Final 3.0.0 test release

Final 3.0.0 test release

[Zoltan Hidvegi]

I've put together the final 3.0.0 release but before I release it I'd like
some other people to try it.  At the moment only the patch is available on
ftp://ftp.cs.elte.hu/pub/zsh/testing/zsh-3.0-pre6-final-3.0.0.diff.gz

The file date/size:

   18774 Aug 14 19:07 zsh-3.0-pre6-final-3.0.0.diff.gz

MD5 checksum:

61fcbbcdb9f032525bdaf51de99314ac  zsh-3.0-pre6-final-3.0.0.diff.gz

Do not forget to touch stamp-h.in and configure after patching.

The final zsh-3.0.0 will be released tomorrow and on Friday I'll leave for
a one-week holiday.

Cheers,

Zoltan

Re: Final 3.0.0 test release

[Richard Coleman]

> I've put together the final 3.0.0 release but before I release it I'd like
> some other people to try it.  At the moment only the patch is available on
> ftp://ftp.cs.elte.hu/pub/zsh/testing/zsh-3.0-pre6-final-3.0.0.diff.gz
> 
> The file date/size:
> 
>    18774 Aug 14 19:07 zsh-3.0-pre6-final-3.0.0.diff.gz
> 
> MD5 checksum:
> 
> 61fcbbcdb9f032525bdaf51de99314ac  zsh-3.0-pre6-final-3.0.0.diff.gz
> 
> Do not forget to touch stamp-h.in and configure after patching.
> 
> The final zsh-3.0.0 will be released tomorrow and on Friday I'll leave for
> a one-week holiday.

announce it here first, so the mirror sites can pull it down before
you announce it widely.

Also, are you going to post it to comp.sources.misc, and announce it
on comp.unix.shell?

rc

Re: Final 3.0.0 test release

[Bart Schaefer]

On Aug 14,  7:15pm, Zoltan Hidvegi wrote:
} Subject: Final 3.0.0 test release
}
} I've put together the final 3.0.0 release but before I release it I'd like
} some other people to try it.

I still object to having a configure test for the default behavior of
BSD_ECHO when run as "sh".  This should NOT be something that varies
from one installation to the next!

For one thing, people who already have an older version of zsh linked
to sh will end up with the same (possibly wrong) behavior that they had
before.

For another, it breaks the whole notion of being able to rely on the
consistent behavior of the shell following "emulate -R ...".

} The final zsh-3.0.0 will be released tomorrow and on Friday I'll leave for
} a one-week holiday.

Are you sure you shouldn't wait to do the release until you're back?  So
that the archive maintainer isn't absent for the first ten days of new
non-developer users fooling with it?

-- 
Bart Schaefer                             Brass Lantern Enterprises
http://www.well.com/user/barts            http://www.nbn.com/people/lantern

New male in /home/schaefer:
>N  2 Justin William Schaefer  Sat May 11 03:43  53/4040  "Happy Birthday"

Re: Final 3.0.0 test release

[Steven L Baur]

I'm mirroring the zsh manual on the web at
	http://www.miranova.com/zsh-man/

-- 
steve@miranova.com baur
Unsolicited commercial e-mail will be proofread for $250/hour.
Andrea Seastrand: For your vote on the Telecom bill, I will vote for anyone
except you in November.

Re: Final 3.0.0 test release

[Richard Coleman]

> } I've put together the final 3.0.0 release but before I release it I'd like
> } some other people to try it.
> 
> I still object to having a configure test for the default behavior of
> BSD_ECHO when run as "sh".  This should NOT be something that varies
> from one installation to the next!
> 
> For one thing, people who already have an older version of zsh linked
> to sh will end up with the same (possibly wrong) behavior that they had
> before.
> 
> For another, it breaks the whole notion of being able to rely on the
> consistent behavior of the shell following "emulate -R ...".

I completely agree...

The options for zsh when called as sh, should not depend on
your current configuration.  It is especially bad for us sysadmins
who work on heterogenous environments.... I don't want zsh working
one way on one machine, and another way on another machine.

