From harpchad at opencsw.org  Wed Apr  1 01:47:15 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 31 Mar 2009 18:47:15 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <F8FC4A58-F0D9-4D5B-A574-080E0028CB8F@opencsw.org>

I assume RUNPATHQUOTE=3 is also valid?  I have one package where  
RPATH=/opt/csw/lib/SALIST with =2, looks like $I got expanded.

On Mar 31, 2009, at 4:04 PM, Dagobert Michelsen wrote:

> /usr/ccs/bin/dump -Lv



From harpchad at opencsw.org  Wed Apr  1 01:52:31 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 31 Mar 2009 18:52:31 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>

Phil, can checkpkg detect some of the common cases?  I'm thinking at  
least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those  
would indicate a bad RUNPATHQUOTE value.  James can probably give us  
an idea of what he's coming across the most.


From mwatters at opencsw.org  Wed Apr  1 02:24:57 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 31 Mar 2009 19:24:57 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
Message-ID: <49D2B459.3040702@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chad Harp wrote:
> Phil, can checkpkg detect some of the common cases?  I'm thinking at
> least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those would
> indicate a bad RUNPATHQUOTE value.  James can probably give us an idea
> of what he's coming across the most.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I am not sure if it should go into checkpkg or as a post-install in gar.
put a post-install that will not allow the package to continue until it
matches NOISALIST = 0 or NOISALIST = 1 dump -Lv


if we catch it at the post install it will allow a quicker fix then
waiting for the package to merge and package.



- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknStFkACgkQLrhmsXMSLxck8ACeIvN/nzrlDucP0Rr7ZnESgG6M
N4YAoOpDY06tU1O8hrHD4IyX50Y+pUXP
=JUoL
-----END PGP SIGNATURE-----


From rupert at opencsw.org  Wed Apr  1 07:25:23 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Wed, 1 Apr 2009 07:25:23 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <6af4270903312225q6f57bfc8k63e3c21ffb3379c4@mail.gmail.com>

currently i get:
~$ ssh rupert at login.opencsw.org
PTY allocation request failed on channel 0
Received disconnect from 213.178.77.178: 2: fork failed: Not enough space


On Tue, Mar 31, 2009 at 23:04, Dagobert Michelsen <dam at opencsw.org> wrote:
> Hi,
>
> James is currently in the tedious task of reviewing all
> runtime linker pathes from the packages. To make it
> easier for you as maintainer I have a summary here how
> this works in GAR and how to fix things.
>
> Runtime pathes can be checked with
> ?/usr/ccs/bin/dump -Lv <binary> | <library>
>
> A correct setting with ISALIST should look like
> [7] ? ? RUNPATH ? ? ? ? /opt/csw/lib/$ISALIST:/opt/csw/lib
> [8] ? ? RPATH ? ? ? ? ? /opt/csw/lib/$ISALIST:/opt/csw/lib
>
> A correct setting without ISALIST should look like
> [7] ? ? RUNPATH ? ? ? ? /opt/csw/lib
> [8] ? ? RPATH ? ? ? ? ? /opt/csw/lib
>
> GAR adds a runtime path with $ISALIST by default. To skip this
> entirely in GAR you can use
> ?NOISALIST = 1
> How much shell escaping is necessary depends on if the application
> uses autoconf and/or libtool. The quoting-level can be reduced with
> ?RUNPATHQUOTE = 1
> (quote for one expansion, e. g. only configure) as opposed to the default
> ?RUNPATHQUOTE = 2
> (quote for double expansion, configure and libtool).
>
>
> Best regards
>
> ?-- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From rupert at opencsw.org  Wed Apr  1 07:29:29 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Wed, 1 Apr 2009 07:29:29 +0200
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
References: <ce97543b0903260441k26550f94r96a1f6bb6c95755d@mail.gmail.com>
	<625385e30903271340r4307dd64hd691e4dac6fa46ff@mail.gmail.com>
	<20090327211430.GV47757@bolthole.com>
	<ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>
	<20090329140924.GD30131@bolthole.com>
	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>
	<20090329220733.GM69158@bolthole.com>
	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>
	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>
	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
Message-ID: <6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>

not by me anyway :)
dago, you mind giving me write access to mediawiki?

rupert.


On Tue, Mar 31, 2009 at 12:24, Maciej (Matchek) Blizinski
<maciej at opencsw.org> wrote:
> On Tue, Mar 31, 2009 at 12:39 AM, rupert THURNER <rupert at opencsw.org> wrote:
>>> When I set up "gar" on SourceForge Trac was not available as
>>> hosted application. As it is there now we can happily switch
>>> over from the standard SF bugtracker and MediaWiki to GAR.
>>>
>>> Rupert, Maciej, if you could join in on moving the stuff over :-)
>>
>> for the wiki: the mGAR things are copied. you want to keep the gar stuff as is?
>
> Can already migrated Mediawiki pages be replaced with links to their
> Trac locations?
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From dam at opencsw.org  Wed Apr  1 08:37:21 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 08:37:21 +0200
Subject: [csw-maintainers] BO buildfarm down
In-Reply-To: <49D29240.2070105@opencsw.org>
References: <49D29240.2070105@opencsw.org>
Message-ID: <B65657D6-5AB5-4253-B36E-6B15E6CC4B4A@opencsw.org>

Hi,

Am 31.03.2009 um 23:59 schrieb Roger H?kansson:
> hson at build8s :~/dev/mgar/pkg/zlib/trunk > gmake build
> bash: fork: Not enough space

Am 01.04.2009 um 07:25 schrieb rupert THURNER:
> currently i get:
> ~$ ssh rupert at login.opencsw.org
> PTY allocation request failed on channel 0
> Received disconnect from 213.178.77.178: 2: fork failed: Not enough  
> space

it seems like the T5220 was overloaded due to lack of Ram. As I
already wrote there will be a memory upgrade soon. We are working
on fixing this issue.

In the meantime please use build8s.go.opencsw.org, all accounts should
be also configured there.


Best regards

   -- Dago

From dam at opencsw.org  Wed Apr  1 09:12:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 09:12:08 +0200
Subject: [csw-maintainers] BO buildfarm down
In-Reply-To: <B65657D6-5AB5-4253-B36E-6B15E6CC4B4A@opencsw.org>
References: <49D29240.2070105@opencsw.org>
	<B65657D6-5AB5-4253-B36E-6B15E6CC4B4A@opencsw.org>
Message-ID: <C0BBBFAB-6239-4C18-8E53-98BC55DF603E@opencsw.org>

Hi,

Am 01.04.2009 um 08:37 schrieb Dagobert Michelsen:
> it seems like the T5220 was overloaded due to lack of Ram. As I
> already wrote there will be a memory upgrade soon. We are working
> on fixing this issue.

There were a number of problems:

1. Hudson got too busy:
Apr  1 09:00:40 ncsw tmpfs: WARNING: /zones/hudson/root/tmp: File  
system full, swap space limit exceeded

2. Bug #6682528 was hit: memory leak in nfsmapid if resolv_init() is  
called

3. Overall machine usage due to heavy packaging activity

All this in sum slowed the machine down. As a temporary workaround
I turned off the Hudson zone (it isn't productive yet, Trygve, yes?).

Additionally, to make something good out of the bad I use the
unplanned downtime to bring the machine to a current patchlevel
and updated all CSW packages to current/.


The machine will be available again in a few hours. I let you know.


Sorry for the inconvenience

   -- Dago


From skayser at opencsw.org  Wed Apr  1 10:22:27 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 01 Apr 2009 10:22:27 +0200
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
References: <ce97543b0903260441k26550f94r96a1f6bb6c95755d@mail.gmail.com>	<625385e30903271340r4307dd64hd691e4dac6fa46ff@mail.gmail.com>	<20090327211430.GV47757@bolthole.com>	<ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>	<20090329140924.GD30131@bolthole.com>	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>	<20090329220733.GM69158@bolthole.com>	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
	<6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
Message-ID: <49D32443.1030804@opencsw.org>

rupert THURNER wrote:
> not by me anyway :)
> dago, you mind giving me write access to mediawiki?

Done, you now have editor rights.

Dago, what about changing gar.opencsw.org so that it redirects to the
Trac installation, now that Rupert has moved the remaining up-to-date docs?

Sebastian


From dam at opencsw.org  Wed Apr  1 10:51:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 10:51:07 +0200
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <49D32443.1030804@opencsw.org>
References: <ce97543b0903260441k26550f94r96a1f6bb6c95755d@mail.gmail.com>	<625385e30903271340r4307dd64hd691e4dac6fa46ff@mail.gmail.com>	<20090327211430.GV47757@bolthole.com>	<ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>	<20090329140924.GD30131@bolthole.com>	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>	<20090329220733.GM69158@bolthole.com>	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
	<6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
	<49D32443.1030804@opencsw.org>
Message-ID: <631F4BE9-F055-41CE-972F-1C44202C19F1@opencsw.org>

Hi Sebastian,

Am 01.04.2009 um 10:22 schrieb Sebastian Kayser:
> Dago, what about changing gar.opencsw.org so that it redirects to the
> Trac installation, now that Rupert has moved the remaining up-to- 
> date docs?

Done.


Best regards

   -- Dago


From james at opencsw.org  Wed Apr  1 11:03:20 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 09:03:20 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
Message-ID: <20090401.9032000.316239962@gyor.oxdrove.co.uk>

On 01/04/09, 00:52:31, Chad Harp <harpchad at opencsw.org> wrote regarding Re:
[csw-maintainers] Runtime linker pathes in GAR:

> Phil, can checkpkg detect some of the common cases?  I'm thinking at
> least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those
> would indicate a bad RUNPATHQUOTE value.  James can probably give us
> an idea of what he's coming across the most.

/opt/csw/lib/\$ISALIST
/opt/csw/lib/\SALIST
/opt/csw/lib/\$$ISALIST
/opt/csw/lib/SALIST
/opt/csw/lib/-R/


but checking for bad paths is the wrong way to find errors.  I check
for good paths (directory exists and/or is valid dynamic linker token)
and throw up anything else.


The nice thing about using LD_OPTIONS is it doesn't need escaping
(more than initially) and can put $ISALIST first without libtool
interfering.




James.


From dam at opencsw.org  Wed Apr  1 12:05:35 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 12:05:35 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <20090401.9032000.316239962@gyor.oxdrove.co.uk>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
Message-ID: <BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>

Hi James,

Am 01.04.2009 um 11:03 schrieb James Lee:
> On 01/04/09, 00:52:31, Chad Harp <harpchad at opencsw.org> wrote  
> regarding Re:
> [csw-maintainers] Runtime linker pathes in GAR:
>
>> Phil, can checkpkg detect some of the common cases?  I'm thinking at
>> least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those
>> would indicate a bad RUNPATHQUOTE value.  James can probably give us
>> an idea of what he's coming across the most.
>
> /opt/csw/lib/\$ISALIST
> /opt/csw/lib/\SALIST
> /opt/csw/lib/\$$ISALIST
> /opt/csw/lib/SALIST
> /opt/csw/lib/-R/
>
> but checking for bad paths is the wrong way to find errors.  I check
> for good paths (directory exists and/or is valid dynamic linker token)
> and throw up anything else.
>
> The nice thing about using LD_OPTIONS is it doesn't need escaping
> (more than initially) and can put $ISALIST first without libtool
> interfering.

Maybe this is the one occasion where using LD_OPTIONS does
more good than bad? Just for '-R'?


Best regards

   -- Dago


From james at opencsw.org  Wed Apr  1 12:20:55 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 10:20:55 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
Message-ID: <20090401.10205500.1310557855@gyor.oxdrove.co.uk>

On 01/04/09, 11:05:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
Re: [csw-maintainers] Runtime linker pathes in GAR:

> > The nice thing about using LD_OPTIONS is it doesn't need escaping
> > (more than initially) and can put $ISALIST first without libtool
> > interfering.

> Maybe this is the one occasion where using LD_OPTIONS does
> more good than bad? Just for '-R'?

I suspect that on these occasions LD_OPTIONS wasn't used.




James.







From james at opencsw.org  Wed Apr  1 12:43:36 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 10:43:36 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <20090401.10205500.1310557855@gyor.oxdrove.co.uk>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
	<20090401.10205500.1310557855@gyor.oxdrove.co.uk>
Message-ID: <20090401.10433600.467498167@gyor.oxdrove.co.uk>

On 01/04/09, 11:20:55, James Lee <james at opencsw.org> wrote regarding Re: 
[csw-maintainers] Runtime linker pathes in GAR:

> On 01/04/09, 11:05:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding
> Re: [csw-maintainers] Runtime linker pathes in GAR:

> > > The nice thing about using LD_OPTIONS is it doesn't need escaping
> > > (more than initially) and can put $ISALIST first without libtool
> > > interfering.

> > Maybe this is the one occasion where using LD_OPTIONS does
> > more good than bad? Just for '-R'?

> I suspect that on these occasions LD_OPTIONS wasn't used.

PS... Sorry I am muddling occasions, I'm thinking the subject occasion
is "we have a lot of bad RPATHs" but we've moved to the topic to "how
to easily set an $ in the first element of an RPATH", and yes LD_OPTIONS
does it.



James.


From dam at opencsw.org  Wed Apr  1 13:21:50 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 13:21:50 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <20090401.10433600.467498167@gyor.oxdrove.co.uk>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
	<20090401.10205500.1310557855@gyor.oxdrove.co.uk>
	<20090401.10433600.467498167@gyor.oxdrove.co.uk>
Message-ID: <0A949260-DD06-4FC0-A9EB-35785EC4411F@opencsw.org>

Hi James,

Am 01.04.2009 um 12:43 schrieb James Lee:
> On 01/04/09, 11:20:55, James Lee <james at opencsw.org> wrote regarding  
> Re:
> [csw-maintainers] Runtime linker pathes in GAR:
>
>> On 01/04/09, 11:05:35, Dagobert Michelsen <dam at opencsw.org> wrote  
>> regarding
>> Re: [csw-maintainers] Runtime linker pathes in GAR:
>
>>>> The nice thing about using LD_OPTIONS is it doesn't need escaping
>>>> (more than initially) and can put $ISALIST first without libtool
>>>> interfering.
>
>>> Maybe this is the one occasion where using LD_OPTIONS does
>>> more good than bad? Just for '-R'?
>
>> I suspect that on these occasions LD_OPTIONS wasn't used.

I disabled the use of LD_OPTIONS after the last discussion that it
was a Bad Thing(tm).

> PS... Sorry I am muddling occasions, I'm thinking the subject occasion
> is "we have a lot of bad RPATHs" but we've moved to the topic to "how
> to easily set an $ in the first element of an RPATH", and yes  
> LD_OPTIONS
> does it.

Ok, is it agreed then that using LD_OPTIONS for -R is a Good Thing(tm)?
I'll remove the -R from LDFLAGS and put enable LD_OPTIONS for this  
option
only again.


Best regards

   -- Dago


From skayser at opencsw.org  Wed Apr  1 13:23:43 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 01 Apr 2009 13:23:43 +0200
Subject: [csw-maintainers] mGAR cpan builds: satisfying non-installed
	dependencies?
Message-ID: <49D34EBF.2020702@opencsw.org>

Hi,

are there any thoughts underway to enable straight-forward builds of
CPAN packages that have not-yet-installed dependencies (which need to be
built also)?

>From looking at the category.mk for the cpan category it seems as if
this had been working with spool* directories of GAR, but as mGAR has a
separate $(DESTDIR) for each package it doesn't work with mGAR.

$ svn pg svn:externals .
gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2

$ ggrep -C1 ^PERL5LIB gar/categories/cpan/category.mk
# Enable scripts to see prereqs
PERL5LIB  = $(DESTDIR)$(libdir)/perl/csw
PERL5LIB := $(PERL5LIB):$(DESTDIR)$(datadir)/perl/csw
export PERL5LIB

I just built ack [1] which relied on File-Next (not yet in the catalog,
nor on the build farm). To satisfy this dependency I built File-Next,
manually copied it to a temporary directory, and pointed PERL5LIB to
this directory when building ack.

This worked, but maybe someone has ideas on how to tweak GAR so that
this installation and finding of dependencies becomes an automated
process :)

Sebastian

[1] http://betterthangrep.com/


From dam at opencsw.org  Wed Apr  1 14:32:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 14:32:55 +0200
Subject: [csw-maintainers] Update: BO Buildfarm
Message-ID: <63668C3B-E4A2-49D2-94E9-399C1CFC605F@opencsw.org>

Hi,

the real cause for the downtime was bldcat:

> Inspecting ./gcc4g++-4.3.3,REV=2009.03.31-SunOS5.8-sparc-CSW.pkg.gz
> Use of uninitialized value in concatenation (.) or string at /opt/ 
> csw/bin/bldcat line 81.
> Use of uninitialized value in concatenation (.) or string at /opt/ 
> csw/bin/bldcat line 81.
> Can't open /tmp/bldcat.1301.1238588867//pkginfo: No such file or  
> directory at /opt/csw/bin/bldcat line 81.

/tmp was filled with old stuff. There should be a number of changes
to bldcat. Peter:

- change string from \s+ to [^-]+
- install trap handler so that all temporary files are cleaned up on
   exit, regardless why.

Until this is fixed I disabled catalog generation. I'll look that up as
soon as I can. There are still a number of patches running in. Please
stand by.


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  1 15:11:05 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 15:11:05 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <F8FC4A58-F0D9-4D5B-A574-080E0028CB8F@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<F8FC4A58-F0D9-4D5B-A574-080E0028CB8F@opencsw.org>
Message-ID: <EBC98680-E60F-4D7D-A236-560AE199E15A@opencsw.org>

Hi Chad,

Am 01.04.2009 um 01:47 schrieb Chad Harp:
> I assume RUNPATHQUOTE=3 is also valid?  I have one package where  
> RPATH=/opt/csw/lib/SALIST with =2, looks like $I got expanded.

No, that is not valid, just 1 and 2. It looks like this whole
system needs a thorough rewrite.


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  1 15:18:35 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 15:18:35 +0200
Subject: [csw-maintainers] Another cyclic dependency
Message-ID: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>

Hi,

and another interesting dependency chain, this time three
packages long:

> Trying to install dependancy commons_collect
> No existing install of CSWajccollect found. Installing...
> Pre-existing local file commons_collect-3.2.1,REV=2009.03.26- 
> SunOS5.8-all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_config
> No existing install of CSWajcconfig found. Installing...
> Pre-existing local file commons_config-1.6,REV=2009.03.27-SunOS5.8- 
> all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_digester
> No existing install of CSWajcdigester found. Installing...
> Pre-existing local file commons_digester-2.0,REV=2009.03.26-SunOS5.8- 
> all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_collect
> No existing install of CSWajccollect found. Installing...
> Pre-existing local file commons_collect-3.2.1,REV=2009.03.26- 
> SunOS5.8-all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_config
> No existing install of CSWajcconfig found. Installing...
> Pre-existing local file commons_config-1.6,REV=2009.03.27-SunOS5.8- 
> all-CSW.pkg.gz matches checksum
...


Best regards

   -- Dago


From william at wbonnet.net  Wed Apr  1 15:16:24 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 15:16:24 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
Message-ID: <49D36928.6060801@wbonnet.net>

Hi Dago

Tanks for the information, i fix it tonight

Cheers

> Hi,
>
> and another interesting dependency chain, this time three
> packages long:
>
>> Trying to install dependancy commons_collect
>> No existing install of CSWajccollect found. Installing...
>> Pre-existing local file 
>> commons_collect-3.2.1,REV=2009.03.26-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_config
>> No existing install of CSWajcconfig found. Installing...
>> Pre-existing local file 
>> commons_config-1.6,REV=2009.03.27-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_digester
>> No existing install of CSWajcdigester found. Installing...
>> Pre-existing local file 
>> commons_digester-2.0,REV=2009.03.26-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_collect
>> No existing install of CSWajccollect found. Installing...
>> Pre-existing local file 
>> commons_collect-3.2.1,REV=2009.03.26-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_config
>> No existing install of CSWajcconfig found. Installing...
>> Pre-existing local file 
>> commons_config-1.6,REV=2009.03.27-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
> ...
>
>
> Best regards
>
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From james at opencsw.org  Wed Apr  1 15:24:24 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 13:24:24 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <0A949260-DD06-4FC0-A9EB-35785EC4411F@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
	<20090401.10205500.1310557855@gyor.oxdrove.co.uk>
	<20090401.10433600.467498167@gyor.oxdrove.co.uk>
	<0A949260-DD06-4FC0-A9EB-35785EC4411F@opencsw.org>
Message-ID: <20090401.13242400.652740328@gyor.oxdrove.co.uk>

On 01/04/09, 12:21:50, Dagobert Michelsen <dam at opencsw.org> wrote regarding
Re: [csw-maintainers] Runtime linker pathes in GAR:

> Ok, is it agreed then that using LD_OPTIONS for -R is a Good Thing(tm)?

LD_OPTIONS is a useful tool.  As is LD_LIBRARY_PATH for linking to
libraries in temporary locations (but don't run with it as RPATH
should know best).

The problem with LD_OPTIONS is it sets the same parameters for all
which might not be appropriate.  One of your executables might need,
e.g., /opt/csw/bdb4/lib first but maybe not all would - although,
assuming the right lib is eventually found, it only wastes the linker's
and user's time, just as those bad RPATHs cause the runtime linker to
look for files that don't exist.

Use LD_OPTIONS if it works.



James.


From james at opencsw.org  Wed Apr  1 15:40:17 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 13:40:17 GMT
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
Message-ID: <20090401.13401700.4235995713@gyor.oxdrove.co.uk>

On 01/04/09, 14:18:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
[csw-maintainers] Another cyclic dependency:


>From CSWggettextrt, needs libiconv which is in CSWiconv:

$ ldd /opt/csw/bin/ggettext
        libintl.so.8 =>  /opt/csw/lib/libintl.so.8
        libsec.so.1 =>   /lib/libsec.so.1
        libc.so.1 =>     /lib/libc.so.1
        libiconv.so.2 =>         /opt/csw/lib/libiconv.so.2
        libavl.so.1 =>   /lib/libavl.so.1
        libm.so.2 =>     /lib/libm.so.2



>From CSWiconv, needs libintl which is in CSWggettextrt:

$ ldd /opt/csw/bin/iconv
        libiconv.so.2 =>         /opt/csw/lib/i386/libiconv.so.2
        libintl.so.3 =>  /opt/csw/lib/i386/libintl.so.3
        libc.so.1 =>     /lib/libc.so.1
        libm.so.2 =>     /lib/libm.so.2


Shouldn't the installer be tolerant of dependancy loops and just 
install both when either is requested?




James.


From dam at opencsw.org  Wed Apr  1 16:04:33 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 16:04:33 +0200
Subject: [csw-maintainers] Update: BO Buildfarm available again
In-Reply-To: <63668C3B-E4A2-49D2-94E9-399C1CFC605F@opencsw.org>
References: <63668C3B-E4A2-49D2-94E9-399C1CFC605F@opencsw.org>
Message-ID: <6139E2DD-DB5F-427D-9CC1-A1D505CF36B6@opencsw.org>

Hi,

Am 01.04.2009 um 14:32 schrieb Dagobert Michelsen:
> /tmp was filled with old stuff.

I have now a workaround in place until the course is fixed.
Everything should be fine again. Please let me know if
you encounter any other issues.


Sorry for the inconvenience

   -- Dago


From william at wbonnet.net  Wed Apr  1 16:16:01 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 16:16:01 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <20090401.13401700.4235995713@gyor.oxdrove.co.uk>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
Message-ID: <49D37721.8030707@wbonnet.net>

Hi

> >From CSWggettextrt, needs libiconv which is in CSWiconv:
> >From CSWiconv, needs libintl which is in CSWggettextrt:
>   
This one can easily be solved by splitting pckage into CSWiconv and 
CSWiconvrt

Cheers


From Darin.Perusich at cognigencorp.com  Wed Apr  1 16:39:07 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 01 Apr 2009 10:39:07 -0400
Subject: [csw-maintainers] install mtx
Message-ID: <49D37C8B.20201@cognigencorp.com>

Can someone install the mtx package across the build farm?

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From harpchad at opencsw.org  Wed Apr  1 17:03:11 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 01 Apr 2009 10:03:11 -0500
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <20090401.13401700.4235995713@gyor.oxdrove.co.uk>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
Message-ID: <49D3822F.6030002@opencsw.org>

An excerpt from the inconv docs:

After installing GNU libiconv for the first time, it is recommended to 
recompile and reinstall GNU gettext, so that it can take advantage of 
libiconv.

On systems other than GNU/Linux, the iconv program will be 
internationalized only if GNU gettext has been built and installed 
before GNU libiconv. This means that the first time GNU libiconv is 
installed, we have a circular dependency between the GNU libiconv and 
GNU gettext packages, which can be resolved by building and installing 
either

     * first libiconv, then gettext, then libiconv again,

or (on systems supporting shared libraries, excluding AIX)

     * first gettext, then libiconv, then gettext again.



What a mess, so it seems it's intended and we have to live with it.  Can 
the install utils support co-dependent packages and James suggested?

James Lee wrote:
> On 01/04/09, 14:18:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
> [csw-maintainers] Another cyclic dependency:
> 
> 
>>From CSWggettextrt, needs libiconv which is in CSWiconv:
> 
> $ ldd /opt/csw/bin/ggettext
>         libintl.so.8 =>  /opt/csw/lib/libintl.so.8
>         libsec.so.1 =>   /lib/libsec.so.1
>         libc.so.1 =>     /lib/libc.so.1
>         libiconv.so.2 =>         /opt/csw/lib/libiconv.so.2
>         libavl.so.1 =>   /lib/libavl.so.1
>         libm.so.2 =>     /lib/libm.so.2
> 
> 
> 
>>From CSWiconv, needs libintl which is in CSWggettextrt:
> 
> $ ldd /opt/csw/bin/iconv
>         libiconv.so.2 =>         /opt/csw/lib/i386/libiconv.so.2
>         libintl.so.3 =>  /opt/csw/lib/i386/libintl.so.3
>         libc.so.1 =>     /lib/libc.so.1
>         libm.so.2 =>     /lib/libm.so.2
> 
> 
> Shouldn't the installer be tolerant of dependancy loops and just 
> install both when either is requested?
> 
> 
> 
> 
> James.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From james at opencsw.org  Wed Apr  1 17:06:01 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 15:06:01 GMT
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <49D37721.8030707@wbonnet.net>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
	<49D37721.8030707@wbonnet.net>
Message-ID: <20090401.15060100.1316498277@gyor.oxdrove.co.uk>

On 01/04/09, 15:16:01, William Bonnet <william at wbonnet.net> wrote regarding
Re: [csw-maintainers] Another cyclic dependency:

> > >From CSWggettextrt, needs libiconv which is in CSWiconv:
> > >From CSWiconv, needs libintl which is in CSWggettextrt:
> >
> This one can easily be solved by splitting pckage into CSWiconv and
> CSWiconvrt

I think you'll find the libraries are also interdependent.  Granted
one doesn't "run" libraries and superior packages can pull both as
has been happening since 2003.

http://www.opencsw.org/bugtrack/view.php?id=173

It can be easily solved without splitting by breaking the depend loop
on install which would also handle all cases not just this one.



James.


From william at wbonnet.net  Wed Apr  1 17:12:41 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 17:12:41 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <49D3822F.6030002@opencsw.org>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
	<49D3822F.6030002@opencsw.org>
Message-ID: <49D38469.2090802@wbonnet.net>

Hi
>
>
> What a mess, so it seems it's intended and we have to live with it.  
> Can the install utils support co-dependent packages and James suggested?
BTW it is the same kind of mess for some java libs...


cheers


From Darin.Perusich at cognigencorp.com  Wed Apr  1 17:24:16 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 01 Apr 2009 11:24:16 -0400
Subject: [csw-maintainers] install star
Message-ID: <49D38720.2040703@cognigencorp.com>

Can star be installed across the build farm please?

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From harpchad at opencsw.org  Wed Apr  1 17:27:32 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 01 Apr 2009 10:27:32 -0500
Subject: [csw-maintainers] glib 2.20.0 in testing
Message-ID: <49D387E4.5020405@opencsw.org>

It's a manual download/install until the cataloger is fixed:

# glib2-2.20.0,REV=2009.03.31-SunOS5.8-i386-CSW.pkg.gz
# glib2-2.20.0,REV=2009.03.31-SunOS5.8-sparc-CSW.pkg.gz
# glib2_devel-2.20.0,REV=2009.03.31-SunOS5.8-i386-CSW.pkg.gz
# glib2_devel-2.20.0,REV=2009.03.31-SunOS5.8-sparc-CSW.pkg.gz


From amaier at opencsw.org  Wed Apr  1 17:58:36 2009
From: amaier at opencsw.org (Alexander Maier)
Date: Wed, 1 Apr 2009 17:58:36 +0200
Subject: [csw-maintainers] Font packages
In-Reply-To: <ce97543b0903310320n2ef5c363ica24fb750a880eea@mail.gmail.com>
References: <ce97543b0903310320n2ef5c363ica24fb750a880eea@mail.gmail.com>
Message-ID: <8642B5FA-EA51-4702-A07B-32641C37B1D4@opencsw.org>

Am 31.03.2009 um 12:20 schrieb Maciej (Matchek) Blizinski:

> How do we go about adding font packages? Let's suppose I have a font
> what I want to package. Let it be Terminus[1], for the sake of the
> exercise. Can it be somehow integrated with Sun-provided X server in
> Solaris 10? Are there established practices about including fonts in
> OpenCSW?
>


I think one have to distinguish between fonts for "classic X apps"  
with local/remote xfs and fonts for apps using libfreetype/fontconfig  
(based on GTK or Qt).

In the latter case I would suggest to put these fonts into /opt/csw/ 
share/fonts (e.g. Microsoft TrueType fonts or Red Hat Liberation fonts).

Terminus seems to be a classic X11/Motif font, so it would probably be  
better off somewhere below /opt/csw/X11 (as Phil has suggested).

-Alexander




From dam at opencsw.org  Wed Apr  1 18:11:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 18:11:40 +0200
Subject: [csw-maintainers] install mtx
In-Reply-To: <49D37C8B.20201@cognigencorp.com>
References: <49D37C8B.20201@cognigencorp.com>
Message-ID: <8812DEA1-EC3F-449B-ABEB-064DB1375471@opencsw.org>

Hi Darin,

Am 01.04.2009 um 16:39 schrieb Darin Perusich:
> Can someone install the mtx package across the build farm?

Am 01.04.2009 um 17:24 schrieb Darin Perusich:
> Can star be installed across the build farm please?

Doing this now. Please mail to
   buildfarm at lists.opencsw.org
next time as there are possibly buildfarm admins who are not on
maintainers at .


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  1 18:26:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 18:26:17 +0200
Subject: [csw-maintainers] install mtx
In-Reply-To: <8812DEA1-EC3F-449B-ABEB-064DB1375471@opencsw.org>
References: <49D37C8B.20201@cognigencorp.com>
	<8812DEA1-EC3F-449B-ABEB-064DB1375471@opencsw.org>
Message-ID: <D7B0F3DD-D5EE-4F1A-8687-1A2FD51733A0@opencsw.org>

Hi,

Am 01.04.2009 um 18:11 schrieb Dagobert Michelsen:
> Am 01.04.2009 um 16:39 schrieb Darin Perusich:
>> Can someone install the mtx package across the build farm?
>
> Am 01.04.2009 um 17:24 schrieb Darin Perusich:
>> Can star be installed across the build farm please?
>
> Doing this now.

Done.


Best regards

   -- Dago


From phil at bolthole.com  Wed Apr  1 18:45:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 1 Apr 2009 09:45:45 -0700
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <631F4BE9-F055-41CE-972F-1C44202C19F1@opencsw.org>
References: <ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>
	<20090329140924.GD30131@bolthole.com>
	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>
	<20090329220733.GM69158@bolthole.com>
	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>
	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>
	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
	<6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
	<49D32443.1030804@opencsw.org>
	<631F4BE9-F055-41CE-972F-1C44202C19F1@opencsw.org>
Message-ID: <20090401164545.GA19124@bolthole.com>

On Wed, Apr 01, 2009 at 10:51:07AM +0200, Dagobert Michelsen wrote:
> Hi Sebastian,
>
> Am 01.04.2009 um 10:22 schrieb Sebastian Kayser:
>> Dago, what about changing gar.opencsw.org so that it redirects to the
>> Trac installation, now that Rupert has moved the remaining up-to-date 
>> docs?
>
> Done.

i glanced at gar.opencsw.org.

It mentions "mGar is completely copied, GAR stuff not yet completely."
but it doesnt say what "mGar" *is*.

May i sugest that you now officially "deprecate" and devalue references to
the old gar version(s). Move all of that stuff to a "Historical" category at
the bottom.

having "GAR versions" as the #2 link under "Documentation", just adds
unneccessary confusion, in my opinion.

I would also like to request that the "trac admins who aren't Dago",
finally give us an up to date "Getting started with GAR" wiki page there?
;-)
[Dago has been meaning to for a long time, but he's too busy doing all
 the other stuff he does for us. Help him out! :-) ]


From skayser at opencsw.org  Wed Apr  1 19:11:37 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 01 Apr 2009 19:11:37 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <49D3A049.2040102@opencsw.org>

Dagobert Michelsen wrote:
> James is currently in the tedious task of reviewing all
> runtime linker pathes from the packages.

Just for clarification: is it only about correcting those incorrectly
quoted RUNPATHs or are we supposed to also adjust the RUNPATH (via
LINKER_FLAGS or similar) depending on what our applications use (i.e.
drop .../$ISALIST or even /opt/csw/lib completely)?

I personally would prefer to just have GAR corrected WRT to the $ISALIST
quoting issue and not add more intelligence to the build description,
except maybe for applications where the startup delay might have a
performance impact.

Sebastian


From harpchad at opencsw.org  Wed Apr  1 19:15:47 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 01 Apr 2009 12:15:47 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <49D3A143.90709@opencsw.org>

I think I need a better understanding of what I should be doing. 
Several of my packages have path problems due to quoting.  Should I be 
looking for the correct combinations of variable to fix it, or will GAR 
be fixed to deal with it?


From mwatters at opencsw.org  Wed Apr  1 19:33:48 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 01 Apr 2009 12:33:48 -0500
Subject: [csw-maintainers] chrpath now in Testing
Message-ID: <49D3A57C.4070201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ladies and Gentlemen,

I am pleased to announce a new package I copied into testing.

chrpath
This will allow changing the RPATH and RUNPATH values in binaries and libraries.
I have done testing on both sparc and x86 platforms and all seems well.

The main limitation I have found is you can not "grow" the path any larger the
the initial size.

for example:
if the original has RPATH of /opt/csw/lib/SALIST:/opt/csw/lib

you could NOT make it /opt/csw/lib/$ISALIST:/opt/csw/lib
but you CAN make it /opt/csw/lib/$ISALIST

example2:
if the original has RPATH of /opt/csw/lib/\\$ISALIST:/opt/csw/lib:/opt/csw/lib
you can make it /opt/csw/lib/$ISALIST:/opt/csw/lib

you don't need to escape if you use single quotes
chrpath -r '/opt/csw/lib/$ISALIST:/opt/csw/lib' filename
will work beautifully

I had to dig around for the source as the link from FSF is broken.
I downloaded the source from Debian's repository. but the package points
to the FSF site.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknTpXwACgkQLrhmsXMSLxek6QCeKhTMINMz+nrtoQijdCT5tNMd
Yg4An1hK1NmHNzesnACRO7k4RsQ2i7ZI
=Pm8U
-----END PGP SIGNATURE-----


From james at opencsw.org  Wed Apr  1 20:19:08 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 18:19:08 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <49D3A049.2040102@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<49D3A049.2040102@opencsw.org>
Message-ID: <20090401.18190800.2943361049@gyor.oxdrove.co.uk>

On 01/04/09, 18:11:37, Sebastian Kayser <skayser at opencsw.org> wrote
regarding Re: [csw-maintainers] Runtime linker pathes in GAR:

> Just for clarification: is it only about correcting those incorrectly
> quoted RUNPATHs or are we supposed to also adjust the RUNPATH (via
> LINKER_FLAGS or similar) depending on what our applications use (i.e.
> drop .../$ISALIST or even /opt/csw/lib completely)?

You can drop /opt/csw/lib when behind /opt/csw/lib/$ISALIST
because the arch dir for the minimum arch is a link back to
/opt/csw/lib, provided by CSWcommon.

$ ls -l /opt/csw/lib/sparcv8
lrwxrwxrwx   1 root     other          1 Jun 22  2008 
/opt/csw/lib/sparcv8
-> .


This means there will be a match before the end of the $ISALIST.


$ isalist
sparcv9+vis sparcv9 sparcv8plus+vis sparcv8plus sparcv8 sparcv8-fsmuld
sparcv7 sparc

would match at sparcv8 even with no arch libs, so skipping sparcv7
etc. which are never in CSW anyway.  Hence never needs /opt/csw/lib
after.



James.


From james at opencsw.org  Wed Apr  1 20:26:31 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 18:26:31 GMT
Subject: [csw-maintainers] chrpath now in Testing
In-Reply-To: <49D3A57C.4070201@opencsw.org>
References: <49D3A57C.4070201@opencsw.org>
Message-ID: <20090401.18263100.3812230620@gyor.oxdrove.co.uk>

On 01/04/09, 18:33:48, Mike Watters <mwatters at opencsw.org> wrote regarding
[csw-maintainers] chrpath now in Testing:

> chrpath
> This will allow changing the RPATH and RUNPATH values in binaries and
> libraries.
> I have done testing on both sparc and x86 platforms and all seems well.

> The main limitation I have found is you can not "grow" the path any
> larger the the initial size.

> for example:
> if the original has RPATH of /opt/csw/lib/SALIST:/opt/csw/lib

> you could NOT make it /opt/csw/lib/$ISALIST:/opt/csw/lib
> but you CAN make it /opt/csw/lib/$ISALIST

That's fine, as per my previous message, /opt/csw/lib is not needed
after /opt/csw/lib/$ISALIST.

Lots of RPATHS have duplicate directories too so often you will have
the slack to add the missing "$I".




James.


From dam at opencsw.org  Wed Apr  1 21:35:25 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 21:35:25 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <49D3A143.90709@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<49D3A143.90709@opencsw.org>
Message-ID: <7673312B-D309-4AEC-B869-D2C0D3261BB9@opencsw.org>

Hi Chad,

Am 01.04.2009 um 19:15 schrieb Chad Harp:
> I think I need a better understanding of what I should be doing.
> Several of my packages have path problems due to quoting.  Should
> I be looking for the correct combinations of variable to fix it,
> or will GAR be fixed to deal with it?

Ok, now here is the rationale why things are as they are now:

- /opt/csw/lib/$ISALIST is needed if and only if libraries are
   linked that may contain optimizations. Otherwise it is best to
   avoid $ISALIST completely. This can be done by setting
     NOISALIST = 1

- I put -R for runtime linker path in LDFLAGS because in the last
   discussion LD_OPTIONS was considered bad, as it is always applied
   first in linking regardless of other flags. Unfortunately these
   flags are passwd through multiple commands, each with evaluation
   of the argument. Now $ISALIST looks like a variable and must be
   shielded from evaluation by quoting. The amount of quoting depends
   on the number of evaluation, e. g. if libtool is used, a custom
   Makefile, etc.
   The number of quotings can be adjusted by the simple parameter
     RUNPATHQUOTE = 1    for single evaluation
     RUNPATHQUOTE = 2    for double evaluation
   Unfortunately this obviously proved to be not sufficient as the
   bug reports from James show.

   (BTW, if you are interested: Making the literal '$' as prefix is
     _Q = \\\$$\$$       for RUNPATHQUOTE = 1
   and
     _Q = \\\\\\\$$\$$   for RUNPATHQUOTE = 2
   in gar.conf.mk).

- Moving -R to LD_OPTIONS would make things considerable easier,
   because the number of evaluations is fixed as it is not processed
   by multiple commands. It would just be needed to decide if
   $ISALIST should be in or not. NOISALIST is already there for this.
   This would conflict with packages already using LD_OPTIONS:
     amarok
     cups
     daimonin
     exim
     gftp
     gphoto2
     hatari
     hypermail
     kdesvn
     kile
     libgphoto2
     lyx
     mysql-python
     netsnmp
     slrn
     tsclient
     w3m
   To be consistent, EXTRA_LD_OPTIONS should be introduced and the
   Makefiles adjusted accordingly when the GAR modifications have
   been made.


So the GAR integration is possible, but needs some experimentation
in quoting and best use.


Best regards

   -- Dago
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090401/6f35ccbe/attachment.htm>

From william at wbonnet.net  Wed Apr  1 21:56:56 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 21:56:56 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <49D38469.2090802@wbonnet.net>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>	<49D3822F.6030002@opencsw.org>
	<49D38469.2090802@wbonnet.net>
Message-ID: <49D3C708.80504@wbonnet.net>

Hi

I removed dependency from digester to collections. It should "break" the 
loop. I submited the fix to Phil.

Cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Wed Apr  1 22:42:23 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 1 Apr 2009 13:42:23 -0700
Subject: [csw-maintainers] Wanted: opensolaris python lovers
Message-ID: <20090401204223.GF19124@bolthole.com>


So... Anyone running opensolaris, in addition to the regular stuff?
Love python?

I'm thinking it would be a very good thing, if we offered an IPS option for
getting our packages.

However, the existing "automated converter mechanism" sun offers sucks. It totally
misses out on postinstall stuff. (among other drawbacks)

There is apparently a mechanism in IPS, called an "actuator", that
sounds to be a perfect way to trigger postinstall scripts.
Theree's just the small wrinkle, that the IPS team refuses to make people's
lives easier and saner by delivering a pre-written actuator for the task.

So, in the continuing spirit of us taking on what Sun is too short-sighted
to do... Would anyone be interested in augmenting the existing
SVR4-to-IPS converter, to deal with this as well?

Since all IPS related stuff is python, this person would need to have a
good familiarity with python, as well as running opensolaris somewhere.

The downside: a decent-sized chunk of work and responsability
The upside:   gobs of fame and glory ;-)



From mwatters at opencsw.org  Wed Apr  1 23:47:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 01 Apr 2009 16:47:33 -0500
Subject: [csw-maintainers] python 2.6.1 now in testing
Message-ID: <49D3E0F5.2060809@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:

	bug fix: 3510 'import curses' fails in current Python...
			forced curses to use the library in /opt/csw
	bug fix: 3579 RPATH stuff
			added NOISALIST = 1
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknT4PUACgkQLrhmsXMSLxeP6wCeMvcjju/oNRtKUky+XipTnA2O
+50AoOPPCfkd5PhSaysD+VeaK6JeLuPp
=Y8/M
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr  2 00:07:20 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 01 Apr 2009 17:07:20 -0500
Subject: [csw-maintainers] mysql-python now in testing
Message-ID: <49D3E598.60303@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	isalist fix

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknT5ZgACgkQLrhmsXMSLxepuACfbjpI6klFYyGdidU4iFhpwfdy
6p0An3CDd08aGv5WgVf3mbVP2xy6nDeU
=d5Fp
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Apr  2 08:52:38 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 2 Apr 2009 08:52:38 +0200
Subject: [csw-maintainers] mGAR cpan builds: satisfying non-installed
	dependencies?
In-Reply-To: <49D34EBF.2020702@opencsw.org>
References: <49D34EBF.2020702@opencsw.org>
Message-ID: <1BD7044E-489F-4EB7-BCF5-3907AFC782F9@opencsw.org>

Hi Sebastian,

Am 01.04.2009 um 13:23 schrieb Sebastian Kayser:
> are there any thoughts underway to enable straight-forward builds of
> CPAN packages that have not-yet-installed dependencies (which need  
> to be
> built also)?
>
>> From looking at the category.mk for the cpan category it seems as if
> this had been working with spool* directories of GAR, but as mGAR  
> has a
> separate $(DESTDIR) for each package it doesn't work with mGAR.
>
> $ svn pg svn:externals .
> gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2
>
> $ ggrep -C1 ^PERL5LIB gar/categories/cpan/category.mk
> # Enable scripts to see prereqs
> PERL5LIB  = $(DESTDIR)$(libdir)/perl/csw
> PERL5LIB := $(PERL5LIB):$(DESTDIR)$(datadir)/perl/csw
> export PERL5LIB
>
> I just built ack [1] which relied on File-Next (not yet in the  
> catalog,
> nor on the build farm). To satisfy this dependency I built File-Next,
> manually copied it to a temporary directory, and pointed PERL5LIB to
> this directory when building ack.

The official way is to build the dependency, release it, have it
installed on the farm and continue with the next package.

Maybe we find in October ("CPAN Coverage") a more viable solution.
I guess with this runtime pathes and the stable release there are
other things to be done first...


Best regards

   -- Dago


From skayser at opencsw.org  Thu Apr  2 14:22:34 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 02 Apr 2009 14:22:34 +0200
Subject: [csw-maintainers] mGAR cpan builds: satisfying non-installed
 dependencies?
In-Reply-To: <1BD7044E-489F-4EB7-BCF5-3907AFC782F9@opencsw.org>
References: <49D34EBF.2020702@opencsw.org>
	<1BD7044E-489F-4EB7-BCF5-3907AFC782F9@opencsw.org>
Message-ID: <49D4AE0A.80102@opencsw.org>

Dagobert Michelsen wrote:
> Am 01.04.2009 um 13:23 schrieb Sebastian Kayser:
>> are there any thoughts underway to enable straight-forward builds of
>> CPAN packages that have not-yet-installed dependencies (which need  
>> to be
>> built also)?
>>
>>> From looking at the category.mk for the cpan category it seems as if
>> this had been working with spool* directories of GAR, but as mGAR  
>> has a
>> separate $(DESTDIR) for each package it doesn't work with mGAR.
>>
>> $ svn pg svn:externals .
>> gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2
>>
>> $ ggrep -C1 ^PERL5LIB gar/categories/cpan/category.mk
>> # Enable scripts to see prereqs
>> PERL5LIB  = $(DESTDIR)$(libdir)/perl/csw
>> PERL5LIB := $(PERL5LIB):$(DESTDIR)$(datadir)/perl/csw
>> export PERL5LIB
>>
>> I just built ack [1] which relied on File-Next (not yet in the  
>> catalog,
>> nor on the build farm). To satisfy this dependency I built File-Next,
>> manually copied it to a temporary directory, and pointed PERL5LIB to
>> this directory when building ack.
> 
> The official way is to build the dependency, release it, have it
> installed on the farm and continue with the next package.
> 
> Maybe we find in October ("CPAN Coverage") a more viable solution.
> I guess with this runtime pathes and the stable release there are
> other things to be done first...

Indeed. Was just wondering whether there might have been a more elegant
way to do it ATM. Thanks for the response.

Sebastian


From dam at opencsw.org  Thu Apr  2 14:22:37 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 2 Apr 2009 14:22:37 +0200
Subject: [csw-maintainers] RPATH in GAR
Message-ID: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>

Hi,

I noticed that some of you are already running ahead working
around the misconceptions of the current build system and
fixing runtime linker pathes superfast. Thanks for your
engagement! However, I think a more general solution in GAR
itself is a more viable solution, both in terms of
standardization and also easier to handle.

The rewritten module for runtime linker pathes is in effect
since r4157/r4158. See
   <http://apps.sourceforge.net/trac/gar/browser/csw/mgar/gar/v2/gar.conf.mk#L435 
 >
for details.

The usage of the rewritten module is documented at
   <http://apps.sourceforge.net/trac/gar/wiki/RuntimeLinkerPathes>

In the easiest case it should be sufficient to do just a
rebuild and everything should be fine. However, the new
structure gives you fine-grained control of the pathes
to be added.

Feel free to ask if there are any issues.


Best regards

   -- Dago


From mwatters at opencsw.org  Thu Apr  2 17:02:09 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 10:02:09 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
Message-ID: <49D4D371.7070102@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Maciej Blizi?ski wrote:
> On Wed, Apr 1, 2009 at 10:47 PM, Mike Watters <mwatters at opencsw.org> wrote:
>> Changes:
>>
>>        bug fix: 3510 'import curses' fails in current Python...
>>                        forced curses to use the library in /opt/csw
> 
> There's still a problem with importing 'curses' module, even though
> it's now linked against CSW ncurses library.
> 
> vsol01 ~ # pkgparam CSWpython VERSION
> 2.6.1,REV=2009.04.01
> vsol01 ~ # ldd /opt/csw/lib/python2.6/lib-dynload/_cursesmodule.so
>         libncurses.so.5 =>       /opt/csw/lib/libncurses.so.5
> vsol01 ~ # python
> Python 2.6.1 (r261:67515, Apr  1 2009, 16:27:46) [C] on sunos5
> Type "help", "copyright", "credits" or "license" for more information.
>>>> import curses
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
>   File "/opt/csw/lib/python2.6/curses/__init__.py", line 15, in <module>
>     from _curses import *
> ImportError: ld.so.1: python: fatal: relocation error: file
> /opt/csw/lib/python2.6/lib-dynload/_cursesmodule.so: symbol w32addch:
> referenced symbol not found
> _______________________________________________
> users mailing list
> users at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/users

ARRGGHHH!!

after a little more searching,
- -- wchgat symbol is in /opt/csw/lib/libncurses.so,
but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so

I will try to get this sorted out in the next few days.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEUEARECAAYFAknU03EACgkQLrhmsXMSLxe0TwCgn2fxcYMglnjrJekijLE9wvsq
gJQAl1RD8FqBq/Hrk+U+Vrfx4nC9rv8=
=WDao
-----END PGP SIGNATURE-----


From phil at bolthole.com  Thu Apr  2 18:23:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Apr 2009 09:23:25 -0700
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4D371.7070102@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
Message-ID: <20090402162325.GG19124@bolthole.com>

On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
> ARRGGHHH!!
> 
> after a little more searching,
> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so

thatas for widecha support. if it isnt there it may be bug in our curses.

or, you need to quit mixing curses headers with ncurses libraries.

-I/opt/csw/include/ncurses

if i remember correctly.


From mwatters at opencsw.org  Thu Apr  2 18:27:48 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 11:27:48 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <20090402162325.GG19124@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
Message-ID: <49D4E784.7070605@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>> ARRGGHHH!!
>>
>> after a little more searching,
>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
> 
> thatas for widecha support. if it isnt there it may be bug in our curses.
> 
> or, you need to quit mixing curses headers with ncurses libraries.
> 
> -I/opt/csw/include/ncurses
> 
> if i remember correctly.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

for some reason, /opt/csw/include/ncurses does not exist?
it exists as /opt/csw/include/ncursesw

and for the record, I have it in the recipe  still have issues  ;(
I will check the ncurses site for bugs and will file a bug against
the ncurses header location.

;)

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknU54QACgkQLrhmsXMSLxcMNQCfY203H7quhK6VWwODJZz5UiYS
JlMAn0CYCbSYwqUFbmsFNOwvxObmGCxo
=uIlm
-----END PGP SIGNATURE-----


From phil at bolthole.com  Thu Apr  2 18:34:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Apr 2009 09:34:57 -0700
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4E784.7070605@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
Message-ID: <20090402163457.GI19124@bolthole.com>

On Thu, Apr 02, 2009 at 11:27:48AM -0500, Mike Watters wrote:

> for some reason, /opt/csw/include/ncurses does not exist?
> it exists as /opt/csw/include/ncursesw

sounds like a takeover maintainer bug

> 
> and for the record, I have it in the recipe  still have issues  ;(
> I will check the ncurses site for bugs and will file a bug against
> the ncurses header location.

just have your pkg use sun curses


From mwatters at opencsw.org  Thu Apr  2 18:36:58 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 11:36:58 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <20090402162325.GG19124@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
Message-ID: <49D4E9AA.8050604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>> ARRGGHHH!!
>>
>> after a little more searching,
>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
> 
> thatas for widecha support. if it isnt there it may be bug in our curses.
> 
> or, you need to quit mixing curses headers with ncurses libraries.
> 
> -I/opt/csw/include/ncurses
> 
> if i remember correctly.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

AH-HA!

it seems there are 2 sets of ncurses libs
libncurses and libncursesw (wide support)
My Headers are picking up the "Wide" support
but my link is against the normal.

this is explained in
http://www.opencsw.org/mantis/view.php?id=3457
though I am still unclear as to why the headers
are missing for the "regular" ncurses.

I will fix and get a new version in testing.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknU6akACgkQLrhmsXMSLxeAywCeJTavy9igOA4zkngRuIMXQraR
WG4AoNlHRySPVahEvYXrKpe1TZQDFeaB
=QQmS
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr  2 19:05:59 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 12:05:59 -0500
Subject: [csw-maintainers] libssh2 now in testing
Message-ID: <49D4F077.4040205@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	update to version 1.1

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknU8HcACgkQLrhmsXMSLxedJACgiXu9pAqNMKX5Y+QAvut4ioEw
x9gAoJalAoAeHg2isdIvzaW20zzZUzmt
=W+1/
-----END PGP SIGNATURE-----


From skayser at opencsw.org  Thu Apr  2 19:55:34 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 02 Apr 2009 19:55:34 +0200
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4E9AA.8050604@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>
	<49D4E9AA.8050604@opencsw.org>
Message-ID: <49D4FC16.6060508@opencsw.org>

Mike Watters wrote:
> Philip Brown wrote:
>> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>>> ARRGGHHH!!
>>>
>>> after a little more searching,
>>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
>> thatas for widecha support. if it isnt there it may be bug in our curses.
> 
>> or, you need to quit mixing curses headers with ncurses libraries.
> 
>> -I/opt/csw/include/ncurses
> 
>> if i remember correctly.
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> AH-HA!
> 
> it seems there are 2 sets of ncurses libs
> libncurses and libncursesw (wide support)
> My Headers are picking up the "Wide" support
> but my link is against the normal.
> 
> this is explained in
> http://www.opencsw.org/mantis/view.php?id=3457
> though I am still unclear as to why the headers
> are missing for the "regular" ncurses.
> 
> I will fix and get a new version in testing.

I didn't get my head around why your linkage went like it did quite yet.
Shouldn't it have bailed out with linking errors (for not finding
w32addch) during building already? James?? ;)

Dago recently built our ncurses with wide character support (thanks
btw.), so that we now also have w variants of the ncurses libraries in
addition to the regular ones (not binary compatible but as they have a w
suffix there's no harm done). Both in the same package.

Coming to the include files. According to the ncurses README:

If you configure using the --enable-widec option, a "w" is appended to
the library names (e.g., libncursesw.a), and the resulting libraries
support wide-characters, e.g., via a UTF-8 locale.  The _corresponding
header files are compatible with the non-wide-character configuration_;
wide-character features are provided by ifdef's in the header files.

I don't exactly know how to read the header bit WRT packages building
against these headers. Can non-wide and wide applications both use the
same headers and ./configure sets #defines in case wide-support is
requested by an app (that's how i would read it)?

At least we should have the original include/ncurses directory back in
the package - presumably filled with the header files of the non-wide
build (as a safe option). And then in addition have include/ncursesw
populated with the header files of the wide variant?

Sebastian


From mwatters at opencsw.org  Thu Apr  2 20:31:59 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 13:31:59 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4FC16.6060508@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>	<49D4E9AA.8050604@opencsw.org>
	<49D4FC16.6060508@opencsw.org>
Message-ID: <49D5049F.10201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Sebastian Kayser wrote:
> Mike Watters wrote:
>> Philip Brown wrote:
>>> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>>>> ARRGGHHH!!
>>>>
>>>> after a little more searching,
>>>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>>>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
>>> thatas for widecha support. if it isnt there it may be bug in our curses.
>>> or, you need to quit mixing curses headers with ncurses libraries.
>>> -I/opt/csw/include/ncurses
>>> if i remember correctly.
>>> _______________________________________________
>>> maintainers mailing list
>>> maintainers at lists.opencsw.org
>>> https://lists.opencsw.org/mailman/listinfo/maintainers
>> AH-HA!
>>
>> it seems there are 2 sets of ncurses libs
>> libncurses and libncursesw (wide support)
>> My Headers are picking up the "Wide" support
>> but my link is against the normal.
>>
>> this is explained in
>> http://www.opencsw.org/mantis/view.php?id=3457
>> though I am still unclear as to why the headers
>> are missing for the "regular" ncurses.
>>
>> I will fix and get a new version in testing.
> 
> I didn't get my head around why your linkage went like it did quite yet.
> Shouldn't it have bailed out with linking errors (for not finding
> w32addch) during building already? James?? ;)
> 
> Dago recently built our ncurses with wide character support (thanks
> btw.), so that we now also have w variants of the ncurses libraries in
> addition to the regular ones (not binary compatible but as they have a w
> suffix there's no harm done). Both in the same package.
> 
> Coming to the include files. According to the ncurses README:
> 
> If you configure using the --enable-widec option, a "w" is appended to
> the library names (e.g., libncursesw.a), and the resulting libraries
> support wide-characters, e.g., via a UTF-8 locale.  The _corresponding
> header files are compatible with the non-wide-character configuration_;
> wide-character features are provided by ifdef's in the header files.
> 
> I don't exactly know how to read the header bit WRT packages building
> against these headers. Can non-wide and wide applications both use the
> same headers and ./configure sets #defines in case wide-support is
> requested by an app (that's how i would read it)?
> 
> At least we should have the original include/ncurses directory back in
> the package - presumably filled with the header files of the non-wide
> build (as a safe option). And then in addition have include/ncursesw
> populated with the header files of the wide variant?
> 
> Sebastian
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I have not figured out exactly how this is happening,
I may be totally off base, but looking in the libraries and the error messages
from the import I can't see any other way these errors should/could pop up.

I just updated my sparc sandbox and will do some more intensive testing
to figure it out.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknVBJ8ACgkQLrhmsXMSLxeROACePW2N3j760Pl4o9BLsEHPbhIt
3hMAoOVL9pRa7URKeZVlhAogUHRK/S8c
=5NgB
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr  2 22:08:16 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 15:08:16 -0500
Subject: [csw-maintainers] pulled gcc4 from testing
Message-ID: <49D51B30.8030006@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

problems with the libraries, not sure exactly what happened.
I am rebuilding using gcc to get all the packages in one recipe


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknVGy8ACgkQLrhmsXMSLxfuSACfQ3RnASa+qf3Aqsp1DhAnEb40
MEIAoJxjpEcora3UrMrE5+tkJmso8Fsp
=paFp
-----END PGP SIGNATURE-----


From rupert at opencsw.org  Fri Apr  3 12:53:35 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Fri, 3 Apr 2009 12:53:35 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
Message-ID: <6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>

hi,

we tried to upgrade our csw local zone resently, and it seems
cswclassutils uses an additional path than the pre-announced /opt/csw.
we have a readonly global zone and got the /opt/csw free to do a mount
loopback:

/etc/opt/csw/init.d/csw.smf.sample
/opt/csw/share/doc/cswclassutils/LICENSE
/opt/csw/share/doc/cswclassutils/README.CSW
/usr/sadm/install/scripts/i.cswcpsampleconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/i.cswinitsmf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswinitsmf> for writing: (30) Read-only
file system
/usr/sadm/install/scripts/i.cswpreserveconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswpreserveconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/i.cswusergroup
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswusergroup> for writing: (30) Read-only
file system
/usr/sadm/install/scripts/r.cswcpsampleconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswcpsampleconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/r.cswinitsmf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswinitsmf> for writing: (30) Read-only
file system
/usr/sadm/install/scripts/r.cswpreserveconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswpreserveconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/r.cswusergroup
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswusergroup> for writing: (30) Read-only
file system
[ verifying class <none> ]
ERROR: attribute verification of
</usr/sadm/install/scripts/i.cswcpsampleconf> failed
? ?pathname does not exist
ERROR: attribute verification of </usr/sadm/install/scripts/i.cswinitsmf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/i.cswpreserveconf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/i.cswusergroup> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/r.cswcpsampleconf> failed
? ?pathname does not exist
ERROR: attribute verification of </usr/sadm/install/scripts/r.cswinitsmf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/r.cswpreserveconf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/r.cswusergroup> failed
? ?pathname does not exist


From bonivart at opencsw.org  Fri Apr  3 12:57:43 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Fri, 3 Apr 2009 12:57:43 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
Message-ID: <625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>

On Fri, Apr 3, 2009 at 12:53 PM, rupert THURNER <rupert at opencsw.org> wrote:
> hi,
>
> we tried to upgrade our csw local zone resently, and it seems
> cswclassutils uses an additional path than the pre-announced /opt/csw.
> we have a readonly global zone and got the /opt/csw free to do a mount
> loopback:
>
> /etc/opt/csw/init.d/csw.smf.sample
> /opt/csw/share/doc/cswclassutils/LICENSE
> /opt/csw/share/doc/cswclassutils/README.CSW
> /usr/sadm/install/scripts/i.cswcpsampleconf
> pkgadd.ORG: ERROR: unable to open
> </usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
> Read-only file system

As far as I know class action scripts only work from
/usr/sadm/install/scripts so you need to install CSWcswclassutils from
the global zone, then the dependency will be fulfilled in all
non-global zones.

-- 
/peter


From mwatters at opencsw.org  Fri Apr  3 22:03:11 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 03 Apr 2009 15:03:11 -0500
Subject: [csw-maintainers] New Python in Testing
Message-ID: <49D66B7F.3000101@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	FIX: bug 3510 "...import curses fails" ( actually fixed this time )
	     I am able to import EVERY module that is bundled with python ;)

	FIX: bug 3579 RPATH problems.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknWa38ACgkQLrhmsXMSLxdL8ACgslSu0C2V0rpIIXKTDpEgCixQ
XNEAoIezAYzu+tEsKh5TqJY5rABM0/jH
=H3p7
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Fri Apr  3 22:13:47 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 03 Apr 2009 15:13:47 -0500
Subject: [csw-maintainers] php5 5.2.9 now in testing
Message-ID: <49D66DFB.7010408@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	fixed bugs
	3587: Extra Paths in package
	3499: xsl module broken
	3479: uncomment extensions

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknWbfsACgkQLrhmsXMSLxe9EQCguZ3K7fV4ozplhN7+WAfOsJRb
3DEAoN5GM48Tn79v+2csfnbKtCOQm1at
=+nuL
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sat Apr  4 18:27:24 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 04 Apr 2009 11:27:24 -0500
Subject: [csw-maintainers] pysqlite and pysqlite2 now in testing
Message-ID: <49D78A6C.1080502@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	recompile to fix isalist

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknXimwACgkQLrhmsXMSLxer9ACg5nx6CwXY7IR7Qn/sLNcJ+fx7
cdkAoJZPjh1G2dV+BfNA2DYIcCU59dm0
=RAaL
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sun Apr  5 09:32:26 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 05 Apr 2009 02:32:26 -0500
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
Message-ID: <49D85E8A.4040709@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Is there a way we can get the summary in addition to the link.

Example:

Here is a summary of your open CSW Mantis issues.  If you are not the
maintainer of these projects please inform <bug-admin at opencsw.org>.

**********************************************************************
These reports have severity "major" or above:
- ---------------------------------------------
Summary: summary major bug 1
   Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}


**********************************************************************
These reports have severity "minor" or below:
- ---------------------------------------------
Summary: summary minor bug 1
   Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}





- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknYXooACgkQLrhmsXMSLxd9IwCg28u6/9fZ5kBS3HivHpWKvVh7
VXUAmwdwkBQY22l9NFWT/N8VLU6XdBde
=yjGi
-----END PGP SIGNATURE-----


From hson at opencsw.org  Sun Apr  5 10:24:18 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Sun, 05 Apr 2009 10:24:18 +0200
Subject: [csw-maintainers] zlib package in /testing
Message-ID: <49D86AB2.5080005@opencsw.org>

I've put a zlib package in /testing which is the same version as the 
current one, but this is built with gar plus have a 64bit library.

Unless someone finds any trouble with it I'll release it before easter


From james at opencsw.org  Sun Apr  5 10:51:23 2009
From: james at opencsw.org (James Lee)
Date: Sun, 05 Apr 2009 08:51:23 GMT
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <49D85E8A.4040709@opencsw.org>
References: <49D85E8A.4040709@opencsw.org>
Message-ID: <20090405.8512300.4214488583@gyor.oxdrove.co.uk>

On 05/04/09, 08:32:26, Mike Watters <mwatters at opencsw.org> wrote regarding 
[csw-maintainers] RFE: Mantis Bug Reminder:

> Is there a way we can get the summary in addition to the link.

...

> Summary: summary major bug 1
>    Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}

The number works for me but if you have trouble learning phone
books it can be added.

You are supposed to be working on this list so it should familiar
and more importantly short or not there!



James.



From james at opencsw.org  Sun Apr  5 10:57:29 2009
From: james at opencsw.org (James Lee)
Date: Sun, 05 Apr 2009 08:57:29 GMT
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D86AB2.5080005@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
Message-ID: <20090405.8572900.3105252976@gyor.oxdrove.co.uk>

On 05/04/09, 09:24:18, "Roger H?kansson" <hson at opencsw.org> wrote regarding 
[csw-maintainers] zlib package in /testing:

> I've put a zlib package in /testing which is the same version as the
> current one, but this is built with gar plus have a 64bit library.

The exists has 64 bit libs (both sparcv9 and amd64)

Please set an SONAME.

There is no need for an RPATH as it doesn't use any CSW libs.

$ dump -Lv /usr/lib/libz.so | egrep 'RPATH|SONAME'
[4]	SONAME          libz.so.1




James.






From hson at opencsw.org  Sun Apr  5 11:42:57 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Sun, 05 Apr 2009 11:42:57 +0200
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <20090405.8572900.3105252976@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
Message-ID: <49D87D21.4050606@opencsw.org>

James Lee wrote:
> On 05/04/09, 09:24:18, "Roger H?kansson" <hson at opencsw.org> wrote regarding 
> [csw-maintainers] zlib package in /testing:
> 
>> I've put a zlib package in /testing which is the same version as the
>> current one, but this is built with gar plus have a 64bit library.
> 
> The exists has 64 bit libs (both sparcv9 and amd64)

How could I miss that...

> There is no need for an RPATH as it doesn't use any CSW libs.


Is it really a problem if it has RPATH set? (and its correctly set)
"Unsetting" (in general, not just for this package) it would mean that 
each gar package (which don't use any CSW libs) would have to have a 
LD_OPTION rewrite rule, since gar sets it by default?

I have no problem doing this, but I'm just thinking if were creating 
unnecessary work.

 > Please set an SONAME.
 >
> $ dump -Lv /usr/lib/libz.so | egrep 'RPATH|SONAME'
> [4]	SONAME          libz.so.1
> 

Fixed.



From mwatters at opencsw.org  Sun Apr  5 17:33:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 05 Apr 2009 10:33:49 -0500
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <20090405.8512300.4214488583@gyor.oxdrove.co.uk>
References: <49D85E8A.4040709@opencsw.org>
	<20090405.8512300.4214488583@gyor.oxdrove.co.uk>
Message-ID: <49D8CF5D.3080205@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

James Lee wrote:
> On 05/04/09, 08:32:26, Mike Watters <mwatters at opencsw.org> wrote regarding 
> [csw-maintainers] RFE: Mantis Bug Reminder:
> 
>> Is there a way we can get the summary in addition to the link.
> 
> ...
> 
>> Summary: summary major bug 1
>>    Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}
> 
> The number works for me but if you have trouble learning phone
> books it can be added.
> 
> You are supposed to be working on this list so it should familiar
> and more importantly short or not there!
> 
> 
> 
> James.
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

it is not important.
the bugs are usually not open long enough for me to memorize the number. ;)


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknYz10ACgkQLrhmsXMSLxeS/wCgk431s94W/cOCojpsSPB8bmEr
s/AAnRLhsXCOk0ltYt9qaBvPBFyIM7N5
=u/JN
-----END PGP SIGNATURE-----


From phil at bolthole.com  Sun Apr  5 17:59:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 5 Apr 2009 08:59:45 -0700
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D87D21.4050606@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
Message-ID: <20090405155945.GA9321@bolthole.com>

On Sun, Apr 05, 2009 at 11:42:57AM +0200, Roger H?kansson wrote:
> Is it really a problem if it has RPATH set? (and its correctly set)

well, technically, if it doesnt use any csw libs, you could probably make
it a fully relocatable package.

however, relocatable packages shouldnt have a fixed RPATH set :)

(does gar support making relocatable packages?)


From jgoerzen at opencsw.org  Sun Apr  5 19:01:23 2009
From: jgoerzen at opencsw.org (Jake Goerzen)
Date: Sun, 5 Apr 2009 10:01:23 -0700
Subject: [csw-maintainers] new fltk pkgs in /testing
Message-ID: <315c02ae0904051001u441554a9u38bd240b0b51f37a@mail.gmail.com>

new fltk pkgs are available in /testing:

fltk-1.1.9,REV=2009.04.05-SunOS5.8-i386-CSW.pkg.gz
fltk-1.1.9,REV=2009.04.05-SunOS5.8-sparc-CSW.pkg.gz

These pkgs should resolve the following bugs:

0000976 -  need fltk compiled with OpenGL (Mesa)
0001164 -  Incompatible with flphoto
0001982 -  local paths in config files

Testing and feedback appreciated
Thanks, Jake


From jgoerzen at opencsw.org  Sun Apr  5 19:07:02 2009
From: jgoerzen at opencsw.org (Jake Goerzen)
Date: Sun, 5 Apr 2009 10:07:02 -0700
Subject: [csw-maintainers] new fltk pkgs in /testing
Message-ID: <315c02ae0904051007va81ee68uc6d3a8500eabcb83@mail.gmail.com>

Sorry, I over looked that 2 of the listed bugs are already closed bugs.
So, this release of fltk pkgs should fix bug id # 0001982

Thanks, Jake


From jgoerzen at opencsw.org  Sun Apr  5 19:26:16 2009
From: jgoerzen at opencsw.org (Jake Goerzen)
Date: Sun, 5 Apr 2009 10:26:16 -0700
Subject: [csw-maintainers] new libsdl pkgs in /testing
Message-ID: <315c02ae0904051026l57e1cb10h740e213303f858a@mail.gmail.com>

new libsdl pkgs available in /testing:

libsdl-1.2.13,REV=2009.04.04-SunOS5.8-i386-CSW.pkg.gz
libsdl-1.2.13,REV=2009.04.04-SunOS5.8-sparc-CSW.pkg.gz

These pkgs I hope fix missing symlink issues:

0003048 -  libSDL-1.2.so.0 doe snot link the good lib
0003518 -  link libSDL-1.2.so.0 not pointing to the right lib

Thanks for testing, Jake


From dam at opencsw.org  Sun Apr  5 23:00:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 5 Apr 2009 23:00:40 +0200
Subject: [csw-maintainers] Updating all packages on all buildfarm servers to
	current now.
Message-ID: <105A89E8-CB1F-4C47-9062-FD162748DD8D@opencsw.org>

Hi,

I am updating all packages on all buildfarm servers to current now
to reflect the recent packages releases.


Best regards

   -- Dago


From dam at opencsw.org  Sun Apr  5 23:17:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 5 Apr 2009 23:17:17 +0200
Subject: [csw-maintainers] RPATH in GAR
In-Reply-To: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
References: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
Message-ID: <C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>

Hi,

Am 02.04.2009 um 14:22 schrieb Dagobert Michelsen:
> The usage of the rewritten module is documented at
>  <http://apps.sourceforge.net/trac/gar/wiki/RuntimeLinkerPathes>
>
> In the easiest case it should be sufficient to do just a
> rebuild and everything should be fine. However, the new
> structure gives you fine-grained control of the pathes
> to be added.
>
> Feel free to ask if there are any issues.

I found another issue: if an application adds somehow runtime linker
pathes themselves the resulting binary contains pathes similar to
   /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib
beginning with the one or two correct pathes from LD_OPTIONS and then
some redundant stuff added from the application. James proposed to use
some compiler-interposer which strips and reorders certain flags before
passing them to the real compiler. If someone has a less-intrusive idea
then it would be taken very well :-)


Best regards

   -- Dago


From bwalton at opencsw.org  Sun Apr  5 23:36:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 05 Apr 2009 17:36:59 -0400
Subject: [csw-maintainers] RPATH in GAR
In-Reply-To: <C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>
References: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
	<C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>
Message-ID: <1238967347-sup-9259@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Apr 05 17:17:17 -0400 2009:
> I found another issue: if an application adds somehow runtime linker
> pathes themselves the resulting binary contains pathes similar to
>    /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib

...and ruby is now sticking a /lib before everything else somehow too,
which is now starting to really tick me off.

Other than being wasteful and redundant, is the duplication of the
paths harmful?

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090405/6eeed85c/attachment.asc>

From mwatters at opencsw.org  Mon Apr  6 01:15:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 05 Apr 2009 18:15:33 -0500
Subject: [csw-maintainers] gcc-4.3.3 in testing
Message-ID: <49D93B95.5020307@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

make check results checked into subversion at:
http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/gcc4/trunk/files/test-results/?pathrev=4204

Please test and provide feedback.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknZO5UACgkQLrhmsXMSLxcbGQCggUZpkWVvTT0xJSPqJJp1kx1k
GFEAoObNnhfWMcnT0PIWgcFo1atpkbuj
=56PD
-----END PGP SIGNATURE-----


From phil at bolthole.com  Mon Apr  6 05:10:01 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 5 Apr 2009 20:10:01 -0700
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
Message-ID: <20090406031001.GA1795@bolthole.com>

a reminder for those who may not realize it:

If you want a "summary" of your bugs... just make sure that they are at
least all "assigned" to you... then you can go to "my view", and select
"all projects", or "view all bugs" page. and select project of "all
projects"

you'll then get a summary with bugids AND summary line.


From dam at opencsw.org  Mon Apr  6 09:58:54 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 09:58:54 +0200
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <20090406031001.GA1795@bolthole.com>
References: <20090406031001.GA1795@bolthole.com>
Message-ID: <E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>

Hi,

Am 06.04.2009 um 05:10 schrieb Philip Brown:
> a reminder for those who may not realize it:
>
> If you want a "summary" of your bugs... just make sure that they are  
> at
> least all "assigned" to you... then you can go to "my view", and  
> select
> "all projects", or "view all bugs" page. and select project of "all
> projects"
>
> you'll then get a summary with bugids AND summary line.

...and of course at the bottom of your maintainer homepage an individual
listing of all bugs, e. g. for me
   http://www.opencsw.org/buglist/maintainer.cgi?maintainer=dam


Best regards

   -- Dago


From james at opencsw.org  Mon Apr  6 10:30:51 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 08:30:51 GMT
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D87D21.4050606@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
Message-ID: <20090406.8305100.1838829892@gyor.oxdrove.co.uk>

On 05/04/09, 10:42:57, "Roger H?kansson" <hson at opencsw.org> wrote regarding
Re: [csw-maintainers] zlib package in /testing:

> > There is no need for an RPATH as it doesn't use any CSW libs.

> Is it really a problem if it has RPATH set? (and its correctly set)
> "Unsetting" (in general, not just for this package) it would mean that
> each gar package (which don't use any CSW libs) would have to have a
> LD_OPTION rewrite rule, since gar sets it by default?

> I have no problem doing this, but I'm just thinking if were creating
> unnecessary work.

It's not correct as no CSW libs are opened.

It's not a "problem" but if it's the result of the build system it
suggests we are generally making settings with no thought.  The
current batch of odd RPATHs only causes the link loader to look for
a directory that doesn't exist, but this is less wasted effort than
what happens with $ISALIST which typically looks at all the 64 bit
locations before finding that matching 32 lib.


How difficult is it to do right?  Perhaps the recent batch of odd
paths are telling me that people are using a "sausage machine"
approach to building (throw anything in the top and just wind the
handle).  We are dangerously close to the "I'd like to do the right
thing but the build system won't let me".  The value CSW should add
*is* to take the time to get things right - your individual effort
shared for the benefit of many.



>  > Please set an SONAME.
>  >
> > $ dump -Lv /usr/lib/libz.so | egrep 'RPATH|SONAME'
> > [4] SONAME          libz.so.1
> >

> Fixed.

Thank you.




James.


From james at opencsw.org  Mon Apr  6 10:36:14 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 08:36:14 GMT
Subject: [csw-maintainers] RPATH in GAR
In-Reply-To: <1238967347-sup-9259@ntdws12.chass.utoronto.ca>
References: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
	<C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>
	<1238967347-sup-9259@ntdws12.chass.utoronto.ca>
Message-ID: <20090406.8361400.4103879239@gyor.oxdrove.co.uk>

On 05/04/09, 22:36:59, Ben Walton <bwalton at opencsw.org> wrote regarding Re:
[csw-maintainers] RPATH in GAR:

> > I found another issue: if an application adds somehow runtime linker
> > pathes themselves the resulting binary contains pathes similar to
> >    /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib


> Other than being wasteful and redundant, is the duplication of the
> paths harmful?

No.



James.


From james at opencsw.org  Mon Apr  6 10:36:16 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 08:36:16 GMT
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>
References: <20090406031001.GA1795@bolthole.com>
	<E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>
Message-ID: <20090406.8361600.4264367480@gyor.oxdrove.co.uk>

On 06/04/09, 08:58:54, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
Re: [csw-maintainers] RFE: Mantis Bug Reminder:

> Am 06.04.2009 um 05:10 schrieb Philip Brown:
> > a reminder for those who may not realize it:
> >
> > If you want a "summary" of your bugs... just make sure that they are
> > at
> > least all "assigned" to you... then you can go to "my view", and
> > select
> > "all projects", or "view all bugs" page. and select project of "all
> > projects"
> >
> > you'll then get a summary with bugids AND summary line.

> ...and of course at the bottom of your maintainer homepage an individual
> listing of all bugs, e. g. for me
>    http://www.opencsw.org/buglist/maintainer.cgi?maintainer=dam


I later realised this too, the reminder could just say:
"You have a|some|lots|loads|excessive bugs, go to the mantis web page".
Listing the links just makes it look proportionate.



James.





From rupert at opencsw.org  Mon Apr  6 11:48:38 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 6 Apr 2009 11:48:38 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
Message-ID: <6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>

what would happen if one lives without classutils?


On Fri, Apr 3, 2009 at 12:57, Peter Bonivart <bonivart at opencsw.org> wrote:
> On Fri, Apr 3, 2009 at 12:53 PM, rupert THURNER <rupert at opencsw.org> wrote:
>> hi,
>>
>> we tried to upgrade our csw local zone resently, and it seems
>> cswclassutils uses an additional path than the pre-announced /opt/csw.
>> we have a readonly global zone and got the /opt/csw free to do a mount
>> loopback:
>>
>> /etc/opt/csw/init.d/csw.smf.sample
>> /opt/csw/share/doc/cswclassutils/LICENSE
>> /opt/csw/share/doc/cswclassutils/README.CSW
>> /usr/sadm/install/scripts/i.cswcpsampleconf
>> pkgadd.ORG: ERROR: unable to open
>> </usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
>> Read-only file system
>
> As far as I know class action scripts only work from
> /usr/sadm/install/scripts so you need to install CSWcswclassutils from
> the global zone, then the dependency will be fulfilled in all
> non-global zones.
>
> --
> /peter
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From dam at opencsw.org  Mon Apr  6 14:30:25 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 14:30:25 +0200
Subject: [csw-maintainers] Shutdown of login,
	build* in 30 minutes for upgrade
Message-ID: <3B9E6F40-E42B-4BDB-B1F0-2A242CE43BD4@opencsw.org>

Hi,

I am going to upgrade the T5220 hosting login, build8s, build9s,
build10s, web, hudson and other stuff from 8 GB to 16 GB Ram
in 30 minutes. Please stand by.


Best regards

   -- Dago


From hson at opencsw.org  Mon Apr  6 14:35:24 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Mon, 06 Apr 2009 14:35:24 +0200
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <20090406.8305100.1838829892@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>
	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>
Message-ID: <49D9F70C.9030007@opencsw.org>

James Lee wrote:
 > It's not correct as no CSW libs are opened.
 >
> It's not a "problem" but if it's the result of the build system it
> suggests we are generally making settings with no thought.  The
> current batch of odd RPATHs only causes the link loader to look for
> a directory that doesn't exist, but this is less wasted effort than
> what happens with $ISALIST which typically looks at all the 64 bit
> locations before finding that matching 32 lib.

$ISALIST doesn't just expand to 64bit, you have optimized 32bit versions 
(sparcv8plus, sparcv8plus+vis...) as well?

I take it that you haven't built any packages using gar yet?
The recent batch of odd paths was due to problems with gar where 
LD_OPTIONS somehow got messed up (there have been a number of postings 
about this problem).
Even though I've only been a maintainer for a few months, my 
understanding (from http://opencsw.org/standards/pkg-walkthrough as an 
example) was that LD_OPTIONS was a preferred setting when building 
packages. So all packages I've built (either using gar or manually) have 
LD_OPTIONS set which means RPATH/RUNPATH have been set to 
/opt/csw/lib/$ISALIST:/opt/csw/lib whatever they need.

If LD_OPTIONS shouldn't be used, or should only contain what's needed 
for a particular package, it should be communicated more clearly.

> How difficult is it to do right?  Perhaps the recent batch of odd
> paths are telling me that people are using a "sausage machine"
> approach to building (throw anything in the top and just wind the
> handle).  We are dangerously close to the "I'd like to do the right
> thing but the build system won't let me".  The value CSW should add
> *is* to take the time to get things right - your individual effort
> shared for the benefit of many.

Right now, as far as I can see it, gar lets you set LD_OPTIONS to 
anything you want, but if you don't set it, gar will set it for you.

To reiterate my question:
Is there any actual harm (disregarding the fact that the link loader 
might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without 
finding any match) having RPATH/RUNPATH set?

If not, it is only when a package doesn't use any libs at all from 
/opt/csw/lib, when RPATH/RUNPATH should be changed from the default.

If package A is linked to a lib i package B (both CSW packages) and A i 
packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future 
releases of B which contain optimized libraries won't be used until A is 
repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/csw/lib
And if there is a longer dependency list (A->B->C-D where D is the one 
with optimized libs) it creates much more work for the maintainer of A 
(he must check every lib in the dependency path to see if there are any 
optimized libs).

But then again, if there is no harm in having RPATH/RUNPATH set and it 
only creates much more work for the maintainers, why shouldn't we leave 
it as it is?


From james at opencsw.org  Mon Apr  6 16:34:50 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 14:34:50 GMT
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D9F70C.9030007@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>
	<49D9F70C.9030007@opencsw.org>
Message-ID: <20090406.14345000.1346203350@gyor.oxdrove.co.uk>

On 06/04/09, 13:35:24, "Roger H?kansson" <hson at opencsw.org> wrote regarding
Re: [csw-maintainers] zlib package in /testing:

> > It's not a "problem" but if it's the result of the build system it
> > suggests we are generally making settings with no thought.  The
> > current batch of odd RPATHs only causes the link loader to look for
> > a directory that doesn't exist, but this is less wasted effort than
> > what happens with $ISALIST which typically looks at all the 64 bit
> > locations before finding that matching 32 lib.

> $ISALIST doesn't just expand to 64bit, you have optimized 32bit versions
> (sparcv8plus, sparcv8plus+vis...) as well?

It expands to the isalist.  It tries all in order but if it's a 32 bit
prog there is never a valid match in the 64s, then it runs through the
32s until the match, or not in the case of this first non CSW lib:

$ truss /opt/csw/bin/gsc
execve("/opt/csw/bin/gsc", 0xFFBEFBDC, 0xFFBEFBE4)  argc = 1
resolvepath("/usr/lib/ld.so.1", "/usr/lib/ld.so.1", 1023) = 16
open("/var/ld/ld.config", O_RDONLY)             Err#2 ENOENT
sysinfo(514, "sparcv9+vis2 sparcv9+vis sparcv9 sparcv8plus+vis
sparcv8plus sparcv8 sparcv8-fsmuld sparcv7 sparc", 257) = 98
stat("/opt/csw/lib/sparcv9+vis2/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv9+vis/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv9/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8plus+vis/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8plus/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8-fsmuld/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv7/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparc/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/libc.so.1", 0xFFBEF488)      Err#2 ENOENT
stat("/usr/lib/libc.so.1", 0xFFBEF488)          = 0
resolvepath("/usr/lib/libc.so.1", "/usr/lib/libc.so.1", 1023) = 18
open("/usr/lib/libc.so.1", O_RDONLY)            = 3
...

Lots of Err#2


> Even though I've only been a maintainer for a few months, my
> understanding (from http://opencsw.org/standards/pkg-walkthrough as an
> example) was that LD_OPTIONS was a preferred setting when building
> packages. So all packages I've built (either using gar or manually) have
> LD_OPTIONS set which means RPATH/RUNPATH have been set to
> /opt/csw/lib/$ISALIST:/opt/csw/lib whatever they need.

> If LD_OPTIONS shouldn't be used, or should only contain what's needed
> for a particular package, it should be communicated more clearly.

Once built a package doesn't know how it was built so if it works
use it.

LD_OPTIONS overrides all so is powerful.  Like all powerful weapons
use with caution.

$ man ld
...
     LD_OPTIONS
           A default set of options to ld. LD_OPTIONS  is  inter-
           preted  by ld just as though its value had been placed
           on the command line, immediately  following  the  name
           used to invoke ld, as in:

           ld $LD_OPTIONS ... other-arguments ...


> To reiterate my question:
> Is there any actual harm (disregarding the fact that the link loader
> might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without
> finding any match) having RPATH/RUNPATH set?

No, well it's actual but trivial.


> If not, it is only when a package doesn't use any libs at all from
> /opt/csw/lib, when RPATH/RUNPATH should be changed from the default.

Correct, as is the case with zlib.

> If package A is linked to a lib i package B (both CSW packages) and A i
> packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future
> releases of B which contain optimized libraries won't be used until A is
> repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/csw/lib

Correct.  Add $ISALIST in, unless you know it won't be needed.


> And if there is a longer dependency list (A->B->C-D where D is the one
> with optimized libs) it creates much more work for the maintainer of A
> (he must check every lib in the dependency path to see if there are any
> optimized libs).

D is governed by the RPATH of C.  C by B.



James.


From bonivart at opencsw.org  Mon Apr  6 16:43:45 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Apr 2009 16:43:45 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com> 
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com> 
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
Message-ID: <625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>

On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
> what would happen if one lives without classutils?

Disregarding that you have to force past the dependency requirement it
would be the equivalent of not running pre/post scripts. The packages
that depend on cswclassutils use it for some combo of e.g. init/SMF
support and handling of configuration files, see
http://wiki.opencsw.org/cswclassutils-package for all features.

What's the problem to install it in the global zone? Even if you don't
have root access to it yourself you should be able to make a request
for a package install, right? What would you do with a sparse zone if
you needed a SUNW package that installed in /usr?

-- 
/peter


From dam at opencsw.org  Mon Apr  6 16:49:23 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 16:49:23 +0200
Subject: [csw-maintainers] Shutdown of login,
	build* in 30 minutes for upgrade
In-Reply-To: <3B9E6F40-E42B-4BDB-B1F0-2A242CE43BD4@opencsw.org>
References: <3B9E6F40-E42B-4BDB-B1F0-2A242CE43BD4@opencsw.org>
Message-ID: <3B5B2F86-9763-497F-A8E6-E9511D9C33B2@opencsw.org>

Hi,

Am 06.04.2009 um 14:30 schrieb Dagobert Michelsen:
> I am going to upgrade the T5220 hosting login, build8s, build9s,
> build10s, web, hudson and other stuff from 8 GB to 16 GB Ram
> in 30 minutes. Please stand by.

Sorry for the extended downtime. The issue proved to be more
complicated than anticipated because there already were
DataRam memory in the system and the new memory was Sun
memory. The type of memory is actually compared during POST
and mixing memory leads to immediate shutdown :-( However,
certain bank combinations do work and the system now has
16 GB of Ram.

Happy packaging :-)

   -- Dago


From Darin.Perusich at cognigencorp.com  Mon Apr  6 16:56:31 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 06 Apr 2009 10:56:31 -0400
Subject: [csw-maintainers] inetd.conf entries for solaris 10
Message-ID: <49DA181F.8080305@cognigencorp.com>

Where can I find documentation/examples for dealing with inetd.conf/SMF
entries? I need to convert a few inetd.conf entries and while using
inetconv will give me what I need I'd prefer to provide the manifest.

Thanks

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From bonivart at opencsw.org  Mon Apr  6 17:18:05 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Apr 2009 17:18:05 +0200
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <49DA181F.8080305@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
Message-ID: <625385e30904060818x214be099r1f8d6485da577087@mail.gmail.com>

On Mon, Apr 6, 2009 at 4:56 PM, Darin Perusich
<Darin.Perusich at cognigencorp.com> wrote:
> Where can I find documentation/examples for dealing with inetd.conf/SMF
> entries? I need to convert a few inetd.conf entries and while using
> inetconv will give me what I need I'd prefer to provide the manifest.

I would also be interested in that for my qpopper package.

-- 
/peter


From dam at opencsw.org  Mon Apr  6 18:17:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 18:17:34 +0200
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <20090402163457.GI19124@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
Message-ID: <A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>

Hi,

Am 02.04.2009 um 18:34 schrieb Philip Brown:
> On Thu, Apr 02, 2009 at 11:27:48AM -0500, Mike Watters wrote:
>
>> for some reason, /opt/csw/include/ncurses does not exist?
>> it exists as /opt/csw/include/ncursesw
>
> sounds like a takeover maintainer bug

There is now an updated ncurses in testing which has both
include/ncurses and include/ncursesw.


Best regards

   -- Dago


From phil at bolthole.com  Mon Apr  6 19:14:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 10:14:32 -0700
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <20090406.8361600.4264367480@gyor.oxdrove.co.uk>
References: <20090406031001.GA1795@bolthole.com>
	<E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>
	<20090406.8361600.4264367480@gyor.oxdrove.co.uk>
Message-ID: <20090406171432.GE73510@bolthole.com>

On Mon, Apr 06, 2009 at 08:36:16AM +0000, James Lee wrote:
> > ...and of course at the bottom of your maintainer homepage an individual
> > listing of all bugs, e. g. for me
> >    http://www.opencsw.org/buglist/maintainer.cgi?maintainer=dam
> 
> 
> I later realised this too, the reminder could just say:
> "You have a|some|lots|loads|excessive bugs, go to the mantis web page".
> Listing the links just makes it look proportionate.

Eh.... for people who get A LOT of email...
sometimes, its nice to get an email with the specifics, so it catches your
eye and reminds you about a specific issue that is worth going to look at
"sooner", rather than "later".



From phil at bolthole.com  Mon Apr  6 19:37:46 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 10:37:46 -0700
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <49DA181F.8080305@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
Message-ID: <20090406173746.GL73510@bolthole.com>

On Mon, Apr 06, 2009 at 10:56:31AM -0400, Darin Perusich wrote:
> Where can I find documentation/examples for dealing with inetd.conf/SMF
> entries? I need to convert a few inetd.conf entries and while using
> inetconv will give me what I need I'd prefer to provide the manifest.
> 

in sol10, the inetd.conf stuff is kinda deprecated. 
you run inetdconv (or whatever), to slurp up whatever is in inetd.conf, and
  autogenerate a "real" smf definition for it.

You could just look at what was generated for SMF for some test run of the
above, and then make your package use that directly, instead of messing
with inetd.conf in sol10.

(Note to Peter B: bettter still would be running qpopper, etc, as a full
  demon autorestarted via SMF, if possible, rather than inetd, on sol10)




From Darin.Perusich at cognigencorp.com  Mon Apr  6 20:11:29 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 06 Apr 2009 14:11:29 -0400
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <20090406173746.GL73510@bolthole.com>
References: <49DA181F.8080305@cognigencorp.com>
	<20090406173746.GL73510@bolthole.com>
Message-ID: <49DA45D1.8060008@cognigencorp.com>


Philip Brown wrote:
> On Mon, Apr 06, 2009 at 10:56:31AM -0400, Darin Perusich wrote:
>> Where can I find documentation/examples for dealing with inetd.conf/SMF
>> entries? I need to convert a few inetd.conf entries and while using
>> inetconv will give me what I need I'd prefer to provide the manifest.
>>
> 
> in sol10, the inetd.conf stuff is kinda deprecated. 
> you run inetdconv (or whatever), to slurp up whatever is in inetd.conf, and
>   autogenerate a "real" smf definition for it.

This is my plan and I've already done so.

> You could just look at what was generated for SMF for some test run of the
> above, and then make your package use that directly, instead of messing
> with inetd.conf in sol10.

This is why I'm asking for documentation or an example of creating a
package in such a way. While SMF may be the "proper" way on 10 I'm going
to do what works, has always worked, and still does, init.d script and
inetconv for inetd things.

> (Note to Peter B: bettter still would be running qpopper, etc, as a full
>   demon autorestarted via SMF, if possible, rather than inetd, on sol10)
> 


-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Mon Apr  6 20:21:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 11:21:52 -0700
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <49DA45D1.8060008@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
	<20090406173746.GL73510@bolthole.com>
	<49DA45D1.8060008@cognigencorp.com>
Message-ID: <20090406182152.GP73510@bolthole.com>

On Mon, Apr 06, 2009 at 02:11:29PM -0400, Darin Perusich wrote:
> This is why I'm asking for documentation or an example of creating a
> package in such a way. While SMF may be the "proper" way on 10 I'm going
> to do what works, has always worked, and still does, init.d script and
> inetconv for inetd things.

oh... so to be specific, you're looking at examples of how to handle stuff
in inetd.conf for sol8/9, IN CONTRAST to the solaris 10 way?

That rather clashes with the subject line :-}

I'm not sure we really have a set way. I think most demons we have packages,
either avoid inetd entirely, or just leave it up to the user to 
edit the file when and if they want to.

You COULD make a package for it, that does nothing to the inetd conf.
Or you could make an explicit postinstall script.

Or... you could be "noble", and make a new classutils class action script
for it :)

so on sol8/9 it would auto insert/update/.... inetd.conf, and
  in sol10 it would auto SMF it.

you could allow for a file like for the cswusergroup class....

/opt/csw/etc/pkg/[pkgname]/cswinetdconf

with one or more lines in inetdconf style, that would be the trigger
for the class action script.
Although you would also potentially require some entries for any
"new" /etc/services entries, if they were required.



From bwalton at opencsw.org  Mon Apr  6 20:32:06 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 06 Apr 2009 14:32:06 -0400
Subject: [csw-maintainers] ruby in testing
Message-ID: <1239042541-sup-7577@ntdws12.chass.utoronto.ca>


Hi All,

Updated ruby packages are available from testing.  These address the
RPATH bugs (eg: they now properly pick up any optimized libraries) and
have relocated the libruby.so symlink from rubydev to ruby to avoid
breaking some older packages.

Please let me know if you have any issues with these.

Thanks
-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090406/2773b086/attachment.asc>

From william at wbonnet.net  Mon Apr  6 21:45:49 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 06 Apr 2009 21:45:49 +0200
Subject: [csw-maintainers] Packages age statistics : March '09
Message-ID: <49DA5BED.9010908@wbonnet.net>

Hi

Here is the March update of package age statistics for the current source.

The third column show the delta between this month and previous month, the last one the cummulated delta over the year.

As you can notice packages total from years 2003 to 2008 decreased,
which is good (they have been updated). Delta of packages in 2009 is
+239, which is excellent, the best ever we achieved.

This following statistics show that 136 packages have been update over 
the last month, and 103 new packages added. 

Once again this is excellent. Thanks to you all.

Year    Total      Delta      Year      
1997      1          0          0
1998      1          0          0
2001      3          0          0
2002      4          0          0
2003     23         -2         -3
2004     99         -8         -8
2005    174         -2         -5
2006    184        -16        -20
2007    160        -24        -39
2008    304        -84       -131
2009    361       +239       +383

Cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From rmacduff at opencsw.org  Mon Apr  6 21:53:28 2009
From: rmacduff at opencsw.org (Ross Macduff)
Date: Mon, 06 Apr 2009 15:53:28 -0400
Subject: [csw-maintainers] gsed in testing
Message-ID: <1239047388-sup-4875@frog.chass.utoronto.ca>

Hello,

A revised version of gsed 4.1.5 is in testing.  This version does not
contain ISALIST in RPATH.

Ross


From Darin.Perusich at cognigencorp.com  Mon Apr  6 21:59:00 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 06 Apr 2009 15:59:00 -0400
Subject: [csw-maintainers] migrating inetd.conf entries to SMF for sol10
In-Reply-To: <20090406182152.GP73510@bolthole.com>
References: <49DA181F.8080305@cognigencorp.com>	<20090406173746.GL73510@bolthole.com>	<49DA45D1.8060008@cognigencorp.com>
	<20090406182152.GP73510@bolthole.com>
Message-ID: <49DA5F04.9010002@cognigencorp.com>

I changed the subject to better describe what I'm looking to do :-).

Philip Brown wrote:
> On Mon, Apr 06, 2009 at 02:11:29PM -0400, Darin Perusich wrote:
>> This is why I'm asking for documentation or an example of creating a
>> package in such a way. While SMF may be the "proper" way on 10 I'm going
>> to do what works, has always worked, and still does, init.d script and
>> inetconv for inetd things.
> 
> oh... so to be specific, you're looking at examples of how to handle stuff
> in inetd.conf for sol8/9, IN CONTRAST to the solaris 10 way?

I'm currently using postinstall script to check for the entries and add
them if they don't exist and nothing special for Solaris 10.

> You COULD make a package for it, that does nothing to the inetd conf.
> Or you could make an explicit postinstall script.
> 
> Or... you could be "noble", and make a new classutils class action script
> for it :)

This is what I was looking for, where are the docs/examples explaining
how this classutils works?

> so on sol8/9 it would auto insert/update/.... inetd.conf, and
>   in sol10 it would auto SMF it.
> 
> you could allow for a file like for the cswusergroup class....
> 
> /opt/csw/etc/pkg/[pkgname]/cswinetdconf
> 
> with one or more lines in inetdconf style, that would be the trigger
> for the class action script.
> Although you would also potentially require some entries for any
> "new" /etc/services entries, if they were required.

The existing postinstall already adds to services so that's taken care of.

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From william at wbonnet.net  Mon Apr  6 22:02:14 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 06 Apr 2009 22:02:14 +0200
Subject: [csw-maintainers] Thematics March '09 : Move to GAR
Message-ID: <49DA5FC6.6040700@wbonnet.net>

Hi all

The move to Gar month is now over, thank you all for your effort. As you 
can see from the statistics email, many packages have been added or 
updated (239 packages over March).

The current source contains 1939 packages, including 728 from the gar 
repository (38%). This is about 200 more than at the beginning of the 
month.

It is important to notice that the GAR repository is also having 261 
packages which are not yet released to current ! That would take us to 
2200 packages and 46% of packages from GAR.

Next milestone is to continue the GAR and package update. April 
thematics is package update, which is closely related to the GAR thematics.

Cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Mon Apr  6 22:07:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 13:07:59 -0700
Subject: [csw-maintainers] migrating inetd.conf entries to SMF for sol10
In-Reply-To: <49DA5F04.9010002@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
	<20090406173746.GL73510@bolthole.com>
	<49DA45D1.8060008@cognigencorp.com>
	<20090406182152.GP73510@bolthole.com>
	<49DA5F04.9010002@cognigencorp.com>
Message-ID: <20090406200759.GU73510@bolthole.com>

On Mon, Apr 06, 2009 at 03:59:00PM -0400, Darin Perusich wrote:
> > Or... you could be "noble", and make a new classutils class action script
> > for it :)
> 
> This is what I was looking for, where are the docs/examples explaining
> how this classutils works?

its a solaris standard mechanism.
so, docs.sun.com would be one place to start.

or... you could just look at the existing class action scripts in our
cswclassutils package, and modify one to suit your needs.

the concept is that you write a matching pair of
install/remove scripts which go in /usr/sadm/[whateveritis]

Then at pkgadd/pkgrm time, the script gets called, with a list of files
that are in "class YOURCLASS"

The script can then do basically anything it wants.
It's very similar to a postinstall script, but with a few more "input" and
"output" expectations. nothing too strenous though. again, see our existing
ones.


> > so on sol8/9 it would auto insert/update/.... inetd.conf, and
> >   in sol10 it would auto SMF it.
> > 
> > you could allow for a file like for the cswusergroup class....
> > 
> > /opt/csw/etc/pkg/[pkgname]/cswinetdconf
> > 
> > with one or more lines in inetdconf style, that would be the trigger
> > for the class action script.
> > Although you would also potentially require some entries for any
> > "new" /etc/services entries, if they were required.
> 
> The existing postinstall already adds to services so that's taken care of.

However, if you converted it to a class action script, it would be 
Very Nice for other programs that may like to use inetd. 

Plus, eliminating postinstall scripts is nice for not prompting sysadmins,
"do you realy want to run this script?"


From phil at bolthole.com  Mon Apr  6 22:10:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 13:10:50 -0700
Subject: [csw-maintainers] Thematics March '09 : Move to GAR
In-Reply-To: <49DA5FC6.6040700@wbonnet.net>
References: <49DA5FC6.6040700@wbonnet.net>
Message-ID: <20090406201050.GV73510@bolthole.com>

On Mon, Apr 06, 2009 at 10:02:14PM +0200, William Bonnet wrote:
>...
> It is important to notice that the GAR repository is also having 261  
> packages which are not yet released to current ! That would take us to  
> 2200 packages and 46% of packages from GAR.
>
> Next milestone is to continue the GAR and package update. April  
> thematics is package update, which is closely related to the GAR 
> thematics.

however, one big GAR-related (unstated)milestone stil to be met, is:

"make gar its own downloadable package".

Where the package would consist of basic, more userfriendly tools that
would look something like:

$ gar-build-from-src somesoftware

and gar would then go grab the source, download, configure, compile and
package that, on a non-opencsw.or build box, without further prompting.


The concept being, that "modular gar", would at last be TRUELY modular, and
not have a need to svn-download the whole svn tree to do something useful.



From rupert at opencsw.org  Mon Apr  6 22:17:45 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 6 Apr 2009 22:17:45 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
Message-ID: <6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>

On Mon, Apr 6, 2009 at 16:43, Peter Bonivart <bonivart at opencsw.org> wrote:
> On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
>> what would happen if one lives without classutils?
>
> Disregarding that you have to force past the dependency requirement it
> would be the equivalent of not running pre/post scripts. The packages
> that depend on cswclassutils use it for some combo of e.g. init/SMF
> support and handling of configuration files, see
> http://wiki.opencsw.org/cswclassutils-package for all features.
>
> What's the problem to install it in the global zone? Even if you don't
> have root access to it yourself you should be able to make a request
> for a package install, right? What would you do with a sparse zone if
> you needed a SUNW package that installed in /usr?

we would need to make a request. as we have 1000 machines, and only a
few of them run opencsw it would be denied.

i am wondering how it was done in the past? because we never had write
access and it always seemed to work. also the install description only
states /opt/csw as mount point.

rupert.


From phil at bolthole.com  Mon Apr  6 22:28:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 13:28:17 -0700
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
Message-ID: <20090406202817.GB51303@bolthole.com>

On Mon, Apr 06, 2009 at 10:17:45PM +0200, rupert THURNER wrote:
> On Mon, Apr 6, 2009 at 16:43, Peter Bonivart <bonivart at opencsw.org> wrote:
> > On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
> >> what would happen if one lives without classutils?
> >
> > Disregarding that you have to force past the dependency requirement it
> > would be the equivalent of not running pre/post scripts. The packages
> > that depend on cswclassutils use it for some combo of e.g. init/SMF
> > support and handling of configuration files, see
> > http://wiki.opencsw.org/cswclassutils-package for all features.
> >
> > What's the problem to install it in the global zone? Even if you don't
> > have root access to it yourself you should be able to make a request
> > for a package install, right? What would you do with a sparse zone if
> > you needed a SUNW package that installed in /usr?
> 
> we would need to make a request. as we have 1000 machines, and only a
> few of them run opencsw it would be denied.
> 
> i am wondering how it was done in the past? because we never had write
> access and it always seemed to work. also the install description only
> states /opt/csw as mount point.

"in the past", we didnt use the class script as much.

I would suggest that you make the request. you dont lose much by making the
request, no?

Afer that, as Peter B. said, you would have to run whatever is needful, by
hand.

Which is one reason I have been pushing to have the things called by the
class scripts, in /opt/csw rather than /etc.
This allows for tools to be written to do what is neccessary.

You are basically in the same circumstances as kind of a
 "NFS-shared /opt/csw" model. Perhaps this might motivate you to help
write some of those tools? :-)
The good news is, you would only have to do it once for each cswclassutils
script that we have.



From bonivart at opencsw.org  Mon Apr  6 22:39:58 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Apr 2009 22:39:58 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com> 
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com> 
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com> 
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com> 
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
Message-ID: <625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>

On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
> we would need to make a request. as we have 1000 machines, and only a
> few of them run opencsw it would be denied.

Why would it be denied? It's not like it would hurt the machines not
using CSW, it's something like a 20 KB package containing 10 files
which has to be specifically called to do anything.

Not even my company would deny that request. :-)

-- 
/peter


From rupert at opencsw.org  Mon Apr  6 22:50:31 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 6 Apr 2009 22:50:31 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
Message-ID: <6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>

On Mon, Apr 6, 2009 at 22:39, Peter Bonivart <bonivart at opencsw.org> wrote:
> On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
>> we would need to make a request. as we have 1000 machines, and only a
>> few of them run opencsw it would be denied.
>
> Why would it be denied? It's not like it would hurt the machines not
> using CSW, it's something like a 20 KB package containing 10 files
> which has to be specifically called to do anything.
>

because all machines look the same on the base layer. and upgrading
all machines is a major thing. and waiting for the next operating
system release as well (because it takes so much time).

it was already difficult to get the /opt/csw mountpoint.

but i'll try :)

rupert.


From phil at bolthole.com  Mon Apr  6 23:05:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 14:05:26 -0700
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
	<6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
Message-ID: <20090406210526.GC51303@bolthole.com>

On Mon, Apr 06, 2009 at 10:50:31PM +0200, rupert THURNER wrote:
> On Mon, Apr 6, 2009 at 22:39, Peter Bonivart <bonivart at opencsw.org> wrote:
> > On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
> >> we would need to make a request. as we have 1000 machines, and only a
> >> few of them run opencsw it would be denied.
> >
> > Why would it be denied? It's not like it would hurt the machines not
> > using CSW, it's something like a 20 KB package containing 10 files
> > which has to be specifically called to do anything.
> >
> 
> because all machines look the same on the base layer. and upgrading
> all machines is a major thing. and waiting for the next operating
> system release as well (because it takes so much time).
> 
> it was already difficult to get the /opt/csw mountpoint.
> 
> but i'll try :)
> 

If it's a "these things change only once a year, if you're LUCKY" thing...
then it would almost be worse to be stuck with an older version of the
thing.
your requested change should be more along the lines of "please install a
symlink to point to /etc/csw/[somehackplace]" so that you can deploy
updates to the class utils in a timely fashion.


But i think you'll probably be best off trying my "treat as read-only
/opt/csw nfs share" approach.


From skayser at opencsw.org  Tue Apr  7 00:01:20 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Apr 2009 00:01:20 +0200
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>	<49D4E784.7070605@opencsw.org>	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
Message-ID: <49DA7BB0.6050807@opencsw.org>

Dagobert Michelsen wrote:
> Am 02.04.2009 um 18:34 schrieb Philip Brown:
>> On Thu, Apr 02, 2009 at 11:27:48AM -0500, Mike Watters wrote:
>>
>>> for some reason, /opt/csw/include/ncurses does not exist?
>>> it exists as /opt/csw/include/ncursesw
>> sounds like a takeover maintainer bug
> 
> There is now an updated ncurses in testing which has both
> include/ncurses and include/ncursesw.

Thanks, would you mind pushing that onto the build farm?

Sebastian


From rupert at opencsw.org  Tue Apr  7 00:45:32 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 7 Apr 2009 00:45:32 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <20090406210526.GC51303@bolthole.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
	<6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
	<20090406210526.GC51303@bolthole.com>
Message-ID: <6af4270904061545y730a8a2r30d93cb6569fa9d7@mail.gmail.com>

On Mon, Apr 6, 2009 at 23:05, Philip Brown <phil at bolthole.com> wrote:
> On Mon, Apr 06, 2009 at 10:50:31PM +0200, rupert THURNER wrote:
>> On Mon, Apr 6, 2009 at 22:39, Peter Bonivart <bonivart at opencsw.org> wrote:
>> > On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
>> >> we would need to make a request. as we have 1000 machines, and only a
>> >> few of them run opencsw it would be denied.
>> >
>> > Why would it be denied? It's not like it would hurt the machines not
>> > using CSW, it's something like a 20 KB package containing 10 files
>> > which has to be specifically called to do anything.
>> >
>>
>> because all machines look the same on the base layer. and upgrading
>> all machines is a major thing. and waiting for the next operating
>> system release as well (because it takes so much time).
>>
>> it was already difficult to get the /opt/csw mountpoint.
>>
>> but i'll try :)
>>
>
> If it's a "these things change only once a year, if you're LUCKY" thing...
> then it would almost be worse to be stuck with an older version of the
> thing.
> your requested change should be more along the lines of "please install a
> symlink to point to /etc/csw/[somehackplace]" so that you can deploy
> updates to the class utils in a timely fashion.

will do. but will this be sufficient, as the error was:
/etc/opt/csw/init.d/csw.smf.sample
...
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
Read-only file system

or this was more the kind of follow-up error?

>
> But i think you'll probably be best off trying my "treat as read-only
> /opt/csw nfs share" approach.

here every machine is installed separately, but this is quite
automated. i am not aware that we use nfs. currently we install in a
writable area, and then mount into /opt/csw as stated in your
description. which works very good.


rupert.


From bwalton at opencsw.org  Tue Apr  7 00:55:34 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 06 Apr 2009 18:55:34 -0400
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <49DA7BB0.6050807@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
Message-ID: <1239058478-sup-114@ntdws12.chass.utoronto.ca>

Excerpts from Sebastian Kayser's message of Mon Apr 06 18:01:20 -0400 2009:
> > There is now an updated ncurses in testing which has both
> > include/ncurses and include/ncursesw.
> 
> Thanks, would you mind pushing that onto the build farm?

It likely wouldn't hurt in this case, but the official policy is that
no unreleased packages get installed.  As soon as it's available
through the normal channels, it can be added.

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090406/3d768b0e/attachment.asc>

From phil at bolthole.com  Tue Apr  7 00:55:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 15:55:52 -0700
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061545y730a8a2r30d93cb6569fa9d7@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
	<6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
	<20090406210526.GC51303@bolthole.com>
	<6af4270904061545y730a8a2r30d93cb6569fa9d7@mail.gmail.com>
Message-ID: <20090406225552.GB3314@bolthole.com>

hi Rupert,

I think we are not communicating well, because I'm trying to give you too
much information, which you arent ready for :-)

Simplified approach suggested for you:

since you have /opt/csw mounted from "somewhere else", also mount
    /usr/sadm/install/scripts from 'somewhere else'.

Have that place populated by both the solaris standard, AND csw, scripts
that belong in there.



> >
> > But i think you'll probably be best off trying my "treat as read-only
> > /opt/csw nfs share" approach.
> 
> here every machine is installed separately, but this is quite
> automated. i am not aware that we use nfs. currently we install in a
> writable area, and then mount into /opt/csw as stated in your
> description. which works very good.


From rupert at opencsw.org  Tue Apr  7 01:06:07 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 7 Apr 2009 01:06:07 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <20090406202817.GB51303@bolthole.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<20090406202817.GB51303@bolthole.com>
Message-ID: <6af4270904061606n67fd1a4radfa86efc738069b@mail.gmail.com>

On Mon, Apr 6, 2009 at 22:28, Philip Brown <phil at bolthole.com> wrote:
> On Mon, Apr 06, 2009 at 10:17:45PM +0200, rupert THURNER wrote:
>> On Mon, Apr 6, 2009 at 16:43, Peter Bonivart <bonivart at opencsw.org> wrote:
>> > On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
>> >> what would happen if one lives without classutils?
>> >
>> > Disregarding that you have to force past the dependency requirement it
>> > would be the equivalent of not running pre/post scripts. The packages
>> > that depend on cswclassutils use it for some combo of e.g. init/SMF
>> > support and handling of configuration files, see
>> > http://wiki.opencsw.org/cswclassutils-package for all features.
>> >
>> > What's the problem to install it in the global zone? Even if you don't
>> > have root access to it yourself you should be able to make a request
>> > for a package install, right? What would you do with a sparse zone if
>> > you needed a SUNW package that installed in /usr?
>>
>> we would need to make a request. as we have 1000 machines, and only a
>> few of them run opencsw it would be denied.
>>
>> i am wondering how it was done in the past? because we never had write
>> access and it always seemed to work. also the install description only
>> states /opt/csw as mount point.
>
> "in the past", we didnt use the class script as much.
>
> I would suggest that you make the request. you dont lose much by making the
> request, no?
>
> Afer that, as Peter B. said, you would have to run whatever is needful, by
> hand.
>
> Which is one reason I have been pushing to have the things called by the
> class scripts, in /opt/csw rather than /etc.
> This allows for tools to be written to do what is neccessary.
>
> You are basically in the same circumstances as kind of a
> ?"NFS-shared /opt/csw" model. Perhaps this might motivate you to help
> write some of those tools? :-)
> The good news is, you would only have to do it once for each cswclassutils
> script that we have.

could you point me to an example?

rupert.


From skayser at opencsw.org  Tue Apr  7 01:10:55 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Apr 2009 01:10:55 +0200
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <20090406220708.GD51303@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
	<20090406220708.GD51303@bolthole.com>
Message-ID: <49DA8BFF.9070809@opencsw.org>

Philip Brown wrote:
> On Tue, Apr 07, 2009 at 12:01:20AM +0200, Sebastian Kayser wrote:
>>> There is now an updated ncurses in testing which has both
>>> include/ncurses and include/ncursesw.
>> Thanks, would you mind pushing that onto the build farm?
> 
> that violates the build farm principles.
> only released packages.

Oops, right. No harm intended.

Just installed the package on my local build machine and successfully
tested watch and ncdu builds against regular ncurses. mcabber building
with the ncursesw includes bails out, will have a look at that
tomorrow. Maybe it looks obvious to someone.

source='commands.c' object='commands.o' libtool=no \
        DEPDIR=.deps depmode=none /bin/bash ../depcomp \
        /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..  -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
"/usr/include/iso/wctype_c99.h", line 48: syntax error before or at: wctype
"/usr/include/iso/wctype_c99.h", line 48: warning: undefined or missing type for: wctype
"/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:  empty declaration
"commands.c", line 110: cannot recover from previous errors
cc: acomp failed for commands.c

Sebastian


From phil at bolthole.com  Tue Apr  7 01:24:38 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 16:24:38 -0700
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <49DA8BFF.9070809@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
	<20090406220708.GD51303@bolthole.com>
	<49DA8BFF.9070809@opencsw.org>
Message-ID: <20090406232438.GD3314@bolthole.com>

On Tue, Apr 07, 2009 at 01:10:55AM +0200, Sebastian Kayser wrote:
> 
> Just installed the package on my local build machine and successfully
> tested watch and ncdu builds against regular ncurses. mcabber building
> with the ncursesw includes bails out, will have a look at that
> tomorrow. Maybe it looks obvious to someone.
> 

looks fairly obvious to me. 
Solaris system includes are getting looked at before the stuff in
wncurses dir. This breaks things for ncurses.

So either you fiddle with your -I optoins to the compiler, or you have to
take a look at the source code, to juggle order of whatever is causing
an #include of /usr/include/iso/wctype_c99.h, to come later, instead of
sooner.


(or... simply drop trying to use ncurses, and use the system standard
  curses libs :)



> source='commands.c' object='commands.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..  -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
> "/usr/include/iso/wctype_c99.h", line 48: syntax error before or at: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: undefined or missing type for: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:  empty declaration
> "commands.c", line 110: cannot recover from previous errors
> cc: acomp failed for commands.c
> 


From hson at opencsw.org  Tue Apr  7 08:08:54 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Tue, 07 Apr 2009 08:08:54 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <20090406.14345000.1346203350@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>
	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>
Message-ID: <49DAEDF6.8050404@opencsw.org>

James Lee wrote:
> 
> It expands to the isalist.  It tries all in order but if it's a 32 bit
> prog there is never a valid match in the 64s, then it runs through the
> 32s until the match, or not in the case of this first non CSW lib:
> 

And your solution to that "problem" is?
Add all 32bit paths to RPATH/RUNPATH when building a 32bit app/lib, 
instead of just $ISALIST?

> Once built a package doesn't know how it was built so if it works
> use it.

Parse error...

> LD_OPTIONS overrides all so is powerful.  Like all powerful weapons
> use with caution.
> 

My point was that LD_OPTIONS is used by default when building using gar.
And those broken RPATH/RUNPATHs was due to a bug in gar, not what the 
maintainer might have set (or not set).

>> To reiterate my question:
>> Is there any actual harm (disregarding the fact that the link loader
>> might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without
>> finding any match) having RPATH/RUNPATH set?
> 
> No, well it's actual but trivial.
> 
> 
>> If not, it is only when a package doesn't use any libs at all from
>> /opt/csw/lib, when RPATH/RUNPATH should be changed from the default.
> 
> Correct, as is the case with zlib.

I'm not thinking of just zlib, I'm speaking in a more general term plus 
"should" should have been "could".

Since we have established that its not harmful to have RPATH set even 
when its not used, I can see more problem (as in more work for the 
maintainer in the future) with actively unsetting LD_OPTIONS than not.


>> If package A is linked to a lib i package B (both CSW packages) and A i
>> packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future
>> releases of B which contain optimized libraries won't be used until A is
>> repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/csw/lib
> 
> Correct.  Add $ISALIST in, unless you know it won't be needed.

And my point is that you can't know if its going to be needed or not at 
the time of package (unless you are both maintaner of A and B and know 
that you won't package optimized version of B).

>> And if there is a longer dependency list (A->B->C-D where D is the one
>> with optimized libs) it creates much more work for the maintainer of A
>> (he must check every lib in the dependency path to see if there are any
>> optimized libs).
> 
> D is governed by the RPATH of C.  C by B.

You're correct (at least if C and B are linked with required libraries, 
reference to my post the other week regarding gpdf and libnet), my 
memory was wrong.





From dam at opencsw.org  Tue Apr  7 08:56:05 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 08:56:05 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <49DAEDF6.8050404@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>
	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>
	<49DAEDF6.8050404@opencsw.org>
Message-ID: <C860D70E-A2E4-40AC-BD8F-3428F43E83F9@opencsw.org>

Hi Roger,

Am 07.04.2009 um 08:08 schrieb Roger H?kansson:
> And those broken RPATH/RUNPATHs was due to a bug in gar, not what  
> the maintainer might have set (or not set).

To be precise here: The method $ISALIST was passed was due to
multiple expansion, where the number of expansions depended on
the package and it was the responsibility of the maintainer to
ensure proper quoting. It was my fault to not have made this
clear and give advice on how to properly set it.

> Since we have established that its not harmful to have RPATH set  
> even when its not used, I can see more problem (as in more work for  
> the maintainer in the future) with actively unsetting LD_OPTIONS  
> than not.

It is possible to include other ways of passing -R to the linker
that don't interfere with what the maintainer intended. We can
work on these cases when they occur.

>>> If package A is linked to a lib i package B (both CSW packages)  
>>> and A i
>>> packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future
>>> releases of B which contain optimized libraries won't be used  
>>> until A is
>>> repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/ 
>>> csw/lib
>> Correct.  Add $ISALIST in, unless you know it won't be needed.
>
> And my point is that you can't know if its going to be needed or not  
> at the time of package (unless you are both maintaner of A and B and  
> know that you won't package optimized version of B).

That's why GAR adds $ISALIST per default, both in the old version
and now. If you don't want it you have to turn it off, the unaware gets
optimizations.


Best regards

   -- Dago

From bonivart at opencsw.org  Tue Apr  7 16:11:09 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Tue, 7 Apr 2009 16:11:09 +0200
Subject: [csw-maintainers] /testing perl-ldap 0.39
Message-ID: <625385e30904070711l4a3a4f14uf05bcc8d508b1ee8@mail.gmail.com>

I got a request to update perl-ldap but I'm not a user myself (current
maintainer is Alex Moore who is retired) so I could need some help
with testing.

http://mirror.opencsw.org/testing/pm_ldap-0.39,REV=2009.04.07-SunOS5.8-all-CSW.pkg.gz

-- 
/peter


From dam at opencsw.org  Tue Apr  7 17:23:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 17:23:07 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
Message-ID: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>

Hi,

in the past were requests for installation of experimental packages not
released yet to see how things go. However, the strict "released only"
policy on build8s prohibited this installation. To allow these
"experimental packages" I set up a new zone build8st (t=testing)
where packages from testing may be installed. Building packages
for release is not allowed on this machine! I am thinking of
entering both testing/ and current/ into the repository from pkgutil
and allow csw-group execution of pkgutil. Thought?


Best regards

   -- Dago


From mwatters at opencsw.org  Tue Apr  7 18:26:12 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 11:26:12 -0500
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
Message-ID: <49DB7EA4.2060302@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> in the past were requests for installation of experimental packages not
> released yet to see how things go. However, the strict "released only"
> policy on build8s prohibited this installation. To allow these
> "experimental packages" I set up a new zone build8st (t=testing)
> where packages from testing may be installed. Building packages
> for release is not allowed on this machine! I am thinking of
> entering both testing/ and current/ into the repository from pkgutil
> and allow csw-group execution of pkgutil. Thought?
> 
> 
> Best regards
> 
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Sounds good, one caveat

if all maintainers have access to add / remove packages
we will need a way to coordinate install/remove of packages.



- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbfqMACgkQLrhmsXMSLxd8MQCfXiysqx4Tc6vBBU2xhoUdys0A
AGAAnjGyHTWFhRTQRIUaArEhNRlHYYQC
=w6mQ
-----END PGP SIGNATURE-----


From trygvis at opencsw.org  Tue Apr  7 18:37:56 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 07 Apr 2009 18:37:56 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <49DB7EA4.2060302@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
	<49DB7EA4.2060302@opencsw.org>
Message-ID: <49DB8164.6030605@opencsw.org>

Mike Watters wrote:
> Dagobert Michelsen wrote:
>> Hi,
> 
>> in the past were requests for installation of experimental packages not
>> released yet to see how things go. However, the strict "released only"
>> policy on build8s prohibited this installation. To allow these
>> "experimental packages" I set up a new zone build8st (t=testing)
>> where packages from testing may be installed. Building packages
>> for release is not allowed on this machine! I am thinking of
>> entering both testing/ and current/ into the repository from pkgutil
>> and allow csw-group execution of pkgutil. Thought?

Yay!

> Sounds good, one caveat
> 
> if all maintainers have access to add / remove packages
> we will need a way to coordinate install/remove of packages.

Does it require anything other that just doing it? If it is in testing/,
it should be installed? If it is removed, remove it etc.

--
Trygve


From mwatters at opencsw.org  Tue Apr  7 19:49:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 12:49:52 -0500
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <49DB8164.6030605@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>	<49DB7EA4.2060302@opencsw.org>
	<49DB8164.6030605@opencsw.org>
Message-ID: <49DB9240.30304@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Trygve Laugst?l wrote:
> Mike Watters wrote:
>> Dagobert Michelsen wrote:
>>> Hi,
>>> in the past were requests for installation of experimental packages not
>>> released yet to see how things go. However, the strict "released only"
>>> policy on build8s prohibited this installation. To allow these
>>> "experimental packages" I set up a new zone build8st (t=testing)
>>> where packages from testing may be installed. Building packages
>>> for release is not allowed on this machine! I am thinking of
>>> entering both testing/ and current/ into the repository from pkgutil
>>> and allow csw-group execution of pkgutil. Thought?
> 
> Yay!
> 
>> Sounds good, one caveat
>>
>> if all maintainers have access to add / remove packages
>> we will need a way to coordinate install/remove of packages.
> 
> Does it require anything other that just doing it? If it is in testing/,
> it should be installed? If it is removed, remove it etc.
> 
> --
> Trygve
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

My point is I would like to make sure no one removes any packages another
maintainer is currently testing with.

example, I am currently working on gcc4 packages,
if someone is doing a test compile against a set of libraries using gcc4
I don't think they would appreciate if I uninstalled gcc out from under them.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbkkAACgkQLrhmsXMSLxen7ACfTTXiWmVXwAZbdV8PKuN2Ulue
NNcAn1WEPSdNilFfyJ0zFSyJbRurDJ3S
=QfDd
-----END PGP SIGNATURE-----


From dam at opencsw.org  Tue Apr  7 21:25:00 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 21:25:00 +0200
Subject: [csw-maintainers] Updating all buildfarm servers to current/
Message-ID: <92AAA77E-031F-427A-BB86-1B6814FF1C84@opencsw.org>

Hi,

I am updating all buildfarm servers to current now.


Best regards

   -- Dago


From dam at opencsw.org  Tue Apr  7 21:45:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 21:45:42 +0200
Subject: [csw-maintainers] Updated ncurses released
In-Reply-To: <49DA8BFF.9070809@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
	<20090406220708.GD51303@bolthole.com>
	<49DA8BFF.9070809@opencsw.org>
Message-ID: <BF9FD469-989D-4992-827B-AD445D492AB4@opencsw.org>

Hi,

Am 07.04.2009 um 01:10 schrieb Sebastian Kayser:
> Philip Brown wrote:
>> On Tue, Apr 07, 2009 at 12:01:20AM +0200, Sebastian Kayser wrote:
>>>> There is now an updated ncurses in testing which has both
>>>> include/ncurses and include/ncursesw.
>>> Thanks, would you mind pushing that onto the build farm?
>>
>> that violates the build farm principles.
>> only released packages.
>
> Oops, right. No harm intended.
>
> Just installed the package on my local build machine and successfully
> tested watch and ncdu builds against regular ncurses. mcabber building
> with the ncursesw includes bails out, will have a look at that
> tomorrow. Maybe it looks obvious to someone.
>
> source='commands.c' object='commands.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..   
> -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/ 
> opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/ 
> csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ 
> ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
> "/usr/include/iso/wctype_c99.h", line 48: syntax error before or  
> at: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: undefined or  
> missing type for: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:   
> empty declaration
> "commands.c", line 110: cannot recover from previous errors
> cc: acomp failed for commands.c

The CSWncurses package is fixed, released and installed on all buildfarm
servers.


Best regards

   -- Dago


From skayser at opencsw.org  Tue Apr  7 22:40:55 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Apr 2009 22:40:55 +0200
Subject: [csw-maintainers] Thanks! (was Re: [csw-users] python 2.6.1 now in
	testing)
In-Reply-To: <49D3E0F5.2060809@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
Message-ID: <49DBBA57.3050001@opencsw.org>

Mike Watters wrote:
> Changes:
> 
> 	bug fix: 3510 'import curses' fails in current Python...
> 			forced curses to use the library in /opt/csw
> 	bug fix: 3579 RPATH stuff
> 			added NOISALIST = 1

Btw. Mike, i just happened to be at one of my distant colleagues over at
the development unit today and he was struggling to get Mercurial
running with the sunfreeware.com stack (bailed out with some python _md5
bla issues on commits ... he just wanted to use it, not troubleshoot it).

He was already pondering about alternatives and i told him to give
opencsw.org a try. He was happy to give it a try and even more happy to
see that everything worked out of the box :) He remarked that the
Mercurial over at sunfreeware.com is more recent but for his needs ours
is sufficient ... needless to say that any Mercurial is more useful than
one that is more recent but somehow broken ;)

Thanks to you and Trygvis (who takes care of hg)!

Sebastian


From yann at pleiades.fr.eu.org  Tue Apr  7 23:13:34 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Tue, 07 Apr 2009 23:13:34 +0200
Subject: [csw-maintainers] bash_completion and modification of the version
	numbering scheme
Message-ID: <49DBC1FE.9080803@pleiades.fr.eu.org>

Hi everybody,

You will find in testing a new release of the bash_completion package:
http://buildfarm.opencsw.org/testing/bash_completion-1.0,REV=2009.04.07-SunOS5.8-all-CSW.pkg.gz

In addition to upstream completion, this package provides some 
completion for pkgutil, pkg-get, pkgrm and smf tools. Testing and 
feedback are welcome.

I have however a problem with this package, upstream changed its version 
numbering scheme from $YEAR$MONTH$DAY to standard X.X.

 From what I remember, pkgutil uses the REV field if available so it 
would not be a problem with it but pkg-get doesn't not seem to rely on 
it so will never upgrade the package.
Am I right here and is there a way to correctly handle this problem ?

Thanks in advance.

Yann


From mwatters at opencsw.org  Tue Apr  7 23:27:30 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 16:27:30 -0500
Subject: [csw-maintainers] bash_completion and modification of the
 version numbering scheme
In-Reply-To: <49DBC1FE.9080803@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org>
Message-ID: <49DBC542.30304@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yann Rouillard wrote:
> Hi everybody,
> 
> You will find in testing a new release of the bash_completion package:
> http://buildfarm.opencsw.org/testing/bash_completion-1.0,REV=2009.04.07-SunOS5.8-all-CSW.pkg.gz
> 
> 
> In addition to upstream completion, this package provides some
> completion for pkgutil, pkg-get, pkgrm and smf tools. Testing and
> feedback are welcome.
> 
> I have however a problem with this package, upstream changed its version
> numbering scheme from $YEAR$MONTH$DAY to standard X.X.
> 
> From what I remember, pkgutil uses the REV field if available so it
> would not be a problem with it but pkg-get doesn't not seem to rely on
> it so will never upgrade the package.
> Am I right here and is there a way to correctly handle this problem ?
> 
> Thanks in advance.
> 
> Yann
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Yann,
	I am by no means an expert with Gar.
but if you change or add
# We define upstream file regex so we can be notified of new
# upstream software release

UFILES_REGEX = $(GARNAME)-(\d+(?:\.\d+)*).tar.bz2

to match the "NEW" schema it will pick up the new version as released upstream
HTH


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbxUEACgkQLrhmsXMSLxcFIgCdFIoR5h6NZGJ2r1jPd8pK5/Yd
JJYAmwdXWiACCg83BuO+o/YVbPwCGKzc
=8p1b
-----END PGP SIGNATURE-----


From yann at pleiades.fr.eu.org  Tue Apr  7 23:33:17 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Tue, 07 Apr 2009 23:33:17 +0200
Subject: [csw-maintainers] bash_completion and modification of the
 version numbering scheme
In-Reply-To: <49DBC542.30304@opencsw.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
Message-ID: <49DBC69D.1090407@pleiades.fr.eu.org>


Le 07.04.2009 23:27, Mike Watters a ?crit :
> Yann,
> 	I am by no means an expert with Gar.
> but if you change or add
> # We define upstream file regex so we can be notified of new
> # upstream software release
>
> UFILES_REGEX = $(GARNAME)-(\d+(?:\.\d+)*).tar.bz2
>
> to match the "NEW" schema it will pick up the new version as released upstream
> HTH
>

Hi Mike,

My problem is not with the check upstream code from gar but with pkg-get 
which will not accept to upgrade to the new package at it will always 
think it is older that the currently installed one

Yann




From mwatters at opencsw.org  Wed Apr  8 00:00:40 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 17:00:40 -0500
Subject: [csw-maintainers] bash_completion and modification of the
 version numbering scheme
In-Reply-To: <49DBC69D.1090407@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org>
Message-ID: <49DBCD08.2040805@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yann Rouillard wrote:
> 
> Le 07.04.2009 23:27, Mike Watters a ?crit :
>> Yann,
>>     I am by no means an expert with Gar.
>> but if you change or add
>> # We define upstream file regex so we can be notified of new
>> # upstream software release
>>
>> UFILES_REGEX = $(GARNAME)-(\d+(?:\.\d+)*).tar.bz2
>>
>> to match the "NEW" schema it will pick up the new version as released
>> upstream
>> HTH
>>
> 
> Hi Mike,
> 
> My problem is not with the check upstream code from gar but with pkg-get
> which will not accept to upgrade to the new package at it will always
> think it is older that the currently installed one
> 
> Yann
> 
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

sorry, I miss read the post. You can override the variable(s)
in the package set GARVERSION as x.x and have
SPKG_VERSION be YYYYMMDD (as package build date)


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbzQcACgkQLrhmsXMSLxeKJgCfThwyK9HRYpK9Dh0xhbbzne1J
j/UAn0Qo7lPqpiRIgv22lx47SbbNbWoG
=nC2U
-----END PGP SIGNATURE-----


From bonivart at opencsw.org  Wed Apr  8 00:11:40 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 00:11:40 +0200
Subject: [csw-maintainers] bash_completion and modification of the
	version numbering scheme
In-Reply-To: <49DBCD08.2040805@opencsw.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org> <49DBCD08.2040805@opencsw.org>
Message-ID: <625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>

On Wed, Apr 8, 2009 at 12:00 AM, Mike Watters <mwatters at opencsw.org> wrote:
> sorry, I miss read the post. You can override the variable(s)
> in the package set GARVERSION as x.x and have
> SPKG_VERSION be YYYYMMDD (as package build date)

But why would we fake the upstream version and for how long should we
keep doing that? Some packages, like my geodb and possibly pca,
doesn't have versions at all so we assign something like YYMMDD to it
but in this case the end users would be confused, right?

This would also be really ugly:

20090408,REV=2009.04.08_rev=1.0

-- 
/peter


From bonivart at opencsw.org  Wed Apr  8 00:18:21 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 00:18:21 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
Message-ID: <625385e30904071518l767e9d5gf281d8a6dc9ce79b@mail.gmail.com>

On Tue, Apr 7, 2009 at 5:23 PM, Dagobert Michelsen <dam at opencsw.org> wrote:
> Hi,
>
> in the past were requests for installation of experimental packages not
> released yet to see how things go. However, the strict "released only"
> policy on build8s prohibited this installation. To allow these
> "experimental packages" I set up a new zone build8st (t=testing)
> where packages from testing may be installed. Building packages
> for release is not allowed on this machine! I am thinking of
> entering both testing/ and current/ into the repository from pkgutil
> and allow csw-group execution of pkgutil. Thought?

Is the zone for building or just for installing packages for those who
don't have Solaris 8 machines to test on?

-- 
/peter


From bonivart at opencsw.org  Wed Apr  8 00:23:06 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 00:23:06 +0200
Subject: [csw-maintainers] March 31st - last day of Solaris 8 support from
	Sun
Message-ID: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com>

The day went by quietly here but earlier we have agreed that after
this date we should no longer be required to build for Solaris 8. I
know that Dago has prepared Solaris 9 build machines but is it
officially ok to start using them?

-- 
/peter


From skayser at opencsw.org  Wed Apr  8 02:48:54 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 08 Apr 2009 02:48:54 +0200
Subject: [csw-maintainers] iswblank() and wctype.h woes (was Re: Updated
 ncurses available in testing/)
In-Reply-To: <20090406232438.GD3314@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>	<49D4E784.7070605@opencsw.org>	<20090402163457.GI19124@bolthole.com>	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>	<49DA7BB0.6050807@opencsw.org>	<20090406220708.GD51303@bolthole.com>	<49DA8BFF.9070809@opencsw.org>
	<20090406232438.GD3314@bolthole.com>
Message-ID: <49DBF476.2020409@opencsw.org>

Philip Brown wrote:
> On Tue, Apr 07, 2009 at 01:10:55AM +0200, Sebastian Kayser wrote:
>> Just installed the package on my local build machine and successfully
>> tested watch and ncdu builds against regular ncurses. mcabber building
>> with the ncursesw includes bails out, will have a look at that
>> tomorrow. Maybe it looks obvious to someone.
>>
> 
> looks fairly obvious to me. 
> Solaris system includes are getting looked at before the stuff in
> wncurses dir. This breaks things for ncurses.
> 
> So either you fiddle with your -I optoins to the compiler, or you have to
> take a look at the source code, to juggle order of whatever is causing
> an #include of /usr/include/iso/wctype_c99.h, to come later, instead of
> sooner.

Ok, fixed, multiple reasons. First of all i had been (mis)using a
Solaris 10 machine, so the system headers were different from the build
farm ones. On the build farm the build is fine.

Second - and that was the main "culprit" then - i had worked around the
missing iswblank() on Solaris 8 via the following snippet

# ifndef HAVE_ISWBLANK
# define iswblank(wc) iswctype(wc, wctype("blank"))
# endif

HAVE_ISWBLANK isn't defined by the mcabber autoconf system so it would
just always be included. No problem on Solaris 8 where iswblank() is
missing (and hence not used by any system headers), but on Solaris 10 it
is available and unfortunately i had put the snippet in a place where
the #define replaced system header occurences of iswblank() where
wctype() wasn't yet declared. Thus the observed error messages.

> (or... simply drop trying to use ncurses, and use the system standard
>   curses libs :)

No ncurses issue here ;)

>> source='commands.c' object='commands.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..  -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
>> "/usr/include/iso/wctype_c99.h", line 48: syntax error before or at: wctype
>> "/usr/include/iso/wctype_c99.h", line 48: warning: undefined or missing type for: wctype
>> "/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:  empty declaration
>> "commands.c", line 110: cannot recover from previous errors
>> cc: acomp failed for commands.c

I have now moved the snippet after the includes to wctype.h (which fixes
the build even on the Solaris 10 box) and will check with the mcabber
author whether he can include an autoconf check for iswblank().

Sebastian


From hson at opencsw.org  Wed Apr  8 04:00:17 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 08 Apr 2009 04:00:17 +0200
Subject: [csw-maintainers] Building multiple packages from one source package
Message-ID: <49DC0531.6060006@opencsw.org>

I'm trying to build two packages from one source package and I've looked 
at some other packages (gettext for example), but I can't get it to work 
properly.

The first package CSWhtmldoc-common is build without any problem, but 
CSWhtmldoc fails since I'm trying to have CSWhtmldoc-common as a 
dependency of CSWhtmldoc.

Examining 'depend' file
P CSWfltk fltk - Fast light Tool Kit
P CSWjpeg jpeg - JPEG library and tools by the Independent JPEG Group
P CSWosslrt openssl_rt - Openssl runtime libraries
P CSWpng png - library for Portable Network Graphics format (PNG)
P CSWzlib zlib - Zlib Data Compression Library
P CSWhtmldoc-common htmldoc_common - This package contains the htmldoc 
files common to all architectures.
P CSWcommon common - common files and dirs for CSW packages

system      CSWcommon      common - common files and dirs for CSW packages
application CSWfltk        fltk - Fast light Tool Kit
ERROR: information for "CSWhtmldoc-common" was not found
ERROR: Invalid package CSWhtmldoc-common specified. Check of 
/tmp/htmldoc-1.8.27,REV=2009.04.08-SunOS5.8-sparc-CSW.pkg.gz fails


How do I make this work?


From bonivart at opencsw.org  Wed Apr  8 09:36:29 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 09:36:29 +0200
Subject: [csw-maintainers] Building multiple packages from one source
	package
In-Reply-To: <49DC0531.6060006@opencsw.org>
References: <49DC0531.6060006@opencsw.org>
Message-ID: <625385e30904080036g343cf0e6n421354471b807ffe@mail.gmail.com>

On Wed, Apr 8, 2009 at 4:00 AM, Roger H?kansson <hson at opencsw.org> wrote:
> How do I make this work?

As far as I know you can't pass the checks unless the package you're
depending on is installed on the build server. With our policy to not
install unreleased packages it will not work for split packages. I
disable the checks with:

ENABLE_CHECK=0

And test on another system instead. The cswutils package contains the
checkpkg script. I guess we can use the new build8st for this.

-- 
/peter


From james at opencsw.org  Wed Apr  8 12:05:43 2009
From: james at opencsw.org (James Lee)
Date: Wed, 08 Apr 2009 10:05:43 GMT
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <49DAEDF6.8050404@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>
	<49D9F70C.9030007@opencsw.org>
	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>
	<49DAEDF6.8050404@opencsw.org>
Message-ID: <20090408.10054300.1414585890@gyor.oxdrove.co.uk>

On 07/04/09, 07:08:54, "Roger H?kansson" <hson at opencsw.org> wrote regarding
Re: [csw-maintainers] RPATH/ISALIST (zlib package in /testing):

> > It expands to the isalist.  It tries all in order but if it's a 32 bit
> > prog there is never a valid match in the 64s, then it runs through the
> > 32s until the match, or not in the case of this first non CSW lib:
> >

> And your solution to that "problem" is?

Sun could make 32 progs use a 32 bit isalist, but this isn't "my"
problem, it's Sun's.  All I'm saying is that a good $ISALIST RPATH
causes more failed file look-ups than a bad one.


> Add all 32bit paths to RPATH/RUNPATH when building a 32bit app/lib,
> instead of just $ISALIST?

isalist expansion happens locally, that why/how it works.



> > Once built a package doesn't know how it was built so if it works
> > use it.

> Parse error...

Once a package is built the package doesn't know how the package was
built so if LD_OPTIONS works to make a package that works use
LD_OPTIONS to build the package.




> My point was that LD_OPTIONS is used by default when building using gar.
> And those broken RPATH/RUNPATHs was due to a bug in gar, not what the
> maintainer might have set (or not set).

The package doesn't know how the package was built - GAR is irrelevant
to the result.  The maintainer is responsible for what is in the package.



> >> To reiterate my question:
> >> Is there any actual harm (disregarding the fact that the link loader
> >> might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without
> >> finding any match) having RPATH/RUNPATH set?
> >
> > No, well it's actual but trivial.
> >
> >
> >> If not, it is only when a package doesn't use any libs at all from
> >> /opt/csw/lib, when RPATH/RUNPATH should be changed from the default.
> >
> > Correct, as is the case with zlib.

> I'm not thinking of just zlib, I'm speaking in a more general term plus
> "should" should have been "could".

"should" could have been "could".
"should" is right, but the RPATH needn't be right because the error is
trivial.


switch (person.getType())
{
    case PEDANT:
        message = "must";
        break;
    case MATHEMATICIAN:
        message = "should";
        break;
    case ENGINEER:
        message = "could";
        break;
    case ACCOUNTANT:
        message = "if it makes money";
        break;
    case BANKER:
        message = "if it makes me money";
        break;
    case TEENAGER:
        message = "yeah wotevva";
        break;



> Since we have established that its not harmful to have RPATH set even
> when its not used,

It's harmful but trivial.


> I can see more problem (as in more work for the
> maintainer in the future) with actively unsetting LD_OPTIONS than not.

Sorry I can't see why unsetting something has only been set by yourself
in the first place is more work.  In fact we should always explicitly
either set or unset such controlling environmental variables as a
matter of good practice.




James.


From hson at opencsw.org  Wed Apr  8 13:11:20 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 08 Apr 2009 13:11:20 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <20090408.10054300.1414585890@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>	<49DAEDF6.8050404@opencsw.org>
	<20090408.10054300.1414585890@gyor.oxdrove.co.uk>
Message-ID: <49DC8658.8020908@opencsw.org>

James Lee wrote:
>>I can see more problem (as in more work for the
>> maintainer in the future) with actively unsetting LD_OPTIONS than not.
> 
> Sorry I can't see why unsetting something has only been set by yourself
> in the first place is more work.  In fact we should always explicitly
> either set or unset such controlling environmental variables as a
> matter of good practice.
> 

When building packages with gar, the default (i.e without the maintainer 
doing a thing) is that the RPATH will contain 
/opt/csw/lib/$ISALIST:/opt/csw/lib. If a maintainer is to package 
something with a different RPATH he must set it manually, i.e more work 
than default.


From yann at pleiades.fr.eu.org  Wed Apr  8 14:30:48 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Wed, 08 Apr 2009 14:30:48 +0200
Subject: [csw-maintainers] bash_completion and modification of
 the	version numbering scheme
In-Reply-To: <625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
References: <49DBC1FE.9080803@pleiades.fr.eu.org>
	<49DBC542.30304@opencsw.org>	<49DBC69D.1090407@pleiades.fr.eu.org>
	<49DBCD08.2040805@opencsw.org>
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
Message-ID: <49DC98F8.4060307@pleiades.fr.eu.org>


Le 08.04.2009 00:11, Peter Bonivart a ?crit :
> On Wed, Apr 8, 2009 at 12:00 AM, Mike Watters<mwatters at opencsw.org>  wrote:
>> sorry, I miss read the post. You can override the variable(s)
>> in the package set GARVERSION as x.x and have
>> SPKG_VERSION be YYYYMMDD (as package build date)
>
> But why would we fake the upstream version and for how long should we
> keep doing that? Some packages, like my geodb and possibly pca,
> doesn't have versions at all so we assign something like YYMMDD to it
> but in this case the end users would be confused, right?
>
> This would also be really ugly:
>
> 20090408,REV=2009.04.08_rev=1.0
>

I would also prefer to follow the upstream version numbering for the 
package, that seems closer to the standard 
http://www.opencsw.org/standards/build .

Why pkgutil and pkg-get don't have the same way to compare package 
versions ? It wasn't supposed to be standard ?

Yann


From bonivart at opencsw.org  Wed Apr  8 14:52:40 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 14:52:40 +0200
Subject: [csw-maintainers] bash_completion and modification of the
	version numbering scheme
In-Reply-To: <49DC98F8.4060307@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org> <49DBCD08.2040805@opencsw.org> 
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com> 
	<49DC98F8.4060307@pleiades.fr.eu.org>
Message-ID: <625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>

On Wed, Apr 8, 2009 at 2:30 PM, Yann Rouillard <yann at pleiades.fr.eu.org> wrote:
> Why pkgutil and pkg-get don't have the same way to compare package versions
> ? It wasn't supposed to be standard ?

I don't know how pkg-get does it, but for pkgutil we had a public
discussion on this list where James came with a suggestion and Dago
made a few changes if I remember correctly. I implemented it and I
have described it here in four rules:
http://pkgutil.wikidot.com/get-install-and-configure#toc7 (version
compare method).

The standard also hints about the REV-field becoming the standard way
of comparison for pkg-get, I assume it was written by Phil.

http://www.opencsw.org/standards/build#versioning

"Please note: the ",REV=YYYY.MM.DD" is now Mandatory. It provides a
fixed-format way of telling how recent the package really is, for
version comparison download purposes. At some point, it will be the
primary comparison key for pkg-get.(but not yet)"

-- 
/peter


From yann at pleiades.fr.eu.org  Wed Apr  8 17:53:38 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Wed, 08 Apr 2009 17:53:38 +0200
Subject: [csw-maintainers] bash_completion and modification of
 the	version numbering scheme
In-Reply-To: <625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>
References: <49DBC1FE.9080803@pleiades.fr.eu.org>
	<49DBC542.30304@opencsw.org>	<49DBC69D.1090407@pleiades.fr.eu.org>
	<49DBCD08.2040805@opencsw.org>
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
	<49DC98F8.4060307@pleiades.fr.eu.org>
	<625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>
Message-ID: <49DCC882.6000708@pleiades.fr.eu.org>

Le 08.04.2009 14:52, Peter Bonivart a ?crit :
> The standard also hints about the REV-field becoming the standard way
> of comparison for pkg-get, I assume it was written by Phil.
>
> http://www.opencsw.org/standards/build#versioning
>
> "Please note: the ",REV=YYYY.MM.DD" is now Mandatory. It provides a
> fixed-format way of telling how recent the package really is, for
> version comparison download purposes. At some point, it will be the
> primary comparison key for pkg-get.(but not yet)"
>

You're right, I missed that point.
Phil, do you intend to change the comparison code of pkg-get soon or do 
you think I should rather adopt the 
YEARMONTHDAY,REV=YEARMONTHDAY_rev=X.X package version numbering to be 
able to release bash_completion shortly ?

Yann


From Darin.Perusich at cognigencorp.com  Wed Apr  8 18:04:14 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 08 Apr 2009 12:04:14 -0400
Subject: [csw-maintainers] build8s and build9s inconsistancies
Message-ID: <49DCCAFE.2040900@cognigencorp.com>

I was building a package, amanda, on build8s and everything was going
fine but given that Solaris 8 is no longer supported I figured I'd move
over to the Solaris 9 build machines. The code compiles fine but the
staging of the package is incomplete on build9s, whether I use
'stagepkg' or 'gmake DESTDIR=$PWD/cswstage install' the results are the
same.

The staging fails when setting permission, see below, on some of the
binaries and I have no idea why. I compared the CSW packages and they
are identical on both machine so I'm wondering if its something in the OS.

Has anyone else run across this before?

gmake[5]: Entering directory `/home/darin/build/amanda-2.6.1/common-src'
  (dest=/opt/csw/sbin)
  (chown=amanda)
chown darin:sys
/home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt
chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt:
Not owner
gmake[5]: *** [installperms-data] Error 1
gmake[5]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[4]: *** [install-data-am] Error 2
gmake[4]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[3]: *** [install-am] Error 2
gmake[3]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[2]: *** [install] Error 2
gmake[2]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[1]: *** [install-recursive] Error 1
gmake[1]: Leaving directory `/home/darin/build/amanda-2.6.1'
gmake: *** [install] Error 2


-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Wed Apr  8 18:30:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 09:30:25 -0700
Subject: [csw-maintainers] bash_completion and modification of
	the	version numbering scheme
In-Reply-To: <49DCC882.6000708@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org>
	<49DBCD08.2040805@opencsw.org>
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
	<49DC98F8.4060307@pleiades.fr.eu.org>
	<625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>
	<49DCC882.6000708@pleiades.fr.eu.org>
Message-ID: <20090408163025.GS99983@bolthole.com>

On Wed, Apr 08, 2009 at 05:53:38PM +0200, Yann Rouillard wrote:
> You're right, I missed that point.
> Phil, do you intend to change the comparison code of pkg-get soon or do  
> you think I should rather adopt the  
> YEARMONTHDAY,REV=YEARMONTHDAY_rev=X.X package version numbering to be  
> able to release bash_completion shortly ?

I do intend to change it.... however, just how soon it happens, is somewhat
up in the air :-/

I may have time to add it next week.



From mwatters at opencsw.org  Wed Apr  8 19:44:17 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 12:44:17 -0500
Subject: [csw-maintainers] build8s and build9s inconsistancies
In-Reply-To: <49DCCAFE.2040900@cognigencorp.com>
References: <49DCCAFE.2040900@cognigencorp.com>
Message-ID: <49DCE271.9050304@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Darin Perusich wrote:
> I was building a package, amanda, on build8s and everything was going
> fine but given that Solaris 8 is no longer supported I figured I'd move
> over to the Solaris 9 build machines. The code compiles fine but the
> staging of the package is incomplete on build9s, whether I use
> 'stagepkg' or 'gmake DESTDIR=$PWD/cswstage install' the results are the
> same.
> 
> The staging fails when setting permission, see below, on some of the
> binaries and I have no idea why. I compared the CSW packages and they
> are identical on both machine so I'm wondering if its something in the OS.
> 
> Has anyone else run across this before?
> 
> gmake[5]: Entering directory `/home/darin/build/amanda-2.6.1/common-src'
>   (dest=/opt/csw/sbin)
>   (chown=amanda)
> chown darin:sys
> /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt:
> Not owner
> gmake[5]: *** [installperms-data] Error 1
> gmake[5]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[4]: *** [install-data-am] Error 2
> gmake[4]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[3]: *** [install-am] Error 2
> gmake[3]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[2]: *** [install] Error 2
> gmake[2]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[1]: *** [install-recursive] Error 1
> gmake[1]: Leaving directory `/home/darin/build/amanda-2.6.1'
> gmake: *** [install] Error 2
> 
> 

I can't even get onto build9s or build9x, I get prompted for password.
since home dir's are mounted across the env, I have to assume I don't have an
account on those servers.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc4nEACgkQLrhmsXMSLxfs0QCeISmYi9rNC40Z27vzmGkomKfL
SMQAoKBEs6B7kzxTHgjC19izgfiCs9yk
=lBSx
-----END PGP SIGNATURE-----


From pfelecan at opencsw.org  Wed Apr  8 19:47:32 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 08 Apr 2009 19:47:32 +0200
Subject: [csw-maintainers] static libraries lib*.a in our packages
Message-ID: <x6vdpfkpfv.fsf@abhavana.free.fr>

I cannot recollect precisely if we already had this discussion. Phil and
me, we had an exchange on this subject and I would like to have the
opinion of the other fellow maintainers as our point of view are not
totally convergent.

The questions are:

1. Is a good practice for CSW packages to contain library archives of
   the form lib*.a when we deliver dynamic libraries lib*.so ?

2. What's the potential usage for such libraries?

3. Which mechanism do you use for not building (e.g. ./configure
   --disable-static) or to post-install (in the sense of after make install)
   remove them.

Thank you in advance for your answers and comments.
-- 
Peter


From phil at bolthole.com  Wed Apr  8 20:01:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 11:01:50 -0700
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <x6vdpfkpfv.fsf@abhavana.free.fr>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
Message-ID: <20090408180150.GZ99983@bolthole.com>

On Wed, Apr 08, 2009 at 07:47:32PM +0200, Peter FELECAN wrote:
> I cannot recollect precisely if we already had this discussion. Phil and
> me, we had an exchange on this subject and I would like to have the
> opinion of the other fellow maintainers as our point of view are not
> totally convergent.
> 
> The questions are:
> 
> 1. Is a good practice for CSW packages to contain library archives of
>    the form lib*.a when we deliver dynamic libraries lib*.so ?
> 
> 2. What's the potential usage for such libraries?


To save some back-and-forth on the mailing list, here's a bit of history
for folks who were not around for the original discussions 4(?) years ago
:)

The perspective was that people very rarely ever use static libraries.
nowadays, pretty much every library is a dynamic library.
This helps both for memory sharing, and for ease of updates.
(you only have to update the lib package, not have to recompile
everything).

Occasionally, there may be benefits to making a static library for some
software available for optional use. However, when and if the maintainer
decides there is a benefit to providing a static lib, it is best to do so
in a "devel" subpackage (unless the library is relatively small).

that way, when a library package gets pulled in, solely as a result of
some other package having a dependancy on "libfoo.so", the user does not
needlessly download "libfoo.a" as well.

This is particularly important in the case of some packages, such as
opensp, where the static library size is *4 megabytes*.


so.... that's the background. 
Other folks, please add your thoughts to this :-)


From mwatters at opencsw.org  Wed Apr  8 20:07:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 13:07:52 -0500
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <x6vdpfkpfv.fsf@abhavana.free.fr>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
Message-ID: <49DCE7F8.1040005@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Peter FELECAN wrote:
> I cannot recollect precisely if we already had this discussion. Phil and
> me, we had an exchange on this subject and I would like to have the
> opinion of the other fellow maintainers as our point of view are not
> totally convergent.
> 
> The questions are:
> 
> 1. Is a good practice for CSW packages to contain library archives of
>    the form lib*.a when we deliver dynamic libraries lib*.so ?
> 
> 2. What's the potential usage for such libraries?
> 
> 3. Which mechanism do you use for not building (e.g. ./configure
>    --disable-static) or to post-install (in the sense of after make install)
>    remove them.
> 
> Thank you in advance for your answers and comments.

I would say if we are compiling a library that offers both static and shared
we compile them both.  there are specific instances when a user or an admin
may need to compile a tool statically ( albeit very rare ).

If we are compiling an "application" that offers both static and dynamic.
I would say compile them both and offer the static in a devel package along
with the headers.

My Rational:

There are companies that use small boxes as member interface points,
particularly Credit Card Companies.  these are boxes that live all over the
world, some have dial-up modems attached to them for access, some have x.25
interfaces (no I am not kidding), and bandwidth is at a premium.
The static binary may be bigger then the dynamic one, but on the whole, the one
static binary will be smaller then the possibly multiple packages.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc5/cACgkQLrhmsXMSLxdIfwCfVFuQg9lhvkxL2dWT5do0ILJz
150An3kCntXhYuPpijBXv3NIK+UaYzAG
=sNat
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Wed Apr  8 20:11:52 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Wed, 08 Apr 2009 14:11:52 -0400
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <20090408180150.GZ99983@bolthole.com>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
	<20090408180150.GZ99983@bolthole.com>
Message-ID: <1239214130-sup-790@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Wed Apr 08 14:01:50 -0400 2009:
> Occasionally, there may be benefits to making a static library for some
> software available for optional use. However, when and if the maintainer
> decides there is a benefit to providing a static lib, it is best to do so
> in a "devel" subpackage (unless the library is relatively small).
> 
> Other folks, please add your thoughts to this :-)

This lines up with the Debian policy and sounds quite reasonable to
me.  You shouldn't need the .a file unless you're building something
against it anyway, so the dev package is a good spot for it.

To save others the google, here is the debian policy link:
http://www.debian.org/doc/debian-policy/ch-sharedlibs.html#s-sharedlibs-static

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090408/43175e66/attachment.asc>

From phil at bolthole.com  Wed Apr  8 20:17:04 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 11:17:04 -0700
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <49DCE7F8.1040005@opencsw.org>
References: <x6vdpfkpfv.fsf@abhavana.free.fr> <49DCE7F8.1040005@opencsw.org>
Message-ID: <20090408181704.GA99983@bolthole.com>

On Wed, Apr 08, 2009 at 01:07:52PM -0500, Mike Watters wrote:
> If we are compiling an "application" that offers both static and dynamic.
> I would say compile them both and offer the static in a devel package along
> with the headers.

Err.. did you mean "library"?

we do not offer statically compiled "applications". all of our stuff is
compiled against the static lib.
[i also dont see how an "application" belongs in a devel package]



Peter was not proposing offering our own statically compiled applications;
only offering more static libraries for people to compile things themselves
statically.



From mwatters at opencsw.org  Wed Apr  8 20:18:34 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 13:18:34 -0500
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <1239214130-sup-790@ntdws12.chass.utoronto.ca>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>	<20090408180150.GZ99983@bolthole.com>
	<1239214130-sup-790@ntdws12.chass.utoronto.ca>
Message-ID: <49DCEA7A.30702@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ben Walton wrote:
> Excerpts from Philip Brown's message of Wed Apr 08 14:01:50 -0400 2009:
>> Occasionally, there may be benefits to making a static library for some
>> software available for optional use. However, when and if the maintainer
>> decides there is a benefit to providing a static lib, it is best to do so
>> in a "devel" subpackage (unless the library is relatively small).
>>
>> Other folks, please add your thoughts to this :-)
> 
> This lines up with the Debian policy and sounds quite reasonable to
> me.  You shouldn't need the .a file unless you're building something
> against it anyway, so the dev package is a good spot for it.
> 
> To save others the google, here is the debian policy link:
> http://www.debian.org/doc/debian-policy/ch-sharedlibs.html#s-sharedlibs-static
> 
> -Ben
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Fedora/Red Hat policy follows this convention as well.

and the following line quotes directly and I think it is the best rational.

"A good rule of thumb is if the file is used for development and not needed for
the base package to run properly, it should go in -devel."
http://fedoraproject.org/wiki/Packaging:Guidelines
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc6noACgkQLrhmsXMSLxefPQCfRoH8YnaYTjy3ZSmgkDecW+lW
iwkAoOTkvFhT6YQ04ZbhcFuSvvqeWL3q
=TvKO
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Wed Apr  8 20:21:31 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 13:21:31 -0500
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <20090408181704.GA99983@bolthole.com>
References: <x6vdpfkpfv.fsf@abhavana.free.fr> <49DCE7F8.1040005@opencsw.org>
	<20090408181704.GA99983@bolthole.com>
Message-ID: <49DCEB2B.3080705@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Wed, Apr 08, 2009 at 01:07:52PM -0500, Mike Watters wrote:
>> If we are compiling an "application" that offers both static and dynamic.
>> I would say compile them both and offer the static in a devel package along
>> with the headers.
> 
> Err.. did you mean "library"?
> 
> we do not offer statically compiled "applications". all of our stuff is
> compiled against the static lib.
> [i also dont see how an "application" belongs in a devel package]
> 
> 
> 
> Peter was not proposing offering our own statically compiled applications;
> only offering more static libraries for people to compile things themselves
> statically.
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

sort of,  "application" would be like mysql.

I guess I should have written it as...
If we are compiling an "application", such as mysql, that offer
both static and dynamic libraries, compile both and include the static library
in the devel package.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc6ysACgkQLrhmsXMSLxesRACdEfVe9sp2yBeBFHOvmtOI2GDu
6+QAoJJNuQyHp9v/0sBSBIg2uybecuzO
=QMoW
-----END PGP SIGNATURE-----


From hson at opencsw.org  Wed Apr  8 21:57:42 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 08 Apr 2009 21:57:42 +0200
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
Message-ID: <49DD01B6.2020301@opencsw.org>

I've got the impression that /opt/csw/var is a obsolete place to put 
things and that /var/opt/csw should be used instead.
Shouldn't this be reflected when using gar?
I.e gar.conf.mk should have localstatedir = /var/opt/csw instead of 
$(prefix)/var


From dam at opencsw.org  Wed Apr  8 21:58:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 21:58:45 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <625385e30904071518l767e9d5gf281d8a6dc9ce79b@mail.gmail.com>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
	<625385e30904071518l767e9d5gf281d8a6dc9ce79b@mail.gmail.com>
Message-ID: <F7C23A9B-8CA7-4F80-A453-07F744BBD4F7@opencsw.org>

Hi Peter,

Am 08.04.2009 um 00:18 schrieb Peter Bonivart:
> On Tue, Apr 7, 2009 at 5:23 PM, Dagobert Michelsen <dam at opencsw.org>  
> wrote:
>> in the past were requests for installation of experimental packages  
>> not
>> released yet to see how things go. However, the strict "released  
>> only"
>> policy on build8s prohibited this installation. To allow these
>> "experimental packages" I set up a new zone build8st (t=testing)
>> where packages from testing may be installed. Building packages
>> for release is not allowed on this machine! I am thinking of
>> entering both testing/ and current/ into the repository from pkgutil
>> and allow csw-group execution of pkgutil. Thought?
>
> Is the zone for building or just for installing packages for those who
> don't have Solaris 8 machines to test on?

It is for tests which require installation of packages from testing/,
for whatever this may be useful.


Best regards

    -- Dago


From dam at opencsw.org  Wed Apr  8 22:01:06 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:01:06 +0200
Subject: [csw-maintainers] March 31st - last day of Solaris 8 support
	from Sun
In-Reply-To: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com>
References: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com>
Message-ID: <71D80C94-0EBE-4E99-A281-C487FF52081B@opencsw.org>

Hi Peter,

Am 08.04.2009 um 00:23 schrieb Peter Bonivart:
> The day went by quietly here but earlier we have agreed that after
> this date we should no longer be required to build for Solaris 8. I
> know that Dago has prepared Solaris 9 build machines but is it
> officially ok to start using them?

Yes. Please note that it is no longer required to deliver for
Solaris 8, but it is still preferred. So if building on Solaris 8
is not harder than for Solaris 9, please continue to build on
Solaris 8.


Best regards

    -- Dago


From bonivart at opencsw.org  Wed Apr  8 22:05:05 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 22:05:05 +0200
Subject: [csw-maintainers] March 31st - last day of Solaris 8 support
	from Sun
In-Reply-To: <71D80C94-0EBE-4E99-A281-C487FF52081B@opencsw.org>
References: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com> 
	<71D80C94-0EBE-4E99-A281-C487FF52081B@opencsw.org>
Message-ID: <625385e30904081305o7c2b8d49o45f02be169da7e5@mail.gmail.com>

On Wed, Apr 8, 2009 at 10:01 PM, Dagobert Michelsen <dam at opencsw.org> wrote:
> Yes. Please note that it is no longer required to deliver for
> Solaris 8, but it is still preferred. So if building on Solaris 8
> is not harder than for Solaris 9, please continue to build on
> Solaris 8.

That sounds reasonable. In my case I will try to build on Solaris 8
first and if it works - fine, but I will not try to get upstream to
fix stuff for Solaris 8 like I recently did for ClamAV. Unless it
doesn't work on Solaris 9 either, then I will have to beg again. :-)

-- 
/peter


From dam at opencsw.org  Wed Apr  8 22:14:09 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:14:09 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <49DC8658.8020908@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>	<49DAEDF6.8050404@opencsw.org>
	<20090408.10054300.1414585890@gyor.oxdrove.co.uk>
	<49DC8658.8020908@opencsw.org>
Message-ID: <9FC00626-A3C3-4B3E-BBCE-CE46C42683A4@opencsw.org>

Hi Roger,

Am 08.04.2009 um 13:11 schrieb Roger H?kansson:
> James Lee wrote:
>>> I can see more problem (as in more work for the
>>> maintainer in the future) with actively unsetting LD_OPTIONS than  
>>> not.
>> Sorry I can't see why unsetting something has only been set by  
>> yourself
>> in the first place is more work.  In fact we should always explicitly
>> either set or unset such controlling environmental variables as a
>> matter of good practice.
>
> When building packages with gar, the default (i.e without the  
> maintainer doing a thing) is that the RPATH will contain /opt/csw/ 
> lib/$ISALIST:/opt/csw/lib. If a maintainer is to package something  
> with a different RPATH he must set it manually, i.e more work than  
> default.

This is done because it works in almost all cases. As maintainer you
can choose to optimize this, though. Of course the default action
is open for discussion ;-)


Best regards

   -- Dago

From dam at opencsw.org  Wed Apr  8 22:24:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:24:17 +0200
Subject: [csw-maintainers] [csw-buildfarm] build8s and build9s
	inconsistancies
In-Reply-To: <49DCE271.9050304@opencsw.org>
References: <49DCCAFE.2040900@cognigencorp.com> <49DCE271.9050304@opencsw.org>
Message-ID: <5569F5BD-77D9-4172-AE7B-5A36325EFD65@opencsw.org>

Hi Mike,

Am 08.04.2009 um 19:44 schrieb Mike Watters:
> I can't even get onto build9s or build9x, I get prompted for password.
> since home dir's are mounted across the env, I have to assume I  
> don't have an
> account on those servers.

There was a transitional time when I made a script for Phil for
adding users which had a bug. Phil used it, I fixed the bug and
your and your account alone has been forgotten :-( Fixed now.
If you or anybody else encounters this: please post ASAP.


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  8 22:29:00 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:29:00 +0200
Subject: [csw-maintainers] [csw-buildfarm] build8s and build9s
	inconsistancies
In-Reply-To: <49DCCAFE.2040900@cognigencorp.com>
References: <49DCCAFE.2040900@cognigencorp.com>
Message-ID: <A2D3DB4A-B7CC-4E10-81B6-C89BD42DA88B@opencsw.org>

Hi Darin,

Am 08.04.2009 um 18:04 schrieb Darin Perusich:
> I was building a package, amanda, on build8s and everything was going
> fine but given that Solaris 8 is no longer supported I figured I'd  
> move
> over to the Solaris 9 build machines.

Please continue to build on Solaris 8 per default until there is a
specific reason why you build for Solaris 9. Solaris 8 is no longer
mandatory, but still recommended.

> The code compiles fine but the
> staging of the package is incomplete on build9s, whether I use
> 'stagepkg' or 'gmake DESTDIR=$PWD/cswstage install' the results are  
> the
> same.
>
> The staging fails when setting permission, see below, on some of the
> binaries and I have no idea why. I compared the CSW packages and they
> are identical on both machine so I'm wondering if its something in  
> the OS.
>
> Has anyone else run across this before?
>
> gmake[5]: Entering directory `/home/darin/build/amanda-2.6.1/common- 
> src'
>  (dest=/opt/csw/sbin)
>  (chown=amanda)
> chown darin:sys
> /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/ 
> amgpgcrypt:
> Not owner
> gmake[5]: *** [installperms-data] Error 1
> gmake[5]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[4]: *** [install-data-am] Error 2
> gmake[4]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[3]: *** [install-am] Error 2
> gmake[3]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[2]: *** [install] Error 2
> gmake[2]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[1]: *** [install-recursive] Error 1
> gmake[1]: Leaving directory `/home/darin/build/amanda-2.6.1'
> gmake: *** [install] Error 2

I cannot reproduce this:

> root at login [login]:/root > su - darin
> Sun Microsystems Inc.   SunOS 5.10      Generic January 2005
>
> You are on the OpenCSW Login Server from Baltic Online.
> To get an overview of the setup please read /etc/SETUP.
>
> <login:~> ssh build9s
> Last login: Wed Apr  8 17:37:52 2009 from 192.168.1.2
> Sun Microsystems Inc.   SunOS 5.9       Generic May 2002
> <build9s:~> chown darin:sys /home/darin/build/amanda-2.6.1/cswstage/ 
> opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/ 
> amgpgcrypt: Not owner
> <build9s:~> exit
> logout
> Connection to build9s closed.
> <login:~> ssh build8s
> Last login: Wed Apr  8 22:26:57 2009 from 192.168.1.2
> <build8s:~> chown darin:sys /home/darin/build/amanda-2.6.1/cswstage/ 
> opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/ 
> amgpgcrypt: Not owner


Best regards

   -- Dago



From william at wbonnet.net  Wed Apr  8 22:33:45 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Apr 2009 22:33:45 +0200
Subject: [csw-maintainers] glib updates
Message-ID: <49DD0A29.2030209@wbonnet.net>

Hi

A good news :)

Recent updates libs update like Glib2 solveed the segfault problems i 
had with FF2 last versions. I am checking TB last version right now.

I will certainly release a FF2 update this week.

Cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From dam at opencsw.org  Wed Apr  8 22:38:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:38:43 +0200
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <x6vdpfkpfv.fsf@abhavana.free.fr>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
Message-ID: <3C086376-7F83-4B3E-8618-D763D5382351@opencsw.org>

Hi Peter,

Am 08.04.2009 um 19:47 schrieb Peter FELECAN:
> I cannot recollect precisely if we already had this discussion. Phil  
> and
> me, we had an exchange on this subject and I would like to have the
> opinion of the other fellow maintainers as our point of view are not
> totally convergent.
>
> The questions are:
>
> 1. Is a good practice for CSW packages to contain library archives of
>   the form lib*.a when we deliver dynamic libraries lib*.so ?
>
> 2. What's the potential usage for such libraries?
>
> 3. Which mechanism do you use for not building (e.g. ./configure
>   --disable-static) or to post-install (in the sense of after make  
> install)
>   remove them.

You can skip the build like you described. Additionally, GAR skips  
static
libraries per default during the merge phase (copying stuff from the
install-*/ to pkgroot/). If you want static libs you must set
   MERGE_EXCLUDE_STATICLIBS =      (nothing)
The use of special rules is discouraged.


Best regards

   -- Dago


From phil at bolthole.com  Wed Apr  8 22:45:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 13:45:06 -0700
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <3C086376-7F83-4B3E-8618-D763D5382351@opencsw.org>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
	<3C086376-7F83-4B3E-8618-D763D5382351@opencsw.org>
Message-ID: <20090408204506.GD99983@bolthole.com>

On Wed, Apr 08, 2009 at 10:38:43PM +0200, Dagobert Michelsen wrote:
> You can skip the build like you described. Additionally, GAR skips  
> static
> libraries per default during the merge phase (copying stuff from the
> install-*/ to pkgroot/). If you want static libs you must set
>   MERGE_EXCLUDE_STATICLIBS =      (nothing)
> The use of special rules is discouraged.

that's... really odd naming, again



From dam at opencsw.org  Wed Apr  8 22:54:06 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:54:06 +0200
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
In-Reply-To: <49DD01B6.2020301@opencsw.org>
References: <49DD01B6.2020301@opencsw.org>
Message-ID: <AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>

Hi Roger,

Am 08.04.2009 um 21:57 schrieb Roger H?kansson:
> I've got the impression that /opt/csw/var is a obsolete place to put  
> things and that /var/opt/csw should be used instead.
> Shouldn't this be reflected when using gar?
> I.e gar.conf.mk should have localstatedir = /var/opt/csw instead of $ 
> (prefix)/var

I guess you are right. But this may break existing packages. Maybe  
this is
a candidate for mGAR v3? I would like to have some feedback from more
package maintainers on this change.

This is also partly true for sysconfdir = /opt/csw/etc vs. /etc/opt/csw


Best regards

   -- Dago

From phil at bolthole.com  Wed Apr  8 23:00:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 14:00:56 -0700
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
In-Reply-To: <AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
References: <49DD01B6.2020301@opencsw.org>
	<AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
Message-ID: <20090408210056.GE99983@bolthole.com>

On Wed, Apr 08, 2009 at 10:54:06PM +0200, Dagobert Michelsen wrote:
> Hi Roger,
>
> Am 08.04.2009 um 21:57 schrieb Roger H?kansson:
>> I've got the impression that /opt/csw/var is a obsolete place to put  
>> things and that /var/opt/csw should be used instead.
>> Shouldn't this be reflected when using gar?
>> I.e gar.conf.mk should have localstatedir = /var/opt/csw instead of $ 
>> (prefix)/var
>
> I guess you are right. But this may break existing packages. Maybe this 
> is
> a candidate for mGAR v3? I would like to have some feedback from more
> package maintainers on this change.

then they need to be "broken", and then fixed.



> This is also partly true for sysconfdir = /opt/csw/etc vs. /etc/opt/csw

"partly" is accurate.
Sometimes, it is still appropriate for sysconfdir to be /opt/csw/etc



From bonivart at opencsw.org  Wed Apr  8 23:00:41 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 23:00:41 +0200
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
In-Reply-To: <AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
References: <49DD01B6.2020301@opencsw.org>
	<AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
Message-ID: <625385e30904081400u4a9922dch5ad309b46dad2768@mail.gmail.com>

On Wed, Apr 8, 2009 at 10:54 PM, Dagobert Michelsen <dam at opencsw.org> wrote:
> I guess you are right. But this may break existing packages. Maybe this is
> a candidate for mGAR v3? I would like to have some feedback from more
> package maintainers on this change.
>
> This is also partly true for sysconfdir = /opt/csw/etc vs. /etc/opt/csw

I always set /var/opt/csw and usually /etcopt/csw.

-- 
/peter


From maybird1776 at yahoo.com  Thu Apr  9 00:07:03 2009
From: maybird1776 at yahoo.com (ken mays)
Date: Wed, 8 Apr 2009 15:07:03 -0700 (PDT)
Subject: [csw-maintainers] glib updates
Message-ID: <803218.42804.qm@web111309.mail.gq1.yahoo.com>


> William Bonnet wrote:
> Recent updates like Glib2 solved the segfault problems i had
> with FF2 last versions. I am checking TB last version right now.

Well, well, well...

See, it was not such a bad idea after all.

~ Ken



      


From hson at opencsw.org  Thu Apr  9 04:51:29 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Thu, 09 Apr 2009 04:51:29 +0200
Subject: [csw-maintainers] For the gnome team
Message-ID: <49DD62B1.9050602@opencsw.org>

I've been packaging some packages which were either listed as orphaned 
or where the maintainer is retired, which I think would fall under the 
gnome domain.
Since I personally don't use any gnome packages at all (don't even run X 
at all), I'm handing them off to the gnome team to finish.

gnumeric and libgoffice builds just fine as long as you have libgsf 1.14.
However I've not been able to get around the fact that libgsf-gnome-1.so 
get linked to the installed libgsf instead of the newly built one (see 
my previous posts regarding this).
So in order to build libgsf the current package has to be removed from 
the build server (which I've been testing on my own build servers)

I've also been trying to build libgda and libgnomedb (which are optional 
for gnumeric), but there seems to be some incompatibility from upstream 
so libgnomedb won't built atm.


From mwatters at opencsw.org  Thu Apr  9 07:05:36 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Apr 2009 00:05:36 -0500
Subject: [csw-maintainers] sudo and sudo.ldap in testing
Message-ID: <49DD8220.9060605@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	updated to gar v2 and sudo 1.7.0
	force incomparable between CSWsudo and CSWsudo-ldap

main compile options:

- --with-pam           (use pam authentication rules)
- --with-logging=both  (file and/or syslog for logging)
- --with-ignore-dot    (ignore 'dot' in path for sudo)
- --with-env-editor    (honor users EDITOR environment variable for visudo)

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkndgiAACgkQLrhmsXMSLxd/BACgthcvnihNjVCSFU1Y0JRrmxp1
Z2wAoKzB4M7WxZq9yOTNnG9KLwlM/t4E
=YWL/
-----END PGP SIGNATURE-----


From trygvis at opencsw.org  Thu Apr  9 13:29:12 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 09 Apr 2009 13:29:12 +0200
Subject: [csw-maintainers] Issues with pkg-get 4.2
Message-ID: <49DDDC08.9020605@opencsw.org>

I'm working on updating the maven2 package, but I can't seem to get 
pkg-get to download the package (I tried with just "maven2" as name too).

It does work with the pkg-get in current.

$ sudo pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u 
maven2-2.0.10,REV=2009.04.09
Getting catalog...
--2009-04-09 13:22:24-- 
http://mirror.opencsw.org/opencsw/testing/i386/5.11/catalog
Resolving mirror.opencsw.org... 213.178.77.176
Connecting to mirror.opencsw.org|213.178.77.176|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 29410 (29K) [text/plain]
Saving to: `catalog'

100%[============================================================================================================>] 
29,410      86.5K/s   in 0.3s

2009-04-09 13:22:24 (86.5 KB/s) - `catalog' saved [29410/29410]

Updating catalog file
/var/pkg-get/catalog-mirror.opencsw.org updated

--2009-04-09 13:22:24-- 
http://mirror.opencsw.org/opencsw/testing/i386/5.11/descriptions
Resolving mirror.opencsw.org... 213.178.77.176
Connecting to mirror.opencsw.org|213.178.77.176|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 7114 (6.9K) [text/plain]
Saving to: `descriptions'

100%[============================================================================================================>] 
7,114       --.-K/s   in 0.07s

2009-04-09 13:22:25 (94.3 KB/s) - `descriptions' saved [7114/7114]

Updated description file
Need TWO args to newer_rev

pkg-get version:
$ pkg-get
pkg-get,   by Philip Brown , phil at bolthole.com
  (Internal SCCS code revision @(#) pkg-get 4.2@(#))
Originally from http://www.bolthole.com/solaris/pkg-get.html
[snip]

--
Trygve


From Darin.Perusich at cognigencorp.com  Thu Apr  9 15:57:19 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 09 Apr 2009 09:57:19 -0400
Subject: [csw-maintainers] cswusergroup questions
Message-ID: <49DDFEBF.1020905@cognigencorp.com>

I've been reviewing the cswusergroup class script and have a few quick
questions. The docs recommend using the
/opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
/opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
this location be created as part of that packages?

What are peoples thoughts about being able to set password policies for
the users we're creating? This can be a touchy area but given that some
services will fail if the users password has expired it might be nice to
be able to set them.

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From harpchad at opencsw.org  Thu Apr  9 17:12:12 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 09 Apr 2009 10:12:12 -0500
Subject: [csw-maintainers] New in testing: fontconfig 2.6.0
In-Reply-To: <FD21DEAC-14E7-4479-89B2-3E4BF458738A@opencsw.org>
References: <FD21DEAC-14E7-4479-89B2-3E4BF458738A@opencsw.org>
Message-ID: <49DE104C.1060300@opencsw.org>

Broke SUNWgnome for me, appears that the new font caches aren't 
backwards compatible with the old ones?

running '/usr/bin/fc-cache -f' (SUNWfontconfig version 2.2.3) fixed it, 
but probably broke the CSW version.

Alexander Maier wrote:
> Hi,
> 
> current fontconfig version is now in testing at 
> http://mirror.opencsw.org/testing/ :
> 
> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-i386-CSW.pkg.gz
> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-sparc-CSW.pkg.gz
> 
> Please give it a try.
> 
> -Alexander
> 
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From harpchad at opencsw.org  Thu Apr  9 17:32:16 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 09 Apr 2009 10:32:16 -0500
Subject: [csw-maintainers] New in testing: fontconfig 2.6.0
In-Reply-To: <49DE104C.1060300@opencsw.org>
References: <FD21DEAC-14E7-4479-89B2-3E4BF458738A@opencsw.org>
	<49DE104C.1060300@opencsw.org>
Message-ID: <49DE1500.9060407@opencsw.org>

Looks like the old version created its own separate (-csw) cache, but 
the new one doesn't?

Chad Harp wrote:
> Broke SUNWgnome for me, appears that the new font caches aren't 
> backwards compatible with the old ones?
> 
> running '/usr/bin/fc-cache -f' (SUNWfontconfig version 2.2.3) fixed it, 
> but probably broke the CSW version.
> 
> Alexander Maier wrote:
>> Hi,
>>
>> current fontconfig version is now in testing at 
>> http://mirror.opencsw.org/testing/ :
>>
>> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-i386-CSW.pkg.gz
>> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-sparc-CSW.pkg.gz
>>
>> Please give it a try.
>>
>> -Alexander
>>
>>
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From phil at bolthole.com  Thu Apr  9 18:02:29 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 09:02:29 -0700
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <49DDFEBF.1020905@cognigencorp.com>
References: <49DDFEBF.1020905@cognigencorp.com>
Message-ID: <20090409160229.GG28226@bolthole.com>

On Thu, Apr 09, 2009 at 09:57:19AM -0400, Darin Perusich wrote:
> I've been reviewing the cswusergroup class script and have a few quick
> questions. The docs recommend using the
> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> this location be created as part of that packages?

that would be an option, sure.


> What are peoples thoughts about being able to set password policies for
> the users we're creating? This can be a touchy area but given that some
> services will fail if the users password has expired it might be nice to
> be able to set them.

Leave it to the site admins, i would suggest


From phil at bolthole.com  Thu Apr  9 18:04:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 09:04:33 -0700
Subject: [csw-maintainers] Issues with pkg-get 4.2
In-Reply-To: <49DDDC08.9020605@opencsw.org>
References: <49DDDC08.9020605@opencsw.org>
Message-ID: <20090409160433.GI28226@bolthole.com>

On Thu, Apr 09, 2009 at 01:29:12PM +0200, Trygve Laugst?l wrote:
> pkg-get version:
> $ pkg-get
> pkg-get,   by Philip Brown , phil at bolthole.com
>  (Internal SCCS code revision @(#) pkg-get 4.2@(#))
> Originally from http://www.bolthole.com/solaris/pkg-get.html

please update your pkg-get pkg


From bonivart at opencsw.org  Thu Apr  9 18:11:30 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 9 Apr 2009 18:11:30 +0200
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <20090409160229.GG28226@bolthole.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<20090409160229.GG28226@bolthole.com>
Message-ID: <625385e30904090911s10867d0ex611b15977edac954@mail.gmail.com>

On Thu, Apr 9, 2009 at 6:02 PM, Philip Brown <phil at bolthole.com> wrote:
> On Thu, Apr 09, 2009 at 09:57:19AM -0400, Darin Perusich wrote:
>> I've been reviewing the cswusergroup class script and have a few quick
>> questions. The docs recommend using the
>> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
>> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
>> this location be created as part of that packages?
>
> that would be an option, sure.

If we recommend using that location I think it should be included in CSWcommon.

-- 
/peter


From phil at bolthole.com  Thu Apr  9 18:17:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 09:17:26 -0700
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <625385e30904090911s10867d0ex611b15977edac954@mail.gmail.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<20090409160229.GG28226@bolthole.com>
	<625385e30904090911s10867d0ex611b15977edac954@mail.gmail.com>
Message-ID: <20090409161726.GK28226@bolthole.com>

On Thu, Apr 09, 2009 at 06:11:30PM +0200, Peter Bonivart wrote:
> On Thu, Apr 9, 2009 at 6:02 PM, Philip Brown <phil at bolthole.com> wrote:
> > On Thu, Apr 09, 2009 at 09:57:19AM -0400, Darin Perusich wrote:
> >> I've been reviewing the cswusergroup class script and have a few quick
> >> questions. The docs recommend using the
> >> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> >> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> >> this location be created as part of that packages?
> >
> > that would be an option, sure.
> 
> If we recommend using that location I think it should be included in CSWcommon.

But it only gets used by cswclassutils? :-)



From Darin.Perusich at cognigencorp.com  Thu Apr  9 20:01:17 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 09 Apr 2009 14:01:17 -0400
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <49DDFEBF.1020905@cognigencorp.com>
References: <49DDFEBF.1020905@cognigencorp.com>
Message-ID: <49DE37ED.6070603@cognigencorp.com>

Also, given these are system account what about setting the UID's to
below 100?

Darin Perusich wrote:
> I've been reviewing the cswusergroup class script and have a few quick
> questions. The docs recommend using the
> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> this location be created as part of that packages?
> 
> What are peoples thoughts about being able to set password policies for
> the users we're creating? This can be a touchy area but given that some
> services will fail if the users password has expired it might be nice to
> be able to set them.
> 

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Thu Apr  9 20:07:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 11:07:25 -0700
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <49DE37ED.6070603@cognigencorp.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<49DE37ED.6070603@cognigencorp.com>
Message-ID: <20090409180725.GA16708@bolthole.com>

On Thu, Apr 09, 2009 at 02:01:17PM -0400, Darin Perusich wrote:
> Also, given these are system account what about setting the UID's to
> below 100?

nonono... leave that to the local site admins. 

and/or, tweak cswusergroup to have a config option in csw.conf or
whereever, to specify preferred range of UIDs when it has to add UIDs


> 
> Darin Perusich wrote:
> > I've been reviewing the cswusergroup class script and have a few quick
> > questions. The docs recommend using the
> > /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> > /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> > this location be created as part of that packages?
> > 
> > What are peoples thoughts about being able to set password policies for
> > the users we're creating? This can be a touchy area but given that some
> > services will fail if the users password has expired it might be nice to
> > be able to set them.


From mwatters at opencsw.org  Thu Apr  9 21:13:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Apr 2009 14:13:52 -0500
Subject: [csw-maintainers] updated: sudo in testing
Message-ID: <49DE48F0.9070003@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	repackaged after fixing permissions on the sudo* binaries.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkneSO8ACgkQLrhmsXMSLxc6rwCbBY2B/0Gf1YQpzMIBwL3uwNkZ
xDgAoJKlEWUWLIYckJ98eKSQzzTEk+6i
=/dN8
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Apr  9 22:46:58 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Apr 2009 22:46:58 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <49DD62B1.9050602@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
Message-ID: <C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>

Hi Roger,

Am 09.04.2009 um 04:51 schrieb Roger H?kansson:
> I've been packaging some packages which were either listed as  
> orphaned or where the maintainer is retired, which I think would  
> fall under the gnome domain.
> Since I personally don't use any gnome packages at all (don't even  
> run X at all), I'm handing them off to the gnome team to finish.
>
> gnumeric and libgoffice builds just fine as long as you have libgsf  
> 1.14.
> However I've not been able to get around the fact that libgsf- 
> gnome-1.so get linked to the installed libgsf instead of the newly  
> built one (see my previous posts regarding this).
> So in order to build libgsf the current package has to be removed  
> from the build server (which I've been testing on my own build  
> servers)

I just build it from SVN without any problems, maybe the LD_OPTIONS  
change was
more good than expected?

Anyway, libgsf is now in testing:
   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-i386-CSW.pkg.gz
   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-sparc-CSW.pkg.gz

> I've also been trying to build libgda and libgnomedb (which are  
> optional for gnumeric), but there seems to be some incompatibility  
> from upstream so libgnomedb won't built atm.

Nope, no incompatibility here. Builds absolutely smooth. Packages in  
testing/ now.
Thanks for the good work!

BTW: To get GNOME running it would be nice if we had someone who  
coordinates what
packages need to be built most urgent. Volunteers?


Best regards

   -- Dago

From phil at bolthole.com  Thu Apr  9 22:58:04 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 13:58:04 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
Message-ID: <20090409205804.GC16708@bolthole.com>

On Thu, Apr 09, 2009 at 10:46:58PM +0200, Dagobert Michelsen wrote:
>
> BTW: To get GNOME running it would be nice if we had someone who  
> coordinates what
> packages need to be built most urgent. Volunteers?

Pffft... we dont need a "coordinator". we need someone to
BUILD AND TEST packages :-)

Any volunteers for that?

[Note: Please volunteer by stating,
  "I'm going to have package X dropped in newpkgs by tomorrow", not
"well, yes I'd like to volunteer and I'll make a whole lot of
     packages ... when i have more time...."  ]




From hson at opencsw.org  Fri Apr 10 02:19:32 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Fri, 10 Apr 2009 02:19:32 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
Message-ID: <49DE9094.4070504@opencsw.org>

Dagobert Michelsen wrote:
> I just build it from SVN without any problems, maybe the LD_OPTIONS 
> change was
> more good than expected?
> 
> Anyway, libgsf is now in testing:
>   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-i386-CSW.pkg.gz
>   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-sparc-CSW.pkg.gz

Maybe I was a bit unclear, the package will build just fine, however 
libgsf-gnome-1.so isn't linked to the libgsf-1.so that the build produce, 
but the old one in /opt/csw/lib.
And I've been trying to fix this in various ways but with no luck.
So far the only working solution I've found is to remove libgsf from the 
build machine when packaging.

hson at build8x :~ > dump -Lv 
/home/dam/mgar/pkg/libgsf/trunk/work/pkgroot/opt/csw/lib/libgsf-gnome-1.so.114.0.11
...
[1]     NEEDED          libgsf-1.so.1
...
[16]    SONAME          libgsf-gnome-1.so.114


>> I've also been trying to build libgda and libgnomedb (which are 
>> optional for gnumeric), but there seems to be some incompatibility 
>> from upstream so libgnomedb won't built atm.
> 
> Nope, no incompatibility here. Builds absolutely smooth. Packages in 
> testing/ now.
> Thanks for the good work!

Hmm, I get a bunch of "undefined struct/union member" when building 
libgnomedb on my build machine, but I've installed the libgda I packaged, I 
guess the buildfarm have the older one?
Otherwise its probably some other package that is missing from my build 
machine which is installed on the build farm...


From hson at opencsw.org  Fri Apr 10 02:34:10 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Fri, 10 Apr 2009 02:34:10 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <20090409205804.GC16708@bolthole.com>
References: <49DD62B1.9050602@opencsw.org>	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
Message-ID: <49DE9402.3000301@opencsw.org>

Philip Brown wrote:
> On Thu, Apr 09, 2009 at 10:46:58PM +0200, Dagobert Michelsen wrote:
>> BTW: To get GNOME running it would be nice if we had someone who  
>> coordinates what
>> packages need to be built most urgent. Volunteers?
> 
> Pffft... we dont need a "coordinator". we need someone to
> BUILD AND TEST packages :-)

I think think there is a need for some coordination too.
There is a lot of dependencies for the gnome packages and in order to get 
them in sync with upstream there has to be some kind of synchronized work.


> Any volunteers for that?
> 
> [Note: Please volunteer by stating,
>   "I'm going to have package X dropped in newpkgs by tomorrow", not
> "well, yes I'd like to volunteer and I'll make a whole lot of
>      packages ... when i have more time...."  ]

Well, as I stated in private mails to Phil, I use very few CSW packages 
other than those either already properly maintained (i.e fairly current with 
upstream and with active maintainer), or those I've become maintainer for (I 
don't even use all of those).

I don't even run X as a desktop (Windows on desktop, Linux and Solaris on 
servers) so I can't even test the packages.
But I'm willing to help pushing existing packages into gar to the point a 
package is correctly built.


This far, I've been picking random packages that were either listed as 
orphaned or with a "retired" maintainer.
Right now I'm focusing on packages with many bugs reported (netsnmp have 
been my main focus for instance, now I'm looking at the bacula and 
postgresql packages)


From phil at bolthole.com  Fri Apr 10 03:02:30 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 18:02:30 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <49DE9402.3000301@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
	<49DE9402.3000301@opencsw.org>
Message-ID: <20090410010230.GB31623@bolthole.com>

On Fri, Apr 10, 2009 at 02:34:10AM +0200, Roger H?kansson wrote:
> I think think there is a need for some coordination too.
> There is a lot of dependencies for the gnome packages and in order to get 
> them in sync with upstream there has to be some kind of synchronized 
> work.

To have a need for synchronization, you first need "multiple threads".
A single threaded operation, never needs "synchronization".
(nor does 'zero threads' :-)

Do we have two or more people currently working on compiling GNOME
packages? 
please speak up, because I'm not aware of many people currently working in
parallel on this that need synchronization.



Right now, we just have occasional, sporadic updates of low level GNOME
packages, like glib.
Which is GREAT! But... we have no one i know of, who is committing to build
the whole chain.
Or even a specific large PIECE of the chain.
We're just picking one one or two packages here and there.

This effort can be self-organizing; the main "difficult" step, is people
who are going to do the work.
When someone decides to do the work, then just announce, "I'm building
these packages now".
There; problem of "synchronization" solved.


But what the heck, lets make it easier to remember who said they're doing
what.

http://wiki.opencsw.org/gnome
now exists, so that people planning to work on multiple GNOME packages, can
let others know which ones are being worked on already.

The gnome effort consists of basically two types of components:

1. core libs (10 to 20?)
 
2. a plethora of "auxiliary stuff".


For anyone who needs/wants to work on gnome, the work flow is relatively
straightforward.

1. Pick a gnome lib or two that you wish to work on, that is out of date
 (dont know what to pick? take a look through
       http://wiki.opencsw.org/maintainers/kenmays 
   for packages with "lib" in the name, for ideas)t



2. Take a look at its dependancies, and make sure they are up to date.
   If not, then work on its dependancies.
   Otherwise, just update the lib that you chose.

 If desired, return to step 1, and pick another package.




From mwatters at opencsw.org  Fri Apr 10 04:20:39 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Apr 2009 21:20:39 -0500
Subject: [csw-maintainers] gcc 4.3.3
Message-ID: <49DEACF7.5020803@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I am at a loss,  I have combined the sparc builds in
just about every conceivable way and can not get one package to work on
both solaris 8 and solaris 10. (did not try solaris 9)

Unless someone can provide a solution, (my recipe is up to date in svn)
I will package up OS specific packages on Monday.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknerPcACgkQLrhmsXMSLxcRbwCfcJ/eUlU1aCLsljRb8CoahhvJ
H1gAoKW4N03e4s1TAzJeWjprCOGV3kir
=e+TM
-----END PGP SIGNATURE-----


From phil at bolthole.com  Fri Apr 10 07:11:49 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 22:11:49 -0700
Subject: [csw-maintainers] gcc 4.3.3
In-Reply-To: <49DEACF7.5020803@opencsw.org>
References: <49DEACF7.5020803@opencsw.org>
Message-ID: <20090410051149.GA8309@bolthole.com>

On Thu, Apr 09, 2009 at 09:20:39PM -0500, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I am at a loss,  I have combined the sparc builds in
> just about every conceivable way and can not get one package to work on
> both solaris 8 and solaris 10. (did not try solaris 9)

btw; did you ask on the gcc mailing list?


From maybird1776 at yahoo.com  Fri Apr 10 13:00:21 2009
From: maybird1776 at yahoo.com (ken mays)
Date: Fri, 10 Apr 2009 04:00:21 -0700 (PDT)
Subject: [csw-maintainers] For the gnome team
Message-ID: <144698.96991.qm@web111305.mail.gq1.yahoo.com>


I can assist in the GNOME packaging development - just not right now due to current demands.

~K


--- On Thu, 4/9/09, Philip Brown <phil at bolthole.com> wrote:

> From: Philip Brown <phil at bolthole.com>
> Subject: Re: [csw-maintainers] For the gnome team
> To: maintainers at lists.opencsw.org
> Date: Thursday, April 9, 2009, 9:02 PM
> On Fri, Apr 10, 2009 at 02:34:10AM
> +0200, Roger H?kansson wrote:
> > I think think there is a need for some coordination
> too.
> > There is a lot of dependencies for the gnome packages
> and in order to get 
> > them in sync with upstream there has to be some kind
> of synchronized 
> > work.
> 
> To have a need for synchronization, you first need
> "multiple threads".
> A single threaded operation, never needs
> "synchronization".
> (nor does 'zero threads' :-)
> 
> Do we have two or more people currently working on
> compiling GNOME
> packages? 
> please speak up, because I'm not aware of many people
> currently working in
> parallel on this that need synchronization.
> 
> 
> 
> Right now, we just have occasional, sporadic updates of low
> level GNOME
> packages, like glib.
> Which is GREAT! But... we have no one i know of, who is
> committing to build
> the whole chain.
> Or even a specific large PIECE of the chain.
> We're just picking one one or two packages here and there.
> 
> This effort can be self-organizing; the main "difficult"
> step, is people
> who are going to do the work.
> When someone decides to do the work, then just announce,
> "I'm building
> these packages now".
> There; problem of "synchronization" solved.
> 
> 
> But what the heck, lets make it easier to remember who said
> they're doing
> what.
> 
> http://wiki.opencsw.org/gnome
> now exists, so that people planning to work on multiple
> GNOME packages, can
> let others know which ones are being worked on already.
> 
> The gnome effort consists of basically two types of
> components:
> 
> 1. core libs (10 to 20?)
>  
> 2. a plethora of "auxiliary stuff".
> 
> 
> For anyone who needs/wants to work on gnome, the work flow
> is relatively
> straightforward.
> 
> 1. Pick a gnome lib or two that you wish to work on, that
> is out of date
>  (dont know what to pick? take a look through
> ? ? ???http://wiki.opencsw.org/maintainers/kenmays 
> ???for packages with "lib" in the name, for
> ideas)t
> 
> 
> 
> 2. Take a look at its dependancies, and make sure they are
> up to date.
> ???If not, then work on its dependancies.
> ???Otherwise, just update the lib that you
> chose.
> 
>  If desired, return to step 1, and pick another package.
> 
> 
> 
> -----Inline Attachment Follows-----
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers


      



From mwatters at opencsw.org  Fri Apr 10 22:00:19 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 10 Apr 2009 15:00:19 -0500
Subject: [csw-maintainers] gcc 4.3.3
In-Reply-To: <20090410051149.GA8309@bolthole.com>
References: <49DEACF7.5020803@opencsw.org> <20090410051149.GA8309@bolthole.com>
Message-ID: <49DFA553.8080101@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Apr 09, 2009 at 09:20:39PM -0500, Mike Watters wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> I am at a loss,  I have combined the sparc builds in
>> just about every conceivable way and can not get one package to work on
>> both solaris 8 and solaris 10. (did not try solaris 9)
> 
> btw; did you ask on the gcc mailing list?
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I sent the question to gcc-help mailing list this morning.
I am waiting for an answer.  I will still packages up OS specific packages on
Monday if I don't get a solution.  If the gcc or this list comes back with a
solution, I will re-package.
There are too many packages waiting on amd64 gcc to hold this any longer.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknfpVMACgkQLrhmsXMSLxdV6ACfQRJGVGuwu6kCMffhbCk3eYU+
G8sAoOkB4ur9yxCXCvtqGfjrjhflRir8
=z8VL
-----END PGP SIGNATURE-----


From hson at opencsw.org  Sat Apr 11 02:06:15 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Sat, 11 Apr 2009 02:06:15 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <20090410010230.GB31623@bolthole.com>
References: <49DD62B1.9050602@opencsw.org>	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>	<20090409205804.GC16708@bolthole.com>	<49DE9402.3000301@opencsw.org>
	<20090410010230.GB31623@bolthole.com>
Message-ID: <49DFDEF7.7090309@opencsw.org>

Philip Brown wrote:
> Do we have two or more people currently working on compiling GNOME
> packages? 

