From newsletters at the-financial-news.com  Tue Oct  1 10:02:17 2002
From: newsletters at the-financial-news.com (The Financial News)
Date: Wed Aug  4 00:02:41 2004
Subject: Development countries. News in brief
Message-ID: <MDAEMON-F200210011705.AA055544md50020499117@wtnserver>


The Financial News, October 2002

Production Mini-plants in mobile containers. Co-investment Program

"...Science Network will supply to countries and developing regions the technology and the necessary support for the production in series of Mini-plants in mobile containers (40-foot). The Mini-
plant system is designed in such a way that all the production machinery is fixed on the platform of the container, with all wiring, piping, and installation parts; that is to say, they are fully equipped... and the mini-plant is ready for production."

More than 700 portable production systems: Bakeries, Steel Nails, Welding Electrodes, Tire Retreading, Reinforcement Bar Bending for Construction Framework, Sheeting for Roofing, Ceilings and Fa?ades, Plated Drums, Aluminum Buckets, Injected Polypropylene Housewares, Pressed Melamine Items (Glasses, Cups, Plates, Mugs, etc.), Mufflers, Construction Electrically Welded Mesh, Plastic Bags and Packaging, Mobile units of medical assistance, Sanitary Material, Hypodermic Syringes, Hemostatic Clamps, etc. 
Science Network has started a process of Co-investment for the installation of small Assembly plants to manufacture in series the Mini-plants of portable production on the site, region or country where they may be required. One of the most relevant features is the fact that these plants will be connected to the World Trade System (WTS) with access to more than 50 million raw materials, products and services and automatic transactions for world trade.

Due to financial reasons, involving cost and social impact, the right thing to do is to set up assembly plants in the same countries and regions, using local resources (labor, some equipment, etc.)

Science Network participates with 50% in the investment of each Assembly plant.

For more information: <A HREF="mailto:reports@the-financial-news.com">Mini-plants in mobile containers</A>

By Steven P. Leibacher, The Financial News, Editor


Mini-plantas de produccion en contenedores moviles. Programa de Co-inversion

"...Science Network suministrara a paises y regiones en vias de desarrollo la tecnologia y el apoyo necesario para la fabricacion en serie de Mini-plantas de produccion en contenedores moviles (40-
foot). El sistema de mini-plantas esta dise?ado de forma que todas las maquinas de produccion van instaladas fijas sobre la propia plataforma del contenedor, con el cableado, tuberias e instalaciones; es decir, completamente equipadas... y a partir de ese momento est?n listas para producir." 
Mas de 700 sistemas de produccion portatil: Panaderias, Producci?n de clavos de acero, Electrodos para soldadura, Recauchutado de neumaticos, Curvado de hierro para armaduras de construccion, Lamina perfilada para cubiertas, techos y cerramientos de fachada, Bidones de chapa, Cubos de aluminio, Menaje de polipropileno inyectado, Piezas de melamina prensada (vasos, platos, tazas, cafeteras, etc.) Silenciadores para vehiculos, Malla electrosoldada para la construccion, Bolsas y envases de plastico, Unidades moviles de asistencia medica, Material sanitario (jeringas hipodermicas, Pinzas hemostaticas, etc.)

Science Network ha puesto en marcha un proceso de Co-inversion para la instalacion de peque?as Plantas ensambladoras para fabricar en serie las Mini-plantas de produccion portatil, en el lugar, region o pais que lo necesite. Una de las caracter?sticas relevantes es el hecho de que dichas plantas quedaran conectadas al Sistema del Comercio Mundial (WTS) con acceso a mas de 50 millones de mercancias, materia primas, productos, servicios y las operaciones automaticas de comercio internacional.

Resulta obvio que por razones economicas, de costes y de impacto social, lo apropiado es instalar plantas ensambladoras en los mismos paises y regiones asi como utilizar los recursos locales (mano de obra, ciertos equipamientos, etc.)

Science Network participa al 50% en la inversion de cada Planta ensambladora.

Para recibir mas informacion: <A HREF="mailto:reports@the-financial-news.com">Mini-plantas de produccion en contenedores moviles</A>

Steven P. Leibacher, The Financial News, Editor

-------------------------------------------------------------------------
If you received this in error or would like to be removed from our list, please return us indicating: remove or un-subscribe in 'subject' field, Thanks. <A HREF="mailto:newsletters@the-financial-
news.com"> Editor</A>
? 2002 The Financial News. All rights reserved.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021001/72393d96/attachment.htm
From scottp at dd.com.au  Tue Oct  1 22:37:38 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:41 2004
Subject: where does one put use strict
Message-ID: <4C34CA1C-D5B8-11D6-8360-003065B58CF8@dd.com.au>

Hey Dudes,

I have been working on some code which inserts 'use strict' before the 
package line
eg:
	use strict;
	use warnings;

	package Blah;
	use Another;

Now what concerns me about this is that strict and warning then appears 
outside of the package, and therefore in whatever contains the use, or 
does it ?

What is the effect of putting these outside of the package line ?

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From Martin.Ryan at sensis.com.au  Tue Oct  1 23:28:28 2002
From: Martin.Ryan at sensis.com.au (Ryan, Martin)
Date: Wed Aug  4 00:02:41 2004
Subject: where does one put use strict
Message-ID: <D0C637D367F10D419D50B67AD041C7815CF652@cw-exg-srv04.pacificaccess.com.au>

Hi Scott,

The perlfunc manpage explains in general terms what "use" does before going
on to say this;

               Because this is a wide-open interface, pragmas
               (compiler directives) are also implemented this
               way.  Currently implemented pragmas are:
 
                   use integer;
                   use diagnostics;
                   use sigtrap qw(SEGV BUS);
                   use strict  qw(subs vars refs);
                   use subs    qw(afunc blurfl);
 
               Some of these these pseudo-modules import
               semantics into the current block scope (like
               strict or integer, unlike ordinary modules, which
               import symbols into the current package (which are
               effective through the end of the file).
 
My interpretation of this is that pragamas are effective from when they're
mentioned through to the end of the current block scope which in your case
is the end of the current file.  Earlier up the man page "use" is defined as
equivalent to;
       BEGIN { require Module; import Module LIST; }
So, I'd suggest that when you do a "use blah" and the blah module has a "use
strict" before the package line, you're going to "require" the module, and
the "strict" diretive will be in effect from when it's mentioned through to
the end of the file that has the blah module - blah.pm under normal
circumstances.

