From toby.corkindale at strategicdata.com.au  Sun Apr  3 19:50:46 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Mon, 04 Apr 2011 12:50:46 +1000
Subject: [Melbourne-pm] April Meeting
Message-ID: <4D993206.70001@strategicdata.com.au>

Hello Perlmongers,
The next Melbourne Perlmongers meeting will be on Wednesday the 13th of 
April, and we'll probably be holding it at Strategic Data again.

Does anyone have a talk they would like to give?

Cheers,
Toby

From cosimo.streppone at gmail.com  Tue Apr  5 22:12:28 2011
From: cosimo.streppone at gmail.com (Cosimo Streppone)
Date: Wed, 6 Apr 2011 15:12:28 +1000
Subject: [Melbourne-pm] April Meeting
In-Reply-To: <4D993206.70001@strategicdata.com.au>
References: <4D993206.70001@strategicdata.com.au>
Message-ID: <BANLkTik=VYiZD15bc+tkwNLF+gnPq+a57w@mail.gmail.com>

On Mon, 04 Apr 2011 12:50 +1000, "Toby Corkindale"
<toby.corkindale at strategicdata.com.au> wrote:

> Hello Perlmongers,
> The next Melbourne Perlmongers meeting will be on Wednesday the 13th of
> April, and we'll probably be holding it at Strategic Data again.
>
> Does anyone have a talk they would like to give?

I'm currently on vacation and I haven't prepared anything specific yet.
After our last meeting, I recall there was some interest in our
geographic dns, varnish and plack.

If you don't have anything arranged yet, I could perhaps talk a bit of
those, as in informal chat
and answer your questions. If you want a more structured talk, I can bring some
slides from a previous presentation that I could use as a track.

-- 
Cosimo

From jarich at perltraining.com.au  Tue Apr  5 23:23:41 2011
From: jarich at perltraining.com.au (Jacinta Richardson)
Date: Wed, 06 Apr 2011 16:23:41 +1000
Subject: [Melbourne-pm] April Meeting
In-Reply-To: <4D993206.70001@strategicdata.com.au>
References: <4D993206.70001@strategicdata.com.au>
Message-ID: <4D9C06ED.200@perltraining.com.au>


> Does anyone have a talk they would like to give?

I think Jesse Vincent of Best Practical, (RT, Hiveminder, Jifty and 
other awesomenesss) will be back in town by next Wednesday, so perhaps 
we could ask him to bless us with a talk too? Jesse?

J

From alecclews at gmail.com  Wed Apr  6 03:49:38 2011
From: alecclews at gmail.com (Alec Clews)
Date: Wed, 06 Apr 2011 20:49:38 +1000
Subject: [Melbourne-pm] Fwd: [devops-aus] Announcing Devops Down Under
	2011/Call for proposals
Message-ID: <4D9C4542.8010501@gmail.com>

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

Might be of interest to some folks:

- -------- Original Message --------
Subject: 	[devops-aus] Announcing Devops Down Under 2011/Call for proposals
Date: 	Wed, 6 Apr 2011 18:41:02 +1000
From: 	Chris Bushell <cbushell at thoughtworks.com>
Reply-To: 	devops-aus at googlegroups.com
To: 	undisclosed-recipients:;



Following the success of Devops Down Under 2010, we're pleased to
announce Devops Down Under 2011, to be held on July 22/23 in Melbourne.

Our aim for the conference is to build on the current momentum within
the Devops community and to work towards uniting developers and
sysadmins/operations.

To make this event happen, we need your help. If you have a topic you'd
like to present, a speaker you'd like to see present, or a subject you'd
like to see discussed at the conference, please let us know here
<http://www.surveymonkey.com/s/FNF8MKL>. If you or your organisation is
interesting in sponsoring the event, also please let us know here
<http://www.surveymonkey.com/s/FNF8MKL>.

Registration and ticket details will be available the first week in May,
but for now, please save the date for Devops Down Under 2011 - we look
forward to seeing you there!


Lindsay Holmwood, Chris Bushell, Natalie Drucker

contact at devopsdownunder.org <mailto:contact at devopsdownunder.org>
www.devopsdownunder.org <http://www.devopsdownunder.org>

- -- 
You received this message because you are subscribed to the Google
Groups "DevOps Australia" group.
To post to this group, send email to devops-aus at googlegroups.com.
To unsubscribe from this group, send email to
devops-aus+unsubscribe at googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/devops-aus?hl=en.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk2cRUIACgkQP+e5j5u7/HxkwQCeMMlsJ6igjf3yoFZrLgALxF2g
kGAAn1zP9J/+9bkcCKHn1G7fMbXbXLVu
=NW9M
-----END PGP SIGNATURE-----

From toby.corkindale at strategicdata.com.au  Wed Apr  6 23:00:41 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Thu, 07 Apr 2011 16:00:41 +1000
Subject: [Melbourne-pm] April Meeting
In-Reply-To: <BANLkTik=VYiZD15bc+tkwNLF+gnPq+a57w@mail.gmail.com>
References: <4D993206.70001@strategicdata.com.au>
	<BANLkTik=VYiZD15bc+tkwNLF+gnPq+a57w@mail.gmail.com>
Message-ID: <4D9D5309.3000903@strategicdata.com.au>

On 06/04/11 15:12, Cosimo Streppone wrote:
> On Mon, 04 Apr 2011 12:50 +1000, "Toby Corkindale"
> <toby.corkindale at strategicdata.com.au>  wrote:
>
>> Hello Perlmongers,
>> The next Melbourne Perlmongers meeting will be on Wednesday the 13th of
>> April, and we'll probably be holding it at Strategic Data again.
>>
>> Does anyone have a talk they would like to give?
>
> I'm currently on vacation and I haven't prepared anything specific yet.
> After our last meeting, I recall there was some interest in our
> geographic dns, varnish and plack.
>
> If you don't have anything arranged yet, I could perhaps talk a bit
> of those, as in informal chat and answer your questions. If you want
> a  more structured talk, I can bring some
> slides from a previous presentation that I could use as a track.

Hi Cosimo,
If you're happy to talk on those, then that would be great - we seem to 
do well with the discussion forum way of running meetings. Maybe grab a 
few of your old slides to use as a structure though?

Thanks,
Toby

From ickphum at gmail.com  Thu Apr  7 21:20:08 2011
From: ickphum at gmail.com (Ian Macdonald)
Date: Fri, 8 Apr 2011 14:20:08 +1000
Subject: [Melbourne-pm] Our job ad on jobs.perl.org
Message-ID: <BANLkTik=sjjo7PfcLNPK+pUQgf7=XegoVQ@mail.gmail.com>

Hi,

I'd just like to point out our job ad on jobs.perl.org, as you probably
could have guessed from the subject.

http://jobs.perl.org/job/13990

Any technical questions may as well come straight to me as they'll end up
with me anyway...

Thanks,

Ian Macdonald
Armaguard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/melbourne-pm/attachments/20110408/c2a210d4/attachment.html>

From ts at meme.com.au  Sun Apr 10 00:54:40 2011
From: ts at meme.com.au (Tony Smith)
Date: Sun, 10 Apr 2011 17:54:40 +1000
Subject: [Melbourne-pm] Can May MelbPM meeting host the OSDC AGM?
Message-ID: <C0452E22-1F47-4C2E-954D-A2FD0088BC1B@meme.com.au>

OSDC is trying to move its subsequent Annual General Meetings to coincide with the annual conference late November/early December, but we need to have one more AGM in the interim to meet reporting obligations and get approval for that change.

Hopefully we can get it all done with in 20 minutes, so it will otherwise be a normal Melbourne PM meeting.

OSDC will need confirmation of this arrangement from this week's Melbourne PM meeting so we can get the required notices out on time.