I seem to remember some posts regarding someone joining the "gnome team"...


From rupert at opencsw.org  Sat Apr 11 12:29:49 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 11 Apr 2009 12:29:49 +0200
Subject: [csw-maintainers] python-reportlab 2.3
Message-ID: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com>

hi,

for getting http://trac-hacks.org/wiki/TracWikiPrintPlugin we would
need http://reportlab.org/ v2.3 (http://www.reportlab.org/ftp/) . i
saw there is a package http://opencsw.org/packages/reportlab, v2.1.

as it is not in
http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg, where would
be the source to try it?

rupert.


From pfelecan at opencsw.org  Sat Apr 11 14:41:47 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Sat, 11 Apr 2009 14:41:47 +0200
Subject: [csw-maintainers] gcc 4.3.3
In-Reply-To: <49DFA553.8080101@opencsw.org> (Mike Watters's message of "Fri\,
	10 Apr 2009 15\:00\:19 -0500")
References: <49DEACF7.5020803@opencsw.org>
	<20090410051149.GA8309@bolthole.com> <49DFA553.8080101@opencsw.org>
Message-ID: <x63acfjras.fsf@abhavana.free.fr>

Mike Watters <mwatters at opencsw.org> writes:

> Philip Brown wrote:
>> On Thu, Apr 09, 2009 at 09:20:39PM -0500, Mike Watters wrote:
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>>
>>> I am at a loss,  I have combined the sparc builds in
>>> just about every conceivable way and can not get one package to work on
>>> both solaris 8 and solaris 10. (did not try solaris 9)
>> 
>> btw; did you ask on the gcc mailing list?
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
>
> I sent the question to gcc-help mailing list this morning.
> I am waiting for an answer.  I will still packages up OS specific packages on
> Monday if I don't get a solution.  If the gcc or this list comes back with a
> solution, I will re-package.
> There are too many packages waiting on amd64 gcc to hold this any longer.

Don't wanting to revive old discussions but I still don't understand all
this fuss about 64bit being so critical. Still, I can appreciate your
efforts and hope that you'll find a solution.
-- 
Peter


From dam at opencsw.org  Sun Apr 12 00:34:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Apr 2009 00:34:42 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <20090409205804.GC16708@bolthole.com>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
Message-ID: <038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>

Hi Phil,

Am 09.04.2009 um 22:58 schrieb Philip Brown:
> On Thu, Apr 09, 2009 at 10:46:58PM +0200, Dagobert Michelsen wrote:
>>
>> BTW: To get GNOME running it would be nice if we had someone who
>> coordinates what
>> packages need to be built most urgent. Volunteers?
>
> Pffft... we dont need a "coordinator".

I disagree here. It would help if there was someone who would say:
"We have a roadmap, these 3 libs need updating first, than we
test that application and than proceed to these packages". At
least I wouldn't know where in this vast dependency building
the bottom building blocks are.

> we need someone to BUILD AND TEST packages :-) Any volunteers for  
> that?

These will come if the necessary packages have been identified.


Best regards

   -- Dago


From dam at opencsw.org  Sun Apr 12 00:36:30 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Apr 2009 00:36:30 +0200
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <20090409180725.GA16708@bolthole.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<49DE37ED.6070603@cognigencorp.com>
	<20090409180725.GA16708@bolthole.com>
Message-ID: <49D25B28-7EC6-4034-A433-EA08587265D7@opencsw.org>

Hi Phil,

Am 09.04.2009 um 20:07 schrieb Philip Brown:
> and/or, tweak cswusergroup to have a config option in csw.conf or
> whereever, to specify preferred range of UIDs when it has to add UIDs

That is a very good idea! Please take this as a filed features  
request ;-)


Best regards

   -- Dago


From ihsan at opencsw.org  Sun Apr 12 14:46:26 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 12 Apr 2009 14:46:26 +0200
Subject: [csw-maintainers] problems with new libtool
Message-ID: <49E1E2A2.2010704@opencsw.org>

Hello,

Since libtool was updated on the buildfarm, I'm having troubles to build
 Apache 2.2.

checking whether it is safe to define __EXTENSIONS__... yes
checking for library containing strerror... none required
checking whether system uses EBCDIC... no
performing libtool configuration...
/home/ihsan/gar/csw/mgar/pkg/apache2/trunk/work/build-isa-sparcv8/httpd-2.2.11/srclib/apr/configure:
line 9748: syntax error near unexpected token
`lt_if_append_uniq(lt_decl_varnames,'
/home/ihsan/gar/csw/mgar/pkg/apache2/trunk/work/build-isa-sparcv8/httpd-2.2.11/srclib/apr/configure:
line 9748: `lt_if_append_uniq(lt_decl_varnames, SED, , ,'
configure failed for srclib/apr
gmake[1]: *** [configure-work/build-isa-sparcv8/httpd-2.2.11/configure]
Error 1
gmake[1]: Leaving directory `/home/ihsan/gar/csw/mgar/pkg/apache2/trunk'
gmake: *** [merge-isa-sparcv8] Error 2

With the old libtool, it was running just fine. I'm not sure if this is
related to Apache, so I'm wondering if others are also having troubles.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Mon Apr 13 16:43:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 13 Apr 2009 07:43:31 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
	<038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
Message-ID: <20090413144331.GA51320@bolthole.com>

On Sun, Apr 12, 2009 at 12:34:42AM +0200, Dagobert Michelsen wrote:
> I disagree here. It would help if there was someone who would say:
> "We have a roadmap, these 3 libs need updating first, than we
> test that application and than proceed to these packages". At
> least I wouldn't know where in this vast dependency building
> the bottom building blocks are.

i agree that it would be helpful to have a dependancy tree/map,
particularly noting which are out of date, and which are not.

That is not the same thing as having "a coordinator".

that title implies something with more responsibility.



From phil at bolthole.com  Mon Apr 13 16:54:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 13 Apr 2009 07:54:06 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
	<038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
Message-ID: <20090413145406.GB51320@bolthole.com>

On Sun, Apr 12, 2009 at 12:34:42AM +0200, Dagobert Michelsen wrote:
>> Phil wrote:
>> we need someone to BUILD AND TEST packages :-) Any volunteers for  
>> that?
>
> These will come if the necessary packages have been identified.
>

ok, lets test that theory. 
A good intermediate dependancy target is libgnome.


lower level libraries that we need updated, to update libgnome
packages, in approximate "order":

fontconfig (its actually in testing, but needs to be released!!)
libcairo
libxrender
libpango
libatk
gconf2
orbit2
libbonobo2, libbonoboui
gnomevfs2
libgnome




From dam at opencsw.org  Tue Apr 14 13:18:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 13:18:15 +0200
Subject: [csw-maintainers] libcairo
Message-ID: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>

Hi,

I jumped in on the work on libcairo. William already prepared most of
what is necessary and I push in the latest GAR tweaks and provide 64  
bit.
Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>


Best regards

   -- Dago


From dam at opencsw.org  Tue Apr 14 13:43:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 13:43:40 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
Message-ID: <99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>

Hi,

Am 14.04.2009 um 13:18 schrieb Dagobert Michelsen:
> I jumped in on the work on libcairo. William already prepared most of
> what is necessary and I push in the latest GAR tweaks and provide 64  
> bit.
> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>

New in testing/:

libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz


Best regards

   -- Dago


From Darin.Perusich at cognigencorp.com  Tue Apr 14 22:31:44 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Tue, 14 Apr 2009 16:31:44 -0400
Subject: [csw-maintainers] /testing amanda 2.6.1p1
Message-ID: <49E4F2B0.9000107@cognigencorp.com>

Hello,

I'd like to announce that latest release of Amanda, the world's most
popular Open Source Backup and Archiving software, is available for
testing via the testing repository. For a listing of features see the
Amanda Wiki at http://wiki.zmanda.com/index.php/2.6.1_features. Please
use Mantis to report any issues with this package.

To install Amanda directly from the testing repository run the following:

pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u amanda

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From dam at opencsw.org  Tue Apr 14 22:32:12 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 22:32:12 +0200
Subject: [csw-maintainers] libcurl and 64 bit on x86
Message-ID: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>

Hi,

I am having problems compiling vorbistools in 64 bit on x86:

> Making all in ogg123
> gmake[4]: Entering directory `/home/dam/mgar/pkg/vorbistools/trunk/ 
> work/build-isa-amd64/vorbis-tools-1.2.0/ogg123'
> source='http_transport.c' object='http_transport.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DSYSCONFDIR=\"/opt/csw/etc 
> \" -DLOCALEDIR=\"/opt/csw/share/locale\" -DHAVE_CONFIG_H -I. -I.. -I/ 
> opt/csw/include   -I/opt/csw/include   -I/opt/csw/include -I/opt/csw/ 
> include -D_REENTRANT  -I../include -I../intl  -I/opt/csw/include - 
> D_REENTRANT  -O -xO3 -xarch=amd64 -I/opt/csw/include -c  
> http_transport.c
> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative  
> subscript
> cc: acomp failed for http_transport.c
> gmake[4]: *** [http_transport.o] Error 2

This is

> /*
>  * Verify that the size previously defined and expected for long
>  * is the same as the one reported by sizeof() at compile time.
>  */
>
> typedef char
>   __curl_rule_01__
>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];

CURL_SIZEOF_LONG is 4:

> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4


I guess another set of includes is needed for 64 bit, right?


Best regards

   -- Dago





From harpchad at opencsw.org  Tue Apr 14 22:41:43 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 14 Apr 2009 15:41:43 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
Message-ID: <49E4F507.1040503@opencsw.org>

Possibly, but the current version of curl doesn't include amd64 support 
since openldap doesn't have amd64 libraries (Mantis 3028). So I suspect 
you'd run into linking problems even if you were able to compile.

Did you have any issue with sparcv9?

Dagobert Michelsen wrote:
> Hi,
> 
> I am having problems compiling vorbistools in 64 bit on x86:
> 
>> Making all in ogg123
>> gmake[4]: Entering directory 
>> `/home/dam/mgar/pkg/vorbistools/trunk/work/build-isa-amd64/vorbis-tools-1.2.0/ogg123' 
>>
>> source='http_transport.c' object='http_transport.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc 
>> -DSYSCONFDIR=\"/opt/csw/etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" 
>> -DHAVE_CONFIG_H -I. -I.. -I/opt/csw/include   -I/opt/csw/include   
>> -I/opt/csw/include -I/opt/csw/include -D_REENTRANT  -I../include 
>> -I../intl  -I/opt/csw/include -D_REENTRANT  -O -xO3 -xarch=amd64 
>> -I/opt/csw/include -c http_transport.c
>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative subscript
>> cc: acomp failed for http_transport.c
>> gmake[4]: *** [http_transport.o] Error 2
> 
> This is
> 
>> /*
>>  * Verify that the size previously defined and expected for long
>>  * is the same as the one reported by sizeof() at compile time.
>>  */
>>
>> typedef char
>>   __curl_rule_01__
>>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];
> 
> CURL_SIZEOF_LONG is 4:
> 
>> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4
> 
> 
> I guess another set of includes is needed for 64 bit, right?
> 
> 
> Best regards
> 
>   -- Dago
> 
> 
> 



From dam at opencsw.org  Tue Apr 14 22:59:09 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 22:59:09 +0200
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E4F507.1040503@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
Message-ID: <239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>

Hi Chad,

Am 14.04.2009 um 22:41 schrieb Chad Harp:
> Possibly, but the current version of curl doesn't include amd64  
> support since openldap doesn't have amd64 libraries (Mantis 3028).

Damn. Want to take a look at OpenLDAP?

> So I suspect you'd run into linking problems even if you were able  
> to compile.
>
> Did you have any issue with sparcv9?

Yes, same issue:

> gmake[4]: Entering directory `/home/dam/mgar/pkg/vorbistools/trunk/ 
> work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123'
> source='http_transport.c' object='http_transport.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DSYSCONFDIR=\"/opt/csw/etc 
> \" -DLOCALEDIR=\"/opt/csw/share/locale\" -DHAVE_CONFIG_H -I. -I.. -I/ 
> opt/csw/include   -I/opt/csw/include   -I/opt/csw/include -I/opt/csw/ 
> include -D_REENTRANT  -I../include -I../intl  -I/opt/csw/include - 
> D_REENTRANT  -xO4 -fast -w -fsimple -native -xcg92 -xO3 -xarch=v9 -I/ 
> opt/csw/include -c http_transport.c
> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative  
> subscript
> cc: acomp failed for http_transport.c


Best regards

   -- Dago


From harpchad at opencsw.org  Tue Apr 14 23:57:34 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 14 Apr 2009 16:57:34 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
Message-ID: <49E506CE.50900@opencsw.org>

Ok, I wrote a quick program to test it and am able to replicate this. 
I'll work on a new build to fix the sparcv9.

I'll take a look at openldap and see what's involved.  All I really need 
is the library, I don't want to take on the server and other stuff as I 
don't use it (and wouldn't be able to test it).

Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 14.04.2009 um 22:41 schrieb Chad Harp:
>> Possibly, but the current version of curl doesn't include amd64 
>> support since openldap doesn't have amd64 libraries (Mantis 3028).
> 
> Damn. Want to take a look at OpenLDAP?
> 
>> So I suspect you'd run into linking problems even if you were able to 
>> compile.
>>
>> Did you have any issue with sparcv9?
> 
> Yes, same issue:
> 
>> gmake[4]: Entering directory 
>> `/home/dam/mgar/pkg/vorbistools/trunk/work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123' 
>>
>> source='http_transport.c' object='http_transport.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc 
>> -DSYSCONFDIR=\"/opt/csw/etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" 
>> -DHAVE_CONFIG_H -I. -I.. -I/opt/csw/include   -I/opt/csw/include   
>> -I/opt/csw/include -I/opt/csw/include -D_REENTRANT  -I../include 
>> -I../intl  -I/opt/csw/include -D_REENTRANT  -xO4 -fast -w -fsimple 
>> -native -xcg92 -xO3 -xarch=v9 -I/opt/csw/include -c http_transport.c
>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative subscript
>> cc: acomp failed for http_transport.c
> 
> 
> Best regards
> 
>   -- Dago



From harpchad at opencsw.org  Wed Apr 15 02:06:11 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 14 Apr 2009 19:06:11 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E506CE.50900@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E506CE.50900@opencsw.org>
Message-ID: <72B3CB61-A6E3-4E13-A83D-62FA3BF40103@opencsw.org>

Looks like I'll need to wrap the curlbuild.h

Fedora does this:

#include <bits/wordsize.h>

#if __WORDSIZE == 32
#include "curlbuild-32.h"
#elif __WORDSIZE == 64
#include "curlbuild-64.h"
#else
#error "Unknown word size"
#endif


We don't (I think) have a __WORDSIZE macro anywhere, so I'm thinking  
of this:

#if defined __arch64__ || defined __sparcv9
#include "curlbuild-64.h"
#else
#include "curlbuild-32.h"
#endif

I'm not positive about my macro choices, anybody want to shoot some  
holes in that before I commit (will it still include the 32 bit  
headers when building 32bit on a 64bit platform)?

On Apr 14, 2009, at 4:57 PM, Chad Harp wrote:

> Ok, I wrote a quick program to test it and am able to replicate  
> this. I'll work on a new build to fix the sparcv9.
>
> I'll take a look at openldap and see what's involved.  All I really  
> need is the library, I don't want to take on the server and other  
> stuff as I don't use it (and wouldn't be able to test it).
>
> Dagobert Michelsen wrote:
>> Hi Chad,
>> Am 14.04.2009 um 22:41 schrieb Chad Harp:
>>> Possibly, but the current version of curl doesn't include amd64  
>>> support since openldap doesn't have amd64 libraries (Mantis 3028).
>> Damn. Want to take a look at OpenLDAP?
>>> So I suspect you'd run into linking problems even if you were  
>>> able to compile.
>>>
>>> Did you have any issue with sparcv9?
>> Yes, same issue:
>>> gmake[4]: Entering directory `/home/dam/mgar/pkg/vorbistools/ 
>>> trunk/work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123'
>>> source='http_transport.c' object='http_transport.o' libtool=no \
>>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>>         /opt/studio/SOS11/SUNWspro/bin/cc -DSYSCONFDIR=\"/opt/csw/ 
>>> etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" -DHAVE_CONFIG_H -I. - 
>>> I.. -I/opt/csw/include   -I/opt/csw/include   -I/opt/csw/include - 
>>> I/opt/csw/include -D_REENTRANT  -I../include -I../intl  -I/opt/ 
>>> csw/include -D_REENTRANT  -xO4 -fast -w -fsimple -native -xcg92 - 
>>> xO3 -xarch=v9 -I/opt/csw/include -c http_transport.c
>>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative  
>>> subscript
>>> cc: acomp failed for http_transport.c
>> Best regards
>>   -- Dago
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From james at opencsw.org  Wed Apr 15 11:02:29 2009
From: james at opencsw.org (James Lee)
Date: Wed, 15 Apr 2009 09:02:29 GMT
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
Message-ID: <20090415.9022900.4244226440@gyor.oxdrove.co.uk>

On 14/04/09, 21:32:12, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
[csw-maintainers] libcurl and 64 bit on x86:

> >  * Verify that the size previously defined and expected for long
> >  * is the same as the one reported by sizeof() at compile time.
> >  */
> >
> > typedef char
> >   __curl_rule_01__
> >     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];

> CURL_SIZEOF_LONG is 4:

> > /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4


#define CURL_SIZEOF_LONG sizeof(long)




James.


From harpchad at opencsw.org  Wed Apr 15 15:57:51 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 08:57:51 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415.9022900.4244226440@gyor.oxdrove.co.uk>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<20090415.9022900.4244226440@gyor.oxdrove.co.uk>
Message-ID: <49E5E7DF.1030003@opencsw.org>

James Lee wrote:
> On 14/04/09, 21:32:12, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
> [csw-maintainers] libcurl and 64 bit on x86:
> 
>>>  * Verify that the size previously defined and expected for long
>>>  * is the same as the one reported by sizeof() at compile time.
>>>  */
>>>
>>> typedef char
>>>   __curl_rule_01__
>>>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];
> 
>> CURL_SIZEOF_LONG is 4:
> 
>>> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4
> 
> 
> #define CURL_SIZEOF_LONG sizeof(long)
> 
> 
> 
> 
> James.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

curl does a check to make sure that CURL_SIZEOF_LONG == sizeof(long), so 
   by changing CURL_SIZEOF_LONG to sizeof(long) I'd just be saying
sizeof(long) == sizeof(long). :)

There's a discussion going on over on the curl mailing list as to how to 
make this more friendly towards packaging for multiple architectures, 
but the consensus for now seems to be to wrap the headers.


From phil at bolthole.com  Wed Apr 15 16:03:36 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 07:03:36 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5E7DF.1030003@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<20090415.9022900.4244226440@gyor.oxdrove.co.uk>
	<49E5E7DF.1030003@opencsw.org>
Message-ID: <20090415140336.GA55760@bolthole.com>

On Wed, Apr 15, 2009 at 08:57:51AM -0500, Chad Harp wrote:
>
> curl does a check to make sure that CURL_SIZEOF_LONG == sizeof(long), so  
>   by changing CURL_SIZEOF_LONG to sizeof(long) I'd just be saying
> sizeof(long) == sizeof(long). :)

sounds appropriate, really.

possibly the reason the define for CURL_SIZEOF_LONG exists, is because in
some situations and with some compilers, it must have a fixed number in the
define, rather than an operation-style macro.

but if our compiler is happy with it, then just use it.
although we might need  extra parens, to be safe:

#define CURL_SIZEOF_LONG (sizeof(long))


From james at opencsw.org  Wed Apr 15 16:23:39 2009
From: james at opencsw.org (James Lee)
Date: Wed, 15 Apr 2009 14:23:39 GMT
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5E7DF.1030003@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<20090415.9022900.4244226440@gyor.oxdrove.co.uk>
	<49E5E7DF.1030003@opencsw.org>
Message-ID: <20090415.14233900.2233379359@gyor.oxdrove.co.uk>

On 15/04/09, 14:57:51, Chad Harp <harpchad at opencsw.org> wrote regarding Re:
[csw-maintainers] libcurl and 64 bit on x86:

> >>>  * Verify that the size previously defined and expected for long
> >>>  * is the same as the one reported by sizeof() at compile time.
> >>>  */
> >>>
> >>> typedef char
> >>>   __curl_rule_01__
> >>>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];
> >
> >> CURL_SIZEOF_LONG is 4:
> >
> >>> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4
> >
> >
> > #define CURL_SIZEOF_LONG sizeof(long)


> curl does a check to make sure that CURL_SIZEOF_LONG == sizeof(long) so
>    by changing CURL_SIZEOF_LONG to sizeof(long) I'd just be saying
> sizeof(long) == sizeof(long). :)

It will be.  Another chuck of code we can cut.


> There's a discussion going on over on the curl mailing list as to how to
> make this more friendly towards packaging for multiple architectures,
> but the consensus for now seems to be to wrap the headers.

Do they get paid for code by the yard?  I don't understand what problem
they are trying to solve, at least I thought I did, but this was all
solved decades ago by the sizeof operator.




James.


From harpchad at opencsw.org  Wed Apr 15 16:43:12 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 09:43:12 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
Message-ID: <49E5F280.7070503@opencsw.org>

I've put a new curldevel package in testing (sparc only, since x86 
doesn't have 64-bit support right now).

curldevel-7.19.4,REV=2009.04.15-SunOS5.8-sparc-CSW.pkg.gz

It's wraps curlbuild.h using the following:

/* Allow 32 and 64 bit headers to coexist */
#if defined __arch64__ || defined __sparcv9
#include "curlbuild-64.h"
#else
#include "curlbuild-32.h"
#endif

I opted not to make the changes to curlbuild.h because there were 
several lines (other than the sizeof(long) we discussed) that would have 
to change (see diff below).  I think this will be more easily adapted to 
future versions.



harpchad at build8s (CSW)$ diff curlbuild-32.h curlbuild-64.h
108c108
< #define CURL_PULL_SYS_TYPES_H 1
---
 > /* #undef CURL_PULL_SYS_TYPES_H */
122c122
< #define CURL_PULL_INTTYPES_H 1
---
 > /* #undef CURL_PULL_INTTYPES_H */
128c128
< #define CURL_SIZEOF_LONG 4
---
 > #define CURL_SIZEOF_LONG 8
131c131
< #define CURL_TYPEOF_CURL_OFF_T int64_t
---
 > #define CURL_TYPEOF_CURL_OFF_T long
137c137
< #define CURL_FORMAT_CURL_OFF_T "lld"
---
 > #define CURL_FORMAT_CURL_OFF_T "ld"
140c140
< #define CURL_FORMAT_CURL_OFF_TU "llu"
---
 > #define CURL_FORMAT_CURL_OFF_TU "lu"
143c143
< #define CURL_FORMAT_OFF_T "%lld"
---
 > #define CURL_FORMAT_OFF_T "%ld"
149c149
< #define CURL_SUFFIX_CURL_OFF_T LL
---
 > #define CURL_SUFFIX_CURL_OFF_T L
152c152
< #define CURL_SUFFIX_CURL_OFF_TU ULL
---
 > #define CURL_SUFFIX_CURL_OFF_TU UL


Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 14.04.2009 um 22:41 schrieb Chad Harp:
>> Possibly, but the current version of curl doesn't include amd64 
>> support since openldap doesn't have amd64 libraries (Mantis 3028).
> 
> Damn. Want to take a look at OpenLDAP?
> 
>> So I suspect you'd run into linking problems even if you were able to 
>> compile.
>>
>> Did you have any issue with sparcv9?
> 
> Yes, same issue:
> 
>> gmake[4]: Entering directory 
>> `/home/dam/mgar/pkg/vorbistools/trunk/work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123' 
>>
>> source='http_transport.c' object='http_transport.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc 
>> -DSYSCONFDIR=\"/opt/csw/etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" 
>> -DHAVE_CONFIG_H -I. -I.. -I/opt/csw/include   -I/opt/csw/include   
>> -I/opt/csw/include -I/opt/csw/include -D_REENTRANT  -I../include 
>> -I../intl  -I/opt/csw/include -D_REENTRANT  -xO4 -fast -w -fsimple 
>> -native -xcg92 -xO3 -xarch=v9 -I/opt/csw/include -c http_transport.c
>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative subscript
>> cc: acomp failed for http_transport.c
> 
> 
> Best regards
> 
>   -- Dago



From phil at bolthole.com  Wed Apr 15 16:59:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 07:59:31 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5F280.7070503@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
Message-ID: <20090415145931.GC16035@bolthole.com>

On Wed, Apr 15, 2009 at 09:43:12AM -0500, Chad Harp wrote:
> I opted not to make the changes to curlbuild.h because there were  
> several lines (other than the sizeof(long) we discussed) that would have  
> to change (see diff below).  I think this will be more easily adapted to  
> future versions.

for the record, I think you chose the wrong replacement values. please take
a deeper look into sys/types.h.

for example

> > #define CURL_TYPEOF_CURL_OFF_T long

should be

#define CURL_TYPEOF_CURL_OFF_T offset_t

and CURL_TYPEOF_CURL_OFF_TU should be u_offset_t

This will work cleanly for both 32bit and 64bit.




From harpchad at opencsw.org  Wed Apr 15 17:11:28 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 10:11:28 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415145931.GC16035@bolthole.com>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>	<49E4F507.1040503@opencsw.org>	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
Message-ID: <49E5F920.1050101@opencsw.org>

Philip Brown wrote:
> for the record, I think you chose the wrong replacement values. please take
> a deeper look into sys/types.h.
> 
> for example
> 
>>> #define CURL_TYPEOF_CURL_OFF_T long
> 
> should be
> 
> #define CURL_TYPEOF_CURL_OFF_T offset_t
> 
> and CURL_TYPEOF_CURL_OFF_TU should be u_offset_t
> 
> This will work cleanly for both 32bit and 64bit.
> 
> 

I didn't choose those values but I can propose them upstream.  I simply 
take the two headers generated from the builds and wrap them.  If I try 
to make a custom curlbuild.h I'm going to have to redo it every time 
there is a new curl release.



From phil at bolthole.com  Wed Apr 15 17:40:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 08:40:33 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5F920.1050101@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
Message-ID: <20090415154033.GA96009@bolthole.com>

On Wed, Apr 15, 2009 at 10:11:28AM -0500, Chad Harp wrote:
> I didn't choose those values but I can propose them upstream.  I simply  
> take the two headers generated from the builds and wrap them.  If I try  
> to make a custom curlbuild.h I'm going to have to redo it every time  
> there is a new curl release.
>

right now, you have to add two files, AND patch a header file.

If you do as I suggest, you only have to patch a single header file.
It is less work :)

(however, if you do the patch in a way that is portable,then perhaps
  upstream will accept it in the future and it will be even better)



From harpchad at opencsw.org  Wed Apr 15 17:49:02 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 10:49:02 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415154033.GA96009@bolthole.com>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>	<49E4F507.1040503@opencsw.org>	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>	<49E5F280.7070503@opencsw.org>	<20090415145931.GC16035@bolthole.com>	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
Message-ID: <49E601EE.9020309@opencsw.org>

Philip Brown wrote:
> On Wed, Apr 15, 2009 at 10:11:28AM -0500, Chad Harp wrote:
>> I didn't choose those values but I can propose them upstream.  I simply  
>> take the two headers generated from the builds and wrap them.  If I try  
>> to make a custom curlbuild.h I'm going to have to redo it every time  
>> there is a new curl release.
>>
> 
> right now, you have to add two files, AND patch a header file.
> 
> If you do as I suggest, you only have to patch a single header file.
> It is less work :)
> 
> (however, if you do the patch in a way that is portable,then perhaps
>   upstream will accept it in the future and it will be even better)
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I guess I'm not following, I don't patch anything, I just take the 
headers that curl generates and use one or the other based on the 
compile environment.

What are the appropriate portable format and suffix values for offset_t 
and u_offset_t?


From ihsan at opencsw.org  Wed Apr 15 17:50:04 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 15 Apr 2009 17:50:04 +0200
Subject: [csw-maintainers] wget Paket
In-Reply-To: <AB71266F-5B6B-4969-B5EB-A6107C344891@opencsw.org>
References: <49D235A7.2030605@opencsw.org>
	<AB71266F-5B6B-4969-B5EB-A6107C344891@opencsw.org>
Message-ID: <49E6022C.6020703@opencsw.org>

Hi Dago,

Am 31.3.2009 21:16 Uhr, Dagobert Michelsen schrieb:

>> I was thinking, that I could run a second build process with different
>> "configure" options. You provide already a possibility with the
>> modulations, but I don't need an optimized version. I know this is a
>> special case, but is this possible with Gar?
> 
> Absolutely. I just finished ncurses with a modulation for wide-characters.
> Just take a look at
>  
> <http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/ncurses/trunk/Makefile>

Worked perfectly.
Thanks you.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From dam at opencsw.org  Wed Apr 15 17:56:20 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 15 Apr 2009 17:56:20 +0200
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415154033.GA96009@bolthole.com>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
Message-ID: <A1CE492E-EC3D-4DFF-809D-4A08AB666B65@opencsw.org>

Hi,

Am 15.04.2009 um 17:40 schrieb Philip Brown:
> On Wed, Apr 15, 2009 at 10:11:28AM -0500, Chad Harp wrote:
>> I didn't choose those values but I can propose them upstream.  I  
>> simply
>> take the two headers generated from the builds and wrap them.  If I  
>> try
>> to make a custom curlbuild.h I'm going to have to redo it every time
>> there is a new curl release.
>>
>
> right now, you have to add two files, AND patch a header file.

Well, now there is only one static conditional include. Everything else
is done automatically and needs no work on upstream version bump. Think
of this solution as least upstream dependency and least longterm effort.

The package in testing is working fine.


Best regards

   -- Dago


From valholla75 at gmail.com  Wed Apr 15 19:33:45 2009
From: valholla75 at gmail.com (Mike Watters)
Date: Wed, 15 Apr 2009 12:33:45 -0500
Subject: [csw-maintainers] gcc4 now in testing
Message-ID: <49E61A79.6090805@gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

gcc 4.3.3  NOW IN TESTING

OS specific compiles for:

solaris  8 "sparcv8, sparcv9" and i386
solaris  9 "sparcv8, sparcv9" and i386
solaris 10 "sparcv8, sparcv9" and "i386, amd64"

Results from make checks are checked into subversion
mgar/pkg/gcc4/trunk/files/test-results/

solaris9 test results are coming.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknmGnkACgkQW3VzXti4wcraeACfXKOfb8gYYmtktIuK1AjM9FiW
zIwAnjMmuK9AtF35UAx86PiDQd0bC4a0
=wzjZ
-----END PGP SIGNATURE-----


From phil at bolthole.com  Wed Apr 15 19:44:10 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:44:10 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E601EE.9020309@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
	<49E601EE.9020309@opencsw.org>
Message-ID: <20090415174410.GE976@bolthole.com>

On Wed, Apr 15, 2009 at 10:49:02AM -0500, Chad Harp wrote:
> I guess I'm not following, I don't patch anything, I just take the  
> headers that curl generates and use one or the other based on the  
> compile environment.

ohhhh.



> What are the appropriate portable format and suffix values for offset_t  
> and u_offset_t?

I dont understand the question.



From phil at bolthole.com  Wed Apr 15 19:44:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:44:52 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <A1CE492E-EC3D-4DFF-809D-4A08AB666B65@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
	<A1CE492E-EC3D-4DFF-809D-4A08AB666B65@opencsw.org>
Message-ID: <20090415174452.GF976@bolthole.com>

On Wed, Apr 15, 2009 at 05:56:20PM +0200, Dagobert Michelsen wrote:
> Well, now there is only one static conditional include. Everything else
> is done automatically and needs no work on upstream version bump. Think
> of this solution as least upstream dependency and least longterm effort.

I think the least LONGTERM effort, is to get upstream to have their code
work more portably.

which involves more shortterm work, of writing a patch.



From phil at bolthole.com  Wed Apr 15 19:48:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:48:06 -0700
Subject: [csw-maintainers] gcc4 now in testing
In-Reply-To: <49E61A79.6090805@gmail.com>
References: <49E61A79.6090805@gmail.com>
Message-ID: <20090415174806.GG976@bolthole.com>

On Wed, Apr 15, 2009 at 12:33:45PM -0500, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> gcc 4.3.3  NOW IN TESTING
> 
> OS specific compiles for:
> 
> solaris  8 "sparcv8, sparcv9" and i386
> solaris  9 "sparcv8, sparcv9" and i386
> solaris 10 "sparcv8, sparcv9" and "i386, amd64"

they dont follow prior package path conventions.

there's no gcc4 prefix.



From phil at bolthole.com  Wed Apr 15 19:52:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:52:45 -0700
Subject: [csw-maintainers] gcc4 now in testing
In-Reply-To: <20090415174806.GG976@bolthole.com>
References: <49E61A79.6090805@gmail.com> <20090415174806.GG976@bolthole.com>
Message-ID: <20090415175245.GI976@bolthole.com>

On Wed, Apr 15, 2009 at 10:48:06AM -0700, Philip Brown wrote:
> On Wed, Apr 15, 2009 at 12:33:45PM -0500, Mike Watters wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> > 
> > gcc 4.3.3  NOW IN TESTING
> > 
> > OS specific compiles for:
> > 
> > solaris  8 "sparcv8, sparcv9" and i386
> > solaris  9 "sparcv8, sparcv9" and i386
> > solaris 10 "sparcv8, sparcv9" and "i386, amd64"
> 
> they dont follow prior package path conventions.
> 
> there's no gcc4 prefix.

correction: some do, some dont.

gcc4core has gcc4 prefix.

gcc4corert does not.



From ihsan at opencsw.org  Wed Apr 15 20:38:06 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 15 Apr 2009 20:38:06 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>
Message-ID: <49E6298E.2030801@opencsw.org>

Hello Dago,

Am 14.4.2009 13:43 Uhr, Dagobert Michelsen schrieb:

>> I jumped in on the work on libcairo. William already prepared most of
>> what is necessary and I push in the latest GAR tweaks and provide 64 bit.
>> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>
> 
> New in testing/:
> 
> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz

Could you please install those libraries on build8st, so I could build
rrdtool there?



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From dam at opencsw.org  Wed Apr 15 21:42:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 15 Apr 2009 21:42:46 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <49E6298E.2030801@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>
	<49E6298E.2030801@opencsw.org>
Message-ID: <73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>

Hi Ihsan,

Am 15.04.2009 um 20:38 schrieb Ihsan Dogan:
> Am 14.4.2009 13:43 Uhr, Dagobert Michelsen schrieb:
>
>>> I jumped in on the work on libcairo. William already prepared most  
>>> of
>>> what is necessary and I push in the latest GAR tweaks and provide  
>>> 64 bit.
>>> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/ 
>>> gnome>
>>
>> New in testing/:
>>
>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>
> Could you please install those libraries on build8st, so I could build
> rrdtool there?

Done. This is exactly what build8st is for :-)


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 09:04:59 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 09:04:59 +0200
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
Message-ID: <0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>

Hi Chad,

Am 16.04.2009 um 03:05 schrieb Chad Harp:
> On Feb 24, 2009, at 2:05 PM, Chad Harp wrote:
>>
>> Dagobert Michelsen wrote:
>>> Hi Chad,
>>> I noticed there is an error in the current CSWftype2
>>> 2.3.8,REV=2009.02.16 in
>>>   /opt/csw/include/ft2build.h
>>> with the line
>>>   #include <freetype/config/ftheader.h>
>>> The path is instead
>>>   /opt/csw/include/freetype2/freetype/config/ftheader.h
>>> I guess it should either be in
>>>   /opt/csw/include/freetype2/ft2build.h
>>> or contain
>>>   #include <freetype2/freetype/config/ftheader.h>
>>
>>
>> That looks correct to me, can you send me more details on where  
>> you're getting the error?
>>
>> #include <freetype/config/ftheader.h>
>> with -I/opt/csw/include/freetype2
>> should find
>> /opt/csw/include/freetype2/freetype/config/ftheader.h
>>
>> Perhaps what you're compiling isn't using freetype-config?
>>
>> /opt/csw/bin/freetype-config --cflags
>> -I/opt/csw/include/freetype2 -I/opt/csw/include
>

> I was going through my list of things to look into and ran across  
> this.  I can't remember what the outcome was, did you get past this  
> error?

And:

Am 20.02.2009 um 12:56 schrieb Ihsan Dogan:
> Am 17.2.2009 19:03 Uhr, Dagobert Michelsen schrieb:
>>
>>> brauchst zum Kompilieren die aktuelle freetype2 Version aus /
>>> testing.
>>> Koenntest Du das bitte auf der buildfarm einspielen ?
>>> Ziel ist es, freetype2 und fontconfig nach einer Testphase
>>> gleichzeitig zu releasen.
>>
>> Ok, so freetype2 and fontconfig should be released at the
>> same time and freetype2 must be installed on the buildfarm
>> for building fontconfig.
>>
>> Ihsan, Phil: Should I do it this way violating standards or
>> should these packages be released one after another?
>
> As I understand, we don't have many choices, so it's fine for me.
> Important is, that a notify is sent to the maintainers list.

We have not build8st for this. AFAIK this has not been done yet.
Want to give it a try?


Best regards

   -- Dagp


From dam at opencsw.org  Thu Apr 16 10:05:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 10:05:45 +0200
Subject: [csw-maintainers] Package name now contains commit status
Message-ID: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>

Hi,

according to Bens great work the package name now contains the
repository status if the package has not been fully checked in.

The usual name is
   %{bitname}-%{SPKG_VERSION}%{SPKG_REVSTAMP}-%{SPKG_OSNAME}-%{arch}- 
CSW.pkg

Instead of the CSW there are now three other possible values:
   UNCOMMITTED  Not every file has been checked in or ignored
                ('svn status' not empty)
   NOSVN        The svn binary could not be found
   NOTVERSIONED The tree is not under version control

The change is in effect since r4338.


Best regards

   -- Dago


From ihsan at opencsw.org  Thu Apr 16 12:59:44 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 12:59:44 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>	<49E6298E.2030801@opencsw.org>
	<73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>
Message-ID: <49E70FA0.90700@opencsw.org>

Hi Dago,

Am 15.4.2009 21:42 Uhr, Dagobert Michelsen schrieb:

>>>> I jumped in on the work on libcairo. William already prepared most of
>>>> what is necessary and I push in the latest GAR tweaks and provide 64
>>>> bit.
>>>> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>
>>>
>>> New in testing/:
>>>
>>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>>
>> Could you please install those libraries on build8st, so I could build
>> rrdtool there?
> 
> Done. This is exactly what build8st is for :-)

Thanks.

Is it possible, that the new cairo packages are breaking pango?

Find 3rd-Party Libraries
checking for cairo_font_options_create in -lcairo... yes
checking cairo.h usability... no
checking cairo.h presence... no
checking for cairo.h... no
checking for pkg-config... pkg-config
checking for cairo_font_options_create in -lcairo... yes
checking cairo.h usability... yes
checking cairo.h presence... yes
checking for cairo.h... yes
checking for cairo_svg_surface_create in -lcairo... yes
checking cairo-svg.h usability... yes
checking cairo-svg.h presence... yes
checking for cairo-svg.h... yes
checking for cairo_pdf_surface_create in -lcairo... yes
checking cairo-pdf.h usability... yes
checking cairo-pdf.h presence... yes
checking for cairo-pdf.h... yes
checking for cairo_ps_surface_create in -lcairo... yes
checking cairo-ps.h usability... yes
checking cairo-ps.h presence... yes
checking for cairo-ps.h... yes
checking for pango_cairo_context_set_font_options in -lpango-1.0... no
checking for pkg-config... (cached) pkg-config
sh: gnome-config: not found
configure: WARNING:
----------------------------------------------------------------------------
* I found a copy of pkgconfig, but there is no pangocairo.pc file around.
  You may want to set the PKG_CONFIG_PATH variable to point to its
  location.
----------------------------------------------------------------------------




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From dam at opencsw.org  Thu Apr 16 14:39:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 14:39:34 +0200
Subject: [csw-maintainers] libcairo in testing
Message-ID: <94097AB5-019B-45C1-AF52-0A90A2C3CD56@opencsw.org>

Hi,

there is a libcairo 1.8.6 in testing/. As there are quite some
dependencies I would prefer someone else than me could test it.
The timely release is important, as there are a number of other
packages depending on it (including pango and rrdtool, which is
currently disfunctional in current/).


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 14:58:57 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 14:58:57 +0200
Subject: [csw-maintainers] build8x down
Message-ID: <DB296B4A-4D1C-4198-AEBC-D8D63505B2B0@opencsw.org>

Hi,

one of the mirrored disks of build8x died a few days ago.
On the attempt to replace the drive the Linux mirror driver
crashed the machine :-(

The machine will be available again shortly. You can use
the old build8x1 in the meantime.


Sorry for the inconvenience.

   -- Dago


From dam at opencsw.org  Thu Apr 16 15:28:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 15:28:15 +0200
Subject: [csw-maintainers] New subversion?
Message-ID: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>

Hi,

I remember there was work on an update of subversion which doesn't
seem to have been released nor is one in testing/. Is there an
obstacle in the way? There is need for a current one to sync up
with the SourceForge subversion.


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 15:37:16 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 15:37:16 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <49E70FA0.90700@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>	<49E6298E.2030801@opencsw.org>
	<73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>
	<49E70FA0.90700@opencsw.org>
Message-ID: <09C55997-A791-4810-8FD0-24FC08BCAAE0@opencsw.org>

Hi Ihsan,

Am 16.04.2009 um 12:59 schrieb Ihsan Dogan:
> Is it possible, that the new cairo packages are breaking pango?

Well, you need pangocairo. That is included in pango and requires
a current cairo which hasn't been released yet. After release I'll
happily rebuild pango with all bells and whistles. You may test
the packages in testing/ to further speed up the process ;-)


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 16:10:49 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 16:10:49 +0200
Subject: [csw-maintainers] build8x down
In-Reply-To: <DB296B4A-4D1C-4198-AEBC-D8D63505B2B0@opencsw.org>
References: <DB296B4A-4D1C-4198-AEBC-D8D63505B2B0@opencsw.org>
Message-ID: <835018D9-BA3A-49A1-9605-0685D0209BA0@opencsw.org>

Hi,

Am 16.04.2009 um 14:58 schrieb Dagobert Michelsen:
> one of the mirrored disks of build8x died a few days ago.
> On the attempt to replace the drive the Linux mirror driver
> crashed the machine :-(
>
> The machine will be available again shortly. You can use
> the old build8x1 in the meantime.

build8x is available again.


Best regards

   -- Dago


From harpchad at opencsw.org  Thu Apr 16 16:32:26 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 16 Apr 2009 09:32:26 -0500
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
Message-ID: <49E7417A.7060307@opencsw.org>

Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 16.04.2009 um 03:05 schrieb Chad Harp:
>> On Feb 24, 2009, at 2:05 PM, Chad Harp wrote:
>>>
>>> Dagobert Michelsen wrote:
>>>> Hi Chad,
>>>> I noticed there is an error in the current CSWftype2
>>>> 2.3.8,REV=2009.02.16 in
>>>>   /opt/csw/include/ft2build.h
>>>> with the line
>>>>   #include <freetype/config/ftheader.h>
>>>> The path is instead
>>>>   /opt/csw/include/freetype2/freetype/config/ftheader.h
>>>> I guess it should either be in
>>>>   /opt/csw/include/freetype2/ft2build.h
>>>> or contain
>>>>   #include <freetype2/freetype/config/ftheader.h>
>>>
>>>
>>> That looks correct to me, can you send me more details on where 
>>> you're getting the error?
>>>
>>> #include <freetype/config/ftheader.h>
>>> with -I/opt/csw/include/freetype2
>>> should find
>>> /opt/csw/include/freetype2/freetype/config/ftheader.h
>>>
>>> Perhaps what you're compiling isn't using freetype-config?
>>>
>>> /opt/csw/bin/freetype-config --cflags
>>> -I/opt/csw/include/freetype2 -I/opt/csw/include
>>
> 
>> I was going through my list of things to look into and ran across 
>> this.  I can't remember what the outcome was, did you get past this 
>> error?
> 
> And:
> 
> Am 20.02.2009 um 12:56 schrieb Ihsan Dogan:
>> Am 17.2.2009 19:03 Uhr, Dagobert Michelsen schrieb:
>>>
>>>> brauchst zum Kompilieren die aktuelle freetype2 Version aus /
>>>> testing.
>>>> Koenntest Du das bitte auf der buildfarm einspielen ?
>>>> Ziel ist es, freetype2 und fontconfig nach einer Testphase
>>>> gleichzeitig zu releasen.
>>>
>>> Ok, so freetype2 and fontconfig should be released at the
>>> same time and freetype2 must be installed on the buildfarm
>>> for building fontconfig.
>>>
>>> Ihsan, Phil: Should I do it this way violating standards or
>>> should these packages be released one after another?
>>
>> As I understand, we don't have many choices, so it's fine for me.
>> Important is, that a notify is sent to the maintainers list.
> 
> We have not build8st for this. AFAIK this has not been done yet.
> Want to give it a try?
> 
> 
> Best regards
> 
>   -- Dagp
Freetype was updated to 2.3.8 in February, I think it's already on the 
build farm. (2.3.9 was release in March but I haven't updated yet).

Alexander Maier was working on a new fontconfig, but I don't think it 
has been finished yet.

I guess we should get it sorted out before the Cairo and Pango builds 
get too far along, it be nice if they were built against fontconfig 2.4 
(or later).



From dam at opencsw.org  Thu Apr 16 16:38:19 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 16:38:19 +0200
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <49E7417A.7060307@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
	<49E7417A.7060307@opencsw.org>
Message-ID: <22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>

Hi Chad,

Am 16.04.2009 um 16:32 schrieb Chad Harp:
> Freetype was updated to 2.3.8 in February, I think it's already on  
> the build farm. (2.3.9 was release in March but I haven't updated  
> yet).
>
> Alexander Maier was working on a new fontconfig, but I don't think  
> it has been finished yet.

Alexander, how far did you come? Where is the problem?

> I guess we should get it sorted out before the Cairo and Pango  
> builds get too far along, it be nice if they were built against  
> fontconfig 2.4 (or later).

Yes. I take a look at fontconfig now, this needs to be out of the door.
Alexander, please join in :-)


Best regards

   -- Dago


From harpchad at opencsw.org  Thu Apr 16 17:06:56 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 16 Apr 2009 10:06:56 -0500
Subject: [csw-maintainers] gnutls 2.6.5 in testing
Message-ID: <49E74990.1040407@opencsw.org>

gnutls-2.6.5,REV=2009.04.15-SunOS5.8-i386-CSW.pkg.gz
gnutls-2.6.5,REV=2009.04.15-SunOS5.8-sparc-CSW.pkg.gz
gnutls_devel-2.6.5,REV=2009.04.15-SunOS5.8-i386-CSW.pkg.gz
gnutls_devel-2.6.5,REV=2009.04.15-SunOS5.8-sparc-CSW.pkg.gz


From harpchad at opencsw.org  Thu Apr 16 17:25:34 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 16 Apr 2009 10:25:34 -0500
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
	<49E7417A.7060307@opencsw.org>
	<22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>
Message-ID: <49E74DEE.5090003@opencsw.org>

I just put freetype 2.3.9 in testing.  If we get fontconfig 2.6.0 built 
against that we'll be at the latest stable for both.

Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 16.04.2009 um 16:32 schrieb Chad Harp:
>> Freetype was updated to 2.3.8 in February, I think it's already on the 
>> build farm. (2.3.9 was release in March but I haven't updated yet).
>>
>> Alexander Maier was working on a new fontconfig, but I don't think it 
>> has been finished yet.
> 
> Alexander, how far did you come? Where is the problem?
> 
>> I guess we should get it sorted out before the Cairo and Pango builds 
>> get too far along, it be nice if they were built against fontconfig 
>> 2.4 (or later).
> 
> Yes. I take a look at fontconfig now, this needs to be out of the door.
> Alexander, please join in :-)
> 
> 
> Best regards
> 
>   -- Dago



From dam at opencsw.org  Thu Apr 16 17:29:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 17:29:51 +0200
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <49E74DEE.5090003@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
	<49E7417A.7060307@opencsw.org>
	<22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>
	<49E74DEE.5090003@opencsw.org>
Message-ID: <B8B6F2F5-843E-44FC-81DA-CAC092EC7B44@opencsw.org>

Hi Chad,

Am 16.04.2009 um 17:25 schrieb Chad Harp:
> Dagobert Michelsen wrote:
>> Am 16.04.2009 um 16:32 schrieb Chad Harp:
>>> Freetype was updated to 2.3.8 in February, I think it's already on  
>>> the build farm. (2.3.9 was release in March but I haven't updated  
>>> yet).
>>>
>>> Alexander Maier was working on a new fontconfig, but I don't think  
>>> it has been finished yet.
>>
>> Alexander, how far did you come? Where is the problem?
>>
>>> I guess we should get it sorted out before the Cairo and Pango  
>>> builds get too far along, it be nice if they were built against  
>>> fontconfig 2.4 (or later).
>>
>> Yes. I take a look at fontconfig now, this needs to be out of the  
>> door.
>> Alexander, please join in :-)
>>
>
> I just put freetype 2.3.9 in testing.  If we get fontconfig 2.6.0  
> built against that we'll be at the latest stable for both.

As I just heard Alexander is on vacation and will be back on monday, so
hopefully we have a full set of base libs by mid of next week.


Best regards+

   -- Dago


From mwatters at opencsw.org  Thu Apr 16 17:57:20 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 10:57:20 -0500
Subject: [csw-maintainers] gcc 4.3.3 miracle in testing
Message-ID: <49E75560.9010208@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I have "new" gcc4.3.3 packages in testing

this "should" run on all solaris platforms solaris 8 on
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnVWAACgkQLrhmsXMSLxeNCgCcCdSg6qYEuGbpptTrrdDA+tNJ
ykcAoMdZZJkiET54OGBkwwuIV6pSaqHx
=ERyu
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 16 18:09:21 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 18:09:21 +0200
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
Message-ID: <49E75831.8000401@opencsw.org>

Am 16.4.2009 10:05 Uhr, Dagobert Michelsen schrieb:

> according to Bens great work the package name now contains the
> repository status if the package has not been fully checked in.
> 
> The usual name is
>   %{bitname}-%{SPKG_VERSION}%{SPKG_REVSTAMP}-%{SPKG_OSNAME}-%{arch}-CSW.pkg
> 
> Instead of the CSW there are now three other possible values:
>   UNCOMMITTED  Not every file has been checked in or ignored
>                ('svn status' not empty)
>   NOSVN        The svn binary could not be found
>   NOTVERSIONED The tree is not under version control
> 
> The change is in effect since r4338.

I've just made a commit on build8s and went then to build8x to build the
the x86 version. While the file on build8s was fine, build8x put
UNCOMMITTED into the filename.

ihsan at build8s:~/staging/build-16.Apr.2009$ ll
total 2429
-rw-r--r--   1 ihsan    csw       592155 Apr 16 18:04
nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-UNCOMMITTED.pkg.gz
-rw-r--r--   1 ihsan    csw       622693 Apr 16 18:00
nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz


Bug or feature?




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Thu Apr 16 18:18:03 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 11:18:03 -0500
Subject: [csw-maintainers] please install gcc4* from testing on build8xt and
	build10xt
Message-ID: <49E75A3B.1040206@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


I would like to test the compiler on both platforms.

gcc testing on sparc solaris 9 and solaris 10
I compiled zlib and vim and both tested clean


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnWjoACgkQLrhmsXMSLxfnpwCdEYlNElt5XMQu0JXecTFrnrPF
SQAAnjiaEtWkcFe0QrVdHHwAMS8XOydE
=wvdp
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Thu Apr 16 18:40:06 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Apr 2009 12:40:06 -0400
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <49E75831.8000401@opencsw.org>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
	<49E75831.8000401@opencsw.org>
Message-ID: <1239899850-sup-285@ntdws12.chass.utoronto.ca>

Excerpts from Ihsan Dogan's message of Thu Apr 16 12:09:21 -0400 2009:
> ihsan at build8s:~/staging/build-16.Apr.2009$ ll
> total 2429
> -rw-r--r--   1 ihsan    csw       592155 Apr 16 18:04
> nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-UNCOMMITTED.pkg.gz
> -rw-r--r--   1 ihsan    csw       622693 Apr 16 18:00
> nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz
> 
> 
> Bug or feature?

Depends.  Are there any files listed in your working directory that
show up in `svn status` as ?.  If so, those can produce the
UNCOMMITTED tag.  The best approach here is to either commit the files
or add them to the ignore list.

If there are no stray/uncommitted files and you're still getting that
status, it's a bug.

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090416/9915248e/attachment.asc>

From mwatters at opencsw.org  Thu Apr 16 19:43:27 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 12:43:27 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
Message-ID: <49E76E3F.8000604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> I remember there was work on an update of subversion which doesn't
> seem to have been released nor is one in testing/. Is there an
> obstacle in the way? There is need for a current one to sync up
> with the SourceForge subversion.
> 
> 
> Best regards
> 
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I am packaging it,  I thought I submitted it to Phil
I will go back and see what the status is/was.



- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnbj8ACgkQLrhmsXMSLxftvACfT3/OGSP1dwXt1yF+b3qzChYa
beUAnRe9SFgeLG3gDsdWvHllqEs/nyr5
=f4pS
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr 16 19:46:43 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 12:46:43 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <49E76E3F.8000604@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
	<49E76E3F.8000604@opencsw.org>
Message-ID: <49E76F03.1040802@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mike Watters wrote:
> Dagobert Michelsen wrote:
>> Hi,
> 
>> I remember there was work on an update of subversion which doesn't
>> seem to have been released nor is one in testing/. Is there an
>> obstacle in the way? There is need for a current one to sync up
>> with the SourceForge subversion.
> 
> 
>> Best regards
> 
>>   -- Dago
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> I am packaging it,  I thought I submitted it to Phil
> I will go back and see what the status is/was.
> 
> 
> 
I can't seem to find anything on it in my email.
I will rebuild it today and push it into testing.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnbwMACgkQLrhmsXMSLxdmFQCgzPiddU7PHqHr67ynr7QFwEU8
M5AAn1MBCd+I48ddg4XautWgdy5VFMVT
=Qbwh
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr 16 20:41:01 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 13:41:01 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <49E76F03.1040802@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>	<49E76E3F.8000604@opencsw.org>
	<49E76F03.1040802@opencsw.org>
Message-ID: <49E77BBD.1020204@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mike Watters wrote:
> Mike Watters wrote:
>> Dagobert Michelsen wrote:
>>> Hi,
>>> I remember there was work on an update of subversion which doesn't
>>> seem to have been released nor is one in testing/. Is there an
>>> obstacle in the way? There is need for a current one to sync up
>>> with the SourceForge subversion.
> 
>>> Best regards
>>>   -- Dago
>>> _______________________________________________
>>> maintainers mailing list
>>> maintainers at lists.opencsw.org
>>> https://lists.opencsw.org/mailman/listinfo/maintainers
>> I am packaging it,  I thought I submitted it to Phil
>> I will go back and see what the status is/was.
> 
> 
> 
> I can't seem to find anything on it in my email.
> I will rebuild it today and push it into testing.
> 
_______________________________________________
maintainers mailing list
maintainers at lists.opencsw.org
https://lists.opencsw.org/mailman/listinfo/maintainers

I remember what happened to svn.
It was in testing and I did a ISALIST check on the binaries.
The perl module had a bad ISALIST entry.

I removed it from testing with the intent on recompiling,
but got sidetracked with gcc.  I am rebuilding it right now.

new version 1.6.1


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknne70ACgkQLrhmsXMSLxdcEwCeKeFhxkF1JoOqtq/e9IJQmjfu
ueEAoIEX2cjacAImY0T8du0yjwtDTTj+
=fu73
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Apr 16 22:35:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:35:42 +0200
Subject: [csw-maintainers] OpenCSW & JDK redistribution
References: <31180549-9BB3-4763-BAC5-2D256AA7331C@baltic-online.de>
Message-ID: <8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>

Hi,

after lots of talks to people from Sun about JDK/JRE
redistribution here is a status where we are now:

Anfang der weitergeleiteten E-Mail:
> Von: Dalibor Topic <Dalibor.Topic at Sun.COM>
>
> Hi Dagobert,
>
> Thank you for your call on Tuesday and for taking some time to
> chat today. I'm CC:ing the current maintainer of OpenJDK 6,
> Joe Darcy on this e-mail.
>
> As discussed during the calls we had, I've had another look at
> the different JDK licensing options and discussed your inquiry
> with others inside Sun. After looking at the typical use cases
> for the different licenses, I don't think that any of the existing
> non-open-source JDK licensing options fits OpenCSW's use case
> of distributing easily redistributable stand-alone binary packages
> of the JDK.
>
> According to http://www.opencsw.org/about :
>
> " OpenCSW.org is a community effort to build and distribute binary
> packages (CSW, aka "Community SoftWare" packages) for
> production-grade Solaris releases."
> [snip]
> "We added "Open" to our organizational name name, to indicate that
> the packages will always be freely "open and available to use" by
> anyone. There will never be a cost to use them, or copy them.
> Additionally, it is our goal (although we arent there yet), that
> all packages will have publically open and easily reproducible
> build frameworks."
>
> The only release of the JDK that would really let OpenCSW meet
> that goal and really fits the description of "Open" in "OpenCSW"
> is OpenJDK 6.
>
> So I would strongly encourage the OpenCSW project to look into
> packaging OpenJDK 6 as part of OpenCSW, like Fedora, OpenSUSE,
> Ubuntu, Debian and many other community efforts with similar
> goals and approaches have successfully done so far. I would
> love to be able to list OpenCSW alongside other projects on
> http://openjdk.java.net/install/ !
>
> You can get the latest source code release, OpenJDK 6 b16, at
> http://download.java.net/openjdk/jdk6/ . A description of the
> latest changes is at
> http://blogs.sun.com/darcy/entry/openjdk_6_b16_source_bundle
>
> Unless you need the internal SNMP support, I would not bother
> with the binary plugs on the download page - the OpenJDK build
> should work fine without them, and the production-grade Linux
> distributions like Red Hat Enterprise Linux and CentOS aren't
> packaging them either in their OpenJDK 6 builds.
>
> OpenJDK 6 doesn't come with a plugin or webstart implementation
> (yet). If you'd like to package an independent plugin and
> Webstart implementation, they are available through the IcedTea
> project at http//icedtea.classpath.org .
>
> In order to ensure that OpenCSW users are getting an OpenJDK 6
> build that's compatible with JDK 6, the OpenCSW project can
> apply for access to the JCK under the OpenJDK Community TCK
> License Agreement, and test its own OpenJDK 6 builds against
> the JCK. Please see the OpenJDK Conformance Group web site for
> details at http://openjdk.java.net/groups/conformance/ .
>
> If you decide to package OpenJDK 6, and during your work
> come up with bug fixes, and other improvements, it would be
> nice if you would submit them back to the OpenJDK project using
> the OpenJDK Bugzilla instance at http://bugs.openjdk.java.net .
>
> Happy packaging, and all the best for OpenCSW! And do let us
> know how things go - I'm regularly on #openjdk on irc.oftc.net
> during German afternoons / evenings.
>
> cheers,
> dalibor topic

Additionally, he suggested to make a meta-package with manual
download like some Linux distributions. Any volunteers for
packaging OpenJDK?


Best regards

  -- Dago



From dam at opencsw.org  Thu Apr 16 22:40:50 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:40:50 +0200
Subject: [csw-maintainers] [csw-buildfarm] please install gcc4* from
	testing on build8xt and build10xt
In-Reply-To: <49E75A3B.1040206@opencsw.org>
References: <49E75A3B.1040206@opencsw.org>
Message-ID: <11C9C55E-A03E-40C2-95A1-B2BD8B71DF23@opencsw.org>

Hi Mike,

Am 16.04.2009 um 18:18 schrieb Mike Watters:
> Re: [csw-buildfarm] please install gcc4* from testing on build8xt  
> and build10xt
> I would like to test the compiler on both platforms.

The compiler etc. totals to 354 MB, wow!
build8xt is updating now (the old build8x, actually).
Unfortunately we don't have build10xt right now, build10x is
the global zone. We may move that into a local zone and
add build8xt as extra zone, but more disk space is needed
for this.

Jan: Is there extra space available, like 10 GB on the ESX farm?

> gcc testing on sparc solaris 9 and solaris 10
> I compiled zlib and vim and both tested clean

Cool.


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 22:46:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:46:01 +0200
Subject: [csw-maintainers] Machine renaming: build8x1 (old build8x) is now
	build8xt
Message-ID: <952B97E0-07AC-47B1-8C9D-910EBC6B5E31@opencsw.org>

Hi,

the old build8x (before the migration to the V65z) and former
build8x1 is now called build8xt for package testing. It is
currently used for gcc4 compilation tests. Just in case you
are wondering...


Best regards

   -- Dago


From phil at bolthole.com  Thu Apr 16 22:52:02 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 16 Apr 2009 13:52:02 -0700
Subject: [csw-maintainers] OpenCSW & JDK redistribution
In-Reply-To: <8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>
References: <31180549-9BB3-4763-BAC5-2D256AA7331C@baltic-online.de>
	<8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>
Message-ID: <20090416205202.GE74250@bolthole.com>

On Thu, Apr 16, 2009 at 10:35:42PM +0200, Dagobert Michelsen wrote:
> Hi,
>
> after lots of talks to people from Sun about JDK/JRE
> redistribution here is a status where we are now:

>...


Did you make it clear that we really dont care about having the source code
for the jdk itself; we just want to redistribute it, with full attributions
left to Sun?

I'd rather we redistribute have "FullJDK" than "OpenJDK", if you know what
I mean ;-)


From dam at opencsw.org  Thu Apr 16 22:57:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:57:42 +0200
Subject: [csw-maintainers] OpenCSW & JDK redistribution
In-Reply-To: <20090416205202.GE74250@bolthole.com>
References: <31180549-9BB3-4763-BAC5-2D256AA7331C@baltic-online.de>
	<8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>
	<20090416205202.GE74250@bolthole.com>
Message-ID: <A41FB4A0-9EA9-4198-A6AC-2B608C8F3487@opencsw.org>

Hi Phil,

Am 16.04.2009 um 22:52 schrieb Philip Brown:
> On Thu, Apr 16, 2009 at 10:35:42PM +0200, Dagobert Michelsen wrote:
>> after lots of talks to people from Sun about JDK/JRE
>> redistribution here is a status where we are now:
>
>> ...
>
>
> Did you make it clear that we really dont care about having the  
> source code
> for the jdk itself; we just want to redistribute it, with full  
> attributions
> left to Sun?

Yes, I made that clear. I talked for an hour to Dalibor about this.
There are lots of licenses tailored for different usecases, for
OEM-bundling with software like Veritas does it, for distributions
like SchilliX, but nothing for package archives. This mixup of
licenses is going to be ended with OpenJDK. I guess I am at an
end here, if you have contacts deeper inside Sun to make a
special license for our usecase please do so.

> I'd rather we redistribute have "FullJDK" than "OpenJDK", if you  
> know what
> I mean ;-)

I know what you mean. That's why I already made these beautiful
GAR descriptions for the production JDK/JRE in testing :-(


Best regards

   -- Dago


From ihsan at opencsw.org  Thu Apr 16 22:59:51 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 22:59:51 +0200
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <1239899850-sup-285@ntdws12.chass.utoronto.ca>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>	<49E75831.8000401@opencsw.org>
	<1239899850-sup-285@ntdws12.chass.utoronto.ca>
Message-ID: <49E79C47.1050907@opencsw.org>

Am 16.4.2009 18:40 Uhr, Ben Walton schrieb:

>> ihsan at build8s:~/staging/build-16.Apr.2009$ ll
>> total 2429
>> -rw-r--r--   1 ihsan    csw       592155 Apr 16 18:04
>> nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-UNCOMMITTED.pkg.gz
>> -rw-r--r--   1 ihsan    csw       622693 Apr 16 18:00
>> nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz
>>
>>
>> Bug or feature?
> 
> Depends.  Are there any files listed in your working directory that
> show up in `svn status` as ?.  If so, those can produce the
> UNCOMMITTED tag.  The best approach here is to either commit the files
> or add them to the ignore list.

No, there are not.

> If there are no stray/uncommitted files and you're still getting that
> status, it's a bug.

A few hours later, I've run "gmake package" again and the filename is
now correct. This is a little bit odd, because I haven't change anything.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From ihsan at opencsw.org  Thu Apr 16 23:05:07 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 23:05:07 +0200
Subject: [csw-maintainers] nsd 3.2.1 in testing
Message-ID: <49E79D83.9090305@opencsw.org>

Hello,

I've packaged NSD - Name Server Daemon. NSD is a BSD licensed DNS server
and a real alternative to the widely known Bind.

NSD has been developed by NLnet Labs in cooperation with Ripe. Security
and performance are one of the project goals.

http://mirror.opencsw.org/testing/nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Fri Apr 17 05:20:16 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 22:20:16 -0500
Subject: [csw-maintainers] pysqlite2 now in testing
Message-ID: <49E7F570.3010703@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	update to version 2.4.0
- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknn9XAACgkQLrhmsXMSLxdWEgCfU7WXf1l3XhFNDDiZwTWOu90v
xn4AoI4CFx0XFprkzDQOAgkc7p6z9GMP
=8Nq/
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Fri Apr 17 17:09:38 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 17 Apr 2009 10:09:38 -0500
Subject: [csw-maintainers] new gcc 4.3.3 for Intel in testing
Message-ID: <49E89BB2.7000500@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	fixed linking errors in packages

errors:

/opt/csw/gcc4/bin/cpp <linked pathname>
/opt/csw/gcc4/bin/gcc <linked pathname>
ERROR: attribute verification of </opt/csw/gcc4/bin/gcc> failed
    pathname does not exist
    unable to create link to
</opt/csw/gcc4/bin/i386/i386-pc-solaris2.8-gcc-4.3.3=/opt/csw/bin/isaexec>
/opt/csw/gcc4/bin/gccbug <linked pathname>
/opt/csw/gcc4/bin/gcov <linked pathname>
/opt/csw/gcc4/bin/i386/gcc <linked pathname>
ERROR: attribute verification of </opt/csw/gcc4/bin/i386/gcc> failed
    pathname does not exist
    unable to create link to </opt/csw/gcc4/bin/gcc>
/opt/csw/gcc4/bin/i386/i386-pc-solaris2.8-gcc <linked pathname>


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknom7EACgkQLrhmsXMSLxeEdQCeLA3HbXem01RsAYx3gBhRz7pW
RGoAn3OWz2v+NQ/jvzxSsDhQ0sPF7mDK
=5Nt1
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Fri Apr 17 18:39:35 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 17 Apr 2009 12:39:35 -0400
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <49E79C47.1050907@opencsw.org>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
	<49E75831.8000401@opencsw.org>
	<1239899850-sup-285@ntdws12.chass.utoronto.ca>
	<49E79C47.1050907@opencsw.org>
Message-ID: <1239986336-sup-3099@ntdws12.chass.utoronto.ca>

Excerpts from Ihsan Dogan's message of Thu Apr 16 16:59:51 -0400 2009:
> > If there are no stray/uncommitted files and you're still getting that
> > status, it's a bug.
> 
> A few hours later, I've run "gmake package" again and the filename is
> now correct. This is a little bit odd, because I haven't change anything.

Ok.  That is strange.  I'll have a look at it tonight to see if I can
replicated it.

Thanks
-Ben

-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090417/0f573d24/attachment.asc>

From mwatters at opencsw.org  Fri Apr 17 20:34:58 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 17 Apr 2009 13:34:58 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
Message-ID: <49E8CBD2.2040209@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> I remember there was work on an update of subversion which doesn't
> seem to have been released nor is one in testing/. Is there an
> obstacle in the way? There is need for a current one to sync up
> with the SourceForge subversion.
> 
> 
> Best regards
> 
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Sorry for the delay,  I am having issue getting 1.6.1 to compile.
I tried rebuilding 1.5.6 that I had already packaged, and
it seems something changed on the build servers and it won't compile either.
I look to determine what changed after I get 1.6.1 to compile.
I will keep working on it, but just wanted to give a status of where it is.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknoy9IACgkQLrhmsXMSLxfmvACfTYIiswPZP/t+4EnPOHZ8Gwrk
pgIAoMGpFWtL4JiroXRMnkE3ulVeiCAW
=uMgd
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Fri Apr 17 21:50:59 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 17 Apr 2009 14:50:59 -0500
Subject: [csw-maintainers] sudo 1.7.0 now in testing
Message-ID: <49E8DDA3.1080507@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	fixed some package inconsistencies.
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkno3aMACgkQLrhmsXMSLxcHCwCgz/2AMyHyc17TWPz07d9C1/xu
B18AoJvzbr2AycD6mZ3ROP4qCh0mXf0O
=xIgk
-----END PGP SIGNATURE-----


From william at wbonnet.net  Sat Apr 18 22:33:47 2009
From: william at wbonnet.net (William Bonnet)
Date: Sat, 18 Apr 2009 22:33:47 +0200
Subject: [csw-maintainers] /testing pixman 0.15.2
Message-ID: <49EA392B.7040003@wbonnet.net>

Hi

I have updated pixman package to version 0.15.2. This package now 
includes 64bits librairies

cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From maybird1776 at yahoo.com  Sun Apr 19 02:40:16 2009
From: maybird1776 at yahoo.com (ken mays)
Date: Sat, 18 Apr 2009 17:40:16 -0700 (PDT)
Subject: [csw-maintainers] GNOME: GTK+ 2.16.1 stable release
Message-ID: <46659.65772.qm@web111306.mail.gq1.yahoo.com>


This is the next package needed for building GNOME.
I'll submit a report.

~ Ken



      


From phil at bolthole.com  Sun Apr 19 02:47:16 2009
From: phil at bolthole.com (Philip Brown)
Date: Sat, 18 Apr 2009 17:47:16 -0700
Subject: [csw-maintainers] GNOME: GTK+ 2.16.1 stable release
In-Reply-To: <46659.65772.qm@web111306.mail.gq1.yahoo.com>
References: <46659.65772.qm@web111306.mail.gq1.yahoo.com>
Message-ID: <20090419004716.GA87849@bolthole.com>

On Sat, Apr 18, 2009 at 05:40:16PM -0700, ken mays wrote:
> 
> This is the next package needed for building GNOME.
> I'll submit a report.

not sure what you mean by "report"... but perhaps you might simply update
the wiki page.

wiki.opencsw.org/gnome


From william at wbonnet.net  Sun Apr 19 11:00:21 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 19 Apr 2009 11:00:21 +0200
Subject: [csw-maintainers] GAR and PKG_CONFIG_PATH
Message-ID: <49EAE825.9020700@wbonnet.net>

Hi

I am building X11 libs and i need to set PKG_CONFIG_PATH and i have a 
few problems... i am looking for the good way to define the 
PKG_CONFIG_PATH variable and add to its value /opt/csw/X11/lib/pkgconfig.

I tried either setting

EXTRA_PKGCONFIG_PATH  = /opt/csw/X11/lib/pkgconfig

or

PKG_CONFIG_PATH  = /opt/csw/X11/lib/pkgconfig

in Makefile or category file, but it is not taken in account by GAR 
scripts. The only work around i dound is to set it and export into the 
shell before calling gmake.

So please anyone to explain me what i did wrong ? :)

thanksin advance

cheers
W.

PS: mgar/pkg/x11/libXdmcp is the example package

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From trygvis at opencsw.org  Sun Apr 19 14:11:45 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sun, 19 Apr 2009 14:11:45 +0200
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
Message-ID: <49EB1501.5090006@opencsw.org>

Howdy,

I used some wikidot magic to automatically create the list of packages
on that is shown on [1]. It will now automatically list all the pages
tagged with "package". So if you write some documentation for a package,
please tag it with "package" and it'll show up automatically.

[1]: http://wiki.opencsw.org/packages

--
Trygve


From phil at bolthole.com  Sun Apr 19 17:01:34 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 19 Apr 2009 08:01:34 -0700
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <49EB1501.5090006@opencsw.org>
References: <49EB1501.5090006@opencsw.org>
Message-ID: <20090419150134.GA9243@bolthole.com>

On Sun, Apr 19, 2009 at 02:11:45PM +0200, Trygve Laugst?l wrote:
> Howdy,
> 
> I used some wikidot magic to automatically create the list of packages
> on that is shown on [1]. It will now automatically list all the pages
> tagged with "package". So if you write some documentation for a package,
> please tag it with "package" and it'll show up automatically.
> 
> [1]: http://wiki.opencsw.org/packages

nice trick!


Perhaps you could pick a different title for the page, and possible url,
though?

"packagedocs"  perhaps?



From trygvis at opencsw.org  Sun Apr 19 17:52:19 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sun, 19 Apr 2009 17:52:19 +0200
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <20090419150134.GA9243@bolthole.com>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
Message-ID: <49EB48B3.1080506@opencsw.org>

Philip Brown wrote:
> On Sun, Apr 19, 2009 at 02:11:45PM +0200, Trygve Laugst?l wrote:
>> Howdy,
>>
>> I used some wikidot magic to automatically create the list of packages
>> on that is shown on [1]. It will now automatically list all the pages
>> tagged with "package". So if you write some documentation for a package,
>> please tag it with "package" and it'll show up automatically.
>>
>> [1]: http://wiki.opencsw.org/packages
> 
> nice trick!
> 
> 
> Perhaps you could pick a different title for the page, and possible url,
> though?
> 
> "packagedocs"  perhaps?

I tried to think of a better name, but I couldn't really find a name I 
liked more so I just went with what already was there.

One thing I've been thinking about is creating documentation for users 
and (future) maintainers. If I where to take over a package I would 
appreciate some documentation on how to test the package.

How about creating two tags:
* user-doc: documentation for users (basically what those pages are today)
* maintainer-doc: documentation for maintainers, which would include 
tricks to build the package and how to test it.

--
Trygve


From dam at opencsw.org  Sun Apr 19 22:55:04 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Apr 2009 22:55:04 +0200
Subject: [csw-maintainers] GAR and PKG_CONFIG_PATH
In-Reply-To: <49EAE825.9020700@wbonnet.net>
References: <49EAE825.9020700@wbonnet.net>
Message-ID: <9DAF9057-B813-4CE8-926C-E36EE97A3D00@opencsw.org>

Hi William,

Am 19.04.2009 um 11:00 schrieb William Bonnet:
> I am building X11 libs and i need to set PKG_CONFIG_PATH and i have  
> a few problems... i am looking for the good way to define the  
> PKG_CONFIG_PATH variable and add to its value /opt/csw/X11/lib/ 
> pkgconfig.
>
> I tried either setting
>
> EXTRA_PKGCONFIG_PATH  = /opt/csw/X11/lib/pkgconfig
>
> or
>
> PKG_CONFIG_PATH  = /opt/csw/X11/lib/pkgconfig
>
> in Makefile or category file, but it is not taken in account by GAR  
> scripts.

This is basically ok, however I have some remarks.

You use this in x11/category.mk:

# pkg-config options
PKG_CONFIG_PATH  = /opt/csw/X11/lib/pkgconfig
PKG_CONFIG_PATH += $(DESTDIR)/opt/csw/lib/pkgconfig
PKG_CONFIG_PATH += $(DESTDIR)/opt/csw/X11/lib/pkgconfig

Please avoid DESTDIR as much as possible as it circumvents the standard
workflow of
   build base -> release -> install -> build dependent

There should be something to add category-specific stuff to pkgconfig:
   _CATEGORY_PKG_CONFIG_PATH = $(abspath $(prefix)/X11/lib/$ 
(MM_LIBDIR)/pkgconfig)

I committed the appropriate changes in r4413 and changed x11/category.mk
accordingly.

You can always check the value of PKG_CONFIG_PATH by typing

> build8s% gmake modenv
>           Arch: sparc
>         Kernel: sparcv9
>
> Default ISA 32: sparcv8
> Default ISA 64: sparcv9
>
> Requested ISAs: sparcv8 sparcv9 i386 amd64
>    Needed ISAs: sparcv8 sparcv9
>     Build ISAs: sparcv8 sparcv9
>
>   ISAEXEC dirs: /opt/csw/bin /opt/csw/sbin /opt/csw/libexec
>  ISAEXEC files:
>
>  Merge include:
>  Merge exclude:     /opt/csw/share/info/dir /opt/csw/lib/.*\.la .* 
> \~ /opt/csw/lib/.*\.a
>
>     Modulators: ISA
>    Modulations: isa-sparcv8 isa-sparcv9
>
> Requested compiler flags:
>
> * Modulation isa-sparcv8: ISA=sparcv8
>            PATH = /home/dam/mgar/pkg/x11/libXdmcp/trunk/work/install- 
> isa-sparcv8/opt/csw/bin:/home/dam/mgar/pkg/x11/libXdmcp/trunk/work/ 
> install-isa-sparcv8/opt/csw/bin:/home/dam/mgar/pkg/x11/libXdmcp/ 
> trunk/work/install-isa-sparcv8/opt/csw/sbin:/home/dam/mgar/pkg/x11/ 
> libXdmcp/trunk/work/install-isa-sparcv8/opt/csw/sbin:/opt/csw/bin:/ 
> opt/csw/bin:/opt/csw/sbin:/opt/csw/sbin:/opt/studio/SOS11/SUNWspro/ 
> bin:/home/dam/mgar/pkg/x11/libXdmcp/trunk/gar/bin:/usr/bin:/usr/ 
> sbin:/usr/java/bin:/usr/ccs/bin:/usr/openwin/bin
> PKG_CONFIG_PATH = /opt/csw/lib/pkgconfig:/opt/csw/lib/X11/pkgconfig
>          CFLAGS = -xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION - 
> xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION -xlibmil - 
> errtags=yes -erroff=E_EMPTY_DECLARATION
>        CXXFLAGS = -xlibmil -xlibmopt -features=tmplife -norunpath - 
> xlibmil -xlibmopt -features=tmplife -norunpath -xlibmil -xlibmopt - 
> features=tmplife -norunpath
>        CPPFLAGS =
>         LDFLAGS = -L/opt/csw/lib -R/opt/csw/lib -L/opt/csw/lib -R/ 
> opt/csw/lib -L/opt/csw/lib -R/opt/csw/lib
>         ASFLAGS =
>        OPTFLAGS = -xO3 -xarch=v8
>
> * Modulation isa-sparcv9: ISA=sparcv9
>            PATH = /home/dam/mgar/pkg/x11/libXdmcp/trunk/work/install- 
> isa-sparcv9/opt/csw/bin/sparcv9:/home/dam/mgar/pkg/x11/libXdmcp/ 
> trunk/work/install-isa-sparcv9/opt/csw/bin:/home/dam/mgar/pkg/x11/ 
> libXdmcp/trunk/work/install-isa-sparcv9/opt/csw/sbin/sparcv9:/home/ 
> dam/mgar/pkg/x11/libXdmcp/trunk/work/install-isa-sparcv9/opt/csw/ 
> sbin:/opt/csw/bin/sparcv9:/opt/csw/bin:/opt/csw/sbin/sparcv9:/opt/ 
> csw/sbin:/opt/studio/SOS11/SUNWspro/bin:/home/dam/mgar/pkg/x11/ 
> libXdmcp/trunk/gar/bin:/usr/bin:/usr/sbin:/usr/java/bin:/usr/ccs/ 
> bin:/usr/openwin/bin
> PKG_CONFIG_PATH = /opt/csw/lib/64/pkgconfig:/opt/csw/lib/64/X11/ 
> pkgconfig
>          CFLAGS = -xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION - 
> xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION -xlibmil - 
> errtags=yes -erroff=E_EMPTY_DECLARATION
>        CXXFLAGS = -xlibmil -xlibmopt -features=tmplife -norunpath - 
> xlibmil -xlibmopt -features=tmplife -norunpath -xlibmil -xlibmopt - 
> features=tmplife -norunpath
>        CPPFLAGS =
>         LDFLAGS = -L/opt/csw/lib -R/opt/csw/lib -L/opt/csw/lib -R/ 
> opt/csw/lib -L/opt/csw/lib -R/opt/csw/lib
>         ASFLAGS =
>        OPTFLAGS = -xO3 -xarch=v9
>

> The only work around i dound is to set it and export into the shell  
> before calling gmake.


Well, it is only passed during configure-time:
   <http://apps.sourceforge.net/trac/gar/browser/csw/mgar/gar/v2/gar.conf.mk#L567 
 >
If you need it during build-time too you can set
   EXTRA_BUILD_EXPORTS = PKG_CONFIG_PATH
but it usually shouldn't be necessary.

> So please anyone to explain me what i did wrong ? :)

The code looks like this:
      1     comand # This is for foo-config chaos
   2471 dmichelsen PKG_CONFIG_DIRS ?= $(libdir_install) $ 
(EXTRA_PKG_CONFIG_DIRS)
   2471 dmichelsen PKG_CONFIG_PATH ?= $(call MAKEPATH,$(foreach D,$ 
(PKG_CONFIG_DIRS),$(abspath $D/$(MM_LIBDIR)/pkgconfig)) $(EXTRA_PKGCON
FIG_PATH))

The mistyping of PKGCONFIG is a bug, as pkg-config itself uses  
PKG_CONFIG, this should be used.
This too is fixed in r4413 too (now EXTRA_PKG_CONFIG_PATH, it hadn't  
been used anyway).

Additionally, I removed in r2471 in gar.conf.mk (line 547)
   http://apps.sourceforge.net/trac/gar/changeset/2471
the export as I was not aware that it was really needed in a forked  
process.
This is fixed in r4415.

BTW, there is a much easier way to use dynamic licenses. Just use  
dynamic gspecs
and everything will be done automatically as the license has the  
default name.
I converted this package as example for you and the package builds  
fine now,
however, only 32 bit. For 64 bit the category.mk must have all the  
MM_*-stuff
from gar.conf.mk that differentiates pathes for isa-specific stuff  
during
install and merge. I'll take a look at that tomorrow.


I apologize for this accumulation of inaccuracies and confusion.


Best regards

   -- Dago


From phil at bolthole.com  Mon Apr 20 04:52:38 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 19 Apr 2009 19:52:38 -0700
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <49EB48B3.1080506@opencsw.org>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
	<49EB48B3.1080506@opencsw.org>
Message-ID: <20090420025238.GA87575@bolthole.com>

On Sun, Apr 19, 2009 at 05:52:19PM +0200, Trygve Laugst??l wrote:
> Philip Brown wrote:
>> Perhaps you could pick a different title for the page, and possible url,
>> though?
>>
>> "packagedocs"  perhaps?
>
> I tried to think of a better name, but I couldn't really find a name I  
> liked more so I just went with what already was there.

but what was there, does not suit the purpose :-/
The usage conflicts with the core site's 
http://www.opencsw.org/packages

To my mind, I think that for
http://XXXX.opencsw.org/NAMEHERE
that "NAMEHERE", should ideally be somewhat unique across the organization,
reguardless of which specific sub-site it is currently hosted it.


> One thing I've been thinking about is creating documentation for users  
> and (future) maintainers. If I where to take over a package I would  
> appreciate some documentation on how to test the package.
>
> How about creating two tags:
> * user-doc: documentation for users (basically what those pages are today)
> * maintainer-doc: documentation for maintainers, which would include  
> tricks to build the package and how to test it.

makes sense. ideally, they would somehow automatically have a button or
link to toggle between userlevel and maintainer level docs.

(not the same page; just that each page links to its counterpart)



From dam at opencsw.org  Mon Apr 20 10:47:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 20 Apr 2009 10:47:46 +0200
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <20090420025238.GA87575@bolthole.com>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
	<49EB48B3.1080506@opencsw.org>
	<20090420025238.GA87575@bolthole.com>
Message-ID: <88906185-3CC2-4BC8-9064-5A4C9EFD6BE3@opencsw.org>

Hi,

Am 20.04.2009 um 04:52 schrieb Philip Brown:
> On Sun, Apr 19, 2009 at 05:52:19PM +0200, Trygve Laugst??l wrote:
>> Philip Brown wrote:
>>> Perhaps you could pick a different title for the page, and  
>>> possible url,
>>> though?
>>>
>>> "packagedocs"  perhaps?
>>
>> I tried to think of a better name, but I couldn't really find a  
>> name I
>> liked more so I just went with what already was there.
>
> but what was there, does not suit the purpose :-/
> The usage conflicts with the core site's
> http://www.opencsw.org/packages
>
> To my mind, I think that for
> http://XXXX.opencsw.org/NAMEHERE
> that "NAMEHERE", should ideally be somewhat unique across the  
> organization,
> reguardless of which specific sub-site it is currently hosted it.

Yes. Ideally, opencsw.org/packages and the wikidot package page  
information
should be on one package, with automated content from OpenCSW and wiki
content. Maybe on the new website? BTW, is someone currently working on
it? I remember we had some helping hands offers :-)

>> One thing I've been thinking about is creating documentation for  
>> users
>> and (future) maintainers. If I where to take over a package I would
>> appreciate some documentation on how to test the package.
>>
>> How about creating two tags:
>> * user-doc: documentation for users (basically what those pages are  
>> today)
>> * maintainer-doc: documentation for maintainers, which would include
>> tricks to build the package and how to test it.
>
> makes sense. ideally, they would somehow automatically have a button  
> or
> link to toggle between userlevel and maintainer level docs.
>
> (not the same page; just that each page links to its counterpart)

The maintainer doc should be in the GAR Wiki. It was just moved to
Trac and a tight integration of code and wiki-content is possible
now.


Best regards

   -- Dago

From rupert at opencsw.org  Mon Apr 20 15:29:56 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 20 Apr 2009 15:29:56 +0200
Subject: [csw-maintainers] oracle buys sun ...
Message-ID: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>

wow ... that was unexpected: http://www.google.com/finance?q=NASDAQ%3AJAVA.

rupert.


From maciej at opencsw.org  Mon Apr 20 15:36:42 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 20 Apr 2009 14:36:42 +0100
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
Message-ID: <ce97543b0904200636q692d1955o8e29e77a545d17a6@mail.gmail.com>

On Mon, Apr 20, 2009 at 2:29 PM, rupert THURNER <rupert at opencsw.org> wrote:
> wow ... that was unexpected: http://www.google.com/finance?q=NASDAQ%3AJAVA.

Oracle offered $0.10 more per share. They now own MySQL.


From phil at bolthole.com  Mon Apr 20 15:58:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 20 Apr 2009 06:58:59 -0700
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
Message-ID: <20090420135859.GA81648@bolthole.com>

On Mon, Apr 20, 2009 at 03:29:56PM +0200, rupert THURNER wrote:
> wow ... that was unexpected.

not really. our local unix users' group speculated oracle was the next one
in line, as soon as the IBM deal dropped out :-)

and it's a much better fit than IBM.


on the darker side, oracle now "owns" both berkeleydb and mysql.

sigh.
 


From william at wbonnet.net  Mon Apr 20 16:26:59 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 20 Apr 2009 16:26:59 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <20090420135859.GA81648@bolthole.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
	<20090420135859.GA81648@bolthole.com>
Message-ID: <49EC8633.1070902@wbonnet.net>

Hi
> not really. our local unix users' group speculated oracle was the next one
> in line, as soon as the IBM deal dropped out :-)
>   
I was betting buzz around IBM buying Sun was just a diversionary movement.
> and it's a much better fit than IBM.
>
>
> on the darker side, oracle now "owns" both berkeleydb and mysql.
>   
And glassfish... they baught BEA (thus Weblogic) last year.


From phil at bolthole.com  Mon Apr 20 18:18:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 20 Apr 2009 09:18:26 -0700
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <88906185-3CC2-4BC8-9064-5A4C9EFD6BE3@opencsw.org>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
	<49EB48B3.1080506@opencsw.org>
	<20090420025238.GA87575@bolthole.com>
	<88906185-3CC2-4BC8-9064-5A4C9EFD6BE3@opencsw.org>
Message-ID: <20090420161826.GC72003@bolthole.com>

On Mon, Apr 20, 2009 at 10:47:46AM +0200, Dagobert Michelsen wrote:
> Yes. Ideally, opencsw.org/packages and the wikidot package page  
> information
> should be on one package, with automated content from OpenCSW and wiki
> content. Maybe on the new website? BTW, is someone currently working on
> it? I remember we had some helping hands offers :-)

Errr... not sure we are saying the same thing here :-}

but it would be nice for

http://www.opencsw.org/packages

to be able to provide links for

"click here for user level docs"
"click here for maintainer level docs"



From ihsan at opencsw.org  Tue Apr 21 08:51:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Tue, 21 Apr 2009 08:51:02 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <20090420135859.GA81648@bolthole.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
	<20090420135859.GA81648@bolthole.com>
Message-ID: <49ED6CD6.8070100@opencsw.org>

Am 20.4.2009 15:58 Uhr, Philip Brown schrieb:

>> wow ... that was unexpected.
> 
> not really. our local unix users' group speculated oracle was the next one
> in line, as soon as the IBM deal dropped out :-)

I was expecting, that Fujitsu would buy Sun.

> and it's a much better fit than IBM.

Much better.

> on the darker side, oracle now "owns" both berkeleydb and mysql.

Sun screwed up MySQL already. A few MySQL people founded already a new
company to go on with the MySQL development, so I think, it really does
not matter what Oracle is going to do with MySQL.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Tue Apr 21 14:10:41 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Tue, 21 Apr 2009 14:10:41 +0200
Subject: [csw-maintainers] /testing clamav 0.95.1
Message-ID: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com>

Please help testing ClamAV 0.95.1, especially the milter part that I
don't use myself.

http://mirror.opencsw.org/testing.html

http://mirror.opencsw.org/testing/clamav-0.95.1,REV=2009.04.21-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/clamav-0.95.1,REV=2009.04.21-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/libclamav-0.95.1,REV=2009.04.21-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libclamav-0.95.1,REV=2009.04.21-SunOS5.8-sparc-CSW.pkg.gz

-- 
/peter


From rupert at opencsw.org  Tue Apr 21 18:11:00 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 21 Apr 2009 18:11:00 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <49ED6CD6.8070100@opencsw.org>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
	<20090420135859.GA81648@bolthole.com> <49ED6CD6.8070100@opencsw.org>
Message-ID: <6af4270904210911q5cbad510w8a9d0a3be887f10a@mail.gmail.com>

On Tue, Apr 21, 2009 at 08:51, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Am 20.4.2009 15:58 Uhr, Philip Brown schrieb:
>
>>> wow ... that was unexpected.
>>
>> not really. our local unix users' group speculated oracle was the next one
>> in line, as soon as the IBM deal dropped out :-)
>
> I was expecting, that Fujitsu would buy Sun.
>
>> and it's a much better fit than IBM.
>
> Much better.
>
>> on the darker side, oracle now "owns" both berkeleydb and mysql.
>
> Sun screwed up MySQL already. A few MySQL people founded already a new
> company to go on with the MySQL development, so I think, it really does
> not matter what Oracle is going to do with MySQL.
>
you mean http://askmonty.org/wiki/index.php/MariaDB, foundet by the
founder of mysql, michael widenius?


From ihsan at opencsw.org  Wed Apr 22 09:51:52 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 22 Apr 2009 09:51:52 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <6af4270904210911q5cbad510w8a9d0a3be887f10a@mail.gmail.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>	<20090420135859.GA81648@bolthole.com>
	<49ED6CD6.8070100@opencsw.org>
	<6af4270904210911q5cbad510w8a9d0a3be887f10a@mail.gmail.com>
Message-ID: <49EECC98.5090407@opencsw.org>

Am 21.4.2009 18:11 Uhr, rupert THURNER schrieb:

>>>> wow ... that was unexpected.
>>> not really. our local unix users' group speculated oracle was the next one
>>> in line, as soon as the IBM deal dropped out :-)
>> I was expecting, that Fujitsu would buy Sun.
>>
>>> and it's a much better fit than IBM.
>> Much better.
>>
>>> on the darker side, oracle now "owns" both berkeleydb and mysql.
>> Sun screwed up MySQL already. A few MySQL people founded already a new
>> company to go on with the MySQL development, so I think, it really does
>> not matter what Oracle is going to do with MySQL.
>>
> you mean http://askmonty.org/wiki/index.php/MariaDB, foundet by the
> founder of mysql, michael widenius?

Yes. I just couldn't remember the name anymore.


Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Wed Apr 22 15:19:04 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 22 Apr 2009 15:19:04 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
Message-ID: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>

I have addressed bugs 3632, 3633 and 3634 in the new 1.6 version of
cswclassutils. All bugs refer to the cswinitsmf class action script
and were reported by Yann.

0003632: Service should be disabled synchronously
http://www.opencsw.org/mantis/view.php?id=3632

0003633: Service using init scripts should not be configured to at
boot time when autoenable_daemons=no
http://www.opencsw.org/mantis/view.php?id=3633

0003634: Please add smf service state persistence across upgrade
http://www.opencsw.org/mantis/view.php?id=3634

I will of course try the changes myself but it would be nice if more
people could test it before release.

http://mirror.opencsw.org/testing.html
http://mirror.opencsw.org/testing/cswclassutils-1.6,REV=2009.04.21-SunOS5.8-all-CSW.pkg.gz

I have more bugs reported but they are more of feature requests so
they will have to hold off a while longer. One is a request for a new
CAS for creating texinfo files, I will not do that one myself but will
happily include it into the package if someone contributes.

http://www.opencsw.org/mantis/view.php?id=3485

-- 
/peter


From ihsan at opencsw.org  Wed Apr 22 16:33:04 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 22 Apr 2009 16:33:04 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
Message-ID: <49EF2AA0.3080402@opencsw.org>

Hello Peter,

Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:

> 0003634: Please add smf service state persistence across upgrade
> http://www.opencsw.org/mantis/view.php?id=3634

I've tested only that one and it works fine, but the output might be
confusing:

Installing class <cswinitsmf> ...
Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
Creating manifest ...
Configuring service in SMF ...
CSWclamav is using Service Management Facility. The FMRI is
svc:/network/cswclamav-milter:default
Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
Creating manifest ...
Configuring service in SMF ...
CSWclamav is using Service Management Facility. The FMRI is
svc:/network/cswclamd:default
Starting CSWclamav ...
[ verifying class <cswinitsmf> ]

Clamd was not started, even it said that it was started.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From cyril.plisko at mountall.com  Wed Apr 22 16:39:57 2009
From: cyril.plisko at mountall.com (Cyril Plisko)
Date: Wed, 22 Apr 2009 17:39:57 +0300
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49EF2AA0.3080402@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>
Message-ID: <c7dddeaa0904220739g2348a2a4rfabc30103a27f7d2@mail.gmail.com>

On 4/22/09, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Hello Peter,
>
> Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:
>
>> 0003634: Please add smf service state persistence across upgrade
>> http://www.opencsw.org/mantis/view.php?id=3634
>
> I've tested only that one and it works fine, but the output might be
> confusing:
>
> Installing class <cswinitsmf> ...
> Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamav-milter:default
> Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamd:default
> Starting CSWclamav ...
> [ verifying class <cswinitsmf> ]
>
> Clamd was not started, even it said that it was started.
>
>
>
>
> Ihsan
>
> --
> ihsan at dogan.ch		http://blog.dogan.ch/
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


-- 
Regards,
        Cyril


From bonivart at opencsw.org  Wed Apr 22 16:56:42 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 22 Apr 2009 16:56:42 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49EF2AA0.3080402@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com> 
	<49EF2AA0.3080402@opencsw.org>
Message-ID: <625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>

On Wed, Apr 22, 2009 at 4:33 PM, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Hello Peter,
>
> Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:
>
>> 0003634: Please add smf service state persistence across upgrade
>> http://www.opencsw.org/mantis/view.php?id=3634
>
> I've tested only that one and it works fine, but the output might be
> confusing:
>
> Installing class <cswinitsmf> ...
> Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamav-milter:default
> Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamd:default
> Starting CSWclamav ...
> [ verifying class <cswinitsmf> ]
>
> Clamd was not started, even it said that it was started.

I think this is related to a bug when multiple services are set up
like above. The placement of the start block in the script is wrong.

I will look into it. Thanks for the report.

-- 
/peter


From cyril.plisko at mountall.com  Wed Apr 22 17:05:07 2009
From: cyril.plisko at mountall.com (Cyril Plisko)
Date: Wed, 22 Apr 2009 18:05:07 +0300
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49EF2AA0.3080402@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>
Message-ID: <c7dddeaa0904220805p61efec93w1dd61e9ba63fa5aa@mail.gmail.com>

On 4/22/09, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Hello Peter,
>
> Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:
>
>> 0003634: Please add smf service state persistence across upgrade
>> http://www.opencsw.org/mantis/view.php?id=3634
>
> I've tested only that one and it works fine, but the output might be
> confusing:
>
> Installing class <cswinitsmf> ...
> Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamav-milter:default
> Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamd:default
> Starting CSWclamav ...
> [ verifying class <cswinitsmf> ]
>
> Clamd was not started, even it said that it was started.
>
>
>
>
> Ihsan
>
> --
> ihsan at dogan.ch		http://blog.dogan.ch/
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


-- 
Regards,
        Cyril


From ihsan at opencsw.org  Wed Apr 22 18:56:32 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 22 Apr 2009 18:56:32 +0200
Subject: [csw-maintainers] /testing clamav 0.95.1
In-Reply-To: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com>
References: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com>
Message-ID: <49EF4C40.9010308@opencsw.org>

Hello Peter,

Am 21.4.2009 14:10 Uhr, Peter Bonivart schrieb:

> Please help testing ClamAV 0.95.1, especially the milter part that I
> don't use myself.

Works fine, but I haven't tested the Milter part.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Wed Apr 22 19:51:49 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 22 Apr 2009 19:51:49 +0200
Subject: [csw-maintainers] /testing clamav 0.95.1
In-Reply-To: <49EF4C40.9010308@opencsw.org>
References: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com> 
	<49EF4C40.9010308@opencsw.org>
Message-ID: <625385e30904221051t3cd465e7g33b744800d308b40@mail.gmail.com>

On Wed, Apr 22, 2009 at 6:56 PM, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Works fine, but I haven't tested the Milter part.

Ok, I will release it later this week if no one reports any problems
with it. It's just a respin with new source.

-- 
/peter


From Darin.Perusich at cognigencorp.com  Wed Apr 22 23:31:37 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 22 Apr 2009 17:31:37 -0400
Subject: [csw-maintainers] off topic question about suntar
Message-ID: <49EF8CB9.3040301@cognigencorp.com>

I have an off topic question I'm hoping one of you Solaris hackers can
help me answer. This might be right up Joerg alley since it's star guru ;-).

