4.0.3

4.0.3

[Peter Stephenson]

I am in the process of uploading 4.0.3 to the zsh site.  It takes a long
time because I always make several dozen mistakes, so if anybody's
forgotten to commit anything it's too late.  I haven't altered the links
yet or moved 4.0.2.  It would be useful if a few people could check that
this version works before I announce it elsewhere and upload it to
sourceforge.

The announcement will look something like:


Version 4.0.3 of zsh, the Z-Shell, has been released.  Zsh is an enhanced
UNIX shell related to sh and ksh.  This is a maintenance release of the
stable 4.0 series.  There are very few user-visible changes, mostly
minor enhancements to the completion function system.  Some configuration
and compilation issues have been addressed to improve compatibility with
AIX, MacOS X/Darwin, and QNX Neutrino.

Please send comments and bug reports to <zsh-workers@sunsite.dk>.

The META-FAQ, giving general information about availability of the shell,
is appended.

Peter Stephenson <coordinator@zsh.org>


------------------------
META-FAQ for the Z Shell
------------------------

The latest version of this META-FAQ can be found at any of the FTP sites
listed below.

Author
------

Zsh was originally written by Paul Falstad <pf@zsh.org>.
Zsh is now maintained by the members of the zsh-workers mailing
list <zsh-workers@sunsite.dk>.  The development is currently
coordinated by Peter Stephenson <pws@zsh.org>.  The coordinator
can be contacted at <coordinator@zsh.org>, but matters relating to
the code should generally go to the mailing list.

Availability
------------

Zsh is available from the following anonymous FTP sites.  These mirror
sites are kept frequently up to date.  The sites marked with (H) may be
mirroring ftp.cs.elte.hu instead of the primary site.

      Primary site
            ftp://ftp.zsh.org/pub/zsh/
            http://www.zsh.org/pub/zsh/

      Australia
            ftp://ftp.zsh.org/pub/zsh/
            http://www.zsh.org/pub/zsh/
            ftp://ftp.ips.gov.au/pub/packages/zsh/  (H)

      Denmark
            ftp://sunsite.dk/pub/unix/shells/zsh/

      Finland
            ftp://ftp.funet.fi/pub/unix/shells/zsh/

      France
            ftp://ftp.cenatls.cena.dgac.fr/shells/zsh/

      Germany
            ftp://ftp.fu-berlin.de/pub/unix/shells/zsh/  (H)
            ftp://ftp.gmd.de/packages/zsh/
            ftp://ftp.uni-trier.de/pub/unix/shell/zsh/

      Hungary
            ftp://ftp.cs.elte.hu/pub/zsh/
            http://www.cs.elte.hu/pub/zsh/
            ftp://ftp.kfki.hu/pub/packages/zsh/

      Israel
            ftp://ftp.math.technion.ac.il/pub/zsh/
            http://www.math.technion.ac.il/pub/zsh/

      Italy
            ftp://ftp.unina.it/pub/Unix/pkgs/shell/zsh/

      Japan
            ftp://ftp.nisiq.net/pub/shells/zsh/  (H)
            ftp://ftp.win.ne.jp/pub/shell/zsh/

      Norway
            ftp://ftp.uit.no/pub/unix/shells/zsh/

      Poland
            ftp://sunsite.icm.edu.pl/pub/unix/shells/zsh/

      Romania
            ftp://ftp.roedu.net/pub/mirrors/ftp.zsh.org/pub/zsh/
            ftp://ftp.kappa.ro/pub/mirrors/ftp.zsh.org/pub/zsh/

      Slovenia
            ftp://ftp.siol.net/mirrors/zsh/

      Sweden
            ftp://ftp.lysator.liu.se/pub/unix/zsh/

      UK
            ftp://ftp.net.lut.ac.uk/zsh/
            ftp://sunsite.org.uk/packages/zsh/

      USA
            ftp://uiarchive.uiuc.edu/pub/packages/shells/zsh/
            ftp://ftp.rge.com/pub/shells/zsh/
            ftp://foad.org/pub/zsh/
            http://foad.org/zsh/

Mailing Lists
-------------

Zsh has 3 mailing lists:

      <zsh-announce@sunsite.dk>
            Announcements about releases, major changes in the shell and the
            monthly posting of the Zsh FAQ.  (moderated)

      <zsh-users@sunsite.dk>
            User discussions.

      <zsh-workers@sunsite.dk>
            Hacking, development, bug reports and patches.

To subscribe or unsubscribe, send mail
to the associated administrative address for the mailing list.

