zsh 4.3.3 released

zsh 4.3.3 released

[Peter Stephenson]

I've uploaded version 4.3.3 of zsh to ftp.zsh.org.  If there aren't
complaints too loud I'll send it to Sourceforge and announce it on
Freshmeat tomorrow.  (This time I remembered to check that all functions
are exported where necessary, so we shouldn't have the traditional round
of mod_export additions.)

Release notes (note I didn't hunt too hard for new features, so please
tell me if there's something interesting I missed):

Version 4.3.3 now contains most of the planned support for multibyte
characters.  There are still some problems notably with Unicode combining
characters and composition of characters (notably on MacOS).  Some
expertise in these areas would be useful.  If support is available the
shell will be compiled with multibyte support and will start in multibyte
mode (option MULTIBYTE set).

Apart from the usual bug fixes and enhancements to completion functions,
there are various other new features:

- Users can write shell functions to be used in mathematical expressions.
  See the description of "functions -M" in the zshbuiltin manual page
  and the description of zmathfuncdef in the zshcontrib manual page.
- The zle widget read-command reads a key sequence without executing it.
- The zle widgets auto-suffix-remove and auto-suffix-retain provide
  user-defined control for suffixes (on the lines of AUTO_REMOVE_SLASH).
- Various arrays are available to add a named function to a set
  called where the specific functions chpwd, periodic, precmd and preexec
  are called.  The arrays are called chpwd_functions, etc.  There is a new
  special function zshexit also with a corresponding hook.  See SPECIAL
  FUNCTIONS in the zshmisc manual.
- ";|" at the end of a case statement causes the shell to continue
  looking at later tests (c.f. ";&" which causes an unconditional fall
  through).
- The sched builtin has been enhanced to run events asynchronously
  when the shell is waiting at an interactive prompt.
- There is a new set of functions to handle calendars and dates.
  See the zshcalsys manual page.
- The zsh/parameter module makes available the parameter functrace
  for function backtraces.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070


To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php

To get further information regarding CSR, please visit our Investor Relations page at http://ir.csr.com/csr/about/overview

Re: zsh 4.3.3 released

[Thorsten Kampe]

* Peter Stephenson (Mon, 16 Apr 2007 17:47:02 +0100)
> I've uploaded version 4.3.3 of zsh to ftp.zsh.org.  If there aren't
> complaints too loud I'll send it to Sourceforge and announce it on
> Freshmeat tomorrow.

I freshly compiled and installed on Cygwin and I get this is 
completion
% cd /[TAB]
[1;32m}directory[m}
bin/   cygdrive/   etc/   home/   lib/   proc/   sbin/   tmp/   usr/   
var/

I have this in my .zshenv
G=$'\e[1;32m'

and in my .zshrc:
zstyle ':completion:*' list-prompt "%{$G%}last line %l (%p)%{$r%}  %
{$W%}last match %m%{$r%}"
zstyle ':completion:*:descriptions' format "%{$G%}%d%{$r%}"
zstyle ':completion:*:messages'     format "%{$G%}%d%{$r%}"

4.3.2 was okay. I removed all .zwc files and .zcompdump before 
starting the shell.


Thorsten

Re: zsh 4.3.3 released

[Christoph Bauer]

Hi,


SunOS compiler complains about:

cc -c -I. -I/T/tcl_tk/8.5a5/binaries/suns8/include -DHAVE_CONFIG_H -O  -o utils.o ../../Src/utils.c
"../../Src/utils.c", line 3058: cannot dereference non-pointer type
"../../Src/utils.c", line 3058: operands must have integral type: op "&"
"../../Src/utils.c", line 3059: cannot dereference non-pointer type
cc: acomp failed for ../../Src/utils.c
make[2]: *** [utils.o] Error 2
make[2]: Leaving directory `/projekte/backup/tools/unix/build/zsh/zsh-4.3.3/suns8/Src'
make[1]: *** [modobjs] Error 2
make[1]: Leaving directory `/projekte/backup/tools/unix/build/zsh/zsh-4.3.3/suns8/Src'
make: *** [all] Error 1

Patch:

*** utils.c.orig	Mon Apr 16 15:17:35 2007
--- utils.c	Tue Apr 17 11:52:29 2007
***************
*** 3055,3062 ****
      if (len == 0) {
  	/* NULL is special */
  	return zistype(0, itype);
!     } else if (len == 1 && isascii(*outstr)) {
! 	return zistype(*outstr, itype);
      } else {
  	switch (itype) {
  	case IIDENT:
--- 3055,3062 ----
      if (len == 0) {
  	/* NULL is special */
  	return zistype(0, itype);
!     } else if (len == 1 && isascii(outstr[0])) {
! 	return zistype(outstr[0], itype);
      } else {
  	switch (itype) {
  	case IIDENT:

Christoph Bauer
--  
LMS Deutschland GmbH - Geschäftsführer: Dr. Ing. Urbain Vandeurzen, Dr.-Ing. Norbert Reimann - Sitz: Kaiserslautern - Registergericht: HRB Kaiserslautern 3706
 

Re: zsh 4.3.3 released

[Peter Stephenson]

"Christoph Bauer" wrote:
> SunOS compiler complains about:
> 
> cc -c -I. -I/T/tcl_tk/8.5a5/binaries/suns8/include -DHAVE_CONFIG_H -O  -o u=
> tils.o ../../Src/utils.c
> "../../Src/utils.c", line 3058: cannot dereference non-pointer type
> 
> Patch:

Thanks, VARARR() compiles differently for different compilers so I
missed this.  I've committed it.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070


To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php

To get further information regarding CSR, please visit our Investor Relations page at http://ir.csr.com/csr/about/overview

Re: zsh 4.3.3 released

[Thorsten Kampe]

* Thorsten Kampe (Tue, 17 Apr 2007 00:27:34 +0100)
> * Peter Stephenson (Mon, 16 Apr 2007 17:47:02 +0100)
> > I've uploaded version 4.3.3 of zsh to ftp.zsh.org.  If there aren't
> > complaints too loud I'll send it to Sourceforge and announce it on
> > Freshmeat tomorrow.
> 
> I freshly compiled and installed on Cygwin and I get this is 
> completion
> [Color escape codes in completio description]

Could I have a confirmation for this? Was my informal bug report 
sufficient or do you need a more detailed description?!

Thorsten

Re: zsh 4.3.3 released

[Peter Stephenson]

Thorsten Kampe wrote:
> Could I have a confirmation for this? Was my informal bug report 
> sufficient or do you need a more detailed description?!

I haven't yet understood what the problem actually is, nor reproduced it
yet.  I think you're claiming that escape sequences used to be processed
literally by the terminal output for certain styles, similar to in
prompts, but now aren't.  If so, this should be fixable.  (The
zshcompsys documentation isn't explicit about what escape sequences are
allowed but sort of implies anything that works in a normal prompt
should here.)

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070


To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php

To get further information regarding CSR, please visit our Investor Relations page at http://ir.csr.com/csr/about/overview

Re: zsh 4.3.3 released

[Peter Stephenson]

Peter Stephenson <pws@csr.com> wrote:
> Thorsten Kampe wrote:
> > Could I have a confirmation for this? Was my informal bug report 
> > sufficient or do you need a more detailed description?!
> 
> I haven't yet understood what the problem actually is, nor reproduced it
> yet.

OK, I've found it; it wasn't too hard---all uses of %{ in completion were
basically broken (I'm incrementing the pointer before instead of after
processing the escape character).  Thanks for the report.  See if there are
any problems left after this.

Index: Src/Zle/complist.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/Zle/complist.c,v
retrieving revision 1.102
diff -u -r1.102 complist.c
--- Src/Zle/complist.c	2 Apr 2007 13:58:19 -0000	1.102
+++ Src/Zle/complist.c	17 Apr 2007 14:04:08 -0000
@@ -1046,13 +1046,11 @@
 			tcout(TCUNDERLINEEND);
 		    break;
 		case ZWC('{'):
-		    for (p++; *p && (*p != '%' || p[1] != '}'); p++)
+		    for (; *p && (*p != '%' || p[1] != '}'); p++)
 			if (dopr)
 			    putc(*p == Meta ? *++p ^ 32 : *p, shout);
 		    if (*p)
-			p++;
-		    else
-			p--;
+			p += 2;
 		    break;
 		case ZWC('m'):
 		    if (stat) {


-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070


To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php

To get further information regarding CSR, please visit our Investor Relations page at http://ir.csr.com/csr/about/overview

Re: zsh 4.3.3 released

[Thorsten Kampe]