For those not familiar with the history, the Open Source Developers Conference was originally a Melbourne PM initiative. Following its success we incorporated Open Source Developers Club to oversee the annual conference organisation and to act as an incorporated parent to related groups including Melbourne PM which then dropped its separate incorporation.





Tony Smith
Complex Systems Analyst
Melbourne, Australia
http://www.ynotds.com/
Giving thanks to the space, time, energy, matter and other lives that have allowed me to tell my lies on this old and damp ball of rock.


From toby.corkindale at strategicdata.com.au  Sun Apr 10 18:50:00 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Mon, 11 Apr 2011 11:50:00 +1000
Subject: [Melbourne-pm] Can May MelbPM meeting host the OSDC AGM?
In-Reply-To: <C0452E22-1F47-4C2E-954D-A2FD0088BC1B@meme.com.au>
References: <C0452E22-1F47-4C2E-954D-A2FD0088BC1B@meme.com.au>
Message-ID: <4DA25E48.1010004@strategicdata.com.au>

On 10/04/11 17:54, Tony Smith wrote:
> OSDC is trying to move its subsequent Annual General Meetings to coincide with the annual conference late November/early December, but we need to have one more AGM in the interim to meet reporting obligations and get approval for that change.
>
> Hopefully we can get it all done with in 20 minutes, so it will otherwise be a normal Melbourne PM meeting.
>
> OSDC will need confirmation of this arrangement from this week's Melbourne PM meeting so we can get the required notices out on time.
>
> For those not familiar with the history, the Open Source Developers Conference was originally a Melbourne PM initiative. Following its success we incorporated Open Source Developers Club to oversee the annual conference organisation and to act as an incorporated parent to related groups including Melbourne PM which then dropped its separate incorporation.

Hi Tony,
I don't think the Perlmongers will have any objections to hosting the 
OSDC AGM as such - however we would need a bigger venue - we've had some 
over-capacity meetings here, and I don't think we would fit in the extra 
OSDC crowd as well.

Is anyone on the list able to volunteer a larger meeting space for the 
May meeting?

Toby

From toby.corkindale at strategicdata.com.au  Sun Apr 10 19:02:12 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Mon, 11 Apr 2011 12:02:12 +1000
Subject: [Melbourne-pm] Melbourne Perlmongers - This wednesday!
Message-ID: <4DA26124.5030805@strategicdata.com.au>

Hello Perlmongers,
our next meeting will be at 6:30pm on Wednesday 13th of April - ie. this 
wednesday.

This meeting will have a couple of speakers, although in more of a 
discussion/forum kind of environment.

  * Cosimo Streppone from Opera talking about some or all of Plack, 
Varnish and geographic DNS.
  * Jesse Vincent from Best Practical (and current Pumpking) will be 
discussing what's new and exciting at the bleeding edge of the Perl core.


Strategic Data are hosting the meeting again, at:
Level 2,
51-55 Johnston street,
Fitzroy 3065


See you on Wednesday!
-Toby

From scottp at dd.com.au  Sun Apr 10 19:29:55 2011
From: scottp at dd.com.au (Scott Penrose)
Date: Sun, 10 Apr 2011 21:29:55 -0500 (CDT)
Subject: [Melbourne-pm] Can May MelbPM meeting host the OSDC AGM?
In-Reply-To: <4DA25E48.1010004@strategicdata.com.au>
Message-ID: <2113917922.7536.1302488995345.JavaMail.root@mail-4.01.com>

OSDC is in a position to provide some funds too. If that helps with a location for this event, although generally hired locations are often too expensive. I thought maybe OSDC could provide some sustenance for the next meeting?

Scott

----- Original Message -----
> On 10/04/11 17:54, Tony Smith wrote:
> > OSDC is trying to move its subsequent Annual General Meetings to
> > coincide with the annual conference late November/early December,
> > but we need to have one more AGM in the interim to meet reporting
> > obligations and get approval for that change.
> >
> > Hopefully we can get it all done with in 20 minutes, so it will
> > otherwise be a normal Melbourne PM meeting.
> >
> > OSDC will need confirmation of this arrangement from this week's
> > Melbourne PM meeting so we can get the required notices out on time.
> >
> > For those not familiar with the history, the Open Source Developers
> > Conference was originally a Melbourne PM initiative. Following its
> > success we incorporated Open Source Developers Club to oversee the
> > annual conference organisation and to act as an incorporated parent
> > to related groups including Melbourne PM which then dropped its
> > separate incorporation.
> 
> Hi Tony,
> I don't think the Perlmongers will have any objections to hosting the
> OSDC AGM as such - however we would need a bigger venue - we've had
> some
> over-capacity meetings here, and I don't think we would fit in the
> extra
> OSDC crowd as well.
> 
> Is anyone on the list able to volunteer a larger meeting space for the
> May meeting?
> 
> Toby
> _______________________________________________
> Melbourne-pm mailing list
> Melbourne-pm at pm.org
> http://mail.pm.org/mailman/listinfo/melbourne-pm

-- 
http://scott.dd.com.au/
scottp at dd.com.au



From ts at meme.com.au  Tue Apr 19 14:32:49 2011
From: ts at meme.com.au (Tony Smith)
Date: Wed, 20 Apr 2011 07:32:49 +1000
Subject: [Melbourne-pm] Still seeking OSDC AGM/next MelbPM meeting venue
Message-ID: <2C0FDF55-A907-4B51-88B4-775BF1441C28@meme.com.au>

Having come from a full day of entirely unconnected meetings, I wasn't entirely with it by the Perl Mongers end of last Wednesday. After the meeting I was approached by somebody whose name escapes me suggesting they may be able to approach a former employer whose venue we had previously used and get back to me, but I have heard nothing further.

Trouble is if we are to hold the OSDC AGM in conjunction with the next Perl Mongers meeting we need to get official notice out today. So if I don't get any better suggestions by early afternoon, I'm going to need to start trying pubs which are known to sometimes have suitable space. Even there I could use suggestions as I only ever get to a couple close enough to the city and transport.





Tony Smith
Complex Systems Analyst
Melbourne, Australia
http://www.ynotds.com/
Giving thanks to the space, time, energy, matter and other lives that have allowed me to tell my lies on this old and damp ball of rock.


From daniel at rimspace.net  Tue Apr 19 15:55:08 2011
From: daniel at rimspace.net (Daniel Pittman)
Date: Tue, 19 Apr 2011 15:55:08 -0700
Subject: [Melbourne-pm] Presentation tool that simulates typing on a tty...
Message-ID: <BANLkTi=qihZsY4vyV+VDSLRPg-oWcdBEog@mail.gmail.com>

So, approximately forever ago I saw a presentation given by Damian, in
which he used some presentation thingy that ran in a regular TTY and
simulated typing input and obtaining output ? typos included for free,
but without the risk of actually typing yourself during a
presentation.  (IIRC, it actually waited for a keypress to simulate
each bit of the output, for extra verisimilitude.)

Which now, three or four years later, would actually be really useful
to me.  So, I hope that someone here can point me at it; Google fails
me.

Thanks,
    Daniel
-- 
? Puppet Labs Developer ? http://puppetlabs.com
? Daniel Pittman <daniel at rimspace.net>
? Contact me via gtalk, email, or phone: +1 (503) 893-2285
? Made with 100 percent post-consumer electrons

From alfiejohn at gmail.com  Tue Apr 19 16:02:48 2011
From: alfiejohn at gmail.com (Alfie John)
Date: Wed, 20 Apr 2011 09:02:48 +1000
Subject: [Melbourne-pm] Presentation tool that simulates typing on a
	tty...
In-Reply-To: <BANLkTi=qihZsY4vyV+VDSLRPg-oWcdBEog@mail.gmail.com>
References: <BANLkTi=qihZsY4vyV+VDSLRPg-oWcdBEog@mail.gmail.com>
Message-ID: <BANLkTim5o1xfwy00kJgTp2hsFEc6ATgeog@mail.gmail.com>

Hey Daniel,

Your Googlefu might fail you, but your CPANfu shouldn't:

  http://search.cpan.org/~DCONWAY/

I think you're after IO::Prompt.

