTE CONTINUE-ROUTE STOP-ROUTE ACTIVE-SENSING-ROUTE SYSTEM-RESET-ROUTE META-MESSAGE-TYPE ;; Loading file /root/cm-1.3/midi/ml.fas ... *** - symbol CHANNEL-MESSAGE-CHANNEL has no value this is part of ml.lisp where these are defined: (eval-when #+cltl2 (:load-toplevel :execute :compile-toplevel) #-cltl2 (load eval compile) (proclaim '(inline ;; channel messages make-channel-message channel-message-channel channel-message-opcode channel-message-data1 channel-message-data2 note-off-channel note-off-key note-off-velocity note-on-channel note-on-key note-on-velocity key-pressure-channel key-pressure-key key-pressure-pressure control-change-channel control-change-controller control-change-value program-change-channel program-change-program channel-pressure-channel channel-pressure-pressure pitch-bend-channel pitch-bend-lsb pitch-bend-msb ... etc. (defun channel-message-channel (message) (ldb mi_d:+enc-logical-channel-byte+ message)) -matti mjkoskin@sci.fi From amoroso@mclink.it Mon Mar 29 11:47:24 1999 Received: from mail1.mclink.it (net128-007.mclink.it [195.110.128.7]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id LAA05672 for ; Mon, 29 Mar 1999 11:46:11 -0800 (PST) Received: from net145-106.mclink.it (net145-106.mclink.it [195.110.145.106]) by mail1.mclink.it (8.9.1/8.9.0) with SMTP id VAA09775 for ; Mon, 29 Mar 1999 21:45:55 +0200 (CEST) From: amoroso@mclink.it (Paolo Amoroso) To: clisp-list@seagull.cons.org Subject: Re: Bug in one of the demos of NCLX Date: Mon, 29 Mar 1999 19:45:31 GMT Organization: Paolo Amoroso - Milan, ITALY Message-ID: <3701d4a8.1222318@mail.mclink.it> References: <36fd315d.38577@mail.mclink.it> In-Reply-To: <36fd315d.38577@mail.mclink.it> X-Mailer: Forte Agent 1.5/32.451 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit On Sat, 27 Mar 1999 11:53:35 -0800 (PST), you wrote: > There are problems also with sokoban.lsp. As suggested by Gilbert Baumann, the author of NCLX, it's probably a color map problem. While Sokoban does not run on my system with 256 colors (please note that I use AfterStep 1.5, a notoriously color-hungry window manager), it works just fine with 65K colors. Paolo -- Paolo Amoroso From tkunze@ccrma.stanford.edu Mon Mar 29 14:41:00 1999 Received: from ccrma.stanford.edu (ccrma.Stanford.EDU [36.49.0.84]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id OAA07792 for ; Mon, 29 Mar 1999 14:40:21 -0800 (PST) Received: from ccrma.stanford.edu ([207.97.74.46]) by ccrma.stanford.edu (8.8.8/8.8.8) with ESMTP id OAA19036; Mon, 29 Mar 1999 14:41:30 -0800 (PST) Sender: tkunze@ccrma.stanford.edu Message-ID: <37000055.46F259A1@ccrma.stanford.edu> Date: Mon, 29 Mar 1999 17:36:05 -0500 From: Tobias Kunze X-Mailer: Mozilla 4.05 [en] (X11; I; IRIX 6.5 IP22) MIME-Version: 1.0 To: clisp-list@seagull.cons.org CC: Multiple recipients of list Subject: Re: Newest clisp broken? References: <36FFCC4B.42E5ECD5@sci.fi> Content-Type: multipart/mixed; boundary="------------8F52D787B5A4B36676C6C2BC" This is a multi-part message in MIME format. --------------8F52D787B5A4B36676C6C2BC Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi Matti- I have no problems in my clisp. I believe your clisp breaks at l. 437 in ml.lisp, when (define-message-setf channel-message-channel ...) is called. That is, for some reason this symbol gets evaluated as a symbol instead