* Peter Stephenson (Tue, 17 Apr 2007 15:06:45 +0100)
> Peter Stephenson <pws@csr.com> wrote:
> > Thorsten Kampe wrote:
> > > Could I have a confirmation for this? Was my informal bug report 
> > > sufficient or do you need a more detailed description?!
> > 
> > I haven't yet understood what the problem actually is, nor reproduced it
> > yet.
> 
> OK, I've found it; it wasn't too hard---all uses of %{ in completion were
> basically broken (I'm incrementing the pointer before instead of after
> processing the escape character).  Thanks for the report.  See if there are
> any problems left after this.
> 
> Index: Src/Zle/complist.c
> ===================================================================
> RCS file: /cvsroot/zsh/zsh/Src/Zle/complist.c,v
> retrieving revision 1.102
> diff -u -r1.102 complist.c
> --- Src/Zle/complist.c	2 Apr 2007 13:58:19 -0000	1.102
> +++ Src/Zle/complist.c	17 Apr 2007 14:04:08 -0000
> [...]

Sorry, but how would I have to apply this patch?!

Re: zsh 4.3.3 released

[Thorsten Kampe]

* Thorsten Kampe (Tue, 17 Apr 2007 17:10:12 +0100)
> * Peter Stephenson (Tue, 17 Apr 2007 15:06:45 +0100)
> > Peter Stephenson <pws@csr.com> wrote:
> > > Thorsten Kampe wrote:
> > > > Could I have a confirmation for this? Was my informal bug report 
> > > > sufficient or do you need a more detailed description?!
> > > 
> > > I haven't yet understood what the problem actually is, nor reproduced it
> > > yet.
> > 
> > OK, I've found it; it wasn't too hard---all uses of %{ in completion were
> > basically broken (I'm incrementing the pointer before instead of after
> > processing the escape character).  Thanks for the report.  See if there are
> > any problems left after this.
> > 
> > Index: Src/Zle/complist.c
> > ===================================================================
> > RCS file: /cvsroot/zsh/zsh/Src/Zle/complist.c,v
> > retrieving revision 1.102
> > diff -u -r1.102 complist.c
> > --- Src/Zle/complist.c	2 Apr 2007 13:58:19 -0000	1.102
> > +++ Src/Zle/complist.c	17 Apr 2007 14:04:08 -0000
> > [...]
> 
> Sorry, but how would I have to apply this patch?!

Found it myself: "patch -u < file". With the patch the problem is 
gone. Thanks.

T.

Re: zsh 4.3.3 released

[Wayne Davison]

[-- Attachment #1: Type: text/plain, Size: 439 bytes --]

On Mon, Apr 16, 2007 at 05:47:02PM +0100, Peter Stephenson wrote:
> I've uploaded version 4.3.3 of zsh to ftp.zsh.org.

I just tested out the latest CVS version, and found that I get an error
using the smart-insert-last-word function:

smart-insert-last-word:108: bad math expression: operator expected at `Account fo...'

The attached patch fixes this for me.  Was it ever possible to infix a
comment inside a math expression?

..wayne..

[-- Attachment #2: smart.patch --]
[-- Type: text/x-diff, Size: 532 bytes --]

--- Functions/Zle/smart-insert-last-word	26 Feb 2007 07:43:39 -0000	1.3
+++ Functions/Zle/smart-insert-last-word	17 Apr 2007 18:06:36 -0000
@@ -104,8 +104,8 @@ fi
               found=$lastcmd[(I)$pattern]
           done
       fi
-      (( found-- > 0 &&            # Account for 1-based index
-        (numeric = $#lastcmd - found) ))
+      # The following will account for 1-based index
+      (( found-- > 0 && (numeric = $#lastcmd - found) ))
   fi
 } always {
   HISTNO=$_ilw_hist                # Return to current command

Re: zsh 4.3.3 released

[Peter Stephenson]

Wayne Davison wrote:
> I just tested out the latest CVS version, and found that I get an error
> using the smart-insert-last-word function:
> 
> smart-insert-last-word:108: bad math expression: operator expected at `Accoun
> t fo...'
> 
> The attached patch fixes this for me.  Was it ever possible to infix a
> comment inside a math expression?

No, I don't think the math code has ever tried to parse comments, and
I don't think it should.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070


To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php

To get further information regarding CSR, please visit our Investor Relations page at http://ir.csr.com/csr/about/overview

zsh 4.3.4 released

[Peter Stephenson]

Peter Stephenson <pws@csr.com> wrote:
> I've uploaded version 4.3.3 of zsh to ftp.zsh.org.