Alfie

On Wed, Apr 20, 2011 at 8:55 AM, Daniel Pittman <daniel at rimspace.net> wrote:

> So, approximately forever ago I saw a presentation given by Damian, in
> which he used some presentation thingy that ran in a regular TTY and
> simulated typing input and obtaining output ? typos included for free,
> but without the risk of actually typing yourself during a
> presentation.  (IIRC, it actually waited for a keypress to simulate
> each bit of the output, for extra verisimilitude.)
>
> Which now, three or four years later, would actually be really useful
> to me.  So, I hope that someone here can point me at it; Google fails
> me.
>
> Thanks,
>    Daniel
> --
> ? Puppet Labs Developer ? http://puppetlabs.com
> ? Daniel Pittman <daniel at rimspace.net>
> ? Contact me via gtalk, email, or phone: +1 (503) 893-2285
> ? Made with 100 percent post-consumer electrons
> _______________________________________________
> Melbourne-pm mailing list
> Melbourne-pm at pm.org
> http://mail.pm.org/mailman/listinfo/melbourne-pm
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/melbourne-pm/attachments/20110420/0db78b8c/attachment.html>

From damian at conway.org  Tue Apr 19 16:14:43 2011
From: damian at conway.org (Damian Conway)
Date: Wed, 20 Apr 2011 09:14:43 +1000
Subject: [Melbourne-pm] Presentation tool that simulates typing on a
	tty...
In-Reply-To: <BANLkTi=qihZsY4vyV+VDSLRPg-oWcdBEog@mail.gmail.com>
References: <BANLkTi=qihZsY4vyV+VDSLRPg-oWcdBEog@mail.gmail.com>
Message-ID: <BANLkTikaPrbCcFypOzNrAzYQ9BhH4gw49w@mail.gmail.com>

Daniel Pittman asked:

> So, approximately forever ago I saw a presentation given by Damian, in
> which he used some presentation thingy that ran in a regular TTY and
> simulated typing input and obtaining output ? typos included for free,
> but without the risk of actually typing yourself during a
> presentation. ?(IIRC, it actually waited for a keypress to simulate
> each bit of the output, for extra verisimilitude.)

http://search.cpan.org/~dconway/IO-Prompter-0.001001/lib/IO/Prompter.pm#Simulating_input

The trick is to put a regular:

    use IO::Prompter;

at the top of your program and a sneaky:

    use IO::Prompter <<'END_OF_AUTOMATED_INPUT'
    This will be typed in automatically for you
    Et cetera
    Et cetera
    END_OF_AUTOMATED_INPUT

way down the bottom of the program (e.g. after a hundred empty lines),
where it will never appear on screen.

The IO::Prompter distribution includes a program: demo/demo_fake.pl
which illustrates the technique.

Damian

From melbourne-pm at mjch.net  Tue Apr 19 16:51:59 2011
From: melbourne-pm at mjch.net (Malcolm Herbert)
Date: Wed, 20 Apr 2011 09:51:59 +1000
Subject: [Melbourne-pm] caching CPAN dist files
Message-ID: <1303257119.5447.1443075445@webmail.messagingengine.com>

I'm in the process of putting together a project that I'm expecting to
use a fair number of CPAN modules directly rather than using the system
native packages of them, for portability.  I have a few development and
test instances and a few instances in production.  I'd like to use CPAN
to manage the modules installed on each, however I want to make sure
that they stay in sync with each other and avoid potential dependency
issues should my development and production environments drift too far
apart.  Ideally I'd like to use the CPAN tools to update the production
environment only from the set of modules I've collected and tested on my
dev/test hosts.

Is there a good method for doing this?  I've discovered Local::lib from
the Catalyst book which looks promising for installing modules into
non-system locations which means I can add and update modules without
being root on the host, but I can't see how to get CPAN to only ever
look in local directories for dist files rather than fetching from the
web.

Also, if down the track I discover a need for module xxx, the current
version of which requires a later yyy than I have (ignoring issues of
whether this is a good idea) how would I go about trying to locate the
latest version of xxx that would be satisfied with the version of yyy
that I already have rather than trying to update both?  How long do old
versions of modules stick around in CPAN?

I've been able to avoid precisely this issue with pkgsrc by
pre-emptively downloading all the binary packages as they are available
in one hit, which has meant I've been able to add new packages to older
installs of hosts without issues, at the cost of downloading around 10GB
of data ... do people do similar things with CPAN?  How big is CPAN if I
were to attempt this?  Are there snapshot versions of CPAN out there
which don't change (apart from minor bug fixes)?  How long do they last
in that state?

So many questions ... :)

Regards,
Malcolm

-- 
Malcolm Herbert                                This brain intentionally
mjch at mjch.net                                                left blank


From daniel at rimspace.net  Tue Apr 19 18:08:08 2011
From: daniel at rimspace.net (Daniel Pittman)
Date: Tue, 19 Apr 2011 18:08:08 -0700
Subject: [Melbourne-pm] Presentation tool that simulates typing on a
	tty...
In-Reply-To: <BANLkTikaPrbCcFypOzNrAzYQ9BhH4gw49w@mail.gmail.com>
References: <BANLkTi=qihZsY4vyV+VDSLRPg-oWcdBEog@mail.gmail.com>
	<BANLkTikaPrbCcFypOzNrAzYQ9BhH4gw49w@mail.gmail.com>
Message-ID: <BANLkTi=5eHQQ_rrj9s-VKggo-PYzrsLn+A@mail.gmail.com>

On Tue, Apr 19, 2011 at 16:14, Damian Conway <damian at conway.org> wrote:
> Daniel Pittman asked:
>
>> So, approximately forever ago I saw a presentation given by Damian, in
>> which he used some presentation thingy that ran in a regular TTY and
>> simulated typing input and obtaining output ? typos included for free,
>> but without the risk of actually typing yourself during a
>> presentation. ?(IIRC, it actually waited for a keypress to simulate
>> each bit of the output, for extra verisimilitude.)
>
> http://search.cpan.org/~dconway/IO-Prompter-0.001001/lib/IO/Prompter.pm#Simulating_input
>
> The trick is to put a regular:
> ? ?use IO::Prompter;

Ah!  Thank you very much.

Daniel
-- 
? Puppet Labs Developer ? http://puppetlabs.com
? Daniel Pittman <daniel at rimspace.net>
? Contact me via gtalk, email, or phone: +1 (503) 893-2285
? Made with 100 percent post-consumer electrons

From toby.corkindale at strategicdata.com.au  Tue Apr 19 18:27:47 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Wed, 20 Apr 2011 11:27:47 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <1303257119.5447.1443075445@webmail.messagingengine.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>
Message-ID: <4DAE3693.304@strategicdata.com.au>

I install a custom Perl into /usr/local/$NAME/perl, and then install the 
multitude of CPAN modules in there with it using CPANPLUS.

I have a script that can do the above automatically, and can also update 
the CPAN modules.

After everything is in there, I add/update the files to a Git 
repository, then build a Debian package out of the whole lot.

Just a single, monolithic package that contains everything.

Then that gets installed on development and production machines.

If new modules are required, the process is re-run, and a new single 
package is built.. this gets tested in development alongside the next 
upcoming production release, then release to production alongside the 
next release.

This seems to work fairly well; I can't claim that the idea is mine - it 
was inspired by a talk at OSDC a couple of years ago by Adam Kennedy.

I have intentions to get this cleaned up enough to release as an 
open-source project, but it's not quite there yet.