Cheers,
Martin

-----Original Message-----
From: Scott Penrose
To: melbourne-pm@pm.org
Sent: 2/10/02 13:37
Subject: where does one put use strict

Hey Dudes,

I have been working on some code which inserts 'use strict' before the 
package line
eg:
	use strict;
	use warnings;

	package Blah;
	use Another;

Now what concerns me about this is that strict and warning then appears 
outside of the package, and therefore in whatever contains the use, or 
does it ?

What is the effect of putting these outside of the package line ?

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.

From piers at ompa.net  Wed Oct  2 02:13:24 2002
From: piers at ompa.net (Piers Harding)
Date: Wed Aug  4 00:02:41 2004
Subject: where does one put use strict
In-Reply-To: <4C34CA1C-D5B8-11D6-8360-003065B58CF8@dd.com.au>; from scottp@dd.com.au on Wed, Oct 02, 2002 at 01:37:38PM +1000
References: <4C34CA1C-D5B8-11D6-8360-003065B58CF8@dd.com.au>
Message-ID: <20021002081324.B4038@gnu>


I believe that use strict/warnings are lexically scoped, so if you
switched on outside a package declaration in a file then the would
affect the rest of the file, until you used "no strict/warnings".
Problem is that some pragmatic modules are not lexically scoped and
there isnt much documentation about it.

Cheers.


On Wed, Oct 02, 2002 at 01:37:38PM +1000, Scott Penrose wrote:
> Hey Dudes,
> 
> I have been working on some code which inserts 'use strict' before the 
> package line
> eg:
> 	use strict;
> 	use warnings;
> 
> 	package Blah;
> 	use Another;
> 
> Now what concerns me about this is that strict and warning then appears 
> outside of the package, and therefore in whatever contains the use, or 
> does it ?
> 
> What is the effect of putting these outside of the package line ?
> 
> Scott
> -- 
> Scott Penrose
> VP in charge of Pancakes
> http://linux.dd.com.au/
> scottp@dd.com.au
> 
> Dismaimer: If you receive this email in error - please eat it 
> immediately to prevent it from falling into the wrong hands.

From scottp at dd.com.au  Wed Oct  2 03:57:18 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:42 2004
Subject: where does one put use strict
In-Reply-To: <20021002081324.B4038@gnu>
Message-ID: <F4B3D13C-D5E4-11D6-BC8B-003065B58CF8@dd.com.au>

With a little research and chatting and experimenting I have discovered 
the following.

'use X' before package puts the name space within the previous module 
(the one that called it).
However PRAGMA such as 'use strict', 'use warnings' etc are all File 
Based, specifically to do with compile time options.

That being said however, the only thing this protects is the 
possibility of a non-strict package line.

Thanks for those who replied.

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From jens at cyber.com.au  Wed Oct  2 19:38:52 2002
From: jens at cyber.com.au (Jens Porup)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
In-Reply-To: <F4B3D13C-D5E4-11D6-BC8B-003065B58CF8@dd.com.au>
References: <20021002081324.B4038@gnu> <F4B3D13C-D5E4-11D6-BC8B-003065B58CF8@dd.com.au>
Message-ID: <20021003003852.GC22604@vanilla.office.cyber.com.au>



Here's another question from some code I'm documenting:

Isn't it redundant to say 

use lib '.';

?

Doesn't a Perl program automatically have access to programs
in the current directory?

Jens



From scottp at dd.com.au  Wed Oct  2 19:48:54 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
In-Reply-To: <20021003003852.GC22604@vanilla.office.cyber.com.au>
Message-ID: <E473D11C-D669-11D6-BC8B-003065B58CF8@dd.com.au>


> Doesn't a Perl program automatically have access to programs
> in the current directory?
>
> Jens

It certainly seems to, just tried it.

Scott
-- 
Scott Penrose
Open source developer
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: Open sauce usually ends up never coming out (of the bottle).


From Adam.Clarke at StrategicData.com.au  Wed Oct  2 19:45:45 2002
From: Adam.Clarke at StrategicData.com.au (Adam Clarke)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
References: <20021002081324.B4038@gnu> <F4B3D13C-D5E4-11D6-BC8B-003065B58CF8@dd.com.au> <20021003003852.GC22604@vanilla.office.cyber.com.au>
Message-ID: <3D9B9339.8030703@StrategicData.com.au>

Seems to. Maybe it's a version thing (or just plain redundant :)

Cheers
Adam

Jens Porup wrote:

>Here's another question from some code I'm documenting:
>
>Isn't it redundant to say 
>
>use lib '.';
>
>?
>
>Doesn't a Perl program automatically have access to programs
>in the current directory?
>
>Jens
>
>
>
>  
>





From Martin.Ryan at sensis.com.au  Wed Oct  2 20:31:11 2002
From: Martin.Ryan at sensis.com.au (Ryan, Martin)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
Message-ID: <D0C637D367F10D419D50B67AD041C7815CF655@cw-exg-srv04.pacificaccess.com.au>

> > Doesn't a Perl program automatically have access to programs
> > in the current directory?
>
> It certainly seems to, just tried it.
> 

Is the context such that '.' would ever be removed from @INC?
That's the only reason I could see for "use lib '.'".
Check this out - certainly looks redundant to me...

[107]mryan@host:~> perl -e 'print join("\n", @INC), "\n"'
/usr/libdata/perl/5.00503/mach
/usr/libdata/perl/5.00503
/usr/local/lib/perl5/site_perl/5.005/i386-freebsd
/usr/local/lib/perl5/site_perl/5.005
.
[108]mryan@host:~> perl -e 'use lib "."; print join("\n", @INC), "\n"'
.
/usr/libdata/perl/5.00503/mach
/usr/libdata/perl/5.00503
/usr/local/lib/perl5/site_perl/5.005/i386-freebsd
/usr/local/lib/perl5/site_perl/5.005
.
[109]mryan@host:~>

Cheers,
Martin

From rickm at isite.net.au  Wed Oct  2 20:46:23 2002
From: rickm at isite.net.au (Rick Measham)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
In-Reply-To: 
 <D0C637D367F10D419D50B67AD041C7815CF655@cw-exg-srv04.pacificaccess.com.au>
References: 
 <D0C637D367F10D419D50B67AD041C7815CF655@cw-exg-srv04.pacificaccess.com.au>
Message-ID: <a05010402b9c1516f764e@[192.168.1.128]>

Its not redundant as Martins output proves. Perl looks through @INC 
for its modules. If I have my home-rolled Carp.pm in my local 
directory then it wont be found unless I use '.';

Here it finds the one in site_perl (or wherever it happens to be)
>[107]mryan@host:~> perl -e 'print join("\n", @INC), "\n"'
>/usr/libdata/perl/5.00503/mach
>/usr/libdata/perl/5.00503
>/usr/local/lib/perl5/site_perl/5.005/i386-freebsd
>/usr/local/lib/perl5/site_perl/5.005
>.