I'm using suntar on Solaris 10 to create archives and I've run into an
problem where when LANG is not set or set to a non-UTF-8 value like C,
POSIX, etc, then 'suntar' will throw the message "tar: invalid character
in UTF-8 conversion of 'luke?s.arf.gz'. There are a bunch of files that
have funky names, I don't know who or why, and I get the same error for
each one and they're not included in the archive.

How does suntar know whether or not the file name is UTF-8 and why
doesn't it even matter?

Thanks!

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com



From Joerg.Schilling at fokus.fraunhofer.de  Thu Apr 23 12:18:21 2009
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Thu, 23 Apr 2009 12:18:21 +0200
Subject: [csw-maintainers] off topic question about suntar
In-Reply-To: <49EF8CB9.3040301@cognigencorp.com>
References: <49EF8CB9.3040301@cognigencorp.com>
Message-ID: <49f0406d.7hLAW/yQwf7c3vnm%Joerg.Schilling@fokus.fraunhofer.de>

Darin Perusich <Darin.Perusich at cognigencorp.com> wrote:

> I have an off topic question I'm hoping one of you Solaris hackers can
> help me answer. This might be right up Joerg alley since it's star guru ;-).
>
> I'm using suntar on Solaris 10 to create archives and I've run into an
> problem where when LANG is not set or set to a non-UTF-8 value like C,
> POSIX, etc, then 'suntar' will throw the message "tar: invalid character
> in UTF-8 conversion of 'luke???s.arf.gz'. There are a bunch of files that
> have funky names, I don't know who or why, and I get the same error for
> each one and they're not included in the archive.
>
> How does suntar know whether or not the file name is UTF-8 and why
> doesn't it even matter?

Did you use the -E option to suntar?

In this case and only in this case, the file names will be converted to UTF-8.

Note that we did add the permission to also have untranslated filenames to the 
POSIX standard around 2004 in order to allow tar to be used as backup format.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily


From Darin.Perusich at cognigencorp.com  Thu Apr 23 15:05:02 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 23 Apr 2009 09:05:02 -0400
Subject: [csw-maintainers] off topic question about suntar
In-Reply-To: <49f0406d.7hLAW/yQwf7c3vnm%Joerg.Schilling@fokus.fraunhofer.de>
References: <49EF8CB9.3040301@cognigencorp.com>
	<49f0406d.7hLAW/yQwf7c3vnm%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <49F0677E.2050603@cognigencorp.com>


Joerg Schilling wrote:
> 
> Did you use the -E option to suntar?

Yes -E is being used, here is the command line. I'm forcing it to error
by setting LANG=C.

LANG=C /usr/bin/pfexec /usr/sbin/tar -cpE at bf 256 - -C
/snapshots/client_data . > file.tar
tar: invalid character in UTF-8 conversion of './luke?s.arf.gz'
tar: file (./luke?s.arf.gz): UTF-8 conversion failed.

> In this case and only in this case, the file names will be converted to UTF-8.
> 
> Note that we did add the permission to also have untranslated filenames to the 
> POSIX standard around 2004 in order to allow tar to be used as backup format.

This appears to be limited to Solaris 10 x86. I just tried this on SPARC
system and didn't get any errors. I'll have to submit a bug report.

Thanks

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com



From mwatters at opencsw.org  Thu Apr 23 16:29:46 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 09:29:46 -0500
Subject: [csw-maintainers] php5 now in testing
Message-ID: <49F07B5A.5080608@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	Repackage due to gar bug with dynamic admin files.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknwe1kACgkQLrhmsXMSLxeKjQCgh484pRIo1jRra/iAgyjsz0Xw
HGcAoLkCAA12RnJcUKQ+pepdVeUbZnku
=V3Vl
-----END PGP SIGNATURE-----


From bonivart at opencsw.org  Thu Apr 23 18:53:35 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 23 Apr 2009 18:53:35 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com> 
	<49EF2AA0.3080402@opencsw.org>
	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
Message-ID: <625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>

On Wed, Apr 22, 2009 at 4:56 PM, Peter Bonivart <bonivart at opencsw.org> wrote:
>> Clamd was not started, even it said that it was started.
>
> I think this is related to a bug when multiple services are set up
> like above. The placement of the start block in the script is wrong.
>
> I will look into it. Thanks for the report.

There's a new version in testing to address this problem, please give
it a try if you have the time. It should start all services a package
contains and give better info about what it's doing. It should also
properly stop services on non-SMF systems.

http://mirror.opencsw.org/testing/cswclassutils-1.8,REV=2009.04.23-SunOS5.8-all-CSW.pkg.gz

-- 
/peter


From mwatters at opencsw.org  Thu Apr 23 20:06:40 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 13:06:40 -0500
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0AD05.1070909@nature.berkeley.edu>
References: <49F07B5A.5080608@opencsw.org>
	<49F0AD05.1070909@nature.berkeley.edu>
Message-ID: <49F0AE30.5070406@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Gary Casterline wrote:
> Mike Watters wrote:
>> Changes:
>> 	Repackage due to gar bug with dynamic admin files.
>>
> 
> I don't see mod_php5 or ap2_modphp5 -- will those come later?
> 
>  _Gary
> 
> 
> _______________________________________________
> users mailing list
> users at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/users
> 
> _______________________________________________
> users mailing list
> users at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/users

Gary,	
	I don't maintain those  ;)

Ihsan: do you have any plans on pushing out a new version of those?

I tested running php5 with the existing ap2_modphp5 ... with no issues.
I don't run 1.3x anymore I didn't test against mod_php5

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknwrjAACgkQLrhmsXMSLxdu3ACgk+U7zifhcDr2Qcu5EeqrVlU/
FDUAn0Fv6j5zyrfzXTPt7VBRcsIWYUOD
=Dsz7
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 23 22:35:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 23 Apr 2009 22:35:02 +0200
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0AE30.5070406@opencsw.org>
References: <49F07B5A.5080608@opencsw.org>	<49F0AD05.1070909@nature.berkeley.edu>
	<49F0AE30.5070406@opencsw.org>
Message-ID: <49F0D0F6.6050003@opencsw.org>

Am 23.4.2009 20:06 Uhr, Mike Watters schrieb:

> Gary,	
> 	I don't maintain those  ;)
> 
> Ihsan: do you have any plans on pushing out a new version of those?
> 
> I tested running php5 with the existing ap2_modphp5 ... with no issues.
> I don't run 1.3x anymore I didn't test against mod_php5

I thought, you took these as well.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Thu Apr 23 22:36:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 15:36:49 -0500
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0D0F6.6050003@opencsw.org>
References: <49F07B5A.5080608@opencsw.org>	<49F0AD05.1070909@nature.berkeley.edu>	<49F0AE30.5070406@opencsw.org>
	<49F0D0F6.6050003@opencsw.org>
Message-ID: <49F0D161.5090507@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ihsan Dogan wrote:
> Am 23.4.2009 20:06 Uhr, Mike Watters schrieb:
> 
>> Gary,	
>> 	I don't maintain those  ;)
>>
>> Ihsan: do you have any plans on pushing out a new version of those?
>>
>> I tested running php5 with the existing ap2_modphp5 ... with no issues.
>> I don't run 1.3x anymore I didn't test against mod_php5
> 
> I thought, you took these as well.
> 
> 
> 
> Ihsan
> 

I can not a problem. I have been scatter brain'd lately
I didn't remember if I took them or not.

	I will re-create them in a bit.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknw0WAACgkQLrhmsXMSLxenegCgiHMDlo5/tsVSsBn5fT1wSn23
VGYAnRutS/lQQ7Edewt9SBrKVMdzQavJ
=Uo12
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 23 22:57:45 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 23 Apr 2009 22:57:45 +0200
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0D161.5090507@opencsw.org>
References: <49F07B5A.5080608@opencsw.org>	<49F0AD05.1070909@nature.berkeley.edu>	<49F0AE30.5070406@opencsw.org>	<49F0D0F6.6050003@opencsw.org>
	<49F0D161.5090507@opencsw.org>
Message-ID: <49F0D649.2060004@opencsw.org>

Am 23.4.2009 22:36 Uhr, Mike Watters schrieb:

>>> Gary,	
>>> 	I don't maintain those  ;)
>>>
>>> Ihsan: do you have any plans on pushing out a new version of those?
>>>
>>> I tested running php5 with the existing ap2_modphp5 ... with no issues.
>>> I don't run 1.3x anymore I didn't test against mod_php5
>> I thought, you took these as well.
> 
> I can not a problem. I have been scatter brain'd lately
> I didn't remember if I took them or not.
> 
> 	I will re-create them in a bit.

Thanks.

I think it makes sense, when all PHP packages are maintained by a single
person.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Fri Apr 24 06:40:38 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 23:40:38 -0500
Subject: [csw-maintainers] php5 now in testing
Message-ID: <49F142C6.1060106@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


Changes:
	rebuilt to include the 2.x and 1.3x apache modules in main php5 recipe.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknxQsYACgkQLrhmsXMSLxcNlwCfboUUr4DSznoNDxXMdKte+nnW
vhsAnAljY7T4q8kvsm5wONCOqZwXED94
=npFG
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Sat Apr 25 17:19:49 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sat, 25 Apr 2009 17:19:49 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
Message-ID: <49F32A15.8040901@opencsw.org>

>>> Clamd was not started, even it said that it was started.
>> I think this is related to a bug when multiple services are set up
>> like above. The placement of the start block in the script is wrong.
>>
>> I will look into it. Thanks for the report.
> 
> There's a new version in testing to address this problem, please give
> it a try if you have the time. It should start all services a package
> contains and give better info about what it's doing. It should also
> properly stop services on non-SMF systems.
> 
> http://mirror.opencsw.org/testing/cswclassutils-1.8,REV=2009.04.23-SunOS5.8-all-CSW.pkg.gz

Thanks. I will test it on Sunday.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Sat Apr 25 17:29:19 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Sat, 25 Apr 2009 17:29:19 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49F32A15.8040901@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com> 
	<49EF2AA0.3080402@opencsw.org>
	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com> 
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com> 
	<49F32A15.8040901@opencsw.org>
Message-ID: <625385e30904250829o619cb81ek3aa676b151075fc5@mail.gmail.com>

On Sat, Apr 25, 2009 at 5:19 PM, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Thanks. I will test it on Sunday.

Note that it's now:

http://mirror.opencsw.org/testing/cswclassutils-1.9,REV=2009.04.24-SunOS5.8-all-CSW.pkg.gz

It has a few more bug fixes and I had to comment out the state
persistent code. The last few days I've been working on making
cswinitsmf able to handle multiple services per package and even
though the state persistent feature is great during upgrades it does
not support that, I have notified Yann and he's looking at it so it
will come back soon.

-- 
/peter


From mwatters at opencsw.org  Sun Apr 26 03:10:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Apr 2009 20:10:33 -0500
Subject: [csw-maintainers] roundcube webmail?
Message-ID: <49F3B489.5000604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I tried to send an email to maintainers using roundcube mail.opencsw.org.

I got a bounce saying...

"Your mail to 'maintainers' with the subject

    Update on Subversion

Is being held until the list moderator can review it for approval.

The reason it is being held:

    Post by non-member to a members-only list"
...

is this normal?


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknztIkACgkQLrhmsXMSLxcpQACfQVbMdQ3Fckah3qVAx2SKMsK8
ylYAn1Ln6VSOzbTVLNg3Bu91Un9iWCV5
=pf6u
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sun Apr 26 05:34:53 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Apr 2009 22:34:53 -0500
Subject: [csw-maintainers] RFE mGar
Message-ID: <49F3D65D.4070907@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

if possible, could we get an option to print what is "UNCOMMITTED"

I keep finding myself searching for files that need committed,
deleted, or needing added to ignore.

it would be nice if we had a quick way to find them.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknz1l0ACgkQLrhmsXMSLxfSlQCglF5ls2ibR4TT6L3BgOti4rWI
8YsAnRuGY9L4MjMousTne8oeS/9AdwAc
=w9Oi
-----END PGP SIGNATURE-----


From dam at opencsw.org  Sun Apr 26 09:22:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Apr 2009 09:22:08 +0200
Subject: [csw-maintainers] RFE mGar
In-Reply-To: <49F3D65D.4070907@opencsw.org>
References: <49F3D65D.4070907@opencsw.org>
Message-ID: <CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>

Hi Mike,

Am 26.04.2009 um 05:34 schrieb Mike Watters <mwatters at opencsw.org>:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> if possible, could we get an option to print what is "UNCOMMITTED"
>
> I keep finding myself searching for files that need committed,
> deleted, or needing added to ignore.
>
> it would be nice if we had a quick way to find them.

If you do a 'svn status' and see more than the external gar reference  
the files shown are the ones to be either committed or ignored. We may  
have a target for that, by I guess it is rather straight forward.

Best regards

   -- Dago



From mwatters at localhost.opencsw.org  Sun Apr 26 01:42:29 2009
From: mwatters at localhost.opencsw.org (mwatters)
Date: Sun, 26 Apr 2009 01:42:29 +0200
Subject: [csw-maintainers] Update on Subversion
In-Reply-To: <49F32A15.8040901@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
	<49F32A15.8040901@opencsw.org>
Message-ID: <fc8682fa9ed37a32432e01c976cf92ab@localhost>


I have the issues just about worked out with Subversion.
I also figured out what changed between the "good compile" and 
now.   neon was updated to 0.28.4 from 0.28.3

this is not an issue, but I found the subversion compile
fails when I use --with-neon=$(prefix)

when I remove the line and let the configure "find" it it works
beautifully.  I will check the bug list and file one if it don't exist.

version 1.6.1 of subversion should be in testing shortly.


-- Mike


From trygvis at opencsw.org  Sun Apr 26 14:02:01 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sun, 26 Apr 2009 14:02:01 +0200
Subject: [csw-maintainers] RFE mGar
In-Reply-To: <CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
References: <49F3D65D.4070907@opencsw.org>
	<CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
Message-ID: <49F44D39.40109@opencsw.org>

Dagobert Michelsen wrote:
> Hi Mike,
> 
> Am 26.04.2009 um 05:34 schrieb Mike Watters <mwatters at opencsw.org>:
> 
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> if possible, could we get an option to print what is "UNCOMMITTED"
>>
>> I keep finding myself searching for files that need committed,
>> deleted, or needing added to ignore.
>>
>> it would be nice if we had a quick way to find them.
> 
> If you do a 'svn status' and see more than the external gar reference 
> the files shown are the ones to be either committed or ignored. We may 
> have a target for that, by I guess it is rather straight forward.

I started once on a set of "scm-*" goals as a facade for the operations 
a OpenCSW maintainer is supposed to use, it might be nice to add 
"scm-status" when you're in a package for those not used to SCMs or 
Subversion in particular.

--
Trygve


From bwalton at opencsw.org  Sun Apr 26 14:09:34 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Apr 2009 08:09:34 -0400
Subject: [csw-maintainers] RFE mGar
In-Reply-To: <CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
References: <49F3D65D.4070907@opencsw.org>
	<CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
Message-ID: <1240747607-sup-863@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Apr 26 03:22:08 -0400 2009:

Hi Mike,

> If you do a 'svn status' and see more than the external gar reference  
> the files shown are the ones to be either committed or ignored. We may  
> have a target for that, by I guess it is rather straight forward.

I'll point out specifically that UNCOMMITTED will occur when the
output of `svn status` contains files marked as ?.  Eg, files not
added to svn will trigger this state.  While it might be nice to
allow untracked files to be excepted from this, it's difficult to
properly prune out things like: Makefile is committed, but the new
.postinstall file is still unknown, etc.

HTH.
-Ben

-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090426/811fa9da/attachment.asc>

From ihsan at opencsw.org  Sun Apr 26 23:34:35 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 26 Apr 2009 23:34:35 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904250829o619cb81ek3aa676b151075fc5@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
	<49F32A15.8040901@opencsw.org>
	<625385e30904250829o619cb81ek3aa676b151075fc5@mail.gmail.com>
Message-ID: <49F4D36B.3010101@opencsw.org>

Hello Peter,

Am 25.4.2009 17:29 Uhr, Peter Bonivart schrieb:

>> Thanks. I will test it on Sunday.
> 
> Note that it's now:
> 
> http://mirror.opencsw.org/testing/cswclassutils-1.9,REV=2009.04.24-SunOS5.8-all-CSW.pkg.gz
> 
> It has a few more bug fixes and I had to comment out the state
> persistent code. The last few days I've been working on making
> cswinitsmf able to handle multiple services per package and even
> though the state persistent feature is great during upgrades it does
> not support that, I have notified Yann and he's looking at it so it
> will come back soon.

I just installed 1.9 on my testing system and after deinstallation and
reinstallation of NSD, it started the daemon, even the SMF service was
disabled.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From ihsan at opencsw.org  Sun Apr 26 23:37:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 26 Apr 2009 23:37:02 +0200
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F3B489.5000604@opencsw.org>
References: <49F3B489.5000604@opencsw.org>
Message-ID: <49F4D3FE.7060705@opencsw.org>

Hello Mike,

Am 26.4.2009 3:10 Uhr, Mike Watters schrieb:

> I tried to send an email to maintainers using roundcube mail.opencsw.org.

[...]

Ouuhmm, looks like there is a configuration mistake. The sender address
is wrong. I'll have a look if I can set a predefined domain.

You can fix it with setting your sender address to xyz at opencsw.org.

BTW: Does anybody know a good webmail client? I'm not really satisfied
with Roundcube.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From rupert at opencsw.org  Mon Apr 27 01:24:24 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 27 Apr 2009 01:24:24 +0200
Subject: [csw-maintainers] setuptools does not find cc compiler?
Message-ID: <6af4270904261624w8af7e6en736851dd87ee6394@mail.gmail.com>

today i had time to test edgewall trac. python-2.6.1, svn-1.5.6, trac,
pysvn, sqlite3, apache worked fine.

the only thing i noticed was with setuptools while trying to do a
native genshi install:

Downloading http://ftp.edgewall.com/pub/genshi/Genshi-0.5.1.zip
Processing Genshi-0.5.1.zip
Running Genshi-0.5.1/setup.py -q bdist_egg --dist-dir
/tmp/easy_install-wiThg9/Genshi-0.5.1/egg-dist-tmp-5GkKb1
warning: no previously-included files found matching 'doc/2000ft.graffle'
warning: no previously-included files matching '*' found under
directory 'doc/logo.lineform'
unable to execute /opt/studio/SOS11/SUNWspro/bin/cc: No such file or directory
**********************************************************************
WARNING:
An optional C extension could not be compiled, speedups will not be
available.
**********************************************************************
Adding Genshi 0.5.1 to easy-install.pth file

we have cc in another location, and on the path. also
"CC=/opt/SUNWspro/bin/cc easy_install genshi" resulted in the same
error message.

is this something for upstream?

rupert.


From rupert at opencsw.org  Mon Apr 27 01:25:40 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 27 Apr 2009 01:25:40 +0200
Subject: [csw-maintainers] Installation of <CSWosshclient> partially failed.
Message-ID: <6af4270904261625w33fb63cfy7e0eb842c0550c3@mail.gmail.com>

while trying to install git, i got the following error (but had no
time to test if there is any impact).


Installing openssh_client - OpenSSH Secure Shell as <CSWosshclient>

## Installing part 1 of 1.
/opt/csw/bin/scp
/opt/csw/bin/sftp
/opt/csw/bin/slogin <symbolic link>
/opt/csw/bin/ssh
/opt/csw/bin/ssh-add
/opt/csw/bin/ssh-agent
/opt/csw/bin/ssh-keygen
/opt/csw/bin/ssh-keyscan
/opt/csw/libexec/ssh-keysign
/opt/csw/share/doc/openssh_client/CREDITS
/opt/csw/share/doc/openssh_client/ChangeLog
/opt/csw/share/doc/openssh_client/ChangeLog.gssapi
/opt/csw/share/doc/openssh_client/INSTALL
/opt/csw/share/doc/openssh_client/LICENCE
/opt/csw/share/doc/openssh_client/OVERVIEW
/opt/csw/share/doc/openssh_client/README
/opt/csw/share/doc/openssh_client/README.dns
/opt/csw/share/doc/openssh_client/README.platform
/opt/csw/share/doc/openssh_client/README.privsep
/opt/csw/share/doc/openssh_client/README.smartcard
/opt/csw/share/doc/openssh_client/README.tun
/opt/csw/share/doc/openssh_client/TODO
/opt/csw/share/doc/openssh_client/WARNING.RNG
/opt/csw/share/doc/openssh_client/changelog.CSW
/opt/csw/share/man/man1/scp.1
/opt/csw/share/man/man1/sftp.1
/opt/csw/share/man/man1/slogin.1 <symbolic link>
/opt/csw/share/man/man1/ssh-add.1
/opt/csw/share/man/man1/ssh-agent.1
/opt/csw/share/man/man1/ssh-keygen.1
/opt/csw/share/man/man1/ssh-keyscan.1
/opt/csw/share/man/man1/ssh.1
/opt/csw/share/man/man5/ssh_config.5
/opt/csw/share/man/man8/ssh-keysign.8
[ verifying class <none> ]
[ verifying class <conf> ]
ERROR: attribute verification of </opt/csw/etc/ssh/ssh_config> failed
   pathname does not exist

Installation of <CSWosshclient> partially failed.


From mwatters at opencsw.org  Mon Apr 27 02:05:37 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 26 Apr 2009 19:05:37 -0500
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F4D3FE.7060705@opencsw.org>
References: <49F3B489.5000604@opencsw.org> <49F4D3FE.7060705@opencsw.org>
Message-ID: <49F4F6D1.6050900@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ihsan Dogan wrote:
> Hello Mike,
> 
> Am 26.4.2009 3:10 Uhr, Mike Watters schrieb:
> 
>> I tried to send an email to maintainers using roundcube mail.opencsw.org.
> 
> [...]
> 
> Ouuhmm, looks like there is a configuration mistake. The sender address
> is wrong. I'll have a look if I can set a predefined domain.
> 
> You can fix it with setting your sender address to xyz at opencsw.org.
> 
> BTW: Does anybody know a good webmail client? I'm not really satisfied
> with Roundcube.
> 
> 
> 
> 
> Ihsan
> 
I have always had good luck with squirrel mail
http://squirrelmail.org/

I will look at packaging it up.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn09tEACgkQLrhmsXMSLxfZ7QCgqB6c1t8gFp5B3khV5gmxlbrT
rLUAni5XZRhfXs8q3Xk6ir2YYzJ/CQ7B
=PEr4
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Mon Apr 27 02:14:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Apr 2009 20:14:01 -0400
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F4F6D1.6050900@opencsw.org>
References: <49F3B489.5000604@opencsw.org> <49F4D3FE.7060705@opencsw.org>
	<49F4F6D1.6050900@opencsw.org>
Message-ID: <1240791137-sup-4640@ntdws12.chass.utoronto.ca>

Excerpts from Mike Watters's message of Sun Apr 26 20:05:37 -0400 2009:
> I have always had good luck with squirrel mail
> http://squirrelmail.org/

We use squirrelmail in a few different configs here.  It works, but
I'd never call it 'good.'

Is having Google host the mail through their '...for your domain'
service an option for csw?

> I will look at packaging it up.

Cool.

Thanks
-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090426/87d88c69/attachment.asc>

From phil at bolthole.com  Mon Apr 27 04:19:15 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Apr 2009 19:19:15 -0700
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F4D3FE.7060705@opencsw.org>
References: <49F3B489.5000604@opencsw.org> <49F4D3FE.7060705@opencsw.org>
Message-ID: <20090427021915.GB8187@bolthole.com>

On Sun, Apr 26, 2009 at 11:37:02PM +0200, Ihsan Dogan wrote:
> 
> BTW: Does anybody know a good webmail client? I'm not really satisfied
> with Roundcube.

prayer is pretty good.

no, there really is an IMAP webmap client called "prayer" :-)

www-uxsup.csx.cam.ac.uk/~dpc22/prayer/

it's fast, because it is native C.

"Written entirely in C as HTTP to IMAP Gateway. No scripting languages. "

no perl. no php. NO APACHE!




From mwatters at opencsw.org  Mon Apr 27 05:15:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 26 Apr 2009 22:15:49 -0500
Subject: [csw-maintainers] Subversion 1.6.1 now in testing
Message-ID: <49F52365.1040700@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Bindings included: Perl, Python, Ruby, JavaHL

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn1I2UACgkQLrhmsXMSLxdFjwCfZaZBR5bdFcYr0CJkvb/mbmJF
l3oAoKaIStZsPgqWakBPb3xOdbziuLrc
=IgvO
-----END PGP SIGNATURE-----


From dam at opencsw.org  Mon Apr 27 15:03:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 27 Apr 2009 15:03:01 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <20081107133324.GA2059@zerfleddert.de>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>
	<49137326.3000605@wbonnet.net>
	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>
	<20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
Message-ID: <A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>

Hi Michael,

Am 07.11.2008 um 14:33 schrieb Michael Gernoth:
> My old build-scripts and sources for all my packages can be found at:
> http://csw.informatik.uni-erlangen.de/csw/users/michael/sources/
> The scripts are in the 'specs' subdirectory.

I copied your old build descriptions to the GAR repository in the
respective pkg/<pkg>/trunk/legacy/ folders as a reference if
someone wants to bring a package in GAR.


Best regards

   -- Dago


From yann at pleiades.fr.eu.org  Mon Apr 27 23:10:28 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Mon, 27 Apr 2009 23:10:28 +0200
Subject: [csw-maintainers] Installation of <CSWosshclient> partially
	failed.
In-Reply-To: <6af4270904261625w33fb63cfy7e0eb842c0550c3@mail.gmail.com>
References: <6af4270904261625w33fb63cfy7e0eb842c0550c3@mail.gmail.com>
Message-ID: <49F61F44.3040509@pleiades.fr.eu.org>

Hi,

Could you please open a bug for this problem in mantis 
http://www.opencsw.org/mantis/view.php?id=2652
?

Could you also give me the verbose output of the installation ?
You need to manually download the package and install it using pkgadd 
with the -v option to get the verbose output.

Thanks in advance,

Yann

Le 27.04.2009 01:25, rupert THURNER a ?crit :
> while trying to install git, i got the following error (but had no
> time to test if there is any impact).
>
>
> Installing openssh_client - OpenSSH Secure Shell as<CSWosshclient>
>
> ## Installing part 1 of 1.
> /opt/csw/bin/scp
> /opt/csw/bin/sftp
> /opt/csw/bin/slogin<symbolic link>
> /opt/csw/bin/ssh
> /opt/csw/bin/ssh-add
> /opt/csw/bin/ssh-agent
> /opt/csw/bin/ssh-keygen
> /opt/csw/bin/ssh-keyscan
> /opt/csw/libexec/ssh-keysign
> /opt/csw/share/doc/openssh_client/CREDITS
> /opt/csw/share/doc/openssh_client/ChangeLog
> /opt/csw/share/doc/openssh_client/ChangeLog.gssapi
> /opt/csw/share/doc/openssh_client/INSTALL
> /opt/csw/share/doc/openssh_client/LICENCE
> /opt/csw/share/doc/openssh_client/OVERVIEW
> /opt/csw/share/doc/openssh_client/README
> /opt/csw/share/doc/openssh_client/README.dns
> /opt/csw/share/doc/openssh_client/README.platform
> /opt/csw/share/doc/openssh_client/README.privsep
> /opt/csw/share/doc/openssh_client/README.smartcard
> /opt/csw/share/doc/openssh_client/README.tun
> /opt/csw/share/doc/openssh_client/TODO
> /opt/csw/share/doc/openssh_client/WARNING.RNG
> /opt/csw/share/doc/openssh_client/changelog.CSW
> /opt/csw/share/man/man1/scp.1
> /opt/csw/share/man/man1/sftp.1
> /opt/csw/share/man/man1/slogin.1<symbolic link>
> /opt/csw/share/man/man1/ssh-add.1
> /opt/csw/share/man/man1/ssh-agent.1
> /opt/csw/share/man/man1/ssh-keygen.1
> /opt/csw/share/man/man1/ssh-keyscan.1
> /opt/csw/share/man/man1/ssh.1
> /opt/csw/share/man/man5/ssh_config.5
> /opt/csw/share/man/man8/ssh-keysign.8
> [ verifying class<none>  ]
> [ verifying class<conf>  ]
> ERROR: attribute verification of</opt/csw/etc/ssh/ssh_config>  failed
>     pathname does not exist
>
> Installation of<CSWosshclient>  partially failed.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From dam at opencsw.org  Tue Apr 28 20:06:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Apr 2009 20:06:55 +0200
Subject: [csw-maintainers] Problem on CSWphp5zip
Message-ID: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>

Hi,

I got a problem on lots of PHP 5 packages like CSWphp5zip and  
CSWphp5session:

> /var/sadm/pkg/CSWphp5zip/install/postinstall: /home/mwatters/mgar/ 
> pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/php.ini^Jecho  
> "[===> Running Post Install <===]"^Jecho "   ===> Enabling zip  
> extension"^Jif grep CSW: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall:  $ 
> {PHP_INI}^Jelse^Jperl -i -pe s: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall: .*CSW: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall: CSW: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall:  ${PHP_INI}^Jfi^Jcat  
> <<  
> _EOF_ 
> ^ 
> J 
> ******************************************************************************^ 
> J* NOTICE: Successfully Enabled Extension "zip"^J* in $ 
> {PHP_INI}^J*^J* You will need to restart your web server to finish  
> the  
> install 
> ^ 
> J 
> ******************************************************************************^ 
> J_EOF_^Jexit 0^Jgmake[4]: Leaving directory : not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall: gmake[4]:: not found
> pkgadd: ERROR: postinstall script did not complete successfully
>
> Installation of <CSWphp5zip> failed.

...

> Removal of <CSWphp5session> was successful.
> Analysing special files...
> /var/sadm/pkg/CSWphp5session/install/postinstall: /home/mwatters/ 
> mgar/pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/ 
> php.ini^Jecho "[===> Running Post Install <===]"^Jecho "   ===>  
> Enabling session extension"^Jif grep CSW: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall:  $ 
> {PHP_INI}^Jelse^Jperl -i -pe s: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall: .*CSW: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall: CSW: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall:  $ 
> {PHP_INI}^Jfi^Jcat <<  
> _EOF_ 
> ^ 
> J 
> ******************************************************************************^ 
> J* NOTICE: Successfully Enabled Extension "session"^J* in $ 
> {PHP_INI}^J*^J* You will need to restart your web server to finish  
> the  
> install 
> ^ 
> J 
> ******************************************************************************^ 
> J_EOF_^Jexit 0^Jgmake[4]: Leaving directory : not found
> /var/sadm/pkg/CSWphp5session/install/postinstall: gmake[4]:: not found
> pkgadd: ERROR: postinstall script did not complete successfully


Best regards

-- Dago



From dam at opencsw.org  Tue Apr 28 20:07:18 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Apr 2009 20:07:18 +0200
Subject: [csw-maintainers] Updating all buildservers to current/
Message-ID: <52882627-7DE1-4B38-8709-9C2B3E7EF9B6@opencsw.org>

Hi,

I am updating all buildservers now to current. Please stand by.


Best regards

-- Dago


From mwatters at opencsw.org  Tue Apr 28 20:33:30 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Apr 2009 13:33:30 -0500
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
Message-ID: <49F74BFA.5040301@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> I got a problem on lots of PHP 5 packages like CSWphp5zip and
> CSWphp5session:
> 
>> /var/sadm/pkg/CSWphp5zip/install/postinstall:
>> /home/mwatters/mgar/pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/php.ini^Jecho
>> "[===> Running Post Install <===]"^Jecho "   ===> Enabling zip
>> extension"^Jif grep CSW: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall:  ${PHP_INI}^Jelse^Jperl
>> -i -pe s: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall: .*CSW: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall: CSW: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall:  ${PHP_INI}^Jfi^Jcat <<
>> _EOF_^J******************************************************************************^J*
>> NOTICE: Successfully Enabled Extension "zip"^J* in ${PHP_INI}^J*^J*
>> You will need to restart your web server to finish the
>> install^J******************************************************************************^J_EOF_^Jexit
>> 0^Jgmake[4]: Leaving directory : not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall: gmake[4]:: not found
>> pkgadd: ERROR: postinstall script did not complete successfully
>>
>> Installation of <CSWphp5zip> failed.
> 
> ...
> 
>> Removal of <CSWphp5session> was successful.
>> Analysing special files...
>> /var/sadm/pkg/CSWphp5session/install/postinstall:
>> /home/mwatters/mgar/pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/php.ini^Jecho
>> "[===> Running Post Install <===]"^Jecho "   ===> Enabling session
>> extension"^Jif grep CSW: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: 
>> ${PHP_INI}^Jelse^Jperl -i -pe s: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: .*CSW: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: CSW: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall:  ${PHP_INI}^Jfi^Jcat
>> <<
>> _EOF_^J******************************************************************************^J*
>> NOTICE: Successfully Enabled Extension "session"^J* in
>> ${PHP_INI}^J*^J* You will need to restart your web server to finish
>> the
>> install^J******************************************************************************^J_EOF_^Jexit
>> 0^Jgmake[4]: Leaving directory : not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: gmake[4]:: not found
>> pkgadd: ERROR: postinstall script did not complete successfully
> 
> 
> Best regards
> 
> -- Dago
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Dago,
This was the bug that I re-packaged to fix.  can you verify for me what REV you
 attempted to install?

when I do a pkg-get -d php5_zip I get the "broken" one.  REV=2009.04.18

The "Fixed" one is: REV=2009.04.27

Phil: have you finished processing and pushed out all of the php5 packages yet?

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkn3S/oACgkQLrhmsXMSLxcEEwCfRtfquvRJfojsfTNK+B33ZZQL
VigAnj+z/86Rm0bwWvxPr6js8Lh6Svr0
=3mBe
-----END PGP SIGNATURE-----


From dam at opencsw.org  Tue Apr 28 21:00:13 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Apr 2009 21:00:13 +0200
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <49F74BFA.5040301@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
Message-ID: <3E598EA3-9E4C-45DB-8B96-13D2CC4F6DC1@opencsw.org>

Hi Mike,

Am 28.04.2009 um 20:33 schrieb Mike Watters:
> Dago,
> This was the bug that I re-packaged to fix.  can you verify for me  
> what REV you
> attempted to install?
>
> when I do a pkg-get -d php5_zip I get the "broken" one.   
> REV=2009.04.18

Yep, old one:

build8s% pkginfo -l CSWphp5zip
    PKGINST:  CSWphp5zip
       NAME:  php5_zip - zip Extention for PHP5
   CATEGORY:  application
       ARCH:  sparc
    VERSION:  5.2.9,REV=2009.04.18
     VENDOR:  http://www.php.net/downloads.php packaged for CSW by  
Mike Watters
     PSTAMP:  mwatters at build8s-20090418142017
   INSTDATE:  Apr 28 2009 19:59
    HOTLINE:  http://www.opencsw.org/bugtrack/
      EMAIL:  mwatters at opencsw.org
     STATUS:  partially installed
      FILES:        6 installed pathnames
                    5 shared pathnames
                    5 directories
                    1 executables
                  967 blocks used (approx)

> The "Fixed" one is: REV=2009.04.27
>
> Phil: have you finished processing and pushed out all of the php5  
> packages yet?

Immediately before updating I synced up to rsync.opencsw.org.

Uni Erlangen is still old, too:
   <http://ftp.uni-erlangen.de/pub/mirrors/csw/current/sparc/5.8/>


Best regards

   -- Dago


From phil at bolthole.com  Tue Apr 28 21:15:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Apr 2009 12:15:31 -0700
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <49F74BFA.5040301@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
Message-ID: <20090428191531.GC97557@bolthole.com>

On Tue, Apr 28, 2009 at 01:33:30PM -0500, Mike Watters wrote:
> Phil: have you finished processing and pushed out all of the php5 packages yet?

I have. yesterday.
So, Dago probably just hit mirror lag.


From mwatters at opencsw.org  Tue Apr 28 23:39:18 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Apr 2009 16:39:18 -0500
Subject: [csw-maintainers] New Subversion 1.6.1 in testing
Message-ID: <49F77786.6080308@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	Added

share/doc/subversion/tools/*
share/doc/subversion/contrib/*
bin/svn-tools/*
bin/svn-contrib/*


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkn3d4YACgkQLrhmsXMSLxfvCQCZAdiUMnpsolP8gyq5HQ9bp2j/
bQ4AoK+vq+SSuQhcWDcRBC5KIsmLSaEt
=kMjk
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Wed Apr 29 02:35:25 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Apr 2009 19:35:25 -0500
Subject: [csw-maintainers] solaris 8 on CentOS?
Message-ID: <49F7A0CD.2090405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I remember someone saying they had solaris 8
running on CentOS, sadly, I don't remember whom.

if you still have it, can you post your config.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn3oMwACgkQLrhmsXMSLxc7ygCfYp5iUDJqIiWgVGILDk60W2U9
HFwAn3goksz38l5HJJJvjvARzxHdmpxh
=vEZM
-----END PGP SIGNATURE-----


From dam at opencsw.org  Wed Apr 29 09:13:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 09:13:43 +0200
Subject: [csw-maintainers] Fwd: [bug-notifications] [mutt 0003648]: mutt
	does not work with screen's altscreen because it is compiled
	with slang instead of ncurses
References: <1b25a27cdbd0df04a3965beb5c31788e@www.opencsw.org>
Message-ID: <C097C6F7-20AE-4607-8F12-FB8CAC1D6BC1@opencsw.org>

Hi,

I picked up this discussion between ncurses and slang and I guess
it is a good idea to discuss the ncurses/slang matter publicly
on the list.


Best regards

   -- Dago

Anfang der weitergeleiteten E-Mail:

> Von: Mantis Bug Tracker <noreply at opencsw.org>
> Datum: 29. April 2009 07:59:07 MESZ
> An: bug-notifications at lists.opencsw.org
> Betreff: [bug-notifications] [mutt 0003648]: mutt does not work with  
> screen's altscreen because it is compiled with slang instead of  
> ncurses
> Antwort an: users at lists.opencsw.org
>
>
> The following issue has been SUBMITTED.
> ======================================================================
> http://www.opencsw.org/bugtrack/view.php?id=3648
> ======================================================================
> Reported By:                meunier
> Assigned To:
> ======================================================================
> Project:                    mutt
> Issue ID:                   3648
> Category:                   regular use
> Reproducibility:            always
> Severity:                   block
> Priority:                   normal
> Status:                     new
> ======================================================================
> Date Submitted:             2009-04-29 07:59 CEST
> Last Modified:              2009-04-29 07:59 CEST
> ======================================================================
> Summary:                    mutt does not work with screen's  
> altscreen because
> it is compiled with slang instead of ncurses
> Description:
> /opt/csw/bin/mutt is compiled with slang, not ncurses, while
> /opt/csw/bin/screen is compiled with ncurses, not slang.  So if you  
> use
> /opt/csw/bin/screen and start /opt/csw/bin/mutt inside it, it looks  
> like
> the ncurses library used by screen and the slang library used by  
> mutt fight
> each other in a bad way when screen's altscreen feature is enabled.
>
> Here is a way to reproduce the problem:
>
> [after ssh-ing into a Solaris machine from an xterm]
> $ export $TERMINFO=/opt/csw/share/terminfo/
> $ /opt/csw/bin/infocmp
> #       Reconstructed via infocmp from file:
> /opt/csw/share/terminfo/x/xterm
> [blablabla ... so the correct terminfo database is being used]
> $ ls
> [blablabla]
> $ /opt/csw/bin/emacs
> [exit emacs, the screen returns to its previous content, including  
> showing
> the output of the previous 'ls' command.  Note: this will not work  
> and the
> ouput of the previous 'ls' command will be invisible if your TERMINFO
> environment variable is not set correctly]
> $ ls
> [blablabla]
> $ /opt/csw/bin/mutt
> [exit mutt, the screen returns to its previous content, including  
> showing
> the output of the previous 'ls' command]
> $ /opt/csw/bin/screen
> [screen is cleared]
> $ ls
> [blablabla]
> $ /opt/csw/bin/emacs
> [exit from emacs, the cursor is at the bottom of the xterm and the  
> output
> of the previous 'ls' is not visible anymore]
> $ ls
> [blablabla]
> $ /opt/csw/bin/mutt
> [exit from mutt, the cursor is at the bottom of the xterm and the  
> output
> of the previous 'ls' is not visible anymore]
>
> So far so good.  Emacs and mutt normally use xterm's "alternate  
> screen"
> feature which is why the output of the previous 'ls' command is  
> visible in
> the xterm once emacs or mutt has exited.  Screen, on the other hand,  
> does
> not provide an alternate screen by default, so emacs and mutt just  
> use the
> "regular screen" and the output of the previous 'ls' command is then  
> lost
> when emacs or mutt exits.
>
> Now type:
>
> Control-A :
>
> to get the interactive prompt from 'screen', then type:
>
> altscreen on
>
> then you should get a 'Will do alternate screen switching' from  
> 'screen'.
> This tells 'screen' that it should provide an alternate screen to
> applications like emacs or mutt that normally use xterm's alternate  
> screen
> feature.
>
> Now let's try emacs and mutt again:
>
> $ ls
> [blablabla]
> $ /opt/csw/bin/emacs
> [exit emacs, the screen returns to its previous content, including  
> showing
> the output of the previous 'ls' command, just as if emacs were run  
> from a
> normal shell instead of being run from within 'screen'.  Great,  
> that's what
> I want.]
> $ ls
> [blablabla]
> $ /opt/csw/bin/mutt
> [oops, watch screen and mutt fight for control of the alternate  
> screen...
> You can try to type a quick random combination of
>
> x
>
> and
>
> Control-A "
>
> to tell mutt to exit and screen to give you a list of virtual screens
> (rather than fighting with mutt) but good luck with regaining  
> control of
> your window...]
>
> Now, the fact that emacs works fine in combination with screen's  
> altscreen
> feature but that mutt does not tells me that the problem is with  
> mutt, not
> screen. After investigating a little, I've come to the conclusion  
> that the
> problem is not with the code of mutt itself, but with the fact that
> /opt/csw/bin/mutt uses slang while /opt/csw/bin/screen uses ncurses.  
> In
> fact I have compiled (with gcc) a version of mutt 1.5.19 with  
> ncurses 5.7
> which works perfectly well in the examples above.  On the other hand  
> the
> same version of mutt 1.5.19 compiled with slang 2.1.4 fails just like
> /opt/csw/bin/mutt, flashing the screen and all.  My /opt/csw/bin/ 
> mutt uses
> slang 1.4.8, not slang 2.1.4, but that doesn't seem to make any  
> difference,
> both fail in the same way.
>
> So is there a way to get /opt/csw/bin/screen to be compiled with  
> ncurses
> rather than slang, by any chance?
>
> Thanks,
>
>
> ======================================================================
>
> Issue History
> Date Modified    Username       Field                    Change
> ======================================================================
> 2009-04-29 07:59 meunier        New Issue
> ======================================================================
>
> _______________________________________________
> bug-notifications mailing list
> bug-notifications at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/bug-notifications



From hson at opencsw.org  Wed Apr 29 09:49:40 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 29 Apr 2009 09:49:40 +0200
Subject: [csw-maintainers] solaris 8 on CentOS?
In-Reply-To: <49F7A0CD.2090405@opencsw.org>
References: <49F7A0CD.2090405@opencsw.org>
Message-ID: <49F80694.1000103@opencsw.org>

Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I remember someone saying they had solaris 8
> running on CentOS, sadly, I don't remember whom.
> 
> if you still have it, can you post your config.
> 

It was me.

I'm running CentOS 5.2 with VMware server 2.0 and the virtual machine 
config isn't anything strange at all, mainly guestOS = "solaris8".

I can send you my config file if you wish (or even upload my image)


From dam at opencsw.org  Wed Apr 29 13:57:52 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 13:57:52 +0200
Subject: [csw-maintainers] Fwd:  Broken CSWftype2
References: <66824D49-2D25-4398-B93B-907070806DED@baltic-online.de>
Message-ID: <87C9EAD1-5534-4217-9A01-0BB49F9732F8@opencsw.org>

Hi,

Am 16.04.2009 um 09:04 schrieb Dagobert Michelsen:
> Am 20.02.2009 um 12:56 schrieb Ihsan Dogan:
>> Am 17.2.2009 19:03 Uhr, Dagobert Michelsen schrieb:
>>>
>>>> brauchst zum Kompilieren die aktuelle freetype2 Version aus /
>>>> testing.
>>>> Koenntest Du das bitte auf der buildfarm einspielen ?
>>>> Ziel ist es, freetype2 und fontconfig nach einer Testphase
>>>> gleichzeitig zu releasen.
>>>
>>> Ok, so freetype2 and fontconfig should be released at the
>>> same time and freetype2 must be installed on the buildfarm
>>> for building fontconfig.
>>>
>>> Ihsan, Phil: Should I do it this way violating standards or
>>> should these packages be released one after another?
>>
>> As I understand, we don't have many choices, so it's fine for me.
>> Important is, that a notify is sent to the maintainers list.
>
> We have not build8st for this. AFAIK this has not been done yet.
> Want to give it a try?

I notice that fontconfig has been released, but not freetype2.
Is this ok? Should I recompile now cairo, pango and atk or
wait until freetype2 has been released?


Best regards

  -- Dago



From dam at opencsw.org  Wed Apr 29 14:41:18 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 14:41:18 +0200
Subject: [csw-maintainers] New Subversion 1.6.1 in testing
In-Reply-To: <49F77786.6080308@opencsw.org>
References: <49F77786.6080308@opencsw.org>
Message-ID: <CF7958CE-EC9F-4EDC-AF91-00DF21FEF031@opencsw.org>

Hi Mike,

Am 28.04.2009 um 23:39 schrieb Mike Watters:
> New Subversion 1.6.1 in testing

I mirrored up GAR from SourceForge using svnsync and did some
other testing - works great!


Best regads

   -- Dago


From pfelecan at opencsw.org  Wed Apr 29 17:26:25 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 29 Apr 2009 17:26:25 +0200
Subject: [csw-maintainers] packages names normalization (long)
Message-ID: <x6r5zb8ori.fsf@abhavana.free.fr>

Reading the standards page,
http://www.opencsw.org/standards/pkgcreation, I understand that the
standard for naming CSW splitted packages is:

PKGname	     shortname
------------ ---------
CSWsoftrt    soft_rt
CSWsoftdevel soft_devel
CSWsoftdoc   soft_doc 

Some of the cited rationales are: to optimize the download size of the
required packages, to offer a better granularity in packages
inter-dependencies, &c.

For the moment, this standard is not applied consistently, but I
understand that it is now enforced.

Here is a more developed functional relation with the suffixes cited above:

_rt    : run-time

         Any component needed to run the product or *derived* products:
         shared libraries, &c

_devel : development

         Any component needed for developing *derived*  but not
         needed for running it: headers, static libraries, pkgconfig
         elements, &c

_doc   : documentation

         Any component documenting the product and/or the creation of
         *derived* products.

There is something missing: the nil suffix which, in my opinion,
group the components which are *run*, i.e. binaries, shell scripts, &c

A package with a nil suffix, using the above examples, has the form:

CSWsoft      soft
CSWsoftrt    soft_rt
CSWsoftdevel soft_devel
CSW softdoc  soft_doc

Why do I think that we should have nil suffixed packages names:

1. Installing the product by:

   pkg-get install softrt

   is less intuitive than:

   pkg-get install soft

2. Installing the product's binaries is not always necessary to run
   *derived* products, its run-time is however necessary.

Giving a full example is probably a good way to enhance the
documentation of this standard. Consequently, here is the splitting of
the autogen project:

CSWautogen      autogen
CSWautogenrt    autogen_rt
CSWautogendevel autogen_devel
CSWautogendoc   autogen_doc

and their respective content is (I omitted the /opt/csw prefix):

autogen:

	bin/autogen
	bin/autoopts-config
	bin/columns
	bin/getdefs
	bin/xml2ag
	share/autogen/aginfo.tpl
	share/autogen/aginfo3.tpl
	share/autogen/agman-lib.tpl
	share/autogen/agman1.tpl
	share/autogen/agman3.tpl
	share/autogen/autoopts.m4
	share/autogen/confmacs.tpl
	share/autogen/conftest.tpl
	share/autogen/fsm-macro.tpl
	share/autogen/fsm-trans.tpl
	share/autogen/fsm.tpl
	share/autogen/getopt.tpl
	share/autogen/optcode.tpl
	share/autogen/opthead.tpl
	share/autogen/options.tpl
	share/autogen/optlib.tpl
	share/autogen/optmain.tpl
	share/autogen/rc-sample.tpl
	share/autogen/stdoptions.def
	share/autogen/usage.tpl
	share/man/man1/autogen.1
	share/man/man1/autoopts-config.1
	share/man/man1/columns.1
	share/man/man1/getdefs.1
	share/man/man1/xml2ag.1

autogen_rt:

	lib/libguileopts.so.0.0.1
	lib/libopts.so.25.7.0
	share/doc/autogen/AUTHORS
	share/doc/autogen/COPYING
	share/doc/autogen/ChangeLog
	share/doc/autogen/INSTALL
	share/doc/autogen/NEWS
	share/doc/autogen/README
	share/doc/autogen/THANKS
	share/doc/autogen/TODO

autogen_devel:

	include/autoopts/options.h
	include/autoopts/usage-txt.h
	lib/libguileopts.a
	lib/libopts.a
	lib/pkgconfig
	lib/pkgconfig/autoopts.pc
	share/aclocal/autoopts.m4
	share/aclocal/liboptschk.m4
	share/man/man3/ao_string_tokenize.3
	share/man/man3/configFileLoad.3
	share/man/man3/optionFileLoad.3
	share/man/man3/optionFindNextValue.3
	share/man/man3/optionFindValue.3
	share/man/man3/optionFree.3
	share/man/man3/optionGetValue.3
	share/man/man3/optionLoadLine.3
	share/man/man3/optionNextValue.3
	share/man/man3/optionOnlyUsage.3
	share/man/man3/optionProcess.3
	share/man/man3/optionRestore.3
	share/man/man3/optionSaveFile.3
	share/man/man3/optionSaveState.3
	share/man/man3/optionUnloadNested.3
	share/man/man3/optionVersion.3
	share/man/man3/strequate.3
	share/man/man3/streqvcmp.3
	share/man/man3/streqvmap.3
	share/man/man3/strneqvcmp.3
	share/man/man3/strtransform.3

autogen_doc:

	share/doc/autogen/autogen.dvi
	share/doc/autogen/autogen.pdf
	share/doc/autogen/autogen.ps
	share/info/autogen.info
	share/info/autogen.info-1
	share/info/autogen.info-2

All your comments are welcome and when we agree it should be nice to
update the standards page.
-- 
Peter


From dam at opencsw.org  Wed Apr 29 17:38:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 17:38:40 +0200
Subject: [csw-maintainers] libcairo 1.8.6 available
Message-ID: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>

Hi,

I just put libcairo 1.8.6 compiled against the updated
fontconfig. Please test it so we can move ahead with all
the other stuff depending on it:

libcairo-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
libcairo-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
libcairo_doc-1.8.6,REV=2009.04.29-SunOS5.8-all-CSW.pkg.gz

pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libcairo
pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libcairo

Best regards

   -- Dago


From dam at opencsw.org  Wed Apr 29 17:52:38 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 17:52:38 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6r5zb8ori.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
Message-ID: <4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org>

Hi Peter,

Am 29.04.2009 um 17:26 schrieb Peter FELECAN:
> Reading the standards page,
> http://www.opencsw.org/standards/pkgcreation, I understand that the
> standard for naming CSW splitted packages is:
>
> PKGname	     shortname
> ------------ ---------
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSWsoftdoc   soft_doc
>
> Some of the cited rationales are: to optimize the download size of the
> required packages, to offer a better granularity in packages
> inter-dependencies, &c.
>
> For the moment, this standard is not applied consistently, but I
> understand that it is now enforced.
>
> Here is a more developed functional relation with the suffixes cited  
> above:
>
> _rt    : run-time
>
>         Any component needed to run the product or *derived* products:
>         shared libraries, &c
>
> _devel : development
>
>         Any component needed for developing *derived*  but not
>         needed for running it: headers, static libraries, pkgconfig
>         elements, &c
>
> _doc   : documentation
>
>         Any component documenting the product and/or the creation of
>         *derived* products.
>
> There is something missing: the nil suffix which, in my opinion,
> group the components which are *run*, i.e. binaries, shell scripts, &c

It may also very well be a meta-package to say "I just want to
use this software". In most cases this will be the binaries and
depend on _rt. However, in packages whose sole purpose is the
development of software things get more complicated as I'll
illustrate below.

> A package with a nil suffix, using the above examples, has the form:
>
> CSWsoft      soft
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSW softdoc  soft_doc
>
> Why do I think that we should have nil suffixed packages names:
>
> 1. Installing the product by:
>
>   pkg-get install softrt

_rt-packages will usually be never installed manually, they are
meant to be pulled in as dependencies. Factoring out _rt to
minimize impact of dependencies is one of the reasons that
runtime-stuff is actually split off in a separate package.

>
>
>   is less intuitive than:
>
>   pkg-get install soft
>
> 2. Installing the product's binaries is not always necessary to run
>   *derived* products, its run-time is however necessary.

True.

> Giving a full example is probably a good way to enhance the
> documentation of this standard. Consequently, here is the splitting of
> the autogen project:

Bad example IMHO, as the sole purpose of autogen is to aid
in software development and needed for compilation. No user
will use it.

> CSWautogen      autogen
> CSWautogenrt    autogen_rt
> CSWautogendevel autogen_devel
> CSWautogendoc   autogen_doc
>
> and their respective content is (I omitted the /opt/csw prefix):
>
> autogen:
>
> 	bin/autogen
> 	bin/autoopts-config

By default bin/*-config is put in _devel in GAR.

>
> 	bin/columns
> 	bin/getdefs
> 	bin/xml2ag
> 	share/autogen/aginfo.tpl
> 	share/autogen/aginfo3.tpl
> 	share/autogen/agman-lib.tpl
> 	share/autogen/agman1.tpl
> 	share/autogen/agman3.tpl
> 	share/autogen/autoopts.m4
> 	share/autogen/confmacs.tpl
> 	share/autogen/conftest.tpl
> 	share/autogen/fsm-macro.tpl
> 	share/autogen/fsm-trans.tpl
> 	share/autogen/fsm.tpl
> 	share/autogen/getopt.tpl
> 	share/autogen/optcode.tpl
> 	share/autogen/opthead.tpl
> 	share/autogen/options.tpl
> 	share/autogen/optlib.tpl
> 	share/autogen/optmain.tpl
> 	share/autogen/rc-sample.tpl
> 	share/autogen/stdoptions.def
> 	share/autogen/usage.tpl
> 	share/man/man1/autogen.1
> 	share/man/man1/autoopts-config.1

Same is for the manpage.

>
> 	share/man/man1/columns.1
> 	share/man/man1/getdefs.1
> 	share/man/man1/xml2ag.1
>
> autogen_rt:
>
> 	lib/libguileopts.so.0.0.1
> 	lib/libopts.so.25.7.0
>
> 	share/doc/autogen/AUTHORS
> 	share/doc/autogen/COPYING
> 	share/doc/autogen/ChangeLog
> 	share/doc/autogen/INSTALL
> 	share/doc/autogen/NEWS
> 	share/doc/autogen/README
> 	share/doc/autogen/THANKS
> 	share/doc/autogen/TODO

Docs usually go into the base package if there is no _doc.

> autogen_devel:
>
> 	include/autoopts/options.h
> 	include/autoopts/usage-txt.h
> 	lib/libguileopts.a
> 	lib/libopts.a
> 	lib/pkgconfig
> 	lib/pkgconfig/autoopts.pc
> 	share/aclocal/autoopts.m4
> 	share/aclocal/liboptschk.m4
> 	share/man/man3/ao_string_tokenize.3
> 	share/man/man3/configFileLoad.3
> 	share/man/man3/optionFileLoad.3
> 	share/man/man3/optionFindNextValue.3
> 	share/man/man3/optionFindValue.3
> 	share/man/man3/optionFree.3
> 	share/man/man3/optionGetValue.3
> 	share/man/man3/optionLoadLine.3
> 	share/man/man3/optionNextValue.3
> 	share/man/man3/optionOnlyUsage.3
> 	share/man/man3/optionProcess.3
> 	share/man/man3/optionRestore.3
> 	share/man/man3/optionSaveFile.3
> 	share/man/man3/optionSaveState.3
> 	share/man/man3/optionUnloadNested.3
> 	share/man/man3/optionVersion.3
> 	share/man/man3/strequate.3
> 	share/man/man3/streqvcmp.3
> 	share/man/man3/streqvmap.3
> 	share/man/man3/strneqvcmp.3
> 	share/man/man3/strtransform.3
>
> autogen_doc:
>
> 	share/doc/autogen/autogen.dvi
> 	share/doc/autogen/autogen.pdf
> 	share/doc/autogen/autogen.ps
> 	share/info/autogen.info
> 	share/info/autogen.info-1
> 	share/info/autogen.info-2

texinfo pages are put in the main-package as they can be
thought of another form of manpages.

> All your comments are welcome and when we agree it should be nice to
> update the standards page.

Additionally, there should be license files
   /opt/csw/share/doc/autogen/license
   /opt/csw/share/doc/autogen_rt/license
   /opt/csw/share/doc/autogen_devel/license
   /opt/csw/share/doc/autogen_doc/license
with the contents of share/doc/autogen/COPYING


Best regards

   -- Dago


From pfelecan at opencsw.org  Wed Apr 29 18:17:43 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 29 Apr 2009 18:17:43 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org> (Dagobert
	Michelsen's message of "Wed\, 29 Apr 2009 17\:52\:38 +0200")
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org>
Message-ID: <x6my9z8me0.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

Thank you Dago for your quick and appropriate answer.

>> Giving a full example is probably a good way to enhance the
>> documentation of this standard. Consequently, here is the splitting of
>> the autogen project:
>
> Bad example IMHO, as the sole purpose of autogen is to aid
> in software development and needed for compilation. No user
> will use it.

I know that it's an artificial, viz. inappropriate, example. However, I
choose it because I had a lengthy discussion with Phil about the need to
split it (I wanted to deliver it as an unique package, he had the
opinion that I should split)

>> CSWautogen      autogen
>> CSWautogenrt    autogen_rt
>> CSWautogendevel autogen_devel
>> CSWautogendoc   autogen_doc
>>
>> and their respective content is (I omitted the /opt/csw prefix):
>>
>> autogen:
>>
>> 	bin/autogen
>> 	bin/autoopts-config
>
> By default bin/*-config is put in _devel in GAR.

Right.

>> 	share/man/man1/autoopts-config.1
>
> Same is for the manpage.

Agreed.

>> autogen_rt:
>>
>> 	lib/libguileopts.so.0.0.1
>> 	lib/libopts.so.25.7.0
>>
>> 	share/doc/autogen/AUTHORS
>> 	share/doc/autogen/COPYING
>> 	share/doc/autogen/ChangeLog
>> 	share/doc/autogen/INSTALL
>> 	share/doc/autogen/NEWS
>> 	share/doc/autogen/README
>> 	share/doc/autogen/THANKS
>> 	share/doc/autogen/TODO
>
> Docs usually go into the base package if there is no _doc.

I put this quasi-standard documentation files in the run-time package
because it's installed always and is of interest for everybody (a
little bit like Debian does).

>> autogen_doc:
>>
>> 	share/doc/autogen/autogen.dvi
>> 	share/doc/autogen/autogen.pdf
>> 	share/doc/autogen/autogen.ps
>> 	share/info/autogen.info
>> 	share/info/autogen.info-1
>> 	share/info/autogen.info-2
>
> texinfo pages are put in the main-package as they can be
> thought of another form of manpages.

I'm not sure about this. If I look other packaging sources, again Debian
comes up as an interesting example, the info files are in a separate,
documentation package. It can be argued both ways...

> Additionally, there should be license files
>   /opt/csw/share/doc/autogen/license
>   /opt/csw/share/doc/autogen_rt/license
>   /opt/csw/share/doc/autogen_devel/license
>   /opt/csw/share/doc/autogen_doc/license
> with the contents of share/doc/autogen/COPYING

Right. And they are but not cited in my example.

-- 
Peter


From Darin.Perusich at cognigencorp.com  Wed Apr 29 18:22:03 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 29 Apr 2009 12:22:03 -0400
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6r5zb8ori.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
Message-ID: <49F87EAB.3060100@cognigencorp.com>

One thing I feel needs to be addressed is package naming consistency
between PKGname and shortname. If the short name is soft_rt or soft-rt
shouldn't the PKGname be CSWsoft_rt or CSWsoft-rt? There's no
"technical" reason for this given my reading of the pkginfo manual. This
lack of consistency makes searching for packages with pkginfo or listing
/var/sadm/pkg a really hassle.

A perfect example of this is openSSL or openSSH. The PKGname for the
various component packages is CSWossl* or CSWossh*. One would kind of
expect to the PKGname to be CSWopenssl_rt or CSWopenssh_client. I
realize there are historical reasons for not changed this, I'm simply
using this as examples.

Peter FELECAN wrote:
> Reading the standards page,
> http://www.opencsw.org/standards/pkgcreation, I understand that the
> standard for naming CSW splitted packages is:
> 
> PKGname	     shortname
> ------------ ---------
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSWsoftdoc   soft_doc 
> 
> Some of the cited rationales are: to optimize the download size of the
> required packages, to offer a better granularity in packages
> inter-dependencies, &c.
> 
> For the moment, this standard is not applied consistently, but I
> understand that it is now enforced.
> 
> Here is a more developed functional relation with the suffixes cited above:
> 
> _rt    : run-time
> 
>          Any component needed to run the product or *derived* products:
>          shared libraries, &c
> 
> _devel : development
> 
>          Any component needed for developing *derived*  but not
>          needed for running it: headers, static libraries, pkgconfig
>          elements, &c
> 
> _doc   : documentation
> 
>          Any component documenting the product and/or the creation of
>          *derived* products.
> 
> There is something missing: the nil suffix which, in my opinion,
> group the components which are *run*, i.e. binaries, shell scripts, &c
> 
> A package with a nil suffix, using the above examples, has the form:
> 
> CSWsoft      soft
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSW softdoc  soft_doc
> 
> Why do I think that we should have nil suffixed packages names:
> 
> 1. Installing the product by:
> 
>    pkg-get install softrt
> 
>    is less intuitive than:
> 
>    pkg-get install soft
> 
> 2. Installing the product's binaries is not always necessary to run
>    *derived* products, its run-time is however necessary.
> 
> Giving a full example is probably a good way to enhance the
> documentation of this standard. Consequently, here is the splitting of
> the autogen project:
> 
> CSWautogen      autogen
> CSWautogenrt    autogen_rt
> CSWautogendevel autogen_devel
> CSWautogendoc   autogen_doc
> 
> and their respective content is (I omitted the /opt/csw prefix):
> 
> autogen:
> 
> 	bin/autogen
> 	bin/autoopts-config
> 	bin/columns
> 	bin/getdefs
> 	bin/xml2ag
> 	share/autogen/aginfo.tpl
> 	share/autogen/aginfo3.tpl
> 	share/autogen/agman-lib.tpl
> 	share/autogen/agman1.tpl
> 	share/autogen/agman3.tpl
> 	share/autogen/autoopts.m4
> 	share/autogen/confmacs.tpl
> 	share/autogen/conftest.tpl
> 	share/autogen/fsm-macro.tpl
> 	share/autogen/fsm-trans.tpl
> 	share/autogen/fsm.tpl
> 	share/autogen/getopt.tpl
> 	share/autogen/optcode.tpl
> 	share/autogen/opthead.tpl
> 	share/autogen/options.tpl
> 	share/autogen/optlib.tpl
> 	share/autogen/optmain.tpl
> 	share/autogen/rc-sample.tpl
> 	share/autogen/stdoptions.def
> 	share/autogen/usage.tpl
> 	share/man/man1/autogen.1
> 	share/man/man1/autoopts-config.1
> 	share/man/man1/columns.1
> 	share/man/man1/getdefs.1
> 	share/man/man1/xml2ag.1
> 
> autogen_rt:
> 
> 	lib/libguileopts.so.0.0.1
> 	lib/libopts.so.25.7.0
> 	share/doc/autogen/AUTHORS
> 	share/doc/autogen/COPYING
> 	share/doc/autogen/ChangeLog
> 	share/doc/autogen/INSTALL
> 	share/doc/autogen/NEWS
> 	share/doc/autogen/README
> 	share/doc/autogen/THANKS
> 	share/doc/autogen/TODO
> 
> autogen_devel:
> 
> 	include/autoopts/options.h
> 	include/autoopts/usage-txt.h
> 	lib/libguileopts.a
> 	lib/libopts.a
> 	lib/pkgconfig
> 	lib/pkgconfig/autoopts.pc
> 	share/aclocal/autoopts.m4
> 	share/aclocal/liboptschk.m4
> 	share/man/man3/ao_string_tokenize.3
> 	share/man/man3/configFileLoad.3
> 	share/man/man3/optionFileLoad.3
> 	share/man/man3/optionFindNextValue.3
> 	share/man/man3/optionFindValue.3
> 	share/man/man3/optionFree.3
> 	share/man/man3/optionGetValue.3
> 	share/man/man3/optionLoadLine.3
> 	share/man/man3/optionNextValue.3
> 	share/man/man3/optionOnlyUsage.3
> 	share/man/man3/optionProcess.3
> 	share/man/man3/optionRestore.3
> 	share/man/man3/optionSaveFile.3
> 	share/man/man3/optionSaveState.3
> 	share/man/man3/optionUnloadNested.3
> 	share/man/man3/optionVersion.3
> 	share/man/man3/strequate.3
> 	share/man/man3/streqvcmp.3
> 	share/man/man3/streqvmap.3
> 	share/man/man3/strneqvcmp.3
> 	share/man/man3/strtransform.3
> 
> autogen_doc:
> 
> 	share/doc/autogen/autogen.dvi
> 	share/doc/autogen/autogen.pdf
> 	share/doc/autogen/autogen.ps
> 	share/info/autogen.info
> 	share/info/autogen.info-1
> 	share/info/autogen.info-2
> 
> All your comments are welcome and when we agree it should be nice to
> update the standards page.

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From dam at opencsw.org  Wed Apr 29 18:36:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 18:36:51 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6my9z8me0.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org>
	<x6my9z8me0.fsf@abhavana.free.fr>
Message-ID: <BCE77DAB-3F48-437D-B211-7A5D6CC53242@opencsw.org>

Hi Peter,

Am 29.04.2009 um 18:17 schrieb Peter FELECAN:
> Dagobert Michelsen <dam at opencsw.org> writes:
>
> Thank you Dago for your quick and appropriate answer.
>
>>> Giving a full example is probably a good way to enhance the
>>> documentation of this standard. Consequently, here is the  
>>> splitting of
>>> the autogen project:
>>
>> Bad example IMHO, as the sole purpose of autogen is to aid
>> in software development and needed for compilation. No user
>> will use it.
>
> I know that it's an artificial, viz. inappropriate, example.  
> However, I
> choose it because I had a lengthy discussion with Phil about the  
> need to
> split it (I wanted to deliver it as an unique package, he had the
> opinion that I should split)

To make a long story short: If there are packages that depend
on the .so-files split out _rt, otherwise I would argue to
leave it as one package as it is only valuable to developers
anyway.

Phil, did you have other reasons for the split?


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr 29 18:39:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 18:39:55 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <49F87EAB.3060100@cognigencorp.com>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
Message-ID: <C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>

Hi Darin,

Am 29.04.2009 um 18:22 schrieb Darin Perusich:
> One thing I feel needs to be addressed is package naming consistency
> between PKGname and shortname. If the short name is soft_rt or soft-rt
> shouldn't the PKGname be CSWsoft_rt or CSWsoft-rt? There's no
> "technical" reason for this given my reading of the pkginfo manual.  
> This
> lack of consistency makes searching for packages with pkginfo or  
> listing
> /var/sadm/pkg a really hassle.
>
> A perfect example of this is openSSL or openSSH. The PKGname for the
> various component packages is CSWossl* or CSWossh*. One would kind of
> expect to the PKGname to be CSWopenssl_rt or CSWopenssh_client. I
> realize there are historical reasons for not changed this, I'm simply
> using this as examples.

The Sun "Standard" uses "-" as separator, so it would be more like
CSWopenssl-rt. However, some packages already use no separator like
CSWopensslrt.

I'm all for consistency, but as Phil correctly said changing package
names is kind of bad. If pkg-get/pkgutil were smart enough this *may*
be worked through, but I don't know if it is worth it.


Best regards

   -- Dago


From pfelecan at opencsw.org  Wed Apr 29 18:47:40 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 29 Apr 2009 18:47:40 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org> (Dagobert
	Michelsen's message of "Wed\, 29 Apr 2009 18\:39\:55 +0200")
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
Message-ID: <x6iqkn8l03.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Darin,
>
> Am 29.04.2009 um 18:22 schrieb Darin Perusich:
>> One thing I feel needs to be addressed is package naming consistency
>> between PKGname and shortname. If the short name is soft_rt or soft-rt
>> shouldn't the PKGname be CSWsoft_rt or CSWsoft-rt? There's no
>> "technical" reason for this given my reading of the pkginfo manual.
>> This
>> lack of consistency makes searching for packages with pkginfo or
>> listing
>> /var/sadm/pkg a really hassle.
>>
>> A perfect example of this is openSSL or openSSH. The PKGname for the
>> various component packages is CSWossl* or CSWossh*. One would kind of
>> expect to the PKGname to be CSWopenssl_rt or CSWopenssh_client. I
>> realize there are historical reasons for not changed this, I'm simply
>> using this as examples.
>
> The Sun "Standard" uses "-" as separator, so it would be more like
> CSWopenssl-rt. However, some packages already use no separator like
> CSWopensslrt.

The reason for which '-' cannot be used in catalogue package, vs. sys V
pkg, names is that is a "field separator" used for name, version,
revision, architecture &c; in my understanding, we can use '_' in
catalogue names and also in sys V pkg. If we want to normalize, your
example becomes CSWopenssl_rt with a catalogue name of openssl_rt.

-- 
Peter


From phil at bolthole.com  Wed Apr 29 19:08:24 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:08:24 -0700
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
Message-ID: <20090429170824.GB9202@bolthole.com>

On Wed, Apr 29, 2009 at 06:39:55PM +0200, Dagobert Michelsen wrote:
>
> The Sun "Standard" uses "-" as separator, so it would be more like
> CSWopenssl-rt. However, some packages already use no separator like
> CSWopensslrt.

personally, I like not bothering with any separator at all, for CSWxxx
names, unless the resulting name is somehow misleading/unreadable.



From phil at bolthole.com  Wed Apr 29 19:10:53 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:10:53 -0700
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6r5zb8ori.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
Message-ID: <20090429171053.GC9202@bolthole.com>

PS:

On Wed, Apr 29, 2009 at 05:26:25PM +0200, Peter FELECAN wrote:
> There is something missing: the nil suffix which, in my opinion,
> group the components which are *run*, i.e. binaries, shell scripts, &c
> 
> A package with a nil suffix, using the above examples, has the form:
> 
> CSWsoft      soft
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSW softdoc  soft_doc

and.. some packages are like this already.

It depends on the individual software in question, whether there is, as you
put it, a "nil suffix" variant of the package.

And as Dagobert pointed out, "library" type things arent usually directly
installed via pkg-get. They are usually only pulled in as dependancies.



From phil at bolthole.com  Wed Apr 29 19:11:42 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:11:42 -0700
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6iqkn8l03.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
	<x6iqkn8l03.fsf@abhavana.free.fr>
Message-ID: <20090429171142.GD9202@bolthole.com>

On Wed, Apr 29, 2009 at 06:47:40PM +0200, Peter FELECAN wrote:
> The reason for which '-' cannot be used in catalogue package, vs. sys V
> pkg, names is that is a "field separator" used for name, version,
> revision, architecture &c; in my understanding, we can use '_' in
> catalogue names and also in sys V pkg. If we want to normalize, your
> example becomes CSWopenssl_rt with a catalogue name of openssl_rt.

except that "_" is invalid for SVR4 pkg naming, for some reason, i believe.



From dam at opencsw.org  Wed Apr 29 19:17:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 19:17:43 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <20090429171142.GD9202@bolthole.com>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
	<x6iqkn8l03.fsf@abhavana.free.fr>
	<20090429171142.GD9202@bolthole.com>
Message-ID: <189FC9D7-90BE-4599-B1C6-B16444312B37@opencsw.org>

Hi Phil,

Am 29.04.2009 um 19:11 schrieb Philip Brown:
> On Wed, Apr 29, 2009 at 06:47:40PM +0200, Peter FELECAN wrote:
>> The reason for which '-' cannot be used in catalogue package, vs.  
>> sys V
>> pkg, names is that is a "field separator" used for name, version,
>> revision, architecture &c; in my understanding, we can use '_' in
>> catalogue names and also in sys V pkg. If we want to normalize, your
>> example becomes CSWopenssl_rt with a catalogue name of openssl_rt.
>
> except that "_" is invalid for SVR4 pkg naming, for some reason, i  
> believe.

Yes. From pkginfo(4):

>      PKG
>
>          Abbreviation for the package being installed. All  char-
>          acters in the abbreviation must be alphanumeric. You can
>          also use the - and +  characters  in  the  abbreviation.
>          The first character cannot be numeric, a + or a -.

Best regards

   -- Dago


From phil at bolthole.com  Wed Apr 29 19:37:24 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:37:24 -0700
Subject: [csw-maintainers] Fwd:  Broken CSWftype2
In-Reply-To: <87C9EAD1-5534-4217-9A01-0BB49F9732F8@opencsw.org>
References: <66824D49-2D25-4398-B93B-907070806DED@baltic-online.de>
	<87C9EAD1-5534-4217-9A01-0BB49F9732F8@opencsw.org>
Message-ID: <20090429173724.GF9202@bolthole.com>

On Wed, Apr 29, 2009 at 01:57:52PM +0200, Dagobert Michelsen wrote:
> I notice that fontconfig has been released, but not freetype2.
> Is this ok? Should I recompile now cairo, pango and atk or
> wait until freetype2 has been released?


Hm. weeell... since freetype2 "depends on" fontconfig... if the existing
released version does not work, seems like the most
appropriate thing to do is do a quick re-compile and releaes of freetype2
against the just released fontconfig.

please update build machines with latest freetype2 asap.

Then lets get a new freetype2 released quick?

And yes i think safest to wait until these two are done, given how picky
cairo, pango, etc have been in the past.



From phil at bolthole.com  Wed Apr 29 19:57:40 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:57:40 -0700
Subject: [csw-maintainers] Fwd: [bug-notifications] [mutt 0003648]:	mutt
	does not work with screen's altscreen because it is
	compiled	with slang instead of ncurses
In-Reply-To: <C097C6F7-20AE-4607-8F12-FB8CAC1D6BC1@opencsw.org>
References: <1b25a27cdbd0df04a3965beb5c31788e@www.opencsw.org>
	<C097C6F7-20AE-4607-8F12-FB8CAC1D6BC1@opencsw.org>
Message-ID: <20090429175740.GN9202@bolthole.com>

On Wed, Apr 29, 2009 at 09:13:43AM +0200, Dagobert Michelsen wrote:
> Hi,
>
> I picked up this discussion between ncurses and slang and I guess
> it is a good idea to discuss the ncurses/slang matter publicly
> on the list.

ncurses sucks. use slang.
that's all I have to say on that one :-)



From dam at opencsw.org  Wed Apr 29 20:48:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 20:48:17 +0200
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <20090428191531.GC97557@bolthole.com>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
	<20090428191531.GC97557@bolthole.com>
Message-ID: <270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>

Hi Phil,

Am 28.04.2009 um 21:15 schrieb Philip Brown:

> On Tue, Apr 28, 2009 at 01:33:30PM -0500, Mike Watters wrote:
>> Phil: have you finished processing and pushed out all of the php5  
>> packages yet?
>
> I have. yesterday.
> So, Dago probably just hit mirror lag.

No, there is something broken. I just looked up my rsync primary:

web at web [web]:/home/web/bin > rsync rsync://rsync.opencsw.org/csw/ 
current/sparc/5.8/ | grep php5
-rw-r--r--    1494635 2009/04/27 23:17:07  
ap2_modphp5-5.2.9,REV=2009.04.27-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--    1377061 2009/04/27 23:17:07  
mod_php5-5.2.9,REV=2009.04.27-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--    3056820 2009/04/20 19:14:28 php5-5.2.9,REV=2009.04.18- 
SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      55316 2009/02/23 20:38:46  
php5_apc-3.0.19,REV=2009.02.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      31096 2009/04/20 19:14:28  
php5_bcmath-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      14633 2009/04/20 19:14:29  
php5_bz2-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      18688 2009/04/20 19:14:29  
php5_calendar-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       8435 2009/04/20 19:14:29  
php5_ctype-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      29962 2009/04/20 19:14:30  
php5_curl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      35878 2009/04/20 19:14:30  
php5_dba-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20505 2009/04/20 19:14:30  
php5_dbase-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     516915 2009/04/20 19:14:31  
php5_devel-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      74762 2009/04/20 19:14:31  
php5_dom-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      30733 2009/04/20 19:14:31  
php5_exif-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--        748 2009/03/12 20:30:25  
php5_filter-5.2.6,REV=2009.03.08-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      24655 2009/04/20 19:14:32  
php5_ftp-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      50115 2009/04/20 19:14:32 php5_gd-5.2.9,REV=2009.04.18- 
SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       9399 2009/04/20 19:14:32  
php5_gettext-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20005 2009/04/20 19:14:33  
php5_gmp-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     124240 2009/04/20 19:14:33  
php5_hash-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      21397 2009/04/20 19:14:33  
php5_iconv-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      48371 2009/04/20 19:14:34  
php5_imap-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      17236 2009/04/20 19:14:34  
php5_json-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      25121 2009/04/20 19:14:34  
php5_ldap-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--    1132091 2009/04/20 19:14:35  
php5_mbstring-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      17867 2009/04/20 19:14:35  
php5_mcrypt-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       8978 2009/04/20 19:14:35  
php5_mhash-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      16366 2009/04/20 19:14:35  
php5_mime_magic-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      28450 2009/04/20 19:16:34  
php5_mssql-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      24222 2009/04/20 19:14:36  
php5_mysql-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      51030 2009/04/20 19:14:36  
php5_mysqli-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      27016 2009/04/20 19:14:37  
php5_ncurses-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      34827 2009/04/20 19:14:37  
php5_odbc-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      44545 2009/04/20 19:14:37  
php5_openssl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      12290 2009/04/20 19:14:37  
php5_pcntl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      51806 2009/04/21 18:48:05  
php5_pdo-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      19154 2009/04/21 18:48:06  
php5_pdomysql-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      16195 2009/04/21 18:48:06  
php5_pdoodbc-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      18251 2009/04/21 18:48:06  
php5_pdopgsql-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     217588 2009/04/21 20:32:22  
php5_pdosqlite-5.2.9,REV=2009.04.21-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      54032 2009/04/20 19:14:38  
php5_pgsql-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      13132 2009/04/20 19:14:38  
php5_posix-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      15107 2009/04/20 19:14:38  
php5_pspell-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11961 2009/04/21 01:42:25  
php5_readline-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      37229 2009/04/20 19:14:39  
php5_session-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       9237 2009/04/20 19:14:39  
php5_shmop-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      16712 2009/04/20 19:14:39  
php5_snmp-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     169450 2009/04/20 19:14:40  
php5_soap-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20875 2009/04/20 19:14:40  
php5_sockets-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      35274 2009/04/20 19:14:40  
php5_sqlite-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11806 2009/04/20 19:14:41  
php5_sysvmsg-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       8859 2009/04/20 19:14:41  
php5_sysvsem-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11636 2009/04/20 19:14:41  
php5_sysvshm-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20718 2009/04/20 19:14:42  
php5_tidy-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11246 2009/04/20 19:14:42  
php5_tokenizer-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      19022 2009/04/20 19:14:42  
php5_wddx-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     101274 2009/02/23 20:38:47  
php5_xdebug-2.0.4,REV=2009.02.17-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      15563 2009/04/20 19:14:43  
php5_xmlreader-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      42736 2009/04/20 19:14:43  
php5_xmlrpc-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      14638 2009/04/20 19:14:43  
php5_xmlwriter-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      17480 2009/04/20 19:14:43  
php5_xsl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      57071 2009/04/20 19:14:44  
php5_zip-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz

...old!

Am 28.04.2009 um 20:33 schrieb Mike Watters:
> Dago,
> This was the bug that I re-packaged to fix.  can you verify for me  
> what REV you
> attempted to install?
>
> when I do a pkg-get -d php5_zip I get the "broken" one.   
> REV=2009.04.18
>
> The "Fixed" one is: REV=2009.04.27
>
> Phil: have you finished processing and pushed out all of the php5  
> packages yet?


You did release the package, yes? Or isn't rsync.opencsw.org the  
primary any more?


Best regards

   -- Dago


From phil at bolthole.com  Wed Apr 29 21:03:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 12:03:33 -0700
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
	<20090428191531.GC97557@bolthole.com>
	<270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>
Message-ID: <20090429190332.GQ9202@bolthole.com>

On Wed, Apr 29, 2009 at 08:48:17PM +0200, Dagobert Michelsen wrote:
> Hi Phil,
>
> Am 28.04.2009 um 21:15 schrieb Philip Brown:
>
>> On Tue, Apr 28, 2009 at 01:33:30PM -0500, Mike Watters wrote:
>>> Phil: have you finished processing and pushed out all of the php5  
>>> packages yet?
>>
>> I have. yesterday.
>> So, Dago probably just hit mirror lag.
>
> No, there is something broken. I just looked up my rsync primary:

sorry sorry my mistake.



From phil at bolthole.com  Wed Apr 29 21:16:44 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 12:16:44 -0700
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <20090429190332.GQ9202@bolthole.com>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
	<20090428191531.GC97557@bolthole.com>
	<270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>
	<20090429190332.GQ9202@bolthole.com>
Message-ID: <20090429191641.GR9202@bolthole.com>

> 
> sorry sorry my mistake.

hmm. and the php5 "new" packages I hvae, ar bogus.
eg:

php5_openssl-5.2.9,REV=2009.04.27-SunOS5.8-sparc-UNCOMMITTED.pkg.gz




From william at wbonnet.net  Wed Apr 29 21:25:42 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 21:25:42 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
Message-ID: <49F8A9B6.602@wbonnet.net>

Hi
> I copied your old build descriptions to the GAR repository in the
> respective pkg/<pkg>/trunk/legacy/ folders as a reference if
> someone wants to bring a package in GAR.
libpango is a duplicate of pango and libxrender is a duplicate of 
x11/xrender

I think we should keep the old one

cheers
W.



-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From dam at opencsw.org  Wed Apr 29 21:36:09 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 21:36:09 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <49F8A9B6.602@wbonnet.net>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
Message-ID: <54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>

Hi William,

Am 29.04.2009 um 21:25 schrieb William Bonnet:
>> I copied your old build descriptions to the GAR repository in the
>> respective pkg/<pkg>/trunk/legacy/ folders as a reference if
>> someone wants to bring a package in GAR.
> libpango is a duplicate of pango and libxrender is a duplicate of  
> x11/xrender
>
> I think we should keep the old one

Well, the package is called CSWpango and the software is at pango.org,
so I chose pkg/pango instead of libpango, so I would prefer the new  
one ;-)
And xrender belongs to x11 where you put it and I also think this is a  
good
choice, no?


Best regards

   -- Dago


From william at wbonnet.net  Wed Apr 29 21:42:10 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 21:42:10 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
Message-ID: <49F8AD92.30203@wbonnet.net>

Hi Dagobert
> Well, the package is called CSWpango and the software is at pango.org,
> so I chose pkg/pango instead of libpango, so I would prefer the new 
> one ;-)
I also prefer the new one :)

cheers


-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Wed Apr 29 22:05:20 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 13:05:20 -0700
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>
	<49137326.3000605@wbonnet.net>
	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>
	<20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
Message-ID: <20090429200519.GT9202@bolthole.com>

On Wed, Apr 29, 2009 at 09:36:09PM +0200, Dagobert Michelsen wrote:
> Well, the package is called CSWpango and the software is at pango.org,
> so I chose pkg/pango instead of libpango, so I would prefer the new one 
> ;-)

but we have called "software name" to be "libpango" for a very long time.
and also other distribution call it "libpango".

so i think we should stick with "libpango".

ditto for libxrender.
its a standard name.



From william at wbonnet.net  Wed Apr 29 22:16:41 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 22:16:41 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <20090429200519.GT9202@bolthole.com>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>	<49F8A9B6.602@wbonnet.net>	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
	<20090429200519.GT9202@bolthole.com>
Message-ID: <49F8B5A9.6000208@wbonnet.net>

Hi

> but we have called "software name" to be "libpango" for a very long time.
> and also other distribution call it "libpango".
>
> so i think we should stick with "libpango".
>
> ditto for libxrender.
>   
So we move pango content to libpango and move libxrender under x11, then 
move xrender content into libxrender ?

cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From ihsan at opencsw.org  Wed Apr 29 22:33:27 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 29 Apr 2009 22:33:27 +0200
Subject: [csw-maintainers] libcairo 1.8.6 available
In-Reply-To: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
References: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
Message-ID: <49F8B997.9070705@opencsw.org>

Hi Dago,

Am 29.4.2009 17:38 Uhr, Dagobert Michelsen schrieb:

> I just put libcairo 1.8.6 compiled against the updated
> fontconfig. Please test it so we can move ahead with all
> the other stuff depending on it:
> 
> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.29-SunOS5.8-all-CSW.pkg.gz
> 
> pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libcairo
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libcairo

Could you please install it on build8st?



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Wed Apr 29 22:36:35 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 13:36:35 -0700
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <49F8B5A9.6000208@wbonnet.net>
References: <49137326.3000605@wbonnet.net>
	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>
	<20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
	<20090429200519.GT9202@bolthole.com> <49F8B5A9.6000208@wbonnet.net>
Message-ID: <20090429203635.GV9202@bolthole.com>

On Wed, Apr 29, 2009 at 10:16:41PM +0200, William Bonnet wrote:
> Hi
>
>> but we have called "software name" to be "libpango" for a very long time.
>> and also other distribution call it "libpango".
>>
>> so i think we should stick with "libpango".
>>
>> ditto for libxrender.
>>   
> So we move pango content to libpango

yes

> and move libxrender under x11, then  
> move xrender content into libxrender ?

umm... lost me there. is there separate "libxrender" and "xrender"?

PS: I'll remind you that we're supposed to have "flat namespace" in mGar
now! there should be no "under"!

otherwise, the nice future single-module autobuilder will not be able to
cleanly work!


From william at wbonnet.net  Wed Apr 29 22:50:29 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 22:50:29 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <20090429203635.GV9202@bolthole.com>
References: <49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>	<49F8A9B6.602@wbonnet.net>	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>	<20090429200519.GT9202@bolthole.com>
	<49F8B5A9.6000208@wbonnet.net> <20090429203635.GV9202@bolthole.com>
Message-ID: <49F8BD95.3010106@wbonnet.net>

Hi
>> and move libxrender under x11, then  
>> move xrender content into libxrender ?
>>     
>
> umm... lost me there. is there separate "libxrender" and "xrender"?
>   
yes x11/xrender

cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Wed Apr 29 23:09:36 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 14:09:36 -0700
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <49F8BD95.3010106@wbonnet.net>
References: <20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
	<20090429200519.GT9202@bolthole.com> <49F8B5A9.6000208@wbonnet.net>
	<20090429203635.GV9202@bolthole.com> <49F8BD95.3010106@wbonnet.net>
Message-ID: <20090429210936.GA79653@bolthole.com>

On Wed, Apr 29, 2009 at 10:50:29PM +0200, William Bonnet wrote:
> Hi
>>> and move libxrender under x11, then  move xrender content into 
>>> libxrender ?
>>>     
>>
>> umm... lost me there. is there separate "libxrender" and "xrender"?
>
> yes x11/xrender

well, unless they are SUPPOSED to be separate, then I would request it be
put in the top level, as "libxrender", to match software name



From mwatters at opencsw.org  Thu Apr 30 05:02:45 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 29 Apr 2009 22:02:45 -0500
Subject: [csw-maintainers] alpine 2.00 in testing
Message-ID: <49F914D5.9030008@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Sorry for the delay.
Please let me know how the testing goes.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn5FNUACgkQLrhmsXMSLxejcACgpkGOfhu3B/umUOhif37kbU9y
eR4An3HH1eoJIFk48d0fkCoCXyvsNivC
=74JT
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 30 15:54:42 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 30 Apr 2009 15:54:42 +0200
Subject: [csw-maintainers] libcairo 1.8.6 available
In-Reply-To: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
References: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
Message-ID: <49F9ADA2.3070205@opencsw.org>

Am 29.4.2009 17:38 Uhr, Dagobert Michelsen schrieb:

> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.29-SunOS5.8-all-CSW.pkg.gz

Have you built pango, which is installed on build8st, also with this
cairo version?




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Thu Apr 30 21:12:54 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Apr 2009 12:12:54 -0700
Subject: [csw-maintainers] holey blackhole files, batman! pkginfo oddities
Message-ID: <20090430191254.GG79653@bolthole.com>

Just as a reference to others/for the future;

I had a problem where I was tryign to update CSWglib2.
it got stuck, becaus of zone problems.

i ended up downloading the newer version, and installing manually.

but.... pkg-get kept saying I still needed to "upgrade" it!
?!!...

after use of truss, I discovered an interesting "feature" of pkginfo. (ugh)

there was a leftover magical directory, 

/var/sadm/pkg/.save.CSWglib2

It kept looking in THERE, instead of the proper

/var/sadm/pkg/CSWglib2

for the installed file information.

once i trashed that directory, things finally proceeded properly.


From Darin.Perusich at cognigencorp.com  Thu Apr 30 22:43:42 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 30 Apr 2009 16:43:42 -0400
Subject: [csw-maintainers] relocatable on sparc but not x86
Message-ID: <49FA0D7E.4060103@cognigencorp.com>

I'm staging a package and on SPARC it's not relocatable but on i386 it's
not, why would this be?

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Thu Apr 30 23:34:05 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Apr 2009 14:34:05 -0700
Subject: [csw-maintainers] relocatable on sparc but not x86
In-Reply-To: <49FA0D7E.4060103@cognigencorp.com>
References: <49FA0D7E.4060103@cognigencorp.com>
Message-ID: <20090430213405.GI79653@bolthole.com>

On Thu, Apr 30, 2009 at 04:43:42PM -0400, Darin Perusich wrote:
> I'm staging a package and on SPARC it's not relocatable but on i386 it's
> not, why would this be?
> 

that's kinda equivalent to saying, "it's broken. how can I fix it?" without saying
anything else :-/


at minimum, you need to find out what bits are non-relocatable, and tell us
what they are...



From harpchad at opencsw.org  Wed Apr  1 01:47:15 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 31 Mar 2009 18:47:15 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <F8FC4A58-F0D9-4D5B-A574-080E0028CB8F@opencsw.org>

I assume RUNPATHQUOTE=3 is also valid?  I have one package where  
RPATH=/opt/csw/lib/SALIST with =2, looks like $I got expanded.

On Mar 31, 2009, at 4:04 PM, Dagobert Michelsen wrote:

> /usr/ccs/bin/dump -Lv



From harpchad at opencsw.org  Wed Apr  1 01:52:31 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 31 Mar 2009 18:52:31 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>

Phil, can checkpkg detect some of the common cases?  I'm thinking at  
least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those  
would indicate a bad RUNPATHQUOTE value.  James can probably give us  
an idea of what he's coming across the most.


From mwatters at opencsw.org  Wed Apr  1 02:24:57 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 31 Mar 2009 19:24:57 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
Message-ID: <49D2B459.3040702@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chad Harp wrote:
> Phil, can checkpkg detect some of the common cases?  I'm thinking at
> least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those would
> indicate a bad RUNPATHQUOTE value.  James can probably give us an idea
> of what he's coming across the most.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I am not sure if it should go into checkpkg or as a post-install in gar.
put a post-install that will not allow the package to continue until it
matches NOISALIST = 0 or NOISALIST = 1 dump -Lv


if we catch it at the post install it will allow a quicker fix then
waiting for the package to merge and package.



- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknStFkACgkQLrhmsXMSLxck8ACeIvN/nzrlDucP0Rr7ZnESgG6M
N4YAoOpDY06tU1O8hrHD4IyX50Y+pUXP
=JUoL
-----END PGP SIGNATURE-----


From rupert at opencsw.org  Wed Apr  1 07:25:23 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Wed, 1 Apr 2009 07:25:23 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <6af4270903312225q6f57bfc8k63e3c21ffb3379c4@mail.gmail.com>

currently i get:
~$ ssh rupert at login.opencsw.org
PTY allocation request failed on channel 0
Received disconnect from 213.178.77.178: 2: fork failed: Not enough space


On Tue, Mar 31, 2009 at 23:04, Dagobert Michelsen <dam at opencsw.org> wrote:
> Hi,
>
> James is currently in the tedious task of reviewing all
> runtime linker pathes from the packages. To make it
> easier for you as maintainer I have a summary here how
> this works in GAR and how to fix things.
>
> Runtime pathes can be checked with
> ?/usr/ccs/bin/dump -Lv <binary> | <library>
>
> A correct setting with ISALIST should look like
> [7] ? ? RUNPATH ? ? ? ? /opt/csw/lib/$ISALIST:/opt/csw/lib
> [8] ? ? RPATH ? ? ? ? ? /opt/csw/lib/$ISALIST:/opt/csw/lib
>
> A correct setting without ISALIST should look like
> [7] ? ? RUNPATH ? ? ? ? /opt/csw/lib
> [8] ? ? RPATH ? ? ? ? ? /opt/csw/lib
>
> GAR adds a runtime path with $ISALIST by default. To skip this
> entirely in GAR you can use
> ?NOISALIST = 1
> How much shell escaping is necessary depends on if the application
> uses autoconf and/or libtool. The quoting-level can be reduced with
> ?RUNPATHQUOTE = 1
> (quote for one expansion, e. g. only configure) as opposed to the default
> ?RUNPATHQUOTE = 2
> (quote for double expansion, configure and libtool).
>
>
> Best regards
>
> ?-- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From rupert at opencsw.org  Wed Apr  1 07:29:29 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Wed, 1 Apr 2009 07:29:29 +0200
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
References: <ce97543b0903260441k26550f94r96a1f6bb6c95755d@mail.gmail.com>
	<625385e30903271340r4307dd64hd691e4dac6fa46ff@mail.gmail.com>
	<20090327211430.GV47757@bolthole.com>
	<ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>
	<20090329140924.GD30131@bolthole.com>
	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>
	<20090329220733.GM69158@bolthole.com>
	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>
	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>
	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
Message-ID: <6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>

not by me anyway :)
dago, you mind giving me write access to mediawiki?

rupert.


On Tue, Mar 31, 2009 at 12:24, Maciej (Matchek) Blizinski
<maciej at opencsw.org> wrote:
> On Tue, Mar 31, 2009 at 12:39 AM, rupert THURNER <rupert at opencsw.org> wrote:
>>> When I set up "gar" on SourceForge Trac was not available as
>>> hosted application. As it is there now we can happily switch
>>> over from the standard SF bugtracker and MediaWiki to GAR.
>>>
>>> Rupert, Maciej, if you could join in on moving the stuff over :-)
>>
>> for the wiki: the mGAR things are copied. you want to keep the gar stuff as is?
>
> Can already migrated Mediawiki pages be replaced with links to their
> Trac locations?
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From dam at opencsw.org  Wed Apr  1 08:37:21 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 08:37:21 +0200
Subject: [csw-maintainers] BO buildfarm down
In-Reply-To: <49D29240.2070105@opencsw.org>
References: <49D29240.2070105@opencsw.org>
Message-ID: <B65657D6-5AB5-4253-B36E-6B15E6CC4B4A@opencsw.org>

Hi,

Am 31.03.2009 um 23:59 schrieb Roger H?kansson:
> hson at build8s :~/dev/mgar/pkg/zlib/trunk > gmake build
> bash: fork: Not enough space

Am 01.04.2009 um 07:25 schrieb rupert THURNER:
> currently i get:
> ~$ ssh rupert at login.opencsw.org
> PTY allocation request failed on channel 0
> Received disconnect from 213.178.77.178: 2: fork failed: Not enough  
> space

it seems like the T5220 was overloaded due to lack of Ram. As I
already wrote there will be a memory upgrade soon. We are working
on fixing this issue.

In the meantime please use build8s.go.opencsw.org, all accounts should
be also configured there.


Best regards

   -- Dago

From dam at opencsw.org  Wed Apr  1 09:12:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 09:12:08 +0200
Subject: [csw-maintainers] BO buildfarm down
In-Reply-To: <B65657D6-5AB5-4253-B36E-6B15E6CC4B4A@opencsw.org>
References: <49D29240.2070105@opencsw.org>
	<B65657D6-5AB5-4253-B36E-6B15E6CC4B4A@opencsw.org>
Message-ID: <C0BBBFAB-6239-4C18-8E53-98BC55DF603E@opencsw.org>

Hi,

Am 01.04.2009 um 08:37 schrieb Dagobert Michelsen:
> it seems like the T5220 was overloaded due to lack of Ram. As I
> already wrote there will be a memory upgrade soon. We are working
> on fixing this issue.

There were a number of problems:

1. Hudson got too busy:
Apr  1 09:00:40 ncsw tmpfs: WARNING: /zones/hudson/root/tmp: File  
system full, swap space limit exceeded

2. Bug #6682528 was hit: memory leak in nfsmapid if resolv_init() is  
called

3. Overall machine usage due to heavy packaging activity

All this in sum slowed the machine down. As a temporary workaround
I turned off the Hudson zone (it isn't productive yet, Trygve, yes?).

Additionally, to make something good out of the bad I use the
unplanned downtime to bring the machine to a current patchlevel
and updated all CSW packages to current/.


The machine will be available again in a few hours. I let you know.


Sorry for the inconvenience

   -- Dago


From skayser at opencsw.org  Wed Apr  1 10:22:27 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 01 Apr 2009 10:22:27 +0200
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
References: <ce97543b0903260441k26550f94r96a1f6bb6c95755d@mail.gmail.com>	<625385e30903271340r4307dd64hd691e4dac6fa46ff@mail.gmail.com>	<20090327211430.GV47757@bolthole.com>	<ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>	<20090329140924.GD30131@bolthole.com>	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>	<20090329220733.GM69158@bolthole.com>	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
	<6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
Message-ID: <49D32443.1030804@opencsw.org>

rupert THURNER wrote:
> not by me anyway :)
> dago, you mind giving me write access to mediawiki?

Done, you now have editor rights.

Dago, what about changing gar.opencsw.org so that it redirects to the
Trac installation, now that Rupert has moved the remaining up-to-date docs?

Sebastian


From dam at opencsw.org  Wed Apr  1 10:51:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 10:51:07 +0200
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <49D32443.1030804@opencsw.org>
References: <ce97543b0903260441k26550f94r96a1f6bb6c95755d@mail.gmail.com>	<625385e30903271340r4307dd64hd691e4dac6fa46ff@mail.gmail.com>	<20090327211430.GV47757@bolthole.com>	<ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>	<20090329140924.GD30131@bolthole.com>	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>	<20090329220733.GM69158@bolthole.com>	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
	<6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
	<49D32443.1030804@opencsw.org>
Message-ID: <631F4BE9-F055-41CE-972F-1C44202C19F1@opencsw.org>

Hi Sebastian,

Am 01.04.2009 um 10:22 schrieb Sebastian Kayser:
> Dago, what about changing gar.opencsw.org so that it redirects to the
> Trac installation, now that Rupert has moved the remaining up-to- 
> date docs?