-Toby

On 20/04/11 09:51, Malcolm Herbert wrote:
> I'm in the process of putting together a project that I'm expecting to
> use a fair number of CPAN modules directly rather than using the system
> native packages of them, for portability.  I have a few development and
> test instances and a few instances in production.  I'd like to use CPAN
> to manage the modules installed on each, however I want to make sure
> that they stay in sync with each other and avoid potential dependency
> issues should my development and production environments drift too far
> apart.  Ideally I'd like to use the CPAN tools to update the production
> environment only from the set of modules I've collected and tested on my
> dev/test hosts.
>
> Is there a good method for doing this?  I've discovered Local::lib from
> the Catalyst book which looks promising for installing modules into
> non-system locations which means I can add and update modules without
> being root on the host, but I can't see how to get CPAN to only ever
> look in local directories for dist files rather than fetching from the
> web.
>
> Also, if down the track I discover a need for module xxx, the current
> version of which requires a later yyy than I have (ignoring issues of
> whether this is a good idea) how would I go about trying to locate the
> latest version of xxx that would be satisfied with the version of yyy
> that I already have rather than trying to update both?  How long do old
> versions of modules stick around in CPAN?
>
> I've been able to avoid precisely this issue with pkgsrc by
> pre-emptively downloading all the binary packages as they are available
> in one hit, which has meant I've been able to add new packages to older
> installs of hosts without issues, at the cost of downloading around 10GB
> of data ... do people do similar things with CPAN?  How big is CPAN if I
> were to attempt this?  Are there snapshot versions of CPAN out there
> which don't change (apart from minor bug fixes)?  How long do they last
> in that state?
>
> So many questions ... :)
>
> Regards,
> Malcolm
>


-- 
.signature

From mathew.blair.robertson at gmail.com  Tue Apr 19 18:35:53 2011
From: mathew.blair.robertson at gmail.com (Mathew Robertson)
Date: Wed, 20 Apr 2011 11:35:53 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <4DAE3693.304@strategicdata.com.au>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>
	<4DAE3693.304@strategicdata.com.au>
Message-ID: <BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>

Hi Toby,

I am not familiar with how this would do the dependency checking required ->
how would building a package containing exactly the installed files, differ
from simply creating a tarball?  ie: it appears this technique completely
ignores the OS versioning problems...

regards,
Mathew

On 20 April 2011 11:27, Toby Corkindale <
toby.corkindale at strategicdata.com.au> wrote:

> I install a custom Perl into /usr/local/$NAME/perl, and then install the
> multitude of CPAN modules in there with it using CPANPLUS.
>
> I have a script that can do the above automatically, and can also update
> the CPAN modules.
>
> After everything is in there, I add/update the files to a Git repository,
> then build a Debian package out of the whole lot.
>
> Just a single, monolithic package that contains everything.
>
> Then that gets installed on development and production machines.
>
> If new modules are required, the process is re-run, and a new single
> package is built.. this gets tested in development alongside the next
> upcoming production release, then release to production alongside the next
> release.
>
> This seems to work fairly well; I can't claim that the idea is mine - it
> was inspired by a talk at OSDC a couple of years ago by Adam Kennedy.
>
> I have intentions to get this cleaned up enough to release as an
> open-source project, but it's not quite there yet.
>
> -Toby
>
>
> On 20/04/11 09:51, Malcolm Herbert wrote:
>
>> I'm in the process of putting together a project that I'm expecting to
>> use a fair number of CPAN modules directly rather than using the system
>> native packages of them, for portability.  I have a few development and
>> test instances and a few instances in production.  I'd like to use CPAN
>> to manage the modules installed on each, however I want to make sure
>> that they stay in sync with each other and avoid potential dependency
>> issues should my development and production environments drift too far
>> apart.  Ideally I'd like to use the CPAN tools to update the production
>> environment only from the set of modules I've collected and tested on my
>> dev/test hosts.
>>
>> Is there a good method for doing this?  I've discovered Local::lib from
>> the Catalyst book which looks promising for installing modules into
>> non-system locations which means I can add and update modules without
>> being root on the host, but I can't see how to get CPAN to only ever
>> look in local directories for dist files rather than fetching from the
>> web.
>>
>> Also, if down the track I discover a need for module xxx, the current
>> version of which requires a later yyy than I have (ignoring issues of
>> whether this is a good idea) how would I go about trying to locate the
>> latest version of xxx that would be satisfied with the version of yyy
>> that I already have rather than trying to update both?  How long do old
>> versions of modules stick around in CPAN?
>>
>> I've been able to avoid precisely this issue with pkgsrc by
>> pre-emptively downloading all the binary packages as they are available
>> in one hit, which has meant I've been able to add new packages to older
>> installs of hosts without issues, at the cost of downloading around 10GB
>> of data ... do people do similar things with CPAN?  How big is CPAN if I
>> were to attempt this?  Are there snapshot versions of CPAN out there
>> which don't change (apart from minor bug fixes)?  How long do they last
>> in that state?
>>
>> So many questions ... :)
>>
>> Regards,
>> Malcolm
>>
>>
>
> --
> .signature
>
> _______________________________________________
> Melbourne-pm mailing list
> Melbourne-pm at pm.org
> http://mail.pm.org/mailman/listinfo/melbourne-pm
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/melbourne-pm/attachments/20110420/35f54074/attachment.html>

From toby.corkindale at strategicdata.com.au  Tue Apr 19 18:41:51 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Wed, 20 Apr 2011 11:41:51 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>	<4DAE3693.304@strategicdata.com.au>
	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>
Message-ID: <4DAE39DF.5070800@strategicdata.com.au>

On 20/04/11 11:35, Mathew Robertson wrote:
> Hi Toby,
>
> I am not familiar with how this would do the dependency checking
> required -> how would building a package containing exactly the
> installed files, differ from simply creating a tarball?  ie: it appears
> this technique completely ignores the OS versioning problems...

My method is essentially an OS-versioned tarball.
In the Perl world, you have to accept that you're going to get the 
latest versions of modules, rather than a specific version XXX.
A modern application needs hundreds of modules once you include all the 
sub-dependencies - you would be insane to try and manually vet and 
hand-pick every one.
So my method says "take everything, latest as of now". Then you test 
this and ensure it works for you - if there are breakages (and there 
always are) then you fiddle with the packages involved until its good.

Once you have a "Stable" package, you keep using that until you need to 
upgrade, at which point you start the whole process over, until you have 
another "stable" release.

This gives you the benefits of tracking recent versions of modules from 
CPAN, without having the problems of random module versions introducing 
bugs or changes at difficult points in your development process.

