From malekith at pld.org.pl  Mon Jun  4 13:02:33 2001
From: malekith at pld.org.pl (Michal Moskal)
Date: Mon, 4 Jun 2001 13:02:33 +0200
Subject: metaconf tranformations
Message-ID: <20010604130233.A16246@aleph-0.aleph-0.dhs.org>

Ehlo,

I refined Lukasz's idea about xml->config translation.
What we need is :
 1) validation
 2) descriptions
 3) transformation
It seems to be good idea that everything will go into
a single file, as validation is close to writing descriptions
and/or error messages, and transformation requires validated
document, and it is more obvious what you can expect from validated
document if transformation is so close to validation.

Anyway: waiting for comments.

And, I can write parser for files like this.

<?xml version="1.0"?>
<config>
 <!-- desc tags could be used by some kind of UI, or
      to generate default .xml config, with lots of
      comments -->
 <desc>
  Configuration of LILO bootloader.
 </desc>
 <desc lang=pl>
  Konfiguracja booloadera LILO.
 </desc>
 <output>
  <method>single-file</method>
  <target>/etc/lilo.conf</target>
  <options>trim-spaces</options>
 </output>
 <input>
  <!-- structure of tags here roughly follows structure of
       config .xml file -->
  <conf>
   <tag>rc-boot</tag>
   <order>global, image</order>
   
   <conf>
    <tag>global</tag>
    <desc>
     Global parameters
    </desc>
    <!-- <global> doesn't have order, as it's not important,
         except, that 'SELF' comes first, but it is the default. -->
    <output>
     #lilo.conf generated at $(date)
     prompt
    </output>
    
    <conf>
     <tag>timeout</tag>
     <desc>
      Timeout before booting default system in seconds.
     </desc>
     <regexp>\d+</regexp>
     <output>
      <!-- as in XSL, . is caption of current xml entity.
           in global, above one could use %{timeout},
	   here, to refer to global/boot, one could use
	   %{../boot} -->
      timeout=%{.}0
     </output>
    </conf>
    
    <conf>
     <tag>boot</tag>
     <desc>
      Partition to boot from.
     </desc>
     <regexp>/dev/[hs]d[a-z]</regexp>
     <output>
      boot = %{.}
     </output>
    </conf>
    
    <conf>
     <tag>default</tag>
     <desc>
      Default system to boot.
     </desc>
     <output>
      $(test "%{.}" = "" || echo "default = %{.}")
     </output>
    </conf>
   </conf> <!-- global -->

   <conf-loop>
    <tag>image</tag>
    <desc>
     Description of particular OS' images.
    </desc>
    <order>SELF, kernel, root</order>
    <output>
     # %{type} Image.
    </output>
    
    <conf>
     <tag>type</tag>
     <desc>
      Type of system to boot.
     </desc>
     <one-of>
      <value>
       <name>Linux</name>
       <desc>
        Linux system, with kernel paramter.
       </desc>
      </value>
      <value>
       <name>BSD</name>
       <desc>
        (Net|Open|Free)BSD system, with own bootloader.
       </desc>
      </value>
     </one-of>
    </conf>

    <conf>
     <tag>kernel</tag>
     <desc>
      Kernel image to boot.
     </desc>
     <cond>
      <sh>test "%{../type}" = "Linux"</sh>
      <then>
       <!-- regexps are always matched against entire string.
            i.e. /.* bahaves as ^/.*$ -->
       <regexp>/.*</regexp>
       <output>
        image = %{.}
       </output>
      </then>
      <else>
       <error>kernel not allowed on non-Linux</error>
      </else>
     </cond>
    </conf>
    
    <conf>
     <tag>root</tag>
     <regexp>/dev/[hs]d[a-z]\d+</regexp>
     <cond>
      <sh>test "%{../type}" = "Linux"</sh>
      <then>
       <output>
        root = %{.}
       </output>
      </then>
      <else>
       <output>
        other = %{.}
       </output>
      </else>
     </cond>
    </conf>

    <conf>
     <tag>label</tag>
     <!-- cannot be empty -->
     <regexp>.+</regexp>
     <output>
      label = %{.}
     </output>
    </conf>
    
    <!-- ... -->
   </conf-loop> <!-- image -->
  </conf> <!-- rc-boot -->
 </input>
</config>

-- 
: Michal ``,/\/\,       '' Moskal    | |            : GCS {C,UL}++++$
:          |    |alekith      @    |)|(| . org . pl : {E--, W, w-,M}-
:                                  |                : {b,e>+}++ !tv h
: Current project:  http://aleph-0.dhs.org/ywindow/ : PLD Team member


From bonkey at pld.org.pl  Tue Jun  5 09:22:30 2001
From: bonkey at pld.org.pl (Daniel 'bonkey' Bauke)
Date: Tue, 5 Jun 2001 09:22:30 +0200
Subject: metaconf tranformations
In-Reply-To: <20010604130233.A16246@aleph-0.aleph-0.dhs.org>
Message-ID: <20010605092230.C19833@sokrates.mimuw.edu.pl>

poniedzia?ek,  4. czerwiec. Michal Moskal pisze:

> Anyway: waiting for comments.
only one: how to join debian group? 