Done.


Best regards

   -- Dago


From james at opencsw.org  Wed Apr  1 11:03:20 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 09:03:20 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
Message-ID: <20090401.9032000.316239962@gyor.oxdrove.co.uk>

On 01/04/09, 00:52:31, Chad Harp <harpchad at opencsw.org> wrote regarding Re:
[csw-maintainers] Runtime linker pathes in GAR:

> Phil, can checkpkg detect some of the common cases?  I'm thinking at
> least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those
> would indicate a bad RUNPATHQUOTE value.  James can probably give us
> an idea of what he's coming across the most.

/opt/csw/lib/\$ISALIST
/opt/csw/lib/\SALIST
/opt/csw/lib/\$$ISALIST
/opt/csw/lib/SALIST
/opt/csw/lib/-R/


but checking for bad paths is the wrong way to find errors.  I check
for good paths (directory exists and/or is valid dynamic linker token)
and throw up anything else.


The nice thing about using LD_OPTIONS is it doesn't need escaping
(more than initially) and can put $ISALIST first without libtool
interfering.




James.


From dam at opencsw.org  Wed Apr  1 12:05:35 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 12:05:35 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <20090401.9032000.316239962@gyor.oxdrove.co.uk>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
Message-ID: <BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>

Hi James,

Am 01.04.2009 um 11:03 schrieb James Lee:
> On 01/04/09, 00:52:31, Chad Harp <harpchad at opencsw.org> wrote  
> regarding Re:
> [csw-maintainers] Runtime linker pathes in GAR:
>
>> Phil, can checkpkg detect some of the common cases?  I'm thinking at
>> least /opt/csw/bin/\$ISALIST and /opt/csw/bin/SALIST since those
>> would indicate a bad RUNPATHQUOTE value.  James can probably give us
>> an idea of what he's coming across the most.
>
> /opt/csw/lib/\$ISALIST
> /opt/csw/lib/\SALIST
> /opt/csw/lib/\$$ISALIST
> /opt/csw/lib/SALIST
> /opt/csw/lib/-R/
>
> but checking for bad paths is the wrong way to find errors.  I check
> for good paths (directory exists and/or is valid dynamic linker token)
> and throw up anything else.
>
> The nice thing about using LD_OPTIONS is it doesn't need escaping
> (more than initially) and can put $ISALIST first without libtool
> interfering.

Maybe this is the one occasion where using LD_OPTIONS does
more good than bad? Just for '-R'?


Best regards

   -- Dago


From james at opencsw.org  Wed Apr  1 12:20:55 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 10:20:55 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
Message-ID: <20090401.10205500.1310557855@gyor.oxdrove.co.uk>

On 01/04/09, 11:05:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
Re: [csw-maintainers] Runtime linker pathes in GAR:

> > The nice thing about using LD_OPTIONS is it doesn't need escaping
> > (more than initially) and can put $ISALIST first without libtool
> > interfering.

> Maybe this is the one occasion where using LD_OPTIONS does
> more good than bad? Just for '-R'?

I suspect that on these occasions LD_OPTIONS wasn't used.




James.







From james at opencsw.org  Wed Apr  1 12:43:36 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 10:43:36 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <20090401.10205500.1310557855@gyor.oxdrove.co.uk>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
	<20090401.10205500.1310557855@gyor.oxdrove.co.uk>
Message-ID: <20090401.10433600.467498167@gyor.oxdrove.co.uk>

On 01/04/09, 11:20:55, James Lee <james at opencsw.org> wrote regarding Re: 
[csw-maintainers] Runtime linker pathes in GAR:

> On 01/04/09, 11:05:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding
> Re: [csw-maintainers] Runtime linker pathes in GAR:

> > > The nice thing about using LD_OPTIONS is it doesn't need escaping
> > > (more than initially) and can put $ISALIST first without libtool
> > > interfering.

> > Maybe this is the one occasion where using LD_OPTIONS does
> > more good than bad? Just for '-R'?

> I suspect that on these occasions LD_OPTIONS wasn't used.

PS... Sorry I am muddling occasions, I'm thinking the subject occasion
is "we have a lot of bad RPATHs" but we've moved to the topic to "how
to easily set an $ in the first element of an RPATH", and yes LD_OPTIONS
does it.



James.


From dam at opencsw.org  Wed Apr  1 13:21:50 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 13:21:50 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <20090401.10433600.467498167@gyor.oxdrove.co.uk>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
	<20090401.10205500.1310557855@gyor.oxdrove.co.uk>
	<20090401.10433600.467498167@gyor.oxdrove.co.uk>
Message-ID: <0A949260-DD06-4FC0-A9EB-35785EC4411F@opencsw.org>

Hi James,

Am 01.04.2009 um 12:43 schrieb James Lee:
> On 01/04/09, 11:20:55, James Lee <james at opencsw.org> wrote regarding  
> Re:
> [csw-maintainers] Runtime linker pathes in GAR:
>
>> On 01/04/09, 11:05:35, Dagobert Michelsen <dam at opencsw.org> wrote  
>> regarding
>> Re: [csw-maintainers] Runtime linker pathes in GAR:
>
>>>> The nice thing about using LD_OPTIONS is it doesn't need escaping
>>>> (more than initially) and can put $ISALIST first without libtool
>>>> interfering.
>
>>> Maybe this is the one occasion where using LD_OPTIONS does
>>> more good than bad? Just for '-R'?
>
>> I suspect that on these occasions LD_OPTIONS wasn't used.

I disabled the use of LD_OPTIONS after the last discussion that it
was a Bad Thing(tm).

> PS... Sorry I am muddling occasions, I'm thinking the subject occasion
> is "we have a lot of bad RPATHs" but we've moved to the topic to "how
> to easily set an $ in the first element of an RPATH", and yes  
> LD_OPTIONS
> does it.

Ok, is it agreed then that using LD_OPTIONS for -R is a Good Thing(tm)?
I'll remove the -R from LDFLAGS and put enable LD_OPTIONS for this  
option
only again.


Best regards

   -- Dago


From skayser at opencsw.org  Wed Apr  1 13:23:43 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 01 Apr 2009 13:23:43 +0200
Subject: [csw-maintainers] mGAR cpan builds: satisfying non-installed
	dependencies?
Message-ID: <49D34EBF.2020702@opencsw.org>

Hi,

are there any thoughts underway to enable straight-forward builds of
CPAN packages that have not-yet-installed dependencies (which need to be
built also)?

>From looking at the category.mk for the cpan category it seems as if
this had been working with spool* directories of GAR, but as mGAR has a
separate $(DESTDIR) for each package it doesn't work with mGAR.

$ svn pg svn:externals .
gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2

$ ggrep -C1 ^PERL5LIB gar/categories/cpan/category.mk
# Enable scripts to see prereqs
PERL5LIB  = $(DESTDIR)$(libdir)/perl/csw
PERL5LIB := $(PERL5LIB):$(DESTDIR)$(datadir)/perl/csw
export PERL5LIB

I just built ack [1] which relied on File-Next (not yet in the catalog,
nor on the build farm). To satisfy this dependency I built File-Next,
manually copied it to a temporary directory, and pointed PERL5LIB to
this directory when building ack.

This worked, but maybe someone has ideas on how to tweak GAR so that
this installation and finding of dependencies becomes an automated
process :)

Sebastian

[1] http://betterthangrep.com/


From dam at opencsw.org  Wed Apr  1 14:32:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 14:32:55 +0200
Subject: [csw-maintainers] Update: BO Buildfarm
Message-ID: <63668C3B-E4A2-49D2-94E9-399C1CFC605F@opencsw.org>

Hi,

the real cause for the downtime was bldcat:

> Inspecting ./gcc4g++-4.3.3,REV=2009.03.31-SunOS5.8-sparc-CSW.pkg.gz
> Use of uninitialized value in concatenation (.) or string at /opt/ 
> csw/bin/bldcat line 81.
> Use of uninitialized value in concatenation (.) or string at /opt/ 
> csw/bin/bldcat line 81.
> Can't open /tmp/bldcat.1301.1238588867//pkginfo: No such file or  
> directory at /opt/csw/bin/bldcat line 81.

/tmp was filled with old stuff. There should be a number of changes
to bldcat. Peter:

- change string from \s+ to [^-]+
- install trap handler so that all temporary files are cleaned up on
   exit, regardless why.

Until this is fixed I disabled catalog generation. I'll look that up as
soon as I can. There are still a number of patches running in. Please
stand by.


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  1 15:11:05 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 15:11:05 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <F8FC4A58-F0D9-4D5B-A574-080E0028CB8F@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<F8FC4A58-F0D9-4D5B-A574-080E0028CB8F@opencsw.org>
Message-ID: <EBC98680-E60F-4D7D-A236-560AE199E15A@opencsw.org>

Hi Chad,

Am 01.04.2009 um 01:47 schrieb Chad Harp:
> I assume RUNPATHQUOTE=3 is also valid?  I have one package where  
> RPATH=/opt/csw/lib/SALIST with =2, looks like $I got expanded.

No, that is not valid, just 1 and 2. It looks like this whole
system needs a thorough rewrite.


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  1 15:18:35 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 15:18:35 +0200
Subject: [csw-maintainers] Another cyclic dependency
Message-ID: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>

Hi,

and another interesting dependency chain, this time three
packages long:

> Trying to install dependancy commons_collect
> No existing install of CSWajccollect found. Installing...
> Pre-existing local file commons_collect-3.2.1,REV=2009.03.26- 
> SunOS5.8-all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_config
> No existing install of CSWajcconfig found. Installing...
> Pre-existing local file commons_config-1.6,REV=2009.03.27-SunOS5.8- 
> all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_digester
> No existing install of CSWajcdigester found. Installing...
> Pre-existing local file commons_digester-2.0,REV=2009.03.26-SunOS5.8- 
> all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_collect
> No existing install of CSWajccollect found. Installing...
> Pre-existing local file commons_collect-3.2.1,REV=2009.03.26- 
> SunOS5.8-all-CSW.pkg.gz matches checksum
> Keeping existing file
> Analysing special files...
> Trying to install dependancy commons_config
> No existing install of CSWajcconfig found. Installing...
> Pre-existing local file commons_config-1.6,REV=2009.03.27-SunOS5.8- 
> all-CSW.pkg.gz matches checksum
...


Best regards

   -- Dago


From william at wbonnet.net  Wed Apr  1 15:16:24 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 15:16:24 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
Message-ID: <49D36928.6060801@wbonnet.net>

Hi Dago

Tanks for the information, i fix it tonight

Cheers

> Hi,
>
> and another interesting dependency chain, this time three
> packages long:
>
>> Trying to install dependancy commons_collect
>> No existing install of CSWajccollect found. Installing...
>> Pre-existing local file 
>> commons_collect-3.2.1,REV=2009.03.26-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_config
>> No existing install of CSWajcconfig found. Installing...
>> Pre-existing local file 
>> commons_config-1.6,REV=2009.03.27-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_digester
>> No existing install of CSWajcdigester found. Installing...
>> Pre-existing local file 
>> commons_digester-2.0,REV=2009.03.26-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_collect
>> No existing install of CSWajccollect found. Installing...
>> Pre-existing local file 
>> commons_collect-3.2.1,REV=2009.03.26-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
>> Keeping existing file
>> Analysing special files...
>> Trying to install dependancy commons_config
>> No existing install of CSWajcconfig found. Installing...
>> Pre-existing local file 
>> commons_config-1.6,REV=2009.03.27-SunOS5.8-all-CSW.pkg.gz matches 
>> checksum
> ...
>
>
> Best regards
>
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From james at opencsw.org  Wed Apr  1 15:24:24 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 13:24:24 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <0A949260-DD06-4FC0-A9EB-35785EC4411F@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<EB6882A3-201A-4581-933B-B7CE3A2D9FE6@opencsw.org>
	<20090401.9032000.316239962@gyor.oxdrove.co.uk>
	<BCC8DB08-5ACD-4F5C-B23A-830788981250@opencsw.org>
	<20090401.10205500.1310557855@gyor.oxdrove.co.uk>
	<20090401.10433600.467498167@gyor.oxdrove.co.uk>
	<0A949260-DD06-4FC0-A9EB-35785EC4411F@opencsw.org>
Message-ID: <20090401.13242400.652740328@gyor.oxdrove.co.uk>

On 01/04/09, 12:21:50, Dagobert Michelsen <dam at opencsw.org> wrote regarding
Re: [csw-maintainers] Runtime linker pathes in GAR:

> Ok, is it agreed then that using LD_OPTIONS for -R is a Good Thing(tm)?

LD_OPTIONS is a useful tool.  As is LD_LIBRARY_PATH for linking to
libraries in temporary locations (but don't run with it as RPATH
should know best).

The problem with LD_OPTIONS is it sets the same parameters for all
which might not be appropriate.  One of your executables might need,
e.g., /opt/csw/bdb4/lib first but maybe not all would - although,
assuming the right lib is eventually found, it only wastes the linker's
and user's time, just as those bad RPATHs cause the runtime linker to
look for files that don't exist.

Use LD_OPTIONS if it works.



James.


From james at opencsw.org  Wed Apr  1 15:40:17 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 13:40:17 GMT
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
Message-ID: <20090401.13401700.4235995713@gyor.oxdrove.co.uk>

On 01/04/09, 14:18:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
[csw-maintainers] Another cyclic dependency:


>From CSWggettextrt, needs libiconv which is in CSWiconv:

$ ldd /opt/csw/bin/ggettext
        libintl.so.8 =>  /opt/csw/lib/libintl.so.8
        libsec.so.1 =>   /lib/libsec.so.1
        libc.so.1 =>     /lib/libc.so.1
        libiconv.so.2 =>         /opt/csw/lib/libiconv.so.2
        libavl.so.1 =>   /lib/libavl.so.1
        libm.so.2 =>     /lib/libm.so.2



>From CSWiconv, needs libintl which is in CSWggettextrt:

$ ldd /opt/csw/bin/iconv
        libiconv.so.2 =>         /opt/csw/lib/i386/libiconv.so.2
        libintl.so.3 =>  /opt/csw/lib/i386/libintl.so.3
        libc.so.1 =>     /lib/libc.so.1
        libm.so.2 =>     /lib/libm.so.2


Shouldn't the installer be tolerant of dependancy loops and just 
install both when either is requested?




James.


From dam at opencsw.org  Wed Apr  1 16:04:33 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 16:04:33 +0200
Subject: [csw-maintainers] Update: BO Buildfarm available again
In-Reply-To: <63668C3B-E4A2-49D2-94E9-399C1CFC605F@opencsw.org>
References: <63668C3B-E4A2-49D2-94E9-399C1CFC605F@opencsw.org>
Message-ID: <6139E2DD-DB5F-427D-9CC1-A1D505CF36B6@opencsw.org>

Hi,

Am 01.04.2009 um 14:32 schrieb Dagobert Michelsen:
> /tmp was filled with old stuff.

I have now a workaround in place until the course is fixed.
Everything should be fine again. Please let me know if
you encounter any other issues.


Sorry for the inconvenience

   -- Dago


From william at wbonnet.net  Wed Apr  1 16:16:01 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 16:16:01 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <20090401.13401700.4235995713@gyor.oxdrove.co.uk>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
Message-ID: <49D37721.8030707@wbonnet.net>

Hi

> >From CSWggettextrt, needs libiconv which is in CSWiconv:
> >From CSWiconv, needs libintl which is in CSWggettextrt:
>   
This one can easily be solved by splitting pckage into CSWiconv and 
CSWiconvrt

Cheers


From Darin.Perusich at cognigencorp.com  Wed Apr  1 16:39:07 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 01 Apr 2009 10:39:07 -0400
Subject: [csw-maintainers] install mtx
Message-ID: <49D37C8B.20201@cognigencorp.com>

Can someone install the mtx package across the build farm?

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From harpchad at opencsw.org  Wed Apr  1 17:03:11 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 01 Apr 2009 10:03:11 -0500
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <20090401.13401700.4235995713@gyor.oxdrove.co.uk>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
Message-ID: <49D3822F.6030002@opencsw.org>

An excerpt from the inconv docs:

After installing GNU libiconv for the first time, it is recommended to 
recompile and reinstall GNU gettext, so that it can take advantage of 
libiconv.

On systems other than GNU/Linux, the iconv program will be 
internationalized only if GNU gettext has been built and installed 
before GNU libiconv. This means that the first time GNU libiconv is 
installed, we have a circular dependency between the GNU libiconv and 
GNU gettext packages, which can be resolved by building and installing 
either

     * first libiconv, then gettext, then libiconv again,

or (on systems supporting shared libraries, excluding AIX)

     * first gettext, then libiconv, then gettext again.



What a mess, so it seems it's intended and we have to live with it.  Can 
the install utils support co-dependent packages and James suggested?

James Lee wrote:
> On 01/04/09, 14:18:35, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
> [csw-maintainers] Another cyclic dependency:
> 
> 
>>From CSWggettextrt, needs libiconv which is in CSWiconv:
> 
> $ ldd /opt/csw/bin/ggettext
>         libintl.so.8 =>  /opt/csw/lib/libintl.so.8
>         libsec.so.1 =>   /lib/libsec.so.1
>         libc.so.1 =>     /lib/libc.so.1
>         libiconv.so.2 =>         /opt/csw/lib/libiconv.so.2
>         libavl.so.1 =>   /lib/libavl.so.1
>         libm.so.2 =>     /lib/libm.so.2
> 
> 
> 
>>From CSWiconv, needs libintl which is in CSWggettextrt:
> 
> $ ldd /opt/csw/bin/iconv
>         libiconv.so.2 =>         /opt/csw/lib/i386/libiconv.so.2
>         libintl.so.3 =>  /opt/csw/lib/i386/libintl.so.3
>         libc.so.1 =>     /lib/libc.so.1
>         libm.so.2 =>     /lib/libm.so.2
> 
> 
> Shouldn't the installer be tolerant of dependancy loops and just 
> install both when either is requested?
> 
> 
> 
> 
> James.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From james at opencsw.org  Wed Apr  1 17:06:01 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 15:06:01 GMT
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <49D37721.8030707@wbonnet.net>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>
	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
	<49D37721.8030707@wbonnet.net>
Message-ID: <20090401.15060100.1316498277@gyor.oxdrove.co.uk>

On 01/04/09, 15:16:01, William Bonnet <william at wbonnet.net> wrote regarding
Re: [csw-maintainers] Another cyclic dependency:

> > >From CSWggettextrt, needs libiconv which is in CSWiconv:
> > >From CSWiconv, needs libintl which is in CSWggettextrt:
> >
> This one can easily be solved by splitting pckage into CSWiconv and
> CSWiconvrt

I think you'll find the libraries are also interdependent.  Granted
one doesn't "run" libraries and superior packages can pull both as
has been happening since 2003.

http://www.opencsw.org/bugtrack/view.php?id=173

It can be easily solved without splitting by breaking the depend loop
on install which would also handle all cases not just this one.



James.


From william at wbonnet.net  Wed Apr  1 17:12:41 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 17:12:41 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <49D3822F.6030002@opencsw.org>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>
	<49D3822F.6030002@opencsw.org>
Message-ID: <49D38469.2090802@wbonnet.net>

Hi
>
>
> What a mess, so it seems it's intended and we have to live with it.  
> Can the install utils support co-dependent packages and James suggested?
BTW it is the same kind of mess for some java libs...


cheers


From Darin.Perusich at cognigencorp.com  Wed Apr  1 17:24:16 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 01 Apr 2009 11:24:16 -0400
Subject: [csw-maintainers] install star
Message-ID: <49D38720.2040703@cognigencorp.com>

Can star be installed across the build farm please?

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From harpchad at opencsw.org  Wed Apr  1 17:27:32 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 01 Apr 2009 10:27:32 -0500
Subject: [csw-maintainers] glib 2.20.0 in testing
Message-ID: <49D387E4.5020405@opencsw.org>

It's a manual download/install until the cataloger is fixed:

# glib2-2.20.0,REV=2009.03.31-SunOS5.8-i386-CSW.pkg.gz
# glib2-2.20.0,REV=2009.03.31-SunOS5.8-sparc-CSW.pkg.gz
# glib2_devel-2.20.0,REV=2009.03.31-SunOS5.8-i386-CSW.pkg.gz
# glib2_devel-2.20.0,REV=2009.03.31-SunOS5.8-sparc-CSW.pkg.gz


From amaier at opencsw.org  Wed Apr  1 17:58:36 2009
From: amaier at opencsw.org (Alexander Maier)
Date: Wed, 1 Apr 2009 17:58:36 +0200
Subject: [csw-maintainers] Font packages
In-Reply-To: <ce97543b0903310320n2ef5c363ica24fb750a880eea@mail.gmail.com>
References: <ce97543b0903310320n2ef5c363ica24fb750a880eea@mail.gmail.com>
Message-ID: <8642B5FA-EA51-4702-A07B-32641C37B1D4@opencsw.org>

Am 31.03.2009 um 12:20 schrieb Maciej (Matchek) Blizinski:

> How do we go about adding font packages? Let's suppose I have a font
> what I want to package. Let it be Terminus[1], for the sake of the
> exercise. Can it be somehow integrated with Sun-provided X server in
> Solaris 10? Are there established practices about including fonts in
> OpenCSW?
>


I think one have to distinguish between fonts for "classic X apps"  
with local/remote xfs and fonts for apps using libfreetype/fontconfig  
(based on GTK or Qt).

In the latter case I would suggest to put these fonts into /opt/csw/ 
share/fonts (e.g. Microsoft TrueType fonts or Red Hat Liberation fonts).

Terminus seems to be a classic X11/Motif font, so it would probably be  
better off somewhere below /opt/csw/X11 (as Phil has suggested).

-Alexander




From dam at opencsw.org  Wed Apr  1 18:11:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 18:11:40 +0200
Subject: [csw-maintainers] install mtx
In-Reply-To: <49D37C8B.20201@cognigencorp.com>
References: <49D37C8B.20201@cognigencorp.com>
Message-ID: <8812DEA1-EC3F-449B-ABEB-064DB1375471@opencsw.org>

Hi Darin,

Am 01.04.2009 um 16:39 schrieb Darin Perusich:
> Can someone install the mtx package across the build farm?

Am 01.04.2009 um 17:24 schrieb Darin Perusich:
> Can star be installed across the build farm please?

Doing this now. Please mail to
   buildfarm at lists.opencsw.org
next time as there are possibly buildfarm admins who are not on
maintainers at .


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  1 18:26:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 18:26:17 +0200
Subject: [csw-maintainers] install mtx
In-Reply-To: <8812DEA1-EC3F-449B-ABEB-064DB1375471@opencsw.org>
References: <49D37C8B.20201@cognigencorp.com>
	<8812DEA1-EC3F-449B-ABEB-064DB1375471@opencsw.org>
Message-ID: <D7B0F3DD-D5EE-4F1A-8687-1A2FD51733A0@opencsw.org>

Hi,

Am 01.04.2009 um 18:11 schrieb Dagobert Michelsen:
> Am 01.04.2009 um 16:39 schrieb Darin Perusich:
>> Can someone install the mtx package across the build farm?
>
> Am 01.04.2009 um 17:24 schrieb Darin Perusich:
>> Can star be installed across the build farm please?
>
> Doing this now.

Done.


Best regards

   -- Dago


From phil at bolthole.com  Wed Apr  1 18:45:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 1 Apr 2009 09:45:45 -0700
Subject: [csw-maintainers] On-line project resources
In-Reply-To: <631F4BE9-F055-41CE-972F-1C44202C19F1@opencsw.org>
References: <ce97543b0903290330i1f602fa1p117c8ba1975b8459@mail.gmail.com>
	<20090329140924.GD30131@bolthole.com>
	<ce97543b0903291111y5e583444obaa33be7f5841101@mail.gmail.com>
	<20090329220733.GM69158@bolthole.com>
	<1F8FD5D5-7410-4DEB-B548-FF03BEC89B69@opencsw.org>
	<6af4270903301639t7ecb22f3rd4613e3b31e35f80@mail.gmail.com>
	<ce97543b0903310324g14c4f505n9e5aa3b3310e9961@mail.gmail.com>
	<6af4270903312229w5ab996d7xcdff59894e7b542b@mail.gmail.com>
	<49D32443.1030804@opencsw.org>
	<631F4BE9-F055-41CE-972F-1C44202C19F1@opencsw.org>
Message-ID: <20090401164545.GA19124@bolthole.com>

On Wed, Apr 01, 2009 at 10:51:07AM +0200, Dagobert Michelsen wrote:
> Hi Sebastian,
>
> Am 01.04.2009 um 10:22 schrieb Sebastian Kayser:
>> Dago, what about changing gar.opencsw.org so that it redirects to the
>> Trac installation, now that Rupert has moved the remaining up-to-date 
>> docs?
>
> Done.

i glanced at gar.opencsw.org.

It mentions "mGar is completely copied, GAR stuff not yet completely."
but it doesnt say what "mGar" *is*.

May i sugest that you now officially "deprecate" and devalue references to
the old gar version(s). Move all of that stuff to a "Historical" category at
the bottom.

having "GAR versions" as the #2 link under "Documentation", just adds
unneccessary confusion, in my opinion.

I would also like to request that the "trac admins who aren't Dago",
finally give us an up to date "Getting started with GAR" wiki page there?
;-)
[Dago has been meaning to for a long time, but he's too busy doing all
 the other stuff he does for us. Help him out! :-) ]


From skayser at opencsw.org  Wed Apr  1 19:11:37 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 01 Apr 2009 19:11:37 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <49D3A049.2040102@opencsw.org>

Dagobert Michelsen wrote:
> James is currently in the tedious task of reviewing all
> runtime linker pathes from the packages.

Just for clarification: is it only about correcting those incorrectly
quoted RUNPATHs or are we supposed to also adjust the RUNPATH (via
LINKER_FLAGS or similar) depending on what our applications use (i.e.
drop .../$ISALIST or even /opt/csw/lib completely)?

I personally would prefer to just have GAR corrected WRT to the $ISALIST
quoting issue and not add more intelligence to the build description,
except maybe for applications where the startup delay might have a
performance impact.

Sebastian


From harpchad at opencsw.org  Wed Apr  1 19:15:47 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 01 Apr 2009 12:15:47 -0500
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
Message-ID: <49D3A143.90709@opencsw.org>

I think I need a better understanding of what I should be doing. 
Several of my packages have path problems due to quoting.  Should I be 
looking for the correct combinations of variable to fix it, or will GAR 
be fixed to deal with it?


From mwatters at opencsw.org  Wed Apr  1 19:33:48 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 01 Apr 2009 12:33:48 -0500
Subject: [csw-maintainers] chrpath now in Testing
Message-ID: <49D3A57C.4070201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ladies and Gentlemen,

I am pleased to announce a new package I copied into testing.

chrpath
This will allow changing the RPATH and RUNPATH values in binaries and libraries.
I have done testing on both sparc and x86 platforms and all seems well.

The main limitation I have found is you can not "grow" the path any larger the
the initial size.

for example:
if the original has RPATH of /opt/csw/lib/SALIST:/opt/csw/lib

you could NOT make it /opt/csw/lib/$ISALIST:/opt/csw/lib
but you CAN make it /opt/csw/lib/$ISALIST

example2:
if the original has RPATH of /opt/csw/lib/\\$ISALIST:/opt/csw/lib:/opt/csw/lib
you can make it /opt/csw/lib/$ISALIST:/opt/csw/lib

you don't need to escape if you use single quotes
chrpath -r '/opt/csw/lib/$ISALIST:/opt/csw/lib' filename
will work beautifully

I had to dig around for the source as the link from FSF is broken.
I downloaded the source from Debian's repository. but the package points
to the FSF site.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknTpXwACgkQLrhmsXMSLxek6QCeKhTMINMz+nrtoQijdCT5tNMd
Yg4An1hK1NmHNzesnACRO7k4RsQ2i7ZI
=Pm8U
-----END PGP SIGNATURE-----


From james at opencsw.org  Wed Apr  1 20:19:08 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 18:19:08 GMT
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <49D3A049.2040102@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<49D3A049.2040102@opencsw.org>
Message-ID: <20090401.18190800.2943361049@gyor.oxdrove.co.uk>

On 01/04/09, 18:11:37, Sebastian Kayser <skayser at opencsw.org> wrote
regarding Re: [csw-maintainers] Runtime linker pathes in GAR:

> Just for clarification: is it only about correcting those incorrectly
> quoted RUNPATHs or are we supposed to also adjust the RUNPATH (via
> LINKER_FLAGS or similar) depending on what our applications use (i.e.
> drop .../$ISALIST or even /opt/csw/lib completely)?

You can drop /opt/csw/lib when behind /opt/csw/lib/$ISALIST
because the arch dir for the minimum arch is a link back to
/opt/csw/lib, provided by CSWcommon.

$ ls -l /opt/csw/lib/sparcv8
lrwxrwxrwx   1 root     other          1 Jun 22  2008 
/opt/csw/lib/sparcv8
-> .


This means there will be a match before the end of the $ISALIST.


$ isalist
sparcv9+vis sparcv9 sparcv8plus+vis sparcv8plus sparcv8 sparcv8-fsmuld
sparcv7 sparc

would match at sparcv8 even with no arch libs, so skipping sparcv7
etc. which are never in CSW anyway.  Hence never needs /opt/csw/lib
after.



James.


From james at opencsw.org  Wed Apr  1 20:26:31 2009
From: james at opencsw.org (James Lee)
Date: Wed, 01 Apr 2009 18:26:31 GMT
Subject: [csw-maintainers] chrpath now in Testing
In-Reply-To: <49D3A57C.4070201@opencsw.org>
References: <49D3A57C.4070201@opencsw.org>
Message-ID: <20090401.18263100.3812230620@gyor.oxdrove.co.uk>

On 01/04/09, 18:33:48, Mike Watters <mwatters at opencsw.org> wrote regarding
[csw-maintainers] chrpath now in Testing:

> chrpath
> This will allow changing the RPATH and RUNPATH values in binaries and
> libraries.
> I have done testing on both sparc and x86 platforms and all seems well.

> The main limitation I have found is you can not "grow" the path any
> larger the the initial size.

> for example:
> if the original has RPATH of /opt/csw/lib/SALIST:/opt/csw/lib

> you could NOT make it /opt/csw/lib/$ISALIST:/opt/csw/lib
> but you CAN make it /opt/csw/lib/$ISALIST

That's fine, as per my previous message, /opt/csw/lib is not needed
after /opt/csw/lib/$ISALIST.

Lots of RPATHS have duplicate directories too so often you will have
the slack to add the missing "$I".




James.


From dam at opencsw.org  Wed Apr  1 21:35:25 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Apr 2009 21:35:25 +0200
Subject: [csw-maintainers] Runtime linker pathes in GAR
In-Reply-To: <49D3A143.90709@opencsw.org>
References: <33EBB8DF-3A47-4451-8793-B14A9D923519@opencsw.org>
	<49D3A143.90709@opencsw.org>
Message-ID: <7673312B-D309-4AEC-B869-D2C0D3261BB9@opencsw.org>

Hi Chad,

Am 01.04.2009 um 19:15 schrieb Chad Harp:
> I think I need a better understanding of what I should be doing.
> Several of my packages have path problems due to quoting.  Should
> I be looking for the correct combinations of variable to fix it,
> or will GAR be fixed to deal with it?

Ok, now here is the rationale why things are as they are now:

- /opt/csw/lib/$ISALIST is needed if and only if libraries are
   linked that may contain optimizations. Otherwise it is best to
   avoid $ISALIST completely. This can be done by setting
     NOISALIST = 1

- I put -R for runtime linker path in LDFLAGS because in the last
   discussion LD_OPTIONS was considered bad, as it is always applied
   first in linking regardless of other flags. Unfortunately these
   flags are passwd through multiple commands, each with evaluation
   of the argument. Now $ISALIST looks like a variable and must be
   shielded from evaluation by quoting. The amount of quoting depends
   on the number of evaluation, e. g. if libtool is used, a custom
   Makefile, etc.
   The number of quotings can be adjusted by the simple parameter
     RUNPATHQUOTE = 1    for single evaluation
     RUNPATHQUOTE = 2    for double evaluation
   Unfortunately this obviously proved to be not sufficient as the
   bug reports from James show.

   (BTW, if you are interested: Making the literal '$' as prefix is
     _Q = \\\$$\$$       for RUNPATHQUOTE = 1
   and
     _Q = \\\\\\\$$\$$   for RUNPATHQUOTE = 2
   in gar.conf.mk).

- Moving -R to LD_OPTIONS would make things considerable easier,
   because the number of evaluations is fixed as it is not processed
   by multiple commands. It would just be needed to decide if
   $ISALIST should be in or not. NOISALIST is already there for this.
   This would conflict with packages already using LD_OPTIONS:
     amarok
     cups
     daimonin
     exim
     gftp
     gphoto2
     hatari
     hypermail
     kdesvn
     kile
     libgphoto2
     lyx
     mysql-python
     netsnmp
     slrn
     tsclient
     w3m
   To be consistent, EXTRA_LD_OPTIONS should be introduced and the
   Makefiles adjusted accordingly when the GAR modifications have
   been made.


So the GAR integration is possible, but needs some experimentation
in quoting and best use.


Best regards

   -- Dago
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090401/6f35ccbe/attachment.html>

From william at wbonnet.net  Wed Apr  1 21:56:56 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Apr 2009 21:56:56 +0200
Subject: [csw-maintainers] Another cyclic dependency
In-Reply-To: <49D38469.2090802@wbonnet.net>
References: <409198A0-7BA4-413D-9B30-B808D84AC688@opencsw.org>	<20090401.13401700.4235995713@gyor.oxdrove.co.uk>	<49D3822F.6030002@opencsw.org>
	<49D38469.2090802@wbonnet.net>
Message-ID: <49D3C708.80504@wbonnet.net>

Hi

I removed dependency from digester to collections. It should "break" the 
loop. I submited the fix to Phil.

Cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Wed Apr  1 22:42:23 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 1 Apr 2009 13:42:23 -0700
Subject: [csw-maintainers] Wanted: opensolaris python lovers
Message-ID: <20090401204223.GF19124@bolthole.com>


So... Anyone running opensolaris, in addition to the regular stuff?
Love python?

I'm thinking it would be a very good thing, if we offered an IPS option for
getting our packages.

However, the existing "automated converter mechanism" sun offers sucks. It totally
misses out on postinstall stuff. (among other drawbacks)

There is apparently a mechanism in IPS, called an "actuator", that
sounds to be a perfect way to trigger postinstall scripts.
Theree's just the small wrinkle, that the IPS team refuses to make people's
lives easier and saner by delivering a pre-written actuator for the task.

So, in the continuing spirit of us taking on what Sun is too short-sighted
to do... Would anyone be interested in augmenting the existing
SVR4-to-IPS converter, to deal with this as well?

Since all IPS related stuff is python, this person would need to have a
good familiarity with python, as well as running opensolaris somewhere.

The downside: a decent-sized chunk of work and responsability
The upside:   gobs of fame and glory ;-)



From mwatters at opencsw.org  Wed Apr  1 23:47:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 01 Apr 2009 16:47:33 -0500
Subject: [csw-maintainers] python 2.6.1 now in testing
Message-ID: <49D3E0F5.2060809@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:

	bug fix: 3510 'import curses' fails in current Python...
			forced curses to use the library in /opt/csw
	bug fix: 3579 RPATH stuff
			added NOISALIST = 1
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknT4PUACgkQLrhmsXMSLxeP6wCeMvcjju/oNRtKUky+XipTnA2O
+50AoOPPCfkd5PhSaysD+VeaK6JeLuPp
=Y8/M
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr  2 00:07:20 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 01 Apr 2009 17:07:20 -0500
Subject: [csw-maintainers] mysql-python now in testing
Message-ID: <49D3E598.60303@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	isalist fix

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknT5ZgACgkQLrhmsXMSLxepuACfbjpI6klFYyGdidU4iFhpwfdy
6p0An3CDd08aGv5WgVf3mbVP2xy6nDeU
=d5Fp
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Apr  2 08:52:38 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 2 Apr 2009 08:52:38 +0200
Subject: [csw-maintainers] mGAR cpan builds: satisfying non-installed
	dependencies?
In-Reply-To: <49D34EBF.2020702@opencsw.org>
References: <49D34EBF.2020702@opencsw.org>
Message-ID: <1BD7044E-489F-4EB7-BCF5-3907AFC782F9@opencsw.org>

Hi Sebastian,

Am 01.04.2009 um 13:23 schrieb Sebastian Kayser:
> are there any thoughts underway to enable straight-forward builds of
> CPAN packages that have not-yet-installed dependencies (which need  
> to be
> built also)?
>
>> From looking at the category.mk for the cpan category it seems as if
> this had been working with spool* directories of GAR, but as mGAR  
> has a
> separate $(DESTDIR) for each package it doesn't work with mGAR.
>
> $ svn pg svn:externals .
> gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2
>
> $ ggrep -C1 ^PERL5LIB gar/categories/cpan/category.mk
> # Enable scripts to see prereqs
> PERL5LIB  = $(DESTDIR)$(libdir)/perl/csw
> PERL5LIB := $(PERL5LIB):$(DESTDIR)$(datadir)/perl/csw
> export PERL5LIB
>
> I just built ack [1] which relied on File-Next (not yet in the  
> catalog,
> nor on the build farm). To satisfy this dependency I built File-Next,
> manually copied it to a temporary directory, and pointed PERL5LIB to
> this directory when building ack.

The official way is to build the dependency, release it, have it
installed on the farm and continue with the next package.

Maybe we find in October ("CPAN Coverage") a more viable solution.
I guess with this runtime pathes and the stable release there are
other things to be done first...


Best regards

   -- Dago


From skayser at opencsw.org  Thu Apr  2 14:22:34 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 02 Apr 2009 14:22:34 +0200
Subject: [csw-maintainers] mGAR cpan builds: satisfying non-installed
 dependencies?
In-Reply-To: <1BD7044E-489F-4EB7-BCF5-3907AFC782F9@opencsw.org>
References: <49D34EBF.2020702@opencsw.org>
	<1BD7044E-489F-4EB7-BCF5-3907AFC782F9@opencsw.org>
Message-ID: <49D4AE0A.80102@opencsw.org>

Dagobert Michelsen wrote:
> Am 01.04.2009 um 13:23 schrieb Sebastian Kayser:
>> are there any thoughts underway to enable straight-forward builds of
>> CPAN packages that have not-yet-installed dependencies (which need  
>> to be
>> built also)?
>>
>>> From looking at the category.mk for the cpan category it seems as if
>> this had been working with spool* directories of GAR, but as mGAR  
>> has a
>> separate $(DESTDIR) for each package it doesn't work with mGAR.
>>
>> $ svn pg svn:externals .
>> gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2
>>
>> $ ggrep -C1 ^PERL5LIB gar/categories/cpan/category.mk
>> # Enable scripts to see prereqs
>> PERL5LIB  = $(DESTDIR)$(libdir)/perl/csw
>> PERL5LIB := $(PERL5LIB):$(DESTDIR)$(datadir)/perl/csw
>> export PERL5LIB
>>
>> I just built ack [1] which relied on File-Next (not yet in the  
>> catalog,
>> nor on the build farm). To satisfy this dependency I built File-Next,
>> manually copied it to a temporary directory, and pointed PERL5LIB to
>> this directory when building ack.
> 
> The official way is to build the dependency, release it, have it
> installed on the farm and continue with the next package.
> 
> Maybe we find in October ("CPAN Coverage") a more viable solution.
> I guess with this runtime pathes and the stable release there are
> other things to be done first...

Indeed. Was just wondering whether there might have been a more elegant
way to do it ATM. Thanks for the response.

Sebastian


From dam at opencsw.org  Thu Apr  2 14:22:37 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 2 Apr 2009 14:22:37 +0200
Subject: [csw-maintainers] RPATH in GAR
Message-ID: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>

Hi,

I noticed that some of you are already running ahead working
around the misconceptions of the current build system and
fixing runtime linker pathes superfast. Thanks for your
engagement! However, I think a more general solution in GAR
itself is a more viable solution, both in terms of
standardization and also easier to handle.

The rewritten module for runtime linker pathes is in effect
since r4157/r4158. See
   <http://apps.sourceforge.net/trac/gar/browser/csw/mgar/gar/v2/gar.conf.mk#L435 
 >
for details.

The usage of the rewritten module is documented at
   <http://apps.sourceforge.net/trac/gar/wiki/RuntimeLinkerPathes>

In the easiest case it should be sufficient to do just a
rebuild and everything should be fine. However, the new
structure gives you fine-grained control of the pathes
to be added.

Feel free to ask if there are any issues.


Best regards

   -- Dago


From mwatters at opencsw.org  Thu Apr  2 17:02:09 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 10:02:09 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
Message-ID: <49D4D371.7070102@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Maciej Blizi?ski wrote:
> On Wed, Apr 1, 2009 at 10:47 PM, Mike Watters <mwatters at opencsw.org> wrote:
>> Changes:
>>
>>        bug fix: 3510 'import curses' fails in current Python...
>>                        forced curses to use the library in /opt/csw
> 
> There's still a problem with importing 'curses' module, even though
> it's now linked against CSW ncurses library.
> 
> vsol01 ~ # pkgparam CSWpython VERSION
> 2.6.1,REV=2009.04.01
> vsol01 ~ # ldd /opt/csw/lib/python2.6/lib-dynload/_cursesmodule.so
>         libncurses.so.5 =>       /opt/csw/lib/libncurses.so.5
> vsol01 ~ # python
> Python 2.6.1 (r261:67515, Apr  1 2009, 16:27:46) [C] on sunos5
> Type "help", "copyright", "credits" or "license" for more information.
>>>> import curses
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
>   File "/opt/csw/lib/python2.6/curses/__init__.py", line 15, in <module>
>     from _curses import *
> ImportError: ld.so.1: python: fatal: relocation error: file
> /opt/csw/lib/python2.6/lib-dynload/_cursesmodule.so: symbol w32addch:
> referenced symbol not found
> _______________________________________________
> users mailing list
> users at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/users

ARRGGHHH!!

after a little more searching,
- -- wchgat symbol is in /opt/csw/lib/libncurses.so,
but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so

I will try to get this sorted out in the next few days.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEUEARECAAYFAknU03EACgkQLrhmsXMSLxe0TwCgn2fxcYMglnjrJekijLE9wvsq
gJQAl1RD8FqBq/Hrk+U+Vrfx4nC9rv8=
=WDao
-----END PGP SIGNATURE-----


From phil at bolthole.com  Thu Apr  2 18:23:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Apr 2009 09:23:25 -0700
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4D371.7070102@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
Message-ID: <20090402162325.GG19124@bolthole.com>

On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
> ARRGGHHH!!
> 
> after a little more searching,
> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so

thatas for widecha support. if it isnt there it may be bug in our curses.

or, you need to quit mixing curses headers with ncurses libraries.

-I/opt/csw/include/ncurses

if i remember correctly.


From mwatters at opencsw.org  Thu Apr  2 18:27:48 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 11:27:48 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <20090402162325.GG19124@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
Message-ID: <49D4E784.7070605@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>> ARRGGHHH!!
>>
>> after a little more searching,
>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
> 
> thatas for widecha support. if it isnt there it may be bug in our curses.
> 
> or, you need to quit mixing curses headers with ncurses libraries.
> 
> -I/opt/csw/include/ncurses
> 
> if i remember correctly.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

for some reason, /opt/csw/include/ncurses does not exist?
it exists as /opt/csw/include/ncursesw

and for the record, I have it in the recipe  still have issues  ;(
I will check the ncurses site for bugs and will file a bug against
the ncurses header location.

;)

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknU54QACgkQLrhmsXMSLxcMNQCfY203H7quhK6VWwODJZz5UiYS
JlMAn0CYCbSYwqUFbmsFNOwvxObmGCxo
=uIlm
-----END PGP SIGNATURE-----


From phil at bolthole.com  Thu Apr  2 18:34:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Apr 2009 09:34:57 -0700
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4E784.7070605@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
Message-ID: <20090402163457.GI19124@bolthole.com>

On Thu, Apr 02, 2009 at 11:27:48AM -0500, Mike Watters wrote:

> for some reason, /opt/csw/include/ncurses does not exist?
> it exists as /opt/csw/include/ncursesw

sounds like a takeover maintainer bug

> 
> and for the record, I have it in the recipe  still have issues  ;(
> I will check the ncurses site for bugs and will file a bug against
> the ncurses header location.

just have your pkg use sun curses


From mwatters at opencsw.org  Thu Apr  2 18:36:58 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 11:36:58 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <20090402162325.GG19124@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
Message-ID: <49D4E9AA.8050604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>> ARRGGHHH!!
>>
>> after a little more searching,
>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
> 
> thatas for widecha support. if it isnt there it may be bug in our curses.
> 
> or, you need to quit mixing curses headers with ncurses libraries.
> 
> -I/opt/csw/include/ncurses
> 
> if i remember correctly.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

AH-HA!

it seems there are 2 sets of ncurses libs
libncurses and libncursesw (wide support)
My Headers are picking up the "Wide" support
but my link is against the normal.

this is explained in
http://www.opencsw.org/mantis/view.php?id=3457
though I am still unclear as to why the headers
are missing for the "regular" ncurses.

I will fix and get a new version in testing.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknU6akACgkQLrhmsXMSLxeAywCeJTavy9igOA4zkngRuIMXQraR
WG4AoNlHRySPVahEvYXrKpe1TZQDFeaB
=QQmS
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr  2 19:05:59 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 12:05:59 -0500
Subject: [csw-maintainers] libssh2 now in testing
Message-ID: <49D4F077.4040205@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	update to version 1.1

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknU8HcACgkQLrhmsXMSLxedJACgiXu9pAqNMKX5Y+QAvut4ioEw
x9gAoJalAoAeHg2isdIvzaW20zzZUzmt
=W+1/
-----END PGP SIGNATURE-----


From skayser at opencsw.org  Thu Apr  2 19:55:34 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 02 Apr 2009 19:55:34 +0200
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4E9AA.8050604@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>
	<49D4E9AA.8050604@opencsw.org>
Message-ID: <49D4FC16.6060508@opencsw.org>

Mike Watters wrote:
> Philip Brown wrote:
>> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>>> ARRGGHHH!!
>>>
>>> after a little more searching,
>>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
>> thatas for widecha support. if it isnt there it may be bug in our curses.
> 
>> or, you need to quit mixing curses headers with ncurses libraries.
> 
>> -I/opt/csw/include/ncurses
> 
>> if i remember correctly.
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> AH-HA!
> 
> it seems there are 2 sets of ncurses libs
> libncurses and libncursesw (wide support)
> My Headers are picking up the "Wide" support
> but my link is against the normal.
> 
> this is explained in
> http://www.opencsw.org/mantis/view.php?id=3457
> though I am still unclear as to why the headers
> are missing for the "regular" ncurses.
> 
> I will fix and get a new version in testing.

I didn't get my head around why your linkage went like it did quite yet.
Shouldn't it have bailed out with linking errors (for not finding
w32addch) during building already? James?? ;)

Dago recently built our ncurses with wide character support (thanks
btw.), so that we now also have w variants of the ncurses libraries in
addition to the regular ones (not binary compatible but as they have a w
suffix there's no harm done). Both in the same package.

Coming to the include files. According to the ncurses README:

If you configure using the --enable-widec option, a "w" is appended to
the library names (e.g., libncursesw.a), and the resulting libraries
support wide-characters, e.g., via a UTF-8 locale.  The _corresponding
header files are compatible with the non-wide-character configuration_;
wide-character features are provided by ifdef's in the header files.

I don't exactly know how to read the header bit WRT packages building
against these headers. Can non-wide and wide applications both use the
same headers and ./configure sets #defines in case wide-support is
requested by an app (that's how i would read it)?

At least we should have the original include/ncurses directory back in
the package - presumably filled with the header files of the non-wide
build (as a safe option). And then in addition have include/ncursesw
populated with the header files of the wide variant?

Sebastian


From mwatters at opencsw.org  Thu Apr  2 20:31:59 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 13:31:59 -0500
Subject: [csw-maintainers] [csw-users] python 2.6.1 now in testing
In-Reply-To: <49D4FC16.6060508@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>	<49D4E9AA.8050604@opencsw.org>
	<49D4FC16.6060508@opencsw.org>
Message-ID: <49D5049F.10201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Sebastian Kayser wrote:
> Mike Watters wrote:
>> Philip Brown wrote:
>>> On Thu, Apr 02, 2009 at 10:02:09AM -0500, Mike Watters wrote:
>>>> ARRGGHHH!!
>>>>
>>>> after a little more searching,
>>>> - -- wchgat symbol is in /opt/csw/lib/libncurses.so,
>>>> but the new w32addch is NOT.  It lives in /usr/lib/libcurses.so
>>> thatas for widecha support. if it isnt there it may be bug in our curses.
>>> or, you need to quit mixing curses headers with ncurses libraries.
>>> -I/opt/csw/include/ncurses
>>> if i remember correctly.
>>> _______________________________________________
>>> maintainers mailing list
>>> maintainers at lists.opencsw.org
>>> https://lists.opencsw.org/mailman/listinfo/maintainers
>> AH-HA!
>>
>> it seems there are 2 sets of ncurses libs
>> libncurses and libncursesw (wide support)
>> My Headers are picking up the "Wide" support
>> but my link is against the normal.
>>
>> this is explained in
>> http://www.opencsw.org/mantis/view.php?id=3457
>> though I am still unclear as to why the headers
>> are missing for the "regular" ncurses.
>>
>> I will fix and get a new version in testing.
> 
> I didn't get my head around why your linkage went like it did quite yet.
> Shouldn't it have bailed out with linking errors (for not finding
> w32addch) during building already? James?? ;)
> 
> Dago recently built our ncurses with wide character support (thanks
> btw.), so that we now also have w variants of the ncurses libraries in
> addition to the regular ones (not binary compatible but as they have a w
> suffix there's no harm done). Both in the same package.
> 
> Coming to the include files. According to the ncurses README:
> 
> If you configure using the --enable-widec option, a "w" is appended to
> the library names (e.g., libncursesw.a), and the resulting libraries
> support wide-characters, e.g., via a UTF-8 locale.  The _corresponding
> header files are compatible with the non-wide-character configuration_;
> wide-character features are provided by ifdef's in the header files.
> 
> I don't exactly know how to read the header bit WRT packages building
> against these headers. Can non-wide and wide applications both use the
> same headers and ./configure sets #defines in case wide-support is
> requested by an app (that's how i would read it)?
> 
> At least we should have the original include/ncurses directory back in
> the package - presumably filled with the header files of the non-wide
> build (as a safe option). And then in addition have include/ncursesw
> populated with the header files of the wide variant?
> 
> Sebastian
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I have not figured out exactly how this is happening,
I may be totally off base, but looking in the libraries and the error messages
from the import I can't see any other way these errors should/could pop up.

I just updated my sparc sandbox and will do some more intensive testing
to figure it out.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknVBJ8ACgkQLrhmsXMSLxeROACePW2N3j760Pl4o9BLsEHPbhIt
3hMAoOVL9pRa7URKeZVlhAogUHRK/S8c
=5NgB
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr  2 22:08:16 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Apr 2009 15:08:16 -0500
Subject: [csw-maintainers] pulled gcc4 from testing
Message-ID: <49D51B30.8030006@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

problems with the libraries, not sure exactly what happened.
I am rebuilding using gcc to get all the packages in one recipe


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknVGy8ACgkQLrhmsXMSLxfuSACfQ3RnASa+qf3Aqsp1DhAnEb40
MEIAoJxjpEcora3UrMrE5+tkJmso8Fsp
=paFp
-----END PGP SIGNATURE-----


From rupert at opencsw.org  Fri Apr  3 12:53:35 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Fri, 3 Apr 2009 12:53:35 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
Message-ID: <6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>

hi,

we tried to upgrade our csw local zone resently, and it seems
cswclassutils uses an additional path than the pre-announced /opt/csw.
we have a readonly global zone and got the /opt/csw free to do a mount
loopback:

/etc/opt/csw/init.d/csw.smf.sample
/opt/csw/share/doc/cswclassutils/LICENSE
/opt/csw/share/doc/cswclassutils/README.CSW
/usr/sadm/install/scripts/i.cswcpsampleconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/i.cswinitsmf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswinitsmf> for writing: (30) Read-only
file system
/usr/sadm/install/scripts/i.cswpreserveconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswpreserveconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/i.cswusergroup
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswusergroup> for writing: (30) Read-only
file system
/usr/sadm/install/scripts/r.cswcpsampleconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswcpsampleconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/r.cswinitsmf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswinitsmf> for writing: (30) Read-only
file system
/usr/sadm/install/scripts/r.cswpreserveconf
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswpreserveconf> for writing: (30)
Read-only file system
/usr/sadm/install/scripts/r.cswusergroup
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/r.cswusergroup> for writing: (30) Read-only
file system
[ verifying class <none> ]
ERROR: attribute verification of
</usr/sadm/install/scripts/i.cswcpsampleconf> failed
? ?pathname does not exist
ERROR: attribute verification of </usr/sadm/install/scripts/i.cswinitsmf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/i.cswpreserveconf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/i.cswusergroup> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/r.cswcpsampleconf> failed
? ?pathname does not exist
ERROR: attribute verification of </usr/sadm/install/scripts/r.cswinitsmf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/r.cswpreserveconf> failed
? ?pathname does not exist
ERROR: attribute verification of
</usr/sadm/install/scripts/r.cswusergroup> failed
? ?pathname does not exist


From bonivart at opencsw.org  Fri Apr  3 12:57:43 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Fri, 3 Apr 2009 12:57:43 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
Message-ID: <625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>

On Fri, Apr 3, 2009 at 12:53 PM, rupert THURNER <rupert at opencsw.org> wrote:
> hi,
>
> we tried to upgrade our csw local zone resently, and it seems
> cswclassutils uses an additional path than the pre-announced /opt/csw.
> we have a readonly global zone and got the /opt/csw free to do a mount
> loopback:
>
> /etc/opt/csw/init.d/csw.smf.sample
> /opt/csw/share/doc/cswclassutils/LICENSE
> /opt/csw/share/doc/cswclassutils/README.CSW
> /usr/sadm/install/scripts/i.cswcpsampleconf
> pkgadd.ORG: ERROR: unable to open
> </usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
> Read-only file system

As far as I know class action scripts only work from
/usr/sadm/install/scripts so you need to install CSWcswclassutils from
the global zone, then the dependency will be fulfilled in all
non-global zones.

-- 
/peter


From mwatters at opencsw.org  Fri Apr  3 22:03:11 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 03 Apr 2009 15:03:11 -0500
Subject: [csw-maintainers] New Python in Testing
Message-ID: <49D66B7F.3000101@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	FIX: bug 3510 "...import curses fails" ( actually fixed this time )
	     I am able to import EVERY module that is bundled with python ;)

	FIX: bug 3579 RPATH problems.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknWa38ACgkQLrhmsXMSLxdL8ACgslSu0C2V0rpIIXKTDpEgCixQ
XNEAoIezAYzu+tEsKh5TqJY5rABM0/jH
=H3p7
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Fri Apr  3 22:13:47 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 03 Apr 2009 15:13:47 -0500
Subject: [csw-maintainers] php5 5.2.9 now in testing
Message-ID: <49D66DFB.7010408@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	fixed bugs
	3587: Extra Paths in package
	3499: xsl module broken
	3479: uncomment extensions

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknWbfsACgkQLrhmsXMSLxe9EQCguZ3K7fV4ozplhN7+WAfOsJRb
3DEAoN5GM48Tn79v+2csfnbKtCOQm1at
=+nuL
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sat Apr  4 18:27:24 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 04 Apr 2009 11:27:24 -0500
Subject: [csw-maintainers] pysqlite and pysqlite2 now in testing
Message-ID: <49D78A6C.1080502@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	recompile to fix isalist

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknXimwACgkQLrhmsXMSLxer9ACg5nx6CwXY7IR7Qn/sLNcJ+fx7
cdkAoJZPjh1G2dV+BfNA2DYIcCU59dm0
=RAaL
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sun Apr  5 09:32:26 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 05 Apr 2009 02:32:26 -0500
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
Message-ID: <49D85E8A.4040709@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Is there a way we can get the summary in addition to the link.

Example:

Here is a summary of your open CSW Mantis issues.  If you are not the
maintainer of these projects please inform <bug-admin at opencsw.org>.

**********************************************************************
These reports have severity "major" or above:
- ---------------------------------------------
Summary: summary major bug 1
   Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}


**********************************************************************
These reports have severity "minor" or below:
- ---------------------------------------------
Summary: summary minor bug 1
   Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}





- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknYXooACgkQLrhmsXMSLxd9IwCg28u6/9fZ5kBS3HivHpWKvVh7
VXUAmwdwkBQY22l9NFWT/N8VLU6XdBde
=yjGi
-----END PGP SIGNATURE-----


From hson at opencsw.org  Sun Apr  5 10:24:18 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Sun, 05 Apr 2009 10:24:18 +0200
Subject: [csw-maintainers] zlib package in /testing
Message-ID: <49D86AB2.5080005@opencsw.org>

I've put a zlib package in /testing which is the same version as the 
current one, but this is built with gar plus have a 64bit library.

Unless someone finds any trouble with it I'll release it before easter


From james at opencsw.org  Sun Apr  5 10:51:23 2009
From: james at opencsw.org (James Lee)
Date: Sun, 05 Apr 2009 08:51:23 GMT
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <49D85E8A.4040709@opencsw.org>
References: <49D85E8A.4040709@opencsw.org>
Message-ID: <20090405.8512300.4214488583@gyor.oxdrove.co.uk>

On 05/04/09, 08:32:26, Mike Watters <mwatters at opencsw.org> wrote regarding 
[csw-maintainers] RFE: Mantis Bug Reminder:

> Is there a way we can get the summary in addition to the link.

...

> Summary: summary major bug 1
>    Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}

The number works for me but if you have trouble learning phone
books it can be added.

You are supposed to be working on this list so it should familiar
and more importantly short or not there!



James.



From james at opencsw.org  Sun Apr  5 10:57:29 2009
From: james at opencsw.org (James Lee)
Date: Sun, 05 Apr 2009 08:57:29 GMT
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D86AB2.5080005@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
Message-ID: <20090405.8572900.3105252976@gyor.oxdrove.co.uk>

On 05/04/09, 09:24:18, "Roger H?kansson" <hson at opencsw.org> wrote regarding 
[csw-maintainers] zlib package in /testing:

> I've put a zlib package in /testing which is the same version as the
> current one, but this is built with gar plus have a 64bit library.

The exists has 64 bit libs (both sparcv9 and amd64)

Please set an SONAME.

There is no need for an RPATH as it doesn't use any CSW libs.

$ dump -Lv /usr/lib/libz.so | egrep 'RPATH|SONAME'
[4]	SONAME          libz.so.1




James.






From hson at opencsw.org  Sun Apr  5 11:42:57 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Sun, 05 Apr 2009 11:42:57 +0200
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <20090405.8572900.3105252976@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
Message-ID: <49D87D21.4050606@opencsw.org>

James Lee wrote:
> On 05/04/09, 09:24:18, "Roger H?kansson" <hson at opencsw.org> wrote regarding 
> [csw-maintainers] zlib package in /testing:
> 
>> I've put a zlib package in /testing which is the same version as the
>> current one, but this is built with gar plus have a 64bit library.
> 
> The exists has 64 bit libs (both sparcv9 and amd64)

How could I miss that...

> There is no need for an RPATH as it doesn't use any CSW libs.


Is it really a problem if it has RPATH set? (and its correctly set)
"Unsetting" (in general, not just for this package) it would mean that 
each gar package (which don't use any CSW libs) would have to have a 
LD_OPTION rewrite rule, since gar sets it by default?

I have no problem doing this, but I'm just thinking if were creating 
unnecessary work.

 > Please set an SONAME.
 >
> $ dump -Lv /usr/lib/libz.so | egrep 'RPATH|SONAME'
> [4]	SONAME          libz.so.1
> 

Fixed.



From mwatters at opencsw.org  Sun Apr  5 17:33:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 05 Apr 2009 10:33:49 -0500
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <20090405.8512300.4214488583@gyor.oxdrove.co.uk>
References: <49D85E8A.4040709@opencsw.org>
	<20090405.8512300.4214488583@gyor.oxdrove.co.uk>
Message-ID: <49D8CF5D.3080205@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

James Lee wrote:
> On 05/04/09, 08:32:26, Mike Watters <mwatters at opencsw.org> wrote regarding 
> [csw-maintainers] RFE: Mantis Bug Reminder:
> 
>> Is there a way we can get the summary in addition to the link.
> 
> ...
> 
>> Summary: summary major bug 1
>>    Link: http://www.opencsw.org/bugtrack/view.php?id={bugid}
> 
> The number works for me but if you have trouble learning phone
> books it can be added.
> 
> You are supposed to be working on this list so it should familiar
> and more importantly short or not there!
> 
> 
> 
> James.
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

it is not important.
the bugs are usually not open long enough for me to memorize the number. ;)


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknYz10ACgkQLrhmsXMSLxeS/wCgk431s94W/cOCojpsSPB8bmEr
s/AAnRLhsXCOk0ltYt9qaBvPBFyIM7N5
=u/JN
-----END PGP SIGNATURE-----


From phil at bolthole.com  Sun Apr  5 17:59:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 5 Apr 2009 08:59:45 -0700
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D87D21.4050606@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
Message-ID: <20090405155945.GA9321@bolthole.com>

On Sun, Apr 05, 2009 at 11:42:57AM +0200, Roger H?kansson wrote:
> Is it really a problem if it has RPATH set? (and its correctly set)

well, technically, if it doesnt use any csw libs, you could probably make
it a fully relocatable package.

however, relocatable packages shouldnt have a fixed RPATH set :)

(does gar support making relocatable packages?)


From jgoerzen at opencsw.org  Sun Apr  5 19:01:23 2009
From: jgoerzen at opencsw.org (Jake Goerzen)
Date: Sun, 5 Apr 2009 10:01:23 -0700
Subject: [csw-maintainers] new fltk pkgs in /testing
Message-ID: <315c02ae0904051001u441554a9u38bd240b0b51f37a@mail.gmail.com>

new fltk pkgs are available in /testing:

fltk-1.1.9,REV=2009.04.05-SunOS5.8-i386-CSW.pkg.gz
fltk-1.1.9,REV=2009.04.05-SunOS5.8-sparc-CSW.pkg.gz

These pkgs should resolve the following bugs:

0000976 -  need fltk compiled with OpenGL (Mesa)
0001164 -  Incompatible with flphoto
0001982 -  local paths in config files

Testing and feedback appreciated
Thanks, Jake


From jgoerzen at opencsw.org  Sun Apr  5 19:07:02 2009
From: jgoerzen at opencsw.org (Jake Goerzen)
Date: Sun, 5 Apr 2009 10:07:02 -0700
Subject: [csw-maintainers] new fltk pkgs in /testing
Message-ID: <315c02ae0904051007va81ee68uc6d3a8500eabcb83@mail.gmail.com>

Sorry, I over looked that 2 of the listed bugs are already closed bugs.
So, this release of fltk pkgs should fix bug id # 0001982

Thanks, Jake


From jgoerzen at opencsw.org  Sun Apr  5 19:26:16 2009
From: jgoerzen at opencsw.org (Jake Goerzen)
Date: Sun, 5 Apr 2009 10:26:16 -0700
Subject: [csw-maintainers] new libsdl pkgs in /testing
Message-ID: <315c02ae0904051026l57e1cb10h740e213303f858a@mail.gmail.com>

new libsdl pkgs available in /testing:

libsdl-1.2.13,REV=2009.04.04-SunOS5.8-i386-CSW.pkg.gz
libsdl-1.2.13,REV=2009.04.04-SunOS5.8-sparc-CSW.pkg.gz

These pkgs I hope fix missing symlink issues:

0003048 -  libSDL-1.2.so.0 doe snot link the good lib
0003518 -  link libSDL-1.2.so.0 not pointing to the right lib

Thanks for testing, Jake


From dam at opencsw.org  Sun Apr  5 23:00:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 5 Apr 2009 23:00:40 +0200
Subject: [csw-maintainers] Updating all packages on all buildfarm servers to
	current now.
Message-ID: <105A89E8-CB1F-4C47-9062-FD162748DD8D@opencsw.org>

Hi,

I am updating all packages on all buildfarm servers to current now
to reflect the recent packages releases.


Best regards

   -- Dago


From dam at opencsw.org  Sun Apr  5 23:17:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 5 Apr 2009 23:17:17 +0200
Subject: [csw-maintainers] RPATH in GAR
In-Reply-To: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
References: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
Message-ID: <C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>

Hi,

Am 02.04.2009 um 14:22 schrieb Dagobert Michelsen:
> The usage of the rewritten module is documented at
>  <http://apps.sourceforge.net/trac/gar/wiki/RuntimeLinkerPathes>
>
> In the easiest case it should be sufficient to do just a
> rebuild and everything should be fine. However, the new
> structure gives you fine-grained control of the pathes
> to be added.
>
> Feel free to ask if there are any issues.

I found another issue: if an application adds somehow runtime linker
pathes themselves the resulting binary contains pathes similar to
   /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib
beginning with the one or two correct pathes from LD_OPTIONS and then
some redundant stuff added from the application. James proposed to use
some compiler-interposer which strips and reorders certain flags before
passing them to the real compiler. If someone has a less-intrusive idea
then it would be taken very well :-)


Best regards

   -- Dago


From bwalton at opencsw.org  Sun Apr  5 23:36:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 05 Apr 2009 17:36:59 -0400
Subject: [csw-maintainers] RPATH in GAR
In-Reply-To: <C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>
References: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
	<C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>
Message-ID: <1238967347-sup-9259@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Apr 05 17:17:17 -0400 2009:
> I found another issue: if an application adds somehow runtime linker
> pathes themselves the resulting binary contains pathes similar to
>    /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib

...and ruby is now sticking a /lib before everything else somehow too,
which is now starting to really tick me off.

Other than being wasteful and redundant, is the duplication of the
paths harmful?

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090405/6eeed85c/attachment-0001.asc>

From mwatters at opencsw.org  Mon Apr  6 01:15:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 05 Apr 2009 18:15:33 -0500
Subject: [csw-maintainers] gcc-4.3.3 in testing
Message-ID: <49D93B95.5020307@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

make check results checked into subversion at:
http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/gcc4/trunk/files/test-results/?pathrev=4204

Please test and provide feedback.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknZO5UACgkQLrhmsXMSLxcbGQCggUZpkWVvTT0xJSPqJJp1kx1k
GFEAoObNnhfWMcnT0PIWgcFo1atpkbuj
=56PD
-----END PGP SIGNATURE-----


From phil at bolthole.com  Mon Apr  6 05:10:01 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 5 Apr 2009 20:10:01 -0700
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
Message-ID: <20090406031001.GA1795@bolthole.com>

a reminder for those who may not realize it:

If you want a "summary" of your bugs... just make sure that they are at
least all "assigned" to you... then you can go to "my view", and select
"all projects", or "view all bugs" page. and select project of "all
projects"

you'll then get a summary with bugids AND summary line.


From dam at opencsw.org  Mon Apr  6 09:58:54 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 09:58:54 +0200
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <20090406031001.GA1795@bolthole.com>
References: <20090406031001.GA1795@bolthole.com>
Message-ID: <E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>

Hi,

Am 06.04.2009 um 05:10 schrieb Philip Brown:
> a reminder for those who may not realize it:
>
> If you want a "summary" of your bugs... just make sure that they are  
> at
> least all "assigned" to you... then you can go to "my view", and  
> select
> "all projects", or "view all bugs" page. and select project of "all
> projects"
>
> you'll then get a summary with bugids AND summary line.

...and of course at the bottom of your maintainer homepage an individual
listing of all bugs, e. g. for me
   http://www.opencsw.org/buglist/maintainer.cgi?maintainer=dam


Best regards

   -- Dago


From james at opencsw.org  Mon Apr  6 10:30:51 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 08:30:51 GMT
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D87D21.4050606@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
Message-ID: <20090406.8305100.1838829892@gyor.oxdrove.co.uk>

On 05/04/09, 10:42:57, "Roger H?kansson" <hson at opencsw.org> wrote regarding
Re: [csw-maintainers] zlib package in /testing:

> > There is no need for an RPATH as it doesn't use any CSW libs.

> Is it really a problem if it has RPATH set? (and its correctly set)
> "Unsetting" (in general, not just for this package) it would mean that
> each gar package (which don't use any CSW libs) would have to have a
> LD_OPTION rewrite rule, since gar sets it by default?

> I have no problem doing this, but I'm just thinking if were creating
> unnecessary work.

It's not correct as no CSW libs are opened.

It's not a "problem" but if it's the result of the build system it
suggests we are generally making settings with no thought.  The
current batch of odd RPATHs only causes the link loader to look for
a directory that doesn't exist, but this is less wasted effort than
what happens with $ISALIST which typically looks at all the 64 bit
locations before finding that matching 32 lib.


How difficult is it to do right?  Perhaps the recent batch of odd
paths are telling me that people are using a "sausage machine"
approach to building (throw anything in the top and just wind the
handle).  We are dangerously close to the "I'd like to do the right
thing but the build system won't let me".  The value CSW should add
*is* to take the time to get things right - your individual effort
shared for the benefit of many.



>  > Please set an SONAME.
>  >
> > $ dump -Lv /usr/lib/libz.so | egrep 'RPATH|SONAME'
> > [4] SONAME          libz.so.1
> >

> Fixed.

Thank you.




James.


From james at opencsw.org  Mon Apr  6 10:36:14 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 08:36:14 GMT
Subject: [csw-maintainers] RPATH in GAR
In-Reply-To: <1238967347-sup-9259@ntdws12.chass.utoronto.ca>
References: <441795FF-BED1-4530-B31F-1397FFDEE047@opencsw.org>
	<C12E0588-A276-4ED5-917D-1CBCDB6594EA@opencsw.org>
	<1238967347-sup-9259@ntdws12.chass.utoronto.ca>
Message-ID: <20090406.8361400.4103879239@gyor.oxdrove.co.uk>

On 05/04/09, 22:36:59, Ben Walton <bwalton at opencsw.org> wrote regarding Re:
[csw-maintainers] RPATH in GAR:

> > I found another issue: if an application adds somehow runtime linker
> > pathes themselves the resulting binary contains pathes similar to
> >    /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib


> Other than being wasteful and redundant, is the duplication of the
> paths harmful?

No.



James.


From james at opencsw.org  Mon Apr  6 10:36:16 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 08:36:16 GMT
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>
References: <20090406031001.GA1795@bolthole.com>
	<E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>
Message-ID: <20090406.8361600.4264367480@gyor.oxdrove.co.uk>

On 06/04/09, 08:58:54, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
Re: [csw-maintainers] RFE: Mantis Bug Reminder:

> Am 06.04.2009 um 05:10 schrieb Philip Brown:
> > a reminder for those who may not realize it:
> >
> > If you want a "summary" of your bugs... just make sure that they are
> > at
> > least all "assigned" to you... then you can go to "my view", and
> > select
> > "all projects", or "view all bugs" page. and select project of "all
> > projects"
> >
> > you'll then get a summary with bugids AND summary line.

> ...and of course at the bottom of your maintainer homepage an individual
> listing of all bugs, e. g. for me
>    http://www.opencsw.org/buglist/maintainer.cgi?maintainer=dam


I later realised this too, the reminder could just say:
"You have a|some|lots|loads|excessive bugs, go to the mantis web page".
Listing the links just makes it look proportionate.



James.





From rupert at opencsw.org  Mon Apr  6 11:48:38 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 6 Apr 2009 11:48:38 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
Message-ID: <6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>

what would happen if one lives without classutils?


On Fri, Apr 3, 2009 at 12:57, Peter Bonivart <bonivart at opencsw.org> wrote:
> On Fri, Apr 3, 2009 at 12:53 PM, rupert THURNER <rupert at opencsw.org> wrote:
>> hi,
>>
>> we tried to upgrade our csw local zone resently, and it seems
>> cswclassutils uses an additional path than the pre-announced /opt/csw.
>> we have a readonly global zone and got the /opt/csw free to do a mount
>> loopback:
>>
>> /etc/opt/csw/init.d/csw.smf.sample
>> /opt/csw/share/doc/cswclassutils/LICENSE
>> /opt/csw/share/doc/cswclassutils/README.CSW
>> /usr/sadm/install/scripts/i.cswcpsampleconf
>> pkgadd.ORG: ERROR: unable to open
>> </usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
>> Read-only file system
>
> As far as I know class action scripts only work from
> /usr/sadm/install/scripts so you need to install CSWcswclassutils from
> the global zone, then the dependency will be fulfilled in all
> non-global zones.
>
> --
> /peter
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From dam at opencsw.org  Mon Apr  6 14:30:25 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 14:30:25 +0200
Subject: [csw-maintainers] Shutdown of login,
	build* in 30 minutes for upgrade
Message-ID: <3B9E6F40-E42B-4BDB-B1F0-2A242CE43BD4@opencsw.org>

Hi,

I am going to upgrade the T5220 hosting login, build8s, build9s,
build10s, web, hudson and other stuff from 8 GB to 16 GB Ram
in 30 minutes. Please stand by.


Best regards

   -- Dago


From hson at opencsw.org  Mon Apr  6 14:35:24 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Mon, 06 Apr 2009 14:35:24 +0200
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <20090406.8305100.1838829892@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>
	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>
Message-ID: <49D9F70C.9030007@opencsw.org>

James Lee wrote:
 > It's not correct as no CSW libs are opened.
 >
> It's not a "problem" but if it's the result of the build system it
> suggests we are generally making settings with no thought.  The
> current batch of odd RPATHs only causes the link loader to look for
> a directory that doesn't exist, but this is less wasted effort than
> what happens with $ISALIST which typically looks at all the 64 bit
> locations before finding that matching 32 lib.

$ISALIST doesn't just expand to 64bit, you have optimized 32bit versions 
(sparcv8plus, sparcv8plus+vis...) as well?

I take it that you haven't built any packages using gar yet?
The recent batch of odd paths was due to problems with gar where 
LD_OPTIONS somehow got messed up (there have been a number of postings 
about this problem).
Even though I've only been a maintainer for a few months, my 
understanding (from http://opencsw.org/standards/pkg-walkthrough as an 
example) was that LD_OPTIONS was a preferred setting when building 
packages. So all packages I've built (either using gar or manually) have 
LD_OPTIONS set which means RPATH/RUNPATH have been set to 
/opt/csw/lib/$ISALIST:/opt/csw/lib whatever they need.

If LD_OPTIONS shouldn't be used, or should only contain what's needed 
for a particular package, it should be communicated more clearly.

> How difficult is it to do right?  Perhaps the recent batch of odd
> paths are telling me that people are using a "sausage machine"
> approach to building (throw anything in the top and just wind the
> handle).  We are dangerously close to the "I'd like to do the right
> thing but the build system won't let me".  The value CSW should add
> *is* to take the time to get things right - your individual effort
> shared for the benefit of many.

Right now, as far as I can see it, gar lets you set LD_OPTIONS to 
anything you want, but if you don't set it, gar will set it for you.

To reiterate my question:
Is there any actual harm (disregarding the fact that the link loader 
might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without 
finding any match) having RPATH/RUNPATH set?

If not, it is only when a package doesn't use any libs at all from 
/opt/csw/lib, when RPATH/RUNPATH should be changed from the default.

If package A is linked to a lib i package B (both CSW packages) and A i 
packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future 
releases of B which contain optimized libraries won't be used until A is 
repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/csw/lib
And if there is a longer dependency list (A->B->C-D where D is the one 
with optimized libs) it creates much more work for the maintainer of A 
(he must check every lib in the dependency path to see if there are any 
optimized libs).

But then again, if there is no harm in having RPATH/RUNPATH set and it 
only creates much more work for the maintainers, why shouldn't we leave 
it as it is?


From james at opencsw.org  Mon Apr  6 16:34:50 2009
From: james at opencsw.org (James Lee)
Date: Mon, 06 Apr 2009 14:34:50 GMT
Subject: [csw-maintainers] zlib package in /testing
In-Reply-To: <49D9F70C.9030007@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>
	<49D9F70C.9030007@opencsw.org>
Message-ID: <20090406.14345000.1346203350@gyor.oxdrove.co.uk>

On 06/04/09, 13:35:24, "Roger H?kansson" <hson at opencsw.org> wrote regarding
Re: [csw-maintainers] zlib package in /testing:

> > It's not a "problem" but if it's the result of the build system it
> > suggests we are generally making settings with no thought.  The
> > current batch of odd RPATHs only causes the link loader to look for
> > a directory that doesn't exist, but this is less wasted effort than
> > what happens with $ISALIST which typically looks at all the 64 bit
> > locations before finding that matching 32 lib.

> $ISALIST doesn't just expand to 64bit, you have optimized 32bit versions
> (sparcv8plus, sparcv8plus+vis...) as well?

It expands to the isalist.  It tries all in order but if it's a 32 bit
prog there is never a valid match in the 64s, then it runs through the
32s until the match, or not in the case of this first non CSW lib:

$ truss /opt/csw/bin/gsc
execve("/opt/csw/bin/gsc", 0xFFBEFBDC, 0xFFBEFBE4)  argc = 1
resolvepath("/usr/lib/ld.so.1", "/usr/lib/ld.so.1", 1023) = 16
open("/var/ld/ld.config", O_RDONLY)             Err#2 ENOENT
sysinfo(514, "sparcv9+vis2 sparcv9+vis sparcv9 sparcv8plus+vis
sparcv8plus sparcv8 sparcv8-fsmuld sparcv7 sparc", 257) = 98
stat("/opt/csw/lib/sparcv9+vis2/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv9+vis/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv9/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8plus+vis/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8plus/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv8-fsmuld/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparcv7/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/sparc/libc.so.1", 0xFFBEF488) Err#2 ENOENT
stat("/opt/csw/lib/libc.so.1", 0xFFBEF488)      Err#2 ENOENT
stat("/usr/lib/libc.so.1", 0xFFBEF488)          = 0
resolvepath("/usr/lib/libc.so.1", "/usr/lib/libc.so.1", 1023) = 18
open("/usr/lib/libc.so.1", O_RDONLY)            = 3
...

Lots of Err#2


> Even though I've only been a maintainer for a few months, my
> understanding (from http://opencsw.org/standards/pkg-walkthrough as an
> example) was that LD_OPTIONS was a preferred setting when building
> packages. So all packages I've built (either using gar or manually) have
> LD_OPTIONS set which means RPATH/RUNPATH have been set to
> /opt/csw/lib/$ISALIST:/opt/csw/lib whatever they need.

> If LD_OPTIONS shouldn't be used, or should only contain what's needed
> for a particular package, it should be communicated more clearly.

Once built a package doesn't know how it was built so if it works
use it.

LD_OPTIONS overrides all so is powerful.  Like all powerful weapons
use with caution.

$ man ld
...
     LD_OPTIONS
           A default set of options to ld. LD_OPTIONS  is  inter-
           preted  by ld just as though its value had been placed
           on the command line, immediately  following  the  name
           used to invoke ld, as in:

           ld $LD_OPTIONS ... other-arguments ...


> To reiterate my question:
> Is there any actual harm (disregarding the fact that the link loader
> might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without
> finding any match) having RPATH/RUNPATH set?

No, well it's actual but trivial.


> If not, it is only when a package doesn't use any libs at all from
> /opt/csw/lib, when RPATH/RUNPATH should be changed from the default.

Correct, as is the case with zlib.

> If package A is linked to a lib i package B (both CSW packages) and A i
> packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future
> releases of B which contain optimized libraries won't be used until A is
> repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/csw/lib

Correct.  Add $ISALIST in, unless you know it won't be needed.


> And if there is a longer dependency list (A->B->C-D where D is the one
> with optimized libs) it creates much more work for the maintainer of A
> (he must check every lib in the dependency path to see if there are any
> optimized libs).

D is governed by the RPATH of C.  C by B.



James.


From bonivart at opencsw.org  Mon Apr  6 16:43:45 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Apr 2009 16:43:45 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com> 
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com> 
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
Message-ID: <625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>

On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
> what would happen if one lives without classutils?

Disregarding that you have to force past the dependency requirement it
would be the equivalent of not running pre/post scripts. The packages
that depend on cswclassutils use it for some combo of e.g. init/SMF
support and handling of configuration files, see
http://wiki.opencsw.org/cswclassutils-package for all features.

What's the problem to install it in the global zone? Even if you don't
have root access to it yourself you should be able to make a request
for a package install, right? What would you do with a sparse zone if
you needed a SUNW package that installed in /usr?

-- 
/peter


From dam at opencsw.org  Mon Apr  6 16:49:23 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 16:49:23 +0200
Subject: [csw-maintainers] Shutdown of login,
	build* in 30 minutes for upgrade
In-Reply-To: <3B9E6F40-E42B-4BDB-B1F0-2A242CE43BD4@opencsw.org>
References: <3B9E6F40-E42B-4BDB-B1F0-2A242CE43BD4@opencsw.org>
Message-ID: <3B5B2F86-9763-497F-A8E6-E9511D9C33B2@opencsw.org>

Hi,

Am 06.04.2009 um 14:30 schrieb Dagobert Michelsen:
> I am going to upgrade the T5220 hosting login, build8s, build9s,
> build10s, web, hudson and other stuff from 8 GB to 16 GB Ram
> in 30 minutes. Please stand by.

Sorry for the extended downtime. The issue proved to be more
complicated than anticipated because there already were
DataRam memory in the system and the new memory was Sun
memory. The type of memory is actually compared during POST
and mixing memory leads to immediate shutdown :-( However,
certain bank combinations do work and the system now has
16 GB of Ram.

Happy packaging :-)

   -- Dago


From Darin.Perusich at cognigencorp.com  Mon Apr  6 16:56:31 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 06 Apr 2009 10:56:31 -0400
Subject: [csw-maintainers] inetd.conf entries for solaris 10
Message-ID: <49DA181F.8080305@cognigencorp.com>

Where can I find documentation/examples for dealing with inetd.conf/SMF
entries? I need to convert a few inetd.conf entries and while using
inetconv will give me what I need I'd prefer to provide the manifest.

Thanks

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From bonivart at opencsw.org  Mon Apr  6 17:18:05 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Apr 2009 17:18:05 +0200
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <49DA181F.8080305@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
Message-ID: <625385e30904060818x214be099r1f8d6485da577087@mail.gmail.com>

On Mon, Apr 6, 2009 at 4:56 PM, Darin Perusich
<Darin.Perusich at cognigencorp.com> wrote:
> Where can I find documentation/examples for dealing with inetd.conf/SMF
> entries? I need to convert a few inetd.conf entries and while using
> inetconv will give me what I need I'd prefer to provide the manifest.

I would also be interested in that for my qpopper package.

-- 
/peter


From dam at opencsw.org  Mon Apr  6 18:17:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 6 Apr 2009 18:17:34 +0200
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <20090402163457.GI19124@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
Message-ID: <A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>

Hi,

Am 02.04.2009 um 18:34 schrieb Philip Brown:
> On Thu, Apr 02, 2009 at 11:27:48AM -0500, Mike Watters wrote:
>
>> for some reason, /opt/csw/include/ncurses does not exist?
>> it exists as /opt/csw/include/ncursesw
>
> sounds like a takeover maintainer bug

There is now an updated ncurses in testing which has both
include/ncurses and include/ncursesw.


Best regards

   -- Dago


From phil at bolthole.com  Mon Apr  6 19:14:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 10:14:32 -0700
Subject: [csw-maintainers] RFE: Mantis Bug Reminder
In-Reply-To: <20090406.8361600.4264367480@gyor.oxdrove.co.uk>
References: <20090406031001.GA1795@bolthole.com>
	<E956650D-7BB6-48B3-B60D-9E324F5C920F@opencsw.org>
	<20090406.8361600.4264367480@gyor.oxdrove.co.uk>
Message-ID: <20090406171432.GE73510@bolthole.com>

On Mon, Apr 06, 2009 at 08:36:16AM +0000, James Lee wrote:
> > ...and of course at the bottom of your maintainer homepage an individual
> > listing of all bugs, e. g. for me
> >    http://www.opencsw.org/buglist/maintainer.cgi?maintainer=dam
> 
> 
> I later realised this too, the reminder could just say:
> "You have a|some|lots|loads|excessive bugs, go to the mantis web page".
> Listing the links just makes it look proportionate.

Eh.... for people who get A LOT of email...
sometimes, its nice to get an email with the specifics, so it catches your
eye and reminds you about a specific issue that is worth going to look at
"sooner", rather than "later".



From phil at bolthole.com  Mon Apr  6 19:37:46 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 10:37:46 -0700
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <49DA181F.8080305@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
Message-ID: <20090406173746.GL73510@bolthole.com>

On Mon, Apr 06, 2009 at 10:56:31AM -0400, Darin Perusich wrote:
> Where can I find documentation/examples for dealing with inetd.conf/SMF
> entries? I need to convert a few inetd.conf entries and while using
> inetconv will give me what I need I'd prefer to provide the manifest.
> 

in sol10, the inetd.conf stuff is kinda deprecated. 
you run inetdconv (or whatever), to slurp up whatever is in inetd.conf, and
  autogenerate a "real" smf definition for it.

You could just look at what was generated for SMF for some test run of the
above, and then make your package use that directly, instead of messing
with inetd.conf in sol10.

(Note to Peter B: bettter still would be running qpopper, etc, as a full
  demon autorestarted via SMF, if possible, rather than inetd, on sol10)




From Darin.Perusich at cognigencorp.com  Mon Apr  6 20:11:29 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 06 Apr 2009 14:11:29 -0400
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <20090406173746.GL73510@bolthole.com>
References: <49DA181F.8080305@cognigencorp.com>
	<20090406173746.GL73510@bolthole.com>
Message-ID: <49DA45D1.8060008@cognigencorp.com>


Philip Brown wrote:
> On Mon, Apr 06, 2009 at 10:56:31AM -0400, Darin Perusich wrote:
>> Where can I find documentation/examples for dealing with inetd.conf/SMF
>> entries? I need to convert a few inetd.conf entries and while using
>> inetconv will give me what I need I'd prefer to provide the manifest.
>>
> 
> in sol10, the inetd.conf stuff is kinda deprecated. 
> you run inetdconv (or whatever), to slurp up whatever is in inetd.conf, and
>   autogenerate a "real" smf definition for it.

This is my plan and I've already done so.

> You could just look at what was generated for SMF for some test run of the
> above, and then make your package use that directly, instead of messing
> with inetd.conf in sol10.

This is why I'm asking for documentation or an example of creating a
package in such a way. While SMF may be the "proper" way on 10 I'm going
to do what works, has always worked, and still does, init.d script and
inetconv for inetd things.

> (Note to Peter B: bettter still would be running qpopper, etc, as a full
>   demon autorestarted via SMF, if possible, rather than inetd, on sol10)
> 


-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Mon Apr  6 20:21:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 11:21:52 -0700
Subject: [csw-maintainers] inetd.conf entries for solaris 10
In-Reply-To: <49DA45D1.8060008@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
	<20090406173746.GL73510@bolthole.com>
	<49DA45D1.8060008@cognigencorp.com>
Message-ID: <20090406182152.GP73510@bolthole.com>

On Mon, Apr 06, 2009 at 02:11:29PM -0400, Darin Perusich wrote:
> This is why I'm asking for documentation or an example of creating a
> package in such a way. While SMF may be the "proper" way on 10 I'm going
> to do what works, has always worked, and still does, init.d script and
> inetconv for inetd things.

oh... so to be specific, you're looking at examples of how to handle stuff
in inetd.conf for sol8/9, IN CONTRAST to the solaris 10 way?

That rather clashes with the subject line :-}

I'm not sure we really have a set way. I think most demons we have packages,
either avoid inetd entirely, or just leave it up to the user to 
edit the file when and if they want to.

You COULD make a package for it, that does nothing to the inetd conf.
Or you could make an explicit postinstall script.

Or... you could be "noble", and make a new classutils class action script
for it :)

so on sol8/9 it would auto insert/update/.... inetd.conf, and
  in sol10 it would auto SMF it.

you could allow for a file like for the cswusergroup class....

/opt/csw/etc/pkg/[pkgname]/cswinetdconf

with one or more lines in inetdconf style, that would be the trigger
for the class action script.
Although you would also potentially require some entries for any
"new" /etc/services entries, if they were required.



From bwalton at opencsw.org  Mon Apr  6 20:32:06 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 06 Apr 2009 14:32:06 -0400
Subject: [csw-maintainers] ruby in testing
Message-ID: <1239042541-sup-7577@ntdws12.chass.utoronto.ca>


Hi All,

Updated ruby packages are available from testing.  These address the
RPATH bugs (eg: they now properly pick up any optimized libraries) and
have relocated the libruby.so symlink from rubydev to ruby to avoid
breaking some older packages.

Please let me know if you have any issues with these.

Thanks
-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090406/2773b086/attachment-0001.asc>

From william at wbonnet.net  Mon Apr  6 21:45:49 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 06 Apr 2009 21:45:49 +0200
Subject: [csw-maintainers] Packages age statistics : March '09
Message-ID: <49DA5BED.9010908@wbonnet.net>

Hi

Here is the March update of package age statistics for the current source.

The third column show the delta between this month and previous month, the last one the cummulated delta over the year.

As you can notice packages total from years 2003 to 2008 decreased,
which is good (they have been updated). Delta of packages in 2009 is
+239, which is excellent, the best ever we achieved.

This following statistics show that 136 packages have been update over 
the last month, and 103 new packages added. 

Once again this is excellent. Thanks to you all.

Year    Total      Delta      Year      
1997      1          0          0
1998      1          0          0
2001      3          0          0
2002      4          0          0
2003     23         -2         -3
2004     99         -8         -8
2005    174         -2         -5
2006    184        -16        -20
2007    160        -24        -39
2008    304        -84       -131
2009    361       +239       +383

Cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From rmacduff at opencsw.org  Mon Apr  6 21:53:28 2009
From: rmacduff at opencsw.org (Ross Macduff)
Date: Mon, 06 Apr 2009 15:53:28 -0400
Subject: [csw-maintainers] gsed in testing
Message-ID: <1239047388-sup-4875@frog.chass.utoronto.ca>

Hello,

A revised version of gsed 4.1.5 is in testing.  This version does not
contain ISALIST in RPATH.

Ross


From Darin.Perusich at cognigencorp.com  Mon Apr  6 21:59:00 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 06 Apr 2009 15:59:00 -0400
Subject: [csw-maintainers] migrating inetd.conf entries to SMF for sol10
In-Reply-To: <20090406182152.GP73510@bolthole.com>
References: <49DA181F.8080305@cognigencorp.com>	<20090406173746.GL73510@bolthole.com>	<49DA45D1.8060008@cognigencorp.com>
	<20090406182152.GP73510@bolthole.com>
Message-ID: <49DA5F04.9010002@cognigencorp.com>

I changed the subject to better describe what I'm looking to do :-).

Philip Brown wrote:
> On Mon, Apr 06, 2009 at 02:11:29PM -0400, Darin Perusich wrote:
>> This is why I'm asking for documentation or an example of creating a
>> package in such a way. While SMF may be the "proper" way on 10 I'm going
>> to do what works, has always worked, and still does, init.d script and
>> inetconv for inetd things.
> 
> oh... so to be specific, you're looking at examples of how to handle stuff
> in inetd.conf for sol8/9, IN CONTRAST to the solaris 10 way?

I'm currently using postinstall script to check for the entries and add
them if they don't exist and nothing special for Solaris 10.

> You COULD make a package for it, that does nothing to the inetd conf.
> Or you could make an explicit postinstall script.
> 
> Or... you could be "noble", and make a new classutils class action script
> for it :)

This is what I was looking for, where are the docs/examples explaining
how this classutils works?

> so on sol8/9 it would auto insert/update/.... inetd.conf, and
>   in sol10 it would auto SMF it.
> 
> you could allow for a file like for the cswusergroup class....
> 
> /opt/csw/etc/pkg/[pkgname]/cswinetdconf
> 
> with one or more lines in inetdconf style, that would be the trigger
> for the class action script.
> Although you would also potentially require some entries for any
> "new" /etc/services entries, if they were required.

The existing postinstall already adds to services so that's taken care of.

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From william at wbonnet.net  Mon Apr  6 22:02:14 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 06 Apr 2009 22:02:14 +0200
Subject: [csw-maintainers] Thematics March '09 : Move to GAR
Message-ID: <49DA5FC6.6040700@wbonnet.net>

Hi all

The move to Gar month is now over, thank you all for your effort. As you 
can see from the statistics email, many packages have been added or 
updated (239 packages over March).

The current source contains 1939 packages, including 728 from the gar 
repository (38%). This is about 200 more than at the beginning of the 
month.

It is important to notice that the GAR repository is also having 261 
packages which are not yet released to current ! That would take us to 
2200 packages and 46% of packages from GAR.

Next milestone is to continue the GAR and package update. April 
thematics is package update, which is closely related to the GAR thematics.

Cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Mon Apr  6 22:07:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 13:07:59 -0700
Subject: [csw-maintainers] migrating inetd.conf entries to SMF for sol10
In-Reply-To: <49DA5F04.9010002@cognigencorp.com>
References: <49DA181F.8080305@cognigencorp.com>
	<20090406173746.GL73510@bolthole.com>
	<49DA45D1.8060008@cognigencorp.com>
	<20090406182152.GP73510@bolthole.com>
	<49DA5F04.9010002@cognigencorp.com>
Message-ID: <20090406200759.GU73510@bolthole.com>

On Mon, Apr 06, 2009 at 03:59:00PM -0400, Darin Perusich wrote:
> > Or... you could be "noble", and make a new classutils class action script
> > for it :)
> 
> This is what I was looking for, where are the docs/examples explaining
> how this classutils works?

its a solaris standard mechanism.
so, docs.sun.com would be one place to start.

or... you could just look at the existing class action scripts in our
cswclassutils package, and modify one to suit your needs.

the concept is that you write a matching pair of
install/remove scripts which go in /usr/sadm/[whateveritis]

Then at pkgadd/pkgrm time, the script gets called, with a list of files
that are in "class YOURCLASS"

The script can then do basically anything it wants.
It's very similar to a postinstall script, but with a few more "input" and
"output" expectations. nothing too strenous though. again, see our existing
ones.


> > so on sol8/9 it would auto insert/update/.... inetd.conf, and
> >   in sol10 it would auto SMF it.
> > 
> > you could allow for a file like for the cswusergroup class....
> > 
> > /opt/csw/etc/pkg/[pkgname]/cswinetdconf
> > 
> > with one or more lines in inetdconf style, that would be the trigger
> > for the class action script.
> > Although you would also potentially require some entries for any
> > "new" /etc/services entries, if they were required.
> 
> The existing postinstall already adds to services so that's taken care of.

However, if you converted it to a class action script, it would be 
Very Nice for other programs that may like to use inetd. 

Plus, eliminating postinstall scripts is nice for not prompting sysadmins,
"do you realy want to run this script?"


From phil at bolthole.com  Mon Apr  6 22:10:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 13:10:50 -0700
Subject: [csw-maintainers] Thematics March '09 : Move to GAR
In-Reply-To: <49DA5FC6.6040700@wbonnet.net>
References: <49DA5FC6.6040700@wbonnet.net>
Message-ID: <20090406201050.GV73510@bolthole.com>

On Mon, Apr 06, 2009 at 10:02:14PM +0200, William Bonnet wrote:
>...
> It is important to notice that the GAR repository is also having 261  
> packages which are not yet released to current ! That would take us to  
> 2200 packages and 46% of packages from GAR.
>
> Next milestone is to continue the GAR and package update. April  
> thematics is package update, which is closely related to the GAR 
> thematics.

however, one big GAR-related (unstated)milestone stil to be met, is:

"make gar its own downloadable package".

Where the package would consist of basic, more userfriendly tools that
would look something like:

$ gar-build-from-src somesoftware

and gar would then go grab the source, download, configure, compile and
package that, on a non-opencsw.or build box, without further prompting.


The concept being, that "modular gar", would at last be TRUELY modular, and
not have a need to svn-download the whole svn tree to do something useful.



From rupert at opencsw.org  Mon Apr  6 22:17:45 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 6 Apr 2009 22:17:45 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
Message-ID: <6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>

On Mon, Apr 6, 2009 at 16:43, Peter Bonivart <bonivart at opencsw.org> wrote:
> On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
>> what would happen if one lives without classutils?
>
> Disregarding that you have to force past the dependency requirement it
> would be the equivalent of not running pre/post scripts. The packages
> that depend on cswclassutils use it for some combo of e.g. init/SMF
> support and handling of configuration files, see
> http://wiki.opencsw.org/cswclassutils-package for all features.
>
> What's the problem to install it in the global zone? Even if you don't
> have root access to it yourself you should be able to make a request
> for a package install, right? What would you do with a sparse zone if
> you needed a SUNW package that installed in /usr?

we would need to make a request. as we have 1000 machines, and only a
few of them run opencsw it would be denied.

i am wondering how it was done in the past? because we never had write
access and it always seemed to work. also the install description only
states /opt/csw as mount point.

rupert.


From phil at bolthole.com  Mon Apr  6 22:28:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 13:28:17 -0700
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
Message-ID: <20090406202817.GB51303@bolthole.com>

On Mon, Apr 06, 2009 at 10:17:45PM +0200, rupert THURNER wrote:
> On Mon, Apr 6, 2009 at 16:43, Peter Bonivart <bonivart at opencsw.org> wrote:
> > On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
> >> what would happen if one lives without classutils?
> >
> > Disregarding that you have to force past the dependency requirement it
> > would be the equivalent of not running pre/post scripts. The packages
> > that depend on cswclassutils use it for some combo of e.g. init/SMF
> > support and handling of configuration files, see
> > http://wiki.opencsw.org/cswclassutils-package for all features.
> >
> > What's the problem to install it in the global zone? Even if you don't
> > have root access to it yourself you should be able to make a request
> > for a package install, right? What would you do with a sparse zone if
> > you needed a SUNW package that installed in /usr?
> 
> we would need to make a request. as we have 1000 machines, and only a
> few of them run opencsw it would be denied.
> 
> i am wondering how it was done in the past? because we never had write
> access and it always seemed to work. also the install description only
> states /opt/csw as mount point.

"in the past", we didnt use the class script as much.

I would suggest that you make the request. you dont lose much by making the
request, no?

Afer that, as Peter B. said, you would have to run whatever is needful, by
hand.

Which is one reason I have been pushing to have the things called by the
class scripts, in /opt/csw rather than /etc.
This allows for tools to be written to do what is neccessary.

You are basically in the same circumstances as kind of a
 "NFS-shared /opt/csw" model. Perhaps this might motivate you to help
write some of those tools? :-)
The good news is, you would only have to do it once for each cswclassutils
script that we have.



From bonivart at opencsw.org  Mon Apr  6 22:39:58 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Apr 2009 22:39:58 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com> 
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com> 
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com> 
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com> 
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
Message-ID: <625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>

On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
> we would need to make a request. as we have 1000 machines, and only a
> few of them run opencsw it would be denied.