of taken as name for define-setf-method: *** - symbol CHANNEL-MESSAGE-CHANNEL has no value In other words, it seems as if define-message-setf, defined on l. 389, is broken in your clisp. Anyway, in order to try it out on a scaled-down version of cm, could you compile- load the attached file in a vanilla clisp? Here is what it should look like: tkunze@ulysses:~ [86]> clisp -q > (compile-file "~/tmp/matti.lisp") Compiling file /usr/people/tkunze/tmp/matti.lisp ... Compilation of file /usr/people/tkunze/tmp/matti.lisp is finished. 0 errors, 0 warnings #P"/usr/people/tkunze/tmp/matti.fas" ; NIL ; NIL > (load *) ;; Loading file /usr/people/tkunze/tmp/matti.fas ... "-- proclaiming --" "-- defining define-message-setf --" "-- defining message-setf --" ;; Loading of file /usr/people/tkunze/tmp/matti.fas is finished. T > (lisp-implementation-version) "1998-09-09 (September 1998)" > -Tobias ______________________________________________________________________ Tobias Kunze tkunze@ccrma.stanford.edu CCRMA, Stanford University http://www-ccrma.stanford.edu/~tkunze --------------8F52D787B5A4B36676C6C2BC Content-Type: text/plain; charset=us-ascii; name="matti.lisp" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="matti.lisp" ;;; ;;; File: matti.lisp ;;; ;;; Summary: test matti's cm/clisp problem ;;; ;;; Author: Tobias Kunze ;;; e-Mail: tkunze@ccrma.stanford.edu ;;; Org: CCRMA, Stanford University ;;; ;;; orig-Date: 29-Mar-99 at 17:16:52 ;;; Last-Mod: 29-Mar-99 at 17:29:22 by Tobias Kunze ;;; ;;; Revision: ;;; ;;; Description: ;;; ;;; Changes: ;;; dd-Mmm-yy tk (in-package :user) ;;; ;;; (eval-when #+cltl2 (:load-toplevel) #-cltl2 (load) (print "-- proclaiming --")) (eval-when #+cltl2 (:load-toplevel :execute :compile-toplevel) #-cltl2 (load eval compile) (proclaim '(inline channel-message-channel))) ;;; ;;; (eval-when #+cltl2 (:load-toplevel) #-cltl2 (load) (print "-- defining define-message-setf --")) (eval-when (compile load eval) (defmacro define-message-setf (accessor bytespec) `(#-(and cltl2 excl) define-setf-method #+(and cltl2 excl) define-setf-expander ,accessor (message) (multiple-value-bind (temps vals stores store-form access-form) (#-(and cltl2 excl) get-setf-method #+(and cltl2 excl) get-setf-expansion message) (let ((val (gensym)) (msg (first stores))) (values temps vals (list val) ,`(list 'let* (list (list msg (list 'dpb val ',bytespec access-form))) store-form val) ,`(list 'ldb ',bytespec access-form)))))) ) ;;; ;;; (eval-when #+cltl2 (:load-toplevel) #-cltl2 (load) (print "-- defining message-setf --")) (define-message-setf channel-message-channel (byte 14 18)) ;;; ;;; -*- EOF -*- --------------8F52D787B5A4B36676C6C2BC-- From sds@goems.com Tue Mar 30 06:14:11 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id GAA17335 for ; Tue, 30 Mar 1999 06:13:20 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Tue, 30 Mar 1999 09:13:23 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Tue, 30 Mar 1999 09:13:22 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id JAA29653; Tue, 30 Mar 1999 09:14:04 -0500 Sender: sds@goems.com To: clisp-list@seagull.cons.org Subject: Re: Newest clisp broken? References: <37000055.46F259A1@ccrma.stanford.edu> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Tobias Kunze's message of "Tue, 30 Mar 1999 00:05:31 -0800 (PST)" Date: 30 Mar 1999 