Here it will find the one in the local directory.
>[108]mryan@host:~> perl -e 'use lib "."; print join("\n", @INC), "\n"'
>.
>/usr/libdata/perl/5.00503/mach
>/usr/libdata/perl/5.00503
>/usr/local/lib/perl5/site_perl/5.005/i386-freebsd
>/usr/local/lib/perl5/site_perl/5.005
>.


Cheers!
Rick Measham
-- 
--------------------------------------------------------
             There are 10 kinds of people:
   those that understand binary, and those that don't.
--------------------------------------------------------
   The day Microsoft makes something that doesn't suck
     is the day they start selling vacuum cleaners
--------------------------------------------------------

From jarich at perltraining.com.au  Wed Oct  2 20:54:27 2002
From: jarich at perltraining.com.au (Jacinta Richardson)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
In-Reply-To: <D0C637D367F10D419D50B67AD041C7815CF655@cw-exg-srv04.pacificaccess.com.au>
Message-ID: <Pine.LNX.3.96.1021003114913.6765A-100000@teddybear.perltraining.com.au>


> > > Doesn't a Perl program automatically have access to programs
> > > in the current directory?

> Is the context such that '.' would ever be removed from @INC?
> That's the only reason I could see for "use lib '.'".

Nope pay attention to below.  The @INC is searched in order.

> [107]mryan@host:~> perl -e 'print join("\n", @INC), "\n"'
> /usr/libdata/perl/5.00503/mach
> /usr/libdata/perl/5.00503
> /usr/local/lib/perl5/site_perl/5.005/i386-freebsd
> /usr/local/lib/perl5/site_perl/5.005
> .

. is ONLY searched after everything else has been checked.

> [108]mryan@host:~> perl -e 'use lib "."; print join("\n", @INC), "\n"'
> .
> /usr/libdata/perl/5.00503/mach
> /usr/libdata/perl/5.00503
> /usr/local/lib/perl5/site_perl/5.005/i386-freebsd
> /usr/local/lib/perl5/site_perl/5.005
> .

Now . is checked first.
This means that if you've got your own version of CGI.pm (perhaps
because you can't get the sysadmin to upgrade theirs) and it's in the same
location as your script then your script will use your version.  Now
usually it's recommended that you created a separate directory for all
your library files, but you get the idea.

A less obvious problem that this fixes is if you call a library after a
system module you didn't know existed (for example "Config.pm").  If
you've added 
	use lib '.';
to your scripts you'll safely use YOUR Config.pm and all will be good.  If
you don't add the use lib, you'll start to wonder why your Config.pm isn't
being loaded.

	Jacinta

--
   ("`-''-/").___..--''"`-._          |  Jacinta Richardson	    |
    `6_ 6  )   `-.  (     ).`-.__.`)  |  Perl Training Australia    |
    (_Y_.)'  ._   )  `._ `. ``-..-'   |      +613 9354 6001 	    |  
  _..`--'_..-_/  /--'_.' ,'           | contact@perltraining.com.au |
(il),-''  (li),'  ((!.-'              |   www.perltraining.com.au   |


From Martin.Ryan at sensis.com.au  Wed Oct  2 21:15:29 2002
From: Martin.Ryan at sensis.com.au (Ryan, Martin)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
Message-ID: <D0C637D367F10D419D50B67AD041C7815CF656@cw-exg-srv04.pacificaccess.com.au>

> > Is the context such that '.' would ever be removed from @INC?
> > That's the only reason I could see for "use lib '.'".
> 
> Nope pay attention to below. 

[ Sits up straight with focused gaze...] :-) 

> The @INC is searched in order.

[ my output deleted ]

> Now . is checked first.

[ Excellent pair of examples where this is relevant deleted ]

Of course!  Thanks for your explanation - I found it very illuminating.

Cheers,
Martin

From max9249 at bs.ch  Thu Oct  3 16:03:35 2002
From: max9249 at bs.ch (Customer Care Center)
Date: Wed Aug  4 00:02:42 2004
Subject: Want an Investment Backed By Real Property?
Message-ID: <00002cf20954$00002a63$00000b25@gw.britishcouncil.org>

An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021003/fa38c389/attachment.htm
From martin at gleeson.com  Thu Oct  3 04:51:25 2002
From: martin at gleeson.com (Martin Gleeson)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
In-Reply-To: <3D9B9339.8030703@StrategicData.com.au>
References: <20021002081324.B4038@gnu>
 <F4B3D13C-D5E4-11D6-BC8B-003065B58CF8@dd.com.au>
 <20021003003852.GC22604@vanilla.office.cyber.com.au>
 <3D9B9339.8030703@StrategicData.com.au>
Message-ID: <p05111b01b9c1c1b33e37@[61.68.210.14]>

At 10:45 AM +1000 3/10/02, Adam Clarke wrote:
>Jens Porup wrote:
>>Here's another question from some code I'm documenting:
>>
>>Isn't it redundant to say
>>use lib '.';
>>
>>Doesn't a Perl program automatically have access to programs
>>in the current directory?
>>
>Seems to. Maybe it's a version thing (or just plain redundant :)
>
>Cheers
>Adam

Hi Folks,

@INC comes from the environment variable PERLLIB, which has a default
value that is set when perl is compiled, so you may or may not have
'.' in it by default. It would depend on your sysadmin and whether or
not they simply hit 'enter' in response to the compile questions.
You can add to PERLLIB in your shell rc file if you want to automatically
include your own lib directory and/or '.'.

As an aside, having '.' in any search path is seen as a Bad Thing(tm)
from a security point of view. People have been caught out by someone's
own version of 'ls' that they've put in their home directory on a shared
machine. The same thing would apply to their own version of 'strict.pm',
for example. Let your imagination wander...

Cheers,
Marty.
-- 
Martin Gleeson
http://martin.gleeson.com/

From jens at cyber.com.au  Thu Oct  3 17:01:21 2002
From: jens at cyber.com.au (Jens Porup)
Date: Wed Aug  4 00:02:42 2004
Subject: use lib '.';
In-Reply-To: <D0C637D367F10D419D50B67AD041C7815CF656@cw-exg-srv04.pacificaccess.com.au>
References: <D0C637D367F10D419D50B67AD041C7815CF656@cw-exg-srv04.pacificaccess.com.au>
Message-ID: <20021003220121.GA4229@vanilla.office.cyber.com.au>



Here's another question on this thread:

what is the difference between:

use lib '.';

	and...

use '.';

?