Toby


> On 20 April 2011 11:27, Toby Corkindale
> <toby.corkindale at strategicdata.com.au
> <mailto:toby.corkindale at strategicdata.com.au>> wrote:
>
>     I install a custom Perl into /usr/local/$NAME/perl, and then install
>     the multitude of CPAN modules in there with it using CPANPLUS.
>
>     I have a script that can do the above automatically, and can also
>     update the CPAN modules.
>
>     After everything is in there, I add/update the files to a Git
>     repository, then build a Debian package out of the whole lot.
>
>     Just a single, monolithic package that contains everything.
>
>     Then that gets installed on development and production machines.
>
>     If new modules are required, the process is re-run, and a new single
>     package is built.. this gets tested in development alongside the
>     next upcoming production release, then release to production
>     alongside the next release.
>
>     This seems to work fairly well; I can't claim that the idea is mine
>     - it was inspired by a talk at OSDC a couple of years ago by Adam
>     Kennedy.
>
>     I have intentions to get this cleaned up enough to release as an
>     open-source project, but it's not quite there yet.
>
>     -Toby
>
>
>     On 20/04/11 09:51, Malcolm Herbert wrote:
>
>         I'm in the process of putting together a project that I'm
>         expecting to
>         use a fair number of CPAN modules directly rather than using the
>         system
>         native packages of them, for portability.  I have a few
>         development and
>         test instances and a few instances in production.  I'd like to
>         use CPAN
>         to manage the modules installed on each, however I want to make sure
>         that they stay in sync with each other and avoid potential
>         dependency
>         issues should my development and production environments drift
>         too far
>         apart.  Ideally I'd like to use the CPAN tools to update the
>         production
>         environment only from the set of modules I've collected and
>         tested on my
>         dev/test hosts.
>
>         Is there a good method for doing this?  I've discovered
>         Local::lib from
>         the Catalyst book which looks promising for installing modules into
>         non-system locations which means I can add and update modules
>         without
>         being root on the host, but I can't see how to get CPAN to only ever
>         look in local directories for dist files rather than fetching
>         from the
>         web.
>
>         Also, if down the track I discover a need for module xxx, the
>         current
>         version of which requires a later yyy than I have (ignoring
>         issues of
>         whether this is a good idea) how would I go about trying to
>         locate the
>         latest version of xxx that would be satisfied with the version
>         of yyy
>         that I already have rather than trying to update both?  How long
>         do old
>         versions of modules stick around in CPAN?
>
>         I've been able to avoid precisely this issue with pkgsrc by
>         pre-emptively downloading all the binary packages as they are
>         available
>         in one hit, which has meant I've been able to add new packages
>         to older
>         installs of hosts without issues, at the cost of downloading
>         around 10GB
>         of data ... do people do similar things with CPAN?  How big is
>         CPAN if I
>         were to attempt this?  Are there snapshot versions of CPAN out there
>         which don't change (apart from minor bug fixes)?  How long do
>         they last
>         in that state?
>
>         So many questions ... :)
>
>         Regards,
>         Malcolm
>
>
>
>     --
>     .signature
>
>     _______________________________________________
>     Melbourne-pm mailing list
>     Melbourne-pm at pm.org <mailto:Melbourne-pm at pm.org>
>     http://mail.pm.org/mailman/listinfo/melbourne-pm
>
>


-- 
.signature

From toby.corkindale at strategicdata.com.au  Tue Apr 19 19:11:03 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Wed, 20 Apr 2011 12:11:03 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <1303265033.20416.1443113201@webmail.messagingengine.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>
	<4DAE3693.304@strategicdata.com.au>
	<1303265033.20416.1443113201@webmail.messagingengine.com>
Message-ID: <4DAE40B7.6020702@strategicdata.com.au>

On 20/04/11 12:03, Malcolm Herbert wrote:
>
> On Wed, 20 Apr 2011 11:27 +1000, "Toby Corkindale"
> <toby.corkindale at strategicdata.com.au>  wrote:
>> I install a custom Perl into /usr/local/$NAME/perl, and then install the
>> multitude of CPAN modules in there with it using CPANPLUS.
>>
>> I have a script that can do the above automatically, and can also update
>> the CPAN modules.
>>
>> After everything is in there, I add/update the files to a Git
>> repository, then build a Debian package out of the whole lot.
>>
>> Just a single, monolithic package that contains everything.
>
> Interesting idea, and something I'll look into for my project ... the
> main kink I have to deal with is that I might be using Linux in
> production however my development environment so far is NetBSD - I may
> also not have root access on the production host, hence the need for a
> non-root install method ...

Does NetBSD have any virtualisation support - perhaps you could run 
Linux in a VM?
I think it's important to have a very similar setup for development and 
production, or else all kinds of little bugs can creep in - and 
discovering them only on production is not the way to do it!

> I had intended to use the system perl in each case, but now that I think
> of it it's very unlikely that base versions of the core modules would be
> exactly the same, so compiling my own perl binary for each environment
> may still be the best course ...

Cheers, I hope it works out for you.
One thing to note - I recommend manually setting a CPAN mirror, and 
setting it to the same machine everywhere you build, so as to ensure 
you're hitting the same versions of things. (Some mirrors lag by a few 
days or weeks!)

And it pays to have a proxy server set up, so that you're only 
downloading things once :)

From javier at candeira.com  Wed Apr 20 06:12:14 2011
From: javier at candeira.com (Javier Candeira)
Date: Wed, 20 Apr 2011 23:12:14 +1000
Subject: [Melbourne-pm] Perl tutor sought
Message-ID: <BANLkTikjtcx-RhcBT8w_9vWzDijt43A8zA@mail.gmail.com>

I'm looking for help understanding and debugging a website I've
inherited running Debian Sarge/Apache-Perl (Apache 1.3 with mod-perl 1
baked in)/Slash from cvs/Perl 5.8.4/Template-Toolkit and mysql of
similar vintage on a separate machine.

The help would take the form of 1-on-1 tutoring on a weekly basis,
centering our sessions on concrete tasks where I've already done some
preparatory work (and likely met a roadblock). I'm in Melbourne with
good access to the CBD, and though further sessions could be held
remotely via screen sharing and voip, at the beginning we'd need to
work side by side.

I don't have much of a budget, but I can pay tutoring rates.

If you think you can help, please write introducing yourself,
describing your availability and rates, to javier at candeira dot com.

Thanks,

Javier Candeira

From ddick at iinet.net.au  Wed Apr 20 13:48:04 2011
From: ddick at iinet.net.au (David Dick)
Date: Thu, 21 Apr 2011 06:48:04 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>	<4DAE3693.304@strategicdata.com.au>
	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>
Message-ID: <4DAF4684.8070504@iinet.net.au>

On 20/04/11 11:35, Mathew Robertson wrote:
> Hi Toby,
>
> I am not familiar with how this would do the dependency checking
> required -> how would building a package containing exactly the
> installed files, differ from simply creating a tarball?  ie: it appears
> this technique completely ignores the OS versioning problems...

well, technically since toby is creating a debian package, the 
difference is he has wrapped a tarball inside an arball :)

But the practical difference is, creating any sort of native os package 
is the first step to calculating dependencies.  Once you have a debian 
package (or a rpm, or whatever), the next step is to examine your build 
directory for dependencies and list them in the package.

For example, run 'file' on everything in the build directory, grep for 
'shared object' or 'executable', run 'objdump -p' on that, discover the 
library names that aren't in the build directory and run 'dpkg -S' on 
those libraries.  That list (plus version numbers from 'dpkg -s' if you 
want) goes straight into the control file in the debian package.

From mathew.blair.robertson at gmail.com  Wed Apr 20 15:14:02 2011
From: mathew.blair.robertson at gmail.com (Mathew Robertson)
Date: Thu, 21 Apr 2011 08:14:02 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <4DAF4684.8070504@iinet.net.au>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>
	<4DAE3693.304@strategicdata.com.au>
	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>
	<4DAF4684.8070504@iinet.net.au>
Message-ID: <BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>

On 21 April 2011 06:48, David Dick <ddick at iinet.net.au> wrote:

> On 20/04/11 11:35, Mathew Robertson wrote:
>
>> Hi Toby,
>>
>> I am not familiar with how this would do the dependency checking
>> required -> how would building a package containing exactly the
>> installed files, differ from simply creating a tarball?  ie: it appears
>> this technique completely ignores the OS versioning problems...
>>
>
> well, technically since toby is creating a debian package, the difference
> is he has wrapped a tarball inside an arball :)
>
> But the practical difference is, creating any sort of native os package is
> the first step to calculating dependencies.  Once you have a debian package
> (or a rpm, or whatever), the next step is to examine your build directory
> for dependencies and list them in the package.
>
> For example, run 'file' on everything in the build directory, grep for
> 'shared object' or 'executable', run 'objdump -p' on that, discover the
> library names that aren't in the build directory and run 'dpkg -S' on those
> libraries.  That list (plus version numbers from 'dpkg -s' if you want) goes
> straight into the control file in the debian package.
>
>
I do already understand the requirement for determining dependencies, and so
my question.... If you stick every package into an OS specific target (in
this example, /usr/local/$NAME/perl) using the packages' installer, you end
up with a list of files that are dependent on that OS... and so "how is this
different from a tarball?".