09:14:03 -0500 Message-ID: Lines: 33 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <37000055.46F259A1@ccrma.stanford.edu> >>>> On the subject of "Re: Newest clisp broken?" >>>> Sent on Tue, 30 Mar 1999 00:05:31 -0800 (PST) >>>> Honorable Tobias Kunze writes: >> >> I have no problems in my clisp. I believe your clisp breaks >> at l. 437 in ml.lisp, when >> >> (define-message-setf channel-message-channel ...) of course: `define-message-setf' is broken: clisp doesn't have `define-setf-method' as it was replaced with `define-setf-expander' in ANSI CL. Please replace (defmacro define-message-setf (accessor bytespec) `(#-(and cltl2 excl) define-setf-method #+(and cltl2 excl) define-setf-expander ,accessor (message) with (defmacro define-message-setf (accessor bytespec) `(#-ansi-cl define-setf-method #+ansi-cl define-setf-expander ,accessor (message) and calll CLISP with the "-a" switch. -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Type louder, please. From tkunze@ccrma.stanford.edu Tue Mar 30 07:00:51 1999 Received: from ccrma.stanford.edu (ccrma.Stanford.EDU [36.49.0.84]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id HAA17919 for ; Tue, 30 Mar 1999 07:00:48 -0800 (PST) Received: from ccrma.stanford.edu ([207.97.74.43]) by ccrma.stanford.edu (8.8.8/8.8.8) with ESMTP id HAA25898; Tue, 30 Mar 1999 07:02:00 -0800 (PST) Sender: tkunze@ccrma.stanford.edu Message-ID: <3700E641.DC1BA83F@ccrma.stanford.edu> Date: Tue, 30 Mar 1999 09:57:05 -0500 From: Tobias Kunze X-Mailer: Mozilla 4.05 [en] (X11; I; IRIX 6.5 IP22) MIME-Version: 1.0 To: clisp-list@seagull.cons.org CC: Multiple recipients of list , Heinrich Taube Subject: Re: Newest clisp broken? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit or probably better (defmacro define-message-setf (accessor bytespec) `(#-(or ansi-cl cltl2) ansi-cl define-setf-method #+(or ansi-cl cltl2) define-setf-expander ,accessor (message) as it is also in cltl2. Thanks a lot, Sam! -Tobias Sam Steingold wrote: > > >>>> In message <37000055.46F259A1@ccrma.stanford.edu> > >>>> On the subject of "Re: Newest clisp broken?" > >>>> Sent on Tue, 30 Mar 1999 00:05:31 -0800 (PST) > >>>> Honorable Tobias Kunze writes: > >> > >> I have no problems in my clisp. I believe your clisp breaks > >> at l. 437 in ml.lisp, when > >> > >> (define-message-setf channel-message-channel ...) > > of course: `define-message-setf' is broken: clisp doesn't have > `define-setf-method' as it was replaced with `define-setf-expander' in > ANSI CL. Please replace > > (defmacro define-message-setf (accessor bytespec) > `(#-(and cltl2 excl) define-setf-method > #+(and cltl2 excl) define-setf-expander > ,accessor (message) > > with > > (defmacro define-message-setf (accessor bytespec) > `(#-ansi-cl define-setf-method > #+ansi-cl define-setf-expander > ,accessor (message) > > and calll CLISP with the "-a" switch. > > -- > Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux > Micros**t is not the answer. Micros**t is a question, and the answer is Linux, > (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. > Type louder, please. From mjkoskin@sci.fi Tue Mar 30 07:16:55 1999 Received: from ds9.sci.fi (ds9.sci.fi [195.74.0.54]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id HAA18134 for ; Tue, 30 Mar 1999 07:14:14 -0800 (PST) Received: from sci.fi (MCCCXCIX.dyn.saunalahti.fi [195.197.4.199]) by ds9.sci.fi (8.9.1/8.9.1) with ESMTP id SAA27397 for ; Tue, 30 Mar 1999 18:15:33 +0300 (EET DST) Sender: root@ds9.sci.fi Message-ID: <3700EA10.6AE224C4@sci.fi> Date: Tue, 30 Mar 1999 18:13:20 +0300 From: Matti Koskinen X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.35 i586) MIME-Version: 1.0 To: clisp-list@seagull.cons.org Subject: Re: Newest clisp broken? References: <37000055.46F259A1@ccrma.stanford.