Very interesting discussion so far....

Jens




From scottp at dd.com.au  Sun Oct  6 17:53:52 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:42 2004
Subject: Next Weeks Meeting
Message-ID: <7BFA5019-D97E-11D6-975A-003065B58CF8@dd.com.au>

Hey Dudes,

Unfortunately Damian has not contacted me back about next week, so I 
will assume he is probably doing conferences and generally not 
available. I will try him again for next month.

Therefore we are looking for talks for this week.

Currently I have some volunteers, but if you have anything else, small, 
medium or large, please email me :-)

What is your favorite module. Why do you use it. What can you not live 
without.
A talk can be as simple as a 10 minute presentation of a killer module 
you just can't do without.
Use this as an opportunity to get practice talking.

* Spreadsheet::WriteExcel
* Activestate's Perl Development Kit (specifically, the PerlApp tool)
* Class::Multimethod
* Hardware::iButtons

I also have a short demonstration of Dallas iButtons.

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From jh_lists at fastmail.fm  Sun Oct  6 18:08:09 2002
From: jh_lists at fastmail.fm (JP Howard)
Date: Wed Aug  4 00:02:42 2004
Subject: Next Weeks Meeting
Message-ID: <20021006230809.AAE3593707@server2.fastmail.fm>

On Mon, 7 Oct 2002 08:53:52 +1000, "Scott Penrose" <scottp@dd.com.au>
said:
> Currently I have some volunteers, but if you have anything else, small, 
> medium or large, please email me :-)
> 
> What is your favorite module. Why do you use it. What can you not live 
> without.
> A talk can be as simple as a 10 minute presentation of a killer module 
> you just can't do without.

I can do 15 minutes on Power Vim For Perlers. Vim is a really great
editor for Perl code, but it takes some tweaking...

From Adam.Clarke at StrategicData.com.au  Sun Oct  6 18:51:47 2002
From: Adam.Clarke at StrategicData.com.au (Adam Clarke)
Date: Wed Aug  4 00:02:42 2004
Subject: Next Weeks Meeting
References: <20021006230809.AAE3593707@server2.fastmail.fm>
Message-ID: <3DA0CC93.3050903@StrategicData.com.au>

I would love to hear that one. Got my vote.

Cheers
Adam

JP Howard wrote:

>
>I can do 15 minutes on Power Vim For Perlers. Vim is a really great
>editor for Perl code, but it takes some tweaking...
>
>  
>



From gblist at excite.com  Mon Oct  7 08:53:05 2002
From: gblist at excite.com (Greg)
Date: Wed Aug  4 00:02:42 2004
Subject: Perl Tk on Redhat 8.0
Message-ID: <20021007135305.1AAA73DC9@xmxpita.excite.com>

 
Hi there

I am trying to install Perl-Tk on Redhat 8.0.  I downloaded perl-Tk-800.024-1.src.rpm from rpmfind.net but it fails to install as per below.  The graphical installer also bails midway through an attempted install with no message. any help appreciated

Greg


[root@Workstation root]# rpm -iv perl-Tk-800.024-1.src.rpm
warning: perl-Tk-800.024-1.src.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
warning: user builder does not exist - using root
warning: user builder does not exist - using root
warning: user builder does not exist - using root
warning: user builder does not exist - using root
warning: user builder does not exist - using root
perl-Tk-800.024-1




------------------------------------------------
Changed your e-mail?  Keep your contacts!  Use this free e-mail change of address service from Return Path.  Register now!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021007/acf68fa7/attachment.htm
From gcross at alphalink.com.au  Mon Oct  7 18:22:18 2002
From: gcross at alphalink.com.au (Graeme Cross)
Date: Wed Aug  4 00:02:42 2004
Subject: Perl Tk on Redhat 8.0
In-Reply-To: <20021007135305.1AAA73DC9@xmxpita.excite.com>
References: <20021007135305.1AAA73DC9@xmxpita.excite.com>
Message-ID: <200210080922.18277.gcross@alphalink.com.au>

On Mon, 7 Oct 2002 11:53 pm, Greg wrote:
>
> I am trying to install Perl-Tk on Redhat 8.0.  I downloaded
> perl-Tk-800.024-1.src.rpm from rpmfind.net but it fails to install as per
> below.  The graphical installer also bails midway through an attempted
> install with no message. any help appreciated
>
> [root@Workstation root]# rpm -iv perl-Tk-800.024-1.src.rpm
> warning: perl-Tk-800.024-1.src.rpm: V3 DSA signature: NOKEY, key ID
> f9651d5a warning: user builder does not exist - using root
> warning: user builder does not exist - using root
> warning: user builder does not exist - using root
> warning: user builder does not exist - using root
> warning: user builder does not exist - using root
> perl-Tk-800.024-1
>

An odd error message. 

Are you sure that it hasn't installed anything? Have a look in the 
subdirectories in /usr/src/redhat/ -- has it installed a tarball, SPEC file 
or patches?

- Graeme

-- 
Graeme Cross <gcross@alphalink.com.au>


From jens at cyber.com.au  Tue Oct  8 00:27:59 2002
From: jens at cyber.com.au (Jens Porup)
Date: Wed Aug  4 00:02:42 2004
Subject: Damian speaking in November...
In-Reply-To: <20021007135305.1AAA73DC9@xmxpita.excite.com>
References: <20021007135305.1AAA73DC9@xmxpita.excite.com>
Message-ID: <20021008052759.GC18198@vanilla.office.cyber.com.au>


By the looks of things:

	http://www.yetanother.org/damian/diary_latest.html

Damian may not be especially eager to speak for us in November.

	"That means that for the rest of 2002 I will not be reading or
	contributing to Perl mailing lists or newsgroups, will not be
	maintaining my numerous CPAN modules, will not be travelling or
	speaking, and will not even be regularly reading my Perl-
	specific) email. In fact, the only Perl-related activities
	I'll still be undertaking will be those related to the Perl 6
	design work (that is, I'll continue to help Larry and the design
	team, and to write the Exegeses for the Apocalypses)."



From jarich at perltraining.com.au  Tue Oct  8 02:22:55 2002
From: jarich at perltraining.com.au (Jacinta Richardson)
Date: Wed Aug  4 00:02:42 2004
Subject: Tomorrrow Night?
Message-ID: <Pine.LNX.3.96.1021008172224.1114A-100000@teddybear.perltraining.com.au>


What's scheduled?