Why would it be denied? It's not like it would hurt the machines not
using CSW, it's something like a 20 KB package containing 10 files
which has to be specifically called to do anything.

Not even my company would deny that request. :-)

-- 
/peter


From rupert at opencsw.org  Mon Apr  6 22:50:31 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 6 Apr 2009 22:50:31 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
Message-ID: <6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>

On Mon, Apr 6, 2009 at 22:39, Peter Bonivart <bonivart at opencsw.org> wrote:
> On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
>> we would need to make a request. as we have 1000 machines, and only a
>> few of them run opencsw it would be denied.
>
> Why would it be denied? It's not like it would hurt the machines not
> using CSW, it's something like a 20 KB package containing 10 files
> which has to be specifically called to do anything.
>

because all machines look the same on the base layer. and upgrading
all machines is a major thing. and waiting for the next operating
system release as well (because it takes so much time).

it was already difficult to get the /opt/csw mountpoint.

but i'll try :)

rupert.


From phil at bolthole.com  Mon Apr  6 23:05:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 14:05:26 -0700
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
	<6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
Message-ID: <20090406210526.GC51303@bolthole.com>

On Mon, Apr 06, 2009 at 10:50:31PM +0200, rupert THURNER wrote:
> On Mon, Apr 6, 2009 at 22:39, Peter Bonivart <bonivart at opencsw.org> wrote:
> > On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
> >> we would need to make a request. as we have 1000 machines, and only a
> >> few of them run opencsw it would be denied.
> >
> > Why would it be denied? It's not like it would hurt the machines not
> > using CSW, it's something like a 20 KB package containing 10 files
> > which has to be specifically called to do anything.
> >
> 
> because all machines look the same on the base layer. and upgrading
> all machines is a major thing. and waiting for the next operating
> system release as well (because it takes so much time).
> 
> it was already difficult to get the /opt/csw mountpoint.
> 
> but i'll try :)
> 

If it's a "these things change only once a year, if you're LUCKY" thing...
then it would almost be worse to be stuck with an older version of the
thing.
your requested change should be more along the lines of "please install a
symlink to point to /etc/csw/[somehackplace]" so that you can deploy
updates to the class utils in a timely fashion.


But i think you'll probably be best off trying my "treat as read-only
/opt/csw nfs share" approach.


From skayser at opencsw.org  Tue Apr  7 00:01:20 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Apr 2009 00:01:20 +0200
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>	<49D4E784.7070605@opencsw.org>	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
Message-ID: <49DA7BB0.6050807@opencsw.org>

Dagobert Michelsen wrote:
> Am 02.04.2009 um 18:34 schrieb Philip Brown:
>> On Thu, Apr 02, 2009 at 11:27:48AM -0500, Mike Watters wrote:
>>
>>> for some reason, /opt/csw/include/ncurses does not exist?
>>> it exists as /opt/csw/include/ncursesw
>> sounds like a takeover maintainer bug
> 
> There is now an updated ncurses in testing which has both
> include/ncurses and include/ncursesw.

Thanks, would you mind pushing that onto the build farm?

Sebastian


From rupert at opencsw.org  Tue Apr  7 00:45:32 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 7 Apr 2009 00:45:32 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <20090406210526.GC51303@bolthole.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
	<6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
	<20090406210526.GC51303@bolthole.com>
Message-ID: <6af4270904061545y730a8a2r30d93cb6569fa9d7@mail.gmail.com>

On Mon, Apr 6, 2009 at 23:05, Philip Brown <phil at bolthole.com> wrote:
> On Mon, Apr 06, 2009 at 10:50:31PM +0200, rupert THURNER wrote:
>> On Mon, Apr 6, 2009 at 22:39, Peter Bonivart <bonivart at opencsw.org> wrote:
>> > On Mon, Apr 6, 2009 at 10:17 PM, rupert THURNER <rupert at opencsw.org> wrote:
>> >> we would need to make a request. as we have 1000 machines, and only a
>> >> few of them run opencsw it would be denied.
>> >
>> > Why would it be denied? It's not like it would hurt the machines not
>> > using CSW, it's something like a 20 KB package containing 10 files
>> > which has to be specifically called to do anything.
>> >
>>
>> because all machines look the same on the base layer. and upgrading
>> all machines is a major thing. and waiting for the next operating
>> system release as well (because it takes so much time).
>>
>> it was already difficult to get the /opt/csw mountpoint.
>>
>> but i'll try :)
>>
>
> If it's a "these things change only once a year, if you're LUCKY" thing...
> then it would almost be worse to be stuck with an older version of the
> thing.
> your requested change should be more along the lines of "please install a
> symlink to point to /etc/csw/[somehackplace]" so that you can deploy
> updates to the class utils in a timely fashion.

will do. but will this be sufficient, as the error was:
/etc/opt/csw/init.d/csw.smf.sample
...
pkgadd.ORG: ERROR: unable to open
</usr/sadm/install/scripts/i.cswcpsampleconf> for writing: (30)
Read-only file system

or this was more the kind of follow-up error?

>
> But i think you'll probably be best off trying my "treat as read-only
> /opt/csw nfs share" approach.

here every machine is installed separately, but this is quite
automated. i am not aware that we use nfs. currently we install in a
writable area, and then mount into /opt/csw as stated in your
description. which works very good.


rupert.


From bwalton at opencsw.org  Tue Apr  7 00:55:34 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 06 Apr 2009 18:55:34 -0400
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <49DA7BB0.6050807@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
Message-ID: <1239058478-sup-114@ntdws12.chass.utoronto.ca>

Excerpts from Sebastian Kayser's message of Mon Apr 06 18:01:20 -0400 2009:
> > There is now an updated ncurses in testing which has both
> > include/ncurses and include/ncursesw.
> 
> Thanks, would you mind pushing that onto the build farm?

It likely wouldn't hurt in this case, but the official policy is that
no unreleased packages get installed.  As soon as it's available
through the normal channels, it can be added.

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090406/3d768b0e/attachment-0001.asc>

From phil at bolthole.com  Tue Apr  7 00:55:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 15:55:52 -0700
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <6af4270904061545y730a8a2r30d93cb6569fa9d7@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<625385e30904061339p55fda948g20691085795e2134@mail.gmail.com>
	<6af4270904061350t31bbe547h4fcd31937492f88@mail.gmail.com>
	<20090406210526.GC51303@bolthole.com>
	<6af4270904061545y730a8a2r30d93cb6569fa9d7@mail.gmail.com>
Message-ID: <20090406225552.GB3314@bolthole.com>

hi Rupert,

I think we are not communicating well, because I'm trying to give you too
much information, which you arent ready for :-)

Simplified approach suggested for you:

since you have /opt/csw mounted from "somewhere else", also mount
    /usr/sadm/install/scripts from 'somewhere else'.

Have that place populated by both the solaris standard, AND csw, scripts
that belong in there.



> >
> > But i think you'll probably be best off trying my "treat as read-only
> > /opt/csw nfs share" approach.
> 
> here every machine is installed separately, but this is quite
> automated. i am not aware that we use nfs. currently we install in a
> writable area, and then mount into /opt/csw as stated in your
> description. which works very good.


From rupert at opencsw.org  Tue Apr  7 01:06:07 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 7 Apr 2009 01:06:07 +0200
Subject: [csw-maintainers] cswclassutils install failed
In-Reply-To: <20090406202817.GB51303@bolthole.com>
References: <94E2A58978FC324196A142DE8399AB7801A9FF0C@chsa1556.share.beluni.net>
	<6af4270904030353h53b4c21dqbad3dcb3372b5105@mail.gmail.com>
	<625385e30904030357u8e8e51aqf42e7732e146d313@mail.gmail.com>
	<6af4270904060248p672ea03cv3010a2798d286ac4@mail.gmail.com>
	<625385e30904060743y2cce4831k16784e55c91eadac@mail.gmail.com>
	<6af4270904061317ibbd4afeka545248de6dd3262@mail.gmail.com>
	<20090406202817.GB51303@bolthole.com>
Message-ID: <6af4270904061606n67fd1a4radfa86efc738069b@mail.gmail.com>

On Mon, Apr 6, 2009 at 22:28, Philip Brown <phil at bolthole.com> wrote:
> On Mon, Apr 06, 2009 at 10:17:45PM +0200, rupert THURNER wrote:
>> On Mon, Apr 6, 2009 at 16:43, Peter Bonivart <bonivart at opencsw.org> wrote:
>> > On Mon, Apr 6, 2009 at 11:48 AM, rupert THURNER <rupert at opencsw.org> wrote:
>> >> what would happen if one lives without classutils?
>> >
>> > Disregarding that you have to force past the dependency requirement it
>> > would be the equivalent of not running pre/post scripts. The packages
>> > that depend on cswclassutils use it for some combo of e.g. init/SMF
>> > support and handling of configuration files, see
>> > http://wiki.opencsw.org/cswclassutils-package for all features.
>> >
>> > What's the problem to install it in the global zone? Even if you don't
>> > have root access to it yourself you should be able to make a request
>> > for a package install, right? What would you do with a sparse zone if
>> > you needed a SUNW package that installed in /usr?
>>
>> we would need to make a request. as we have 1000 machines, and only a
>> few of them run opencsw it would be denied.
>>
>> i am wondering how it was done in the past? because we never had write
>> access and it always seemed to work. also the install description only
>> states /opt/csw as mount point.
>
> "in the past", we didnt use the class script as much.
>
> I would suggest that you make the request. you dont lose much by making the
> request, no?
>
> Afer that, as Peter B. said, you would have to run whatever is needful, by
> hand.
>
> Which is one reason I have been pushing to have the things called by the
> class scripts, in /opt/csw rather than /etc.
> This allows for tools to be written to do what is neccessary.
>
> You are basically in the same circumstances as kind of a
> ?"NFS-shared /opt/csw" model. Perhaps this might motivate you to help
> write some of those tools? :-)
> The good news is, you would only have to do it once for each cswclassutils
> script that we have.

could you point me to an example?

rupert.


From skayser at opencsw.org  Tue Apr  7 01:10:55 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Apr 2009 01:10:55 +0200
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <20090406220708.GD51303@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
	<20090406220708.GD51303@bolthole.com>
Message-ID: <49DA8BFF.9070809@opencsw.org>

Philip Brown wrote:
> On Tue, Apr 07, 2009 at 12:01:20AM +0200, Sebastian Kayser wrote:
>>> There is now an updated ncurses in testing which has both
>>> include/ncurses and include/ncursesw.
>> Thanks, would you mind pushing that onto the build farm?
> 
> that violates the build farm principles.
> only released packages.

Oops, right. No harm intended.

Just installed the package on my local build machine and successfully
tested watch and ncdu builds against regular ncurses. mcabber building
with the ncursesw includes bails out, will have a look at that
tomorrow. Maybe it looks obvious to someone.

source='commands.c' object='commands.o' libtool=no \
        DEPDIR=.deps depmode=none /bin/bash ../depcomp \
        /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..  -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
"/usr/include/iso/wctype_c99.h", line 48: syntax error before or at: wctype
"/usr/include/iso/wctype_c99.h", line 48: warning: undefined or missing type for: wctype
"/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:  empty declaration
"commands.c", line 110: cannot recover from previous errors
cc: acomp failed for commands.c

Sebastian


From phil at bolthole.com  Tue Apr  7 01:24:38 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Apr 2009 16:24:38 -0700
Subject: [csw-maintainers] Updated ncurses available in testing/
In-Reply-To: <49DA8BFF.9070809@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
	<20090406220708.GD51303@bolthole.com>
	<49DA8BFF.9070809@opencsw.org>
Message-ID: <20090406232438.GD3314@bolthole.com>

On Tue, Apr 07, 2009 at 01:10:55AM +0200, Sebastian Kayser wrote:
> 
> Just installed the package on my local build machine and successfully
> tested watch and ncdu builds against regular ncurses. mcabber building
> with the ncursesw includes bails out, will have a look at that
> tomorrow. Maybe it looks obvious to someone.
> 

looks fairly obvious to me. 
Solaris system includes are getting looked at before the stuff in
wncurses dir. This breaks things for ncurses.

So either you fiddle with your -I optoins to the compiler, or you have to
take a look at the source code, to juggle order of whatever is causing
an #include of /usr/include/iso/wctype_c99.h, to come later, instead of
sooner.


(or... simply drop trying to use ncurses, and use the system standard
  curses libs :)



> source='commands.c' object='commands.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..  -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
> "/usr/include/iso/wctype_c99.h", line 48: syntax error before or at: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: undefined or missing type for: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:  empty declaration
> "commands.c", line 110: cannot recover from previous errors
> cc: acomp failed for commands.c
> 


From hson at opencsw.org  Tue Apr  7 08:08:54 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Tue, 07 Apr 2009 08:08:54 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <20090406.14345000.1346203350@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>
	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>
Message-ID: <49DAEDF6.8050404@opencsw.org>

James Lee wrote:
> 
> It expands to the isalist.  It tries all in order but if it's a 32 bit
> prog there is never a valid match in the 64s, then it runs through the
> 32s until the match, or not in the case of this first non CSW lib:
> 

And your solution to that "problem" is?
Add all 32bit paths to RPATH/RUNPATH when building a 32bit app/lib, 
instead of just $ISALIST?

> Once built a package doesn't know how it was built so if it works
> use it.

Parse error...

> LD_OPTIONS overrides all so is powerful.  Like all powerful weapons
> use with caution.
> 

My point was that LD_OPTIONS is used by default when building using gar.
And those broken RPATH/RUNPATHs was due to a bug in gar, not what the 
maintainer might have set (or not set).

>> To reiterate my question:
>> Is there any actual harm (disregarding the fact that the link loader
>> might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without
>> finding any match) having RPATH/RUNPATH set?
> 
> No, well it's actual but trivial.
> 
> 
>> If not, it is only when a package doesn't use any libs at all from
>> /opt/csw/lib, when RPATH/RUNPATH should be changed from the default.
> 
> Correct, as is the case with zlib.

I'm not thinking of just zlib, I'm speaking in a more general term plus 
"should" should have been "could".

Since we have established that its not harmful to have RPATH set even 
when its not used, I can see more problem (as in more work for the 
maintainer in the future) with actively unsetting LD_OPTIONS than not.


>> If package A is linked to a lib i package B (both CSW packages) and A i
>> packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future
>> releases of B which contain optimized libraries won't be used until A is
>> repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/csw/lib
> 
> Correct.  Add $ISALIST in, unless you know it won't be needed.

And my point is that you can't know if its going to be needed or not at 
the time of package (unless you are both maintaner of A and B and know 
that you won't package optimized version of B).

>> And if there is a longer dependency list (A->B->C-D where D is the one
>> with optimized libs) it creates much more work for the maintainer of A
>> (he must check every lib in the dependency path to see if there are any
>> optimized libs).
> 
> D is governed by the RPATH of C.  C by B.

You're correct (at least if C and B are linked with required libraries, 
reference to my post the other week regarding gpdf and libnet), my 
memory was wrong.





From dam at opencsw.org  Tue Apr  7 08:56:05 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 08:56:05 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <49DAEDF6.8050404@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>
	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>
	<49DAEDF6.8050404@opencsw.org>
Message-ID: <C860D70E-A2E4-40AC-BD8F-3428F43E83F9@opencsw.org>

Hi Roger,

Am 07.04.2009 um 08:08 schrieb Roger H?kansson:
> And those broken RPATH/RUNPATHs was due to a bug in gar, not what  
> the maintainer might have set (or not set).

To be precise here: The method $ISALIST was passed was due to
multiple expansion, where the number of expansions depended on
the package and it was the responsibility of the maintainer to
ensure proper quoting. It was my fault to not have made this
clear and give advice on how to properly set it.

> Since we have established that its not harmful to have RPATH set  
> even when its not used, I can see more problem (as in more work for  
> the maintainer in the future) with actively unsetting LD_OPTIONS  
> than not.

It is possible to include other ways of passing -R to the linker
that don't interfere with what the maintainer intended. We can
work on these cases when they occur.

>>> If package A is linked to a lib i package B (both CSW packages)  
>>> and A i
>>> packaged with RPATH/RUNPATH set to only /opt/csw/lib, any future
>>> releases of B which contain optimized libraries won't be used  
>>> until A is
>>> repackaged with RPATH/RUNPATH set to /opt/csw/lib/$ISALIST:/opt/ 
>>> csw/lib
>> Correct.  Add $ISALIST in, unless you know it won't be needed.
>
> And my point is that you can't know if its going to be needed or not  
> at the time of package (unless you are both maintaner of A and B and  
> know that you won't package optimized version of B).

That's why GAR adds $ISALIST per default, both in the old version
and now. If you don't want it you have to turn it off, the unaware gets
optimizations.


Best regards

   -- Dago

From bonivart at opencsw.org  Tue Apr  7 16:11:09 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Tue, 7 Apr 2009 16:11:09 +0200
Subject: [csw-maintainers] /testing perl-ldap 0.39
Message-ID: <625385e30904070711l4a3a4f14uf05bcc8d508b1ee8@mail.gmail.com>

I got a request to update perl-ldap but I'm not a user myself (current
maintainer is Alex Moore who is retired) so I could need some help
with testing.

http://mirror.opencsw.org/testing/pm_ldap-0.39,REV=2009.04.07-SunOS5.8-all-CSW.pkg.gz

-- 
/peter


From dam at opencsw.org  Tue Apr  7 17:23:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 17:23:07 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
Message-ID: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>

Hi,

in the past were requests for installation of experimental packages not
released yet to see how things go. However, the strict "released only"
policy on build8s prohibited this installation. To allow these
"experimental packages" I set up a new zone build8st (t=testing)
where packages from testing may be installed. Building packages
for release is not allowed on this machine! I am thinking of
entering both testing/ and current/ into the repository from pkgutil
and allow csw-group execution of pkgutil. Thought?


Best regards

   -- Dago


From mwatters at opencsw.org  Tue Apr  7 18:26:12 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 11:26:12 -0500
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
Message-ID: <49DB7EA4.2060302@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> in the past were requests for installation of experimental packages not
> released yet to see how things go. However, the strict "released only"
> policy on build8s prohibited this installation. To allow these
> "experimental packages" I set up a new zone build8st (t=testing)
> where packages from testing may be installed. Building packages
> for release is not allowed on this machine! I am thinking of
> entering both testing/ and current/ into the repository from pkgutil
> and allow csw-group execution of pkgutil. Thought?
> 
> 
> Best regards
> 
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Sounds good, one caveat

if all maintainers have access to add / remove packages
we will need a way to coordinate install/remove of packages.



- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbfqMACgkQLrhmsXMSLxd8MQCfXiysqx4Tc6vBBU2xhoUdys0A
AGAAnjGyHTWFhRTQRIUaArEhNRlHYYQC
=w6mQ
-----END PGP SIGNATURE-----


From trygvis at opencsw.org  Tue Apr  7 18:37:56 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 07 Apr 2009 18:37:56 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <49DB7EA4.2060302@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
	<49DB7EA4.2060302@opencsw.org>
Message-ID: <49DB8164.6030605@opencsw.org>

Mike Watters wrote:
> Dagobert Michelsen wrote:
>> Hi,
> 
>> in the past were requests for installation of experimental packages not
>> released yet to see how things go. However, the strict "released only"
>> policy on build8s prohibited this installation. To allow these
>> "experimental packages" I set up a new zone build8st (t=testing)
>> where packages from testing may be installed. Building packages
>> for release is not allowed on this machine! I am thinking of
>> entering both testing/ and current/ into the repository from pkgutil
>> and allow csw-group execution of pkgutil. Thought?

Yay!

> Sounds good, one caveat
> 
> if all maintainers have access to add / remove packages
> we will need a way to coordinate install/remove of packages.

Does it require anything other that just doing it? If it is in testing/,
it should be installed? If it is removed, remove it etc.

--
Trygve


From mwatters at opencsw.org  Tue Apr  7 19:49:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 12:49:52 -0500
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <49DB8164.6030605@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>	<49DB7EA4.2060302@opencsw.org>
	<49DB8164.6030605@opencsw.org>
Message-ID: <49DB9240.30304@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Trygve Laugst?l wrote:
> Mike Watters wrote:
>> Dagobert Michelsen wrote:
>>> Hi,
>>> in the past were requests for installation of experimental packages not
>>> released yet to see how things go. However, the strict "released only"
>>> policy on build8s prohibited this installation. To allow these
>>> "experimental packages" I set up a new zone build8st (t=testing)
>>> where packages from testing may be installed. Building packages
>>> for release is not allowed on this machine! I am thinking of
>>> entering both testing/ and current/ into the repository from pkgutil
>>> and allow csw-group execution of pkgutil. Thought?
> 
> Yay!
> 
>> Sounds good, one caveat
>>
>> if all maintainers have access to add / remove packages
>> we will need a way to coordinate install/remove of packages.
> 
> Does it require anything other that just doing it? If it is in testing/,
> it should be installed? If it is removed, remove it etc.
> 
> --
> Trygve
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

My point is I would like to make sure no one removes any packages another
maintainer is currently testing with.

example, I am currently working on gcc4 packages,
if someone is doing a test compile against a set of libraries using gcc4
I don't think they would appreciate if I uninstalled gcc out from under them.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbkkAACgkQLrhmsXMSLxen7ACfTTXiWmVXwAZbdV8PKuN2Ulue
NNcAn1WEPSdNilFfyJ0zFSyJbRurDJ3S
=QfDd
-----END PGP SIGNATURE-----


From dam at opencsw.org  Tue Apr  7 21:25:00 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 21:25:00 +0200
Subject: [csw-maintainers] Updating all buildfarm servers to current/
Message-ID: <92AAA77E-031F-427A-BB86-1B6814FF1C84@opencsw.org>

Hi,

I am updating all buildfarm servers to current now.


Best regards

   -- Dago


From dam at opencsw.org  Tue Apr  7 21:45:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 7 Apr 2009 21:45:42 +0200
Subject: [csw-maintainers] Updated ncurses released
In-Reply-To: <49DA8BFF.9070809@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>
	<49D4D371.7070102@opencsw.org>
	<20090402162325.GG19124@bolthole.com>
	<49D4E784.7070605@opencsw.org>
	<20090402163457.GI19124@bolthole.com>
	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>
	<49DA7BB0.6050807@opencsw.org>
	<20090406220708.GD51303@bolthole.com>
	<49DA8BFF.9070809@opencsw.org>
Message-ID: <BF9FD469-989D-4992-827B-AD445D492AB4@opencsw.org>

Hi,

Am 07.04.2009 um 01:10 schrieb Sebastian Kayser:
> Philip Brown wrote:
>> On Tue, Apr 07, 2009 at 12:01:20AM +0200, Sebastian Kayser wrote:
>>>> There is now an updated ncurses in testing which has both
>>>> include/ncurses and include/ncursesw.
>>> Thanks, would you mind pushing that onto the build farm?
>>
>> that violates the build farm principles.
>> only released packages.
>
> Oops, right. No harm intended.
>
> Just installed the package on my local build machine and successfully
> tested watch and ncdu builds against regular ncurses. mcabber building
> with the ncursesw includes bails out, will have a look at that
> tomorrow. Maybe it looks obvious to someone.
>
> source='commands.c' object='commands.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..   
> -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/ 
> opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/ 
> csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ 
> ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
> "/usr/include/iso/wctype_c99.h", line 48: syntax error before or  
> at: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: undefined or  
> missing type for: wctype
> "/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:   
> empty declaration
> "commands.c", line 110: cannot recover from previous errors
> cc: acomp failed for commands.c

The CSWncurses package is fixed, released and installed on all buildfarm
servers.


Best regards

   -- Dago


From skayser at opencsw.org  Tue Apr  7 22:40:55 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Apr 2009 22:40:55 +0200
Subject: [csw-maintainers] Thanks! (was Re: [csw-users] python 2.6.1 now in
	testing)
In-Reply-To: <49D3E0F5.2060809@opencsw.org>
References: <49D3E0F5.2060809@opencsw.org>
Message-ID: <49DBBA57.3050001@opencsw.org>

Mike Watters wrote:
> Changes:
> 
> 	bug fix: 3510 'import curses' fails in current Python...
> 			forced curses to use the library in /opt/csw
> 	bug fix: 3579 RPATH stuff
> 			added NOISALIST = 1

Btw. Mike, i just happened to be at one of my distant colleagues over at
the development unit today and he was struggling to get Mercurial
running with the sunfreeware.com stack (bailed out with some python _md5
bla issues on commits ... he just wanted to use it, not troubleshoot it).

He was already pondering about alternatives and i told him to give
opencsw.org a try. He was happy to give it a try and even more happy to
see that everything worked out of the box :) He remarked that the
Mercurial over at sunfreeware.com is more recent but for his needs ours
is sufficient ... needless to say that any Mercurial is more useful than
one that is more recent but somehow broken ;)

Thanks to you and Trygvis (who takes care of hg)!

Sebastian


From yann at pleiades.fr.eu.org  Tue Apr  7 23:13:34 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Tue, 07 Apr 2009 23:13:34 +0200
Subject: [csw-maintainers] bash_completion and modification of the version
	numbering scheme
Message-ID: <49DBC1FE.9080803@pleiades.fr.eu.org>

Hi everybody,

You will find in testing a new release of the bash_completion package:
http://buildfarm.opencsw.org/testing/bash_completion-1.0,REV=2009.04.07-SunOS5.8-all-CSW.pkg.gz

In addition to upstream completion, this package provides some 
completion for pkgutil, pkg-get, pkgrm and smf tools. Testing and 
feedback are welcome.

I have however a problem with this package, upstream changed its version 
numbering scheme from $YEAR$MONTH$DAY to standard X.X.

 From what I remember, pkgutil uses the REV field if available so it 
would not be a problem with it but pkg-get doesn't not seem to rely on 
it so will never upgrade the package.
Am I right here and is there a way to correctly handle this problem ?

Thanks in advance.

Yann


From mwatters at opencsw.org  Tue Apr  7 23:27:30 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 16:27:30 -0500
Subject: [csw-maintainers] bash_completion and modification of the
 version numbering scheme
In-Reply-To: <49DBC1FE.9080803@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org>
Message-ID: <49DBC542.30304@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yann Rouillard wrote:
> Hi everybody,
> 
> You will find in testing a new release of the bash_completion package:
> http://buildfarm.opencsw.org/testing/bash_completion-1.0,REV=2009.04.07-SunOS5.8-all-CSW.pkg.gz
> 
> 
> In addition to upstream completion, this package provides some
> completion for pkgutil, pkg-get, pkgrm and smf tools. Testing and
> feedback are welcome.
> 
> I have however a problem with this package, upstream changed its version
> numbering scheme from $YEAR$MONTH$DAY to standard X.X.
> 
> From what I remember, pkgutil uses the REV field if available so it
> would not be a problem with it but pkg-get doesn't not seem to rely on
> it so will never upgrade the package.
> Am I right here and is there a way to correctly handle this problem ?
> 
> Thanks in advance.
> 
> Yann
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Yann,
	I am by no means an expert with Gar.
but if you change or add
# We define upstream file regex so we can be notified of new
# upstream software release

UFILES_REGEX = $(GARNAME)-(\d+(?:\.\d+)*).tar.bz2

to match the "NEW" schema it will pick up the new version as released upstream
HTH


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbxUEACgkQLrhmsXMSLxcFIgCdFIoR5h6NZGJ2r1jPd8pK5/Yd
JJYAmwdXWiACCg83BuO+o/YVbPwCGKzc
=8p1b
-----END PGP SIGNATURE-----


From yann at pleiades.fr.eu.org  Tue Apr  7 23:33:17 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Tue, 07 Apr 2009 23:33:17 +0200
Subject: [csw-maintainers] bash_completion and modification of the
 version numbering scheme
In-Reply-To: <49DBC542.30304@opencsw.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
Message-ID: <49DBC69D.1090407@pleiades.fr.eu.org>


Le 07.04.2009 23:27, Mike Watters a ?crit :
> Yann,
> 	I am by no means an expert with Gar.
> but if you change or add
> # We define upstream file regex so we can be notified of new
> # upstream software release
>
> UFILES_REGEX = $(GARNAME)-(\d+(?:\.\d+)*).tar.bz2
>
> to match the "NEW" schema it will pick up the new version as released upstream
> HTH
>

Hi Mike,

My problem is not with the check upstream code from gar but with pkg-get 
which will not accept to upgrade to the new package at it will always 
think it is older that the currently installed one

Yann




From mwatters at opencsw.org  Wed Apr  8 00:00:40 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Apr 2009 17:00:40 -0500
Subject: [csw-maintainers] bash_completion and modification of the
 version numbering scheme
In-Reply-To: <49DBC69D.1090407@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org>
Message-ID: <49DBCD08.2040805@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yann Rouillard wrote:
> 
> Le 07.04.2009 23:27, Mike Watters a ?crit :
>> Yann,
>>     I am by no means an expert with Gar.
>> but if you change or add
>> # We define upstream file regex so we can be notified of new
>> # upstream software release
>>
>> UFILES_REGEX = $(GARNAME)-(\d+(?:\.\d+)*).tar.bz2
>>
>> to match the "NEW" schema it will pick up the new version as released
>> upstream
>> HTH
>>
> 
> Hi Mike,
> 
> My problem is not with the check upstream code from gar but with pkg-get
> which will not accept to upgrade to the new package at it will always
> think it is older that the currently installed one
> 
> Yann
> 
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

sorry, I miss read the post. You can override the variable(s)
in the package set GARVERSION as x.x and have
SPKG_VERSION be YYYYMMDD (as package build date)


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknbzQcACgkQLrhmsXMSLxeKJgCfThwyK9HRYpK9Dh0xhbbzne1J
j/UAn0Qo7lPqpiRIgv22lx47SbbNbWoG
=nC2U
-----END PGP SIGNATURE-----


From bonivart at opencsw.org  Wed Apr  8 00:11:40 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 00:11:40 +0200
Subject: [csw-maintainers] bash_completion and modification of the
	version numbering scheme
In-Reply-To: <49DBCD08.2040805@opencsw.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org> <49DBCD08.2040805@opencsw.org>
Message-ID: <625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>

On Wed, Apr 8, 2009 at 12:00 AM, Mike Watters <mwatters at opencsw.org> wrote:
> sorry, I miss read the post. You can override the variable(s)
> in the package set GARVERSION as x.x and have
> SPKG_VERSION be YYYYMMDD (as package build date)

But why would we fake the upstream version and for how long should we
keep doing that? Some packages, like my geodb and possibly pca,
doesn't have versions at all so we assign something like YYMMDD to it
but in this case the end users would be confused, right?

This would also be really ugly:

20090408,REV=2009.04.08_rev=1.0

-- 
/peter


From bonivart at opencsw.org  Wed Apr  8 00:18:21 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 00:18:21 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
Message-ID: <625385e30904071518l767e9d5gf281d8a6dc9ce79b@mail.gmail.com>

On Tue, Apr 7, 2009 at 5:23 PM, Dagobert Michelsen <dam at opencsw.org> wrote:
> Hi,
>
> in the past were requests for installation of experimental packages not
> released yet to see how things go. However, the strict "released only"
> policy on build8s prohibited this installation. To allow these
> "experimental packages" I set up a new zone build8st (t=testing)
> where packages from testing may be installed. Building packages
> for release is not allowed on this machine! I am thinking of
> entering both testing/ and current/ into the repository from pkgutil
> and allow csw-group execution of pkgutil. Thought?

Is the zone for building or just for installing packages for those who
don't have Solaris 8 machines to test on?

-- 
/peter


From bonivart at opencsw.org  Wed Apr  8 00:23:06 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 00:23:06 +0200
Subject: [csw-maintainers] March 31st - last day of Solaris 8 support from
	Sun
Message-ID: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com>

The day went by quietly here but earlier we have agreed that after
this date we should no longer be required to build for Solaris 8. I
know that Dago has prepared Solaris 9 build machines but is it
officially ok to start using them?

-- 
/peter


From skayser at opencsw.org  Wed Apr  8 02:48:54 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 08 Apr 2009 02:48:54 +0200
Subject: [csw-maintainers] iswblank() and wctype.h woes (was Re: Updated
 ncurses available in testing/)
In-Reply-To: <20090406232438.GD3314@bolthole.com>
References: <49D3E0F5.2060809@opencsw.org>	<f95457130904020110q5bcf8cb5v869d5db76e60c113@mail.gmail.com>	<49D4D371.7070102@opencsw.org>	<20090402162325.GG19124@bolthole.com>	<49D4E784.7070605@opencsw.org>	<20090402163457.GI19124@bolthole.com>	<A191532C-3235-4B79-B9FC-369D2C9A16D5@opencsw.org>	<49DA7BB0.6050807@opencsw.org>	<20090406220708.GD51303@bolthole.com>	<49DA8BFF.9070809@opencsw.org>
	<20090406232438.GD3314@bolthole.com>
Message-ID: <49DBF476.2020409@opencsw.org>

Philip Brown wrote:
> On Tue, Apr 07, 2009 at 01:10:55AM +0200, Sebastian Kayser wrote:
>> Just installed the package on my local build machine and successfully
>> tested watch and ncdu builds against regular ncurses. mcabber building
>> with the ncursesw includes bails out, will have a look at that
>> tomorrow. Maybe it looks obvious to someone.
>>
> 
> looks fairly obvious to me. 
> Solaris system includes are getting looked at before the stuff in
> wncurses dir. This breaks things for ncurses.
> 
> So either you fiddle with your -I optoins to the compiler, or you have to
> take a look at the source code, to juggle order of whatever is causing
> an #include of /usr/include/iso/wctype_c99.h, to come later, instead of
> sooner.

Ok, fixed, multiple reasons. First of all i had been (mis)using a
Solaris 10 machine, so the system headers were different from the build
farm ones. On the build farm the build is fine.

Second - and that was the main "culprit" then - i had worked around the
missing iswblank() on Solaris 8 via the following snippet

# ifndef HAVE_ISWBLANK
# define iswblank(wc) iswctype(wc, wctype("blank"))
# endif

HAVE_ISWBLANK isn't defined by the mcabber autoconf system so it would
just always be included. No problem on Solaris 8 where iswblank() is
missing (and hence not used by any system headers), but on Solaris 10 it
is available and unfortunately i had put the snippet in a place where
the #define replaced system header occurences of iswblank() where
wctype() wasn't yet declared. Thus the observed error messages.

> (or... simply drop trying to use ncurses, and use the system standard
>   curses libs :)

No ncurses issue here ;)

>> source='commands.c' object='commands.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc -DHAVE_CONFIG_H -I. -I..  -I/opt/csw/include/glib-2.0 -I/opt/csw/lib/glib-2.0/include   -I/opt/csw/include -I/opt/csw/include  -I/opt/csw/lib/ncursesw -I/opt/csw/include  -xO3 -xarch=386 -features=no%extinl -I/opt/csw/lib/ncursesw -I/opt/csw/include -I/opt/csw/lib -D_GNU_SOURCE -c commands.c
>> "/usr/include/iso/wctype_c99.h", line 48: syntax error before or at: wctype
>> "/usr/include/iso/wctype_c99.h", line 48: warning: undefined or missing type for: wctype
>> "/usr/include/iso/wctype_c99.h", line 48: warning: syntax error:  empty declaration
>> "commands.c", line 110: cannot recover from previous errors
>> cc: acomp failed for commands.c

I have now moved the snippet after the includes to wctype.h (which fixes
the build even on the Solaris 10 box) and will check with the mcabber
author whether he can include an autoconf check for iswblank().

Sebastian


From hson at opencsw.org  Wed Apr  8 04:00:17 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 08 Apr 2009 04:00:17 +0200
Subject: [csw-maintainers] Building multiple packages from one source package
Message-ID: <49DC0531.6060006@opencsw.org>

I'm trying to build two packages from one source package and I've looked 
at some other packages (gettext for example), but I can't get it to work 
properly.

The first package CSWhtmldoc-common is build without any problem, but 
CSWhtmldoc fails since I'm trying to have CSWhtmldoc-common as a 
dependency of CSWhtmldoc.

Examining 'depend' file
P CSWfltk fltk - Fast light Tool Kit
P CSWjpeg jpeg - JPEG library and tools by the Independent JPEG Group
P CSWosslrt openssl_rt - Openssl runtime libraries
P CSWpng png - library for Portable Network Graphics format (PNG)
P CSWzlib zlib - Zlib Data Compression Library
P CSWhtmldoc-common htmldoc_common - This package contains the htmldoc 
files common to all architectures.
P CSWcommon common - common files and dirs for CSW packages

system      CSWcommon      common - common files and dirs for CSW packages
application CSWfltk        fltk - Fast light Tool Kit
ERROR: information for "CSWhtmldoc-common" was not found
ERROR: Invalid package CSWhtmldoc-common specified. Check of 
/tmp/htmldoc-1.8.27,REV=2009.04.08-SunOS5.8-sparc-CSW.pkg.gz fails


How do I make this work?


From bonivart at opencsw.org  Wed Apr  8 09:36:29 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 09:36:29 +0200
Subject: [csw-maintainers] Building multiple packages from one source
	package
In-Reply-To: <49DC0531.6060006@opencsw.org>
References: <49DC0531.6060006@opencsw.org>
Message-ID: <625385e30904080036g343cf0e6n421354471b807ffe@mail.gmail.com>

On Wed, Apr 8, 2009 at 4:00 AM, Roger H?kansson <hson at opencsw.org> wrote:
> How do I make this work?

As far as I know you can't pass the checks unless the package you're
depending on is installed on the build server. With our policy to not
install unreleased packages it will not work for split packages. I
disable the checks with:

ENABLE_CHECK=0

And test on another system instead. The cswutils package contains the
checkpkg script. I guess we can use the new build8st for this.

-- 
/peter


From james at opencsw.org  Wed Apr  8 12:05:43 2009
From: james at opencsw.org (James Lee)
Date: Wed, 08 Apr 2009 10:05:43 GMT
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <49DAEDF6.8050404@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>
	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>
	<49D87D21.4050606@opencsw.org>
	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>
	<49D9F70C.9030007@opencsw.org>
	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>
	<49DAEDF6.8050404@opencsw.org>
Message-ID: <20090408.10054300.1414585890@gyor.oxdrove.co.uk>

On 07/04/09, 07:08:54, "Roger H?kansson" <hson at opencsw.org> wrote regarding
Re: [csw-maintainers] RPATH/ISALIST (zlib package in /testing):

> > It expands to the isalist.  It tries all in order but if it's a 32 bit
> > prog there is never a valid match in the 64s, then it runs through the
> > 32s until the match, or not in the case of this first non CSW lib:
> >

> And your solution to that "problem" is?

Sun could make 32 progs use a 32 bit isalist, but this isn't "my"
problem, it's Sun's.  All I'm saying is that a good $ISALIST RPATH
causes more failed file look-ups than a bad one.


> Add all 32bit paths to RPATH/RUNPATH when building a 32bit app/lib,
> instead of just $ISALIST?

isalist expansion happens locally, that why/how it works.



> > Once built a package doesn't know how it was built so if it works
> > use it.

> Parse error...

Once a package is built the package doesn't know how the package was
built so if LD_OPTIONS works to make a package that works use
LD_OPTIONS to build the package.




> My point was that LD_OPTIONS is used by default when building using gar.
> And those broken RPATH/RUNPATHs was due to a bug in gar, not what the
> maintainer might have set (or not set).

The package doesn't know how the package was built - GAR is irrelevant
to the result.  The maintainer is responsible for what is in the package.



> >> To reiterate my question:
> >> Is there any actual harm (disregarding the fact that the link loader
> >> might have to traverse /opt/csw/lib/$ISALIST:/opt/csw/lib without
> >> finding any match) having RPATH/RUNPATH set?
> >
> > No, well it's actual but trivial.
> >
> >
> >> If not, it is only when a package doesn't use any libs at all from
> >> /opt/csw/lib, when RPATH/RUNPATH should be changed from the default.
> >
> > Correct, as is the case with zlib.

> I'm not thinking of just zlib, I'm speaking in a more general term plus
> "should" should have been "could".

"should" could have been "could".
"should" is right, but the RPATH needn't be right because the error is
trivial.


switch (person.getType())
{
    case PEDANT:
        message = "must";
        break;
    case MATHEMATICIAN:
        message = "should";
        break;
    case ENGINEER:
        message = "could";
        break;
    case ACCOUNTANT:
        message = "if it makes money";
        break;
    case BANKER:
        message = "if it makes me money";
        break;
    case TEENAGER:
        message = "yeah wotevva";
        break;



> Since we have established that its not harmful to have RPATH set even
> when its not used,

It's harmful but trivial.


> I can see more problem (as in more work for the
> maintainer in the future) with actively unsetting LD_OPTIONS than not.

Sorry I can't see why unsetting something has only been set by yourself
in the first place is more work.  In fact we should always explicitly
either set or unset such controlling environmental variables as a
matter of good practice.




James.


From hson at opencsw.org  Wed Apr  8 13:11:20 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 08 Apr 2009 13:11:20 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <20090408.10054300.1414585890@gyor.oxdrove.co.uk>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>	<49DAEDF6.8050404@opencsw.org>
	<20090408.10054300.1414585890@gyor.oxdrove.co.uk>
Message-ID: <49DC8658.8020908@opencsw.org>

James Lee wrote:
>>I can see more problem (as in more work for the
>> maintainer in the future) with actively unsetting LD_OPTIONS than not.
> 
> Sorry I can't see why unsetting something has only been set by yourself
> in the first place is more work.  In fact we should always explicitly
> either set or unset such controlling environmental variables as a
> matter of good practice.
> 

When building packages with gar, the default (i.e without the maintainer 
doing a thing) is that the RPATH will contain 
/opt/csw/lib/$ISALIST:/opt/csw/lib. If a maintainer is to package 
something with a different RPATH he must set it manually, i.e more work 
than default.


From yann at pleiades.fr.eu.org  Wed Apr  8 14:30:48 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Wed, 08 Apr 2009 14:30:48 +0200
Subject: [csw-maintainers] bash_completion and modification of
 the	version numbering scheme
In-Reply-To: <625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
References: <49DBC1FE.9080803@pleiades.fr.eu.org>
	<49DBC542.30304@opencsw.org>	<49DBC69D.1090407@pleiades.fr.eu.org>
	<49DBCD08.2040805@opencsw.org>
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
Message-ID: <49DC98F8.4060307@pleiades.fr.eu.org>


Le 08.04.2009 00:11, Peter Bonivart a ?crit :
> On Wed, Apr 8, 2009 at 12:00 AM, Mike Watters<mwatters at opencsw.org>  wrote:
>> sorry, I miss read the post. You can override the variable(s)
>> in the package set GARVERSION as x.x and have
>> SPKG_VERSION be YYYYMMDD (as package build date)
>
> But why would we fake the upstream version and for how long should we
> keep doing that? Some packages, like my geodb and possibly pca,
> doesn't have versions at all so we assign something like YYMMDD to it
> but in this case the end users would be confused, right?
>
> This would also be really ugly:
>
> 20090408,REV=2009.04.08_rev=1.0
>

I would also prefer to follow the upstream version numbering for the 
package, that seems closer to the standard 
http://www.opencsw.org/standards/build .

Why pkgutil and pkg-get don't have the same way to compare package 
versions ? It wasn't supposed to be standard ?

Yann


From bonivart at opencsw.org  Wed Apr  8 14:52:40 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 14:52:40 +0200
Subject: [csw-maintainers] bash_completion and modification of the
	version numbering scheme
In-Reply-To: <49DC98F8.4060307@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org> <49DBCD08.2040805@opencsw.org> 
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com> 
	<49DC98F8.4060307@pleiades.fr.eu.org>
Message-ID: <625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>

On Wed, Apr 8, 2009 at 2:30 PM, Yann Rouillard <yann at pleiades.fr.eu.org> wrote:
> Why pkgutil and pkg-get don't have the same way to compare package versions
> ? It wasn't supposed to be standard ?

I don't know how pkg-get does it, but for pkgutil we had a public
discussion on this list where James came with a suggestion and Dago
made a few changes if I remember correctly. I implemented it and I
have described it here in four rules:
http://pkgutil.wikidot.com/get-install-and-configure#toc7 (version
compare method).

The standard also hints about the REV-field becoming the standard way
of comparison for pkg-get, I assume it was written by Phil.

http://www.opencsw.org/standards/build#versioning

"Please note: the ",REV=YYYY.MM.DD" is now Mandatory. It provides a
fixed-format way of telling how recent the package really is, for
version comparison download purposes. At some point, it will be the
primary comparison key for pkg-get.(but not yet)"

-- 
/peter


From yann at pleiades.fr.eu.org  Wed Apr  8 17:53:38 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Wed, 08 Apr 2009 17:53:38 +0200
Subject: [csw-maintainers] bash_completion and modification of
 the	version numbering scheme
In-Reply-To: <625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>
References: <49DBC1FE.9080803@pleiades.fr.eu.org>
	<49DBC542.30304@opencsw.org>	<49DBC69D.1090407@pleiades.fr.eu.org>
	<49DBCD08.2040805@opencsw.org>
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
	<49DC98F8.4060307@pleiades.fr.eu.org>
	<625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>
Message-ID: <49DCC882.6000708@pleiades.fr.eu.org>

Le 08.04.2009 14:52, Peter Bonivart a ?crit :
> The standard also hints about the REV-field becoming the standard way
> of comparison for pkg-get, I assume it was written by Phil.
>
> http://www.opencsw.org/standards/build#versioning
>
> "Please note: the ",REV=YYYY.MM.DD" is now Mandatory. It provides a
> fixed-format way of telling how recent the package really is, for
> version comparison download purposes. At some point, it will be the
> primary comparison key for pkg-get.(but not yet)"
>

You're right, I missed that point.
Phil, do you intend to change the comparison code of pkg-get soon or do 
you think I should rather adopt the 
YEARMONTHDAY,REV=YEARMONTHDAY_rev=X.X package version numbering to be 
able to release bash_completion shortly ?

Yann


From Darin.Perusich at cognigencorp.com  Wed Apr  8 18:04:14 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 08 Apr 2009 12:04:14 -0400
Subject: [csw-maintainers] build8s and build9s inconsistancies
Message-ID: <49DCCAFE.2040900@cognigencorp.com>

I was building a package, amanda, on build8s and everything was going
fine but given that Solaris 8 is no longer supported I figured I'd move
over to the Solaris 9 build machines. The code compiles fine but the
staging of the package is incomplete on build9s, whether I use
'stagepkg' or 'gmake DESTDIR=$PWD/cswstage install' the results are the
same.

The staging fails when setting permission, see below, on some of the
binaries and I have no idea why. I compared the CSW packages and they
are identical on both machine so I'm wondering if its something in the OS.

Has anyone else run across this before?

gmake[5]: Entering directory `/home/darin/build/amanda-2.6.1/common-src'
  (dest=/opt/csw/sbin)
  (chown=amanda)
chown darin:sys
/home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt
chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt:
Not owner
gmake[5]: *** [installperms-data] Error 1
gmake[5]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[4]: *** [install-data-am] Error 2
gmake[4]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[3]: *** [install-am] Error 2
gmake[3]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[2]: *** [install] Error 2
gmake[2]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
gmake[1]: *** [install-recursive] Error 1
gmake[1]: Leaving directory `/home/darin/build/amanda-2.6.1'
gmake: *** [install] Error 2


-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Wed Apr  8 18:30:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 09:30:25 -0700
Subject: [csw-maintainers] bash_completion and modification of
	the	version numbering scheme
In-Reply-To: <49DCC882.6000708@pleiades.fr.eu.org>
References: <49DBC1FE.9080803@pleiades.fr.eu.org> <49DBC542.30304@opencsw.org>
	<49DBC69D.1090407@pleiades.fr.eu.org>
	<49DBCD08.2040805@opencsw.org>
	<625385e30904071511m2cdff35bkf6509a8d1288811d@mail.gmail.com>
	<49DC98F8.4060307@pleiades.fr.eu.org>
	<625385e30904080552w2c410965mf48a7f21e6201fba@mail.gmail.com>
	<49DCC882.6000708@pleiades.fr.eu.org>
Message-ID: <20090408163025.GS99983@bolthole.com>

On Wed, Apr 08, 2009 at 05:53:38PM +0200, Yann Rouillard wrote:
> You're right, I missed that point.
> Phil, do you intend to change the comparison code of pkg-get soon or do  
> you think I should rather adopt the  
> YEARMONTHDAY,REV=YEARMONTHDAY_rev=X.X package version numbering to be  
> able to release bash_completion shortly ?

I do intend to change it.... however, just how soon it happens, is somewhat
up in the air :-/

I may have time to add it next week.



From mwatters at opencsw.org  Wed Apr  8 19:44:17 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 12:44:17 -0500
Subject: [csw-maintainers] build8s and build9s inconsistancies
In-Reply-To: <49DCCAFE.2040900@cognigencorp.com>
References: <49DCCAFE.2040900@cognigencorp.com>
Message-ID: <49DCE271.9050304@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Darin Perusich wrote:
> I was building a package, amanda, on build8s and everything was going
> fine but given that Solaris 8 is no longer supported I figured I'd move
> over to the Solaris 9 build machines. The code compiles fine but the
> staging of the package is incomplete on build9s, whether I use
> 'stagepkg' or 'gmake DESTDIR=$PWD/cswstage install' the results are the
> same.
> 
> The staging fails when setting permission, see below, on some of the
> binaries and I have no idea why. I compared the CSW packages and they
> are identical on both machine so I'm wondering if its something in the OS.
> 
> Has anyone else run across this before?
> 
> gmake[5]: Entering directory `/home/darin/build/amanda-2.6.1/common-src'
>   (dest=/opt/csw/sbin)
>   (chown=amanda)
> chown darin:sys
> /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt:
> Not owner
> gmake[5]: *** [installperms-data] Error 1
> gmake[5]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[4]: *** [install-data-am] Error 2
> gmake[4]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[3]: *** [install-am] Error 2
> gmake[3]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[2]: *** [install] Error 2
> gmake[2]: Leaving directory `/home/darin/build/amanda-2.6.1/common-src'
> gmake[1]: *** [install-recursive] Error 1
> gmake[1]: Leaving directory `/home/darin/build/amanda-2.6.1'
> gmake: *** [install] Error 2
> 
> 

I can't even get onto build9s or build9x, I get prompted for password.
since home dir's are mounted across the env, I have to assume I don't have an
account on those servers.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc4nEACgkQLrhmsXMSLxfs0QCeISmYi9rNC40Z27vzmGkomKfL
SMQAoKBEs6B7kzxTHgjC19izgfiCs9yk
=lBSx
-----END PGP SIGNATURE-----


From pfelecan at opencsw.org  Wed Apr  8 19:47:32 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 08 Apr 2009 19:47:32 +0200
Subject: [csw-maintainers] static libraries lib*.a in our packages
Message-ID: <x6vdpfkpfv.fsf@abhavana.free.fr>

I cannot recollect precisely if we already had this discussion. Phil and
me, we had an exchange on this subject and I would like to have the
opinion of the other fellow maintainers as our point of view are not
totally convergent.

The questions are:

1. Is a good practice for CSW packages to contain library archives of
   the form lib*.a when we deliver dynamic libraries lib*.so ?

2. What's the potential usage for such libraries?

3. Which mechanism do you use for not building (e.g. ./configure
   --disable-static) or to post-install (in the sense of after make install)
   remove them.

Thank you in advance for your answers and comments.
-- 
Peter


From phil at bolthole.com  Wed Apr  8 20:01:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 11:01:50 -0700
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <x6vdpfkpfv.fsf@abhavana.free.fr>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
Message-ID: <20090408180150.GZ99983@bolthole.com>

On Wed, Apr 08, 2009 at 07:47:32PM +0200, Peter FELECAN wrote:
> I cannot recollect precisely if we already had this discussion. Phil and
> me, we had an exchange on this subject and I would like to have the
> opinion of the other fellow maintainers as our point of view are not
> totally convergent.
> 
> The questions are:
> 
> 1. Is a good practice for CSW packages to contain library archives of
>    the form lib*.a when we deliver dynamic libraries lib*.so ?
> 
> 2. What's the potential usage for such libraries?


To save some back-and-forth on the mailing list, here's a bit of history
for folks who were not around for the original discussions 4(?) years ago
:)

The perspective was that people very rarely ever use static libraries.
nowadays, pretty much every library is a dynamic library.
This helps both for memory sharing, and for ease of updates.
(you only have to update the lib package, not have to recompile
everything).

Occasionally, there may be benefits to making a static library for some
software available for optional use. However, when and if the maintainer
decides there is a benefit to providing a static lib, it is best to do so
in a "devel" subpackage (unless the library is relatively small).

that way, when a library package gets pulled in, solely as a result of
some other package having a dependancy on "libfoo.so", the user does not
needlessly download "libfoo.a" as well.

This is particularly important in the case of some packages, such as
opensp, where the static library size is *4 megabytes*.


so.... that's the background. 
Other folks, please add your thoughts to this :-)


From mwatters at opencsw.org  Wed Apr  8 20:07:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 13:07:52 -0500
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <x6vdpfkpfv.fsf@abhavana.free.fr>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
Message-ID: <49DCE7F8.1040005@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Peter FELECAN wrote:
> I cannot recollect precisely if we already had this discussion. Phil and
> me, we had an exchange on this subject and I would like to have the
> opinion of the other fellow maintainers as our point of view are not
> totally convergent.
> 
> The questions are:
> 
> 1. Is a good practice for CSW packages to contain library archives of
>    the form lib*.a when we deliver dynamic libraries lib*.so ?
> 
> 2. What's the potential usage for such libraries?
> 
> 3. Which mechanism do you use for not building (e.g. ./configure
>    --disable-static) or to post-install (in the sense of after make install)
>    remove them.
> 
> Thank you in advance for your answers and comments.

I would say if we are compiling a library that offers both static and shared
we compile them both.  there are specific instances when a user or an admin
may need to compile a tool statically ( albeit very rare ).

If we are compiling an "application" that offers both static and dynamic.
I would say compile them both and offer the static in a devel package along
with the headers.

My Rational:

There are companies that use small boxes as member interface points,
particularly Credit Card Companies.  these are boxes that live all over the
world, some have dial-up modems attached to them for access, some have x.25
interfaces (no I am not kidding), and bandwidth is at a premium.
The static binary may be bigger then the dynamic one, but on the whole, the one
static binary will be smaller then the possibly multiple packages.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc5/cACgkQLrhmsXMSLxdIfwCfVFuQg9lhvkxL2dWT5do0ILJz
150An3kCntXhYuPpijBXv3NIK+UaYzAG
=sNat
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Wed Apr  8 20:11:52 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Wed, 08 Apr 2009 14:11:52 -0400
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <20090408180150.GZ99983@bolthole.com>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
	<20090408180150.GZ99983@bolthole.com>
Message-ID: <1239214130-sup-790@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Wed Apr 08 14:01:50 -0400 2009:
> Occasionally, there may be benefits to making a static library for some
> software available for optional use. However, when and if the maintainer
> decides there is a benefit to providing a static lib, it is best to do so
> in a "devel" subpackage (unless the library is relatively small).
> 
> Other folks, please add your thoughts to this :-)

This lines up with the Debian policy and sounds quite reasonable to
me.  You shouldn't need the .a file unless you're building something
against it anyway, so the dev package is a good spot for it.

To save others the google, here is the debian policy link:
http://www.debian.org/doc/debian-policy/ch-sharedlibs.html#s-sharedlibs-static

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090408/43175e66/attachment-0001.asc>

From phil at bolthole.com  Wed Apr  8 20:17:04 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 11:17:04 -0700
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <49DCE7F8.1040005@opencsw.org>
References: <x6vdpfkpfv.fsf@abhavana.free.fr> <49DCE7F8.1040005@opencsw.org>
Message-ID: <20090408181704.GA99983@bolthole.com>

On Wed, Apr 08, 2009 at 01:07:52PM -0500, Mike Watters wrote:
> If we are compiling an "application" that offers both static and dynamic.
> I would say compile them both and offer the static in a devel package along
> with the headers.

Err.. did you mean "library"?

we do not offer statically compiled "applications". all of our stuff is
compiled against the static lib.
[i also dont see how an "application" belongs in a devel package]



Peter was not proposing offering our own statically compiled applications;
only offering more static libraries for people to compile things themselves
statically.



From mwatters at opencsw.org  Wed Apr  8 20:18:34 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 13:18:34 -0500
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <1239214130-sup-790@ntdws12.chass.utoronto.ca>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>	<20090408180150.GZ99983@bolthole.com>
	<1239214130-sup-790@ntdws12.chass.utoronto.ca>
Message-ID: <49DCEA7A.30702@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ben Walton wrote:
> Excerpts from Philip Brown's message of Wed Apr 08 14:01:50 -0400 2009:
>> Occasionally, there may be benefits to making a static library for some
>> software available for optional use. However, when and if the maintainer
>> decides there is a benefit to providing a static lib, it is best to do so
>> in a "devel" subpackage (unless the library is relatively small).
>>
>> Other folks, please add your thoughts to this :-)
> 
> This lines up with the Debian policy and sounds quite reasonable to
> me.  You shouldn't need the .a file unless you're building something
> against it anyway, so the dev package is a good spot for it.
> 
> To save others the google, here is the debian policy link:
> http://www.debian.org/doc/debian-policy/ch-sharedlibs.html#s-sharedlibs-static
> 
> -Ben
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Fedora/Red Hat policy follows this convention as well.

and the following line quotes directly and I think it is the best rational.

"A good rule of thumb is if the file is used for development and not needed for
the base package to run properly, it should go in -devel."
http://fedoraproject.org/wiki/Packaging:Guidelines
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc6noACgkQLrhmsXMSLxefPQCfRoH8YnaYTjy3ZSmgkDecW+lW
iwkAoOTkvFhT6YQ04ZbhcFuSvvqeWL3q
=TvKO
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Wed Apr  8 20:21:31 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Apr 2009 13:21:31 -0500
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <20090408181704.GA99983@bolthole.com>
References: <x6vdpfkpfv.fsf@abhavana.free.fr> <49DCE7F8.1040005@opencsw.org>
	<20090408181704.GA99983@bolthole.com>
Message-ID: <49DCEB2B.3080705@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Wed, Apr 08, 2009 at 01:07:52PM -0500, Mike Watters wrote:
>> If we are compiling an "application" that offers both static and dynamic.
>> I would say compile them both and offer the static in a devel package along
>> with the headers.
> 
> Err.. did you mean "library"?
> 
> we do not offer statically compiled "applications". all of our stuff is
> compiled against the static lib.
> [i also dont see how an "application" belongs in a devel package]
> 
> 
> 
> Peter was not proposing offering our own statically compiled applications;
> only offering more static libraries for people to compile things themselves
> statically.
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

sort of,  "application" would be like mysql.

I guess I should have written it as...
If we are compiling an "application", such as mysql, that offer
both static and dynamic libraries, compile both and include the static library
in the devel package.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknc6ysACgkQLrhmsXMSLxesRACdEfVe9sp2yBeBFHOvmtOI2GDu
6+QAoJJNuQyHp9v/0sBSBIg2uybecuzO
=QMoW
-----END PGP SIGNATURE-----


From hson at opencsw.org  Wed Apr  8 21:57:42 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 08 Apr 2009 21:57:42 +0200
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
Message-ID: <49DD01B6.2020301@opencsw.org>

I've got the impression that /opt/csw/var is a obsolete place to put 
things and that /var/opt/csw should be used instead.
Shouldn't this be reflected when using gar?
I.e gar.conf.mk should have localstatedir = /var/opt/csw instead of 
$(prefix)/var


From dam at opencsw.org  Wed Apr  8 21:58:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 21:58:45 +0200
Subject: [csw-maintainers] New zone with experimental packages: build8st
In-Reply-To: <625385e30904071518l767e9d5gf281d8a6dc9ce79b@mail.gmail.com>
References: <F92083D6-25AD-4A35-8E61-84B9D7401730@opencsw.org>
	<625385e30904071518l767e9d5gf281d8a6dc9ce79b@mail.gmail.com>
Message-ID: <F7C23A9B-8CA7-4F80-A453-07F744BBD4F7@opencsw.org>

Hi Peter,

Am 08.04.2009 um 00:18 schrieb Peter Bonivart:
> On Tue, Apr 7, 2009 at 5:23 PM, Dagobert Michelsen <dam at opencsw.org>  
> wrote:
>> in the past were requests for installation of experimental packages  
>> not
>> released yet to see how things go. However, the strict "released  
>> only"
>> policy on build8s prohibited this installation. To allow these
>> "experimental packages" I set up a new zone build8st (t=testing)
>> where packages from testing may be installed. Building packages
>> for release is not allowed on this machine! I am thinking of
>> entering both testing/ and current/ into the repository from pkgutil
>> and allow csw-group execution of pkgutil. Thought?
>
> Is the zone for building or just for installing packages for those who
> don't have Solaris 8 machines to test on?

It is for tests which require installation of packages from testing/,
for whatever this may be useful.


Best regards

    -- Dago


From dam at opencsw.org  Wed Apr  8 22:01:06 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:01:06 +0200
Subject: [csw-maintainers] March 31st - last day of Solaris 8 support
	from Sun
In-Reply-To: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com>
References: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com>
Message-ID: <71D80C94-0EBE-4E99-A281-C487FF52081B@opencsw.org>

Hi Peter,

Am 08.04.2009 um 00:23 schrieb Peter Bonivart:
> The day went by quietly here but earlier we have agreed that after
> this date we should no longer be required to build for Solaris 8. I
> know that Dago has prepared Solaris 9 build machines but is it
> officially ok to start using them?

Yes. Please note that it is no longer required to deliver for
Solaris 8, but it is still preferred. So if building on Solaris 8
is not harder than for Solaris 9, please continue to build on
Solaris 8.


Best regards

    -- Dago


From bonivart at opencsw.org  Wed Apr  8 22:05:05 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 22:05:05 +0200
Subject: [csw-maintainers] March 31st - last day of Solaris 8 support
	from Sun
In-Reply-To: <71D80C94-0EBE-4E99-A281-C487FF52081B@opencsw.org>
References: <625385e30904071523p6625cd43vfc8bc6a11a83f234@mail.gmail.com> 
	<71D80C94-0EBE-4E99-A281-C487FF52081B@opencsw.org>
Message-ID: <625385e30904081305o7c2b8d49o45f02be169da7e5@mail.gmail.com>

On Wed, Apr 8, 2009 at 10:01 PM, Dagobert Michelsen <dam at opencsw.org> wrote:
> Yes. Please note that it is no longer required to deliver for
> Solaris 8, but it is still preferred. So if building on Solaris 8
> is not harder than for Solaris 9, please continue to build on
> Solaris 8.

That sounds reasonable. In my case I will try to build on Solaris 8
first and if it works - fine, but I will not try to get upstream to
fix stuff for Solaris 8 like I recently did for ClamAV. Unless it
doesn't work on Solaris 9 either, then I will have to beg again. :-)

-- 
/peter


From dam at opencsw.org  Wed Apr  8 22:14:09 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:14:09 +0200
Subject: [csw-maintainers] RPATH/ISALIST (zlib package in /testing)
In-Reply-To: <49DC8658.8020908@opencsw.org>
References: <49D86AB2.5080005@opencsw.org>	<20090405.8572900.3105252976@gyor.oxdrove.co.uk>	<49D87D21.4050606@opencsw.org>	<20090406.8305100.1838829892@gyor.oxdrove.co.uk>	<49D9F70C.9030007@opencsw.org>	<20090406.14345000.1346203350@gyor.oxdrove.co.uk>	<49DAEDF6.8050404@opencsw.org>
	<20090408.10054300.1414585890@gyor.oxdrove.co.uk>
	<49DC8658.8020908@opencsw.org>
Message-ID: <9FC00626-A3C3-4B3E-BBCE-CE46C42683A4@opencsw.org>

Hi Roger,

Am 08.04.2009 um 13:11 schrieb Roger H?kansson:
> James Lee wrote:
>>> I can see more problem (as in more work for the
>>> maintainer in the future) with actively unsetting LD_OPTIONS than  
>>> not.
>> Sorry I can't see why unsetting something has only been set by  
>> yourself
>> in the first place is more work.  In fact we should always explicitly
>> either set or unset such controlling environmental variables as a
>> matter of good practice.
>
> When building packages with gar, the default (i.e without the  
> maintainer doing a thing) is that the RPATH will contain /opt/csw/ 
> lib/$ISALIST:/opt/csw/lib. If a maintainer is to package something  
> with a different RPATH he must set it manually, i.e more work than  
> default.

This is done because it works in almost all cases. As maintainer you
can choose to optimize this, though. Of course the default action
is open for discussion ;-)


Best regards

   -- Dago

From dam at opencsw.org  Wed Apr  8 22:24:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:24:17 +0200
Subject: [csw-maintainers] [csw-buildfarm] build8s and build9s
	inconsistancies
In-Reply-To: <49DCE271.9050304@opencsw.org>
References: <49DCCAFE.2040900@cognigencorp.com> <49DCE271.9050304@opencsw.org>
Message-ID: <5569F5BD-77D9-4172-AE7B-5A36325EFD65@opencsw.org>

Hi Mike,

Am 08.04.2009 um 19:44 schrieb Mike Watters:
> I can't even get onto build9s or build9x, I get prompted for password.
> since home dir's are mounted across the env, I have to assume I  
> don't have an
> account on those servers.

There was a transitional time when I made a script for Phil for
adding users which had a bug. Phil used it, I fixed the bug and
your and your account alone has been forgotten :-( Fixed now.
If you or anybody else encounters this: please post ASAP.


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr  8 22:29:00 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:29:00 +0200
Subject: [csw-maintainers] [csw-buildfarm] build8s and build9s
	inconsistancies
In-Reply-To: <49DCCAFE.2040900@cognigencorp.com>
References: <49DCCAFE.2040900@cognigencorp.com>
Message-ID: <A2D3DB4A-B7CC-4E10-81B6-C89BD42DA88B@opencsw.org>

Hi Darin,

Am 08.04.2009 um 18:04 schrieb Darin Perusich:
> I was building a package, amanda, on build8s and everything was going
> fine but given that Solaris 8 is no longer supported I figured I'd  
> move
> over to the Solaris 9 build machines.

Please continue to build on Solaris 8 per default until there is a
specific reason why you build for Solaris 9. Solaris 8 is no longer
mandatory, but still recommended.

> The code compiles fine but the
> staging of the package is incomplete on build9s, whether I use
> 'stagepkg' or 'gmake DESTDIR=$PWD/cswstage install' the results are  
> the
> same.
>
> The staging fails when setting permission, see below, on some of the
> binaries and I have no idea why. I compared the CSW packages and they
> are identical on both machine so I'm wondering if its something in  
> the OS.
>
> Has anyone else run across this before?
>
> gmake[5]: Entering directory `/home/darin/build/amanda-2.6.1/common- 
> src'
>  (dest=/opt/csw/sbin)
>  (chown=amanda)
> chown darin:sys
> /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/ 
> amgpgcrypt:
> Not owner
> gmake[5]: *** [installperms-data] Error 1
> gmake[5]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[4]: *** [install-data-am] Error 2
> gmake[4]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[3]: *** [install-am] Error 2
> gmake[3]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[2]: *** [install] Error 2
> gmake[2]: Leaving directory `/home/darin/build/amanda-2.6.1/common- 
> src'
> gmake[1]: *** [install-recursive] Error 1
> gmake[1]: Leaving directory `/home/darin/build/amanda-2.6.1'
> gmake: *** [install] Error 2

I cannot reproduce this:

> root at login [login]:/root > su - darin
> Sun Microsystems Inc.   SunOS 5.10      Generic January 2005
>
> You are on the OpenCSW Login Server from Baltic Online.
> To get an overview of the setup please read /etc/SETUP.
>
> <login:~> ssh build9s
> Last login: Wed Apr  8 17:37:52 2009 from 192.168.1.2
> Sun Microsystems Inc.   SunOS 5.9       Generic May 2002
> <build9s:~> chown darin:sys /home/darin/build/amanda-2.6.1/cswstage/ 
> opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/ 
> amgpgcrypt: Not owner
> <build9s:~> exit
> logout
> Connection to build9s closed.
> <login:~> ssh build8s
> Last login: Wed Apr  8 22:26:57 2009 from 192.168.1.2
> <build8s:~> chown darin:sys /home/darin/build/amanda-2.6.1/cswstage/ 
> opt/csw/sbin/amgpgcrypt
> chown: /home/darin/build/amanda-2.6.1/cswstage/opt/csw/sbin/ 
> amgpgcrypt: Not owner


Best regards

   -- Dago



From william at wbonnet.net  Wed Apr  8 22:33:45 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Apr 2009 22:33:45 +0200
Subject: [csw-maintainers] glib updates
Message-ID: <49DD0A29.2030209@wbonnet.net>

Hi

A good news :)

Recent updates libs update like Glib2 solveed the segfault problems i 
had with FF2 last versions. I am checking TB last version right now.

I will certainly release a FF2 update this week.

Cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From dam at opencsw.org  Wed Apr  8 22:38:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:38:43 +0200
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <x6vdpfkpfv.fsf@abhavana.free.fr>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
Message-ID: <3C086376-7F83-4B3E-8618-D763D5382351@opencsw.org>

Hi Peter,

Am 08.04.2009 um 19:47 schrieb Peter FELECAN:
> I cannot recollect precisely if we already had this discussion. Phil  
> and
> me, we had an exchange on this subject and I would like to have the
> opinion of the other fellow maintainers as our point of view are not
> totally convergent.
>
> The questions are:
>
> 1. Is a good practice for CSW packages to contain library archives of
>   the form lib*.a when we deliver dynamic libraries lib*.so ?
>
> 2. What's the potential usage for such libraries?
>
> 3. Which mechanism do you use for not building (e.g. ./configure
>   --disable-static) or to post-install (in the sense of after make  
> install)
>   remove them.

You can skip the build like you described. Additionally, GAR skips  
static
libraries per default during the merge phase (copying stuff from the
install-*/ to pkgroot/). If you want static libs you must set
   MERGE_EXCLUDE_STATICLIBS =      (nothing)
The use of special rules is discouraged.


Best regards

   -- Dago


From phil at bolthole.com  Wed Apr  8 22:45:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 13:45:06 -0700
Subject: [csw-maintainers] static libraries lib*.a in our packages
In-Reply-To: <3C086376-7F83-4B3E-8618-D763D5382351@opencsw.org>
References: <x6vdpfkpfv.fsf@abhavana.free.fr>
	<3C086376-7F83-4B3E-8618-D763D5382351@opencsw.org>
Message-ID: <20090408204506.GD99983@bolthole.com>

On Wed, Apr 08, 2009 at 10:38:43PM +0200, Dagobert Michelsen wrote:
> You can skip the build like you described. Additionally, GAR skips  
> static
> libraries per default during the merge phase (copying stuff from the
> install-*/ to pkgroot/). If you want static libs you must set
>   MERGE_EXCLUDE_STATICLIBS =      (nothing)
> The use of special rules is discouraged.

that's... really odd naming, again



From dam at opencsw.org  Wed Apr  8 22:54:06 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 8 Apr 2009 22:54:06 +0200
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
In-Reply-To: <49DD01B6.2020301@opencsw.org>
References: <49DD01B6.2020301@opencsw.org>
Message-ID: <AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>

Hi Roger,

Am 08.04.2009 um 21:57 schrieb Roger H?kansson:
> I've got the impression that /opt/csw/var is a obsolete place to put  
> things and that /var/opt/csw should be used instead.
> Shouldn't this be reflected when using gar?
> I.e gar.conf.mk should have localstatedir = /var/opt/csw instead of $ 
> (prefix)/var

I guess you are right. But this may break existing packages. Maybe  
this is
a candidate for mGAR v3? I would like to have some feedback from more
package maintainers on this change.

This is also partly true for sysconfdir = /opt/csw/etc vs. /etc/opt/csw


Best regards

   -- Dago

From phil at bolthole.com  Wed Apr  8 23:00:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 8 Apr 2009 14:00:56 -0700
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
In-Reply-To: <AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
References: <49DD01B6.2020301@opencsw.org>
	<AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
Message-ID: <20090408210056.GE99983@bolthole.com>

On Wed, Apr 08, 2009 at 10:54:06PM +0200, Dagobert Michelsen wrote:
> Hi Roger,
>
> Am 08.04.2009 um 21:57 schrieb Roger H?kansson:
>> I've got the impression that /opt/csw/var is a obsolete place to put  
>> things and that /var/opt/csw should be used instead.
>> Shouldn't this be reflected when using gar?
>> I.e gar.conf.mk should have localstatedir = /var/opt/csw instead of $ 
>> (prefix)/var
>
> I guess you are right. But this may break existing packages. Maybe this 
> is
> a candidate for mGAR v3? I would like to have some feedback from more
> package maintainers on this change.

then they need to be "broken", and then fixed.



> This is also partly true for sysconfdir = /opt/csw/etc vs. /etc/opt/csw

"partly" is accurate.
Sometimes, it is still appropriate for sysconfdir to be /opt/csw/etc



From bonivart at opencsw.org  Wed Apr  8 23:00:41 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Apr 2009 23:00:41 +0200
Subject: [csw-maintainers] /opt/csw/var and /var/opt/csw
In-Reply-To: <AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
References: <49DD01B6.2020301@opencsw.org>
	<AACB8D86-2FC4-4801-850F-3756DF6341D6@opencsw.org>
Message-ID: <625385e30904081400u4a9922dch5ad309b46dad2768@mail.gmail.com>

On Wed, Apr 8, 2009 at 10:54 PM, Dagobert Michelsen <dam at opencsw.org> wrote:
> I guess you are right. But this may break existing packages. Maybe this is
> a candidate for mGAR v3? I would like to have some feedback from more
> package maintainers on this change.
>
> This is also partly true for sysconfdir = /opt/csw/etc vs. /etc/opt/csw

I always set /var/opt/csw and usually /etcopt/csw.

-- 
/peter


From maybird1776 at yahoo.com  Thu Apr  9 00:07:03 2009
From: maybird1776 at yahoo.com (ken mays)
Date: Wed, 8 Apr 2009 15:07:03 -0700 (PDT)
Subject: [csw-maintainers] glib updates
Message-ID: <803218.42804.qm@web111309.mail.gq1.yahoo.com>


> William Bonnet wrote:
> Recent updates like Glib2 solved the segfault problems i had
> with FF2 last versions. I am checking TB last version right now.

Well, well, well...

See, it was not such a bad idea after all.

~ Ken



      


From hson at opencsw.org  Thu Apr  9 04:51:29 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Thu, 09 Apr 2009 04:51:29 +0200
Subject: [csw-maintainers] For the gnome team
Message-ID: <49DD62B1.9050602@opencsw.org>

I've been packaging some packages which were either listed as orphaned 
or where the maintainer is retired, which I think would fall under the 
gnome domain.
Since I personally don't use any gnome packages at all (don't even run X 
at all), I'm handing them off to the gnome team to finish.

gnumeric and libgoffice builds just fine as long as you have libgsf 1.14.
However I've not been able to get around the fact that libgsf-gnome-1.so 
get linked to the installed libgsf instead of the newly built one (see 
my previous posts regarding this).
So in order to build libgsf the current package has to be removed from 
the build server (which I've been testing on my own build servers)

I've also been trying to build libgda and libgnomedb (which are optional 
for gnumeric), but there seems to be some incompatibility from upstream 
so libgnomedb won't built atm.


From mwatters at opencsw.org  Thu Apr  9 07:05:36 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Apr 2009 00:05:36 -0500
Subject: [csw-maintainers] sudo and sudo.ldap in testing
Message-ID: <49DD8220.9060605@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	updated to gar v2 and sudo 1.7.0
	force incomparable between CSWsudo and CSWsudo-ldap

main compile options:

- --with-pam           (use pam authentication rules)
- --with-logging=both  (file and/or syslog for logging)
- --with-ignore-dot    (ignore 'dot' in path for sudo)
- --with-env-editor    (honor users EDITOR environment variable for visudo)

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkndgiAACgkQLrhmsXMSLxd/BACgthcvnihNjVCSFU1Y0JRrmxp1
Z2wAoKzB4M7WxZq9yOTNnG9KLwlM/t4E
=YWL/
-----END PGP SIGNATURE-----


From trygvis at opencsw.org  Thu Apr  9 13:29:12 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 09 Apr 2009 13:29:12 +0200
Subject: [csw-maintainers] Issues with pkg-get 4.2
Message-ID: <49DDDC08.9020605@opencsw.org>

I'm working on updating the maven2 package, but I can't seem to get 
pkg-get to download the package (I tried with just "maven2" as name too).

It does work with the pkg-get in current.

$ sudo pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u 
maven2-2.0.10,REV=2009.04.09
Getting catalog...
--2009-04-09 13:22:24-- 
http://mirror.opencsw.org/opencsw/testing/i386/5.11/catalog
Resolving mirror.opencsw.org... 213.178.77.176
Connecting to mirror.opencsw.org|213.178.77.176|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 29410 (29K) [text/plain]
Saving to: `catalog'

100%[============================================================================================================>] 
29,410      86.5K/s   in 0.3s

2009-04-09 13:22:24 (86.5 KB/s) - `catalog' saved [29410/29410]

Updating catalog file
/var/pkg-get/catalog-mirror.opencsw.org updated

--2009-04-09 13:22:24-- 
http://mirror.opencsw.org/opencsw/testing/i386/5.11/descriptions
Resolving mirror.opencsw.org... 213.178.77.176
Connecting to mirror.opencsw.org|213.178.77.176|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 7114 (6.9K) [text/plain]
Saving to: `descriptions'

100%[============================================================================================================>] 
7,114       --.-K/s   in 0.07s

2009-04-09 13:22:25 (94.3 KB/s) - `descriptions' saved [7114/7114]

Updated description file
Need TWO args to newer_rev

pkg-get version:
$ pkg-get
pkg-get,   by Philip Brown , phil at bolthole.com
  (Internal SCCS code revision @(#) pkg-get 4.2@(#))
Originally from http://www.bolthole.com/solaris/pkg-get.html
[snip]

--
Trygve


From Darin.Perusich at cognigencorp.com  Thu Apr  9 15:57:19 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 09 Apr 2009 09:57:19 -0400
Subject: [csw-maintainers] cswusergroup questions
Message-ID: <49DDFEBF.1020905@cognigencorp.com>

I've been reviewing the cswusergroup class script and have a few quick
questions. The docs recommend using the
/opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
/opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
this location be created as part of that packages?

What are peoples thoughts about being able to set password policies for
the users we're creating? This can be a touchy area but given that some
services will fail if the users password has expired it might be nice to
be able to set them.

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From harpchad at opencsw.org  Thu Apr  9 17:12:12 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 09 Apr 2009 10:12:12 -0500
Subject: [csw-maintainers] New in testing: fontconfig 2.6.0
In-Reply-To: <FD21DEAC-14E7-4479-89B2-3E4BF458738A@opencsw.org>
References: <FD21DEAC-14E7-4479-89B2-3E4BF458738A@opencsw.org>
Message-ID: <49DE104C.1060300@opencsw.org>

Broke SUNWgnome for me, appears that the new font caches aren't 
backwards compatible with the old ones?

running '/usr/bin/fc-cache -f' (SUNWfontconfig version 2.2.3) fixed it, 
but probably broke the CSW version.

Alexander Maier wrote:
> Hi,
> 
> current fontconfig version is now in testing at 
> http://mirror.opencsw.org/testing/ :
> 
> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-i386-CSW.pkg.gz
> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-sparc-CSW.pkg.gz
> 
> Please give it a try.
> 
> -Alexander
> 
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From harpchad at opencsw.org  Thu Apr  9 17:32:16 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 09 Apr 2009 10:32:16 -0500
Subject: [csw-maintainers] New in testing: fontconfig 2.6.0
In-Reply-To: <49DE104C.1060300@opencsw.org>
References: <FD21DEAC-14E7-4479-89B2-3E4BF458738A@opencsw.org>
	<49DE104C.1060300@opencsw.org>
Message-ID: <49DE1500.9060407@opencsw.org>

Looks like the old version created its own separate (-csw) cache, but 
the new one doesn't?

Chad Harp wrote:
> Broke SUNWgnome for me, appears that the new font caches aren't 
> backwards compatible with the old ones?
> 
> running '/usr/bin/fc-cache -f' (SUNWfontconfig version 2.2.3) fixed it, 
> but probably broke the CSW version.
> 
> Alexander Maier wrote:
>> Hi,
>>
>> current fontconfig version is now in testing at 
>> http://mirror.opencsw.org/testing/ :
>>
>> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-i386-CSW.pkg.gz
>> fontconfig-2.6.0,REV=2009.03.30-SunOS5.8-sparc-CSW.pkg.gz
>>
>> Please give it a try.
>>
>> -Alexander
>>
>>
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From phil at bolthole.com  Thu Apr  9 18:02:29 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 09:02:29 -0700
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <49DDFEBF.1020905@cognigencorp.com>
References: <49DDFEBF.1020905@cognigencorp.com>
Message-ID: <20090409160229.GG28226@bolthole.com>

On Thu, Apr 09, 2009 at 09:57:19AM -0400, Darin Perusich wrote:
> I've been reviewing the cswusergroup class script and have a few quick
> questions. The docs recommend using the
> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> this location be created as part of that packages?

that would be an option, sure.


> What are peoples thoughts about being able to set password policies for
> the users we're creating? This can be a touchy area but given that some
> services will fail if the users password has expired it might be nice to
> be able to set them.

Leave it to the site admins, i would suggest


From phil at bolthole.com  Thu Apr  9 18:04:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 09:04:33 -0700
Subject: [csw-maintainers] Issues with pkg-get 4.2
In-Reply-To: <49DDDC08.9020605@opencsw.org>
References: <49DDDC08.9020605@opencsw.org>
Message-ID: <20090409160433.GI28226@bolthole.com>

On Thu, Apr 09, 2009 at 01:29:12PM +0200, Trygve Laugst?l wrote:
> pkg-get version:
> $ pkg-get
> pkg-get,   by Philip Brown , phil at bolthole.com
>  (Internal SCCS code revision @(#) pkg-get 4.2@(#))
> Originally from http://www.bolthole.com/solaris/pkg-get.html

please update your pkg-get pkg


From bonivart at opencsw.org  Thu Apr  9 18:11:30 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 9 Apr 2009 18:11:30 +0200
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <20090409160229.GG28226@bolthole.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<20090409160229.GG28226@bolthole.com>
Message-ID: <625385e30904090911s10867d0ex611b15977edac954@mail.gmail.com>

On Thu, Apr 9, 2009 at 6:02 PM, Philip Brown <phil at bolthole.com> wrote:
> On Thu, Apr 09, 2009 at 09:57:19AM -0400, Darin Perusich wrote:
>> I've been reviewing the cswusergroup class script and have a few quick
>> questions. The docs recommend using the
>> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
>> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
>> this location be created as part of that packages?
>
> that would be an option, sure.

If we recommend using that location I think it should be included in CSWcommon.

-- 
/peter


From phil at bolthole.com  Thu Apr  9 18:17:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 09:17:26 -0700
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <625385e30904090911s10867d0ex611b15977edac954@mail.gmail.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<20090409160229.GG28226@bolthole.com>
	<625385e30904090911s10867d0ex611b15977edac954@mail.gmail.com>
Message-ID: <20090409161726.GK28226@bolthole.com>

On Thu, Apr 09, 2009 at 06:11:30PM +0200, Peter Bonivart wrote:
> On Thu, Apr 9, 2009 at 6:02 PM, Philip Brown <phil at bolthole.com> wrote:
> > On Thu, Apr 09, 2009 at 09:57:19AM -0400, Darin Perusich wrote:
> >> I've been reviewing the cswusergroup class script and have a few quick
> >> questions. The docs recommend using the
> >> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> >> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> >> this location be created as part of that packages?
> >
> > that would be an option, sure.
> 
> If we recommend using that location I think it should be included in CSWcommon.

But it only gets used by cswclassutils? :-)



From Darin.Perusich at cognigencorp.com  Thu Apr  9 20:01:17 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 09 Apr 2009 14:01:17 -0400
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <49DDFEBF.1020905@cognigencorp.com>
References: <49DDFEBF.1020905@cognigencorp.com>
Message-ID: <49DE37ED.6070603@cognigencorp.com>

Also, given these are system account what about setting the UID's to
below 100?

Darin Perusich wrote:
> I've been reviewing the cswusergroup class script and have a few quick
> questions. The docs recommend using the
> /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> this location be created as part of that packages?
> 
> What are peoples thoughts about being able to set password policies for
> the users we're creating? This can be a touchy area but given that some
> services will fail if the users password has expired it might be nice to
> be able to set them.
> 

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Thu Apr  9 20:07:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 11:07:25 -0700
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <49DE37ED.6070603@cognigencorp.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<49DE37ED.6070603@cognigencorp.com>
Message-ID: <20090409180725.GA16708@bolthole.com>

On Thu, Apr 09, 2009 at 02:01:17PM -0400, Darin Perusich wrote:
> Also, given these are system account what about setting the UID's to
> below 100?

nonono... leave that to the local site admins. 

and/or, tweak cswusergroup to have a config option in csw.conf or
whereever, to specify preferred range of UIDs when it has to add UIDs


> 
> Darin Perusich wrote:
> > I've been reviewing the cswusergroup class script and have a few quick
> > questions. The docs recommend using the
> > /opt/csw/etc/pkg/[pkgname]/cswusergroup yet the location
> > /opt/csw/etc/pkg doesn't exist. Maybe I should submit a bug but should
> > this location be created as part of that packages?
> > 
> > What are peoples thoughts about being able to set password policies for
> > the users we're creating? This can be a touchy area but given that some
> > services will fail if the users password has expired it might be nice to
> > be able to set them.


From mwatters at opencsw.org  Thu Apr  9 21:13:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Apr 2009 14:13:52 -0500
Subject: [csw-maintainers] updated: sudo in testing
Message-ID: <49DE48F0.9070003@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	repackaged after fixing permissions on the sudo* binaries.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkneSO8ACgkQLrhmsXMSLxc6rwCbBY2B/0Gf1YQpzMIBwL3uwNkZ
xDgAoJKlEWUWLIYckJ98eKSQzzTEk+6i
=/dN8
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Apr  9 22:46:58 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Apr 2009 22:46:58 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <49DD62B1.9050602@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
Message-ID: <C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>

Hi Roger,

Am 09.04.2009 um 04:51 schrieb Roger H?kansson:
> I've been packaging some packages which were either listed as  
> orphaned or where the maintainer is retired, which I think would  
> fall under the gnome domain.
> Since I personally don't use any gnome packages at all (don't even  
> run X at all), I'm handing them off to the gnome team to finish.
>
> gnumeric and libgoffice builds just fine as long as you have libgsf  
> 1.14.
> However I've not been able to get around the fact that libgsf- 
> gnome-1.so get linked to the installed libgsf instead of the newly  
> built one (see my previous posts regarding this).
> So in order to build libgsf the current package has to be removed  
> from the build server (which I've been testing on my own build  
> servers)

I just build it from SVN without any problems, maybe the LD_OPTIONS  
change was
more good than expected?

Anyway, libgsf is now in testing:
   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-i386-CSW.pkg.gz
   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-sparc-CSW.pkg.gz

> I've also been trying to build libgda and libgnomedb (which are  
> optional for gnumeric), but there seems to be some incompatibility  
> from upstream so libgnomedb won't built atm.

Nope, no incompatibility here. Builds absolutely smooth. Packages in  
testing/ now.
Thanks for the good work!

BTW: To get GNOME running it would be nice if we had someone who  
coordinates what
packages need to be built most urgent. Volunteers?


Best regards

   -- Dago

From phil at bolthole.com  Thu Apr  9 22:58:04 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 13:58:04 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
Message-ID: <20090409205804.GC16708@bolthole.com>

On Thu, Apr 09, 2009 at 10:46:58PM +0200, Dagobert Michelsen wrote:
>
> BTW: To get GNOME running it would be nice if we had someone who  
> coordinates what
> packages need to be built most urgent. Volunteers?

Pffft... we dont need a "coordinator". we need someone to
BUILD AND TEST packages :-)

Any volunteers for that?

[Note: Please volunteer by stating,
  "I'm going to have package X dropped in newpkgs by tomorrow", not
"well, yes I'd like to volunteer and I'll make a whole lot of
     packages ... when i have more time...."  ]




From hson at opencsw.org  Fri Apr 10 02:19:32 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Fri, 10 Apr 2009 02:19:32 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
Message-ID: <49DE9094.4070504@opencsw.org>

Dagobert Michelsen wrote:
> I just build it from SVN without any problems, maybe the LD_OPTIONS 
> change was
> more good than expected?
> 
> Anyway, libgsf is now in testing:
>   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-i386-CSW.pkg.gz
>   libgsf-1.14.11,REV=2009.04.09-SunOS5.8-sparc-CSW.pkg.gz

Maybe I was a bit unclear, the package will build just fine, however 
libgsf-gnome-1.so isn't linked to the libgsf-1.so that the build produce, 
but the old one in /opt/csw/lib.
And I've been trying to fix this in various ways but with no luck.
So far the only working solution I've found is to remove libgsf from the 
build machine when packaging.

hson at build8x :~ > dump -Lv 
/home/dam/mgar/pkg/libgsf/trunk/work/pkgroot/opt/csw/lib/libgsf-gnome-1.so.114.0.11
...
[1]     NEEDED          libgsf-1.so.1
...
[16]    SONAME          libgsf-gnome-1.so.114


>> I've also been trying to build libgda and libgnomedb (which are 
>> optional for gnumeric), but there seems to be some incompatibility 
>> from upstream so libgnomedb won't built atm.
> 
> Nope, no incompatibility here. Builds absolutely smooth. Packages in 
> testing/ now.
> Thanks for the good work!

Hmm, I get a bunch of "undefined struct/union member" when building 
libgnomedb on my build machine, but I've installed the libgda I packaged, I 
guess the buildfarm have the older one?
Otherwise its probably some other package that is missing from my build 
machine which is installed on the build farm...


From hson at opencsw.org  Fri Apr 10 02:34:10 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Fri, 10 Apr 2009 02:34:10 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <20090409205804.GC16708@bolthole.com>
References: <49DD62B1.9050602@opencsw.org>	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
Message-ID: <49DE9402.3000301@opencsw.org>

Philip Brown wrote:
> On Thu, Apr 09, 2009 at 10:46:58PM +0200, Dagobert Michelsen wrote:
>> BTW: To get GNOME running it would be nice if we had someone who  
>> coordinates what
>> packages need to be built most urgent. Volunteers?
> 
> Pffft... we dont need a "coordinator". we need someone to
> BUILD AND TEST packages :-)

I think think there is a need for some coordination too.
There is a lot of dependencies for the gnome packages and in order to get 
them in sync with upstream there has to be some kind of synchronized work.


> Any volunteers for that?
> 
> [Note: Please volunteer by stating,
>   "I'm going to have package X dropped in newpkgs by tomorrow", not
> "well, yes I'd like to volunteer and I'll make a whole lot of
>      packages ... when i have more time...."  ]

Well, as I stated in private mails to Phil, I use very few CSW packages 
other than those either already properly maintained (i.e fairly current with 
upstream and with active maintainer), or those I've become maintainer for (I 
don't even use all of those).

I don't even run X as a desktop (Windows on desktop, Linux and Solaris on 
servers) so I can't even test the packages.
But I'm willing to help pushing existing packages into gar to the point a 
package is correctly built.


This far, I've been picking random packages that were either listed as 
orphaned or with a "retired" maintainer.
Right now I'm focusing on packages with many bugs reported (netsnmp have 
been my main focus for instance, now I'm looking at the bacula and 
postgresql packages)


From phil at bolthole.com  Fri Apr 10 03:02:30 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 18:02:30 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <49DE9402.3000301@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
	<49DE9402.3000301@opencsw.org>
Message-ID: <20090410010230.GB31623@bolthole.com>

On Fri, Apr 10, 2009 at 02:34:10AM +0200, Roger H?kansson wrote:
> I think think there is a need for some coordination too.
> There is a lot of dependencies for the gnome packages and in order to get 
> them in sync with upstream there has to be some kind of synchronized 
> work.

To have a need for synchronization, you first need "multiple threads".
A single threaded operation, never needs "synchronization".
(nor does 'zero threads' :-)

Do we have two or more people currently working on compiling GNOME
packages? 
please speak up, because I'm not aware of many people currently working in
parallel on this that need synchronization.



Right now, we just have occasional, sporadic updates of low level GNOME
packages, like glib.
Which is GREAT! But... we have no one i know of, who is committing to build
the whole chain.
Or even a specific large PIECE of the chain.
We're just picking one one or two packages here and there.

This effort can be self-organizing; the main "difficult" step, is people
who are going to do the work.
When someone decides to do the work, then just announce, "I'm building
these packages now".
There; problem of "synchronization" solved.


But what the heck, lets make it easier to remember who said they're doing
what.

http://wiki.opencsw.org/gnome
now exists, so that people planning to work on multiple GNOME packages, can
let others know which ones are being worked on already.

The gnome effort consists of basically two types of components:

1. core libs (10 to 20?)
 
2. a plethora of "auxiliary stuff".


For anyone who needs/wants to work on gnome, the work flow is relatively
straightforward.

1. Pick a gnome lib or two that you wish to work on, that is out of date
 (dont know what to pick? take a look through
       http://wiki.opencsw.org/maintainers/kenmays 
   for packages with "lib" in the name, for ideas)t



2. Take a look at its dependancies, and make sure they are up to date.
   If not, then work on its dependancies.
   Otherwise, just update the lib that you chose.

 If desired, return to step 1, and pick another package.




From mwatters at opencsw.org  Fri Apr 10 04:20:39 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Apr 2009 21:20:39 -0500
Subject: [csw-maintainers] gcc 4.3.3
Message-ID: <49DEACF7.5020803@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I am at a loss,  I have combined the sparc builds in
just about every conceivable way and can not get one package to work on
both solaris 8 and solaris 10. (did not try solaris 9)

Unless someone can provide a solution, (my recipe is up to date in svn)
I will package up OS specific packages on Monday.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknerPcACgkQLrhmsXMSLxcRbwCfcJ/eUlU1aCLsljRb8CoahhvJ
H1gAoKW4N03e4s1TAzJeWjprCOGV3kir
=e+TM
-----END PGP SIGNATURE-----


From phil at bolthole.com  Fri Apr 10 07:11:49 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Apr 2009 22:11:49 -0700
Subject: [csw-maintainers] gcc 4.3.3
In-Reply-To: <49DEACF7.5020803@opencsw.org>
References: <49DEACF7.5020803@opencsw.org>
Message-ID: <20090410051149.GA8309@bolthole.com>

On Thu, Apr 09, 2009 at 09:20:39PM -0500, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I am at a loss,  I have combined the sparc builds in
> just about every conceivable way and can not get one package to work on
> both solaris 8 and solaris 10. (did not try solaris 9)

btw; did you ask on the gcc mailing list?


From maybird1776 at yahoo.com  Fri Apr 10 13:00:21 2009
From: maybird1776 at yahoo.com (ken mays)
Date: Fri, 10 Apr 2009 04:00:21 -0700 (PDT)
Subject: [csw-maintainers] For the gnome team
Message-ID: <144698.96991.qm@web111305.mail.gq1.yahoo.com>


I can assist in the GNOME packaging development - just not right now due to current demands.

~K


--- On Thu, 4/9/09, Philip Brown <phil at bolthole.com> wrote:

> From: Philip Brown <phil at bolthole.com>
> Subject: Re: [csw-maintainers] For the gnome team
> To: maintainers at lists.opencsw.org
> Date: Thursday, April 9, 2009, 9:02 PM
> On Fri, Apr 10, 2009 at 02:34:10AM
> +0200, Roger H?kansson wrote:
> > I think think there is a need for some coordination
> too.
> > There is a lot of dependencies for the gnome packages
> and in order to get 
> > them in sync with upstream there has to be some kind
> of synchronized 
> > work.
> 
> To have a need for synchronization, you first need
> "multiple threads".
> A single threaded operation, never needs
> "synchronization".
> (nor does 'zero threads' :-)
> 
> Do we have two or more people currently working on
> compiling GNOME
> packages? 
> please speak up, because I'm not aware of many people
> currently working in
> parallel on this that need synchronization.
> 
> 
> 
> Right now, we just have occasional, sporadic updates of low
> level GNOME
> packages, like glib.
> Which is GREAT! But... we have no one i know of, who is
> committing to build
> the whole chain.
> Or even a specific large PIECE of the chain.
> We're just picking one one or two packages here and there.
> 
> This effort can be self-organizing; the main "difficult"
> step, is people
> who are going to do the work.
> When someone decides to do the work, then just announce,
> "I'm building
> these packages now".
> There; problem of "synchronization" solved.
> 
> 
> But what the heck, lets make it easier to remember who said
> they're doing
> what.
> 
> http://wiki.opencsw.org/gnome
> now exists, so that people planning to work on multiple
> GNOME packages, can
> let others know which ones are being worked on already.
> 
> The gnome effort consists of basically two types of
> components:
> 
> 1. core libs (10 to 20?)
>  
> 2. a plethora of "auxiliary stuff".
> 
> 
> For anyone who needs/wants to work on gnome, the work flow
> is relatively
> straightforward.
> 
> 1. Pick a gnome lib or two that you wish to work on, that
> is out of date
>  (dont know what to pick? take a look through
> ? ? ???http://wiki.opencsw.org/maintainers/kenmays 
> ???for packages with "lib" in the name, for
> ideas)t
> 
> 
> 
> 2. Take a look at its dependancies, and make sure they are
> up to date.
> ???If not, then work on its dependancies.
> ???Otherwise, just update the lib that you
> chose.
> 
>  If desired, return to step 1, and pick another package.
> 
> 
> 
> -----Inline Attachment Follows-----
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers


      



From mwatters at opencsw.org  Fri Apr 10 22:00:19 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 10 Apr 2009 15:00:19 -0500
Subject: [csw-maintainers] gcc 4.3.3
In-Reply-To: <20090410051149.GA8309@bolthole.com>
References: <49DEACF7.5020803@opencsw.org> <20090410051149.GA8309@bolthole.com>
Message-ID: <49DFA553.8080101@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Apr 09, 2009 at 09:20:39PM -0500, Mike Watters wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> I am at a loss,  I have combined the sparc builds in
>> just about every conceivable way and can not get one package to work on
>> both solaris 8 and solaris 10. (did not try solaris 9)
> 
> btw; did you ask on the gcc mailing list?
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I sent the question to gcc-help mailing list this morning.
I am waiting for an answer.  I will still packages up OS specific packages on
Monday if I don't get a solution.  If the gcc or this list comes back with a
solution, I will re-package.
There are too many packages waiting on amd64 gcc to hold this any longer.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknfpVMACgkQLrhmsXMSLxdV6ACfQRJGVGuwu6kCMffhbCk3eYU+
G8sAoOkB4ur9yxCXCvtqGfjrjhflRir8
=z8VL
-----END PGP SIGNATURE-----


From hson at opencsw.org  Sat Apr 11 02:06:15 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Sat, 11 Apr 2009 02:06:15 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <20090410010230.GB31623@bolthole.com>
References: <49DD62B1.9050602@opencsw.org>	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>	<20090409205804.GC16708@bolthole.com>	<49DE9402.3000301@opencsw.org>
	<20090410010230.GB31623@bolthole.com>
Message-ID: <49DFDEF7.7090309@opencsw.org>

Philip Brown wrote:
> Do we have two or more people currently working on compiling GNOME
> packages? 