I've uploaded 4.3.4 with fixes for the problems noted in 4.3.3 (thanks for
the responses).

(add-zsh-hook was in the 4.3.3 tar file but not in CVS, since I forgot to
commit it, so the difference there is in the archive but not the
distribution itself.  I still need to document this in the zshcontrib
manual.)


2007-04-19  Peter Stephenson  <pws@csr.com>

	* unposted: Functions/Misc/add-zsh-hook: should have been
	added ages ago but wasn't.

	* README, Config/version.mk, Etc/FAQ.yo: release 4.3.4.

2007-04-17  Wayne Davison  <wayned@users.sourceforge.net>

	* users/11424: Functions/Zle/smart-insert-last-word: fixed
	parse error caused by a comment in a math expression.

2007-04-17  Peter Stephenson  <pws@csr.com>

	* users/11421: Src/Zle/complist.c: use of %{...%} in
	completion list formatting was broken.

	* Christoph Bauer: users/11416: Src/utils.c: bad use
	of VARARR confused some compilers.

2007-04-17  Geoff Wing  <gcw@zsh.org>

	* unposted: LICENCE: update year to 2007



To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php

To get further information regarding CSR, please visit our Investor Relations page at http://ir.csr.com/csr/about/overview

Re: zsh 4.3.4 released

[Mikael Magnusson]

On 19/04/07, Peter Stephenson <pws@csr.com> wrote:
> Peter Stephenson <pws@csr.com> wrote:
> > I've uploaded version 4.3.3 of zsh to ftp.zsh.org.
>
> I've uploaded 4.3.4 with fixes for the problems noted in 4.3.3 (thanks for
> the responses).

I found a weird problem yesterday, if i type 'which svmode' after
sourcing [1] (from zsh -f), zsh allocates all free memory then ooms.
(so set ulimit -v 50000 or so before trying). The strange thing is it
works fine if i cat the file and paste it in the terminal.

I tried reproducing it with a smaller file with dummy functions but
that didn't crash. I don't have debug symbols in right now but here's
a hopefully partly useful backtrace,
(gdb) bt
#0  0x080d9d20 in taddnl ()
#1  0x080dbda0 in gettext2 ()
#2  0x080dcba9 in getpermtext ()
#3  0x0807f7b8 in printshfuncnode ()
#4  0x0805c2ec in bin_whence ()
#5  0x0805335e in execbuiltin ()
#6  0x080724b4 in execcmd ()
#7  0x08072715 in execpline2 ()
#8  0x08072cd3 in execpline ()
#9  0x08074705 in execlist ()
#10 0x080748ea in execode ()
#11 0x0808901d in loop ()
#12 0x0808be98 in zsh_main ()
#13 0x08052d0e in main ()


[1] http://mikachu.ath.cx/slask/dot-zshurxvt
-- 
Mikael Magnusson

Re: zsh 4.3.4 released

[Brian K. White]

----- Original Message ----- 
From: "Mikael Magnusson" <mikachu@gmail.com>
To: "Peter Stephenson" <pws@csr.com>
Cc: "Zsh users list" <zsh-users@sunsite.dk>
Sent: Friday, April 20, 2007 5:40 PM
Subject: Re: zsh 4.3.4 released


> On 19/04/07, Peter Stephenson <pws@csr.com> wrote:
>> Peter Stephenson <pws@csr.com> wrote:
>> > I've uploaded version 4.3.3 of zsh to ftp.zsh.org.
>>
>> I've uploaded 4.3.4 with fixes for the problems noted in 4.3.3 (thanks 
>> for
>> the responses).

Builds and installs clean on cygwin.

I can offer little in the way of testing though, as it's so feature rich and 
I barely use it.
but at least ./configure, make, make install (just like that, no options or 
environment variables) offered no errors and the result runs.