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Tobias Kunze wrote: > > > tkunze@ulysses:~ [86]> clisp -q > > > (compile-file "~/tmp/matti.lisp") > > Compiling file /usr/people/tkunze/tmp/matti.lisp ... > > Compilation of file /usr/people/tkunze/tmp/matti.lisp is finished. > 0 errors, 0 warnings > #P"/usr/people/tkunze/tmp/matti.fas" ; > NIL ; > NIL > > (load *) > ;; Loading file /usr/people/tkunze/tmp/matti.fas ... > "-- proclaiming --" > "-- defining define-message-setf --" > "-- defining message-setf --" > ;; Loading of file /usr/people/tkunze/tmp/matti.fas is finished. > T > > (lisp-implementation-version) > "1998-09-09 (September 1998)" > > > > -Tobias i get exactly the same output [2]> (compile-file "matti.lisp") Compiling file /home/mjkoskin/midi-comp/matti.lisp ... Compilation of file /home/mjkoskin/midi-comp/matti.lisp is finished. 0 errors, 0 warnings #P"/home/mjkoskin/midi-comp/matti.fas" ; NIL ; NIL [3]> (load *) ;; Loading file /home/mjkoskin/midi-comp/matti.fas ... "-- proclaiming --" "-- defining define-message-setf --" "-- defining message-setf --" ;; Loading of file /home/mjkoskin/midi-comp/matti.fas is finished. T [4]> (lisp-implementation-version) "1999-01-08 (January 1999)" [5]> -matti mjkoskin@sci.fi From tkunze@ccrma.stanford.edu Tue Mar 30 08:08:23 1999 Received: from ccrma.stanford.edu (ccrma.Stanford.EDU [36.49.0.84]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA18943 for ; Tue, 30 Mar 1999 08:08:18 -0800 (PST) Received: from ccrma.stanford.edu ([207.97.74.38]) by ccrma.stanford.edu (8.8.8/8.8.8) with ESMTP id IAA26448 for ; Tue, 30 Mar 1999 08:09:42 -0800 (PST) Sender: tkunze@ccrma.stanford.edu Message-ID: <3700F620.8C547D34@ccrma.stanford.edu> Date: Tue, 30 Mar 1999 11:04:48 -0500 From: Tobias Kunze X-Mailer: Mozilla 4.05 [en] (X11; I; IRIX 6.5 IP22) MIME-Version: 1.0 To: clisp-list@seagull.cons.org Subject: Re: Newest clisp broken? References: <3700EA10.6AE224C4@sci.fi> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit > i get exactly the same output *that* is weird. Why does it work now if, as Sam said, define-setf-method is undefined in the newest clisp? what does * (describe 'define-setf-expander) * (describe 'define-setf-method) and * *features* return for you? From taube@uiuc.edu Tue Mar 30 08:12:12 1999 Received: from staff1.cso.uiuc.edu (root@staff1.cso.uiuc.edu [128.174.5.59]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id IAA19046 for ; Tue, 30 Mar 1999 08:12:10 -0800 (PST) Received: from [128.174.92.78] (camil28.music.uiuc.edu [128.174.92.78]) by staff1.cso.uiuc.edu (8.8.8/8.8.8) with ESMTP id KAA20714; Tue, 30 Mar 1999 10:13:22 -0600 (CST) X-Sender: taube@ux1.cso.uiuc.edu Message-Id: In-Reply-To: <3700EA10.6AE224C4@sci.fi> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 30 Mar 1999 10:09:41 -0600 To: Matti Koskinen , Tobias Kunze From: Rick Taube Subject: Re: Newest clisp