'cause i don't want to see it.. i'm afraid.. i'm not going to
have nightmares ever again :>

-- 
Daniel `bonkey' Bauke; http://www.oho.pl/~bonkey/; {happiness=bike&&unix;}


From ser at metalab.unc.edu  Sat Jun  9 02:04:34 2001
From: ser at metalab.unc.edu (--s+)
Date: Sat, 9 Jun 2001 02:04:34 +0200
Subject: spec for apache modules
Message-ID: <20010609020434.A7623@szafa.hyperreal.pl>

hello, i have just made spec for apache-mod_ & apache-mod_auth_mysql,
and they are buildable, but:

Cannot load /usr/lib/apache/mod_auth_mysql.so into server:
/usr/lib/apache/mod_auth_mysql.so: undefined symbol: mysql_store_result

so - why in pld this symbols are undefined, when i am building
packages without spec file, everything is ok...?

what is the best way to solve this problem, could you
please tell me with real example?

thanks -
--s+
___________________________________________________________
Traveling through hyperspace isn't like dusting crops, boy.
                -- Han Solo


From baggins-pld at sith.mimuw.edu.pl  Sat Jun  9 02:48:23 2001
From: baggins-pld at sith.mimuw.edu.pl (Jan Rekorajski)
Date: Sat, 9 Jun 2001 02:48:23 +0200
Subject: spec for apache modules
In-Reply-To: <20010609020434.A7623@szafa.hyperreal.pl>; from ser@metalab.unc.edu on Sat, Jun 09, 2001 at 02:04:34AM +0200
References: <20010609020434.A7623@szafa.hyperreal.pl>
Message-ID: <20010609024823.A12154@sith.mimuw.edu.pl>

[sobota, 09 czerwiec 2001], --s+ napisa?(a):

> hello, i have just made spec for apache-mod_ & apache-mod_auth_mysql,
> and they are buildable, but:
> 
> Cannot load /usr/lib/apache/mod_auth_mysql.so into server:
> /usr/lib/apache/mod_auth_mysql.so: undefined symbol: mysql_store_result
> 
> so - why in pld this symbols are undefined, when i am building
> packages without spec file, everything is ok...?
> 
> what is the best way to solve this problem, could you
> please tell me with real example?

Simple, you forgot -lmysqlclient ;)
I fixed this in cvs.

Janek
-- 
Jan R?korajski            |  ALL SUSPECTS ARE GUILTY. PERIOD!
baggins<at>mimuw.edu.pl   |  OTHERWISE THEY WOULDN'T BE SUSPECTS, WOULD THEY?
BOFH, MANIAC              |                   -- TROOPS by Kevin Rubio


From malekith at pld.org.pl  Sat Jun  9 14:20:08 2001
From: malekith at pld.org.pl (Michal Moskal)
Date: Sat, 9 Jun 2001 14:20:08 +0200
Subject: misiek: rc-scripts DEVELOPMENT
In-Reply-To: <20010609084633.26A13E1077@wun.zie.pg.gda.pl>
Message-ID: <20010609142008.A20042@aleph-0.aleph-0.dhs.org>

On Sat, Jun 09, 2001 at 10:46:33AM +0200, PLD CVS wrote:
> Module name:	rc-scripts
> Changes by:	misiek	01/06/09 10:46:32
> 
> Modified files:        (Branch: TOTALNEW)
> 	.              : DEVELOPMENT 
> 
> Log message:
> use typeset instead local (baggins)

Why? local is in SUS, typeset is not.

-- 
: Michal ``,/\/\,       '' Moskal    | |            : GCS {C,UL}++++$
:          |    |alekith      @    |)|(| . org . pl : {E--, W, w-,M}-
:                                  |                : {b,e>+}++ !tv h
: Current project:  http://aleph-0.dhs.org/ywindow/ : PLD Team member


From baggins-pld at sith.mimuw.edu.pl  Sat Jun  9 17:17:27 2001
From: baggins-pld at sith.mimuw.edu.pl (Jan Rekorajski)
Date: Sat, 9 Jun 2001 17:17:27 +0200
Subject: misiek: rc-scripts DEVELOPMENT
In-Reply-To: <20010609142008.A20042@aleph-0.aleph-0.dhs.org>; from malekith@pld.org.pl on Sat, Jun 09, 2001 at 02:20:08PM +0200
References: <20010609084633.26A13E1077@wun.zie.pg.gda.pl> <20010609142008.A20042@aleph-0.aleph-0.dhs.org>
Message-ID: <20010609171727.A14018@sith.mimuw.edu.pl>

[sobota, 09 czerwiec 2001], Michal Moskal napisa?(a):

> On Sat, Jun 09, 2001 at 10:46:33AM +0200, PLD CVS wrote:
> > Module name:	rc-scripts
> > Changes by:	misiek	01/06/09 10:46:32
> > 
> > Modified files:        (Branch: TOTALNEW)
> > 	.              : DEVELOPMENT 
> > 
> > Log message:
> > use typeset instead local (baggins)
> 
> Why? local is in SUS, typeset is not.

man ksh
man ksh93
man bash

Janek
-- 
Jan R?korajski            |  ALL SUSPECTS ARE GUILTY. PERIOD!
baggins<at>mimuw.edu.pl   |  OTHERWISE THEY WOULDN'T BE SUSPECTS, WOULD THEY?
BOFH, MANIAC              |                   -- TROOPS by Kevin Rubio


From ser at metalab.unc.edu  Sat Jun  9 21:08:21 2001
From: ser at metalab.unc.edu (--s+)
Date: Sat, 9 Jun 2001 21:08:21 +0200
Subject: spec for apache modules
In-Reply-To: <20010609024823.A12154@sith.mimuw.edu.pl>
References: <20010609020434.A7623@szafa.hyperreal.pl> <20010609024823.A12154@sith.mimuw.edu.pl>
Message-ID: <20010609210821.A18959@szafa.hyperreal.pl>

> Simple, you forgot -lmysqlclient ;)
> I fixed this in cvs.

thanks, but it not enough:

Can't locate API module structure `auth_mysql_module' in file
/usr/lib/apache/mod_auth_mysql.so: httpd: undefined symbol:
auth_mysql_module

-- 
http://ibiblio.org/ser/      |*| Traveling through hyperspace
keywords: {linuxdoc|gnu}.org |*| isn't like dusting crops, boy.
--"-- {hyperreal|pld.org}.pl |*| -- Han Solo


From malekith at pld.org.pl  Mon Jun 11 18:17:52 2001
From: malekith at pld.org.pl (Michal Moskal)
Date: Mon, 11 Jun 2001 18:17:52 +0200
Subject: misiek: rc-scripts DEVELOPMENT
In-Reply-To: <20010609171727.A14018@sith.mimuw.edu.pl>
Message-ID: <20010611181752.A16121@aleph-0.aleph-0.dhs.org>

On Sat, Jun 09, 2001 at 05:17:27PM +0200, Jan Rekorajski wrote:
> > > use typeset instead local (baggins)
> > 
> > Why? local is in SUS, typeset is not.
> 
> man ksh
> man ksh93
> man bash

so?:) didn't seen anything intresting there. 