MinGW was another story:

bkw@tv /cygdrive/d/src/zsh-4.3.4
$ ./configure:
[...]
zsh configuration
-----------------
zsh version               : 4.3.4
host operating system     : i686-pc-cygwin
source code location      : .
compiler                  : gcc
preprocessor flags        :
executable compiler flags :  -Wall -Wmissing-prototypes -O2
executable linker flags   :   -s
library flags             : -lm
installation basename     : zsh
binary install path       : /usr/local/bin
man page install path     : /usr/local/man
info install path         : /usr/local/info
functions install path    : /usr/local/share/zsh/4.3.4/functions
See config.modules for installed modules and functions.


bkw@tv /cygdrive/d/src/zsh-4.3.4
$ make
make[1]: Entering directory `/usr/local/src/zsh-4.3.4/Src'
cd .. && /bin/sh $top_srcdir/Src/mkmakemod.sh Src Makemod
creating Src/Makemod.in
config.status: creating Src/Makemod
make[2]: Entering directory `/usr/local/src/zsh-4.3.4/Src'
Makemod:532: *** multiple target patterns.  Stop.
make[2]: Leaving directory `/usr/local/src/zsh-4.3.4/Src'
make[2]: Entering directory `/usr/local/src/zsh-4.3.4/Src'
make[2]: Makemod: No such file or directory
make[2]: *** No rule to make target `Makemod'.  Stop.
make[2]: Leaving directory `/usr/local/src/zsh-4.3.4/Src'
make[1]: *** [headers] Error 2
make[1]: Leaving directory `/usr/local/src/zsh-4.3.4/Src'
make: *** [all] Error 1

bkw@tv /usr/local/src/zsh-4.3.4
$

configure logs here: rsync of the build attempt
http://brian.aljex.com/mingw/src/zsh-4.3.4/

But MinGW is such an odd environment this doesn't mean much. Plus I'm still 
just beginning to try this build so I'm not actually yelling for help yet as 
there are tons of things I haven't tried yet. (like join the mingw mail list 
even)
Mostly just wanted to report no problems on cygwin.
For the record the way I'm using mingw, so far, is to use cygwin with 
mingw's bin prepended to PATH, rather than msys, and it's working ok to 
build some other dead simple utils at least.

Brian K. White    brian@aljex.com    http://www.myspace.com/KEYofR
+++++[>+++[>+++++>+++++++<<-]<-]>>+.>.+++++.+++++++.-.[>+<---]>++.
filePro  BBx    Linux  SCO  FreeBSD    #callahans  Satriani  Filk!

Re: zsh 4.3.4 released

[Peter Stephenson]

"Mikael Magnusson" <mikachu@gmail.com> wrote:
> I found a weird problem yesterday, if i type 'which svmode' after
> sourcing [1] (from zsh -f), zsh allocates all free memory then ooms.
> (so set ulimit -v 50000 or so before trying). The strange thing is it
> works fine if i cat the file and paste it in the terminal.
> 
> [1] http://mikachu.ath.cx/slask/dot-zshurxvt

I haven't been able to get this to happen on various Linux systems or
Solaris 8.  The trace implies its happening in the code that recreates the
function in text for output.  Finding out where the final taddnl() is being
called from might help.  It's possibly an inconsistency with allocation
that's been hidden up to now.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR PLC, Churchill House, Cambridge Business Park, Cowley Road
Cambridge, CB4 0WZ, UK                          Tel: +44 (0)1223 692070


To access the latest news from CSR copy this link into a web browser:  http://www.csr.com/email_sig.php

To get further information regarding CSR, please visit our Investor Relations page at http://ir.csr.com/csr/about/overview

Re: zsh 4.3.4 released

[Mikael Magnusson]

On 23/04/07, Peter Stephenson <pws@csr.com> wrote:
> "Mikael Magnusson" <mikachu@gmail.com> wrote:
> > I found a weird problem yesterday, if i type 'which svmode' after
> > sourcing [1] (from zsh -f), zsh allocates all free memory then ooms.
> > (so set ulimit -v 50000 or so before trying). The strange thing is it
> > works fine if i cat the file and paste it in the terminal.
> >
> > [1] http://mikachu.ath.cx/slask/dot-zshurxvt
>
> I haven't been able to get this to happen on various Linux systems or
> Solaris 8.  The trace implies its happening in the code that recreates the
> function in text for output.  Finding out where the final taddnl() is being
> called from might help.  It's possibly an inconsistency with allocation
> that's been hidden up to now.

