From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 6335 invoked from network); 31 Dec 1996 08:41:14 -0000 Received: from euclid.skiles.gatech.edu (list@130.207.146.50) by coral.primenet.com.au with SMTP; 31 Dec 1996 08:41:14 -0000 Received: (from list@localhost) by euclid.skiles.gatech.edu (8.7.3/8.7.3) id DAA04668; Tue, 31 Dec 1996 03:45:28 -0500 (EST) Resent-Date: Tue, 31 Dec 1996 03:45:28 -0500 (EST) Date: Tue, 31 Dec 1996 11:47:28 +0300 (MSK) From: Andrej Borsenkow X-Sender: bor@itsrm1.mow.sni.de Reply-To: borsenkow.msk@sni.de To: Zoltan Hidvegi cc: Zsh hacking and development Subject: Re: zsh-3.0.3-test1 and zsh-3.1.1-test1 In-Reply-To: <199612310239.DAA07466@hzoli.ppp.cs.elte.hu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"l22fh3.0.q81.dADoo"@euclid> Resent-From: zsh-workers@math.gatech.edu X-Mailing-List: archive/latest/2682 X-Loop: zsh-workers@math.gatech.edu Precedence: list Resent-Sender: zsh-workers-request@math.gatech.edu I experienced some problems with 3.1.1-test1. I am using Sinix 5.42 with native C-DS compiler and GNU make 3.75. I have to use gmake, because I make both static and dynamic version from single source and our make doesn't handle VPATH :-( Here is the story: I. Static version. 1. gmake doesn't make anything for $(MODBINS) (aka modules-bltin) target; that is, zsh just doesn't include any compiled modules. Here is excerpt from running gmake (note also sed error): .... cc -W0 -c -I.. -I. -I../../zsh-3.1.1/Src -IModules -IZle -I../../zsh-3.1.1/Src/Zle -DHAVE_CONFIG_H -O2 -o hist.o ../../zsh-3.1.1/Src/hist.c Creating `bltinmods.list'. sed: Cannot open modules-bltin: No such file or directory sed: Label too long: :1;/\\$/{N;s/\\\n//;b1;};s/ / /g;/^ *$/d;/^[^#:>]*>[^#:>]*$/{s/>.*//;p;} `zshpaths.h' is up to date. Creating `zshxmods.h'. sed: Cannot open modules-bltin: No such file or directory sed: Label too long: :1;/\\$/{N;s/\\\n//;b1;};s/ / /g;/^ *$/d;/^[^#:>]*>[^#:>]*$/{s/>.*//;p;} cc -W0 -c -I.. -I. -I../../zsh-3.1.1/Src -IModules -IZle -I../../zsh-3.1.1/Src/Zle -DHAVE_CONFIG_H -O2 -o init.o ../../zsh-3.1.1/Src/init.c ...... If I go into Src and do "gmake modules-bltin" it tells me "modules-bltin is up to date". Funny enough, our native make works in this case ;) 2. Some problems with texi documentation ... Making info file `zsh.info' from `../../zsh-3.1.1/Doc/zsh.texi'. ../../zsh-3.1.1/Doc/zsh.texi:5295: This `table' doesn't have a matching `@end table'. ../../zsh-3.1.1/Doc/zsh.texi:6835: Unknown command `br'. ../../zsh-3.1.1/Doc/zsh.texi:6835: Misplaced `{'. ../../zsh-3.1.1/Doc/zsh.texi:6835: Misplaced `}'. Info documentation cannot be compiled without makeinfo. You can find precompiled info files in zsh-doc.tar.gz. ... I have texinfo-3.9 (makeinfo 1.67). II. Dynamic version 1. The same problem with mudules-bltin 2. The same problem with makeinfo 3. libzsh is not linked with any library; as I already said, we have static versions of some libraries (notably termcap; libc is also spplitted in static and dynamic parts). It generates run time error for any function, defined in this static libraries. Is there any objection against linking zsh against -lzsh and libzsh against $(LIBS)? If it won't break things somewhere else, it will add one more plattform on which zsh compiles out of the box. 4. For some reasons compctl.so fails to load. I can load all other modules just fine. I don't know, how it was in 3.1.0, because this versions didn't run on my system (couldn't resolve trashzle at runtime). I have no idea, how to debug it - it all happens in dlload(); could it possibly be related to the problem with modules-bltin? greetings ------------------------------------------------------------------------- Andrej Borsenkow Fax: +7 (095) 252 01 05 SNI ITS Moscow Tel: +7 (095) 252 13 88 NERV: borsenkow.msk E-Mail: borsenkow.msk@sni.de -------------------------------------------------------------------------