	Jacinta

--
   ("`-''-/").___..--''"`-._          |  Jacinta Richardson	    |
    `6_ 6  )   `-.  (     ).`-.__.`)  |  Perl Training Australia    |
    (_Y_.)'  ._   )  `._ `. ``-..-'   |      +613 9354 6001 	    |  
  _..`--'_..-_/  /--'_.' ,'           | contact@perltraining.com.au |
(il),-''  (li),'  ((!.-'              |   www.perltraining.com.au   |


From martin at gleeson.com  Mon Oct  7 17:23:28 2002
From: martin at gleeson.com (Martin Gleeson)
Date: Wed Aug  4 00:02:42 2004
Subject: The Perl Journal
Message-ID: <p05111b03b9c7b929eb74@[210.10.135.180]>

Hi Folks,

The Perl Journal wants to be resurrected, but needs subscribers. They've
got almost 900, but need 3000 to be viable. I subscribed to the original
TPJ from the beginning, and highly recommend subscribing.

http://www.tpj.com/

Cheers,
Marty.
-- 
Martin Gleeson
http://martin.gleeson.com/

From hock at AKH-WIEN.AC.AT  Tue Oct  8 16:58:58 2002
From: hock at AKH-WIEN.AC.AT (Sales)
Date: Wed Aug  4 00:02:42 2004
Subject: Have you heard of a no risk investment?
Message-ID: <00004e1a10c0$00007aa1$000053a5@mailgatepi.smals-mvm.be>

An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021008/35d7601e/attachment.htm
From gblist at excite.com  Tue Oct  8 06:35:24 2002
From: gblist at excite.com (Greg)
Date: Wed Aug  4 00:02:42 2004
Subject: Perl Tk on Redhat 8.0
Message-ID: <20021008113524.2DFC3133C8@xmxpita.excite.com>

 Graham

Thanks for your suggestion. The following files exist in SOURCES and SPECS, but the RPM, SRPM and SOURCES directories are empty.

Greg

[root@Workstation SOURCES]# ls
perl-Tk-man_section.patch  perl-Tk-paths.patch
perl-Tk-misc.patch         Tk-800.024.tar.gz
[root@Workstation SOURCES]#
[root@Workstation SPECS]# ls
perl-Tk.spec
[root@Workstation SPECS]#







Hi there

I am trying to install Perl-Tk on Redhat 8.0. I downloaded perl-Tk-800.024-1.src.rpm from rpmfind.net but it fails to install as per 
>below. The graphical installer also bails midway through an attempted install with no message. any help appreciated

Greg


[root@Workstation root]# rpm -iv perl-Tk-800.024-1.src.rpm
warning: perl-Tk-800.024-1.src.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
warning: user builder does not exist - using root
warning: user builder does not exist - using root
warning: user builder does not exist - using root
warning: user builder does not exist - using root
warning: user builder does not exist - using root
perl-Tk-800.024-1


------------------------------------------------
Changed your e-mail?  Keep your contacts!  Use this free e-mail change of address service from Return Path.  Register now!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021008/9bab1e33/attachment.htm
From gcross at alphalink.com.au  Tue Oct  8 09:13:00 2002
From: gcross at alphalink.com.au (Graeme Cross)
Date: Wed Aug  4 00:02:42 2004
Subject: Perl Tk on Redhat 8.0
In-Reply-To: <20021008113524.2DFC3133C8@xmxpita.excite.com>
References: <20021008113524.2DFC3133C8@xmxpita.excite.com>
Message-ID: <200210090013.00955.gcross@alphalink.com.au>

On Tue, 8 Oct 2002 09:35 pm, Greg wrote:
>  Graham
>
> Thanks for your suggestion. The following files exist in SOURCES and SPECS,
> but the RPM, SRPM and SOURCES directories are empty.
>
> Greg
>
> [root@Workstation SOURCES]# ls
> perl-Tk-man_section.patch  perl-Tk-paths.patch
> perl-Tk-misc.patch         Tk-800.024.tar.gz
> [root@Workstation SOURCES]#
> [root@Workstation SPECS]# ls
> perl-Tk.spec
> [root@Workstation SPECS]#
>
>
> I am trying to install Perl-Tk on Redhat 8.0. I downloaded
> perl-Tk-800.024-1.src.rpm from rpmfind.net but it fails to install as per
>
> >below. The graphical installer also bails midway through an attempted
> > install with no message. any help appreciated
>

Greg:

RPM packages come in two flavours: the common binary RPM (which you install 
with a "rpm -ivh" command) and a source RPM, which contains the original 
source, patches for your Red Hat/Mandrake/SuSe distribution and a SPEC file 
which has all the information on how to build the binary RPM package(s).

You have tried to install a source RPM rather than a binary RPM, so you have 
to build the binary RPM(s) from the spec file and sources.

>From memory[1], it's something like:

	cd /usr/src/redhat
	rpm -bb SPECS/perl-Tk.spec

You should then end up with at least one RPM in the /usr/src/redhat/RPMS 
directory, which you then install with the usual rpm -ivh command.

If you can't get it working and it's on a laptop, bring it to the meeting 
tomorrow night and I will give you a hand with it.

Cheers
Graeme

[1] Have a read of the rpm manpage for the precise details and options

-- 
Graeme Cross <gcross@alphalink.com.au>


From scottp at dd.com.au  Tue Oct  8 18:40:01 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:42 2004
Subject: Tonight - Meeting
Message-ID: <435ADCF4-DB17-11D6-975A-003065B58CF8@dd.com.au>

Hey ho,

The perl meeting is tonight, not next week. I wrote the 'what is on 
next week' last week, but forgot to push send, sorry to those I have 
confused.

The schedule for the meeting tonight is as follows:

	- Advanced VIM for Perl - JP Howard

	- More Advanced VIM for Perl - Scott Penrose

	- Spreadsheet::WriteExcel - Graeme Cross

	- Hardware::iButtons - Scott Penrose

When:
	6:30pm Wednesday 9th - Today !
	Doors open from 6:00 to 6:45
	
Where:
	myinternet
	Level 8, 14 Blackwood Street
	North Melbourne

Scott
-- 
Scott Penrose
Welcome to the Digital Dimension
http://www.dd.com.au/
scottp@dd.com.au

Dismaimer: Contents of this mail and signature are bound to change 
randomly. Whilst every attempt has been made to control said 
randomness, the author wishes to remain blameless for the number of 
eggs that damn chicken laid. Oh and I don't want to hear about 
butterflies either.