Here it is with debug symbols,
taddnl () at text.c:42
42	    *tptr++ = c;
(gdb) bt
#0  taddnl () at text.c:42
#1  0x080dbe30 in gettext2 (state=0xafa162a0) at text.c:583
#2  0x080dcc39 in getpermtext (prog=0x8156710, c=0x8156754) at text.c:133
#3  0x0807f7f8 in printshfuncnode (hn=0x8156a68,
printflags=-1478367808) at hashtable.c:889
#4  0x0805c2ec in bin_whence (nam=0xa7edc820 "which", argv=0xafa164f0,
ops=0xafa16540, func=0)
    at builtin.c:3050
#5  0x0805335e in execbuiltin (args=0xa7edc820, bn=0x80f1728) at builtin.c:438
#6  0x080724c4 in execcmd (state=0xafa1ba40, input=0, output=0,
how=18, last1=2) at exec.c:2670
#7  0x08072725 in execpline2 (state=0xafa1ba40, pcode=0, how=18,
input=0, output=0, last1=0)
    at exec.c:1343
#8  0x08072ce3 in execpline (state=0xafa1ba40, slcode=33554432,
how=18, last1=0) at exec.c:1129
#9  0x08074715 in execlist (state=0xafa1ba40, dont_change_job=0,
exiting=0) at exec.c:935
#10 0x080748fa in execode (p=0xa7edc738, dont_change_job=0, exiting=0)
at exec.c:793
#11 0x0808904d in loop (toplevel=1, justonce=0) at init.c:180
#12 0x0808bec8 in zsh_main (argc=1, argv=0xafa1bb74) at init.c:1347
#13 0x08052d0e in main (argc=0, argv=0x0) at ./main.c:93

Here's another one,
Program received signal SIGSEGV, Segmentation fault.
0x080d0ce4 in taddchr (c=9) at text.c:42
42	    *tptr++ = c;
(gdb) bt
#0  0x080d0ce4 in taddchr (c=9) at text.c:42
#1  0x080d0ed9 in taddnl () at text.c:104
#2  0x080d1df7 in gettext2 (state=0xafcafe10) at text.c:583
#3  0x080d0f8a in getpermtext (prog=0x814c6f8, c=0x814c73c) at text.c:133
#4  0x08081057 in printshfuncnode (hn=0x814ca50, printflags=32) at
hashtable.c:889
#5  0x080619f7 in bin_whence (nam=0xa7f44790 "which", argv=0xafcb00a0,
ops=0xafcb00f0, func=0)
    at builtin.c:3050
#6  0x080536bd in execbuiltin (args=0xa7f44768, bn=0x80e73a8) at builtin.c:438
#7  0x0807496c in execcmd (state=0xafcb04d0, input=0, output=0,
how=18, last1=2) at exec.c:2670
#8  0x08070cd1 in execpline2 (state=0xafcb04d0, pcode=131, how=18,
input=0, output=0, last1=0)
    at exec.c:1343
#9  0x0806fc28 in execpline (state=0xafcb04d0, slcode=4098, how=18,
last1=0) at exec.c:1129
#10 0x0806f355 in execlist (state=0xafcb04d0, dont_change_job=0,
exiting=0) at exec.c:935
#11 0x0806ef63 in execode (p=0xa7f446a8, dont_change_job=0, exiting=0)
at exec.c:793
#12 0x0808996c in loop (toplevel=1, justonce=0) at init.c:180
#13 0x0808c470 in zsh_main (argc=1, argv=0xafcb0604) at init.c:1347
#14 0x08052d12 in main (argc=1, argv=0xafcb0604) at ./main.c:93

the first is with O3 and the second with O0 (in order of remembering
how to debug :)

-- 
Mikael Magnusson

RE: zsh 4.3.4 released

[Christoph Bauer]

Two short remarks:

  - HP-UX 11: I need to define _INCLUDE__STDC_A1_SOURCE to compile zsh
    with HP's cc.
    (set CFLAGS="$OTHER_CFLAGS -D_INCLUDE__STDC_A1_SOURCE")
    
  - on SGI zsh-4.3.4 crashes immediatly:

Starting program: /projekte/backup/tools/unix/sgi6n3/bin/zsh-4.3.4 

Program received signal SIGSEGV, Segmentation fault.
0x1005b328 in optlookup ()
(gdb) where
#0  0x1005b328 in optlookup ()
#1  0x1005b0c4 in bin_setopt ()
#2  0x1000e8f8 in execbuiltin ()
#3  0x1000e8f8 in execbuiltin ()
Previous frame identical to this frame (corrupt stack?)

   zsh-4.3.2 runs fine.

Regards,

Christoph Bauer
--  
LMS Deutschland GmbH - Geschäftsführer: Dr. Ing. Urbain Vandeurzen, Dr.-Ing. Norbert Reimann - Sitz: Kaiserslautern - Registergericht: HRB Kaiserslautern 3706