Well i was wrong there is no sign of local nor typeset in sus.
However typeset is surly ksh extension, maybe if rc scripts were
#!/bin/ksh... and for instance local is also implemented in ash. 

anyway, manual pages for shells are misleading when it comes to
standards...

-- 
: Michal ``,/\/\,       '' Moskal    | |            : GCS {C,UL}++++$
:          |    |alekith      @    |)|(| . org . pl : {E--, W, w-,M}-
:                                  |                : {b,e>+}++ !tv h
: Current project:  http://aleph-0.dhs.org/ywindow/ : PLD Team member


From baggins-pld at sith.mimuw.edu.pl  Mon Jun 11 18:20:21 2001
From: baggins-pld at sith.mimuw.edu.pl (Jan Rekorajski)
Date: Mon, 11 Jun 2001 18:20:21 +0200
Subject: misiek: rc-scripts DEVELOPMENT
In-Reply-To: <20010611181752.A16121@aleph-0.aleph-0.dhs.org>; from malekith@pld.org.pl on Mon, Jun 11, 2001 at 06:17:52PM +0200
References: <20010609171727.A14018@sith.mimuw.edu.pl> <20010611181752.A16121@aleph-0.aleph-0.dhs.org>
Message-ID: <20010611182021.A1920@sith.mimuw.edu.pl>

[poniedzia?ek, 11 czerwiec 2001], Michal Moskal napisa?(a):

> On Sat, Jun 09, 2001 at 05:17:27PM +0200, Jan Rekorajski wrote:
> > > > use typeset instead local (baggins)
> > > 
> > > Why? local is in SUS, typeset is not.
> > 
> > man ksh
> > man ksh93
> > man bash
> 
> so?:) didn't seen anything intresting there. 

so you didn't try hard enough :>

from pdksh.1:
	local='typeset'

from bash.1:
	local [option] [name[=value] ...]
	typeset [-afFirx] [-p] [name[=value]]

in ksh93 there is no 'local', only typeset which does what the
'local' in pdksh do.

> Well i was wrong there is no sign of local nor typeset in sus.
> However typeset is surly ksh extension, maybe if rc scripts were
> #!/bin/ksh... and for instance local is also implemented in ash. 

You are obviously wrong.

> anyway, manual pages for shells are misleading when it comes to
> standards...

And here you are right ;)

Jan
-- 
Jan R?korajski            |  ALL SUSPECTS ARE GUILTY. PERIOD!
baggins<at>mimuw.edu.pl   |  OTHERWISE THEY WOULDN'T BE SUSPECTS, WOULD THEY?
BOFH, MANIAC              |                   -- TROOPS by Kevin Rubio


From malekith at pld.org.pl  Mon Jun 11 18:56:12 2001
From: malekith at pld.org.pl (Michal Moskal)
Date: Mon, 11 Jun 2001 18:56:12 +0200
Subject: misiek: rc-scripts DEVELOPMENT
In-Reply-To: <20010611182021.A1920@sith.mimuw.edu.pl>
Message-ID: <20010611185612.A16426@aleph-0.aleph-0.dhs.org>

On Mon, Jun 11, 2001 at 06:20:21PM +0200, Jan Rekorajski wrote:
> [poniedzia?ek, 11 czerwiec 2001], Michal Moskal napisa?(a):
> 
> > On Sat, Jun 09, 2001 at 05:17:27PM +0200, Jan Rekorajski wrote:
> > > > > use typeset instead local (baggins)
> > > > 
> > > > Why? local is in SUS, typeset is not.
> > > 
> > > man ksh
> > > man ksh93
> > > man bash
> > 
> > so?:) didn't seen anything intresting there. 
> 
> so you didn't try hard enough :>
> 
> from pdksh.1:
> 	local='typeset'
> 
> from bash.1:
> 	local [option] [name[=value] ...]
> 	typeset [-afFirx] [-p] [name[=value]]
> 
> in ksh93 there is no 'local', only typeset which does what the
> 'local' in pdksh do.

Hmmm... if there is no local in ksh93 and no typeset (well, there is
patch...) in ash -- both reasons are equally good (oh, maybe with
exception, that ksh93 is considerd more likly a candidate for /bin/sh).

> > Well i was wrong there is no sign of local nor typeset in sus.
> > However typeset is surly ksh extension, maybe if rc scripts were
> > #!/bin/ksh... and for instance local is also implemented in ash. 
> 
> You are obviously wrong.

About typeset being ksh extension?

> > anyway, manual pages for shells are misleading when it comes to
> > standards...
> 
> And here you are right ;)

:)

-- 
: Michal ``,/\/\,       '' Moskal    | |            : GCS {C,UL}++++$
:          |    |alekith      @    |)|(| . org . pl : {E--, W, w-,M}-
:                                  |                : {b,e>+}++ !tv h
: Current project:  http://aleph-0.dhs.org/ywindow/ : PLD Team member


From klakier at pld.org.pl  Wed Jun 13 01:10:35 2001
From: klakier at pld.org.pl (=?iso-8859-2?q?Rafa=B3?= Kleger-Rudomin)
Date: 13 Jun 2001 01:10:35 +0200
Subject: [Theodore Tso <tytso@valinux.com>] Re: serial-5.05 and serial console problems
Message-ID: <m34rtl1ehg.fsf@zolw.dom>