One of the reasons people install third party shells is to get
consistency across their various platforms.

rc

Re: BSD_ECHO

[Zoltan Hidvegi]

Richard wrote:
[...]
> > I still object to having a configure test for the default behavior of
> > BSD_ECHO when run as "sh".  This should NOT be something that varies
> > from one installation to the next!
[...]
> I completely agree...
> 
> The options for zsh when called as sh, should not depend on
> your current configuration.  It is especially bad for us sysadmins
> who work on heterogenous environments.... I don't want zsh working
> one way on one machine, and another way on another machine.

Seing the big opposition I'll remove that configure check.  However this
was not a completely unreasonable change.  I use zsh as /bin/sh on Linux
where the default /bin/sh is bash.  I bet that without BSD_ECHO some
scripts will break.

On the other hand someone else reported that on an other system zsh broke
some scripts because of BSD_ECHO.  It is possible to turn on BSD_ECHO iff
zsh is invoked as /bin/sh (with full pathname) and configure detected that
/bin/sh use BSD_ECHO but it would not be set by emulate or alternatively is
would be always set or unset by emulate.  On systems where zsh is installed
as /bin/sh the system administrator probably knows enough about how to set
the desired default and others won't be affected by this change.

I know that prefect solution is not possible.  Even POSIX did not dare to
define the default echo behaviour.  If one wants to write a portable script
it must begin with something like #! /usr/local/bin/zsh -f.  We have 7
different OS'es in an NFS/NIS network where zsh is very useful as portable
script language (and as a login shell of course).

Zoltan

Re: Final 3.0.0 test release

[Zoltan Hidvegi]

> Are you sure you shouldn't wait to do the release until you're back?  So
> that the archive maintainer isn't absent for the first ten days of new
> non-developer users fooling with it?

Even if I'm here a new release should not happen within ten days of the
first release.  We have to wait for the bug reports to accumulate.  That
ten day is quite good for that.  In the last two weeks there were no really
serious bugreports from zsh-workers or from anywhere else.  I do not think
that an other ten day would bring up enough problems to justify this delay.

Zoltan

Re: Final 3.0.0 test release

[Zoltan Hidvegi]

> I'm mirroring the zsh manual on the web at
> 	http://www.miranova.com/zsh-man/

It can also be found in http://www.cs.elte.hu/zsh-manual/
The FAQ is in http://www.cs.elte.hu/zsh-manual/FAQ

Zoltan

Re: Final 3.0.0 test release

[Steven L Baur]

>>>>> "Zoltan" == Zoltan Hidvegi <hzoli@cs.elte.hu> writes:

sb> I'm mirroring the zsh manual on the web at
sb> http://www.miranova.com/zsh-man/

Zoltan> It can also be found in http://www.cs.elte.hu/zsh-manual/
Zoltan> The FAQ is in http://www.cs.elte.hu/zsh-manual/FAQ

Perhaps this could be added to the META-FAQ?
-- 
steve@miranova.com baur
Unsolicited commercial e-mail will be proofread for $250/hour.
Andrea Seastrand: For your vote on the Telecom bill, I will vote for anyone
except you in November.

Re: Final 3.0.0 test release

[Mark Borges]

>> On 14 Aug 1996 18:14:37 -0700,
>> Steven L Baur(sb) wrote:
>>>>>> "Zoltan" == Zoltan Hidvegi <hzoli@cs.elte.hu> writes:
sb> I'm mirroring the zsh manual on the web at
sb> http://www.miranova.com/zsh-man/

Zoltan> It can also be found in http://www.cs.elte.hu/zsh-manual/
Zoltan> The FAQ is in http://www.cs.elte.hu/zsh-manual/FAQ

Note that although the FAQ at mal.com is currently out of date it has
the advantage of being hyperlinked to the zsh manual. It should
be fairly simple to get it synced up again. I was mainly waiting for
the dust to settle on the documentation, which I guess has happened
now.

sb> Perhaps this could be added to the META-FAQ?

With all this mirroring going on, I'm wondering if the mal.com
connection is too slow for most? (it's only a shared T1 I believe). If
so, maybe I should be the one mirroring. Or perhaps I should just give
up the ghost and pass on the torch. I haven't had much time to give
the documentation project lately. I'm sorry if it's starting to show.