      <zsh-announce-subscribe@sunsite.dk>
      <zsh-users-subscribe@sunsite.dk>
      <zsh-workers-subscribe@sunsite.dk>

      <zsh-announce-unsubscribe@sunsite.dk>
      <zsh-users-unsubscribe@sunsite.dk>
      <zsh-workers-unsubscribe@sunsite.dk>

YOU ONLY NEED TO JOIN ONE OF THE MAILING LISTS AS THEY ARE NESTED.
All submissions to zsh-announce are automatically forwarded to
zsh-users.  All submissions to zsh-users are automatically
forwarded to zsh-workers.

If you have problems subscribing/unsubscribing to any of the mailing
lists, send mail to <listmaster@zsh.org>.  The mailing lists are
maintained by Karsten Thygesen <karthy@kom.auc.dk>.

The mailing lists are archived; the archives can be accessed via the
administrative addresses listed above.  There is also a hypertext
archive, maintained by Geoff Wing <gcw@zsh.org>, available at
http://www.zsh.org/mla/.

The Zsh FAQ
-----------

Zsh has a list of Frequently Asked Questions (FAQ), maintained by
Peter Stephenson <pws@zsh.org>.  It is regularly posted to the
newsgroup comp.unix.shell and the zsh-announce mailing list.
The latest version can be found at any of the Zsh FTP sites, or at
http://www.zsh.org/FAQ/.  The contact address for FAQ-related matters
is <faqmaster@zsh.org>.

The Zsh Web Page
----------------

Zsh has a web page which is located at http://www.zsh.org/.  This is
maintained by Karsten Thygesen <karthy@zsh.org>, of SunSITE Denmark.
The contact address for web-related matters is <webmaster@zsh.org>.

The Zsh Userguide
-----------------

A userguide is currently in preparation.  It is intended to complement the
manual, with explanations and hints on issues where the manual can be
cabbalistic, hierographic, or downright mystifying (for example, the word
`hierographic' does not exist).  It can be viewed in its current state at
http://zsh.sunsite.dk/Guide/.  At the time of writing, chapters
dealing with startup files and their contents and the new completion system
were essentially complete.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Science Park, Milton Road,
Cambridge, CB4 0WH, UK                          Tel: +44 (0)1223 392070


**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************

Re: 4.0.3

[Bart Schaefer]

On Oct 24,  6:30pm, Peter Stephenson wrote:
}
} It would be useful if a few people could check that
} this version works before I announce it elsewhere and upload it to
} sourceforge.

I haven't tried the tarball (obviously, given that it's not there yet)
but the head of the 4-0-patches branch compiles and passes "make check"
on RedHat 5.2 and 6.2.

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net   

Re: 4.0.3

[Trond Eivind Glomsrød]

Bart Schaefer <schaefer@brasslantern.com> writes:

> On Oct 24,  6:30pm, Peter Stephenson wrote:
> }
> } It would be useful if a few people could check that
> } this version works before I announce it elsewhere and upload it to
> } sourceforge.
> 
> I haven't tried the tarball (obviously, given that it's not there yet)
> but the head of the 4-0-patches branch compiles and passes "make check"
> on RedHat 5.2 and 6.2.

The 4.0.3 tarball passes "make check" on Red Hat Linux 7.2.

-- 
Trond Eivind Glomsrød
Red Hat, Inc.

Re[2]: 4.0.3

[S. Cowles]

>> On Oct 24,  6:30pm, Peter Stephenson wrote:
>> }
>> } It would be useful if a few people could check that
>> } this version works before I announce it elsewhere and upload it to
>> } sourceforge.
>>

Passes make check on the following version of cygwin with only two
(expected) failures.

CYGWIN_NT-5.0 BUBO6 1.3.4s(0.46/3/2) 20011003 22:37:17 i686 unknown

./C02cond.ztst: starting.
Warning: Not testing [[ -b blockdevice ]] (no devices found)
Test ./C02cond.ztst failed: bad status 1, expected 0 from:
  [[ -g modish && ! -g zerolength ]]
Was testing: -g cond
./C02cond.ztst: test failed.
./D03procsubst.ztst: starting.
Test ./D03procsubst.ztst failed: bad status 1, expected 0 from:
  paste <(cut -f1 FILE1) <(cut -f3 FILE2)
Error output:
ZTST_execchunk:2: doesn't look like your system supports FIFOs.
Was testing: <(...) substitution
./D03procsubst.ztst: test failed.