Perl has native versioning support and you can use the Makefile.PL
(Build.PL, META.yml, etc) to get your dependencies checked for you, without
having to make OS/brand specific install scripts.   [ Since this is Perl,
neither "dpkg" or "objdump", etc. are applicable **]

Toby's answer was perfectly acceptable -> the build process creates an
OS-versioned target, at the expense of being cross-platform like the actual
modules being used.

Mathew

** I'm not sure why you mentioned those... is there a specific example where
the standard Perl packaging tools dont cover some use cases?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/melbourne-pm/attachments/20110421/7b6e8234/attachment.html>

From melbourne-pm at mjch.net  Wed Apr 20 16:03:58 2011
From: melbourne-pm at mjch.net (Malcolm Herbert)
Date: Thu, 21 Apr 2011 09:03:58 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com><4DAE3693.304@strategicdata.com.au><BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com><4DAF4684.8070504@iinet.net.au>
	<BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>
Message-ID: <1303340638.16547.1443494389@webmail.messagingengine.com>

the other caveat here is that if you don't have admin privileges on the
target host, creating a package won't help you as you probably won't
have access to the tools required to install it ... 

Regards,
Malcolm

-- 
Malcolm Herbert                                This brain intentionally
mjch at mjch.net                                                left blank


From jarich at perltraining.com.au  Wed Apr 20 17:27:25 2011
From: jarich at perltraining.com.au (Jacinta Richardson)
Date: Thu, 21 Apr 2011 10:27:25 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <1303257119.5447.1443075445@webmail.messagingengine.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>
Message-ID: <4DAF79ED.4090204@perltraining.com.au>

Malcolm Herbert wrote:
> I'm in the process of putting together a project that I'm expecting to
> use a fair number of CPAN modules directly rather than using the system
> native packages of them, for portability.  I have a few development and
> test instances and a few instances in production.  I'd like to use CPAN
> to manage the modules installed on each, however I want to make sure
> that they stay in sync with each other and avoid potential dependency
> issues should my development and production environments drift too far
> apart.  Ideally I'd like to use the CPAN tools to update the production
> environment only from the set of modules I've collected and tested on my
> dev/test hosts.

An alternative solution is as follows, although this can be overkill.

CPAN::Mini allows you to create a mirror of CPAN. The latest and 
greatest of everything right now. App::cpanminus is CPAN::Mini aware, so 
you can tell it to install from your mirror rather than CPAN directly. 
Then you can use local::lib to refer to handle having everything 
installed into a local location (App::cpanminus is local::lib aware too).

One not-so-obvious gotcha is that if you're using perlbrew as well, then 
you need to install App::cpanminus for each new version of Perl you 
install with perlbrew. I'm not exactly sure why, but App::cpanminus 
forms an opinion on your version of Perl when you install it; and even 
if you then invoke it later with a different (younger) version of Perl 
in your path, it'll still baulk on version dependencies.

All the best,

Jacinta

From toby.corkindale at strategicdata.com.au  Wed Apr 20 17:51:23 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Thu, 21 Apr 2011 10:51:23 +1000
Subject: [Melbourne-pm] Still seeking OSDC AGM/next MelbPM meeting venue
In-Reply-To: <2C0FDF55-A907-4B51-88B4-775BF1441C28@meme.com.au>
References: <2C0FDF55-A907-4B51-88B4-775BF1441C28@meme.com.au>
Message-ID: <4DAF7F8B.2080805@strategicdata.com.au>

On 20/04/11 07:32, Tony Smith wrote:
> Having come from a full day of entirely unconnected meetings, I
> wasn't entirely with it by the Perl Mongers end of last Wednesday.
> After the meeting I was approached by somebody whose name escapes me
> suggesting they may be able to approach a former employer whose venue
> we had previously used and get back to me, but I have heard nothing
> further.

I believe you spoke to Ken Eley?

> Trouble is if we are to hold the OSDC AGM in conjunction with the
> next Perl Mongers meeting we need to get official notice out today.
> So if I don't get any better suggestions by early afternoon, I'm
> going to need to start trying pubs which are known to sometimes have
> suitable space. Even there I could use suggestions as I only ever get
> to a couple close enough to the city and transport.

I haven't heard of any alternate venues myself, so at the moment I 
assume MelbPM will go ahead in Fitzroy next month - but as you saw, we 
don't really have the space here for the combined attendees of both groups.

Sorry,
Toby

From keley at internode.on.net  Wed Apr 20 23:00:18 2011
From: keley at internode.on.net (Ken Eley)
Date: Thu, 21 Apr 2011 16:00:18 +1000
Subject: [Melbourne-pm] Still seeking OSDC AGM/next MelbPM meeting venue
In-Reply-To: <4DAF7F8B.2080805@strategicdata.com.au>
References: <2C0FDF55-A907-4B51-88B4-775BF1441C28@meme.com.au>
	<4DAF7F8B.2080805@strategicdata.com.au>
Message-ID: <4DAFC7F2.10400@internode.on.net>

Yes, I spoke with you after the meeting Tony.  Sorry I did not get back 
to you.  I've made initial contact with a dev where I used to work, but 
have had no response yet.  Would you like me to continue to pursue it?

Perhaps we could continue this directly, until it is resolved.

My email address is keley at internode.on.net

Ken Eley
9836 3376
0402 890 467

On 21/04/11 10:51, Toby Corkindale wrote:
> On 20/04/11 07:32, Tony Smith wrote:
>> Having come from a full day of entirely unconnected meetings, I
>> wasn't entirely with it by the Perl Mongers end of last Wednesday.
>> After the meeting I was approached by somebody whose name escapes me
>> suggesting they may be able to approach a former employer whose venue
>> we had previously used and get back to me, but I have heard nothing
>> further.
>
> I believe you spoke to Ken Eley?
>
>


From ddick at iinet.net.au  Thu Apr 21 03:13:17 2011
From: ddick at iinet.net.au (David Dick)
Date: Thu, 21 Apr 2011 20:13:17 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>	<4DAE3693.304@strategicdata.com.au>	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>	<4DAF4684.8070504@iinet.net.au>
	<BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>
Message-ID: <4DB0033D.2050605@iinet.net.au>

On 21/04/11 08:14, Mathew Robertson wrote:
>
> Perl has native versioning support and you can use the Makefile.PL
> (Build.PL, META.yml, etc) to get your dependencies checked for you,
> without having to make OS/brand specific install scripts.   [ Since this
> is Perl, neither "dpkg" or "objdump", etc. are applicable **]
>
> ** I'm not sure why you mentioned those... is there a specific example
> where the standard Perl packaging tools dont cover some use cases?
>

sure, you have built a package depending on a os library such as 
libmysqlclient.so, and you would like the native operating system to 
take care of security fixes etc, but you want to signal to the os that 
your package depends on those libraries and therefore do not allow a 
user to uninstall the library (in this case libmysqlclient.so) OR 
upgrade it to a incompatible version.

If your package is 100% pure perl, then of course this is un-necessary.

From javier at candeira.com  Fri Apr 22 00:12:45 2011
From: javier at candeira.com (Javier Candeira)
Date: Fri, 22 Apr 2011 17:12:45 +1000
Subject: [Melbourne-pm] Still seeking OSDC AGM/next MelbPM meeting venue
In-Reply-To: <4DAFC7F2.10400@internode.on.net>
References: <2C0FDF55-A907-4B51-88B4-775BF1441C28@meme.com.au>
	<4DAF7F8B.2080805@strategicdata.com.au>
	<4DAFC7F2.10400@internode.on.net>
Message-ID: <BANLkTikVQiZVEqFT0-0EbOEDgfWsb8V0Mg@mail.gmail.com>

The Melbourne Python Users Group have the use of a big classroom at
RMIT. I've asked the MPUG organisers about their contact there, and
this is what Andy Song, who teaches a course that uses Perl at RMIT,
had to say:

> It should not be a big problem to arrange a meeting room inside of RMIT.  Please let me know the details such as the date, number of attendees, requirement for audio/video equipment etc.