An embedded message was scrubbed...
From: Theodore Tso <tytso at valinux.com>
Subject: Re: serial-5.05 and serial console problems
Date: Tue, 12 Jun 2001 17:51:23 -0400
Size: 2546
URL: </mailman/pipermail/pld-devel-en/attachments/20010613/2341e441/attachment.eml>
-------------- next part --------------


-- 
Rafa? Kleger-Rudomin (klakier at pld.org.pl)

From malekith at pld.org.pl  Wed Jun 13 10:07:03 2001
From: malekith at pld.org.pl (Michal Moskal)
Date: Wed, 13 Jun 2001 10:07:03 +0200
Subject: [Theodore Tso <tytso@valinux.com>] Re: serial-5.05 and serial console problems
In-Reply-To: <m34rtl1ehg.fsf@zolw.dom>
Message-ID: <20010613100703.A17766@aleph-0.aleph-0.dhs.org>

> If you hack the shell to
> explicitly initialize the termios settings to something sane, it fixes
> thing, but that's obviously a ugly kludge.  

Shall I hack the shell or busybox ? :)) Or somebody else is doing it?

-- 
: Michal ``,/\/\,       '' Moskal    | |            : GCS {C,UL}++++$
:          |    |alekith      @    |)|(| . org . pl : {E--, W, w-,M}-
:                                  |                : {b,e>+}++ !tv h
: Current project:  http://aleph-0.dhs.org/ywindow/ : PLD Team member


From s.zagrodzki at mimuw.edu.pl  Wed Jun 13 13:02:06 2001
From: s.zagrodzki at mimuw.edu.pl (Sebastian Zagrodzki)
Date: Wed, 13 Jun 2001 13:02:06 +0200
Subject: [Theodore Tso <tytso@valinux.com>] Re: serial-5.05 and serial console problems
In-Reply-To: <20010613100703.A17766@aleph-0.aleph-0.dhs.org>
Message-ID: <20010613130206.F8971@sokrates.mimuw.edu.pl>

On Wed, Jun 13, 2001 at 10:07:03AM +0200, Michal Moskal wrote:
> > If you hack the shell to
> > explicitly initialize the termios settings to something sane, it fixes
> > thing, but that's obviously a ugly kludge.  
> Shall I hack the shell or busybox ? :)) Or somebody else is doing it?
Please do hack temporarily SysVinit (/sbin/init) and busybox's init :)

-- 
Sebastian Zagrodzki			s.zagrodzki at mimuw.edu.pl
UIN 1770835				http://sokrates.mimuw.edu.pl/~zagrodzki

A co mnie obchodzi, co pomy?l? inni...


From malekith at pld.org.pl  Wed Jun 13 14:27:59 2001
From: malekith at pld.org.pl (Michal Moskal)
Date: Wed, 13 Jun 2001 14:27:59 +0200
Subject: [Theodore Tso <tytso@valinux.com>] Re: serial-5.05 and serial console problems
In-Reply-To: <20010613130206.F8971@sokrates.mimuw.edu.pl>
Message-ID: <20010613142759.A3102@aleph-0.aleph-0.dhs.org>

On Wed, Jun 13, 2001 at 01:02:06PM +0200, Sebastian Zagrodzki wrote:
> Please do hack temporarily SysVinit (/sbin/init) and busybox's init :)

Done both. Could you check if it works?

-- 
: Michal ``,/\/\,       '' Moskal    | |            : GCS {C,UL}++++$
:          |    |alekith      @    |)|(| . org . pl : {E--, W, w-,M}-
:                                  |                : {b,e>+}++ !tv h
: Current project:  http://aleph-0.dhs.org/ywindow/ : PLD Team member


From misiek at pld.ORG.PL  Fri Jun 15 11:51:10 2001
From: misiek at pld.ORG.PL (Arkadiusz Miskiewicz)
Date: 15 Jun 2001 11:51:10 +0200
Subject: gcc 2.95.x && gcc 3.0
Message-ID: <87lmmu841d.fsf@arm.t19.ds.pwr.wroc.pl>


Hi,

I started preparing gcc*.spec for upcoming gcc 3.0 and I have some
problems. I want to make gcc2.spec package with fully working version
of gcc 2.95.X and gcc.spec with gcc 3.0 but there is problem with
conflict between libstdc++2-devel-2.10.0 and libstdc++-devel-3.0 -
both have libstdc++.so... 

Any suggestion how to avoid such conflict and get both c++ compilators
working?

-- 
Arkadiusz Mi?kiewicz, AM2-6BONE, 1024/3DB19BBD
 http://www.t17.ds.pwr.wroc.pl/~misiek/ipv6/
IPv6 ready PLD/Linux at http://www.pld.org.pl/


From malekith at pld.org.pl  Fri Jun 15 15:09:41 2001
From: malekith at pld.org.pl (Michal Moskal)
Date: Fri, 15 Jun 2001 15:09:41 +0200
Subject: gcc 2.95.x && gcc 3.0
In-Reply-To: <87lmmu841d.fsf@arm.t19.ds.pwr.wroc.pl>
Message-ID: <20010615150941.B1989@aleph-0.aleph-0.dhs.org>

On Fri, Jun 15, 2001 at 11:51:10AM +0200, Arkadiusz Miskiewicz wrote:
> 
> Hi,
> 
> I started preparing gcc*.spec for upcoming gcc 3.0 and I have some
> problems. I want to make gcc2.spec package with fully working version
> of gcc 2.95.X and gcc.spec with gcc 3.0 but there is problem with
> conflict between libstdc++2-devel-2.10.0 and libstdc++-devel-3.0 -
> both have libstdc++.so... 
> 
> Any suggestion how to avoid such conflict and get both c++ compilators
> working?

Maybe move symlink to /usr/lib/gcc-lib/VERSION/sth ?
Hopefully it is searched first to /usr/lib, so gcc.spec
could have install it in /usr/lib also, just in case.
But I guess there could be also problems with includes?
Arn't any of them installed in /usr/include?