From scottp at dd.com.au  Tue Oct  8 18:42:05 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:42 2004
Subject: Tonight - Meeting
In-Reply-To: <435ADCF4-DB17-11D6-975A-003065B58CF8@dd.com.au>
Message-ID: <8D55DA5E-DB17-11D6-975A-003065B58CF8@dd.com.au>

I have added another short talk about JEdit and perl.
>
> 	- Advanced VIM for Perl - JP Howard
>
> 	- More Advanced VIM for Perl - Scott Penrose
>
> 	- JEdit for Perl - Scott Penrose
>
> 	- Spreadsheet::WriteExcel - Graeme Cross
>
> 	- Hardware::iButtons - Scott Penrose
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From hock at AKH-WIEN.AC.AT  Fri Oct 11 02:44:19 2002
From: hock at AKH-WIEN.AC.AT (US Telecenter)
Date: Wed Aug  4 00:02:42 2004
Subject: Crystal Clear Conference Calls
Message-ID: <000054223c47$000064be$00002449@mailgatepi.smals-mvm.be>

An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021010/be9a5f1f/attachment.htm
From scottp at dd.com.au  Sat Oct 12 04:51:04 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:42 2004
Subject: Generic File Access
Message-ID: <1F834FFC-DDC8-11D6-9CDE-003065B58CF8@dd.com.au>

Hey Dudes

I have a major need for some generic file access API in Perl.
A generic access API would have the one polymorphic interface to SSH 
(SCP), FTP, WebDAV, Local etc.

The following are the options I have.
	- XS mapping to Gnome VFS (Gnome's virtual file system)
	- XS mapping to CURL (Curl is a set of libs and command lines to do 
this)
	- Write a higher level API (DBI style) to do this.

My problems (and advantages) with each are as follows.

VFS
===
	- Very complicated
	- Very big
	- Good as it is so complete
	- Lots of call backs

Curl
====
	- Behind in crucial implementations (SCP/SSH is still not supported)
	- Simpler

Roll my own
===========
	- Could be simple, depending on level
	- I have written a basic implementation but as soon as I
	  read the VFS functionality I realise that others could
	  require much more than I could provide

So I have to think about what the best approach is. I was looking for 
either some sensible feedback, how other people have dealt with the 
situation and if any other solutions are out there.

Any ideas ? Feedback ? Other alternatives ?

:-)

Scott
-- 
Scott Penrose
Open source developer
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: Open sauce usually ends up never coming out (of the bottle).


From piers at ompa.net  Sun Oct 13 01:29:48 2002
From: piers at ompa.net (Piers Harding)
Date: Wed Aug  4 00:02:42 2004
Subject: Generic File Access
In-Reply-To: <1F834FFC-DDC8-11D6-9CDE-003065B58CF8@dd.com.au>; from scottp@dd.com.au on Sat, Oct 12, 2002 at 07:51:04PM +1000
References: <1F834FFC-DDC8-11D6-9CDE-003065B58CF8@dd.com.au>
Message-ID: <20021013072948.A18890@gnu>

Sounds like VFS is the one - trouble is that it would be a lot of XS
work.

Inline has an AUTOWRAP facility to generate the attaching stubs ( in
Inline::C ) - it may help out here?

Cheers.



On Sat, Oct 12, 2002 at 07:51:04PM +1000, Scott Penrose wrote:
> Hey Dudes
> 
> I have a major need for some generic file access API in Perl.
> A generic access API would have the one polymorphic interface to SSH 
> (SCP), FTP, WebDAV, Local etc.
> 
> The following are the options I have.
> 	- XS mapping to Gnome VFS (Gnome's virtual file system)
> 	- XS mapping to CURL (Curl is a set of libs and command lines to do 
> this)
> 	- Write a higher level API (DBI style) to do this.
> 
> My problems (and advantages) with each are as follows.
> 
> VFS
> ===
> 	- Very complicated
> 	- Very big
> 	- Good as it is so complete
> 	- Lots of call backs
> 
> Curl
> ====
> 	- Behind in crucial implementations (SCP/SSH is still not supported)
> 	- Simpler
> 
> Roll my own
> ===========
> 	- Could be simple, depending on level
> 	- I have written a basic implementation but as soon as I
> 	  read the VFS functionality I realise that others could
> 	  require much more than I could provide
> 
> So I have to think about what the best approach is. I was looking for 
> either some sensible feedback, how other people have dealt with the 
> situation and if any other solutions are out there.
> 
> Any ideas ? Feedback ? Other alternatives ?
> 
> :-)
> 
> Scott
> -- 
> Scott Penrose
> Open source developer
> http://linux.dd.com.au/
> scottp@dd.com.au
> 
> Dismaimer: Open sauce usually ends up never coming out (of the bottle).

From scottp at myinternet.com.au  Sun Oct 13 17:39:41 2002
From: scottp at myinternet.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:42 2004
Subject: November Talk - Perl 6 overview - Damian Conway
Message-ID: <A9DB202C-DEFC-11D6-9321-003065B58CF8@myinternet.com.au>

Hey dudes,

This is just an early heads up for the November Talk.

	Date:		Wednesday November 13th
	Time:		6pm (NOTE - 6pm, not 6:30) (doors open till 6:15 only)
	Place:		myinternet, Level 8, 14 Blackwood Street, North Melbourne
	Topic:		Perl 6 Overview
	Presenter:	Damian Conway
	Length:		3 hours

More details to come.

Scott
-- 
Scott Penrose
Open source developer
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: Open sauce usually ends up never coming out (of the bottle).



From D.Feingold at century-management.ie  Wed Oct 16 12:13:45 2002
From: D.Feingold at century-management.ie (E-Business Update)
Date: Wed Aug  4 00:02:42 2004
Subject: Conference calls/best quality/$.18 per minute!
Message-ID: <000060da6899$0000173f$00002c04@mx1.skima.is>

An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021015/9fee7000/attachment.htm
From Adam.Clarke at StrategicData.com.au  Wed Oct 16 21:42:42 2002
From: Adam.Clarke at StrategicData.com.au (Adam Clarke)
Date: Wed Aug  4 00:02:42 2004
Subject: Cross Platform Perl
Message-ID: <3DAE23A2.5070604@StrategicData.com.au>

Hi all,

I'm trying to write a script that runs on Linux (Unix) and Win32. I want 
to /use/ a module when on Win32 (to access the registry) that doesn't 
exist in Unixy perl. Is there a way to skip a use statement at runtime 
based on platform or is the only/best way to make a full build 
(MakeMaker thing) and handle the differences there somehow. If the 
latter, anyone got some simple pointers.