-- 

S. Cowles
BioInfoGen, Inc.
1600 Adams Drive
Menlo Park, CA 94025-1448

scowles@earthlink.net
5102901165@messaging.sprintpcs.com

RE: Re[2]: 4.0.3

[Borsenkow Andrej]

> 
> ./C02cond.ztst: starting.
> Warning: Not testing [[ -b blockdevice ]] (no devices found)
> Test ./C02cond.ztst failed: bad status 1, expected 0 from:
>   [[ -g modish && ! -g zerolength ]]
> Was testing: -g cond
> ./C02cond.ztst: test failed.
> ./D03procsubst.ztst: starting.
> Test ./D03procsubst.ztst failed: bad status 1, expected 0 from:
>   paste <(cut -f1 FILE1) <(cut -f3 FILE2)
> Error output:
> ZTST_execchunk:2: doesn't look like your system supports FIFOs.
> Was testing: <(...) substitution
> ./D03procsubst.ztst: test failed.
> 

These are 15099 and unposted from Bart referring to 14826 (ChangeLog
from 2001-06-09). Should it go into 4.0? I was unsure about 15099 but
obviously nothing better has emerged since then.

-andrej

Re: 4.0.3

[Mads Martin Joergensen]

* Peter Stephenson <pws@csr.com> [Oct 24. 2001 19:31]:
> Version 4.0.3 of zsh, the Z-Shell, has been released.  Zsh is an enhanced

How come the following stops working in 4.0.3, according to the manpage
it is correct:

(mmj@staudinger) ~> ulimit -Sm unlimited
ulimit: bad option: -m
zsh: exit 1

-- 
Mads Martin Joergensen, http://mmj.dk
"Why make things difficult, when it is possible to make them cryptic
 and totally illogic, with just a little bit more effort."
                                -- A. P. J.

Re: 4.0.3

[Peter Stephenson]

Mads Martin Joergensen wrote:
> How come the following stops working in 4.0.3, according to the manpage
> it is correct:
> 
> (mmj@staudinger) ~> ulimit -Sm unlimited
> ulimit: bad option: -m
> zsh: exit 1

-m corresponds to the internal definition RLIMIT_RSS, see
/usr/include/sys/resource.h.

Are you on a system where RLIMIT_VMEM == RLIMIT_RSS?  If so, this is the
effect of 16033.  (If not, nothing should have changed since 4.0.2.)
Probably the best thing to do is handle this as a special case, as follows
--- this applies to both branches.

I changed the output message for the case in question, since it obviously
doesn't make sense to discriminate between virtual and physical memory here.

I can't really test this here, I don't have RLIMIT_RSS.  Not sure if this
is enough to upgrade 4.0.3 to 4.0.4, but at least I haven't announced 4.0.3
yet.

Index: Src/Builtins/rlimits.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/Builtins/rlimits.c,v
retrieving revision 1.3
diff -u -r1.3 rlimits.c
--- Src/Builtins/rlimits.c	2001/10/22 17:18:29	1.3
+++ Src/Builtins/rlimits.c	2001/10/25 10:49:25
@@ -44,12 +44,6 @@
 
 # include "rlimits.h"
 
-/* If RLIMIT_VMEM and RLIMIT_RSS are defined and equal, drop support *
- * for RLIMIT_RSS.  Observed on QNX Neutrino 6.1.0.                 */
-#if defined(RLIMIT_RSS) && defined(RLIMIT_VMEM) && (RLIMIT_RSS == RLIMIT_VMEM)
-#undef RLIMIT_RSS
-#endif
-
 # if defined(RLIM_T_IS_QUAD_T) || defined(RLIM_T_IS_LONG_LONG) || defined(RLIM_T_IS_UNSIGNED)
 static rlim_t
 zstrtorlimt(const char *s, char **t, int base)
@@ -174,7 +168,9 @@
 	if (limit != RLIM_INFINITY)
 	    limit /= 512;
 	break;
-# ifdef RLIMIT_RSS
+/* If RLIMIT_VMEM and RLIMIT_RSS are defined and equal, avoid *
+ * duplicate case statement.  Observed on QNX Neutrino 6.1.0. */
+# if defined(RLIMIT_RSS) && (!defined(RLIMIT_VMEM) || RLIMIT_VMEM != RLIMIT_RSS)
     case RLIMIT_RSS:
 	if (head)
 	    printf("resident set size (kbytes) ");
@@ -205,7 +201,11 @@
 # ifdef RLIMIT_VMEM
     case RLIMIT_VMEM:
 	if (head)
+#  if defined(RLIMIT_RSS) && RLIMIT_VMEM == RLIMIT_RSS
+	    printf("memory size (kb)           ");
+#  else
 	    printf("virtual memory size (kb)   ");
+#  endif
 	if (limit != RLIM_INFINITY)
 	    limit /= 1024;
 	break;


-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Science Park, Milton Road,
Cambridge, CB4 0WH, UK                          Tel: +44 (0)1223 392070


**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************

RE: 4.0.3

[Borsenkow Andrej]

> 
> I can't really test this here, I don't have RLIMIT_RSS.  Not sure if
this
> is enough to upgrade 4.0.3 to 4.0.4, but at least I haven't announced
> 4.0.3
> yet.