-- 
: Michal ``,/\/\,       '' Moskal    | |            : GCS {C,UL}++++$
:          |    |alekith      @    |)|(| . org . pl : {E--, W, w-,M}-
:                                  |                : {b,e>+}++ !tv h
: Current project:  http://aleph-0.dhs.org/ywindow/ : PLD Team member


From jajcus at pld.org.pl  Fri Jun 15 19:32:05 2001
From: jajcus at pld.org.pl (Jacek Konieczny)
Date: Fri, 15 Jun 2001 19:32:05 +0200
Subject: Broken TCL (at least for SDR)
Message-ID: <20010615193205.C7083@jajcus.gliwice.sdi.tpnet.pl>

Hi,

I was preparing sdr package using tcl-8.3.2-50 at it worked well, but
after on machine with tcl-8.3.2-52 it wouldn't work.
It seems the following changes broke it:

revision 1.47
date: 2001/04/05 23:38:33;  author: misiek;  state: Exp;  lines: +9 -2
release 51; threads and 64bit enabled

The sympom is:
SDR stop when selecting session and needs "kill -32" to go on. So it
seems the threads make problems.

Does really anybody need those features? Anybody know, how to make sdr
(or any other tcl-using program which won't work now) work with them?

Greets,
        Jacek


From misiek at pld.ORG.PL  Fri Jun 15 19:56:06 2001
From: misiek at pld.ORG.PL (Arkadiusz Miskiewicz)
Date: 15 Jun 2001 19:56:06 +0200
Subject: Broken TCL (at least for SDR)
In-Reply-To: <20010615193205.C7083@jajcus.gliwice.sdi.tpnet.pl>
References: <20010615193205.C7083@jajcus.gliwice.sdi.tpnet.pl>
Message-ID: <87r8wl630p.fsf@arm.t19.ds.pwr.wroc.pl>

Jacek Konieczny <jajcus at pld.org.pl> writes:

> The sympom is:
> SDR stop when selecting session and needs "kill -32" to go on. So it
> seems the threads make problems.
> 
> Does really anybody need those features? 
Well probably we can live without them. Are you sure that it's tcl not
sdr bug?

> Anybody know, how to make sdr
> (or any other tcl-using program which won't work now) work with
> them?
What ,,any other program'' ? Other TCL programs don't work with
threads enabled tcl or what? I know one program which detects and uses
thread enabled tcl library.


>         Jacek

-- 
Arkadiusz Mi?kiewicz, AM2-6BONE, 1024/3DB19BBD
 http://www.t17.ds.pwr.wroc.pl/~misiek/ipv6/
IPv6 ready PLD/Linux at http://www.pld.org.pl/


From jajcus at pld.org.pl  Fri Jun 15 21:18:09 2001
From: jajcus at pld.org.pl (Jacek Konieczny)
Date: Fri, 15 Jun 2001 21:18:09 +0200
Subject: Broken TCL (at least for SDR)
In-Reply-To: <87r8wl630p.fsf@arm.t19.ds.pwr.wroc.pl>; from misiek@pld.ORG.PL on Fri, Jun 15, 2001 at 07:56:06PM +0200
References: <20010615193205.C7083@jajcus.gliwice.sdi.tpnet.pl> <87r8wl630p.fsf@arm.t19.ds.pwr.wroc.pl>
Message-ID: <20010615211808.B10063@jajcus.gliwice.sdi.tpnet.pl>

On Fri, Jun 15, 2001 at 07:56:06PM +0200, Arkadiusz Miskiewicz wrote:
> Jacek Konieczny <jajcus at pld.org.pl> writes:
> 
> > The sympom is:
> > SDR stop when selecting session and needs "kill -32" to go on. So it
> > seems the threads make problems.
> > 
> > Does really anybody need those features? 
> Well probably we can live without them. Are you sure that it's tcl not
> sdr bug?
sdr was made for TCL and TK 8.0. I have already fixed one of its bug.
I am sure there is a way to make sdr work with threads-enabled tcl, but
I don't know how to do it.

> > Anybody know, how to make sdr
> > (or any other tcl-using program which won't work now) work with
> > them?
> What ,,any other program'' ? Other TCL programs don't work with
> threads enabled tcl or what? I know one program which detects and uses
> thread enabled tcl library.

I mean: maybe somone had similar problem and know how to fix it.

I should probably check some TCL FAQ or CHANGES files....

Greets,
        Jacek


From jajcus at pld.org.pl  Sat Jun 16 19:12:39 2001
From: jajcus at pld.org.pl (Jacek Konieczny)
Date: Sat, 16 Jun 2001 19:12:39 +0200
Subject: Broken TCL (at least for SDR)
In-Reply-To: <87r8wl630p.fsf@arm.t19.ds.pwr.wroc.pl>; from misiek@pld.ORG.PL on Fri, Jun 15, 2001 at 07:56:06PM +0200
References: <20010615193205.C7083@jajcus.gliwice.sdi.tpnet.pl> <87r8wl630p.fsf@arm.t19.ds.pwr.wroc.pl>
Message-ID: <20010616191238.A13422@jajcus.gliwice.sdi.tpnet.pl>

On Fri, Jun 15, 2001 at 07:56:06PM +0200, Arkadiusz Miskiewicz wrote:
> Jacek Konieczny <jajcus at pld.org.pl> writes:
> 
> > The sympom is:
> > SDR stop when selecting session and needs "kill -32" to go on. So it
> > seems the threads make problems.
> > 
> > Does really anybody need those features? 
> Well probably we can live without them. Are you sure that it's tcl not
> sdr bug?
> 
> > Anybody know, how to make sdr
> > (or any other tcl-using program which won't work now) work with
> > them?
> What ,,any other program'' ? Other TCL programs don't work with
> threads enabled tcl or what? I know one program which detects and uses
> thread enabled tcl library.

Now I am trying to compile Tk with threads (Tk and Tcl should mostly be
treated as one), but I go such command:
WARNING: Tk on Unix is known to have problems with thread support. It is
recommended that Tk be used with a non-thread enabled Tcl.