Cheers
Adam




From pjf at perltraining.com.au  Wed Oct 16 22:01:33 2002
From: pjf at perltraining.com.au (Paul Fenwick)
Date: Wed Aug  4 00:02:42 2004
Subject: Cross Platform Perl
In-Reply-To: <3DAE23A2.5070604@StrategicData.com.au>
References: <3DAE23A2.5070604@StrategicData.com.au>
Message-ID: <20021017030133.GD27913@mukc.org.au>

G'day Adam,

On Thu, Oct 17, 2002 at 12:42:42PM +1000, Adam Clarke wrote:

> I'm trying to write a script that runs on Linux (Unix) and Win32. I want 
> to /use/ a module when on Win32 (to access the registry) that doesn't 
> exist in Unixy perl. Is there a way to skip a use statement at runtime 
> based on platform or is the only/best way to make a full build 
> (MakeMaker thing) and handle the differences there somehow. If the 
> latter, anyone got some simple pointers.

Glad you asked!  Perl has a variable called $^O (or $OSNAME when
using English) that gives you the name of the operating system
you're running under.

So you can do the following:

BEGIN {		# So we run at the same time as most use statements.
	if ($^O eq 'MSWin32') {
		eval { use My::Module; }
	}
}

The eval is needed because usually use statements get executed
immediately, regardless of context.  The BEGIN means that your
chunk of code is executed "at compile time".  If this isn't important,
you can leave the BEGIN block out.

There's lots of code out there which does different things depending
upon the OS.  Take a look at File::Copy's source code
(perldoc -m File::Copy) to see a lot of this in action.

Cheers,

	Paul

-- 
Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
Director of Training                   | Ph:  +61 3 9354 6001
Perl Training Australia                | Fax: +61 3 9354 2681

From Adam.Clarke at StrategicData.com.au  Wed Oct 16 22:21:39 2002
From: Adam.Clarke at StrategicData.com.au (Adam Clarke)
Date: Wed Aug  4 00:02:42 2004
Subject: Cross Platform Perl
References: <3DAE23A2.5070604@StrategicData.com.au> <20021017030133.GD27913@mukc.org.au>
Message-ID: <3DAE2CC3.2060504@StrategicData.com.au>

One question / clarification (based on my attempt to get this to work).

Shouldn't the below acrually read

    eval 'use My::Module;' ;

instead of the {}'s coz I kept getting the same problem as beffore the 
other way?

Paul Fenwick wrote:
 > BEGIN {        # So we run at the same time as most use statements.
 >    if ($^O eq 'MSWin32') {
 >        eval { use My::Module; }
 >    }
 > }




From rob at cowsnet.com.au  Wed Oct 16 22:23:34 2002
From: rob at cowsnet.com.au (Rob Casey)
Date: Wed Aug  4 00:02:42 2004
Subject: Cross Platform Perl
In-Reply-To: <3DAE23A2.5070604@StrategicData.com.au>
Message-ID: <004e01c2758c$949f1500$38662fcb@hewison.com.au>

As a starting point ... This is how I would do it, making use of the
eval statement with a string argument.

eval "use Win32::Registry" if $^O =~ /Win32/;
die 'Cannot load Win32::Registry module - ', $@ if $@;

Regards,
Rob


Rob Casey
Business Manager, Senior IT Consultant
Cowsnet Internet and Professional Services
http://www.cowsnet.com.au

 

-----Original Message-----
From: owner-melbourne-pm@pm.org [mailto:owner-melbourne-pm@pm.org] On
Behalf Of Adam Clarke
Sent: Thursday, 17 October 2002 12:43 PM
To: melbourne-pm@pm.org
Subject: Cross Platform Perl


Hi all,

I'm trying to write a script that runs on Linux (Unix) and Win32. I want

to /use/ a module when on Win32 (to access the registry) that doesn't 
exist in Unixy perl. Is there a way to skip a use statement at runtime 
based on platform or is the only/best way to make a full build 
(MakeMaker thing) and handle the differences there somehow. If the 
latter, anyone got some simple pointers.

Cheers
Adam



-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3222 bytes
Desc: not available
Url : http://mail.pm.org/archives/melbourne-pm/attachments/20021017/b2009de0/smime.bin
From rob at cowsnet.com.au  Wed Oct 16 22:27:37 2002
From: rob at cowsnet.com.au (Rob Casey)
Date: Wed Aug  4 00:02:42 2004
Subject: Cross Platform Perl
In-Reply-To: <20021017030133.GD27913@mukc.org.au>
Message-ID: <005601c2758d$2586b960$38662fcb@hewison.com.au>

Just a follow up to this ...

> BEGIN {		# So we run at the same time as most use
statements.
> 	if ($^O eq 'MSWin32') {
> 		eval { use My::Module; }
> 	}
> }

This will fail - As the compiler will find this use statement and
attempt to include the module at this point.  To make use of dynamic
loading of modules, the eval statement must called with a string
argument as this string is not interpreted at compile time, but at the
time of execution.

Regards,
Rob


Rob Casey
Business Manager, Senior IT Consultant
Cowsnet Internet and Professional Services
http://www.cowsnet.com.au

 

-----Original Message-----
From: owner-melbourne-pm@pm.org [mailto:owner-melbourne-pm@pm.org] On
Behalf Of Paul Fenwick
Sent: Thursday, 17 October 2002 1:02 PM
To: Adam Clarke
Cc: melbourne-pm@pm.org
Subject: Re: Cross Platform Perl


G'day Adam,

On Thu, Oct 17, 2002 at 12:42:42PM +1000, Adam Clarke wrote:

> I'm trying to write a script that runs on Linux (Unix) and Win32. I
want 
> to /use/ a module when on Win32 (to access the registry) that doesn't 
> exist in Unixy perl. Is there a way to skip a use statement at runtime

> based on platform or is the only/best way to make a full build 
> (MakeMaker thing) and handle the differences there somehow. If the 
> latter, anyone got some simple pointers.

Glad you asked!  Perl has a variable called $^O (or $OSNAME when
using English) that gives you the name of the operating system
you're running under.

So you can do the following:

BEGIN {		# So we run at the same time as most use statements.
	if ($^O eq 'MSWin32') {
		eval { use My::Module; }
	}
}

The eval is needed because usually use statements get executed
immediately, regardless of context.  The BEGIN means that your
chunk of code is executed "at compile time".  If this isn't important,
you can leave the BEGIN block out.

There's lots of code out there which does different things depending
upon the OS.  Take a look at File::Copy's source code
(perldoc -m File::Copy) to see a lot of this in action.