> I can advertise the event internally.   There are many staff/students interested in Perl.

His email address is andy.song at rmit education australia.

Cheers, Javier C

From tjc at wintrmute.net  Mon Apr 25 01:23:14 2011
From: tjc at wintrmute.net (Toby Wintermute)
Date: Mon, 25 Apr 2011 18:23:14 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <4DB0033D.2050605@iinet.net.au>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>
	<4DAE3693.304@strategicdata.com.au>
	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>
	<4DAF4684.8070504@iinet.net.au>
	<BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>
	<4DB0033D.2050605@iinet.net.au>
Message-ID: <BANLkTik0V_wR3DhH=Q-zfJXx=aDWmJsxDQ@mail.gmail.com>

On 21 April 2011 20:13, David Dick <ddick at iinet.net.au> wrote:
> On 21/04/11 08:14, Mathew Robertson wrote:
>>
>> Perl has native versioning support and you can use the Makefile.PL
>> (Build.PL, META.yml, etc) to get your dependencies checked for you,
>> without having to make OS/brand specific install scripts. ? [ Since this
>> is Perl, neither "dpkg" or "objdump", etc. are applicable **]
>>
>> ** I'm not sure why you mentioned those... is there a specific example
>> where the standard Perl packaging tools dont cover some use cases?
>>
>
> sure, you have built a package depending on a os library such as
> libmysqlclient.so, and you would like the native operating system to take
> care of security fixes etc, but you want to signal to the os that your
> package depends on those libraries and therefore do not allow a user to
> uninstall the library (in this case libmysqlclient.so) OR upgrade it to a
> incompatible version.

In my case, there is a hand-built list of all those dependencies in
the Debian package (for things like postgresql-client, mysql-client,
libssl, libgd, libjpeg, libpng, etc etc).

Toby

From ddick at iinet.net.au  Mon Apr 25 04:07:02 2011
From: ddick at iinet.net.au (David Dick)
Date: Mon, 25 Apr 2011 21:07:02 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <BANLkTik0V_wR3DhH=Q-zfJXx=aDWmJsxDQ@mail.gmail.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>	<4DAE3693.304@strategicdata.com.au>	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>	<4DAF4684.8070504@iinet.net.au>	<BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>	<4DB0033D.2050605@iinet.net.au>
	<BANLkTik0V_wR3DhH=Q-zfJXx=aDWmJsxDQ@mail.gmail.com>
Message-ID: <4DB555D6.1050400@iinet.net.au>

On 25/04/11 18:23, Toby Wintermute wrote:
> In my case, there is a hand-built list of all those dependencies in
> the Debian package (for things like postgresql-client, mysql-client,
> libssl, libgd, libjpeg, libpng, etc etc).

heh, i just checked my build process.  The debian package has got 11 
packages listed as dependencies for dynamic libraries, most definitely 
manageable for a manual process.  But i built my rpm packager first 
(served as the basis for the debian packager) which probably tainted my 
methodology... it is tracking 71 dynamic libraries in dependencies, 
that's starting to get a tad tedious to track manually. :)

From jarich at perltraining.com.au  Mon Apr 25 04:09:14 2011
From: jarich at perltraining.com.au (Jacinta Richardson)
Date: Mon, 25 Apr 2011 21:09:14 +1000
Subject: [Melbourne-pm] Still seeking OSDC AGM/next MelbPM meeting venue
In-Reply-To: <4DAF7F8B.2080805@strategicdata.com.au>
References: <2C0FDF55-A907-4B51-88B4-775BF1441C28@meme.com.au>
	<4DAF7F8B.2080805@strategicdata.com.au>
Message-ID: <4DB5565A.8080904@perltraining.com.au>

Toby Corkindale wrote:
>
> I haven't heard of any alternate venues myself, so at the moment I 
> assume MelbPM will go ahead in Fitzroy next month - but as you saw, we 
> don't really have the space here for the combined attendees of both 
> groups.
>
This is probably the best assumption. There must be an AGM announcement 
at least 21 days before the AGM, and the next meet is less than 21 days 
from now, so our meeting is no longer appropriate.

Those who are interested though, should expect an AGM announcement very 
soon though, as OSDClub Inc is legally required to hold the AGM sometime 
in May. The announcement will come with information on what to do if 
you'd like to get involved in the OSDClub committee, and may even give 
an estimation for how much work such a position might be. At the AGM 
there should be a president's and treasurer's report covering what the 
Club has been up to and I would personally love to see it also include a 
summary of what the Club hopes to achieve in the future. If I'm lucky, 
I'll even be in town whenever the AGM is held to find out!

J

From mathew.blair.robertson at gmail.com  Mon Apr 25 06:49:09 2011
From: mathew.blair.robertson at gmail.com (Mathew Robertson)
Date: Mon, 25 Apr 2011 23:49:09 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <BANLkTik0V_wR3DhH=Q-zfJXx=aDWmJsxDQ@mail.gmail.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>
	<4DAE3693.304@strategicdata.com.au>
	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>
	<4DAF4684.8070504@iinet.net.au>
	<BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>
	<4DB0033D.2050605@iinet.net.au>
	<BANLkTik0V_wR3DhH=Q-zfJXx=aDWmJsxDQ@mail.gmail.com>
Message-ID: <BANLkTinmofDS6c+iY4AvudTgE+OdBJs9pA@mail.gmail.com>

>
> >> Perl has native versioning support and you can use the Makefile.PL
> >> (Build.PL, META.yml, etc) to get your dependencies checked for you,
> >> without having to make OS/brand specific install scripts.   [ Since this
> >> is Perl, neither "dpkg" or "objdump", etc. are applicable **]
> >>
> >> ** I'm not sure why you mentioned those... is there a specific example
> >> where the standard Perl packaging tools dont cover some use cases?
> >>
> >
> > sure, you have built a package depending on a os library such as
> > libmysqlclient.so, and you would like the native operating system to take
> > care of security fixes etc, but you want to signal to the os that your
> > package depends on those libraries and therefore do not allow a user to
> > uninstall the library (in this case libmysqlclient.so) OR upgrade it to a
> > incompatible version.
>
> In my case, there is a hand-built list of all those dependencies in
> the Debian package (for things like postgresql-client, mysql-client,
> libssl, libgd, libjpeg, libpng, etc etc).
>

None of these make much sense with depending on the OS-specific versions...
you should instead be relying on the equivalent Perl/CPAN module, ie:
DBD::mysql, DBD::Pg, GD::Simple, etc.

There might be some merit in using the package manager to manage
dependencies for a core library that needs wrapping ** but a given Perl
application should only need to depend on its required Perl modules.

** Even in this case, you dont need to create OS/Vendor specific packages,
cf Perl integration with Gentoo where dependencies are tracked across the
Perl-OS boundary.

In fact the example of security updates, is exactly why you dont create your
own OS-specific package, and instead use the built-in support for
versioning.