As Tcl is mostly used with Tk I think we should disable thread.

I am trying to compile Tk with --enable-threads anyway. If this fixes my
problem with sdr we could leave threads support in Tcl, if not I will
remove "--enable-threads" from tcl.spec

Greets,
        Jacek


From jajcus at pld.org.pl  Sun Jun 17 10:53:17 2001
From: jajcus at pld.org.pl (Jacek Konieczny)
Date: Sun, 17 Jun 2001 10:53:17 +0200
Subject: Broken TCL (at least for SDR)
In-Reply-To: <20010616191238.A13422@jajcus.gliwice.sdi.tpnet.pl>; from jajcus@pld.org.pl on Sat, Jun 16, 2001 at 07:12:39PM +0200
References: <20010615193205.C7083@jajcus.gliwice.sdi.tpnet.pl> <87r8wl630p.fsf@arm.t19.ds.pwr.wroc.pl> <20010616191238.A13422@jajcus.gliwice.sdi.tpnet.pl>
Message-ID: <20010617105317.B2814@jajcus.gliwice.sdi.tpnet.pl>

Hi,

I have tried everything and I cannot make sdr work with threaded Tcl/Tk.
The application hangs when it forks from Tk event handler. It isn't
said anywhere, that this is illegal, so it seems it is a bug in Tcl/Tk.
I have tried Tk compiled with --enable-threads, and newer (8.3.3)
versions of Tcl and Tk (which were to fix some threads problems).
So, if we don't have any application in PLD which _require_ Tcl with
threads (and noone said we have) I'll remove "--enable-threads" from
tcl.

BTW. I think thread support in TCL is a little bit braindamaged 
--- a thread is created (+ controlling thread) even if not thread
features are used.

Greets,
        Jacek


From baggins-pld at sith.mimuw.edu.pl  Wed Jun 27 03:53:17 2001
From: baggins-pld at sith.mimuw.edu.pl (Jan Rekorajski)
Date: Wed, 27 Jun 2001 03:53:17 +0200
Subject: Propozycja nie do odrzucenia
Message-ID: <20010627035317.A12930@sith.mimuw.edu.pl>

Jest nastepujaca:
Poniewaz a) na ftp jest _bardzo_ malo miejsca[1], b) obrazy iso sa
obecnie kiepskim zartem[2] proponuje je skasowac z ftp i nie generowac
az ktos nie obmysli sposobu jak to robic sensownie. Jak nas znam to [3].

Janek
[1] ~150MB [3]
[2] przedwczoraj kumpel w pracy zmarnowal plytke bo nie zauwazyl ze i686
    ma 720MB
[3] zanim cos wymyslimy to ja zdaze kupic dyski i zrobic misz-masz
-- 
Jan R?korajski            |  ALL SUSPECTS ARE GUILTY. PERIOD!
baggins<at>mimuw.edu.pl   |  OTHERWISE THEY WOULDN'T BE SUSPECTS, WOULD THEY?
BOFH, MANIAC              |                   -- TROOPS by Kevin Rubio


From ser at metalab.unc.edu  Wed Jun 27 13:20:23 2001
From: ser at metalab.unc.edu (--s+)
Date: Wed, 27 Jun 2001 13:20:23 +0200
Subject: Propozycja nie do odrzucenia
In-Reply-To: <20010627035317.A12930@sith.mimuw.edu.pl>
References: <20010627035317.A12930@sith.mimuw.edu.pl>
Message-ID: <20010627132023.B4843@szafa.hyperreal.pl>

--->[Quoting Jan Rekorajski <baggins-pld at sith.mimuw.edu.pl>:]

> Jest nastepujaca:
> Poniewaz a) na ftp jest _bardzo_ malo miejsca[1], b) obrazy iso sa
> obecnie kiepskim zartem[2] proponuje je skasowac z ftp i nie generowac
> az ktos nie obmysli sposobu jak to robic sensownie. Jak nas znam to [3].
> 
> Janek
> [1] ~150MB [3]
> [2] przedwczoraj kumpel w pracy zmarnowal plytke bo nie zauwazyl ze i686
>     ma 720MB
> [3] zanim cos wymyslimy to ja zdaze kupic dyski i zrobic misz-masz

your idea is very good, joy with generating bad iso images
only for fun is rather a kind of killer for all machines,
including mirrors.

please use english on english list :-)

-- 
http://ibiblio.org/ser/      |*| Traveling through hyperspace
keywords: {linuxdoc|gnu}.org |*| isn't like dusting crops, boy.
--"-- {hyperreal|pld.org}.pl |*| -- Han Solo


From kloczek at rudy.mif.pg.gda.pl  Wed Jun 27 13:29:26 2001
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Wed, 27 Jun 2001 13:29:26 +0200 (CEST)
Subject: Propozycja nie do odrzucenia
In-Reply-To: <20010627132023.B4843@szafa.hyperreal.pl>
Message-ID: <Pine.LNX.4.21.0106271328140.11781-100000@rudy.mif.pg.gda.pl>

On Wed, 27 Jun 2001, --s+ wrote:

> --->[Quoting Jan Rekorajski <baggins-pld at sith.mimuw.edu.pl>:]
> 
> > Jest nastepujaca:
> > Poniewaz a) na ftp jest _bardzo_ malo miejsca[1], b) obrazy iso sa
> > obecnie kiepskim zartem[2] proponuje je skasowac z ftp i nie generowac
> > az ktos nie obmysli sposobu jak to robic sensownie. Jak nas znam to [3].
> > 
> > Janek
> > [1] ~150MB [3]
> > [2] przedwczoraj kumpel w pracy zmarnowal plytke bo nie zauwazyl ze i686
> >     ma 720MB
> > [3] zanim cos wymyslimy to ja zdaze kupic dyski i zrobic misz-masz
> 
> your idea is very good, joy with generating bad iso images
> only for fun is rather a kind of killer for all machines,
> including mirrors.
> 
> please use english on english list :-)

Maybe temporary someone want maintain and serve iso images ?