broken? Cc: clisp-list@seagull.cons.org CLISP isnt "broken" its just that the new release is more ansi compliant than the earlier one so the common music sources need to get updated. Sam Shepard was kind enough to send me a trace of the problem definitions and ill fix them asap. From sds@goems.com Tue Mar 30 09:16:58 1999 Received: from mail.eaglets.com ([208.235.77.228]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id JAA20085 ; Tue, 30 Mar 1999 09:16:38 -0800 (PST) Received: by mail.eaglets.com from localhost (router,SLMail V3.1); Tue, 30 Mar 1999 12:16:49 -0500 Received: by mail.eaglets.com from eho.eaglets.com [208.235.77.238] (SLmail 3.1.2948 (Release Build)); Tue, 30 Mar 1999 12:16:49 -0500 Received: (from sds@localhost) by eho.eaglets.com (8.9.3/8.9.3) id MAA06099; Tue, 30 Mar 1999 12:17:34 -0500 Sender: sds@goems.com To: clisp-list@seagull.cons.org Cc: Multiple recipients of list Subject: Re: Newest clisp broken? References: <3700F620.8C547D34@ccrma.stanford.edu> Return-Receipt-To: sds@goems.com Reply-To: sds@goems.com X-Disclaimer: You should not expect anyone to agree with me. X-Attribution: Sam X-No-Archive: Yes Mail-Copies-To: never From: Sam Steingold In-Reply-To: Tobias Kunze's message of "Tue, 30 Mar 1999 08:12:29 -0800 (PST)" Date: 30 Mar 1999 12:17:34 -0500 Message-ID: Lines: 20 X-Mailer: Gnus v5.7/Emacs 20.3 >>>> In message <3700F620.8C547D34@ccrma.stanford.edu> >>>> On the subject of "Re: Newest clisp broken?" >>>> Sent on Tue, 30 Mar 1999 08:12:29 -0800 (PST) >>>> Honorable Tobias Kunze writes: >> > i get exactly the same output >> >> *that* is weird. Why does it work now if, as Sam said, >> define-setf-method is undefined in the newest clisp? I wasn't clear enough. sorry. `define-setf-method' is defined in the package SYSTEM (for backwards compatibility), but is not exported to COMMON-LISP and COMMON-LISP-USER. -- Sam Steingold (http://www.goems.com/~sds) running RedHat5.2 GNU/Linux Micros**t is not the answer. Micros**t is a question, and the answer is Linux, (http://www.linux.org) the choice of the GNU (http://www.gnu.org) generation. Why use Windows, when there are Doors? From tkunze@ccrma.stanford.edu Tue Mar 30 09:49:35 1999 Received: from ccrma.stanford.edu (ccrma.Stanford.EDU [36.49.0.84]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id JAA20441 for ; Tue, 30 Mar 1999 09:48:39 -0800 (PST) Received: from ccrma.stanford.edu ([207.97.74.44]) by ccrma.stanford.edu (8.8.8/8.8.8) with ESMTP id JAA27432 for ; Tue, 30 Mar 1999 09:49:58 -0800 (PST) Sender: tkunze@ccrma.stanford.edu Message-ID: <37010D9B.764DD1A2@ccrma.stanford.edu> Date: Tue, 30 Mar 1999 12:44:59 -0500 From: Tobias Kunze X-Mailer: Mozilla 4.05 [en] (X11; I; IRIX 6.5 IP22) MIME-Version: 1.0 To: clisp-list@seagull.cons.org Subject: Re: Newest clisp broken? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit hm why does my little example work for matti but cm as a whole not? From tkunze@ccrma.stanford.edu Tue Mar 30 09:50:26 1999 Received: from ccrma.stanford.edu (ccrma.Stanford.EDU [36.49.0.84]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id JAA20455 for ; Tue, 30 Mar 1999 09:50:05 -0800 (PST) Received: from ccrma.stanford.edu ([207.97.74.44]) by ccrma.stanford.edu (8.8.8/8.8.8) with ESMTP id JAA27457 for ; Tue, 30 Mar 1999 09:51:31 -0800 (PST) Sender: tkunze@ccrma.stanford.edu