Is it already tagged in CVS?

Re: 4.0.3

[Mads Martin Joergensen]

* Peter Stephenson <pws@csr.com> [Oct 25. 2001 12:55]:
> > (mmj@staudinger) ~> ulimit -Sm unlimited
> > ulimit: bad option: -m
> > zsh: exit 1
> 
> -m corresponds to the internal definition RLIMIT_RSS, see
> /usr/include/sys/resource.h.
> 
> Are you on a system where RLIMIT_VMEM == RLIMIT_RSS?  If so, this is the
> effect of 16033.  (If not, nothing should have changed since 4.0.2.)
> Probably the best thing to do is handle this as a special case, as follows
> --- this applies to both branches.
> 
> I changed the output message for the case in question, since it obviously
> doesn't make sense to discriminate between virtual and physical memory here.
> 
> I can't really test this here, I don't have RLIMIT_RSS.  Not sure if this
> is enough to upgrade 4.0.3 to 4.0.4, but at least I haven't announced 4.0.3
> yet.

With the patch applied, it is happy again, thanks. But I cannot seem to
find the two RLIMITs in question either on my SuSE Linux 7.3 system. It
might very well be that I'm a moron not looking in the right places
though.

I would not justify a 4.0.4 with this, when 4.0.3 as you pointed out not
have been announced yet. There's of course factors like CVS that have to
be taken into question, which I am unaware of.

-- 
Mads Martin Joergensen, http://mmj.dk
"Why make things difficult, when it is possible to make them cryptic
 and totally illogic, with just a little bit more effort."
                                -- A. P. J.

Re: 4.0.3

[Peter Stephenson]

Borsenkow Andrej wrote:
> > 
> > I can't really test this here, I don't have RLIMIT_RSS.  Not sure if
> this
> > is enough to upgrade 4.0.3 to 4.0.4, but at least I haven't announced
> > 4.0.3
> > yet.
> 
> Is it already tagged in CVS?

I've committed it on 4.1.  If it seems to be working, I'll commit it on
4.0, too.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Science Park, Milton Road,
Cambridge, CB4 0WH, UK                          Tel: +44 (0)1223 392070


**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************

RE: 4.0.3

[Borsenkow Andrej]

> > >
> > > I can't really test this here, I don't have RLIMIT_RSS.  Not sure
if
> > this
> > > is enough to upgrade 4.0.3 to 4.0.4, but at least I haven't
announced
> > > 4.0.3
> > > yet.
> >
> > Is it already tagged in CVS?
> 
> I've committed it on 4.1.  If it seems to be working, I'll commit it
on
> 4.0, too.
> 

No, I meant - if 4.0.3 is not yet tagged, you can apply patch as part of
4.0.3. Else it has to be 4.0.4. At least if I understand how CVS works.

-andrej

Re: 4.0.3

[Trond Eivind Glomsrød]

Borsenkow Andrej <Andrej.Borsenkow@mow.siemens.ru> writes:

> No, I meant - if 4.0.3 is not yet tagged, you can apply patch as part of
> 4.0.3. Else it has to be 4.0.4. At least if I understand how CVS works.

4.0.3 was available on public ftp servers yesterday.

-- 
Trond Eivind Glomsrød
Red Hat, Inc.

Re: 4.0.3

[Peter Stephenson]

Borsenkow Andrej wrote:
> No, I meant - if 4.0.3 is not yet tagged, you can apply patch as part of
> 4.0.3. Else it has to be 4.0.4. At least if I understand how CVS works.

You can move the tag, but it's trivial just to turn it into 4.0.4 and it
avoids confusion since there is already something called 4.0.3 out there.
It's tempting not to bother, but with some of these niggly little things
you spend more time replying to people who've found the problem than
providing a fixed version.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Science Park, Milton Road,
Cambridge, CB4 0WH, UK                          Tel: +44 (0)1223 392070


**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************