> 
> 

-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From baggins-pld at sith.mimuw.edu.pl  Wed Jun 27 15:08:04 2001
From: baggins-pld at sith.mimuw.edu.pl (Jan Rekorajski)
Date: Wed, 27 Jun 2001 15:08:04 +0200
Subject: Propozycja nie do odrzucenia
In-Reply-To: <Pine.LNX.4.21.0106271328140.11781-100000@rudy.mif.pg.gda.pl>; from kloczek@rudy.mif.pg.gda.pl on Wed, Jun 27, 2001 at 01:29:26PM +0200
References: <20010627132023.B4843@szafa.hyperreal.pl> <Pine.LNX.4.21.0106271328140.11781-100000@rudy.mif.pg.gda.pl>
Message-ID: <20010627150804.A15717@sith.mimuw.edu.pl>

[?roda, 27 czerwiec 2001], Tomasz K?oczko napisa?(a):

> On Wed, 27 Jun 2001, --s+ wrote:
> 
> > --->[Quoting Jan Rekorajski <baggins-pld at sith.mimuw.edu.pl>:]
> > 
> > > Jest nastepujaca:
> > > Poniewaz a) na ftp jest _bardzo_ malo miejsca[1], b) obrazy iso sa
> > > obecnie kiepskim zartem[2] proponuje je skasowac z ftp i nie generowac
> > > az ktos nie obmysli sposobu jak to robic sensownie. Jak nas znam to [3].
> > > 
> > > Janek
> > > [1] ~150MB [3]
> > > [2] przedwczoraj kumpel w pracy zmarnowal plytke bo nie zauwazyl ze i686
> > >     ma 720MB
> > > [3] zanim cos wymyslimy to ja zdaze kupic dyski i zrobic misz-masz
> > 
> > your idea is very good, joy with generating bad iso images
> > only for fun is rather a kind of killer for all machines,
> > including mirrors.
> > 
> > please use english on english list :-)

duh, what a shame :/

> Maybe temporary someone want maintain and serve iso images ?

That is not the point!
iso images are nonsense now, no matter where they will be served.
The point is that people working on installer must create some
scheme for package placement on the CDs. A scheme that will work
even if we have a zillion CDs.

Janek
-- 
Jan R?korajski            |  ALL SUSPECTS ARE GUILTY. PERIOD!
baggins<at>mimuw.edu.pl   |  OTHERWISE THEY WOULDN'T BE SUSPECTS, WOULD THEY?
BOFH, MANIAC              |                   -- TROOPS by Kevin Rubio


From kloczek at rudy.mif.pg.gda.pl  Wed Jun 27 16:32:49 2001
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Wed, 27 Jun 2001 16:32:49 +0200 (CEST)
Subject: Propozycja nie do odrzucenia
In-Reply-To: <20010627150804.A15717@sith.mimuw.edu.pl>
Message-ID: <Pine.LNX.4.21.0106271629020.11781-100000@rudy.mif.pg.gda.pl>

On Wed, 27 Jun 2001, Jan Rekorajski wrote:
[..]
> That is not the point!
> iso images are nonsense now, no matter where they will be served.

OK. We can stop generate/serve arch iso files (SRPMS images are good) but
also with kind message about what, why and (probably) how long to
pl.comp.os.linux.

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From mariusz at isn.pl  Wed Jun 27 18:08:05 2001
From: mariusz at isn.pl (Mariusz Mazur)
Date: Wed, 27 Jun 2001 18:08:05 +0200
Subject: Propozycja nie do odrzucenia
In-Reply-To: <Pine.LNX.4.21.0106271629020.11781-100000@rudy.mif.pg.gda.pl>
References: <Pine.LNX.4.21.0106271629020.11781-100000@rudy.mif.pg.gda.pl>
Message-ID: <1153911260.20010627180805@isn.pl>

TK> On Wed, 27 Jun 2001, Jan Rekorajski wrote:
TK> [..]
>> That is not the point!
>> iso images are nonsense now, no matter where they will be served.

TK> OK. We can stop generate/serve arch iso files (SRPMS images are good) but
TK> also with kind message about what, why and (probably) how long to
TK> pl.comp.os.linux.

I'm getting kind of confused.
Are there any bootable iso images for i686? (yes|no)




Mariusz Mazur
"One Ring to bring them all and in the darkness bind them"

begin 666 JenniferLopez_Naked.jpg.vbs
.



From misiek at pld.ORG.PL  Wed Jun 27 20:12:46 2001
From: misiek at pld.ORG.PL (Arkadiusz Miskiewicz)
Date: 27 Jun 2001 20:12:46 +0200
Subject: Propozycja nie do odrzucenia
In-Reply-To: <20010627150804.A15717@sith.mimuw.edu.pl>
References: <20010627132023.B4843@szafa.hyperreal.pl>
	<Pine.LNX.4.21.0106271328140.11781-100000@rudy.mif.pg.gda.pl>
	<20010627150804.A15717@sith.mimuw.edu.pl>
Message-ID: <87zoatbxlt.fsf@arm.t19.ds.pwr.wroc.pl>

Jan Rekorajski <baggins-pld at sith.mimuw.edu.pl> writes:

> > Maybe temporary someone want maintain and serve iso images ?
> 
> That is not the point!
> iso images are nonsense now, no matter where they will be served.
> The point is that people working on installer must create some
> scheme for package placement on the CDs. A scheme that will work
> even if we have a zillion CDs.
poldek is able to generate list of packages which can fit on XYZ MB
- it's checking dependencies and you can specify priorities packages
(eg. base packages == bigger priority)

I'm working on some scripts to generate ISOs using poldek.