Message-ID: <37010DF9.6C2FA27E@ccrma.stanford.edu> Date: Tue, 30 Mar 1999 12:46:33 -0500 From: Tobias Kunze X-Mailer: Mozilla 4.05 [en] (X11; I; IRIX 6.5 IP22) MIME-Version: 1.0 To: clisp-list@seagull.cons.org Subject: sorry References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit sorry--i meant to reply to Rick Taube only but as usual was too fast with my mouse... :/ From haible@ilog.fr Tue Mar 30 09:58:39 1999 Received: from sceaux.ilog.fr (sceaux.ilog.fr [193.55.64.10]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id JAA20575 for ; Tue, 30 Mar 1999 09:58:23 -0800 (PST) Received: from ilog.ilog.fr ([172.17.4.22]) by sceaux.ilog.fr (8.9.1/8.9.1) with ESMTP id TAA04741 for ; Tue, 30 Mar 1999 19:59:15 +0200 (MET DST) Received: from jaures.ilog.fr (jaures.ilog.fr [172.16.1.81]) by ilog.ilog.fr (8.8.8/8.7.3) with ESMTP id TAA14483; Tue, 30 Mar 1999 19:58:13 +0200 (MET DST) From: Bruno Haible Received: (from haible@localhost) by jaures.ilog.fr (8.9.0/8.9.0) id TAA19387; Tue, 30 Mar 1999 19:59:08 +0200 (MET DST) Date: Tue, 30 Mar 1999 19:59:08 +0200 (MET DST) Message-Id: <199903301759.TAA19387@jaures.ilog.fr> To: clisp-list@seagull.cons.org Subject: Re: communication via unix pipes In-Reply-To: References: Alexander Vorobiev writes: > Unfortunately, this didn't work: > > *** - file #P"/proc/5395/fd/pipe:[1339598]" does not exist That's because these symlinks in /proc/self/fd/ are really faked. They point into void. And unfortunately clisp tries to follow the link, in order to get the absolute pathname of the "file", for the benefit of the function `truename'. > Is it possible to make clisp deal with such "special" files properly? Instead of special casing them, clisp should be changed to treat symlinks pointing into void like regular files, and let the OS give an error if needed. > Or maybe it is possible to avoid using /dev/fd/* files and to connect streams > directly to open file descriptors? Trying to open /dev/fd/* files is perfectly ok; file descriptors as numbers are a Unix specific hack. > And the last question about C to lisp communication: should I use some > ffi tools to convert data for transmitting to/from parent process > written in C? `read-integer' and `write-integer' will be your friends (present in recent clisp snapshots). If you want to write portable programs, open the stream with :element-type '(unsigned-byte 8) and use `read-byte' and `write-byte'. > If I could choose means of communication I would definitely use corba (ilu) > - it's amazing! pity that clisp does not have corba bindings yet. Yes, really. The spec for CORBA bindings is already there, thanks to Franz and Harlequin. See ftp://ftp.omg.org/pub/docs/orbos/98-10-16.ps ftp://ftp.omg.org/pub/docs/orbos/98-10-16.pdf You have the choice between ILU and MICO. Bruno ! To unsubscribe from the clisp-list mailing list, send mail to ! ! listproc@clisp.cons.org ! ! including the two words "unsubscribe clisp-list" as message body. ! From mjkoskin@sci.fi Tue Mar 30 16:38:02 1999 Received: from laude.saunalahti.fi (root@laude.saunalahti.fi [195.74.0.61]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id QAA25084 for ; Tue, 30 Mar 1999 16:37:54 -0800 (PST) Received: from pefletti.saunalahti.fi (root@pefletti [195.74.0.53]) by laude.saunalahti.fi (8.9.1/8.9.1) with ESMTP id VAA28156 for ; Tue, 30 