I seem to remember some posts regarding someone joining the "gnome team"...


From rupert at opencsw.org  Sat Apr 11 12:29:49 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 11 Apr 2009 12:29:49 +0200
Subject: [csw-maintainers] python-reportlab 2.3
Message-ID: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com>

hi,

for getting http://trac-hacks.org/wiki/TracWikiPrintPlugin we would
need http://reportlab.org/ v2.3 (http://www.reportlab.org/ftp/) . i
saw there is a package http://opencsw.org/packages/reportlab, v2.1.

as it is not in
http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg, where would
be the source to try it?

rupert.


From pfelecan at opencsw.org  Sat Apr 11 14:41:47 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Sat, 11 Apr 2009 14:41:47 +0200
Subject: [csw-maintainers] gcc 4.3.3
In-Reply-To: <49DFA553.8080101@opencsw.org> (Mike Watters's message of "Fri\,
	10 Apr 2009 15\:00\:19 -0500")
References: <49DEACF7.5020803@opencsw.org>
	<20090410051149.GA8309@bolthole.com> <49DFA553.8080101@opencsw.org>
Message-ID: <x63acfjras.fsf@abhavana.free.fr>

Mike Watters <mwatters at opencsw.org> writes:

> Philip Brown wrote:
>> On Thu, Apr 09, 2009 at 09:20:39PM -0500, Mike Watters wrote:
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>>
>>> I am at a loss,  I have combined the sparc builds in
>>> just about every conceivable way and can not get one package to work on
>>> both solaris 8 and solaris 10. (did not try solaris 9)
>> 
>> btw; did you ask on the gcc mailing list?
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
>
> I sent the question to gcc-help mailing list this morning.
> I am waiting for an answer.  I will still packages up OS specific packages on
> Monday if I don't get a solution.  If the gcc or this list comes back with a
> solution, I will re-package.
> There are too many packages waiting on amd64 gcc to hold this any longer.

Don't wanting to revive old discussions but I still don't understand all
this fuss about 64bit being so critical. Still, I can appreciate your
efforts and hope that you'll find a solution.
-- 
Peter


From dam at opencsw.org  Sun Apr 12 00:34:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Apr 2009 00:34:42 +0200
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <20090409205804.GC16708@bolthole.com>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
Message-ID: <038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>

Hi Phil,

Am 09.04.2009 um 22:58 schrieb Philip Brown:
> On Thu, Apr 09, 2009 at 10:46:58PM +0200, Dagobert Michelsen wrote:
>>
>> BTW: To get GNOME running it would be nice if we had someone who
>> coordinates what
>> packages need to be built most urgent. Volunteers?
>
> Pffft... we dont need a "coordinator".

I disagree here. It would help if there was someone who would say:
"We have a roadmap, these 3 libs need updating first, than we
test that application and than proceed to these packages". At
least I wouldn't know where in this vast dependency building
the bottom building blocks are.

> we need someone to BUILD AND TEST packages :-) Any volunteers for  
> that?

These will come if the necessary packages have been identified.


Best regards

   -- Dago


From dam at opencsw.org  Sun Apr 12 00:36:30 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Apr 2009 00:36:30 +0200
Subject: [csw-maintainers] cswusergroup questions
In-Reply-To: <20090409180725.GA16708@bolthole.com>
References: <49DDFEBF.1020905@cognigencorp.com>
	<49DE37ED.6070603@cognigencorp.com>
	<20090409180725.GA16708@bolthole.com>
Message-ID: <49D25B28-7EC6-4034-A433-EA08587265D7@opencsw.org>

Hi Phil,

Am 09.04.2009 um 20:07 schrieb Philip Brown:
> and/or, tweak cswusergroup to have a config option in csw.conf or
> whereever, to specify preferred range of UIDs when it has to add UIDs

That is a very good idea! Please take this as a filed features  
request ;-)


Best regards

   -- Dago


From ihsan at opencsw.org  Sun Apr 12 14:46:26 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 12 Apr 2009 14:46:26 +0200
Subject: [csw-maintainers] problems with new libtool
Message-ID: <49E1E2A2.2010704@opencsw.org>

Hello,

Since libtool was updated on the buildfarm, I'm having troubles to build
 Apache 2.2.

checking whether it is safe to define __EXTENSIONS__... yes
checking for library containing strerror... none required
checking whether system uses EBCDIC... no
performing libtool configuration...
/home/ihsan/gar/csw/mgar/pkg/apache2/trunk/work/build-isa-sparcv8/httpd-2.2.11/srclib/apr/configure:
line 9748: syntax error near unexpected token
`lt_if_append_uniq(lt_decl_varnames,'
/home/ihsan/gar/csw/mgar/pkg/apache2/trunk/work/build-isa-sparcv8/httpd-2.2.11/srclib/apr/configure:
line 9748: `lt_if_append_uniq(lt_decl_varnames, SED, , ,'
configure failed for srclib/apr
gmake[1]: *** [configure-work/build-isa-sparcv8/httpd-2.2.11/configure]
Error 1
gmake[1]: Leaving directory `/home/ihsan/gar/csw/mgar/pkg/apache2/trunk'
gmake: *** [merge-isa-sparcv8] Error 2

With the old libtool, it was running just fine. I'm not sure if this is
related to Apache, so I'm wondering if others are also having troubles.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Mon Apr 13 16:43:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 13 Apr 2009 07:43:31 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
	<038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
Message-ID: <20090413144331.GA51320@bolthole.com>

On Sun, Apr 12, 2009 at 12:34:42AM +0200, Dagobert Michelsen wrote:
> I disagree here. It would help if there was someone who would say:
> "We have a roadmap, these 3 libs need updating first, than we
> test that application and than proceed to these packages". At
> least I wouldn't know where in this vast dependency building
> the bottom building blocks are.

i agree that it would be helpful to have a dependancy tree/map,
particularly noting which are out of date, and which are not.

That is not the same thing as having "a coordinator".

that title implies something with more responsibility.



From phil at bolthole.com  Mon Apr 13 16:54:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 13 Apr 2009 07:54:06 -0700
Subject: [csw-maintainers] For the gnome team
In-Reply-To: <038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
References: <49DD62B1.9050602@opencsw.org>
	<C360A4C4-F78A-4C0F-AB91-A02EA6317628@opencsw.org>
	<20090409205804.GC16708@bolthole.com>
	<038CE865-6179-4708-8C0E-EEA413D653B3@opencsw.org>
Message-ID: <20090413145406.GB51320@bolthole.com>

On Sun, Apr 12, 2009 at 12:34:42AM +0200, Dagobert Michelsen wrote:
>> Phil wrote:
>> we need someone to BUILD AND TEST packages :-) Any volunteers for  
>> that?
>
> These will come if the necessary packages have been identified.
>

ok, lets test that theory. 
A good intermediate dependancy target is libgnome.


lower level libraries that we need updated, to update libgnome
packages, in approximate "order":

fontconfig (its actually in testing, but needs to be released!!)
libcairo
libxrender
libpango
libatk
gconf2
orbit2
libbonobo2, libbonoboui
gnomevfs2
libgnome




From dam at opencsw.org  Tue Apr 14 13:18:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 13:18:15 +0200
Subject: [csw-maintainers] libcairo
Message-ID: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>

Hi,

I jumped in on the work on libcairo. William already prepared most of
what is necessary and I push in the latest GAR tweaks and provide 64  
bit.
Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>


Best regards

   -- Dago


From dam at opencsw.org  Tue Apr 14 13:43:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 13:43:40 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
Message-ID: <99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>

Hi,

Am 14.04.2009 um 13:18 schrieb Dagobert Michelsen:
> I jumped in on the work on libcairo. William already prepared most of
> what is necessary and I push in the latest GAR tweaks and provide 64  
> bit.
> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>

New in testing/:

libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz


Best regards

   -- Dago


From Darin.Perusich at cognigencorp.com  Tue Apr 14 22:31:44 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Tue, 14 Apr 2009 16:31:44 -0400
Subject: [csw-maintainers] /testing amanda 2.6.1p1
Message-ID: <49E4F2B0.9000107@cognigencorp.com>

Hello,

I'd like to announce that latest release of Amanda, the world's most
popular Open Source Backup and Archiving software, is available for
testing via the testing repository. For a listing of features see the
Amanda Wiki at http://wiki.zmanda.com/index.php/2.6.1_features. Please
use Mantis to report any issues with this package.

To install Amanda directly from the testing repository run the following:

pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u amanda

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From dam at opencsw.org  Tue Apr 14 22:32:12 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 22:32:12 +0200
Subject: [csw-maintainers] libcurl and 64 bit on x86
Message-ID: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>

Hi,

I am having problems compiling vorbistools in 64 bit on x86:

> Making all in ogg123
> gmake[4]: Entering directory `/home/dam/mgar/pkg/vorbistools/trunk/ 
> work/build-isa-amd64/vorbis-tools-1.2.0/ogg123'
> source='http_transport.c' object='http_transport.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DSYSCONFDIR=\"/opt/csw/etc 
> \" -DLOCALEDIR=\"/opt/csw/share/locale\" -DHAVE_CONFIG_H -I. -I.. -I/ 
> opt/csw/include   -I/opt/csw/include   -I/opt/csw/include -I/opt/csw/ 
> include -D_REENTRANT  -I../include -I../intl  -I/opt/csw/include - 
> D_REENTRANT  -O -xO3 -xarch=amd64 -I/opt/csw/include -c  
> http_transport.c
> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative  
> subscript
> cc: acomp failed for http_transport.c
> gmake[4]: *** [http_transport.o] Error 2

This is

> /*
>  * Verify that the size previously defined and expected for long
>  * is the same as the one reported by sizeof() at compile time.
>  */
>
> typedef char
>   __curl_rule_01__
>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];

CURL_SIZEOF_LONG is 4:

> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4


I guess another set of includes is needed for 64 bit, right?


Best regards

   -- Dago





From harpchad at opencsw.org  Tue Apr 14 22:41:43 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 14 Apr 2009 15:41:43 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
Message-ID: <49E4F507.1040503@opencsw.org>

Possibly, but the current version of curl doesn't include amd64 support 
since openldap doesn't have amd64 libraries (Mantis 3028). So I suspect 
you'd run into linking problems even if you were able to compile.

Did you have any issue with sparcv9?

Dagobert Michelsen wrote:
> Hi,
> 
> I am having problems compiling vorbistools in 64 bit on x86:
> 
>> Making all in ogg123
>> gmake[4]: Entering directory 
>> `/home/dam/mgar/pkg/vorbistools/trunk/work/build-isa-amd64/vorbis-tools-1.2.0/ogg123' 
>>
>> source='http_transport.c' object='http_transport.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc 
>> -DSYSCONFDIR=\"/opt/csw/etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" 
>> -DHAVE_CONFIG_H -I. -I.. -I/opt/csw/include   -I/opt/csw/include   
>> -I/opt/csw/include -I/opt/csw/include -D_REENTRANT  -I../include 
>> -I../intl  -I/opt/csw/include -D_REENTRANT  -O -xO3 -xarch=amd64 
>> -I/opt/csw/include -c http_transport.c
>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative subscript
>> cc: acomp failed for http_transport.c
>> gmake[4]: *** [http_transport.o] Error 2
> 
> This is
> 
>> /*
>>  * Verify that the size previously defined and expected for long
>>  * is the same as the one reported by sizeof() at compile time.
>>  */
>>
>> typedef char
>>   __curl_rule_01__
>>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];
> 
> CURL_SIZEOF_LONG is 4:
> 
>> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4
> 
> 
> I guess another set of includes is needed for 64 bit, right?
> 
> 
> Best regards
> 
>   -- Dago
> 
> 
> 



From dam at opencsw.org  Tue Apr 14 22:59:09 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 14 Apr 2009 22:59:09 +0200
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E4F507.1040503@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
Message-ID: <239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>

Hi Chad,

Am 14.04.2009 um 22:41 schrieb Chad Harp:
> Possibly, but the current version of curl doesn't include amd64  
> support since openldap doesn't have amd64 libraries (Mantis 3028).

Damn. Want to take a look at OpenLDAP?

> So I suspect you'd run into linking problems even if you were able  
> to compile.
>
> Did you have any issue with sparcv9?

Yes, same issue:

> gmake[4]: Entering directory `/home/dam/mgar/pkg/vorbistools/trunk/ 
> work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123'
> source='http_transport.c' object='http_transport.o' libtool=no \
>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>         /opt/studio/SOS11/SUNWspro/bin/cc -DSYSCONFDIR=\"/opt/csw/etc 
> \" -DLOCALEDIR=\"/opt/csw/share/locale\" -DHAVE_CONFIG_H -I. -I.. -I/ 
> opt/csw/include   -I/opt/csw/include   -I/opt/csw/include -I/opt/csw/ 
> include -D_REENTRANT  -I../include -I../intl  -I/opt/csw/include - 
> D_REENTRANT  -xO4 -fast -w -fsimple -native -xcg92 -xO3 -xarch=v9 -I/ 
> opt/csw/include -c http_transport.c
> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative  
> subscript
> cc: acomp failed for http_transport.c


Best regards

   -- Dago


From harpchad at opencsw.org  Tue Apr 14 23:57:34 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 14 Apr 2009 16:57:34 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
Message-ID: <49E506CE.50900@opencsw.org>

Ok, I wrote a quick program to test it and am able to replicate this. 
I'll work on a new build to fix the sparcv9.

I'll take a look at openldap and see what's involved.  All I really need 
is the library, I don't want to take on the server and other stuff as I 
don't use it (and wouldn't be able to test it).

Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 14.04.2009 um 22:41 schrieb Chad Harp:
>> Possibly, but the current version of curl doesn't include amd64 
>> support since openldap doesn't have amd64 libraries (Mantis 3028).
> 
> Damn. Want to take a look at OpenLDAP?
> 
>> So I suspect you'd run into linking problems even if you were able to 
>> compile.
>>
>> Did you have any issue with sparcv9?
> 
> Yes, same issue:
> 
>> gmake[4]: Entering directory 
>> `/home/dam/mgar/pkg/vorbistools/trunk/work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123' 
>>
>> source='http_transport.c' object='http_transport.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc 
>> -DSYSCONFDIR=\"/opt/csw/etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" 
>> -DHAVE_CONFIG_H -I. -I.. -I/opt/csw/include   -I/opt/csw/include   
>> -I/opt/csw/include -I/opt/csw/include -D_REENTRANT  -I../include 
>> -I../intl  -I/opt/csw/include -D_REENTRANT  -xO4 -fast -w -fsimple 
>> -native -xcg92 -xO3 -xarch=v9 -I/opt/csw/include -c http_transport.c
>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative subscript
>> cc: acomp failed for http_transport.c
> 
> 
> Best regards
> 
>   -- Dago



From harpchad at opencsw.org  Wed Apr 15 02:06:11 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Tue, 14 Apr 2009 19:06:11 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E506CE.50900@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E506CE.50900@opencsw.org>
Message-ID: <72B3CB61-A6E3-4E13-A83D-62FA3BF40103@opencsw.org>

Looks like I'll need to wrap the curlbuild.h

Fedora does this:

#include <bits/wordsize.h>

#if __WORDSIZE == 32
#include "curlbuild-32.h"
#elif __WORDSIZE == 64
#include "curlbuild-64.h"
#else
#error "Unknown word size"
#endif


We don't (I think) have a __WORDSIZE macro anywhere, so I'm thinking  
of this:

#if defined __arch64__ || defined __sparcv9
#include "curlbuild-64.h"
#else
#include "curlbuild-32.h"
#endif

I'm not positive about my macro choices, anybody want to shoot some  
holes in that before I commit (will it still include the 32 bit  
headers when building 32bit on a 64bit platform)?

On Apr 14, 2009, at 4:57 PM, Chad Harp wrote:

> Ok, I wrote a quick program to test it and am able to replicate  
> this. I'll work on a new build to fix the sparcv9.
>
> I'll take a look at openldap and see what's involved.  All I really  
> need is the library, I don't want to take on the server and other  
> stuff as I don't use it (and wouldn't be able to test it).
>
> Dagobert Michelsen wrote:
>> Hi Chad,
>> Am 14.04.2009 um 22:41 schrieb Chad Harp:
>>> Possibly, but the current version of curl doesn't include amd64  
>>> support since openldap doesn't have amd64 libraries (Mantis 3028).
>> Damn. Want to take a look at OpenLDAP?
>>> So I suspect you'd run into linking problems even if you were  
>>> able to compile.
>>>
>>> Did you have any issue with sparcv9?
>> Yes, same issue:
>>> gmake[4]: Entering directory `/home/dam/mgar/pkg/vorbistools/ 
>>> trunk/work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123'
>>> source='http_transport.c' object='http_transport.o' libtool=no \
>>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>>         /opt/studio/SOS11/SUNWspro/bin/cc -DSYSCONFDIR=\"/opt/csw/ 
>>> etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" -DHAVE_CONFIG_H -I. - 
>>> I.. -I/opt/csw/include   -I/opt/csw/include   -I/opt/csw/include - 
>>> I/opt/csw/include -D_REENTRANT  -I../include -I../intl  -I/opt/ 
>>> csw/include -D_REENTRANT  -xO4 -fast -w -fsimple -native -xcg92 - 
>>> xO3 -xarch=v9 -I/opt/csw/include -c http_transport.c
>>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative  
>>> subscript
>>> cc: acomp failed for http_transport.c
>> Best regards
>>   -- Dago
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From james at opencsw.org  Wed Apr 15 11:02:29 2009
From: james at opencsw.org (James Lee)
Date: Wed, 15 Apr 2009 09:02:29 GMT
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
Message-ID: <20090415.9022900.4244226440@gyor.oxdrove.co.uk>

On 14/04/09, 21:32:12, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
[csw-maintainers] libcurl and 64 bit on x86:

> >  * Verify that the size previously defined and expected for long
> >  * is the same as the one reported by sizeof() at compile time.
> >  */
> >
> > typedef char
> >   __curl_rule_01__
> >     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];

> CURL_SIZEOF_LONG is 4:

> > /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4


#define CURL_SIZEOF_LONG sizeof(long)




James.


From harpchad at opencsw.org  Wed Apr 15 15:57:51 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 08:57:51 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415.9022900.4244226440@gyor.oxdrove.co.uk>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<20090415.9022900.4244226440@gyor.oxdrove.co.uk>
Message-ID: <49E5E7DF.1030003@opencsw.org>

James Lee wrote:
> On 14/04/09, 21:32:12, Dagobert Michelsen <dam at opencsw.org> wrote regarding 
> [csw-maintainers] libcurl and 64 bit on x86:
> 
>>>  * Verify that the size previously defined and expected for long
>>>  * is the same as the one reported by sizeof() at compile time.
>>>  */
>>>
>>> typedef char
>>>   __curl_rule_01__
>>>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];
> 
>> CURL_SIZEOF_LONG is 4:
> 
>>> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4
> 
> 
> #define CURL_SIZEOF_LONG sizeof(long)
> 
> 
> 
> 
> James.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

curl does a check to make sure that CURL_SIZEOF_LONG == sizeof(long), so 
   by changing CURL_SIZEOF_LONG to sizeof(long) I'd just be saying
sizeof(long) == sizeof(long). :)

There's a discussion going on over on the curl mailing list as to how to 
make this more friendly towards packaging for multiple architectures, 
but the consensus for now seems to be to wrap the headers.


From phil at bolthole.com  Wed Apr 15 16:03:36 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 07:03:36 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5E7DF.1030003@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<20090415.9022900.4244226440@gyor.oxdrove.co.uk>
	<49E5E7DF.1030003@opencsw.org>
Message-ID: <20090415140336.GA55760@bolthole.com>

On Wed, Apr 15, 2009 at 08:57:51AM -0500, Chad Harp wrote:
>
> curl does a check to make sure that CURL_SIZEOF_LONG == sizeof(long), so  
>   by changing CURL_SIZEOF_LONG to sizeof(long) I'd just be saying
> sizeof(long) == sizeof(long). :)

sounds appropriate, really.

possibly the reason the define for CURL_SIZEOF_LONG exists, is because in
some situations and with some compilers, it must have a fixed number in the
define, rather than an operation-style macro.

but if our compiler is happy with it, then just use it.
although we might need  extra parens, to be safe:

#define CURL_SIZEOF_LONG (sizeof(long))


From james at opencsw.org  Wed Apr 15 16:23:39 2009
From: james at opencsw.org (James Lee)
Date: Wed, 15 Apr 2009 14:23:39 GMT
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5E7DF.1030003@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<20090415.9022900.4244226440@gyor.oxdrove.co.uk>
	<49E5E7DF.1030003@opencsw.org>
Message-ID: <20090415.14233900.2233379359@gyor.oxdrove.co.uk>

On 15/04/09, 14:57:51, Chad Harp <harpchad at opencsw.org> wrote regarding Re:
[csw-maintainers] libcurl and 64 bit on x86:

> >>>  * Verify that the size previously defined and expected for long
> >>>  * is the same as the one reported by sizeof() at compile time.
> >>>  */
> >>>
> >>> typedef char
> >>>   __curl_rule_01__
> >>>     [CurlchkszEQ(long, CURL_SIZEOF_LONG)];
> >
> >> CURL_SIZEOF_LONG is 4:
> >
> >>> /opt/csw/include/curl/curlbuild.h:#define CURL_SIZEOF_LONG 4
> >
> >
> > #define CURL_SIZEOF_LONG sizeof(long)


> curl does a check to make sure that CURL_SIZEOF_LONG == sizeof(long) so
>    by changing CURL_SIZEOF_LONG to sizeof(long) I'd just be saying
> sizeof(long) == sizeof(long). :)

It will be.  Another chuck of code we can cut.


> There's a discussion going on over on the curl mailing list as to how to
> make this more friendly towards packaging for multiple architectures,
> but the consensus for now seems to be to wrap the headers.

Do they get paid for code by the yard?  I don't understand what problem
they are trying to solve, at least I thought I did, but this was all
solved decades ago by the sizeof operator.




James.


From harpchad at opencsw.org  Wed Apr 15 16:43:12 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 09:43:12 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
Message-ID: <49E5F280.7070503@opencsw.org>

I've put a new curldevel package in testing (sparc only, since x86 
doesn't have 64-bit support right now).

curldevel-7.19.4,REV=2009.04.15-SunOS5.8-sparc-CSW.pkg.gz

It's wraps curlbuild.h using the following:

/* Allow 32 and 64 bit headers to coexist */
#if defined __arch64__ || defined __sparcv9
#include "curlbuild-64.h"
#else
#include "curlbuild-32.h"
#endif

I opted not to make the changes to curlbuild.h because there were 
several lines (other than the sizeof(long) we discussed) that would have 
to change (see diff below).  I think this will be more easily adapted to 
future versions.



harpchad at build8s (CSW)$ diff curlbuild-32.h curlbuild-64.h
108c108
< #define CURL_PULL_SYS_TYPES_H 1
---
 > /* #undef CURL_PULL_SYS_TYPES_H */
122c122
< #define CURL_PULL_INTTYPES_H 1
---
 > /* #undef CURL_PULL_INTTYPES_H */
128c128
< #define CURL_SIZEOF_LONG 4
---
 > #define CURL_SIZEOF_LONG 8
131c131
< #define CURL_TYPEOF_CURL_OFF_T int64_t
---
 > #define CURL_TYPEOF_CURL_OFF_T long
137c137
< #define CURL_FORMAT_CURL_OFF_T "lld"
---
 > #define CURL_FORMAT_CURL_OFF_T "ld"
140c140
< #define CURL_FORMAT_CURL_OFF_TU "llu"
---
 > #define CURL_FORMAT_CURL_OFF_TU "lu"
143c143
< #define CURL_FORMAT_OFF_T "%lld"
---
 > #define CURL_FORMAT_OFF_T "%ld"
149c149
< #define CURL_SUFFIX_CURL_OFF_T LL
---
 > #define CURL_SUFFIX_CURL_OFF_T L
152c152
< #define CURL_SUFFIX_CURL_OFF_TU ULL
---
 > #define CURL_SUFFIX_CURL_OFF_TU UL


Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 14.04.2009 um 22:41 schrieb Chad Harp:
>> Possibly, but the current version of curl doesn't include amd64 
>> support since openldap doesn't have amd64 libraries (Mantis 3028).
> 
> Damn. Want to take a look at OpenLDAP?
> 
>> So I suspect you'd run into linking problems even if you were able to 
>> compile.
>>
>> Did you have any issue with sparcv9?
> 
> Yes, same issue:
> 
>> gmake[4]: Entering directory 
>> `/home/dam/mgar/pkg/vorbistools/trunk/work/build-isa-sparcv9/vorbis-tools-1.2.0/ogg123' 
>>
>> source='http_transport.c' object='http_transport.o' libtool=no \
>>         DEPDIR=.deps depmode=none /bin/bash ../depcomp \
>>         /opt/studio/SOS11/SUNWspro/bin/cc 
>> -DSYSCONFDIR=\"/opt/csw/etc\" -DLOCALEDIR=\"/opt/csw/share/locale\" 
>> -DHAVE_CONFIG_H -I. -I.. -I/opt/csw/include   -I/opt/csw/include   
>> -I/opt/csw/include -I/opt/csw/include -D_REENTRANT  -I../include 
>> -I../intl  -I/opt/csw/include -D_REENTRANT  -xO4 -fast -w -fsimple 
>> -native -xcg92 -xO3 -xarch=v9 -I/opt/csw/include -c http_transport.c
>> "/opt/csw/include/curl/curlrules.h", line 134: zero or negative subscript
>> cc: acomp failed for http_transport.c
> 
> 
> Best regards
> 
>   -- Dago



From phil at bolthole.com  Wed Apr 15 16:59:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 07:59:31 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5F280.7070503@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
Message-ID: <20090415145931.GC16035@bolthole.com>

On Wed, Apr 15, 2009 at 09:43:12AM -0500, Chad Harp wrote:
> I opted not to make the changes to curlbuild.h because there were  
> several lines (other than the sizeof(long) we discussed) that would have  
> to change (see diff below).  I think this will be more easily adapted to  
> future versions.

for the record, I think you chose the wrong replacement values. please take
a deeper look into sys/types.h.

for example

> > #define CURL_TYPEOF_CURL_OFF_T long

should be

#define CURL_TYPEOF_CURL_OFF_T offset_t

and CURL_TYPEOF_CURL_OFF_TU should be u_offset_t

This will work cleanly for both 32bit and 64bit.




From harpchad at opencsw.org  Wed Apr 15 17:11:28 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 10:11:28 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415145931.GC16035@bolthole.com>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>	<49E4F507.1040503@opencsw.org>	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
Message-ID: <49E5F920.1050101@opencsw.org>

Philip Brown wrote:
> for the record, I think you chose the wrong replacement values. please take
> a deeper look into sys/types.h.
> 
> for example
> 
>>> #define CURL_TYPEOF_CURL_OFF_T long
> 
> should be
> 
> #define CURL_TYPEOF_CURL_OFF_T offset_t
> 
> and CURL_TYPEOF_CURL_OFF_TU should be u_offset_t
> 
> This will work cleanly for both 32bit and 64bit.
> 
> 

I didn't choose those values but I can propose them upstream.  I simply 
take the two headers generated from the builds and wrap them.  If I try 
to make a custom curlbuild.h I'm going to have to redo it every time 
there is a new curl release.



From phil at bolthole.com  Wed Apr 15 17:40:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 08:40:33 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E5F920.1050101@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
Message-ID: <20090415154033.GA96009@bolthole.com>

On Wed, Apr 15, 2009 at 10:11:28AM -0500, Chad Harp wrote:
> I didn't choose those values but I can propose them upstream.  I simply  
> take the two headers generated from the builds and wrap them.  If I try  
> to make a custom curlbuild.h I'm going to have to redo it every time  
> there is a new curl release.
>

right now, you have to add two files, AND patch a header file.

If you do as I suggest, you only have to patch a single header file.
It is less work :)

(however, if you do the patch in a way that is portable,then perhaps
  upstream will accept it in the future and it will be even better)



From harpchad at opencsw.org  Wed Apr 15 17:49:02 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Wed, 15 Apr 2009 10:49:02 -0500
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415154033.GA96009@bolthole.com>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>	<49E4F507.1040503@opencsw.org>	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>	<49E5F280.7070503@opencsw.org>	<20090415145931.GC16035@bolthole.com>	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
Message-ID: <49E601EE.9020309@opencsw.org>

Philip Brown wrote:
> On Wed, Apr 15, 2009 at 10:11:28AM -0500, Chad Harp wrote:
>> I didn't choose those values but I can propose them upstream.  I simply  
>> take the two headers generated from the builds and wrap them.  If I try  
>> to make a custom curlbuild.h I'm going to have to redo it every time  
>> there is a new curl release.
>>
> 
> right now, you have to add two files, AND patch a header file.
> 
> If you do as I suggest, you only have to patch a single header file.
> It is less work :)
> 
> (however, if you do the patch in a way that is portable,then perhaps
>   upstream will accept it in the future and it will be even better)
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I guess I'm not following, I don't patch anything, I just take the 
headers that curl generates and use one or the other based on the 
compile environment.

What are the appropriate portable format and suffix values for offset_t 
and u_offset_t?


From ihsan at opencsw.org  Wed Apr 15 17:50:04 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 15 Apr 2009 17:50:04 +0200
Subject: [csw-maintainers] wget Paket
In-Reply-To: <AB71266F-5B6B-4969-B5EB-A6107C344891@opencsw.org>
References: <49D235A7.2030605@opencsw.org>
	<AB71266F-5B6B-4969-B5EB-A6107C344891@opencsw.org>
Message-ID: <49E6022C.6020703@opencsw.org>

Hi Dago,

Am 31.3.2009 21:16 Uhr, Dagobert Michelsen schrieb:

>> I was thinking, that I could run a second build process with different
>> "configure" options. You provide already a possibility with the
>> modulations, but I don't need an optimized version. I know this is a
>> special case, but is this possible with Gar?
> 
> Absolutely. I just finished ncurses with a modulation for wide-characters.
> Just take a look at
>  
> <http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/ncurses/trunk/Makefile>

Worked perfectly.
Thanks you.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From dam at opencsw.org  Wed Apr 15 17:56:20 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 15 Apr 2009 17:56:20 +0200
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <20090415154033.GA96009@bolthole.com>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
Message-ID: <A1CE492E-EC3D-4DFF-809D-4A08AB666B65@opencsw.org>

Hi,

Am 15.04.2009 um 17:40 schrieb Philip Brown:
> On Wed, Apr 15, 2009 at 10:11:28AM -0500, Chad Harp wrote:
>> I didn't choose those values but I can propose them upstream.  I  
>> simply
>> take the two headers generated from the builds and wrap them.  If I  
>> try
>> to make a custom curlbuild.h I'm going to have to redo it every time
>> there is a new curl release.
>>
>
> right now, you have to add two files, AND patch a header file.

Well, now there is only one static conditional include. Everything else
is done automatically and needs no work on upstream version bump. Think
of this solution as least upstream dependency and least longterm effort.

The package in testing is working fine.


Best regards

   -- Dago


From valholla75 at gmail.com  Wed Apr 15 19:33:45 2009
From: valholla75 at gmail.com (Mike Watters)
Date: Wed, 15 Apr 2009 12:33:45 -0500
Subject: [csw-maintainers] gcc4 now in testing
Message-ID: <49E61A79.6090805@gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

gcc 4.3.3  NOW IN TESTING

OS specific compiles for:

solaris  8 "sparcv8, sparcv9" and i386
solaris  9 "sparcv8, sparcv9" and i386
solaris 10 "sparcv8, sparcv9" and "i386, amd64"

Results from make checks are checked into subversion
mgar/pkg/gcc4/trunk/files/test-results/

solaris9 test results are coming.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknmGnkACgkQW3VzXti4wcraeACfXKOfb8gYYmtktIuK1AjM9FiW
zIwAnjMmuK9AtF35UAx86PiDQd0bC4a0
=wzjZ
-----END PGP SIGNATURE-----


From phil at bolthole.com  Wed Apr 15 19:44:10 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:44:10 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <49E601EE.9020309@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
	<49E601EE.9020309@opencsw.org>
Message-ID: <20090415174410.GE976@bolthole.com>

On Wed, Apr 15, 2009 at 10:49:02AM -0500, Chad Harp wrote:
> I guess I'm not following, I don't patch anything, I just take the  
> headers that curl generates and use one or the other based on the  
> compile environment.

ohhhh.



> What are the appropriate portable format and suffix values for offset_t  
> and u_offset_t?

I dont understand the question.



From phil at bolthole.com  Wed Apr 15 19:44:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:44:52 -0700
Subject: [csw-maintainers] libcurl and 64 bit on x86
In-Reply-To: <A1CE492E-EC3D-4DFF-809D-4A08AB666B65@opencsw.org>
References: <1D6D5B97-4B34-4A93-B50A-35B5670AA6A0@opencsw.org>
	<49E4F507.1040503@opencsw.org>
	<239BA55E-7143-4C9D-A2D4-71E1F071173C@opencsw.org>
	<49E5F280.7070503@opencsw.org>
	<20090415145931.GC16035@bolthole.com>
	<49E5F920.1050101@opencsw.org>
	<20090415154033.GA96009@bolthole.com>
	<A1CE492E-EC3D-4DFF-809D-4A08AB666B65@opencsw.org>
Message-ID: <20090415174452.GF976@bolthole.com>

On Wed, Apr 15, 2009 at 05:56:20PM +0200, Dagobert Michelsen wrote:
> Well, now there is only one static conditional include. Everything else
> is done automatically and needs no work on upstream version bump. Think
> of this solution as least upstream dependency and least longterm effort.

I think the least LONGTERM effort, is to get upstream to have their code
work more portably.

which involves more shortterm work, of writing a patch.



From phil at bolthole.com  Wed Apr 15 19:48:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:48:06 -0700
Subject: [csw-maintainers] gcc4 now in testing
In-Reply-To: <49E61A79.6090805@gmail.com>
References: <49E61A79.6090805@gmail.com>
Message-ID: <20090415174806.GG976@bolthole.com>

On Wed, Apr 15, 2009 at 12:33:45PM -0500, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> gcc 4.3.3  NOW IN TESTING
> 
> OS specific compiles for:
> 
> solaris  8 "sparcv8, sparcv9" and i386
> solaris  9 "sparcv8, sparcv9" and i386
> solaris 10 "sparcv8, sparcv9" and "i386, amd64"

they dont follow prior package path conventions.

there's no gcc4 prefix.



From phil at bolthole.com  Wed Apr 15 19:52:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Apr 2009 10:52:45 -0700
Subject: [csw-maintainers] gcc4 now in testing
In-Reply-To: <20090415174806.GG976@bolthole.com>
References: <49E61A79.6090805@gmail.com> <20090415174806.GG976@bolthole.com>
Message-ID: <20090415175245.GI976@bolthole.com>

On Wed, Apr 15, 2009 at 10:48:06AM -0700, Philip Brown wrote:
> On Wed, Apr 15, 2009 at 12:33:45PM -0500, Mike Watters wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> > 
> > gcc 4.3.3  NOW IN TESTING
> > 
> > OS specific compiles for:
> > 
> > solaris  8 "sparcv8, sparcv9" and i386
> > solaris  9 "sparcv8, sparcv9" and i386
> > solaris 10 "sparcv8, sparcv9" and "i386, amd64"
> 
> they dont follow prior package path conventions.
> 
> there's no gcc4 prefix.

correction: some do, some dont.

gcc4core has gcc4 prefix.

gcc4corert does not.



From ihsan at opencsw.org  Wed Apr 15 20:38:06 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 15 Apr 2009 20:38:06 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>
Message-ID: <49E6298E.2030801@opencsw.org>

Hello Dago,

Am 14.4.2009 13:43 Uhr, Dagobert Michelsen schrieb:

>> I jumped in on the work on libcairo. William already prepared most of
>> what is necessary and I push in the latest GAR tweaks and provide 64 bit.
>> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>
> 
> New in testing/:
> 
> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz

Could you please install those libraries on build8st, so I could build
rrdtool there?



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From dam at opencsw.org  Wed Apr 15 21:42:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 15 Apr 2009 21:42:46 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <49E6298E.2030801@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>
	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>
	<49E6298E.2030801@opencsw.org>
Message-ID: <73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>

Hi Ihsan,

Am 15.04.2009 um 20:38 schrieb Ihsan Dogan:
> Am 14.4.2009 13:43 Uhr, Dagobert Michelsen schrieb:
>
>>> I jumped in on the work on libcairo. William already prepared most  
>>> of
>>> what is necessary and I push in the latest GAR tweaks and provide  
>>> 64 bit.
>>> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/ 
>>> gnome>
>>
>> New in testing/:
>>
>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>
> Could you please install those libraries on build8st, so I could build
> rrdtool there?

Done. This is exactly what build8st is for :-)


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 09:04:59 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 09:04:59 +0200
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
Message-ID: <0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>

Hi Chad,

Am 16.04.2009 um 03:05 schrieb Chad Harp:
> On Feb 24, 2009, at 2:05 PM, Chad Harp wrote:
>>
>> Dagobert Michelsen wrote:
>>> Hi Chad,
>>> I noticed there is an error in the current CSWftype2
>>> 2.3.8,REV=2009.02.16 in
>>>   /opt/csw/include/ft2build.h
>>> with the line
>>>   #include <freetype/config/ftheader.h>
>>> The path is instead
>>>   /opt/csw/include/freetype2/freetype/config/ftheader.h
>>> I guess it should either be in
>>>   /opt/csw/include/freetype2/ft2build.h
>>> or contain
>>>   #include <freetype2/freetype/config/ftheader.h>
>>
>>
>> That looks correct to me, can you send me more details on where  
>> you're getting the error?
>>
>> #include <freetype/config/ftheader.h>
>> with -I/opt/csw/include/freetype2
>> should find
>> /opt/csw/include/freetype2/freetype/config/ftheader.h
>>
>> Perhaps what you're compiling isn't using freetype-config?
>>
>> /opt/csw/bin/freetype-config --cflags
>> -I/opt/csw/include/freetype2 -I/opt/csw/include
>

> I was going through my list of things to look into and ran across  
> this.  I can't remember what the outcome was, did you get past this  
> error?

And:

Am 20.02.2009 um 12:56 schrieb Ihsan Dogan:
> Am 17.2.2009 19:03 Uhr, Dagobert Michelsen schrieb:
>>
>>> brauchst zum Kompilieren die aktuelle freetype2 Version aus /
>>> testing.
>>> Koenntest Du das bitte auf der buildfarm einspielen ?
>>> Ziel ist es, freetype2 und fontconfig nach einer Testphase
>>> gleichzeitig zu releasen.
>>
>> Ok, so freetype2 and fontconfig should be released at the
>> same time and freetype2 must be installed on the buildfarm
>> for building fontconfig.
>>
>> Ihsan, Phil: Should I do it this way violating standards or
>> should these packages be released one after another?
>
> As I understand, we don't have many choices, so it's fine for me.
> Important is, that a notify is sent to the maintainers list.

We have not build8st for this. AFAIK this has not been done yet.
Want to give it a try?


Best regards

   -- Dagp


From dam at opencsw.org  Thu Apr 16 10:05:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 10:05:45 +0200
Subject: [csw-maintainers] Package name now contains commit status
Message-ID: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>

Hi,

according to Bens great work the package name now contains the
repository status if the package has not been fully checked in.

The usual name is
   %{bitname}-%{SPKG_VERSION}%{SPKG_REVSTAMP}-%{SPKG_OSNAME}-%{arch}- 
CSW.pkg

Instead of the CSW there are now three other possible values:
   UNCOMMITTED  Not every file has been checked in or ignored
                ('svn status' not empty)
   NOSVN        The svn binary could not be found
   NOTVERSIONED The tree is not under version control

The change is in effect since r4338.


Best regards

   -- Dago


From ihsan at opencsw.org  Thu Apr 16 12:59:44 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 12:59:44 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>	<49E6298E.2030801@opencsw.org>
	<73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>
Message-ID: <49E70FA0.90700@opencsw.org>

Hi Dago,

Am 15.4.2009 21:42 Uhr, Dagobert Michelsen schrieb:

>>>> I jumped in on the work on libcairo. William already prepared most of
>>>> what is necessary and I push in the latest GAR tweaks and provide 64
>>>> bit.
>>>> Anybody else up for Gnome? Look here: <http://wiki.opencsw.org/gnome>
>>>
>>> New in testing/:
>>>
>>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>>> libcairo-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>>> libcairo_devel-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-i386-CSW.pkg.gz
>>> libcairo_doc-1.8.6,REV=2009.04.14-SunOS5.8-sparc-CSW.pkg.gz
>>
>> Could you please install those libraries on build8st, so I could build
>> rrdtool there?
> 
> Done. This is exactly what build8st is for :-)

Thanks.

Is it possible, that the new cairo packages are breaking pango?

Find 3rd-Party Libraries
checking for cairo_font_options_create in -lcairo... yes
checking cairo.h usability... no
checking cairo.h presence... no
checking for cairo.h... no
checking for pkg-config... pkg-config
checking for cairo_font_options_create in -lcairo... yes
checking cairo.h usability... yes
checking cairo.h presence... yes
checking for cairo.h... yes
checking for cairo_svg_surface_create in -lcairo... yes
checking cairo-svg.h usability... yes
checking cairo-svg.h presence... yes
checking for cairo-svg.h... yes
checking for cairo_pdf_surface_create in -lcairo... yes
checking cairo-pdf.h usability... yes
checking cairo-pdf.h presence... yes
checking for cairo-pdf.h... yes
checking for cairo_ps_surface_create in -lcairo... yes
checking cairo-ps.h usability... yes
checking cairo-ps.h presence... yes
checking for cairo-ps.h... yes
checking for pango_cairo_context_set_font_options in -lpango-1.0... no
checking for pkg-config... (cached) pkg-config
sh: gnome-config: not found
configure: WARNING:
----------------------------------------------------------------------------
* I found a copy of pkgconfig, but there is no pangocairo.pc file around.
  You may want to set the PKG_CONFIG_PATH variable to point to its
  location.
----------------------------------------------------------------------------




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From dam at opencsw.org  Thu Apr 16 14:39:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 14:39:34 +0200
Subject: [csw-maintainers] libcairo in testing
Message-ID: <94097AB5-019B-45C1-AF52-0A90A2C3CD56@opencsw.org>

Hi,

there is a libcairo 1.8.6 in testing/. As there are quite some
dependencies I would prefer someone else than me could test it.
The timely release is important, as there are a number of other
packages depending on it (including pango and rrdtool, which is
currently disfunctional in current/).


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 14:58:57 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 14:58:57 +0200
Subject: [csw-maintainers] build8x down
Message-ID: <DB296B4A-4D1C-4198-AEBC-D8D63505B2B0@opencsw.org>

Hi,

one of the mirrored disks of build8x died a few days ago.
On the attempt to replace the drive the Linux mirror driver
crashed the machine :-(

The machine will be available again shortly. You can use
the old build8x1 in the meantime.


Sorry for the inconvenience.

   -- Dago


From dam at opencsw.org  Thu Apr 16 15:28:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 15:28:15 +0200
Subject: [csw-maintainers] New subversion?
Message-ID: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>

Hi,

I remember there was work on an update of subversion which doesn't
seem to have been released nor is one in testing/. Is there an
obstacle in the way? There is need for a current one to sync up
with the SourceForge subversion.


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 15:37:16 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 15:37:16 +0200
Subject: [csw-maintainers] libcairo
In-Reply-To: <49E70FA0.90700@opencsw.org>
References: <9AE98745-6EC8-4CCD-B80F-C01C013C7ADA@opencsw.org>	<99616050-AAF9-40D3-9E1B-069C83BACBB7@opencsw.org>	<49E6298E.2030801@opencsw.org>
	<73FC0484-49A3-4258-BD27-9298B3CE6146@opencsw.org>
	<49E70FA0.90700@opencsw.org>
Message-ID: <09C55997-A791-4810-8FD0-24FC08BCAAE0@opencsw.org>

Hi Ihsan,

Am 16.04.2009 um 12:59 schrieb Ihsan Dogan:
> Is it possible, that the new cairo packages are breaking pango?

Well, you need pangocairo. That is included in pango and requires
a current cairo which hasn't been released yet. After release I'll
happily rebuild pango with all bells and whistles. You may test
the packages in testing/ to further speed up the process ;-)


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 16:10:49 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 16:10:49 +0200
Subject: [csw-maintainers] build8x down
In-Reply-To: <DB296B4A-4D1C-4198-AEBC-D8D63505B2B0@opencsw.org>
References: <DB296B4A-4D1C-4198-AEBC-D8D63505B2B0@opencsw.org>
Message-ID: <835018D9-BA3A-49A1-9605-0685D0209BA0@opencsw.org>

Hi,

Am 16.04.2009 um 14:58 schrieb Dagobert Michelsen:
> one of the mirrored disks of build8x died a few days ago.
> On the attempt to replace the drive the Linux mirror driver
> crashed the machine :-(
>
> The machine will be available again shortly. You can use
> the old build8x1 in the meantime.

build8x is available again.


Best regards

   -- Dago


From harpchad at opencsw.org  Thu Apr 16 16:32:26 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 16 Apr 2009 09:32:26 -0500
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
Message-ID: <49E7417A.7060307@opencsw.org>

Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 16.04.2009 um 03:05 schrieb Chad Harp:
>> On Feb 24, 2009, at 2:05 PM, Chad Harp wrote:
>>>
>>> Dagobert Michelsen wrote:
>>>> Hi Chad,
>>>> I noticed there is an error in the current CSWftype2
>>>> 2.3.8,REV=2009.02.16 in
>>>>   /opt/csw/include/ft2build.h
>>>> with the line
>>>>   #include <freetype/config/ftheader.h>
>>>> The path is instead
>>>>   /opt/csw/include/freetype2/freetype/config/ftheader.h
>>>> I guess it should either be in
>>>>   /opt/csw/include/freetype2/ft2build.h
>>>> or contain
>>>>   #include <freetype2/freetype/config/ftheader.h>
>>>
>>>
>>> That looks correct to me, can you send me more details on where 
>>> you're getting the error?
>>>
>>> #include <freetype/config/ftheader.h>
>>> with -I/opt/csw/include/freetype2
>>> should find
>>> /opt/csw/include/freetype2/freetype/config/ftheader.h
>>>
>>> Perhaps what you're compiling isn't using freetype-config?
>>>
>>> /opt/csw/bin/freetype-config --cflags
>>> -I/opt/csw/include/freetype2 -I/opt/csw/include
>>
> 
>> I was going through my list of things to look into and ran across 
>> this.  I can't remember what the outcome was, did you get past this 
>> error?
> 
> And:
> 
> Am 20.02.2009 um 12:56 schrieb Ihsan Dogan:
>> Am 17.2.2009 19:03 Uhr, Dagobert Michelsen schrieb:
>>>
>>>> brauchst zum Kompilieren die aktuelle freetype2 Version aus /
>>>> testing.
>>>> Koenntest Du das bitte auf der buildfarm einspielen ?
>>>> Ziel ist es, freetype2 und fontconfig nach einer Testphase
>>>> gleichzeitig zu releasen.
>>>
>>> Ok, so freetype2 and fontconfig should be released at the
>>> same time and freetype2 must be installed on the buildfarm
>>> for building fontconfig.
>>>
>>> Ihsan, Phil: Should I do it this way violating standards or
>>> should these packages be released one after another?
>>
>> As I understand, we don't have many choices, so it's fine for me.
>> Important is, that a notify is sent to the maintainers list.
> 
> We have not build8st for this. AFAIK this has not been done yet.
> Want to give it a try?
> 
> 
> Best regards
> 
>   -- Dagp
Freetype was updated to 2.3.8 in February, I think it's already on the 
build farm. (2.3.9 was release in March but I haven't updated yet).

Alexander Maier was working on a new fontconfig, but I don't think it 
has been finished yet.

I guess we should get it sorted out before the Cairo and Pango builds 
get too far along, it be nice if they were built against fontconfig 2.4 
(or later).



From dam at opencsw.org  Thu Apr 16 16:38:19 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 16:38:19 +0200
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <49E7417A.7060307@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
	<49E7417A.7060307@opencsw.org>
Message-ID: <22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>

Hi Chad,

Am 16.04.2009 um 16:32 schrieb Chad Harp:
> Freetype was updated to 2.3.8 in February, I think it's already on  
> the build farm. (2.3.9 was release in March but I haven't updated  
> yet).
>
> Alexander Maier was working on a new fontconfig, but I don't think  
> it has been finished yet.

Alexander, how far did you come? Where is the problem?

> I guess we should get it sorted out before the Cairo and Pango  
> builds get too far along, it be nice if they were built against  
> fontconfig 2.4 (or later).

Yes. I take a look at fontconfig now, this needs to be out of the door.
Alexander, please join in :-)


Best regards

   -- Dago


From harpchad at opencsw.org  Thu Apr 16 17:06:56 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 16 Apr 2009 10:06:56 -0500
Subject: [csw-maintainers] gnutls 2.6.5 in testing
Message-ID: <49E74990.1040407@opencsw.org>

gnutls-2.6.5,REV=2009.04.15-SunOS5.8-i386-CSW.pkg.gz
gnutls-2.6.5,REV=2009.04.15-SunOS5.8-sparc-CSW.pkg.gz
gnutls_devel-2.6.5,REV=2009.04.15-SunOS5.8-i386-CSW.pkg.gz
gnutls_devel-2.6.5,REV=2009.04.15-SunOS5.8-sparc-CSW.pkg.gz


From harpchad at opencsw.org  Thu Apr 16 17:25:34 2009
From: harpchad at opencsw.org (Chad Harp)
Date: Thu, 16 Apr 2009 10:25:34 -0500
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
	<49E7417A.7060307@opencsw.org>
	<22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>
Message-ID: <49E74DEE.5090003@opencsw.org>

I just put freetype 2.3.9 in testing.  If we get fontconfig 2.6.0 built 
against that we'll be at the latest stable for both.

Dagobert Michelsen wrote:
> Hi Chad,
> 
> Am 16.04.2009 um 16:32 schrieb Chad Harp:
>> Freetype was updated to 2.3.8 in February, I think it's already on the 
>> build farm. (2.3.9 was release in March but I haven't updated yet).
>>
>> Alexander Maier was working on a new fontconfig, but I don't think it 
>> has been finished yet.
> 
> Alexander, how far did you come? Where is the problem?
> 
>> I guess we should get it sorted out before the Cairo and Pango builds 
>> get too far along, it be nice if they were built against fontconfig 
>> 2.4 (or later).
> 
> Yes. I take a look at fontconfig now, this needs to be out of the door.
> Alexander, please join in :-)
> 
> 
> Best regards
> 
>   -- Dago



From dam at opencsw.org  Thu Apr 16 17:29:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 17:29:51 +0200
Subject: [csw-maintainers] Broken CSWftype2
In-Reply-To: <49E74DEE.5090003@opencsw.org>
References: <C3227F27-D523-4512-8CB4-31218530741A@opencsw.org>
	<49A45313.5080806@opencsw.org>
	<78F962BF-F6CE-4E60-9003-73163A11D67E@opencsw.org>
	<0502628F-119B-4562-B7EC-3888B38D1B7B@opencsw.org>
	<49E7417A.7060307@opencsw.org>
	<22D107A7-41FD-4182-91DA-1A170D7A0F7E@opencsw.org>
	<49E74DEE.5090003@opencsw.org>
Message-ID: <B8B6F2F5-843E-44FC-81DA-CAC092EC7B44@opencsw.org>

Hi Chad,

Am 16.04.2009 um 17:25 schrieb Chad Harp:
> Dagobert Michelsen wrote:
>> Am 16.04.2009 um 16:32 schrieb Chad Harp:
>>> Freetype was updated to 2.3.8 in February, I think it's already on  
>>> the build farm. (2.3.9 was release in March but I haven't updated  
>>> yet).
>>>
>>> Alexander Maier was working on a new fontconfig, but I don't think  
>>> it has been finished yet.
>>
>> Alexander, how far did you come? Where is the problem?
>>
>>> I guess we should get it sorted out before the Cairo and Pango  
>>> builds get too far along, it be nice if they were built against  
>>> fontconfig 2.4 (or later).
>>
>> Yes. I take a look at fontconfig now, this needs to be out of the  
>> door.
>> Alexander, please join in :-)
>>
>
> I just put freetype 2.3.9 in testing.  If we get fontconfig 2.6.0  
> built against that we'll be at the latest stable for both.

As I just heard Alexander is on vacation and will be back on monday, so
hopefully we have a full set of base libs by mid of next week.


Best regards+

   -- Dago


From mwatters at opencsw.org  Thu Apr 16 17:57:20 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 10:57:20 -0500
Subject: [csw-maintainers] gcc 4.3.3 miracle in testing
Message-ID: <49E75560.9010208@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I have "new" gcc4.3.3 packages in testing

this "should" run on all solaris platforms solaris 8 on
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnVWAACgkQLrhmsXMSLxeNCgCcCdSg6qYEuGbpptTrrdDA+tNJ
ykcAoMdZZJkiET54OGBkwwuIV6pSaqHx
=ERyu
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 16 18:09:21 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 18:09:21 +0200
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
Message-ID: <49E75831.8000401@opencsw.org>

Am 16.4.2009 10:05 Uhr, Dagobert Michelsen schrieb:

> according to Bens great work the package name now contains the
> repository status if the package has not been fully checked in.
> 
> The usual name is
>   %{bitname}-%{SPKG_VERSION}%{SPKG_REVSTAMP}-%{SPKG_OSNAME}-%{arch}-CSW.pkg
> 
> Instead of the CSW there are now three other possible values:
>   UNCOMMITTED  Not every file has been checked in or ignored
>                ('svn status' not empty)
>   NOSVN        The svn binary could not be found
>   NOTVERSIONED The tree is not under version control
> 
> The change is in effect since r4338.

I've just made a commit on build8s and went then to build8x to build the
the x86 version. While the file on build8s was fine, build8x put
UNCOMMITTED into the filename.

ihsan at build8s:~/staging/build-16.Apr.2009$ ll
total 2429
-rw-r--r--   1 ihsan    csw       592155 Apr 16 18:04
nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-UNCOMMITTED.pkg.gz
-rw-r--r--   1 ihsan    csw       622693 Apr 16 18:00
nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz


Bug or feature?




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Thu Apr 16 18:18:03 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 11:18:03 -0500
Subject: [csw-maintainers] please install gcc4* from testing on build8xt and
	build10xt
Message-ID: <49E75A3B.1040206@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


I would like to test the compiler on both platforms.

gcc testing on sparc solaris 9 and solaris 10
I compiled zlib and vim and both tested clean


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnWjoACgkQLrhmsXMSLxfnpwCdEYlNElt5XMQu0JXecTFrnrPF
SQAAnjiaEtWkcFe0QrVdHHwAMS8XOydE
=wvdp
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Thu Apr 16 18:40:06 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Apr 2009 12:40:06 -0400
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <49E75831.8000401@opencsw.org>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
	<49E75831.8000401@opencsw.org>
Message-ID: <1239899850-sup-285@ntdws12.chass.utoronto.ca>

Excerpts from Ihsan Dogan's message of Thu Apr 16 12:09:21 -0400 2009:
> ihsan at build8s:~/staging/build-16.Apr.2009$ ll
> total 2429
> -rw-r--r--   1 ihsan    csw       592155 Apr 16 18:04
> nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-UNCOMMITTED.pkg.gz
> -rw-r--r--   1 ihsan    csw       622693 Apr 16 18:00
> nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz
> 
> 
> Bug or feature?

Depends.  Are there any files listed in your working directory that
show up in `svn status` as ?.  If so, those can produce the
UNCOMMITTED tag.  The best approach here is to either commit the files
or add them to the ignore list.

If there are no stray/uncommitted files and you're still getting that
status, it's a bug.

-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090416/9915248e/attachment-0001.asc>

From mwatters at opencsw.org  Thu Apr 16 19:43:27 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 12:43:27 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
Message-ID: <49E76E3F.8000604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> I remember there was work on an update of subversion which doesn't
> seem to have been released nor is one in testing/. Is there an
> obstacle in the way? There is need for a current one to sync up
> with the SourceForge subversion.
> 
> 
> Best regards
> 
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I am packaging it,  I thought I submitted it to Phil
I will go back and see what the status is/was.



- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnbj8ACgkQLrhmsXMSLxftvACfT3/OGSP1dwXt1yF+b3qzChYa
beUAnRe9SFgeLG3gDsdWvHllqEs/nyr5
=f4pS
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr 16 19:46:43 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 12:46:43 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <49E76E3F.8000604@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
	<49E76E3F.8000604@opencsw.org>
Message-ID: <49E76F03.1040802@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mike Watters wrote:
> Dagobert Michelsen wrote:
>> Hi,
> 
>> I remember there was work on an update of subversion which doesn't
>> seem to have been released nor is one in testing/. Is there an
>> obstacle in the way? There is need for a current one to sync up
>> with the SourceForge subversion.
> 
> 
>> Best regards
> 
>>   -- Dago
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> I am packaging it,  I thought I submitted it to Phil
> I will go back and see what the status is/was.
> 
> 
> 
I can't seem to find anything on it in my email.
I will rebuild it today and push it into testing.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknnbwMACgkQLrhmsXMSLxdmFQCgzPiddU7PHqHr67ynr7QFwEU8
M5AAn1MBCd+I48ddg4XautWgdy5VFMVT
=Qbwh
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Apr 16 20:41:01 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 13:41:01 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <49E76F03.1040802@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>	<49E76E3F.8000604@opencsw.org>
	<49E76F03.1040802@opencsw.org>
Message-ID: <49E77BBD.1020204@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mike Watters wrote:
> Mike Watters wrote:
>> Dagobert Michelsen wrote:
>>> Hi,
>>> I remember there was work on an update of subversion which doesn't
>>> seem to have been released nor is one in testing/. Is there an
>>> obstacle in the way? There is need for a current one to sync up
>>> with the SourceForge subversion.
> 
>>> Best regards
>>>   -- Dago
>>> _______________________________________________
>>> maintainers mailing list
>>> maintainers at lists.opencsw.org
>>> https://lists.opencsw.org/mailman/listinfo/maintainers
>> I am packaging it,  I thought I submitted it to Phil
>> I will go back and see what the status is/was.
> 
> 
> 
> I can't seem to find anything on it in my email.
> I will rebuild it today and push it into testing.
> 
_______________________________________________
maintainers mailing list
maintainers at lists.opencsw.org
https://lists.opencsw.org/mailman/listinfo/maintainers

I remember what happened to svn.
It was in testing and I did a ISALIST check on the binaries.
The perl module had a bad ISALIST entry.

I removed it from testing with the intent on recompiling,
but got sidetracked with gcc.  I am rebuilding it right now.

new version 1.6.1


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknne70ACgkQLrhmsXMSLxdcEwCeKeFhxkF1JoOqtq/e9IJQmjfu
ueEAoIEX2cjacAImY0T8du0yjwtDTTj+
=fu73
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Apr 16 22:35:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:35:42 +0200
Subject: [csw-maintainers] OpenCSW & JDK redistribution
References: <31180549-9BB3-4763-BAC5-2D256AA7331C@baltic-online.de>
Message-ID: <8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>

Hi,

after lots of talks to people from Sun about JDK/JRE
redistribution here is a status where we are now:

Anfang der weitergeleiteten E-Mail:
> Von: Dalibor Topic <Dalibor.Topic at Sun.COM>
>
> Hi Dagobert,
>
> Thank you for your call on Tuesday and for taking some time to
> chat today. I'm CC:ing the current maintainer of OpenJDK 6,
> Joe Darcy on this e-mail.
>
> As discussed during the calls we had, I've had another look at
> the different JDK licensing options and discussed your inquiry
> with others inside Sun. After looking at the typical use cases
> for the different licenses, I don't think that any of the existing
> non-open-source JDK licensing options fits OpenCSW's use case
> of distributing easily redistributable stand-alone binary packages
> of the JDK.
>
> According to http://www.opencsw.org/about :
>
> " OpenCSW.org is a community effort to build and distribute binary
> packages (CSW, aka "Community SoftWare" packages) for
> production-grade Solaris releases."
> [snip]
> "We added "Open" to our organizational name name, to indicate that
> the packages will always be freely "open and available to use" by
> anyone. There will never be a cost to use them, or copy them.
> Additionally, it is our goal (although we arent there yet), that
> all packages will have publically open and easily reproducible
> build frameworks."
>
> The only release of the JDK that would really let OpenCSW meet
> that goal and really fits the description of "Open" in "OpenCSW"
> is OpenJDK 6.
>
> So I would strongly encourage the OpenCSW project to look into
> packaging OpenJDK 6 as part of OpenCSW, like Fedora, OpenSUSE,
> Ubuntu, Debian and many other community efforts with similar
> goals and approaches have successfully done so far. I would
> love to be able to list OpenCSW alongside other projects on
> http://openjdk.java.net/install/ !
>
> You can get the latest source code release, OpenJDK 6 b16, at
> http://download.java.net/openjdk/jdk6/ . A description of the
> latest changes is at
> http://blogs.sun.com/darcy/entry/openjdk_6_b16_source_bundle
>
> Unless you need the internal SNMP support, I would not bother
> with the binary plugs on the download page - the OpenJDK build
> should work fine without them, and the production-grade Linux
> distributions like Red Hat Enterprise Linux and CentOS aren't
> packaging them either in their OpenJDK 6 builds.
>
> OpenJDK 6 doesn't come with a plugin or webstart implementation
> (yet). If you'd like to package an independent plugin and
> Webstart implementation, they are available through the IcedTea
> project at http//icedtea.classpath.org .
>
> In order to ensure that OpenCSW users are getting an OpenJDK 6
> build that's compatible with JDK 6, the OpenCSW project can
> apply for access to the JCK under the OpenJDK Community TCK
> License Agreement, and test its own OpenJDK 6 builds against
> the JCK. Please see the OpenJDK Conformance Group web site for
> details at http://openjdk.java.net/groups/conformance/ .
>
> If you decide to package OpenJDK 6, and during your work
> come up with bug fixes, and other improvements, it would be
> nice if you would submit them back to the OpenJDK project using
> the OpenJDK Bugzilla instance at http://bugs.openjdk.java.net .
>
> Happy packaging, and all the best for OpenCSW! And do let us
> know how things go - I'm regularly on #openjdk on irc.oftc.net
> during German afternoons / evenings.
>
> cheers,
> dalibor topic

Additionally, he suggested to make a meta-package with manual
download like some Linux distributions. Any volunteers for
packaging OpenJDK?


Best regards

  -- Dago



From dam at opencsw.org  Thu Apr 16 22:40:50 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:40:50 +0200
Subject: [csw-maintainers] [csw-buildfarm] please install gcc4* from
	testing on build8xt and build10xt
In-Reply-To: <49E75A3B.1040206@opencsw.org>
References: <49E75A3B.1040206@opencsw.org>
Message-ID: <11C9C55E-A03E-40C2-95A1-B2BD8B71DF23@opencsw.org>

Hi Mike,

Am 16.04.2009 um 18:18 schrieb Mike Watters:
> Re: [csw-buildfarm] please install gcc4* from testing on build8xt  
> and build10xt
> I would like to test the compiler on both platforms.

The compiler etc. totals to 354 MB, wow!
build8xt is updating now (the old build8x, actually).
Unfortunately we don't have build10xt right now, build10x is
the global zone. We may move that into a local zone and
add build8xt as extra zone, but more disk space is needed
for this.

Jan: Is there extra space available, like 10 GB on the ESX farm?

> gcc testing on sparc solaris 9 and solaris 10
> I compiled zlib and vim and both tested clean

Cool.


Best regards

   -- Dago


From dam at opencsw.org  Thu Apr 16 22:46:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:46:01 +0200
Subject: [csw-maintainers] Machine renaming: build8x1 (old build8x) is now
	build8xt
Message-ID: <952B97E0-07AC-47B1-8C9D-910EBC6B5E31@opencsw.org>

Hi,

the old build8x (before the migration to the V65z) and former
build8x1 is now called build8xt for package testing. It is
currently used for gcc4 compilation tests. Just in case you
are wondering...


Best regards

   -- Dago


From phil at bolthole.com  Thu Apr 16 22:52:02 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 16 Apr 2009 13:52:02 -0700
Subject: [csw-maintainers] OpenCSW & JDK redistribution
In-Reply-To: <8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>
References: <31180549-9BB3-4763-BAC5-2D256AA7331C@baltic-online.de>
	<8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>
Message-ID: <20090416205202.GE74250@bolthole.com>

On Thu, Apr 16, 2009 at 10:35:42PM +0200, Dagobert Michelsen wrote:
> Hi,
>
> after lots of talks to people from Sun about JDK/JRE
> redistribution here is a status where we are now:

>...


Did you make it clear that we really dont care about having the source code
for the jdk itself; we just want to redistribute it, with full attributions
left to Sun?

I'd rather we redistribute have "FullJDK" than "OpenJDK", if you know what
I mean ;-)


From dam at opencsw.org  Thu Apr 16 22:57:42 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 16 Apr 2009 22:57:42 +0200
Subject: [csw-maintainers] OpenCSW & JDK redistribution
In-Reply-To: <20090416205202.GE74250@bolthole.com>
References: <31180549-9BB3-4763-BAC5-2D256AA7331C@baltic-online.de>
	<8DD275CB-DCE2-4F73-B3D9-44BEF223DD8F@opencsw.org>
	<20090416205202.GE74250@bolthole.com>
Message-ID: <A41FB4A0-9EA9-4198-A6AC-2B608C8F3487@opencsw.org>

Hi Phil,

Am 16.04.2009 um 22:52 schrieb Philip Brown:
> On Thu, Apr 16, 2009 at 10:35:42PM +0200, Dagobert Michelsen wrote:
>> after lots of talks to people from Sun about JDK/JRE
>> redistribution here is a status where we are now:
>
>> ...
>
>
> Did you make it clear that we really dont care about having the  
> source code
> for the jdk itself; we just want to redistribute it, with full  
> attributions
> left to Sun?

Yes, I made that clear. I talked for an hour to Dalibor about this.
There are lots of licenses tailored for different usecases, for
OEM-bundling with software like Veritas does it, for distributions
like SchilliX, but nothing for package archives. This mixup of
licenses is going to be ended with OpenJDK. I guess I am at an
end here, if you have contacts deeper inside Sun to make a
special license for our usecase please do so.

> I'd rather we redistribute have "FullJDK" than "OpenJDK", if you  
> know what
> I mean ;-)

I know what you mean. That's why I already made these beautiful
GAR descriptions for the production JDK/JRE in testing :-(


Best regards

   -- Dago


From ihsan at opencsw.org  Thu Apr 16 22:59:51 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 22:59:51 +0200
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <1239899850-sup-285@ntdws12.chass.utoronto.ca>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>	<49E75831.8000401@opencsw.org>
	<1239899850-sup-285@ntdws12.chass.utoronto.ca>
Message-ID: <49E79C47.1050907@opencsw.org>

Am 16.4.2009 18:40 Uhr, Ben Walton schrieb:

>> ihsan at build8s:~/staging/build-16.Apr.2009$ ll
>> total 2429
>> -rw-r--r--   1 ihsan    csw       592155 Apr 16 18:04
>> nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-UNCOMMITTED.pkg.gz
>> -rw-r--r--   1 ihsan    csw       622693 Apr 16 18:00
>> nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz
>>
>>
>> Bug or feature?
> 
> Depends.  Are there any files listed in your working directory that
> show up in `svn status` as ?.  If so, those can produce the
> UNCOMMITTED tag.  The best approach here is to either commit the files
> or add them to the ignore list.

No, there are not.

> If there are no stray/uncommitted files and you're still getting that
> status, it's a bug.

A few hours later, I've run "gmake package" again and the filename is
now correct. This is a little bit odd, because I haven't change anything.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From ihsan at opencsw.org  Thu Apr 16 23:05:07 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 16 Apr 2009 23:05:07 +0200
Subject: [csw-maintainers] nsd 3.2.1 in testing
Message-ID: <49E79D83.9090305@opencsw.org>

Hello,

I've packaged NSD - Name Server Daemon. NSD is a BSD licensed DNS server
and a real alternative to the widely known Bind.

NSD has been developed by NLnet Labs in cooperation with Ripe. Security
and performance are one of the project goals.

http://mirror.opencsw.org/testing/nsd-3.2.1,REV=2009.04.16-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/nsd-3.2.1,REV=2009.04.16-SunOS5.8-sparc-CSW.pkg.gz



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Fri Apr 17 05:20:16 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 16 Apr 2009 22:20:16 -0500
Subject: [csw-maintainers] pysqlite2 now in testing
Message-ID: <49E7F570.3010703@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	update to version 2.4.0
- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknn9XAACgkQLrhmsXMSLxdWEgCfU7WXf1l3XhFNDDiZwTWOu90v
xn4AoI4CFx0XFprkzDQOAgkc7p6z9GMP
=8Nq/
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Fri Apr 17 17:09:38 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 17 Apr 2009 10:09:38 -0500
Subject: [csw-maintainers] new gcc 4.3.3 for Intel in testing
Message-ID: <49E89BB2.7000500@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	fixed linking errors in packages

errors:

/opt/csw/gcc4/bin/cpp <linked pathname>
/opt/csw/gcc4/bin/gcc <linked pathname>
ERROR: attribute verification of </opt/csw/gcc4/bin/gcc> failed
    pathname does not exist
    unable to create link to
</opt/csw/gcc4/bin/i386/i386-pc-solaris2.8-gcc-4.3.3=/opt/csw/bin/isaexec>
/opt/csw/gcc4/bin/gccbug <linked pathname>
/opt/csw/gcc4/bin/gcov <linked pathname>
/opt/csw/gcc4/bin/i386/gcc <linked pathname>
ERROR: attribute verification of </opt/csw/gcc4/bin/i386/gcc> failed
    pathname does not exist
    unable to create link to </opt/csw/gcc4/bin/gcc>
/opt/csw/gcc4/bin/i386/i386-pc-solaris2.8-gcc <linked pathname>


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknom7EACgkQLrhmsXMSLxeEdQCeLA3HbXem01RsAYx3gBhRz7pW
RGoAn3OWz2v+NQ/jvzxSsDhQ0sPF7mDK
=5Nt1
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Fri Apr 17 18:39:35 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 17 Apr 2009 12:39:35 -0400
Subject: [csw-maintainers] Package name now contains commit status
In-Reply-To: <49E79C47.1050907@opencsw.org>
References: <CED91C04-231C-4952-AB1B-2F68F9C49C39@opencsw.org>
	<49E75831.8000401@opencsw.org>
	<1239899850-sup-285@ntdws12.chass.utoronto.ca>
	<49E79C47.1050907@opencsw.org>
Message-ID: <1239986336-sup-3099@ntdws12.chass.utoronto.ca>

Excerpts from Ihsan Dogan's message of Thu Apr 16 16:59:51 -0400 2009:
> > If there are no stray/uncommitted files and you're still getting that
> > status, it's a bug.
> 
> A few hours later, I've run "gmake package" again and the filename is
> now correct. This is a little bit odd, because I haven't change anything.

Ok.  That is strange.  I'll have a look at it tonight to see if I can
replicated it.

Thanks
-Ben

-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090417/0f573d24/attachment-0001.asc>

From mwatters at opencsw.org  Fri Apr 17 20:34:58 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 17 Apr 2009 13:34:58 -0500
Subject: [csw-maintainers] New subversion?
In-Reply-To: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
References: <F04BE5AC-8114-4330-AAD3-7724F9BB690D@opencsw.org>
Message-ID: <49E8CBD2.2040209@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> I remember there was work on an update of subversion which doesn't
> seem to have been released nor is one in testing/. Is there an
> obstacle in the way? There is need for a current one to sync up
> with the SourceForge subversion.
> 
> 
> Best regards
> 
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Sorry for the delay,  I am having issue getting 1.6.1 to compile.
I tried rebuilding 1.5.6 that I had already packaged, and
it seems something changed on the build servers and it won't compile either.
I look to determine what changed after I get 1.6.1 to compile.
I will keep working on it, but just wanted to give a status of where it is.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknoy9IACgkQLrhmsXMSLxfmvACfTYIiswPZP/t+4EnPOHZ8Gwrk
pgIAoMGpFWtL4JiroXRMnkE3ulVeiCAW
=uMgd
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Fri Apr 17 21:50:59 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 17 Apr 2009 14:50:59 -0500
Subject: [csw-maintainers] sudo 1.7.0 now in testing
Message-ID: <49E8DDA3.1080507@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

changes:
	fixed some package inconsistencies.
- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkno3aMACgkQLrhmsXMSLxcHCwCgz/2AMyHyc17TWPz07d9C1/xu
B18AoJvzbr2AycD6mZ3ROP4qCh0mXf0O
=xIgk
-----END PGP SIGNATURE-----


From william at wbonnet.net  Sat Apr 18 22:33:47 2009
From: william at wbonnet.net (William Bonnet)
Date: Sat, 18 Apr 2009 22:33:47 +0200
Subject: [csw-maintainers] /testing pixman 0.15.2
Message-ID: <49EA392B.7040003@wbonnet.net>

Hi

I have updated pixman package to version 0.15.2. This package now 
includes 64bits librairies

cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From maybird1776 at yahoo.com  Sun Apr 19 02:40:16 2009
From: maybird1776 at yahoo.com (ken mays)
Date: Sat, 18 Apr 2009 17:40:16 -0700 (PDT)
Subject: [csw-maintainers] GNOME: GTK+ 2.16.1 stable release
Message-ID: <46659.65772.qm@web111306.mail.gq1.yahoo.com>


This is the next package needed for building GNOME.
I'll submit a report.

~ Ken



      


From phil at bolthole.com  Sun Apr 19 02:47:16 2009
From: phil at bolthole.com (Philip Brown)
Date: Sat, 18 Apr 2009 17:47:16 -0700
Subject: [csw-maintainers] GNOME: GTK+ 2.16.1 stable release
In-Reply-To: <46659.65772.qm@web111306.mail.gq1.yahoo.com>
References: <46659.65772.qm@web111306.mail.gq1.yahoo.com>
Message-ID: <20090419004716.GA87849@bolthole.com>

On Sat, Apr 18, 2009 at 05:40:16PM -0700, ken mays wrote:
> 
> This is the next package needed for building GNOME.
> I'll submit a report.

not sure what you mean by "report"... but perhaps you might simply update
the wiki page.

wiki.opencsw.org/gnome


From william at wbonnet.net  Sun Apr 19 11:00:21 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 19 Apr 2009 11:00:21 +0200
Subject: [csw-maintainers] GAR and PKG_CONFIG_PATH
Message-ID: <49EAE825.9020700@wbonnet.net>

Hi

I am building X11 libs and i need to set PKG_CONFIG_PATH and i have a 
few problems... i am looking for the good way to define the 
PKG_CONFIG_PATH variable and add to its value /opt/csw/X11/lib/pkgconfig.

I tried either setting

EXTRA_PKGCONFIG_PATH  = /opt/csw/X11/lib/pkgconfig

or

PKG_CONFIG_PATH  = /opt/csw/X11/lib/pkgconfig

in Makefile or category file, but it is not taken in account by GAR 
scripts. The only work around i dound is to set it and export into the 
shell before calling gmake.

So please anyone to explain me what i did wrong ? :)

thanksin advance

cheers
W.

PS: mgar/pkg/x11/libXdmcp is the example package

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From trygvis at opencsw.org  Sun Apr 19 14:11:45 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sun, 19 Apr 2009 14:11:45 +0200
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
Message-ID: <49EB1501.5090006@opencsw.org>

Howdy,

I used some wikidot magic to automatically create the list of packages
on that is shown on [1]. It will now automatically list all the pages
tagged with "package". So if you write some documentation for a package,
please tag it with "package" and it'll show up automatically.

[1]: http://wiki.opencsw.org/packages

--
Trygve


From phil at bolthole.com  Sun Apr 19 17:01:34 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 19 Apr 2009 08:01:34 -0700
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <49EB1501.5090006@opencsw.org>
References: <49EB1501.5090006@opencsw.org>
Message-ID: <20090419150134.GA9243@bolthole.com>

On Sun, Apr 19, 2009 at 02:11:45PM +0200, Trygve Laugst?l wrote:
> Howdy,
> 
> I used some wikidot magic to automatically create the list of packages
> on that is shown on [1]. It will now automatically list all the pages
> tagged with "package". So if you write some documentation for a package,
> please tag it with "package" and it'll show up automatically.
> 
> [1]: http://wiki.opencsw.org/packages

nice trick!


Perhaps you could pick a different title for the page, and possible url,
though?

"packagedocs"  perhaps?



From trygvis at opencsw.org  Sun Apr 19 17:52:19 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sun, 19 Apr 2009 17:52:19 +0200
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <20090419150134.GA9243@bolthole.com>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
Message-ID: <49EB48B3.1080506@opencsw.org>

Philip Brown wrote:
> On Sun, Apr 19, 2009 at 02:11:45PM +0200, Trygve Laugst?l wrote:
>> Howdy,
>>
>> I used some wikidot magic to automatically create the list of packages
>> on that is shown on [1]. It will now automatically list all the pages
>> tagged with "package". So if you write some documentation for a package,
>> please tag it with "package" and it'll show up automatically.
>>
>> [1]: http://wiki.opencsw.org/packages
> 
> nice trick!
> 
> 
> Perhaps you could pick a different title for the page, and possible url,
> though?
> 
> "packagedocs"  perhaps?

I tried to think of a better name, but I couldn't really find a name I 
liked more so I just went with what already was there.

One thing I've been thinking about is creating documentation for users 
and (future) maintainers. If I where to take over a package I would 
appreciate some documentation on how to test the package.

How about creating two tags:
* user-doc: documentation for users (basically what those pages are today)
* maintainer-doc: documentation for maintainers, which would include 
tricks to build the package and how to test it.

--
Trygve


From dam at opencsw.org  Sun Apr 19 22:55:04 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Apr 2009 22:55:04 +0200
Subject: [csw-maintainers] GAR and PKG_CONFIG_PATH
In-Reply-To: <49EAE825.9020700@wbonnet.net>
References: <49EAE825.9020700@wbonnet.net>
Message-ID: <9DAF9057-B813-4CE8-926C-E36EE97A3D00@opencsw.org>

Hi William,

Am 19.04.2009 um 11:00 schrieb William Bonnet:
> I am building X11 libs and i need to set PKG_CONFIG_PATH and i have  
> a few problems... i am looking for the good way to define the  
> PKG_CONFIG_PATH variable and add to its value /opt/csw/X11/lib/ 
> pkgconfig.
>
> I tried either setting
>
> EXTRA_PKGCONFIG_PATH  = /opt/csw/X11/lib/pkgconfig
>
> or
>
> PKG_CONFIG_PATH  = /opt/csw/X11/lib/pkgconfig
>
> in Makefile or category file, but it is not taken in account by GAR  
> scripts.

This is basically ok, however I have some remarks.

You use this in x11/category.mk:

# pkg-config options
PKG_CONFIG_PATH  = /opt/csw/X11/lib/pkgconfig
PKG_CONFIG_PATH += $(DESTDIR)/opt/csw/lib/pkgconfig
PKG_CONFIG_PATH += $(DESTDIR)/opt/csw/X11/lib/pkgconfig

Please avoid DESTDIR as much as possible as it circumvents the standard
workflow of
   build base -> release -> install -> build dependent

There should be something to add category-specific stuff to pkgconfig:
   _CATEGORY_PKG_CONFIG_PATH = $(abspath $(prefix)/X11/lib/$ 
(MM_LIBDIR)/pkgconfig)

I committed the appropriate changes in r4413 and changed x11/category.mk
accordingly.

You can always check the value of PKG_CONFIG_PATH by typing

> build8s% gmake modenv
>           Arch: sparc
>         Kernel: sparcv9
>
> Default ISA 32: sparcv8
> Default ISA 64: sparcv9
>
> Requested ISAs: sparcv8 sparcv9 i386 amd64
>    Needed ISAs: sparcv8 sparcv9
>     Build ISAs: sparcv8 sparcv9
>
>   ISAEXEC dirs: /opt/csw/bin /opt/csw/sbin /opt/csw/libexec
>  ISAEXEC files:
>
>  Merge include:
>  Merge exclude:     /opt/csw/share/info/dir /opt/csw/lib/.*\.la .* 
> \~ /opt/csw/lib/.*\.a
>
>     Modulators: ISA
>    Modulations: isa-sparcv8 isa-sparcv9
>
> Requested compiler flags:
>
> * Modulation isa-sparcv8: ISA=sparcv8
>            PATH = /home/dam/mgar/pkg/x11/libXdmcp/trunk/work/install- 
> isa-sparcv8/opt/csw/bin:/home/dam/mgar/pkg/x11/libXdmcp/trunk/work/ 
> install-isa-sparcv8/opt/csw/bin:/home/dam/mgar/pkg/x11/libXdmcp/ 
> trunk/work/install-isa-sparcv8/opt/csw/sbin:/home/dam/mgar/pkg/x11/ 
> libXdmcp/trunk/work/install-isa-sparcv8/opt/csw/sbin:/opt/csw/bin:/ 
> opt/csw/bin:/opt/csw/sbin:/opt/csw/sbin:/opt/studio/SOS11/SUNWspro/ 
> bin:/home/dam/mgar/pkg/x11/libXdmcp/trunk/gar/bin:/usr/bin:/usr/ 
> sbin:/usr/java/bin:/usr/ccs/bin:/usr/openwin/bin
> PKG_CONFIG_PATH = /opt/csw/lib/pkgconfig:/opt/csw/lib/X11/pkgconfig
>          CFLAGS = -xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION - 
> xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION -xlibmil - 
> errtags=yes -erroff=E_EMPTY_DECLARATION
>        CXXFLAGS = -xlibmil -xlibmopt -features=tmplife -norunpath - 
> xlibmil -xlibmopt -features=tmplife -norunpath -xlibmil -xlibmopt - 
> features=tmplife -norunpath
>        CPPFLAGS =
>         LDFLAGS = -L/opt/csw/lib -R/opt/csw/lib -L/opt/csw/lib -R/ 
> opt/csw/lib -L/opt/csw/lib -R/opt/csw/lib
>         ASFLAGS =
>        OPTFLAGS = -xO3 -xarch=v8
>
> * Modulation isa-sparcv9: ISA=sparcv9
>            PATH = /home/dam/mgar/pkg/x11/libXdmcp/trunk/work/install- 
> isa-sparcv9/opt/csw/bin/sparcv9:/home/dam/mgar/pkg/x11/libXdmcp/ 
> trunk/work/install-isa-sparcv9/opt/csw/bin:/home/dam/mgar/pkg/x11/ 
> libXdmcp/trunk/work/install-isa-sparcv9/opt/csw/sbin/sparcv9:/home/ 
> dam/mgar/pkg/x11/libXdmcp/trunk/work/install-isa-sparcv9/opt/csw/ 
> sbin:/opt/csw/bin/sparcv9:/opt/csw/bin:/opt/csw/sbin/sparcv9:/opt/ 
> csw/sbin:/opt/studio/SOS11/SUNWspro/bin:/home/dam/mgar/pkg/x11/ 
> libXdmcp/trunk/gar/bin:/usr/bin:/usr/sbin:/usr/java/bin:/usr/ccs/ 
> bin:/usr/openwin/bin
> PKG_CONFIG_PATH = /opt/csw/lib/64/pkgconfig:/opt/csw/lib/64/X11/ 
> pkgconfig
>          CFLAGS = -xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION - 
> xlibmil -errtags=yes -erroff=E_EMPTY_DECLARATION -xlibmil - 
> errtags=yes -erroff=E_EMPTY_DECLARATION
>        CXXFLAGS = -xlibmil -xlibmopt -features=tmplife -norunpath - 
> xlibmil -xlibmopt -features=tmplife -norunpath -xlibmil -xlibmopt - 
> features=tmplife -norunpath
>        CPPFLAGS =
>         LDFLAGS = -L/opt/csw/lib -R/opt/csw/lib -L/opt/csw/lib -R/ 
> opt/csw/lib -L/opt/csw/lib -R/opt/csw/lib
>         ASFLAGS =
>        OPTFLAGS = -xO3 -xarch=v9
>

> The only work around i dound is to set it and export into the shell  
> before calling gmake.


Well, it is only passed during configure-time:
   <http://apps.sourceforge.net/trac/gar/browser/csw/mgar/gar/v2/gar.conf.mk#L567 
 >
If you need it during build-time too you can set
   EXTRA_BUILD_EXPORTS = PKG_CONFIG_PATH
but it usually shouldn't be necessary.

> So please anyone to explain me what i did wrong ? :)

The code looks like this:
      1     comand # This is for foo-config chaos
   2471 dmichelsen PKG_CONFIG_DIRS ?= $(libdir_install) $ 
(EXTRA_PKG_CONFIG_DIRS)
   2471 dmichelsen PKG_CONFIG_PATH ?= $(call MAKEPATH,$(foreach D,$ 
(PKG_CONFIG_DIRS),$(abspath $D/$(MM_LIBDIR)/pkgconfig)) $(EXTRA_PKGCON
FIG_PATH))

The mistyping of PKGCONFIG is a bug, as pkg-config itself uses  
PKG_CONFIG, this should be used.
This too is fixed in r4413 too (now EXTRA_PKG_CONFIG_PATH, it hadn't  
been used anyway).

Additionally, I removed in r2471 in gar.conf.mk (line 547)
   http://apps.sourceforge.net/trac/gar/changeset/2471
the export as I was not aware that it was really needed in a forked  
process.
This is fixed in r4415.

BTW, there is a much easier way to use dynamic licenses. Just use  
dynamic gspecs
and everything will be done automatically as the license has the  
default name.
I converted this package as example for you and the package builds  
fine now,
however, only 32 bit. For 64 bit the category.mk must have all the  
MM_*-stuff
from gar.conf.mk that differentiates pathes for isa-specific stuff  
during
install and merge. I'll take a look at that tomorrow.


I apologize for this accumulation of inaccuracies and confusion.


Best regards

   -- Dago


From phil at bolthole.com  Mon Apr 20 04:52:38 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 19 Apr 2009 19:52:38 -0700
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <49EB48B3.1080506@opencsw.org>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
	<49EB48B3.1080506@opencsw.org>
Message-ID: <20090420025238.GA87575@bolthole.com>

On Sun, Apr 19, 2009 at 05:52:19PM +0200, Trygve Laugst??l wrote:
> Philip Brown wrote:
>> Perhaps you could pick a different title for the page, and possible url,
>> though?
>>
>> "packagedocs"  perhaps?
>
> I tried to think of a better name, but I couldn't really find a name I  
> liked more so I just went with what already was there.

but what was there, does not suit the purpose :-/
The usage conflicts with the core site's 
http://www.opencsw.org/packages

To my mind, I think that for
http://XXXX.opencsw.org/NAMEHERE
that "NAMEHERE", should ideally be somewhat unique across the organization,
reguardless of which specific sub-site it is currently hosted it.


> One thing I've been thinking about is creating documentation for users  
> and (future) maintainers. If I where to take over a package I would  
> appreciate some documentation on how to test the package.
>
> How about creating two tags:
> * user-doc: documentation for users (basically what those pages are today)
> * maintainer-doc: documentation for maintainers, which would include  
> tricks to build the package and how to test it.

makes sense. ideally, they would somehow automatically have a button or
link to toggle between userlevel and maintainer level docs.

(not the same page; just that each page links to its counterpart)



From dam at opencsw.org  Mon Apr 20 10:47:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 20 Apr 2009 10:47:46 +0200
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <20090420025238.GA87575@bolthole.com>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
	<49EB48B3.1080506@opencsw.org>
	<20090420025238.GA87575@bolthole.com>
Message-ID: <88906185-3CC2-4BC8-9064-5A4C9EFD6BE3@opencsw.org>

Hi,

Am 20.04.2009 um 04:52 schrieb Philip Brown:
> On Sun, Apr 19, 2009 at 05:52:19PM +0200, Trygve Laugst??l wrote:
>> Philip Brown wrote:
>>> Perhaps you could pick a different title for the page, and  
>>> possible url,
>>> though?
>>>
>>> "packagedocs"  perhaps?
>>
>> I tried to think of a better name, but I couldn't really find a  
>> name I
>> liked more so I just went with what already was there.
>
> but what was there, does not suit the purpose :-/
> The usage conflicts with the core site's
> http://www.opencsw.org/packages
>
> To my mind, I think that for
> http://XXXX.opencsw.org/NAMEHERE
> that "NAMEHERE", should ideally be somewhat unique across the  
> organization,
> reguardless of which specific sub-site it is currently hosted it.

Yes. Ideally, opencsw.org/packages and the wikidot package page  
information
should be on one package, with automated content from OpenCSW and wiki
content. Maybe on the new website? BTW, is someone currently working on
it? I remember we had some helping hands offers :-)

>> One thing I've been thinking about is creating documentation for  
>> users
>> and (future) maintainers. If I where to take over a package I would
>> appreciate some documentation on how to test the package.
>>
>> How about creating two tags:
>> * user-doc: documentation for users (basically what those pages are  
>> today)
>> * maintainer-doc: documentation for maintainers, which would include
>> tricks to build the package and how to test it.
>
> makes sense. ideally, they would somehow automatically have a button  
> or
> link to toggle between userlevel and maintainer level docs.
>
> (not the same page; just that each page links to its counterpart)

The maintainer doc should be in the GAR Wiki. It was just moved to
Trac and a tight integration of code and wiki-content is possible
now.


Best regards

   -- Dago

From rupert at opencsw.org  Mon Apr 20 15:29:56 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 20 Apr 2009 15:29:56 +0200
Subject: [csw-maintainers] oracle buys sun ...
Message-ID: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>

wow ... that was unexpected: http://www.google.com/finance?q=NASDAQ%3AJAVA.

rupert.


From maciej at opencsw.org  Mon Apr 20 15:36:42 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 20 Apr 2009 14:36:42 +0100
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
Message-ID: <ce97543b0904200636q692d1955o8e29e77a545d17a6@mail.gmail.com>

On Mon, Apr 20, 2009 at 2:29 PM, rupert THURNER <rupert at opencsw.org> wrote:
> wow ... that was unexpected: http://www.google.com/finance?q=NASDAQ%3AJAVA.

Oracle offered $0.10 more per share. They now own MySQL.


From phil at bolthole.com  Mon Apr 20 15:58:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 20 Apr 2009 06:58:59 -0700
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
Message-ID: <20090420135859.GA81648@bolthole.com>

On Mon, Apr 20, 2009 at 03:29:56PM +0200, rupert THURNER wrote:
> wow ... that was unexpected.

not really. our local unix users' group speculated oracle was the next one
in line, as soon as the IBM deal dropped out :-)

and it's a much better fit than IBM.


on the darker side, oracle now "owns" both berkeleydb and mysql.

sigh.
 


From william at wbonnet.net  Mon Apr 20 16:26:59 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 20 Apr 2009 16:26:59 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <20090420135859.GA81648@bolthole.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
	<20090420135859.GA81648@bolthole.com>
Message-ID: <49EC8633.1070902@wbonnet.net>

Hi
> not really. our local unix users' group speculated oracle was the next one
> in line, as soon as the IBM deal dropped out :-)
>   
I was betting buzz around IBM buying Sun was just a diversionary movement.
> and it's a much better fit than IBM.
>
>
> on the darker side, oracle now "owns" both berkeleydb and mysql.
>   
And glassfish... they baught BEA (thus Weblogic) last year.


From phil at bolthole.com  Mon Apr 20 18:18:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 20 Apr 2009 09:18:26 -0700
Subject: [csw-maintainers] Package documentation on wiki.opencsw.org
In-Reply-To: <88906185-3CC2-4BC8-9064-5A4C9EFD6BE3@opencsw.org>
References: <49EB1501.5090006@opencsw.org> <20090419150134.GA9243@bolthole.com>
	<49EB48B3.1080506@opencsw.org>
	<20090420025238.GA87575@bolthole.com>
	<88906185-3CC2-4BC8-9064-5A4C9EFD6BE3@opencsw.org>
Message-ID: <20090420161826.GC72003@bolthole.com>

On Mon, Apr 20, 2009 at 10:47:46AM +0200, Dagobert Michelsen wrote:
> Yes. Ideally, opencsw.org/packages and the wikidot package page  
> information
> should be on one package, with automated content from OpenCSW and wiki
> content. Maybe on the new website? BTW, is someone currently working on
> it? I remember we had some helping hands offers :-)

Errr... not sure we are saying the same thing here :-}

but it would be nice for

http://www.opencsw.org/packages

to be able to provide links for

"click here for user level docs"
"click here for maintainer level docs"



From ihsan at opencsw.org  Tue Apr 21 08:51:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Tue, 21 Apr 2009 08:51:02 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <20090420135859.GA81648@bolthole.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
	<20090420135859.GA81648@bolthole.com>
Message-ID: <49ED6CD6.8070100@opencsw.org>

Am 20.4.2009 15:58 Uhr, Philip Brown schrieb:

>> wow ... that was unexpected.
> 
> not really. our local unix users' group speculated oracle was the next one
> in line, as soon as the IBM deal dropped out :-)

I was expecting, that Fujitsu would buy Sun.

> and it's a much better fit than IBM.

Much better.

> on the darker side, oracle now "owns" both berkeleydb and mysql.

Sun screwed up MySQL already. A few MySQL people founded already a new
company to go on with the MySQL development, so I think, it really does
not matter what Oracle is going to do with MySQL.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Tue Apr 21 14:10:41 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Tue, 21 Apr 2009 14:10:41 +0200
Subject: [csw-maintainers] /testing clamav 0.95.1
Message-ID: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com>

Please help testing ClamAV 0.95.1, especially the milter part that I
don't use myself.

http://mirror.opencsw.org/testing.html

http://mirror.opencsw.org/testing/clamav-0.95.1,REV=2009.04.21-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/clamav-0.95.1,REV=2009.04.21-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/libclamav-0.95.1,REV=2009.04.21-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libclamav-0.95.1,REV=2009.04.21-SunOS5.8-sparc-CSW.pkg.gz

-- 
/peter


From rupert at opencsw.org  Tue Apr 21 18:11:00 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 21 Apr 2009 18:11:00 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <49ED6CD6.8070100@opencsw.org>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>
	<20090420135859.GA81648@bolthole.com> <49ED6CD6.8070100@opencsw.org>
Message-ID: <6af4270904210911q5cbad510w8a9d0a3be887f10a@mail.gmail.com>

On Tue, Apr 21, 2009 at 08:51, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Am 20.4.2009 15:58 Uhr, Philip Brown schrieb:
>
>>> wow ... that was unexpected.
>>
>> not really. our local unix users' group speculated oracle was the next one
>> in line, as soon as the IBM deal dropped out :-)
>
> I was expecting, that Fujitsu would buy Sun.
>
>> and it's a much better fit than IBM.
>
> Much better.
>
>> on the darker side, oracle now "owns" both berkeleydb and mysql.
>
> Sun screwed up MySQL already. A few MySQL people founded already a new
> company to go on with the MySQL development, so I think, it really does
> not matter what Oracle is going to do with MySQL.
>
you mean http://askmonty.org/wiki/index.php/MariaDB, foundet by the
founder of mysql, michael widenius?


From ihsan at opencsw.org  Wed Apr 22 09:51:52 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 22 Apr 2009 09:51:52 +0200
Subject: [csw-maintainers] oracle buys sun ...
In-Reply-To: <6af4270904210911q5cbad510w8a9d0a3be887f10a@mail.gmail.com>
References: <6af4270904200629w637d591w7f8368ffaa7d7a1a@mail.gmail.com>	<20090420135859.GA81648@bolthole.com>
	<49ED6CD6.8070100@opencsw.org>
	<6af4270904210911q5cbad510w8a9d0a3be887f10a@mail.gmail.com>
Message-ID: <49EECC98.5090407@opencsw.org>

Am 21.4.2009 18:11 Uhr, rupert THURNER schrieb:

>>>> wow ... that was unexpected.
>>> not really. our local unix users' group speculated oracle was the next one
>>> in line, as soon as the IBM deal dropped out :-)
>> I was expecting, that Fujitsu would buy Sun.
>>
>>> and it's a much better fit than IBM.
>> Much better.
>>
>>> on the darker side, oracle now "owns" both berkeleydb and mysql.
>> Sun screwed up MySQL already. A few MySQL people founded already a new
>> company to go on with the MySQL development, so I think, it really does
>> not matter what Oracle is going to do with MySQL.
>>
> you mean http://askmonty.org/wiki/index.php/MariaDB, foundet by the
> founder of mysql, michael widenius?

Yes. I just couldn't remember the name anymore.


Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Wed Apr 22 15:19:04 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 22 Apr 2009 15:19:04 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
Message-ID: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>

I have addressed bugs 3632, 3633 and 3634 in the new 1.6 version of
cswclassutils. All bugs refer to the cswinitsmf class action script
and were reported by Yann.

0003632: Service should be disabled synchronously
http://www.opencsw.org/mantis/view.php?id=3632

0003633: Service using init scripts should not be configured to at
boot time when autoenable_daemons=no
http://www.opencsw.org/mantis/view.php?id=3633

0003634: Please add smf service state persistence across upgrade
http://www.opencsw.org/mantis/view.php?id=3634

I will of course try the changes myself but it would be nice if more
people could test it before release.

http://mirror.opencsw.org/testing.html
http://mirror.opencsw.org/testing/cswclassutils-1.6,REV=2009.04.21-SunOS5.8-all-CSW.pkg.gz

I have more bugs reported but they are more of feature requests so
they will have to hold off a while longer. One is a request for a new
CAS for creating texinfo files, I will not do that one myself but will
happily include it into the package if someone contributes.

http://www.opencsw.org/mantis/view.php?id=3485

-- 
/peter


From ihsan at opencsw.org  Wed Apr 22 16:33:04 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 22 Apr 2009 16:33:04 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
Message-ID: <49EF2AA0.3080402@opencsw.org>

Hello Peter,

Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:

> 0003634: Please add smf service state persistence across upgrade
> http://www.opencsw.org/mantis/view.php?id=3634

I've tested only that one and it works fine, but the output might be
confusing:

Installing class <cswinitsmf> ...
Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
Creating manifest ...
Configuring service in SMF ...
CSWclamav is using Service Management Facility. The FMRI is
svc:/network/cswclamav-milter:default
Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
Creating manifest ...
Configuring service in SMF ...
CSWclamav is using Service Management Facility. The FMRI is
svc:/network/cswclamd:default
Starting CSWclamav ...
[ verifying class <cswinitsmf> ]

Clamd was not started, even it said that it was started.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From cyril.plisko at mountall.com  Wed Apr 22 16:39:57 2009
From: cyril.plisko at mountall.com (Cyril Plisko)
Date: Wed, 22 Apr 2009 17:39:57 +0300
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49EF2AA0.3080402@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>
Message-ID: <c7dddeaa0904220739g2348a2a4rfabc30103a27f7d2@mail.gmail.com>

On 4/22/09, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Hello Peter,
>
> Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:
>
>> 0003634: Please add smf service state persistence across upgrade
>> http://www.opencsw.org/mantis/view.php?id=3634
>
> I've tested only that one and it works fine, but the output might be
> confusing:
>
> Installing class <cswinitsmf> ...
> Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamav-milter:default
> Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamd:default
> Starting CSWclamav ...
> [ verifying class <cswinitsmf> ]
>
> Clamd was not started, even it said that it was started.
>
>
>
>
> Ihsan
>
> --
> ihsan at dogan.ch		http://blog.dogan.ch/
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


-- 
Regards,
        Cyril


From bonivart at opencsw.org  Wed Apr 22 16:56:42 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 22 Apr 2009 16:56:42 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49EF2AA0.3080402@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com> 
	<49EF2AA0.3080402@opencsw.org>
Message-ID: <625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>

On Wed, Apr 22, 2009 at 4:33 PM, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Hello Peter,
>
> Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:
>
>> 0003634: Please add smf service state persistence across upgrade
>> http://www.opencsw.org/mantis/view.php?id=3634
>
> I've tested only that one and it works fine, but the output might be
> confusing:
>
> Installing class <cswinitsmf> ...
> Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamav-milter:default
> Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamd:default
> Starting CSWclamav ...
> [ verifying class <cswinitsmf> ]
>
> Clamd was not started, even it said that it was started.