Mathew
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.pm.org/pipermail/melbourne-pm/attachments/20110425/0ce64b51/attachment.html>

From ddick at iinet.net.au  Mon Apr 25 14:34:12 2011
From: ddick at iinet.net.au (David Dick)
Date: Tue, 26 Apr 2011 07:34:12 +1000
Subject: [Melbourne-pm] caching CPAN dist files
In-Reply-To: <BANLkTinmofDS6c+iY4AvudTgE+OdBJs9pA@mail.gmail.com>
References: <1303257119.5447.1443075445@webmail.messagingengine.com>	<4DAE3693.304@strategicdata.com.au>	<BANLkTinD3uXR7HFsevbuT-A=3gxSfCCZog@mail.gmail.com>	<4DAF4684.8070504@iinet.net.au>	<BANLkTimXfsPswX8xw6Md-8-5w-jP_aeSVA@mail.gmail.com>	<4DB0033D.2050605@iinet.net.au>	<BANLkTik0V_wR3DhH=Q-zfJXx=aDWmJsxDQ@mail.gmail.com>
	<BANLkTinmofDS6c+iY4AvudTgE+OdBJs9pA@mail.gmail.com>
Message-ID: <4DB5E8D4.3010405@iinet.net.au>

On 25/04/11 23:49, Mathew Robertson wrote:
>     In my case, there is a hand-built list of all those dependencies in
>     the Debian package (for things like postgresql-client, mysql-client,
>     libssl, libgd, libjpeg, libpng, etc etc).
>
> None of these make much sense with depending on the OS-specific
> versions... you should instead be relying on the equivalent Perl/CPAN
> module, ie: DBD::mysql, DBD::Pg, GD::Simple, etc.

I'm not quite sure what you mean.  When you build DBD::mysql, (or 
DBD::Pg, GD, etc) you create a file auto/DBD/mysql/mysql.so (or 
auto/DBD/Pg/Pg.so, auto/GD/GD.so) which is dynamically linked to 
libmysqlclient.so (or libpq.so, libgd.so, etc)

So the auto/DBD/mysql/mysql.so (or whatever) file goes into your package 
to be deployed into (for example) /opt/$company/$product/lib/, but you 
need to ensure that when the package is deployed to production/whatever, 
that libmysqlclient.so (or whatever) actually exists on the filesystem 
(for example in /usr/lib) for your library to link to. Hence, the 
dependency tracking comes in handy.

From scottp at dd.com.au  Tue Apr 26 19:48:12 2011
From: scottp at dd.com.au (Scott Penrose)
Date: Wed, 27 Apr 2011 12:48:12 +1000
Subject: [Melbourne-pm] Still seeking OSDC AGM/next MelbPM meeting venue
In-Reply-To: <4DB5565A.8080904@perltraining.com.au>
References: <2C0FDF55-A907-4B51-88B4-775BF1441C28@meme.com.au>
	<4DAF7F8B.2080805@strategicdata.com.au>
	<4DB5565A.8080904@perltraining.com.au>
Message-ID: <E5E014EB-4316-42CC-9E37-90AA0EEA45ED@dd.com.au>

Due to the time lines etc, I think it is best if we aim for June. I am talking it over with the existing exec not to see if we can come up with a better location. Since it is further away then, we might be able to encourage some other groups along and some other talks. If it is ok with you guys, can we use the Melbourne PM day, thus be a OSDClub type meeting with AGM in it.

I am looking around CBD for some other locations too. Maybe we could even talk with one of the guys at Editure, or put some money into  a back room at a pub or similar. Any suggestions. 

Scott

On 25/04/2011, at 9:09 PM, Jacinta Richardson wrote:

> Toby Corkindale wrote:
>> 
>> I haven't heard of any alternate venues myself, so at the moment I assume MelbPM will go ahead in Fitzroy next month - but as you saw, we don't really have the space here for the combined attendees of both groups.
>> 
> This is probably the best assumption. There must be an AGM announcement at least 21 days before the AGM, and the next meet is less than 21 days from now, so our meeting is no longer appropriate.
> 
> Those who are interested though, should expect an AGM announcement very soon though, as OSDClub Inc is legally required to hold the AGM sometime in May. The announcement will come with information on what to do if you'd like to get involved in the OSDClub committee, and may even give an estimation for how much work such a position might be. At the AGM there should be a president's and treasurer's report covering what the Club has been up to and I would personally love to see it also include a summary of what the Club hopes to achieve in the future. If I'm lucky, I'll even be in town whenever the AGM is held to find out!
> 
> J
> _______________________________________________
> Melbourne-pm mailing list
> Melbourne-pm at pm.org
> http://mail.pm.org/mailman/listinfo/melbourne-pm


From tconnors at astro.swin.edu.au  Tue Apr 26 23:17:03 2011
From: tconnors at astro.swin.edu.au (Tim Connors)
Date: Wed, 27 Apr 2011 16:17:03 +1000 (EST)
Subject: [Melbourne-pm] Obfuscating passwords in configurations
In-Reply-To: <AB2A216A-B10E-40A1-A5A6-A7E4833A6C38@dd.com.au>
References: <201012011650.50254.shlomif@iglu.org.il>
	<AB2A216A-B10E-40A1-A5A6-A7E4833A6C38@dd.com.au>
Message-ID: <alpine.DEB.2.00.1104271612170.23562@dirac.rather.puzzling.org>

On Thu, 2 Dec 2010, Scott Penrose wrote:

> On 02/12/2010, at 1:50 AM, Shlomi Fish wrote:
> > I didn't do it, because this doesn't actually add protection. Anyone who's
> > acquired permissions to read your rc file will be able to run fetchmail as you
> > anyway?and if it's your password they're after, they'd be able to rip the
> > necessary decoder out of the fetchmail code itself to get it.
>
> This is 100% correct, and yet completely wrong.
>
> Some real world examples are:
>
> * Fake or even Real Video cameras. They act as a deterrent
> * Deadlocks on your house, when you have windows
>
> The down side of encrypting (and really, it is just obfuscating in this case) your password is you may get a false sense of security, e.g. you might post it on the net in a forum an example of config.
>
> The upside of encrypting (obfuscating) is that it protects against accidental finding.
>
> Subversion, GIT and many other command line tools in unix obfuscate their passwords. These are mature projects who have thought about the issues. (mind you they have also covered the security too, by recommending things like SSH keys).
>
> If you were running a simple disk scan, or helping someone manage disk issue/disk space, you won't be accidentally giving away your passwords.

Or if you were to cat ~/.fetchmailrc to your screen - someone looking over
your shoulder would have a reasonable chance of remembering a password in
the 15 seconds it was on the screen, if the password isn't too complex,
whereas they would have a heck of a lot harder time remembering a base64
encoded version.  Just like GUI programs display "****", should not config
files at least go to some effort to make `cat` not display the password in
cleartext?  Just like in GUIs such as firefox, you can trivially check the
box that says "shows cleartext", you can run:

say "The decrypted password is: " . $self->config->password;

But protecting, by default, against someone looking over your shoulder as
they walk past is a worthy cause.


-- 
Tim Connors

From toby.corkindale at strategicdata.com.au  Thu Apr 28 01:32:13 2011
From: toby.corkindale at strategicdata.com.au (Toby Corkindale)
Date: Thu, 28 Apr 2011 18:32:13 +1000
Subject: [Melbourne-pm] Perl 5.14
Message-ID: <4DB9260D.8030807@strategicdata.com.au>

It is just coincidence, but once it is today in the rest of the 
inferior-timezoned world, we should be seeing releases of both Perl 5.14 
and Ubuntu 11.04!

Which will you be installing first?

-Toby

From andrew at sericyb.com.au  Thu Apr 28 08:05:58 2011
From: andrew at sericyb.com.au (Andrew Pam)
Date: Fri, 29 Apr 2011 01:05:58 +1000
Subject: [Melbourne-pm] Perl 5.14
In-Reply-To: <4DB9260D.8030807@strategicdata.com.au>
References: <4DB9260D.8030807@strategicdata.com.au>
Message-ID: <4DB98256.1060909@sericyb.com.au>

On 28/04/11 18:32, Toby Corkindale wrote:
> It is just coincidence, but once it is today in the rest of the
> inferior-timezoned world, we should be seeing releases of both Perl 5.14
> and Ubuntu 11.04!

And also Slackware 13.37!

Andrew
-- 
Andrew Pam <andrew at sericyb.com.au>
Serious Cybernetics