-- 
  -mb-

Re: Final 3.0.0 test release

[Marc Martinez]

Zoltan Hidvegi writes:
: From: Zoltan Hidvegi <hzoli@cs.elte.hu>
: Resent-Date: Wed, 14 Aug 1996 13:16:03 -0400 (EDT)
: Resent-Message-ID: <"kbAhh1.0.PO1.IdW4o"@euclid>
: Resent-From: zsh-workers@math.gatech.edu
: Resent-Sender: zsh-workers-request@math.gatech.edu
: Date: Wed, 14 Aug 1996 19:15:06 +0200 (MET DST)
: Posted-Date: Wed, 14 Aug 1996 13:25:41 -0400
: Organization: Dept. of Comp. Sci., Eotvos University, Budapest, Hungary
: Phone: (36 1)2669833 ext: 2667, home phone: (36 1) 2752368
: X-Mailer: ELM [version 2.4ME+ PL16 (25)]
: MIME-Version: 1.0
: Content-Type: text/plain; charset=US-ASCII
: Content-Transfer-Encoding: 7bit
: X-Mailing-List: <zsh-workers@math.gatech.edu> archive/latest/1971
: X-Loop: zsh-workers@math.gatech.edu
: Precedence: list
: Lines: 20
: Xref: smartsun.smartnet.net zsh-workers:697
: X-Gnus-Article-Number: 697   Wed Aug 14 17:20:56 1996
: To: zsh-workers@math.gatech.edu (Zsh workers list)
: Subject: Final 3.0.0 test release
: 
: I've put together the final 3.0.0 release but before I release it I'd like
: some other people to try it.  At the moment only the patch is available on
: ftp://ftp.cs.elte.hu/pub/zsh/testing/zsh-3.0-pre6-final-3.0.0.diff.gz
: 

After applying this patch, the command 'echotc clear' only outputs
'no', without clearing the screen.  3.0-pre6 Does The Right Thing(tm)
and clears the screen.

-- 
 Marc Martinez       "Sleep is unnecessary in the presence of more espresso."
 lastxit@mindport.net      Key fingerprint:   
 PGP public key available        47 AD 25 FF C2 B7 F8 57
  by finger or keyserver         C5 B6 2E B3 5E 98 A5 DE        Fnord.

Re: Final 3.0.0 test release

[Zoltan Hidvegi]

> After applying this patch, the command 'echotc clear' only outputs
> 'no', without clearing the screen.  3.0-pre6 Does The Right Thing(tm)
> and clears the screen.

There was no change that might cause such a problem.  I'm almost sure that
you installedd ncurses recently in a wrong way.  In that case read INSTALL
in the ncurses disribution carefully and either restore your original
termcap.h or link termcap.so to ncurses.so.

Zoltan

Re: Final 3.0.0 test release

[Steven L Baur]

>>>>> "mb" == Mark Borges <mdb@cdc.noaa.gov> writes:

sb> Perhaps this could be added to the META-FAQ?

mb> With all this mirroring going on, I'm wondering if the mal.com
mb> connection is too slow for most? (it's only a shared T1 I
mb> believe). If so, maybe I should be the one mirroring. Or perhaps I
mb> should just give up the ghost and pass on the torch. I haven't had
mb> much time to give the documentation project lately. I'm sorry if
mb> it's starting to show.

My addition of a mirror is more a result of the lingering chaos from
last weekend's power outage than from anything you did (or did not
do).  Mirrors and redundancy are A Good Thing.
-- 
steve@miranova.com baur
Unsolicited commercial e-mail will be proofread for $250/hour.
Andrea Seastrand: For your vote on the Telecom bill, I will vote for anyone
except you in November.

BSD_ECHO

[Zefram]

|+ #ifdef SH_USE_BSD_ECHO
|      {"bsdecho", 		0,    0,    OPT_EMULATE|OPT_SH},
|+ #else
|+     {"bsdecho", 		0,    0,    0},
|+ #endif

There are two problems with this.  Firstly, BSD_ECHO should have the
OPT_EMULATE flag, regardless of its default state.  Secondly, as I've
said before, it's a really bad idea to try to emulate the native sh --
on some OSes, such as the one I'm using here, the results of the
configure test depend on what the installer's $PATH is set to.  I
suggest you revert this patch for the final final 3.0 release, or at
least add the OPT_EMULATE flag to the second case.

-zefram