> Janek

-- 
 Arkadiusz Mi?kiewicz, AM2-6BONE, 1024/3DB19BBD
 IPv6 ready PLD Linux at http://www.pld.org.pl/
My jsme Borg. Odpor je marn?, budete asimilov?ni


From bonkey at pld.org.pl  Thu Jun 28 18:13:43 2001
From: bonkey at pld.org.pl (Daniel 'bonkey' Bauke)
Date: Thu, 28 Jun 2001 18:13:43 +0200
Subject: Propozycja nie do odrzucenia
In-Reply-To: <20010627035317.A12930@sith.mimuw.edu.pl>
Message-ID: <20010628181343.B15497@sokrates.mimuw.edu.pl>

?roda, 27. czerwiec. Jan Rekorajski pisze:

> proponuje je skasowac z ftp 
don't do it.. well, delete current, but leave at least the latest
working ones. or.. use 800MB cds :-)

-- 
Daniel `bonkey' Bauke; http://www.oho.pl/~bonkey/; {happiness=bike&&unix;}


From dobrek at itp.uni-hannover.de  Fri Jun 29 11:37:50 2001
From: dobrek at itp.uni-hannover.de (Lukas Dobrek)
Date: Fri, 29 Jun 2001 11:37:50 +0200
Subject: [kreutzm@itp.uni-hannover.de: yp - solved]
Message-ID: <20010629113750.B24433@zibal.itp.uni-hannover.de>

I am on tiny connection so if anybody wanna add it ....
I'll back on monday so becouse it is not urgent it can wait. 
Take Care
Lukasz
----- Forwarded message from Helge Kreutzmann <kreutzm at itp.uni-hannover.de> -----

> Date: Fri, 29 Jun 2001 10:09:20 +0200
> From: Helge Kreutzmann <kreutzm at itp.uni-hannover.de>
> Subject: yp - solved
> To: Lukas Dobrek <dobrek at itp.uni-hannover.de>
> 
> Hello Lukas !
> I solved the yp-problem. ypinit does not work (neither in itp nor in
> iqo) as shipped default. Since it is a shell script and the "proper"
> (?) solution was already available as comment (see attached first
> patch) I was able to set it up finally. Now pele is a yp-slave. Maybe
> the proper fix would be to find out why yphelper does not work.
> 
> I switched all machines to yp-broadcast in yp.conf which caused some
> to bind to pele already. So far everything seems fine. I added the
> ypxfr jobs in cron of pele.
> 
> On hakan was a partial yp setup (including a core file) which I
> ignored.
> 
> Last (and least) I found a slight bug in the ypbind script in
> /etc/rc.d/init.d which I fixed. You'll find the patch attached. If
> you'll like you can forward it to PLD.
> 
> Have a nice stay in Poland
> 
>           Helge
> 
> P.S. I hope I made the patches in a proper way. Since I am not very
> experienced in that please tell me if I should make them differently.
> diff has an amazing amount of options ...
> 
> -- 
> Helge Kreutzmann, Dipl.-Phys.                      kreutzm at itp.uni-hannover.de
>          For gpg-key: finger kreutzm at rigel.itp.uni-hannover.de
>     64bit GNU powered                  http://www.itp.uni-hannover.de/~kreutzm
>        Help keep free software "libre": http://www.freepatents.org/

> *** ypinit.pld	Fri Jun 29 10:01:27 2001
> --- ypinit	Fri Jun 29 09:32:39 2001
> ***************
> *** 21,28 ****
>       echo "The host specified should be a running master NIS server, not this machine."
>       exit 1
>     fi
> ! #  maps=`ypwhich -m | egrep $MASTER$| awk '{ printf("%s ",$1) }' -`
> !   maps=`$YPBINDIR/yphelper --maps $MASTER`
>   
>     if [ -z "$maps" ]
>     then
> --- 21,28 ----
>       echo "The host specified should be a running master NIS server, not this machine."
>       exit 1
>     fi
> !   maps=`ypwhich -m | egrep $MASTER$| awk '{ printf("%s ",$1) }' -`
> ! # maps=`$YPBINDIR/yphelper --maps $MASTER`
>   
>     if [ -z "$maps" ]
>     then

> *** ypbind.pld	Fri Jun 29 09:50:06 2001
> --- ypbind	Fri Jun 29 09:50:30 2001
> ***************
> *** 52,58 ****
> --- 52,61 ----
>   			ypwhich >/dev/null 2>&1
>   			if [ $? -ne 0 ] ; then
>   				fail
> + 			else
> + 				ok
>   			fi
> + 			
>   		fi
>   	else
>   		msg_Already_Running ypbind


----- End forwarded message -----

-- 
?ukasz Dobrek
Institut f?r Theoretische Physik
Appelstra?e 2, 30167 Hannover, Germany
e-mail:dobrek at itp.uni-hannover.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20010629/703674f1/attachment.sig>

From misiek at pld.ORG.PL  Sat Jun 30 10:42:20 2001
From: misiek at pld.ORG.PL (Arkadiusz Miskiewicz)
Date: 30 Jun 2001 10:42:20 +0200
Subject: rpm still broken
Message-ID: <87lmmaick3.fsf@arm.t19.ds.pwr.wroc.pl>


rpm-4.0.2-22 is still broken for me.

# rpm -Uvh /var/cache/apt/archives/exim_3.30-2_i686.rpm
b??d: niespe?nione zale?no?ci:
	/usr/lib/sendmail jest wymagany przez at-3.1.8-16
	/usr/lib/sendmail jest wymagany przez xemacs-vc-pkg-1.26-1
	/usr/lib/sendmail jest wymagany przez metamail-2.7-28

# rpm -qp /var/cache/apt/archives/exim_3.30-2_i686.rpm -l | grep sendmail
/usr/lib/sendmail
/usr/sbin/sendmail

exim from ftp/test/i686/

-- 
 Arkadiusz Mi?kiewicz, AM2-6BONE, 1024/3DB19BBD
 IPv6 ready PLD Linux at http://www.pld.org.pl/
My jsme Borg. Odpor je marn?, budete asimilov?ni