I think this is related to a bug when multiple services are set up
like above. The placement of the start block in the script is wrong.

I will look into it. Thanks for the report.

-- 
/peter


From cyril.plisko at mountall.com  Wed Apr 22 17:05:07 2009
From: cyril.plisko at mountall.com (Cyril Plisko)
Date: Wed, 22 Apr 2009 18:05:07 +0300
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49EF2AA0.3080402@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>
Message-ID: <c7dddeaa0904220805p61efec93w1dd61e9ba63fa5aa@mail.gmail.com>

On 4/22/09, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Hello Peter,
>
> Am 22.4.2009 15:19 Uhr, Peter Bonivart schrieb:
>
>> 0003634: Please add smf service state persistence across upgrade
>> http://www.opencsw.org/mantis/view.php?id=3634
>
> I've tested only that one and it works fine, but the output might be
> confusing:
>
> Installing class <cswinitsmf> ...
> Creating service script in /var/opt/csw/svc/method/svc-cswclamav-milter ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamav-milter:default
> Creating service script in /var/opt/csw/svc/method/svc-cswclamd ...
> Creating manifest ...
> Configuring service in SMF ...
> CSWclamav is using Service Management Facility. The FMRI is
> svc:/network/cswclamd:default
> Starting CSWclamav ...
> [ verifying class <cswinitsmf> ]
>
> Clamd was not started, even it said that it was started.
>
>
>
>
> Ihsan
>
> --
> ihsan at dogan.ch		http://blog.dogan.ch/
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


-- 
Regards,
        Cyril


From ihsan at opencsw.org  Wed Apr 22 18:56:32 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 22 Apr 2009 18:56:32 +0200
Subject: [csw-maintainers] /testing clamav 0.95.1
In-Reply-To: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com>
References: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com>
Message-ID: <49EF4C40.9010308@opencsw.org>

Hello Peter,

Am 21.4.2009 14:10 Uhr, Peter Bonivart schrieb:

> Please help testing ClamAV 0.95.1, especially the milter part that I
> don't use myself.

Works fine, but I haven't tested the Milter part.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Wed Apr 22 19:51:49 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 22 Apr 2009 19:51:49 +0200
Subject: [csw-maintainers] /testing clamav 0.95.1
In-Reply-To: <49EF4C40.9010308@opencsw.org>
References: <625385e30904210510m7afb408dhe30c430d40808d62@mail.gmail.com> 
	<49EF4C40.9010308@opencsw.org>
Message-ID: <625385e30904221051t3cd465e7g33b744800d308b40@mail.gmail.com>

On Wed, Apr 22, 2009 at 6:56 PM, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Works fine, but I haven't tested the Milter part.

Ok, I will release it later this week if no one reports any problems
with it. It's just a respin with new source.

-- 
/peter


From Darin.Perusich at cognigencorp.com  Wed Apr 22 23:31:37 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 22 Apr 2009 17:31:37 -0400
Subject: [csw-maintainers] off topic question about suntar
Message-ID: <49EF8CB9.3040301@cognigencorp.com>

I have an off topic question I'm hoping one of you Solaris hackers can
help me answer. This might be right up Joerg alley since it's star guru ;-).

I'm using suntar on Solaris 10 to create archives and I've run into an
problem where when LANG is not set or set to a non-UTF-8 value like C,
POSIX, etc, then 'suntar' will throw the message "tar: invalid character
in UTF-8 conversion of 'luke?s.arf.gz'. There are a bunch of files that
have funky names, I don't know who or why, and I get the same error for
each one and they're not included in the archive.

How does suntar know whether or not the file name is UTF-8 and why
doesn't it even matter?

Thanks!

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com



From Joerg.Schilling at fokus.fraunhofer.de  Thu Apr 23 12:18:21 2009
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Thu, 23 Apr 2009 12:18:21 +0200
Subject: [csw-maintainers] off topic question about suntar
In-Reply-To: <49EF8CB9.3040301@cognigencorp.com>
References: <49EF8CB9.3040301@cognigencorp.com>
Message-ID: <49f0406d.7hLAW/yQwf7c3vnm%Joerg.Schilling@fokus.fraunhofer.de>

Darin Perusich <Darin.Perusich at cognigencorp.com> wrote:

> I have an off topic question I'm hoping one of you Solaris hackers can
> help me answer. This might be right up Joerg alley since it's star guru ;-).
>
> I'm using suntar on Solaris 10 to create archives and I've run into an
> problem where when LANG is not set or set to a non-UTF-8 value like C,
> POSIX, etc, then 'suntar' will throw the message "tar: invalid character
> in UTF-8 conversion of 'luke???s.arf.gz'. There are a bunch of files that
> have funky names, I don't know who or why, and I get the same error for
> each one and they're not included in the archive.
>
> How does suntar know whether or not the file name is UTF-8 and why
> doesn't it even matter?

Did you use the -E option to suntar?

In this case and only in this case, the file names will be converted to UTF-8.

Note that we did add the permission to also have untranslated filenames to the 
POSIX standard around 2004 in order to allow tar to be used as backup format.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily


From Darin.Perusich at cognigencorp.com  Thu Apr 23 15:05:02 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 23 Apr 2009 09:05:02 -0400
Subject: [csw-maintainers] off topic question about suntar
In-Reply-To: <49f0406d.7hLAW/yQwf7c3vnm%Joerg.Schilling@fokus.fraunhofer.de>
References: <49EF8CB9.3040301@cognigencorp.com>
	<49f0406d.7hLAW/yQwf7c3vnm%Joerg.Schilling@fokus.fraunhofer.de>
Message-ID: <49F0677E.2050603@cognigencorp.com>


Joerg Schilling wrote:
> 
> Did you use the -E option to suntar?

Yes -E is being used, here is the command line. I'm forcing it to error
by setting LANG=C.

LANG=C /usr/bin/pfexec /usr/sbin/tar -cpE at bf 256 - -C
/snapshots/client_data . > file.tar
tar: invalid character in UTF-8 conversion of './luke?s.arf.gz'
tar: file (./luke?s.arf.gz): UTF-8 conversion failed.

> In this case and only in this case, the file names will be converted to UTF-8.
> 
> Note that we did add the permission to also have untranslated filenames to the 
> POSIX standard around 2004 in order to allow tar to be used as backup format.

This appears to be limited to Solaris 10 x86. I just tried this on SPARC
system and didn't get any errors. I'll have to submit a bug report.

Thanks

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com



From mwatters at opencsw.org  Thu Apr 23 16:29:46 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 09:29:46 -0500
Subject: [csw-maintainers] php5 now in testing
Message-ID: <49F07B5A.5080608@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	Repackage due to gar bug with dynamic admin files.

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknwe1kACgkQLrhmsXMSLxeKjQCgh484pRIo1jRra/iAgyjsz0Xw
HGcAoLkCAA12RnJcUKQ+pepdVeUbZnku
=V3Vl
-----END PGP SIGNATURE-----


From bonivart at opencsw.org  Thu Apr 23 18:53:35 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 23 Apr 2009 18:53:35 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com> 
	<49EF2AA0.3080402@opencsw.org>
	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
Message-ID: <625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>

On Wed, Apr 22, 2009 at 4:56 PM, Peter Bonivart <bonivart at opencsw.org> wrote:
>> Clamd was not started, even it said that it was started.
>
> I think this is related to a bug when multiple services are set up
> like above. The placement of the start block in the script is wrong.
>
> I will look into it. Thanks for the report.

There's a new version in testing to address this problem, please give
it a try if you have the time. It should start all services a package
contains and give better info about what it's doing. It should also
properly stop services on non-SMF systems.

http://mirror.opencsw.org/testing/cswclassutils-1.8,REV=2009.04.23-SunOS5.8-all-CSW.pkg.gz

-- 
/peter


From mwatters at opencsw.org  Thu Apr 23 20:06:40 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 13:06:40 -0500
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0AD05.1070909@nature.berkeley.edu>
References: <49F07B5A.5080608@opencsw.org>
	<49F0AD05.1070909@nature.berkeley.edu>
Message-ID: <49F0AE30.5070406@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Gary Casterline wrote:
> Mike Watters wrote:
>> Changes:
>> 	Repackage due to gar bug with dynamic admin files.
>>
> 
> I don't see mod_php5 or ap2_modphp5 -- will those come later?
> 
>  _Gary
> 
> 
> _______________________________________________
> users mailing list
> users at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/users
> 
> _______________________________________________
> users mailing list
> users at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/users

Gary,	
	I don't maintain those  ;)

Ihsan: do you have any plans on pushing out a new version of those?

I tested running php5 with the existing ap2_modphp5 ... with no issues.
I don't run 1.3x anymore I didn't test against mod_php5

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknwrjAACgkQLrhmsXMSLxdu3ACgk+U7zifhcDr2Qcu5EeqrVlU/
FDUAn0Fv6j5zyrfzXTPt7VBRcsIWYUOD
=Dsz7
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 23 22:35:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 23 Apr 2009 22:35:02 +0200
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0AE30.5070406@opencsw.org>
References: <49F07B5A.5080608@opencsw.org>	<49F0AD05.1070909@nature.berkeley.edu>
	<49F0AE30.5070406@opencsw.org>
Message-ID: <49F0D0F6.6050003@opencsw.org>

Am 23.4.2009 20:06 Uhr, Mike Watters schrieb:

> Gary,	
> 	I don't maintain those  ;)
> 
> Ihsan: do you have any plans on pushing out a new version of those?
> 
> I tested running php5 with the existing ap2_modphp5 ... with no issues.
> I don't run 1.3x anymore I didn't test against mod_php5

I thought, you took these as well.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Thu Apr 23 22:36:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 15:36:49 -0500
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0D0F6.6050003@opencsw.org>
References: <49F07B5A.5080608@opencsw.org>	<49F0AD05.1070909@nature.berkeley.edu>	<49F0AE30.5070406@opencsw.org>
	<49F0D0F6.6050003@opencsw.org>
Message-ID: <49F0D161.5090507@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ihsan Dogan wrote:
> Am 23.4.2009 20:06 Uhr, Mike Watters schrieb:
> 
>> Gary,	
>> 	I don't maintain those  ;)
>>
>> Ihsan: do you have any plans on pushing out a new version of those?
>>
>> I tested running php5 with the existing ap2_modphp5 ... with no issues.
>> I don't run 1.3x anymore I didn't test against mod_php5
> 
> I thought, you took these as well.
> 
> 
> 
> Ihsan
> 

I can not a problem. I have been scatter brain'd lately
I didn't remember if I took them or not.

	I will re-create them in a bit.


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknw0WAACgkQLrhmsXMSLxenegCgiHMDlo5/tsVSsBn5fT1wSn23
VGYAnRutS/lQQ7Edewt9SBrKVMdzQavJ
=Uo12
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 23 22:57:45 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 23 Apr 2009 22:57:45 +0200
Subject: [csw-maintainers] [csw-users] php5 now in testing
In-Reply-To: <49F0D161.5090507@opencsw.org>
References: <49F07B5A.5080608@opencsw.org>	<49F0AD05.1070909@nature.berkeley.edu>	<49F0AE30.5070406@opencsw.org>	<49F0D0F6.6050003@opencsw.org>
	<49F0D161.5090507@opencsw.org>
Message-ID: <49F0D649.2060004@opencsw.org>

Am 23.4.2009 22:36 Uhr, Mike Watters schrieb:

>>> Gary,	
>>> 	I don't maintain those  ;)
>>>
>>> Ihsan: do you have any plans on pushing out a new version of those?
>>>
>>> I tested running php5 with the existing ap2_modphp5 ... with no issues.
>>> I don't run 1.3x anymore I didn't test against mod_php5
>> I thought, you took these as well.
> 
> I can not a problem. I have been scatter brain'd lately
> I didn't remember if I took them or not.
> 
> 	I will re-create them in a bit.

Thanks.

I think it makes sense, when all PHP packages are maintained by a single
person.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From mwatters at opencsw.org  Fri Apr 24 06:40:38 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Apr 2009 23:40:38 -0500
Subject: [csw-maintainers] php5 now in testing
Message-ID: <49F142C6.1060106@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


Changes:
	rebuilt to include the 2.x and 1.3x apache modules in main php5 recipe.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknxQsYACgkQLrhmsXMSLxcNlwCfboUUr4DSznoNDxXMdKte+nnW
vhsAnAljY7T4q8kvsm5wONCOqZwXED94
=npFG
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Sat Apr 25 17:19:49 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sat, 25 Apr 2009 17:19:49 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
Message-ID: <49F32A15.8040901@opencsw.org>

>>> Clamd was not started, even it said that it was started.
>> I think this is related to a bug when multiple services are set up
>> like above. The placement of the start block in the script is wrong.
>>
>> I will look into it. Thanks for the report.
> 
> There's a new version in testing to address this problem, please give
> it a try if you have the time. It should start all services a package
> contains and give better info about what it's doing. It should also
> properly stop services on non-SMF systems.
> 
> http://mirror.opencsw.org/testing/cswclassutils-1.8,REV=2009.04.23-SunOS5.8-all-CSW.pkg.gz

Thanks. I will test it on Sunday.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Sat Apr 25 17:29:19 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Sat, 25 Apr 2009 17:29:19 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <49F32A15.8040901@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com> 
	<49EF2AA0.3080402@opencsw.org>
	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com> 
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com> 
	<49F32A15.8040901@opencsw.org>
Message-ID: <625385e30904250829o619cb81ek3aa676b151075fc5@mail.gmail.com>

On Sat, Apr 25, 2009 at 5:19 PM, Ihsan Dogan <ihsan at opencsw.org> wrote:
> Thanks. I will test it on Sunday.

Note that it's now:

http://mirror.opencsw.org/testing/cswclassutils-1.9,REV=2009.04.24-SunOS5.8-all-CSW.pkg.gz

It has a few more bug fixes and I had to comment out the state
persistent code. The last few days I've been working on making
cswinitsmf able to handle multiple services per package and even
though the state persistent feature is great during upgrades it does
not support that, I have notified Yann and he's looking at it so it
will come back soon.

-- 
/peter


From mwatters at opencsw.org  Sun Apr 26 03:10:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Apr 2009 20:10:33 -0500
Subject: [csw-maintainers] roundcube webmail?
Message-ID: <49F3B489.5000604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I tried to send an email to maintainers using roundcube mail.opencsw.org.

I got a bounce saying...

"Your mail to 'maintainers' with the subject

    Update on Subversion

Is being held until the list moderator can review it for approval.

The reason it is being held:

    Post by non-member to a members-only list"
...

is this normal?


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknztIkACgkQLrhmsXMSLxcpQACfQVbMdQ3Fckah3qVAx2SKMsK8
ylYAn1Ln6VSOzbTVLNg3Bu91Un9iWCV5
=pf6u
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sun Apr 26 05:34:53 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Apr 2009 22:34:53 -0500
Subject: [csw-maintainers] RFE mGar
Message-ID: <49F3D65D.4070907@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

if possible, could we get an option to print what is "UNCOMMITTED"

I keep finding myself searching for files that need committed,
deleted, or needing added to ignore.

it would be nice if we had a quick way to find them.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAknz1l0ACgkQLrhmsXMSLxfSlQCglF5ls2ibR4TT6L3BgOti4rWI
8YsAnRuGY9L4MjMousTne8oeS/9AdwAc
=w9Oi
-----END PGP SIGNATURE-----


From dam at opencsw.org  Sun Apr 26 09:22:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Apr 2009 09:22:08 +0200
Subject: [csw-maintainers] RFE mGar
In-Reply-To: <49F3D65D.4070907@opencsw.org>
References: <49F3D65D.4070907@opencsw.org>
Message-ID: <CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>

Hi Mike,

Am 26.04.2009 um 05:34 schrieb Mike Watters <mwatters at opencsw.org>:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> if possible, could we get an option to print what is "UNCOMMITTED"
>
> I keep finding myself searching for files that need committed,
> deleted, or needing added to ignore.
>
> it would be nice if we had a quick way to find them.

If you do a 'svn status' and see more than the external gar reference  
the files shown are the ones to be either committed or ignored. We may  
have a target for that, by I guess it is rather straight forward.

Best regards

   -- Dago



From mwatters at localhost.opencsw.org  Sun Apr 26 01:42:29 2009
From: mwatters at localhost.opencsw.org (mwatters)
Date: Sun, 26 Apr 2009 01:42:29 +0200
Subject: [csw-maintainers] Update on Subversion
In-Reply-To: <49F32A15.8040901@opencsw.org>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
	<49F32A15.8040901@opencsw.org>
Message-ID: <fc8682fa9ed37a32432e01c976cf92ab@localhost>


I have the issues just about worked out with Subversion.
I also figured out what changed between the "good compile" and 
now.   neon was updated to 0.28.4 from 0.28.3

this is not an issue, but I found the subversion compile
fails when I use --with-neon=$(prefix)

when I remove the line and let the configure "find" it it works
beautifully.  I will check the bug list and file one if it don't exist.

version 1.6.1 of subversion should be in testing shortly.


-- Mike


From trygvis at opencsw.org  Sun Apr 26 14:02:01 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sun, 26 Apr 2009 14:02:01 +0200
Subject: [csw-maintainers] RFE mGar
In-Reply-To: <CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
References: <49F3D65D.4070907@opencsw.org>
	<CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
Message-ID: <49F44D39.40109@opencsw.org>

Dagobert Michelsen wrote:
> Hi Mike,
> 
> Am 26.04.2009 um 05:34 schrieb Mike Watters <mwatters at opencsw.org>:
> 
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> if possible, could we get an option to print what is "UNCOMMITTED"
>>
>> I keep finding myself searching for files that need committed,
>> deleted, or needing added to ignore.
>>
>> it would be nice if we had a quick way to find them.
> 
> If you do a 'svn status' and see more than the external gar reference 
> the files shown are the ones to be either committed or ignored. We may 
> have a target for that, by I guess it is rather straight forward.

I started once on a set of "scm-*" goals as a facade for the operations 
a OpenCSW maintainer is supposed to use, it might be nice to add 
"scm-status" when you're in a package for those not used to SCMs or 
Subversion in particular.

--
Trygve


From bwalton at opencsw.org  Sun Apr 26 14:09:34 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Apr 2009 08:09:34 -0400
Subject: [csw-maintainers] RFE mGar
In-Reply-To: <CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
References: <49F3D65D.4070907@opencsw.org>
	<CFFF2AAD-3E52-42BA-8D10-0DAED9BE8D30@opencsw.org>
Message-ID: <1240747607-sup-863@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Apr 26 03:22:08 -0400 2009:

Hi Mike,

> If you do a 'svn status' and see more than the external gar reference  
> the files shown are the ones to be either committed or ignored. We may  
> have a target for that, by I guess it is rather straight forward.

I'll point out specifically that UNCOMMITTED will occur when the
output of `svn status` contains files marked as ?.  Eg, files not
added to svn will trigger this state.  While it might be nice to
allow untracked files to be excepted from this, it's difficult to
properly prune out things like: Makefile is committed, but the new
.postinstall file is still unknown, etc.

HTH.
-Ben

-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090426/811fa9da/attachment-0001.asc>

From ihsan at opencsw.org  Sun Apr 26 23:34:35 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 26 Apr 2009 23:34:35 +0200
Subject: [csw-maintainers] /testing cswclassutils 1.6
In-Reply-To: <625385e30904250829o619cb81ek3aa676b151075fc5@mail.gmail.com>
References: <625385e30904220619o6a1a67fdgb4a6c6e965397aff@mail.gmail.com>
	<49EF2AA0.3080402@opencsw.org>	<625385e30904220756n50be9fb5jcbbc9c0c05e7d25b@mail.gmail.com>
	<625385e30904230953l53339a49gd49a8e875116c8e9@mail.gmail.com>
	<49F32A15.8040901@opencsw.org>
	<625385e30904250829o619cb81ek3aa676b151075fc5@mail.gmail.com>
Message-ID: <49F4D36B.3010101@opencsw.org>

Hello Peter,

Am 25.4.2009 17:29 Uhr, Peter Bonivart schrieb:

>> Thanks. I will test it on Sunday.
> 
> Note that it's now:
> 
> http://mirror.opencsw.org/testing/cswclassutils-1.9,REV=2009.04.24-SunOS5.8-all-CSW.pkg.gz
> 
> It has a few more bug fixes and I had to comment out the state
> persistent code. The last few days I've been working on making
> cswinitsmf able to handle multiple services per package and even
> though the state persistent feature is great during upgrades it does
> not support that, I have notified Yann and he's looking at it so it
> will come back soon.

I just installed 1.9 on my testing system and after deinstallation and
reinstallation of NSD, it started the daemon, even the SMF service was
disabled.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From ihsan at opencsw.org  Sun Apr 26 23:37:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 26 Apr 2009 23:37:02 +0200
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F3B489.5000604@opencsw.org>
References: <49F3B489.5000604@opencsw.org>
Message-ID: <49F4D3FE.7060705@opencsw.org>

Hello Mike,

Am 26.4.2009 3:10 Uhr, Mike Watters schrieb:

> I tried to send an email to maintainers using roundcube mail.opencsw.org.

[...]

Ouuhmm, looks like there is a configuration mistake. The sender address
is wrong. I'll have a look if I can set a predefined domain.

You can fix it with setting your sender address to xyz at opencsw.org.

BTW: Does anybody know a good webmail client? I'm not really satisfied
with Roundcube.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From rupert at opencsw.org  Mon Apr 27 01:24:24 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 27 Apr 2009 01:24:24 +0200
Subject: [csw-maintainers] setuptools does not find cc compiler?
Message-ID: <6af4270904261624w8af7e6en736851dd87ee6394@mail.gmail.com>

today i had time to test edgewall trac. python-2.6.1, svn-1.5.6, trac,
pysvn, sqlite3, apache worked fine.

the only thing i noticed was with setuptools while trying to do a
native genshi install:

Downloading http://ftp.edgewall.com/pub/genshi/Genshi-0.5.1.zip
Processing Genshi-0.5.1.zip
Running Genshi-0.5.1/setup.py -q bdist_egg --dist-dir
/tmp/easy_install-wiThg9/Genshi-0.5.1/egg-dist-tmp-5GkKb1
warning: no previously-included files found matching 'doc/2000ft.graffle'
warning: no previously-included files matching '*' found under
directory 'doc/logo.lineform'
unable to execute /opt/studio/SOS11/SUNWspro/bin/cc: No such file or directory
**********************************************************************
WARNING:
An optional C extension could not be compiled, speedups will not be
available.
**********************************************************************
Adding Genshi 0.5.1 to easy-install.pth file

we have cc in another location, and on the path. also
"CC=/opt/SUNWspro/bin/cc easy_install genshi" resulted in the same
error message.

is this something for upstream?

rupert.


From rupert at opencsw.org  Mon Apr 27 01:25:40 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Mon, 27 Apr 2009 01:25:40 +0200
Subject: [csw-maintainers] Installation of <CSWosshclient> partially failed.
Message-ID: <6af4270904261625w33fb63cfy7e0eb842c0550c3@mail.gmail.com>

while trying to install git, i got the following error (but had no
time to test if there is any impact).


Installing openssh_client - OpenSSH Secure Shell as <CSWosshclient>

## Installing part 1 of 1.
/opt/csw/bin/scp
/opt/csw/bin/sftp
/opt/csw/bin/slogin <symbolic link>
/opt/csw/bin/ssh
/opt/csw/bin/ssh-add
/opt/csw/bin/ssh-agent
/opt/csw/bin/ssh-keygen
/opt/csw/bin/ssh-keyscan
/opt/csw/libexec/ssh-keysign
/opt/csw/share/doc/openssh_client/CREDITS
/opt/csw/share/doc/openssh_client/ChangeLog
/opt/csw/share/doc/openssh_client/ChangeLog.gssapi
/opt/csw/share/doc/openssh_client/INSTALL
/opt/csw/share/doc/openssh_client/LICENCE
/opt/csw/share/doc/openssh_client/OVERVIEW
/opt/csw/share/doc/openssh_client/README
/opt/csw/share/doc/openssh_client/README.dns
/opt/csw/share/doc/openssh_client/README.platform
/opt/csw/share/doc/openssh_client/README.privsep
/opt/csw/share/doc/openssh_client/README.smartcard
/opt/csw/share/doc/openssh_client/README.tun
/opt/csw/share/doc/openssh_client/TODO
/opt/csw/share/doc/openssh_client/WARNING.RNG
/opt/csw/share/doc/openssh_client/changelog.CSW
/opt/csw/share/man/man1/scp.1
/opt/csw/share/man/man1/sftp.1
/opt/csw/share/man/man1/slogin.1 <symbolic link>
/opt/csw/share/man/man1/ssh-add.1
/opt/csw/share/man/man1/ssh-agent.1
/opt/csw/share/man/man1/ssh-keygen.1
/opt/csw/share/man/man1/ssh-keyscan.1
/opt/csw/share/man/man1/ssh.1
/opt/csw/share/man/man5/ssh_config.5
/opt/csw/share/man/man8/ssh-keysign.8
[ verifying class <none> ]
[ verifying class <conf> ]
ERROR: attribute verification of </opt/csw/etc/ssh/ssh_config> failed
   pathname does not exist

Installation of <CSWosshclient> partially failed.


From mwatters at opencsw.org  Mon Apr 27 02:05:37 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 26 Apr 2009 19:05:37 -0500
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F4D3FE.7060705@opencsw.org>
References: <49F3B489.5000604@opencsw.org> <49F4D3FE.7060705@opencsw.org>
Message-ID: <49F4F6D1.6050900@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ihsan Dogan wrote:
> Hello Mike,
> 
> Am 26.4.2009 3:10 Uhr, Mike Watters schrieb:
> 
>> I tried to send an email to maintainers using roundcube mail.opencsw.org.
> 
> [...]
> 
> Ouuhmm, looks like there is a configuration mistake. The sender address
> is wrong. I'll have a look if I can set a predefined domain.
> 
> You can fix it with setting your sender address to xyz at opencsw.org.
> 
> BTW: Does anybody know a good webmail client? I'm not really satisfied
> with Roundcube.
> 
> 
> 
> 
> Ihsan
> 
I have always had good luck with squirrel mail
http://squirrelmail.org/

I will look at packaging it up.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn09tEACgkQLrhmsXMSLxfZ7QCgqB6c1t8gFp5B3khV5gmxlbrT
rLUAni5XZRhfXs8q3Xk6ir2YYzJ/CQ7B
=PEr4
-----END PGP SIGNATURE-----


From bwalton at opencsw.org  Mon Apr 27 02:14:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Apr 2009 20:14:01 -0400
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F4F6D1.6050900@opencsw.org>
References: <49F3B489.5000604@opencsw.org> <49F4D3FE.7060705@opencsw.org>
	<49F4F6D1.6050900@opencsw.org>
Message-ID: <1240791137-sup-4640@ntdws12.chass.utoronto.ca>

Excerpts from Mike Watters's message of Sun Apr 26 20:05:37 -0400 2009:
> I have always had good luck with squirrel mail
> http://squirrelmail.org/

We use squirrelmail in a few different configs here.  It works, but
I'd never call it 'good.'

Is having Google host the mail through their '...for your domain'
service an option for csw?

> I will look at packaging it up.

Cool.

Thanks
-Ben
-- 
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302

GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu
Contact me to arrange for a CAcert assurance meeting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090426/87d88c69/attachment-0001.asc>

From phil at bolthole.com  Mon Apr 27 04:19:15 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Apr 2009 19:19:15 -0700
Subject: [csw-maintainers] roundcube webmail?
In-Reply-To: <49F4D3FE.7060705@opencsw.org>
References: <49F3B489.5000604@opencsw.org> <49F4D3FE.7060705@opencsw.org>
Message-ID: <20090427021915.GB8187@bolthole.com>

On Sun, Apr 26, 2009 at 11:37:02PM +0200, Ihsan Dogan wrote:
> 
> BTW: Does anybody know a good webmail client? I'm not really satisfied
> with Roundcube.

prayer is pretty good.

no, there really is an IMAP webmap client called "prayer" :-)

www-uxsup.csx.cam.ac.uk/~dpc22/prayer/

it's fast, because it is native C.

"Written entirely in C as HTTP to IMAP Gateway. No scripting languages. "

no perl. no php. NO APACHE!




From mwatters at opencsw.org  Mon Apr 27 05:15:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 26 Apr 2009 22:15:49 -0500
Subject: [csw-maintainers] Subversion 1.6.1 now in testing
Message-ID: <49F52365.1040700@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Bindings included: Perl, Python, Ruby, JavaHL

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn1I2UACgkQLrhmsXMSLxdFjwCfZaZBR5bdFcYr0CJkvb/mbmJF
l3oAoKaIStZsPgqWakBPb3xOdbziuLrc
=IgvO
-----END PGP SIGNATURE-----


From dam at opencsw.org  Mon Apr 27 15:03:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 27 Apr 2009 15:03:01 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <20081107133324.GA2059@zerfleddert.de>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>
	<49137326.3000605@wbonnet.net>
	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>
	<20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
Message-ID: <A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>

Hi Michael,

Am 07.11.2008 um 14:33 schrieb Michael Gernoth:
> My old build-scripts and sources for all my packages can be found at:
> http://csw.informatik.uni-erlangen.de/csw/users/michael/sources/
> The scripts are in the 'specs' subdirectory.

I copied your old build descriptions to the GAR repository in the
respective pkg/<pkg>/trunk/legacy/ folders as a reference if
someone wants to bring a package in GAR.


Best regards

   -- Dago


From yann at pleiades.fr.eu.org  Mon Apr 27 23:10:28 2009
From: yann at pleiades.fr.eu.org (Yann Rouillard)
Date: Mon, 27 Apr 2009 23:10:28 +0200
Subject: [csw-maintainers] Installation of <CSWosshclient> partially
	failed.
In-Reply-To: <6af4270904261625w33fb63cfy7e0eb842c0550c3@mail.gmail.com>
References: <6af4270904261625w33fb63cfy7e0eb842c0550c3@mail.gmail.com>
Message-ID: <49F61F44.3040509@pleiades.fr.eu.org>

Hi,

Could you please open a bug for this problem in mantis 
http://www.opencsw.org/mantis/view.php?id=2652
?

Could you also give me the verbose output of the installation ?
You need to manually download the package and install it using pkgadd 
with the -v option to get the verbose output.

Thanks in advance,

Yann

Le 27.04.2009 01:25, rupert THURNER a ?crit :
> while trying to install git, i got the following error (but had no
> time to test if there is any impact).
>
>
> Installing openssh_client - OpenSSH Secure Shell as<CSWosshclient>
>
> ## Installing part 1 of 1.
> /opt/csw/bin/scp
> /opt/csw/bin/sftp
> /opt/csw/bin/slogin<symbolic link>
> /opt/csw/bin/ssh
> /opt/csw/bin/ssh-add
> /opt/csw/bin/ssh-agent
> /opt/csw/bin/ssh-keygen
> /opt/csw/bin/ssh-keyscan
> /opt/csw/libexec/ssh-keysign
> /opt/csw/share/doc/openssh_client/CREDITS
> /opt/csw/share/doc/openssh_client/ChangeLog
> /opt/csw/share/doc/openssh_client/ChangeLog.gssapi
> /opt/csw/share/doc/openssh_client/INSTALL
> /opt/csw/share/doc/openssh_client/LICENCE
> /opt/csw/share/doc/openssh_client/OVERVIEW
> /opt/csw/share/doc/openssh_client/README
> /opt/csw/share/doc/openssh_client/README.dns
> /opt/csw/share/doc/openssh_client/README.platform
> /opt/csw/share/doc/openssh_client/README.privsep
> /opt/csw/share/doc/openssh_client/README.smartcard
> /opt/csw/share/doc/openssh_client/README.tun
> /opt/csw/share/doc/openssh_client/TODO
> /opt/csw/share/doc/openssh_client/WARNING.RNG
> /opt/csw/share/doc/openssh_client/changelog.CSW
> /opt/csw/share/man/man1/scp.1
> /opt/csw/share/man/man1/sftp.1
> /opt/csw/share/man/man1/slogin.1<symbolic link>
> /opt/csw/share/man/man1/ssh-add.1
> /opt/csw/share/man/man1/ssh-agent.1
> /opt/csw/share/man/man1/ssh-keygen.1
> /opt/csw/share/man/man1/ssh-keyscan.1
> /opt/csw/share/man/man1/ssh.1
> /opt/csw/share/man/man5/ssh_config.5
> /opt/csw/share/man/man8/ssh-keysign.8
> [ verifying class<none>  ]
> [ verifying class<conf>  ]
> ERROR: attribute verification of</opt/csw/etc/ssh/ssh_config>  failed
>     pathname does not exist
>
> Installation of<CSWosshclient>  partially failed.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers



From dam at opencsw.org  Tue Apr 28 20:06:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Apr 2009 20:06:55 +0200
Subject: [csw-maintainers] Problem on CSWphp5zip
Message-ID: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>

Hi,

I got a problem on lots of PHP 5 packages like CSWphp5zip and  
CSWphp5session:

> /var/sadm/pkg/CSWphp5zip/install/postinstall: /home/mwatters/mgar/ 
> pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/php.ini^Jecho  
> "[===> Running Post Install <===]"^Jecho "   ===> Enabling zip  
> extension"^Jif grep CSW: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall:  $ 
> {PHP_INI}^Jelse^Jperl -i -pe s: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall: .*CSW: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall: CSW: not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall:  ${PHP_INI}^Jfi^Jcat  
> <<  
> _EOF_ 
> ^ 
> J 
> ******************************************************************************^ 
> J* NOTICE: Successfully Enabled Extension "zip"^J* in $ 
> {PHP_INI}^J*^J* You will need to restart your web server to finish  
> the  
> install 
> ^ 
> J 
> ******************************************************************************^ 
> J_EOF_^Jexit 0^Jgmake[4]: Leaving directory : not found
> /var/sadm/pkg/CSWphp5zip/install/postinstall: gmake[4]:: not found
> pkgadd: ERROR: postinstall script did not complete successfully
>
> Installation of <CSWphp5zip> failed.

...

> Removal of <CSWphp5session> was successful.
> Analysing special files...
> /var/sadm/pkg/CSWphp5session/install/postinstall: /home/mwatters/ 
> mgar/pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/ 
> php.ini^Jecho "[===> Running Post Install <===]"^Jecho "   ===>  
> Enabling session extension"^Jif grep CSW: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall:  $ 
> {PHP_INI}^Jelse^Jperl -i -pe s: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall: .*CSW: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall: CSW: not found
> /var/sadm/pkg/CSWphp5session/install/postinstall:  $ 
> {PHP_INI}^Jfi^Jcat <<  
> _EOF_ 
> ^ 
> J 
> ******************************************************************************^ 
> J* NOTICE: Successfully Enabled Extension "session"^J* in $ 
> {PHP_INI}^J*^J* You will need to restart your web server to finish  
> the  
> install 
> ^ 
> J 
> ******************************************************************************^ 
> J_EOF_^Jexit 0^Jgmake[4]: Leaving directory : not found
> /var/sadm/pkg/CSWphp5session/install/postinstall: gmake[4]:: not found
> pkgadd: ERROR: postinstall script did not complete successfully


Best regards

-- Dago



From dam at opencsw.org  Tue Apr 28 20:07:18 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Apr 2009 20:07:18 +0200
Subject: [csw-maintainers] Updating all buildservers to current/
Message-ID: <52882627-7DE1-4B38-8709-9C2B3E7EF9B6@opencsw.org>

Hi,

I am updating all buildservers now to current. Please stand by.


Best regards

-- Dago


From mwatters at opencsw.org  Tue Apr 28 20:33:30 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Apr 2009 13:33:30 -0500
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
Message-ID: <49F74BFA.5040301@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dagobert Michelsen wrote:
> Hi,
> 
> I got a problem on lots of PHP 5 packages like CSWphp5zip and
> CSWphp5session:
> 
>> /var/sadm/pkg/CSWphp5zip/install/postinstall:
>> /home/mwatters/mgar/pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/php.ini^Jecho
>> "[===> Running Post Install <===]"^Jecho "   ===> Enabling zip
>> extension"^Jif grep CSW: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall:  ${PHP_INI}^Jelse^Jperl
>> -i -pe s: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall: .*CSW: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall: CSW: not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall:  ${PHP_INI}^Jfi^Jcat <<
>> _EOF_^J******************************************************************************^J*
>> NOTICE: Successfully Enabled Extension "zip"^J* in ${PHP_INI}^J*^J*
>> You will need to restart your web server to finish the
>> install^J******************************************************************************^J_EOF_^Jexit
>> 0^Jgmake[4]: Leaving directory : not found
>> /var/sadm/pkg/CSWphp5zip/install/postinstall: gmake[4]:: not found
>> pkgadd: ERROR: postinstall script did not complete successfully
>>
>> Installation of <CSWphp5zip> failed.
> 
> ...
> 
>> Removal of <CSWphp5session> was successful.
>> Analysing special files...
>> /var/sadm/pkg/CSWphp5session/install/postinstall:
>> /home/mwatters/mgar/pkg/php5/trunk^J#!/bin/sh^JPHP_INI=/opt/csw/php5/lib/php.ini^Jecho
>> "[===> Running Post Install <===]"^Jecho "   ===> Enabling session
>> extension"^Jif grep CSW: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: 
>> ${PHP_INI}^Jelse^Jperl -i -pe s: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: .*CSW: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: CSW: not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall:  ${PHP_INI}^Jfi^Jcat
>> <<
>> _EOF_^J******************************************************************************^J*
>> NOTICE: Successfully Enabled Extension "session"^J* in
>> ${PHP_INI}^J*^J* You will need to restart your web server to finish
>> the
>> install^J******************************************************************************^J_EOF_^Jexit
>> 0^Jgmake[4]: Leaving directory : not found
>> /var/sadm/pkg/CSWphp5session/install/postinstall: gmake[4]:: not found
>> pkgadd: ERROR: postinstall script did not complete successfully
> 
> 
> Best regards
> 
> -- Dago
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Dago,
This was the bug that I re-packaged to fix.  can you verify for me what REV you
 attempted to install?

when I do a pkg-get -d php5_zip I get the "broken" one.  REV=2009.04.18

The "Fixed" one is: REV=2009.04.27

Phil: have you finished processing and pushed out all of the php5 packages yet?

- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkn3S/oACgkQLrhmsXMSLxcEEwCfRtfquvRJfojsfTNK+B33ZZQL
VigAnj+z/86Rm0bwWvxPr6js8Lh6Svr0
=3mBe
-----END PGP SIGNATURE-----


From dam at opencsw.org  Tue Apr 28 21:00:13 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Apr 2009 21:00:13 +0200
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <49F74BFA.5040301@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
Message-ID: <3E598EA3-9E4C-45DB-8B96-13D2CC4F6DC1@opencsw.org>

Hi Mike,

Am 28.04.2009 um 20:33 schrieb Mike Watters:
> Dago,
> This was the bug that I re-packaged to fix.  can you verify for me  
> what REV you
> attempted to install?
>
> when I do a pkg-get -d php5_zip I get the "broken" one.   
> REV=2009.04.18

Yep, old one:

build8s% pkginfo -l CSWphp5zip
    PKGINST:  CSWphp5zip
       NAME:  php5_zip - zip Extention for PHP5
   CATEGORY:  application
       ARCH:  sparc
    VERSION:  5.2.9,REV=2009.04.18
     VENDOR:  http://www.php.net/downloads.php packaged for CSW by  
Mike Watters
     PSTAMP:  mwatters at build8s-20090418142017
   INSTDATE:  Apr 28 2009 19:59
    HOTLINE:  http://www.opencsw.org/bugtrack/
      EMAIL:  mwatters at opencsw.org
     STATUS:  partially installed
      FILES:        6 installed pathnames
                    5 shared pathnames
                    5 directories
                    1 executables
                  967 blocks used (approx)

> The "Fixed" one is: REV=2009.04.27
>
> Phil: have you finished processing and pushed out all of the php5  
> packages yet?

Immediately before updating I synced up to rsync.opencsw.org.

Uni Erlangen is still old, too:
   <http://ftp.uni-erlangen.de/pub/mirrors/csw/current/sparc/5.8/>


Best regards

   -- Dago


From phil at bolthole.com  Tue Apr 28 21:15:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Apr 2009 12:15:31 -0700
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <49F74BFA.5040301@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
Message-ID: <20090428191531.GC97557@bolthole.com>

On Tue, Apr 28, 2009 at 01:33:30PM -0500, Mike Watters wrote:
> Phil: have you finished processing and pushed out all of the php5 packages yet?

I have. yesterday.
So, Dago probably just hit mirror lag.


From mwatters at opencsw.org  Tue Apr 28 23:39:18 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Apr 2009 16:39:18 -0500
Subject: [csw-maintainers] New Subversion 1.6.1 in testing
Message-ID: <49F77786.6080308@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Changes:
	Added

share/doc/subversion/tools/*
share/doc/subversion/contrib/*
bin/svn-tools/*
bin/svn-contrib/*


- --

Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkn3d4YACgkQLrhmsXMSLxfvCQCZAdiUMnpsolP8gyq5HQ9bp2j/
bQ4AoK+vq+SSuQhcWDcRBC5KIsmLSaEt
=kMjk
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Wed Apr 29 02:35:25 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Apr 2009 19:35:25 -0500
Subject: [csw-maintainers] solaris 8 on CentOS?
Message-ID: <49F7A0CD.2090405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I remember someone saying they had solaris 8
running on CentOS, sadly, I don't remember whom.

if you still have it, can you post your config.

- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn3oMwACgkQLrhmsXMSLxc7ygCfYp5iUDJqIiWgVGILDk60W2U9
HFwAn3goksz38l5HJJJvjvARzxHdmpxh
=vEZM
-----END PGP SIGNATURE-----


From dam at opencsw.org  Wed Apr 29 09:13:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 09:13:43 +0200
Subject: [csw-maintainers] Fwd: [bug-notifications] [mutt 0003648]: mutt
	does not work with screen's altscreen because it is compiled
	with slang instead of ncurses
References: <1b25a27cdbd0df04a3965beb5c31788e@www.opencsw.org>
Message-ID: <C097C6F7-20AE-4607-8F12-FB8CAC1D6BC1@opencsw.org>

Hi,

I picked up this discussion between ncurses and slang and I guess
it is a good idea to discuss the ncurses/slang matter publicly
on the list.


Best regards

   -- Dago

Anfang der weitergeleiteten E-Mail:

> Von: Mantis Bug Tracker <noreply at opencsw.org>
> Datum: 29. April 2009 07:59:07 MESZ
> An: bug-notifications at lists.opencsw.org
> Betreff: [bug-notifications] [mutt 0003648]: mutt does not work with  
> screen's altscreen because it is compiled with slang instead of  
> ncurses
> Antwort an: users at lists.opencsw.org
>
>
> The following issue has been SUBMITTED.
> ======================================================================
> http://www.opencsw.org/bugtrack/view.php?id=3648
> ======================================================================
> Reported By:                meunier
> Assigned To:
> ======================================================================
> Project:                    mutt
> Issue ID:                   3648
> Category:                   regular use
> Reproducibility:            always
> Severity:                   block
> Priority:                   normal
> Status:                     new
> ======================================================================
> Date Submitted:             2009-04-29 07:59 CEST
> Last Modified:              2009-04-29 07:59 CEST
> ======================================================================
> Summary:                    mutt does not work with screen's  
> altscreen because
> it is compiled with slang instead of ncurses
> Description:
> /opt/csw/bin/mutt is compiled with slang, not ncurses, while
> /opt/csw/bin/screen is compiled with ncurses, not slang.  So if you  
> use
> /opt/csw/bin/screen and start /opt/csw/bin/mutt inside it, it looks  
> like
> the ncurses library used by screen and the slang library used by  
> mutt fight
> each other in a bad way when screen's altscreen feature is enabled.
>
> Here is a way to reproduce the problem:
>
> [after ssh-ing into a Solaris machine from an xterm]
> $ export $TERMINFO=/opt/csw/share/terminfo/
> $ /opt/csw/bin/infocmp
> #       Reconstructed via infocmp from file:
> /opt/csw/share/terminfo/x/xterm
> [blablabla ... so the correct terminfo database is being used]
> $ ls
> [blablabla]
> $ /opt/csw/bin/emacs
> [exit emacs, the screen returns to its previous content, including  
> showing
> the output of the previous 'ls' command.  Note: this will not work  
> and the
> ouput of the previous 'ls' command will be invisible if your TERMINFO
> environment variable is not set correctly]
> $ ls
> [blablabla]
> $ /opt/csw/bin/mutt
> [exit mutt, the screen returns to its previous content, including  
> showing
> the output of the previous 'ls' command]
> $ /opt/csw/bin/screen
> [screen is cleared]
> $ ls
> [blablabla]
> $ /opt/csw/bin/emacs
> [exit from emacs, the cursor is at the bottom of the xterm and the  
> output
> of the previous 'ls' is not visible anymore]
> $ ls
> [blablabla]
> $ /opt/csw/bin/mutt
> [exit from mutt, the cursor is at the bottom of the xterm and the  
> output
> of the previous 'ls' is not visible anymore]
>
> So far so good.  Emacs and mutt normally use xterm's "alternate  
> screen"
> feature which is why the output of the previous 'ls' command is  
> visible in
> the xterm once emacs or mutt has exited.  Screen, on the other hand,  
> does
> not provide an alternate screen by default, so emacs and mutt just  
> use the
> "regular screen" and the output of the previous 'ls' command is then  
> lost
> when emacs or mutt exits.
>
> Now type:
>
> Control-A :
>
> to get the interactive prompt from 'screen', then type:
>
> altscreen on
>
> then you should get a 'Will do alternate screen switching' from  
> 'screen'.
> This tells 'screen' that it should provide an alternate screen to
> applications like emacs or mutt that normally use xterm's alternate  
> screen
> feature.
>
> Now let's try emacs and mutt again:
>
> $ ls
> [blablabla]
> $ /opt/csw/bin/emacs
> [exit emacs, the screen returns to its previous content, including  
> showing
> the output of the previous 'ls' command, just as if emacs were run  
> from a
> normal shell instead of being run from within 'screen'.  Great,  
> that's what
> I want.]
> $ ls
> [blablabla]
> $ /opt/csw/bin/mutt
> [oops, watch screen and mutt fight for control of the alternate  
> screen...
> You can try to type a quick random combination of
>
> x
>
> and
>
> Control-A "
>
> to tell mutt to exit and screen to give you a list of virtual screens
> (rather than fighting with mutt) but good luck with regaining  
> control of
> your window...]
>
> Now, the fact that emacs works fine in combination with screen's  
> altscreen
> feature but that mutt does not tells me that the problem is with  
> mutt, not
> screen. After investigating a little, I've come to the conclusion  
> that the
> problem is not with the code of mutt itself, but with the fact that
> /opt/csw/bin/mutt uses slang while /opt/csw/bin/screen uses ncurses.  
> In
> fact I have compiled (with gcc) a version of mutt 1.5.19 with  
> ncurses 5.7
> which works perfectly well in the examples above.  On the other hand  
> the
> same version of mutt 1.5.19 compiled with slang 2.1.4 fails just like
> /opt/csw/bin/mutt, flashing the screen and all.  My /opt/csw/bin/ 
> mutt uses
> slang 1.4.8, not slang 2.1.4, but that doesn't seem to make any  
> difference,
> both fail in the same way.
>
> So is there a way to get /opt/csw/bin/screen to be compiled with  
> ncurses
> rather than slang, by any chance?
>
> Thanks,
>
>
> ======================================================================
>
> Issue History
> Date Modified    Username       Field                    Change
> ======================================================================
> 2009-04-29 07:59 meunier        New Issue
> ======================================================================
>
> _______________________________________________
> bug-notifications mailing list
> bug-notifications at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/bug-notifications



From hson at opencsw.org  Wed Apr 29 09:49:40 2009
From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=)
Date: Wed, 29 Apr 2009 09:49:40 +0200
Subject: [csw-maintainers] solaris 8 on CentOS?
In-Reply-To: <49F7A0CD.2090405@opencsw.org>
References: <49F7A0CD.2090405@opencsw.org>
Message-ID: <49F80694.1000103@opencsw.org>

Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I remember someone saying they had solaris 8
> running on CentOS, sadly, I don't remember whom.
> 
> if you still have it, can you post your config.
> 

It was me.

I'm running CentOS 5.2 with VMware server 2.0 and the virtual machine 
config isn't anything strange at all, mainly guestOS = "solaris8".

I can send you my config file if you wish (or even upload my image)


From dam at opencsw.org  Wed Apr 29 13:57:52 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 13:57:52 +0200
Subject: [csw-maintainers] Fwd:  Broken CSWftype2
References: <66824D49-2D25-4398-B93B-907070806DED@baltic-online.de>
Message-ID: <87C9EAD1-5534-4217-9A01-0BB49F9732F8@opencsw.org>

Hi,

Am 16.04.2009 um 09:04 schrieb Dagobert Michelsen:
> Am 20.02.2009 um 12:56 schrieb Ihsan Dogan:
>> Am 17.2.2009 19:03 Uhr, Dagobert Michelsen schrieb:
>>>
>>>> brauchst zum Kompilieren die aktuelle freetype2 Version aus /
>>>> testing.
>>>> Koenntest Du das bitte auf der buildfarm einspielen ?
>>>> Ziel ist es, freetype2 und fontconfig nach einer Testphase
>>>> gleichzeitig zu releasen.
>>>
>>> Ok, so freetype2 and fontconfig should be released at the
>>> same time and freetype2 must be installed on the buildfarm
>>> for building fontconfig.
>>>
>>> Ihsan, Phil: Should I do it this way violating standards or
>>> should these packages be released one after another?
>>
>> As I understand, we don't have many choices, so it's fine for me.
>> Important is, that a notify is sent to the maintainers list.
>
> We have not build8st for this. AFAIK this has not been done yet.
> Want to give it a try?

I notice that fontconfig has been released, but not freetype2.
Is this ok? Should I recompile now cairo, pango and atk or
wait until freetype2 has been released?


Best regards

  -- Dago



From dam at opencsw.org  Wed Apr 29 14:41:18 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 14:41:18 +0200
Subject: [csw-maintainers] New Subversion 1.6.1 in testing
In-Reply-To: <49F77786.6080308@opencsw.org>
References: <49F77786.6080308@opencsw.org>
Message-ID: <CF7958CE-EC9F-4EDC-AF91-00DF21FEF031@opencsw.org>

Hi Mike,

Am 28.04.2009 um 23:39 schrieb Mike Watters:
> New Subversion 1.6.1 in testing

I mirrored up GAR from SourceForge using svnsync and did some
other testing - works great!


Best regads

   -- Dago


From pfelecan at opencsw.org  Wed Apr 29 17:26:25 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 29 Apr 2009 17:26:25 +0200
Subject: [csw-maintainers] packages names normalization (long)
Message-ID: <x6r5zb8ori.fsf@abhavana.free.fr>

Reading the standards page,
http://www.opencsw.org/standards/pkgcreation, I understand that the
standard for naming CSW splitted packages is:

PKGname	     shortname
------------ ---------
CSWsoftrt    soft_rt
CSWsoftdevel soft_devel
CSWsoftdoc   soft_doc 

Some of the cited rationales are: to optimize the download size of the
required packages, to offer a better granularity in packages
inter-dependencies, &c.

For the moment, this standard is not applied consistently, but I
understand that it is now enforced.

Here is a more developed functional relation with the suffixes cited above:

_rt    : run-time

         Any component needed to run the product or *derived* products:
         shared libraries, &c

_devel : development

         Any component needed for developing *derived*  but not
         needed for running it: headers, static libraries, pkgconfig
         elements, &c

_doc   : documentation

         Any component documenting the product and/or the creation of
         *derived* products.

There is something missing: the nil suffix which, in my opinion,
group the components which are *run*, i.e. binaries, shell scripts, &c

A package with a nil suffix, using the above examples, has the form:

CSWsoft      soft
CSWsoftrt    soft_rt
CSWsoftdevel soft_devel
CSW softdoc  soft_doc

Why do I think that we should have nil suffixed packages names:

1. Installing the product by:

   pkg-get install softrt

   is less intuitive than:

   pkg-get install soft

2. Installing the product's binaries is not always necessary to run
   *derived* products, its run-time is however necessary.

Giving a full example is probably a good way to enhance the
documentation of this standard. Consequently, here is the splitting of
the autogen project:

CSWautogen      autogen
CSWautogenrt    autogen_rt
CSWautogendevel autogen_devel
CSWautogendoc   autogen_doc

and their respective content is (I omitted the /opt/csw prefix):

autogen:

	bin/autogen
	bin/autoopts-config
	bin/columns
	bin/getdefs
	bin/xml2ag
	share/autogen/aginfo.tpl
	share/autogen/aginfo3.tpl
	share/autogen/agman-lib.tpl
	share/autogen/agman1.tpl
	share/autogen/agman3.tpl
	share/autogen/autoopts.m4
	share/autogen/confmacs.tpl
	share/autogen/conftest.tpl
	share/autogen/fsm-macro.tpl
	share/autogen/fsm-trans.tpl
	share/autogen/fsm.tpl
	share/autogen/getopt.tpl
	share/autogen/optcode.tpl
	share/autogen/opthead.tpl
	share/autogen/options.tpl
	share/autogen/optlib.tpl
	share/autogen/optmain.tpl
	share/autogen/rc-sample.tpl
	share/autogen/stdoptions.def
	share/autogen/usage.tpl
	share/man/man1/autogen.1
	share/man/man1/autoopts-config.1
	share/man/man1/columns.1
	share/man/man1/getdefs.1
	share/man/man1/xml2ag.1

autogen_rt:

	lib/libguileopts.so.0.0.1
	lib/libopts.so.25.7.0
	share/doc/autogen/AUTHORS
	share/doc/autogen/COPYING
	share/doc/autogen/ChangeLog
	share/doc/autogen/INSTALL
	share/doc/autogen/NEWS
	share/doc/autogen/README
	share/doc/autogen/THANKS
	share/doc/autogen/TODO

autogen_devel:

	include/autoopts/options.h
	include/autoopts/usage-txt.h
	lib/libguileopts.a
	lib/libopts.a
	lib/pkgconfig
	lib/pkgconfig/autoopts.pc
	share/aclocal/autoopts.m4
	share/aclocal/liboptschk.m4
	share/man/man3/ao_string_tokenize.3
	share/man/man3/configFileLoad.3
	share/man/man3/optionFileLoad.3
	share/man/man3/optionFindNextValue.3
	share/man/man3/optionFindValue.3
	share/man/man3/optionFree.3
	share/man/man3/optionGetValue.3
	share/man/man3/optionLoadLine.3
	share/man/man3/optionNextValue.3
	share/man/man3/optionOnlyUsage.3
	share/man/man3/optionProcess.3
	share/man/man3/optionRestore.3
	share/man/man3/optionSaveFile.3
	share/man/man3/optionSaveState.3
	share/man/man3/optionUnloadNested.3
	share/man/man3/optionVersion.3
	share/man/man3/strequate.3
	share/man/man3/streqvcmp.3
	share/man/man3/streqvmap.3
	share/man/man3/strneqvcmp.3
	share/man/man3/strtransform.3

autogen_doc:

	share/doc/autogen/autogen.dvi
	share/doc/autogen/autogen.pdf
	share/doc/autogen/autogen.ps
	share/info/autogen.info
	share/info/autogen.info-1
	share/info/autogen.info-2

All your comments are welcome and when we agree it should be nice to
update the standards page.
-- 
Peter


From dam at opencsw.org  Wed Apr 29 17:38:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 17:38:40 +0200
Subject: [csw-maintainers] libcairo 1.8.6 available
Message-ID: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>

Hi,

I just put libcairo 1.8.6 compiled against the updated
fontconfig. Please test it so we can move ahead with all
the other stuff depending on it:

libcairo-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
libcairo-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
libcairo_doc-1.8.6,REV=2009.04.29-SunOS5.8-all-CSW.pkg.gz

pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libcairo
pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libcairo

Best regards

   -- Dago


From dam at opencsw.org  Wed Apr 29 17:52:38 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 17:52:38 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6r5zb8ori.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
Message-ID: <4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org>

Hi Peter,

Am 29.04.2009 um 17:26 schrieb Peter FELECAN:
> Reading the standards page,
> http://www.opencsw.org/standards/pkgcreation, I understand that the
> standard for naming CSW splitted packages is:
>
> PKGname	     shortname
> ------------ ---------
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSWsoftdoc   soft_doc
>
> Some of the cited rationales are: to optimize the download size of the
> required packages, to offer a better granularity in packages
> inter-dependencies, &c.
>
> For the moment, this standard is not applied consistently, but I
> understand that it is now enforced.
>
> Here is a more developed functional relation with the suffixes cited  
> above:
>
> _rt    : run-time
>
>         Any component needed to run the product or *derived* products:
>         shared libraries, &c
>
> _devel : development
>
>         Any component needed for developing *derived*  but not
>         needed for running it: headers, static libraries, pkgconfig
>         elements, &c
>
> _doc   : documentation
>
>         Any component documenting the product and/or the creation of
>         *derived* products.
>
> There is something missing: the nil suffix which, in my opinion,
> group the components which are *run*, i.e. binaries, shell scripts, &c

It may also very well be a meta-package to say "I just want to
use this software". In most cases this will be the binaries and
depend on _rt. However, in packages whose sole purpose is the
development of software things get more complicated as I'll
illustrate below.

> A package with a nil suffix, using the above examples, has the form:
>
> CSWsoft      soft
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSW softdoc  soft_doc
>
> Why do I think that we should have nil suffixed packages names:
>
> 1. Installing the product by:
>
>   pkg-get install softrt

_rt-packages will usually be never installed manually, they are
meant to be pulled in as dependencies. Factoring out _rt to
minimize impact of dependencies is one of the reasons that
runtime-stuff is actually split off in a separate package.

>
>
>   is less intuitive than:
>
>   pkg-get install soft
>
> 2. Installing the product's binaries is not always necessary to run
>   *derived* products, its run-time is however necessary.

True.

> Giving a full example is probably a good way to enhance the
> documentation of this standard. Consequently, here is the splitting of
> the autogen project:

Bad example IMHO, as the sole purpose of autogen is to aid
in software development and needed for compilation. No user
will use it.

> CSWautogen      autogen
> CSWautogenrt    autogen_rt
> CSWautogendevel autogen_devel
> CSWautogendoc   autogen_doc
>
> and their respective content is (I omitted the /opt/csw prefix):
>
> autogen:
>
> 	bin/autogen
> 	bin/autoopts-config

By default bin/*-config is put in _devel in GAR.

>
> 	bin/columns
> 	bin/getdefs
> 	bin/xml2ag
> 	share/autogen/aginfo.tpl
> 	share/autogen/aginfo3.tpl
> 	share/autogen/agman-lib.tpl
> 	share/autogen/agman1.tpl
> 	share/autogen/agman3.tpl
> 	share/autogen/autoopts.m4
> 	share/autogen/confmacs.tpl
> 	share/autogen/conftest.tpl
> 	share/autogen/fsm-macro.tpl
> 	share/autogen/fsm-trans.tpl
> 	share/autogen/fsm.tpl
> 	share/autogen/getopt.tpl
> 	share/autogen/optcode.tpl
> 	share/autogen/opthead.tpl
> 	share/autogen/options.tpl
> 	share/autogen/optlib.tpl
> 	share/autogen/optmain.tpl
> 	share/autogen/rc-sample.tpl
> 	share/autogen/stdoptions.def
> 	share/autogen/usage.tpl
> 	share/man/man1/autogen.1
> 	share/man/man1/autoopts-config.1

Same is for the manpage.

>
> 	share/man/man1/columns.1
> 	share/man/man1/getdefs.1
> 	share/man/man1/xml2ag.1
>
> autogen_rt:
>
> 	lib/libguileopts.so.0.0.1
> 	lib/libopts.so.25.7.0
>
> 	share/doc/autogen/AUTHORS
> 	share/doc/autogen/COPYING
> 	share/doc/autogen/ChangeLog
> 	share/doc/autogen/INSTALL
> 	share/doc/autogen/NEWS
> 	share/doc/autogen/README
> 	share/doc/autogen/THANKS
> 	share/doc/autogen/TODO

Docs usually go into the base package if there is no _doc.

> autogen_devel:
>
> 	include/autoopts/options.h
> 	include/autoopts/usage-txt.h
> 	lib/libguileopts.a
> 	lib/libopts.a
> 	lib/pkgconfig
> 	lib/pkgconfig/autoopts.pc
> 	share/aclocal/autoopts.m4
> 	share/aclocal/liboptschk.m4
> 	share/man/man3/ao_string_tokenize.3
> 	share/man/man3/configFileLoad.3
> 	share/man/man3/optionFileLoad.3
> 	share/man/man3/optionFindNextValue.3
> 	share/man/man3/optionFindValue.3
> 	share/man/man3/optionFree.3
> 	share/man/man3/optionGetValue.3
> 	share/man/man3/optionLoadLine.3
> 	share/man/man3/optionNextValue.3
> 	share/man/man3/optionOnlyUsage.3
> 	share/man/man3/optionProcess.3
> 	share/man/man3/optionRestore.3
> 	share/man/man3/optionSaveFile.3
> 	share/man/man3/optionSaveState.3
> 	share/man/man3/optionUnloadNested.3
> 	share/man/man3/optionVersion.3
> 	share/man/man3/strequate.3
> 	share/man/man3/streqvcmp.3
> 	share/man/man3/streqvmap.3
> 	share/man/man3/strneqvcmp.3
> 	share/man/man3/strtransform.3
>
> autogen_doc:
>
> 	share/doc/autogen/autogen.dvi
> 	share/doc/autogen/autogen.pdf
> 	share/doc/autogen/autogen.ps
> 	share/info/autogen.info
> 	share/info/autogen.info-1
> 	share/info/autogen.info-2

texinfo pages are put in the main-package as they can be
thought of another form of manpages.

> All your comments are welcome and when we agree it should be nice to
> update the standards page.

Additionally, there should be license files
   /opt/csw/share/doc/autogen/license
   /opt/csw/share/doc/autogen_rt/license
   /opt/csw/share/doc/autogen_devel/license
   /opt/csw/share/doc/autogen_doc/license
with the contents of share/doc/autogen/COPYING


Best regards

   -- Dago


From pfelecan at opencsw.org  Wed Apr 29 18:17:43 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 29 Apr 2009 18:17:43 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org> (Dagobert
	Michelsen's message of "Wed\, 29 Apr 2009 17\:52\:38 +0200")
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org>
Message-ID: <x6my9z8me0.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

Thank you Dago for your quick and appropriate answer.

>> Giving a full example is probably a good way to enhance the
>> documentation of this standard. Consequently, here is the splitting of
>> the autogen project:
>
> Bad example IMHO, as the sole purpose of autogen is to aid
> in software development and needed for compilation. No user
> will use it.

I know that it's an artificial, viz. inappropriate, example. However, I
choose it because I had a lengthy discussion with Phil about the need to
split it (I wanted to deliver it as an unique package, he had the
opinion that I should split)

>> CSWautogen      autogen
>> CSWautogenrt    autogen_rt
>> CSWautogendevel autogen_devel
>> CSWautogendoc   autogen_doc
>>
>> and their respective content is (I omitted the /opt/csw prefix):
>>
>> autogen:
>>
>> 	bin/autogen
>> 	bin/autoopts-config
>
> By default bin/*-config is put in _devel in GAR.

Right.

>> 	share/man/man1/autoopts-config.1
>
> Same is for the manpage.

Agreed.

>> autogen_rt:
>>
>> 	lib/libguileopts.so.0.0.1
>> 	lib/libopts.so.25.7.0
>>
>> 	share/doc/autogen/AUTHORS
>> 	share/doc/autogen/COPYING
>> 	share/doc/autogen/ChangeLog
>> 	share/doc/autogen/INSTALL
>> 	share/doc/autogen/NEWS
>> 	share/doc/autogen/README
>> 	share/doc/autogen/THANKS
>> 	share/doc/autogen/TODO
>
> Docs usually go into the base package if there is no _doc.

I put this quasi-standard documentation files in the run-time package
because it's installed always and is of interest for everybody (a
little bit like Debian does).

>> autogen_doc:
>>
>> 	share/doc/autogen/autogen.dvi
>> 	share/doc/autogen/autogen.pdf
>> 	share/doc/autogen/autogen.ps
>> 	share/info/autogen.info
>> 	share/info/autogen.info-1
>> 	share/info/autogen.info-2
>
> texinfo pages are put in the main-package as they can be
> thought of another form of manpages.

I'm not sure about this. If I look other packaging sources, again Debian
comes up as an interesting example, the info files are in a separate,
documentation package. It can be argued both ways...

> Additionally, there should be license files
>   /opt/csw/share/doc/autogen/license
>   /opt/csw/share/doc/autogen_rt/license
>   /opt/csw/share/doc/autogen_devel/license
>   /opt/csw/share/doc/autogen_doc/license
> with the contents of share/doc/autogen/COPYING

Right. And they are but not cited in my example.

-- 
Peter


From Darin.Perusich at cognigencorp.com  Wed Apr 29 18:22:03 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 29 Apr 2009 12:22:03 -0400
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6r5zb8ori.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
Message-ID: <49F87EAB.3060100@cognigencorp.com>

One thing I feel needs to be addressed is package naming consistency
between PKGname and shortname. If the short name is soft_rt or soft-rt
shouldn't the PKGname be CSWsoft_rt or CSWsoft-rt? There's no
"technical" reason for this given my reading of the pkginfo manual. This
lack of consistency makes searching for packages with pkginfo or listing
/var/sadm/pkg a really hassle.

A perfect example of this is openSSL or openSSH. The PKGname for the
various component packages is CSWossl* or CSWossh*. One would kind of
expect to the PKGname to be CSWopenssl_rt or CSWopenssh_client. I
realize there are historical reasons for not changed this, I'm simply
using this as examples.

Peter FELECAN wrote:
> Reading the standards page,
> http://www.opencsw.org/standards/pkgcreation, I understand that the
> standard for naming CSW splitted packages is:
> 
> PKGname	     shortname
> ------------ ---------
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSWsoftdoc   soft_doc 
> 
> Some of the cited rationales are: to optimize the download size of the
> required packages, to offer a better granularity in packages
> inter-dependencies, &c.
> 
> For the moment, this standard is not applied consistently, but I
> understand that it is now enforced.
> 
> Here is a more developed functional relation with the suffixes cited above:
> 
> _rt    : run-time
> 
>          Any component needed to run the product or *derived* products:
>          shared libraries, &c
> 
> _devel : development
> 
>          Any component needed for developing *derived*  but not
>          needed for running it: headers, static libraries, pkgconfig
>          elements, &c
> 
> _doc   : documentation
> 
>          Any component documenting the product and/or the creation of
>          *derived* products.
> 
> There is something missing: the nil suffix which, in my opinion,
> group the components which are *run*, i.e. binaries, shell scripts, &c
> 
> A package with a nil suffix, using the above examples, has the form:
> 
> CSWsoft      soft
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSW softdoc  soft_doc
> 
> Why do I think that we should have nil suffixed packages names:
> 
> 1. Installing the product by:
> 
>    pkg-get install softrt
> 
>    is less intuitive than:
> 
>    pkg-get install soft
> 
> 2. Installing the product's binaries is not always necessary to run
>    *derived* products, its run-time is however necessary.
> 
> Giving a full example is probably a good way to enhance the
> documentation of this standard. Consequently, here is the splitting of
> the autogen project:
> 
> CSWautogen      autogen
> CSWautogenrt    autogen_rt
> CSWautogendevel autogen_devel
> CSWautogendoc   autogen_doc
> 
> and their respective content is (I omitted the /opt/csw prefix):
> 
> autogen:
> 
> 	bin/autogen
> 	bin/autoopts-config
> 	bin/columns
> 	bin/getdefs
> 	bin/xml2ag
> 	share/autogen/aginfo.tpl
> 	share/autogen/aginfo3.tpl
> 	share/autogen/agman-lib.tpl
> 	share/autogen/agman1.tpl
> 	share/autogen/agman3.tpl
> 	share/autogen/autoopts.m4
> 	share/autogen/confmacs.tpl
> 	share/autogen/conftest.tpl
> 	share/autogen/fsm-macro.tpl
> 	share/autogen/fsm-trans.tpl
> 	share/autogen/fsm.tpl
> 	share/autogen/getopt.tpl
> 	share/autogen/optcode.tpl
> 	share/autogen/opthead.tpl
> 	share/autogen/options.tpl
> 	share/autogen/optlib.tpl
> 	share/autogen/optmain.tpl
> 	share/autogen/rc-sample.tpl
> 	share/autogen/stdoptions.def
> 	share/autogen/usage.tpl
> 	share/man/man1/autogen.1
> 	share/man/man1/autoopts-config.1
> 	share/man/man1/columns.1
> 	share/man/man1/getdefs.1
> 	share/man/man1/xml2ag.1
> 
> autogen_rt:
> 
> 	lib/libguileopts.so.0.0.1
> 	lib/libopts.so.25.7.0
> 	share/doc/autogen/AUTHORS
> 	share/doc/autogen/COPYING
> 	share/doc/autogen/ChangeLog
> 	share/doc/autogen/INSTALL
> 	share/doc/autogen/NEWS
> 	share/doc/autogen/README
> 	share/doc/autogen/THANKS
> 	share/doc/autogen/TODO
> 
> autogen_devel:
> 
> 	include/autoopts/options.h
> 	include/autoopts/usage-txt.h
> 	lib/libguileopts.a
> 	lib/libopts.a
> 	lib/pkgconfig
> 	lib/pkgconfig/autoopts.pc
> 	share/aclocal/autoopts.m4
> 	share/aclocal/liboptschk.m4
> 	share/man/man3/ao_string_tokenize.3
> 	share/man/man3/configFileLoad.3
> 	share/man/man3/optionFileLoad.3
> 	share/man/man3/optionFindNextValue.3
> 	share/man/man3/optionFindValue.3
> 	share/man/man3/optionFree.3
> 	share/man/man3/optionGetValue.3
> 	share/man/man3/optionLoadLine.3
> 	share/man/man3/optionNextValue.3
> 	share/man/man3/optionOnlyUsage.3
> 	share/man/man3/optionProcess.3
> 	share/man/man3/optionRestore.3
> 	share/man/man3/optionSaveFile.3
> 	share/man/man3/optionSaveState.3
> 	share/man/man3/optionUnloadNested.3
> 	share/man/man3/optionVersion.3
> 	share/man/man3/strequate.3
> 	share/man/man3/streqvcmp.3
> 	share/man/man3/streqvmap.3
> 	share/man/man3/strneqvcmp.3
> 	share/man/man3/strtransform.3
> 
> autogen_doc:
> 
> 	share/doc/autogen/autogen.dvi
> 	share/doc/autogen/autogen.pdf
> 	share/doc/autogen/autogen.ps
> 	share/info/autogen.info
> 	share/info/autogen.info-1
> 	share/info/autogen.info-2
> 
> All your comments are welcome and when we agree it should be nice to
> update the standards page.

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From dam at opencsw.org  Wed Apr 29 18:36:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 18:36:51 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6my9z8me0.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<4A72D5B9-C37D-4B78-BC31-B9DE17B48656@opencsw.org>
	<x6my9z8me0.fsf@abhavana.free.fr>
Message-ID: <BCE77DAB-3F48-437D-B211-7A5D6CC53242@opencsw.org>

Hi Peter,

Am 29.04.2009 um 18:17 schrieb Peter FELECAN:
> Dagobert Michelsen <dam at opencsw.org> writes:
>
> Thank you Dago for your quick and appropriate answer.
>
>>> Giving a full example is probably a good way to enhance the
>>> documentation of this standard. Consequently, here is the  
>>> splitting of
>>> the autogen project:
>>
>> Bad example IMHO, as the sole purpose of autogen is to aid
>> in software development and needed for compilation. No user
>> will use it.
>
> I know that it's an artificial, viz. inappropriate, example.  
> However, I
> choose it because I had a lengthy discussion with Phil about the  
> need to
> split it (I wanted to deliver it as an unique package, he had the
> opinion that I should split)

To make a long story short: If there are packages that depend
on the .so-files split out _rt, otherwise I would argue to
leave it as one package as it is only valuable to developers
anyway.

Phil, did you have other reasons for the split?


Best regards

   -- Dago


From dam at opencsw.org  Wed Apr 29 18:39:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 18:39:55 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <49F87EAB.3060100@cognigencorp.com>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
Message-ID: <C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>

Hi Darin,

Am 29.04.2009 um 18:22 schrieb Darin Perusich:
> One thing I feel needs to be addressed is package naming consistency
> between PKGname and shortname. If the short name is soft_rt or soft-rt
> shouldn't the PKGname be CSWsoft_rt or CSWsoft-rt? There's no
> "technical" reason for this given my reading of the pkginfo manual.  
> This
> lack of consistency makes searching for packages with pkginfo or  
> listing
> /var/sadm/pkg a really hassle.
>
> A perfect example of this is openSSL or openSSH. The PKGname for the
> various component packages is CSWossl* or CSWossh*. One would kind of
> expect to the PKGname to be CSWopenssl_rt or CSWopenssh_client. I
> realize there are historical reasons for not changed this, I'm simply
> using this as examples.

The Sun "Standard" uses "-" as separator, so it would be more like
CSWopenssl-rt. However, some packages already use no separator like
CSWopensslrt.

I'm all for consistency, but as Phil correctly said changing package
names is kind of bad. If pkg-get/pkgutil were smart enough this *may*
be worked through, but I don't know if it is worth it.


Best regards

   -- Dago


From pfelecan at opencsw.org  Wed Apr 29 18:47:40 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 29 Apr 2009 18:47:40 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org> (Dagobert
	Michelsen's message of "Wed\, 29 Apr 2009 18\:39\:55 +0200")
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
Message-ID: <x6iqkn8l03.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Darin,
>
> Am 29.04.2009 um 18:22 schrieb Darin Perusich:
>> One thing I feel needs to be addressed is package naming consistency
>> between PKGname and shortname. If the short name is soft_rt or soft-rt
>> shouldn't the PKGname be CSWsoft_rt or CSWsoft-rt? There's no
>> "technical" reason for this given my reading of the pkginfo manual.
>> This
>> lack of consistency makes searching for packages with pkginfo or
>> listing
>> /var/sadm/pkg a really hassle.
>>
>> A perfect example of this is openSSL or openSSH. The PKGname for the
>> various component packages is CSWossl* or CSWossh*. One would kind of
>> expect to the PKGname to be CSWopenssl_rt or CSWopenssh_client. I
>> realize there are historical reasons for not changed this, I'm simply
>> using this as examples.
>
> The Sun "Standard" uses "-" as separator, so it would be more like
> CSWopenssl-rt. However, some packages already use no separator like
> CSWopensslrt.

The reason for which '-' cannot be used in catalogue package, vs. sys V
pkg, names is that is a "field separator" used for name, version,
revision, architecture &c; in my understanding, we can use '_' in
catalogue names and also in sys V pkg. If we want to normalize, your
example becomes CSWopenssl_rt with a catalogue name of openssl_rt.

-- 
Peter


From phil at bolthole.com  Wed Apr 29 19:08:24 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:08:24 -0700
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
Message-ID: <20090429170824.GB9202@bolthole.com>

On Wed, Apr 29, 2009 at 06:39:55PM +0200, Dagobert Michelsen wrote:
>
> The Sun "Standard" uses "-" as separator, so it would be more like
> CSWopenssl-rt. However, some packages already use no separator like
> CSWopensslrt.

personally, I like not bothering with any separator at all, for CSWxxx
names, unless the resulting name is somehow misleading/unreadable.



From phil at bolthole.com  Wed Apr 29 19:10:53 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:10:53 -0700
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6r5zb8ori.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
Message-ID: <20090429171053.GC9202@bolthole.com>

PS:

On Wed, Apr 29, 2009 at 05:26:25PM +0200, Peter FELECAN wrote:
> There is something missing: the nil suffix which, in my opinion,
> group the components which are *run*, i.e. binaries, shell scripts, &c
> 
> A package with a nil suffix, using the above examples, has the form:
> 
> CSWsoft      soft
> CSWsoftrt    soft_rt
> CSWsoftdevel soft_devel
> CSW softdoc  soft_doc

and.. some packages are like this already.

It depends on the individual software in question, whether there is, as you
put it, a "nil suffix" variant of the package.

And as Dagobert pointed out, "library" type things arent usually directly
installed via pkg-get. They are usually only pulled in as dependancies.



From phil at bolthole.com  Wed Apr 29 19:11:42 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:11:42 -0700
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <x6iqkn8l03.fsf@abhavana.free.fr>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
	<x6iqkn8l03.fsf@abhavana.free.fr>
Message-ID: <20090429171142.GD9202@bolthole.com>

On Wed, Apr 29, 2009 at 06:47:40PM +0200, Peter FELECAN wrote:
> The reason for which '-' cannot be used in catalogue package, vs. sys V
> pkg, names is that is a "field separator" used for name, version,
> revision, architecture &c; in my understanding, we can use '_' in
> catalogue names and also in sys V pkg. If we want to normalize, your
> example becomes CSWopenssl_rt with a catalogue name of openssl_rt.

except that "_" is invalid for SVR4 pkg naming, for some reason, i believe.



From dam at opencsw.org  Wed Apr 29 19:17:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 19:17:43 +0200
Subject: [csw-maintainers] packages names normalization (long)
In-Reply-To: <20090429171142.GD9202@bolthole.com>
References: <x6r5zb8ori.fsf@abhavana.free.fr>
	<49F87EAB.3060100@cognigencorp.com>
	<C7177F95-93C7-4277-8CF8-5DFDA5C04D11@opencsw.org>
	<x6iqkn8l03.fsf@abhavana.free.fr>
	<20090429171142.GD9202@bolthole.com>
Message-ID: <189FC9D7-90BE-4599-B1C6-B16444312B37@opencsw.org>

Hi Phil,

Am 29.04.2009 um 19:11 schrieb Philip Brown:
> On Wed, Apr 29, 2009 at 06:47:40PM +0200, Peter FELECAN wrote:
>> The reason for which '-' cannot be used in catalogue package, vs.  
>> sys V
>> pkg, names is that is a "field separator" used for name, version,
>> revision, architecture &c; in my understanding, we can use '_' in
>> catalogue names and also in sys V pkg. If we want to normalize, your
>> example becomes CSWopenssl_rt with a catalogue name of openssl_rt.
>
> except that "_" is invalid for SVR4 pkg naming, for some reason, i  
> believe.

Yes. From pkginfo(4):

>      PKG
>
>          Abbreviation for the package being installed. All  char-
>          acters in the abbreviation must be alphanumeric. You can
>          also use the - and +  characters  in  the  abbreviation.
>          The first character cannot be numeric, a + or a -.

Best regards

   -- Dago


From phil at bolthole.com  Wed Apr 29 19:37:24 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:37:24 -0700
Subject: [csw-maintainers] Fwd:  Broken CSWftype2
In-Reply-To: <87C9EAD1-5534-4217-9A01-0BB49F9732F8@opencsw.org>
References: <66824D49-2D25-4398-B93B-907070806DED@baltic-online.de>
	<87C9EAD1-5534-4217-9A01-0BB49F9732F8@opencsw.org>
Message-ID: <20090429173724.GF9202@bolthole.com>

On Wed, Apr 29, 2009 at 01:57:52PM +0200, Dagobert Michelsen wrote:
> I notice that fontconfig has been released, but not freetype2.
> Is this ok? Should I recompile now cairo, pango and atk or
> wait until freetype2 has been released?


Hm. weeell... since freetype2 "depends on" fontconfig... if the existing
released version does not work, seems like the most
appropriate thing to do is do a quick re-compile and releaes of freetype2
against the just released fontconfig.

please update build machines with latest freetype2 asap.

Then lets get a new freetype2 released quick?

And yes i think safest to wait until these two are done, given how picky
cairo, pango, etc have been in the past.



From phil at bolthole.com  Wed Apr 29 19:57:40 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 10:57:40 -0700
Subject: [csw-maintainers] Fwd: [bug-notifications] [mutt 0003648]:	mutt
	does not work with screen's altscreen because it is
	compiled	with slang instead of ncurses
In-Reply-To: <C097C6F7-20AE-4607-8F12-FB8CAC1D6BC1@opencsw.org>
References: <1b25a27cdbd0df04a3965beb5c31788e@www.opencsw.org>
	<C097C6F7-20AE-4607-8F12-FB8CAC1D6BC1@opencsw.org>
Message-ID: <20090429175740.GN9202@bolthole.com>

On Wed, Apr 29, 2009 at 09:13:43AM +0200, Dagobert Michelsen wrote:
> Hi,
>
> I picked up this discussion between ncurses and slang and I guess
> it is a good idea to discuss the ncurses/slang matter publicly
> on the list.

ncurses sucks. use slang.
that's all I have to say on that one :-)



From dam at opencsw.org  Wed Apr 29 20:48:17 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 20:48:17 +0200
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <20090428191531.GC97557@bolthole.com>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
	<20090428191531.GC97557@bolthole.com>
Message-ID: <270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>

Hi Phil,

Am 28.04.2009 um 21:15 schrieb Philip Brown:

> On Tue, Apr 28, 2009 at 01:33:30PM -0500, Mike Watters wrote:
>> Phil: have you finished processing and pushed out all of the php5  
>> packages yet?
>
> I have. yesterday.
> So, Dago probably just hit mirror lag.

No, there is something broken. I just looked up my rsync primary:

web at web [web]:/home/web/bin > rsync rsync://rsync.opencsw.org/csw/ 
current/sparc/5.8/ | grep php5
-rw-r--r--    1494635 2009/04/27 23:17:07  
ap2_modphp5-5.2.9,REV=2009.04.27-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--    1377061 2009/04/27 23:17:07  
mod_php5-5.2.9,REV=2009.04.27-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--    3056820 2009/04/20 19:14:28 php5-5.2.9,REV=2009.04.18- 
SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      55316 2009/02/23 20:38:46  
php5_apc-3.0.19,REV=2009.02.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      31096 2009/04/20 19:14:28  
php5_bcmath-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      14633 2009/04/20 19:14:29  
php5_bz2-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      18688 2009/04/20 19:14:29  
php5_calendar-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       8435 2009/04/20 19:14:29  
php5_ctype-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      29962 2009/04/20 19:14:30  
php5_curl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      35878 2009/04/20 19:14:30  
php5_dba-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20505 2009/04/20 19:14:30  
php5_dbase-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     516915 2009/04/20 19:14:31  
php5_devel-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      74762 2009/04/20 19:14:31  
php5_dom-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      30733 2009/04/20 19:14:31  
php5_exif-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--        748 2009/03/12 20:30:25  
php5_filter-5.2.6,REV=2009.03.08-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      24655 2009/04/20 19:14:32  
php5_ftp-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      50115 2009/04/20 19:14:32 php5_gd-5.2.9,REV=2009.04.18- 
SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       9399 2009/04/20 19:14:32  
php5_gettext-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20005 2009/04/20 19:14:33  
php5_gmp-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     124240 2009/04/20 19:14:33  
php5_hash-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      21397 2009/04/20 19:14:33  
php5_iconv-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      48371 2009/04/20 19:14:34  
php5_imap-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      17236 2009/04/20 19:14:34  
php5_json-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      25121 2009/04/20 19:14:34  
php5_ldap-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--    1132091 2009/04/20 19:14:35  
php5_mbstring-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      17867 2009/04/20 19:14:35  
php5_mcrypt-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       8978 2009/04/20 19:14:35  
php5_mhash-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      16366 2009/04/20 19:14:35  
php5_mime_magic-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      28450 2009/04/20 19:16:34  
php5_mssql-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      24222 2009/04/20 19:14:36  
php5_mysql-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      51030 2009/04/20 19:14:36  
php5_mysqli-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      27016 2009/04/20 19:14:37  
php5_ncurses-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      34827 2009/04/20 19:14:37  
php5_odbc-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      44545 2009/04/20 19:14:37  
php5_openssl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      12290 2009/04/20 19:14:37  
php5_pcntl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      51806 2009/04/21 18:48:05  
php5_pdo-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      19154 2009/04/21 18:48:06  
php5_pdomysql-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      16195 2009/04/21 18:48:06  
php5_pdoodbc-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      18251 2009/04/21 18:48:06  
php5_pdopgsql-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     217588 2009/04/21 20:32:22  
php5_pdosqlite-5.2.9,REV=2009.04.21-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      54032 2009/04/20 19:14:38  
php5_pgsql-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      13132 2009/04/20 19:14:38  
php5_posix-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      15107 2009/04/20 19:14:38  
php5_pspell-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11961 2009/04/21 01:42:25  
php5_readline-5.2.9,REV=2009.04.20-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      37229 2009/04/20 19:14:39  
php5_session-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       9237 2009/04/20 19:14:39  
php5_shmop-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      16712 2009/04/20 19:14:39  
php5_snmp-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     169450 2009/04/20 19:14:40  
php5_soap-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20875 2009/04/20 19:14:40  
php5_sockets-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      35274 2009/04/20 19:14:40  
php5_sqlite-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11806 2009/04/20 19:14:41  
php5_sysvmsg-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--       8859 2009/04/20 19:14:41  
php5_sysvsem-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11636 2009/04/20 19:14:41  
php5_sysvshm-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      20718 2009/04/20 19:14:42  
php5_tidy-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      11246 2009/04/20 19:14:42  
php5_tokenizer-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      19022 2009/04/20 19:14:42  
php5_wddx-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--     101274 2009/02/23 20:38:47  
php5_xdebug-2.0.4,REV=2009.02.17-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      15563 2009/04/20 19:14:43  
php5_xmlreader-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      42736 2009/04/20 19:14:43  
php5_xmlrpc-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      14638 2009/04/20 19:14:43  
php5_xmlwriter-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      17480 2009/04/20 19:14:43  
php5_xsl-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz
-rw-r--r--      57071 2009/04/20 19:14:44  
php5_zip-5.2.9,REV=2009.04.18-SunOS5.8-sparc-CSW.pkg.gz

...old!

Am 28.04.2009 um 20:33 schrieb Mike Watters:
> Dago,
> This was the bug that I re-packaged to fix.  can you verify for me  
> what REV you
> attempted to install?
>
> when I do a pkg-get -d php5_zip I get the "broken" one.   
> REV=2009.04.18
>
> The "Fixed" one is: REV=2009.04.27
>
> Phil: have you finished processing and pushed out all of the php5  
> packages yet?


You did release the package, yes? Or isn't rsync.opencsw.org the  
primary any more?


Best regards

   -- Dago


From phil at bolthole.com  Wed Apr 29 21:03:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 12:03:33 -0700
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
	<20090428191531.GC97557@bolthole.com>
	<270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>
Message-ID: <20090429190332.GQ9202@bolthole.com>

On Wed, Apr 29, 2009 at 08:48:17PM +0200, Dagobert Michelsen wrote:
> Hi Phil,
>
> Am 28.04.2009 um 21:15 schrieb Philip Brown:
>
>> On Tue, Apr 28, 2009 at 01:33:30PM -0500, Mike Watters wrote:
>>> Phil: have you finished processing and pushed out all of the php5  
>>> packages yet?
>>
>> I have. yesterday.
>> So, Dago probably just hit mirror lag.
>
> No, there is something broken. I just looked up my rsync primary:

sorry sorry my mistake.



From phil at bolthole.com  Wed Apr 29 21:16:44 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 12:16:44 -0700
Subject: [csw-maintainers] Problem on CSWphp5zip
In-Reply-To: <20090429190332.GQ9202@bolthole.com>
References: <9A9D55C5-85A4-47D6-BB63-7994D627F0E2@opencsw.org>
	<49F74BFA.5040301@opencsw.org>
	<20090428191531.GC97557@bolthole.com>
	<270B9501-FE27-42CF-A57A-1660DCA2D40E@opencsw.org>
	<20090429190332.GQ9202@bolthole.com>
Message-ID: <20090429191641.GR9202@bolthole.com>

> 
> sorry sorry my mistake.

hmm. and the php5 "new" packages I hvae, ar bogus.
eg:

php5_openssl-5.2.9,REV=2009.04.27-SunOS5.8-sparc-UNCOMMITTED.pkg.gz




From william at wbonnet.net  Wed Apr 29 21:25:42 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 21:25:42 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
Message-ID: <49F8A9B6.602@wbonnet.net>

Hi
> I copied your old build descriptions to the GAR repository in the
> respective pkg/<pkg>/trunk/legacy/ folders as a reference if
> someone wants to bring a package in GAR.
libpango is a duplicate of pango and libxrender is a duplicate of 
x11/xrender

I think we should keep the old one

cheers
W.



-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From dam at opencsw.org  Wed Apr 29 21:36:09 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Apr 2009 21:36:09 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <49F8A9B6.602@wbonnet.net>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
Message-ID: <54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>

Hi William,

Am 29.04.2009 um 21:25 schrieb William Bonnet:
>> I copied your old build descriptions to the GAR repository in the
>> respective pkg/<pkg>/trunk/legacy/ folders as a reference if
>> someone wants to bring a package in GAR.
> libpango is a duplicate of pango and libxrender is a duplicate of  
> x11/xrender
>
> I think we should keep the old one

Well, the package is called CSWpango and the software is at pango.org,
so I chose pkg/pango instead of libpango, so I would prefer the new  
one ;-)
And xrender belongs to x11 where you put it and I also think this is a  
good
choice, no?


Best regards

   -- Dago


From william at wbonnet.net  Wed Apr 29 21:42:10 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 21:42:10 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
Message-ID: <49F8AD92.30203@wbonnet.net>

Hi Dagobert
> Well, the package is called CSWpango and the software is at pango.org,
> so I chose pkg/pango instead of libpango, so I would prefer the new 
> one ;-)
I also prefer the new one :)

cheers


-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Wed Apr 29 22:05:20 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 13:05:20 -0700
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>
	<49137326.3000605@wbonnet.net>
	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>
	<20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
Message-ID: <20090429200519.GT9202@bolthole.com>

On Wed, Apr 29, 2009 at 09:36:09PM +0200, Dagobert Michelsen wrote:
> Well, the package is called CSWpango and the software is at pango.org,
> so I chose pkg/pango instead of libpango, so I would prefer the new one 
> ;-)

but we have called "software name" to be "libpango" for a very long time.
and also other distribution call it "libpango".

so i think we should stick with "libpango".

ditto for libxrender.
its a standard name.



From william at wbonnet.net  Wed Apr 29 22:16:41 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 22:16:41 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <20090429200519.GT9202@bolthole.com>
References: <E1KyDUv-00008B-0c@c3vjzd1.ch3.sourceforge.com>	<49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>	<49F8A9B6.602@wbonnet.net>	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
	<20090429200519.GT9202@bolthole.com>
Message-ID: <49F8B5A9.6000208@wbonnet.net>

Hi

> but we have called "software name" to be "libpango" for a very long time.
> and also other distribution call it "libpango".
>
> so i think we should stick with "libpango".
>
> ditto for libxrender.
>   
So we move pango content to libpango and move libxrender under x11, then 
move xrender content into libxrender ?

cheers
W.

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From ihsan at opencsw.org  Wed Apr 29 22:33:27 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 29 Apr 2009 22:33:27 +0200
Subject: [csw-maintainers] libcairo 1.8.6 available
In-Reply-To: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
References: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
Message-ID: <49F8B997.9070705@opencsw.org>

Hi Dago,

Am 29.4.2009 17:38 Uhr, Dagobert Michelsen schrieb:

> I just put libcairo 1.8.6 compiled against the updated
> fontconfig. Please test it so we can move ahead with all
> the other stuff depending on it:
> 
> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.29-SunOS5.8-all-CSW.pkg.gz
> 
> pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libcairo
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libcairo

Could you please install it on build8st?



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Wed Apr 29 22:36:35 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 13:36:35 -0700
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <49F8B5A9.6000208@wbonnet.net>
References: <49137326.3000605@wbonnet.net>
	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>
	<20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
	<20090429200519.GT9202@bolthole.com> <49F8B5A9.6000208@wbonnet.net>
Message-ID: <20090429203635.GV9202@bolthole.com>

On Wed, Apr 29, 2009 at 10:16:41PM +0200, William Bonnet wrote:
> Hi
>
>> but we have called "software name" to be "libpango" for a very long time.
>> and also other distribution call it "libpango".
>>
>> so i think we should stick with "libpango".
>>
>> ditto for libxrender.
>>   
> So we move pango content to libpango

yes

> and move libxrender under x11, then  
> move xrender content into libxrender ?

umm... lost me there. is there separate "libxrender" and "xrender"?

PS: I'll remind you that we're supposed to have "flat namespace" in mGar
now! there should be no "under"!

otherwise, the nice future single-module autobuilder will not be able to
cleanly work!


From william at wbonnet.net  Wed Apr 29 22:50:29 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 29 Apr 2009 22:50:29 +0200
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <20090429203635.GV9202@bolthole.com>
References: <49137326.3000605@wbonnet.net>	<FE0FA7BE-734D-4CF4-8226-A304E481FB10@opencsw.org>	<20081107100724.GA24461@zerfleddert.de>	<491441EA.9000302@wbonnet.net>	<20081107133324.GA2059@zerfleddert.de>	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>	<49F8A9B6.602@wbonnet.net>	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>	<20090429200519.GT9202@bolthole.com>
	<49F8B5A9.6000208@wbonnet.net> <20090429203635.GV9202@bolthole.com>
Message-ID: <49F8BD95.3010106@wbonnet.net>

Hi
>> and move libxrender under x11, then  
>> move xrender content into libxrender ?
>>     
>
> umm... lost me there. is there separate "libxrender" and "xrender"?
>   
yes x11/xrender

cheers

-- 
William                  http://www.wbonnet.net

http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix
http://www.opencsw.org 	 Community SoftWare for Solaris
http://www.guses.org     French speaking Solaris User Group




From phil at bolthole.com  Wed Apr 29 23:09:36 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Apr 2009 14:09:36 -0700
Subject: [csw-maintainers] Build descriptions from Michael Gernoth
In-Reply-To: <49F8BD95.3010106@wbonnet.net>
References: <20081107100724.GA24461@zerfleddert.de>
	<491441EA.9000302@wbonnet.net>
	<20081107133324.GA2059@zerfleddert.de>
	<A60B8D95-9BE6-49B8-A58C-DF38EBB3A088@opencsw.org>
	<49F8A9B6.602@wbonnet.net>
	<54981458-C906-495C-9852-23EE41E5BCFB@opencsw.org>
	<20090429200519.GT9202@bolthole.com> <49F8B5A9.6000208@wbonnet.net>
	<20090429203635.GV9202@bolthole.com> <49F8BD95.3010106@wbonnet.net>
Message-ID: <20090429210936.GA79653@bolthole.com>

On Wed, Apr 29, 2009 at 10:50:29PM +0200, William Bonnet wrote:
> Hi
>>> and move libxrender under x11, then  move xrender content into 
>>> libxrender ?
>>>     
>>
>> umm... lost me there. is there separate "libxrender" and "xrender"?
>
> yes x11/xrender

well, unless they are SUPPOSED to be separate, then I would request it be
put in the top level, as "libxrender", to match software name



From mwatters at opencsw.org  Thu Apr 30 05:02:45 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 29 Apr 2009 22:02:45 -0500
Subject: [csw-maintainers] alpine 2.00 in testing
Message-ID: <49F914D5.9030008@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Sorry for the delay.
Please let me know how the testing goes.


- --
Thanks,
Mike

"Any intelligent fool can make things bigger, more complex,
and more violent.  It takes a touch of genius -- and a lot of courage --
to move in the opposite direction."

* Albert Einstein 1879 - 1955
    US German-born Theoretical Physicist
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkn5FNUACgkQLrhmsXMSLxejcACgpkGOfhu3B/umUOhif37kbU9y
eR4An3HH1eoJIFk48d0fkCoCXyvsNivC
=74JT
-----END PGP SIGNATURE-----


From ihsan at opencsw.org  Thu Apr 30 15:54:42 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 30 Apr 2009 15:54:42 +0200
Subject: [csw-maintainers] libcairo 1.8.6 available
In-Reply-To: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
References: <B8C4E107-5D52-4839-A323-74AAA6F76EB8@opencsw.org>
Message-ID: <49F9ADA2.3070205@opencsw.org>

Am 29.4.2009 17:38 Uhr, Dagobert Michelsen schrieb:

> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-i386-CSW.pkg.gz
> libcairo_devel-1.8.6,REV=2009.04.29-SunOS5.8-sparc-CSW.pkg.gz
> libcairo_doc-1.8.6,REV=2009.04.29-SunOS5.8-all-CSW.pkg.gz

Have you built pango, which is installed on build8st, also with this
cairo version?




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Thu Apr 30 21:12:54 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Apr 2009 12:12:54 -0700
Subject: [csw-maintainers] holey blackhole files, batman! pkginfo oddities
Message-ID: <20090430191254.GG79653@bolthole.com>

Just as a reference to others/for the future;

I had a problem where I was tryign to update CSWglib2.
it got stuck, becaus of zone problems.

i ended up downloading the newer version, and installing manually.

but.... pkg-get kept saying I still needed to "upgrade" it!
?!!...

after use of truss, I discovered an interesting "feature" of pkginfo. (ugh)

there was a leftover magical directory, 

/var/sadm/pkg/.save.CSWglib2

It kept looking in THERE, instead of the proper

/var/sadm/pkg/CSWglib2

for the installed file information.

once i trashed that directory, things finally proceeded properly.


From Darin.Perusich at cognigencorp.com  Thu Apr 30 22:43:42 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 30 Apr 2009 16:43:42 -0400
Subject: [csw-maintainers] relocatable on sparc but not x86
Message-ID: <49FA0D7E.4060103@cognigencorp.com>

I'm staging a package and on SPARC it's not relocatable but on i386 it's
not, why would this be?

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From phil at bolthole.com  Thu Apr 30 23:34:05 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Apr 2009 14:34:05 -0700
Subject: [csw-maintainers] relocatable on sparc but not x86
In-Reply-To: <49FA0D7E.4060103@cognigencorp.com>
References: <49FA0D7E.4060103@cognigencorp.com>
Message-ID: <20090430213405.GI79653@bolthole.com>

On Thu, Apr 30, 2009 at 04:43:42PM -0400, Darin Perusich wrote:
> I'm staging a package and on SPARC it's not relocatable but on i386 it's
> not, why would this be?
> 

that's kinda equivalent to saying, "it's broken. how can I fix it?" without saying
anything else :-/


at minimum, you need to find out what bits are non-relocatable, and tell us
what they are...