Mar 1999 21:54:40 +0300 (EET DST) Received: from sci.fi (MCCCXIV.dyn.saunalahti.fi [195.197.4.114]) by pefletti.saunalahti.fi (8.9.1/8.9.1) with ESMTP id VAA25962 for ; Tue, 30 Mar 1999 21:54:18 +0300 (EET DST) Sender: root@pefletti.saunalahti.fi Message-ID: <37011D58.7DE56848@sci.fi> Date: Tue, 30 Mar 1999 21:52:08 +0300 From: Matti Koskinen X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.35 i586) MIME-Version: 1.0 To: Multiple recipients of list Subject: thanks for solving my problem Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit hello all thousands of thanks to all who helped solving the problem with cm+clisp. defining the macro as Sam and Tobias proposed made cm build ok. (this was the easiest part though, now making really music will be much harder :-) thanks -matti mjkoskin@sci.fi From mjkoskin@sci.fi Wed Mar 31 01:10:52 1999 Received: from ds9.sci.fi (ds9.sci.fi [195.74.0.54]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id BAA00508 for ; Wed, 31 Mar 1999 01:10:51 -0800 (PST) Received: from sci.fi (MCDI.dyn.saunalahti.fi [195.197.4.201]) by ds9.sci.fi (8.9.1/8.9.1) with ESMTP id MAA24111 for ; Wed, 31 Mar 1999 12:11:31 +0300 (EET DST) Sender: root@ds9.sci.fi Message-ID: <3701E6C1.7C259FF2@sci.fi> Date: Wed, 31 Mar 1999 12:11:29 +0300 From: Matti Koskinen X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.35 i586) MIME-Version: 1.0 To: clisp-list@seagull.cons.org Subject: Re: Newest clisp broken? References: <37010D9B.764DD1A2@ccrma.stanford.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Tobias Kunze wrote: > > hm why does my little example work for matti > but cm as a whole not? first thanks to all for solving this problem. i got cm built now. still, there is something weird. it doesn't work. stella starts and stella-commands work, but everything in parentheses are silently ignored, but this may be an issue to discuss on cm-list. -matti mjkoskin@sci.fi From murray@uni-paderborn.de Wed Mar 31 09:06:52 1999 Received: from pbinfo (pbinfo.uni-paderborn.de [131.234.22.46]) by seagull.cdrom.com (8.8.8/8.6.6) with ESMTP id JAA06176 for ; Wed, 31 Mar 1999 09:06:40 -0800 (PST) Received: from pc-kb-7 ([131.234.28.72]:1648 "EHLO uni-paderborn.de" ident: "NO-IDENT-SERVICE") by nfsmail.uni-paderborn.de with ESMTP id <52234-25242>; Wed, 31 Mar 1999 19:07:43 +0200 Message-ID: <370255B2.A3FFDE6A@uni-paderborn.de> Date: Wed, 31 Mar 1999 19:04:50 +0200 From: Oliver Niggemann X-Mailer: Mozilla 4.05 [en] (WinNT; I) MIME-Version: 1.0 To: clisp-list@clisp.cons.org Subject: Problem with input Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello, I use CLISP 1999-01-08 for Windows NT 4.0 (german version). When typing input at the clisp-prompt there seems's to be a restriction concerning the maximum number of chars in the input. I can not type input longer than circa 3 lines. This causes a problem when using CLISP together with XEmacs (native code for NT 4.0, version 21.0 b62) as an inferior-lisp-process. Within XEmacs it is now possible to type longer input, but this causes CLISP to crash. Functions like lisp-eval-region do not work either, if the size of the region exceeds the maximum number of allowed chars. What can I do about this restriction? Thanks for any help, Oliver Niggemann ----------------------------- Oliver Niggemann University of Paderborn Germany