Cheers,

	Paul

-- 
Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
Director of Training                   | Ph:  +61 3 9354 6001
Perl Training Australia                | Fax: +61 3 9354 2681

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3222 bytes
Desc: not available
Url : http://mail.pm.org/archives/melbourne-pm/attachments/20021017/5ffe2565/smime.bin
From pjf at perltraining.com.au  Wed Oct 16 22:37:36 2002
From: pjf at perltraining.com.au (Paul Fenwick)
Date: Wed Aug  4 00:02:43 2004
Subject: Cross Platform Perl
In-Reply-To: <005601c2758d$2586b960$38662fcb@hewison.com.au>
References: <20021017030133.GD27913@mukc.org.au> <005601c2758d$2586b960$38662fcb@hewison.com.au>
Message-ID: <20021017033736.GF27913@mukc.org.au>

G'day Everyone,

	Rob and Adam are indeed correct.  You do need a string
eval and not a block eval.  My mistake.  :)

	Cheers,

		Paul

On Thu, Oct 17, 2002 at 01:27:37PM +1000, Rob Casey wrote:
> Just a follow up to this ...
> 
> > BEGIN {		# So we run at the same time as most use
> statements.
> > 	if ($^O eq 'MSWin32') {
> > 		eval { use My::Module; }
> > 	}
> > }
> 
> This will fail - As the compiler will find this use statement and
> attempt to include the module at this point.  To make use of dynamic
> loading of modules, the eval statement must called with a string
> argument as this string is not interpreted at compile time, but at the
> time of execution.
> 
> Regards,
> Rob
> 
> 
> Rob Casey
> Business Manager, Senior IT Consultant
> Cowsnet Internet and Professional Services
> http://www.cowsnet.com.au
> 
>  
> 
> -----Original Message-----
> From: owner-melbourne-pm@pm.org [mailto:owner-melbourne-pm@pm.org] On
> Behalf Of Paul Fenwick
> Sent: Thursday, 17 October 2002 1:02 PM
> To: Adam Clarke
> Cc: melbourne-pm@pm.org
> Subject: Re: Cross Platform Perl
> 
> 
> G'day Adam,
> 
> On Thu, Oct 17, 2002 at 12:42:42PM +1000, Adam Clarke wrote:
> 
> > I'm trying to write a script that runs on Linux (Unix) and Win32. I
> want 
> > to /use/ a module when on Win32 (to access the registry) that doesn't 
> > exist in Unixy perl. Is there a way to skip a use statement at runtime
> 
> > based on platform or is the only/best way to make a full build 
> > (MakeMaker thing) and handle the differences there somehow. If the 
> > latter, anyone got some simple pointers.
> 
> Glad you asked!  Perl has a variable called $^O (or $OSNAME when
> using English) that gives you the name of the operating system
> you're running under.
> 
> So you can do the following:
> 
> BEGIN {		# So we run at the same time as most use statements.
> 	if ($^O eq 'MSWin32') {
> 		eval { use My::Module; }
> 	}
> }
> 
> The eval is needed because usually use statements get executed
> immediately, regardless of context.  The BEGIN means that your
> chunk of code is executed "at compile time".  If this isn't important,
> you can leave the BEGIN block out.
> 
> There's lots of code out there which does different things depending
> upon the OS.  Take a look at File::Copy's source code
> (perldoc -m File::Copy) to see a lot of this in action.
> 
> Cheers,
> 
> 	Paul
> 
> -- 
> Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
> Director of Training                   | Ph:  +61 3 9354 6001
> Perl Training Australia                | Fax: +61 3 9354 2681
> 



-- 
Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
Director of Training                   | Ph:  +61 3 9354 6001
Perl Training Australia                | Fax: +61 3 9354 2681

From scottp at dd.com.au  Wed Oct 16 22:40:21 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Cross Platform Perl
In-Reply-To: <3DAE2CC3.2060504@StrategicData.com.au>
Message-ID: <29B70508-E182-11D6-A5C6-003065B58CF8@dd.com.au>


On Thursday, Oct 17, 2002, at 13:21 Australia/Melbourne, Adam Clarke 
wrote:

> One question / clarification (based on my attempt to get this to work).
>
> Shouldn't the below acrually read
>
>    eval 'use My::Module;' ;

Yes absolutely.

All 'use' are loaded automatically at compile time, EVEN if inside an 
eval {} or BEGIN {} block.
So you have no choice but to do a run time compile by using an eval 
SCALAR.

Scott

> instead of the {}'s coz I kept getting the same problem as beffore the 
> other way?
>
> Paul Fenwick wrote:
> > BEGIN {        # So we run at the same time as most use statements.
> >    if ($^O eq 'MSWin32') {
> >        eval { use My::Module; }
> >    }
> > }
>
>
>
>
>
>
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From rickm at isite.net.au  Wed Oct 16 22:53:16 2002
From: rickm at isite.net.au (rickm@isite.net.au)
Date: Wed Aug  4 00:02:43 2004
Subject: Cross Platform Perl
In-Reply-To: <3DAE23A2.5070604@StrategicData.com.au>
References: <3DAE23A2.5070604@StrategicData.com.au>
Message-ID: <1034826796.3dae342c167e1@webmail.isite.net.au>

G'day Adam,
>From the POD you can replace a 'use' with a 'require' and 'import' inside a 
BEGIN block. Thus you can also add an 'if' in there:

BEGIN {
   if ($platform eq 'Win32') {
      require Module; import Module LIST;
   }
}

Cheers!
Rick Measham

Quoting Adam Clarke <Adam.Clarke@StrategicData.com.au>:
> Hi all,
> 
> I'm trying to write a script that runs on Linux (Unix) and Win32. I want 
> to /use/ a module when on Win32 (to access the registry) that doesn't 
> exist in Unixy perl. Is there a way to skip a use statement at runtime 
> based on platform or is the only/best way to make a full build 
> (MakeMaker thing) and handle the differences there somehow. If the 
> latter, anyone got some simple pointers.
> 
> Cheers
> Adam





From rickm at isite.net.au  Wed Oct 16 22:56:05 2002
From: rickm at isite.net.au (rickm@isite.net.au)
Date: Wed Aug  4 00:02:43 2004
Subject: Cross Platform Perl
In-Reply-To: <20021017033736.GF27913@mukc.org.au>
References: <20021017030133.GD27913@mukc.org.au> <005601c2758d$2586b960$38662fcb@hewison.com.au> <20021017033736.GF27913@mukc.org.au>
Message-ID: <1034826965.3dae34d53bbc6@webmail.isite.net.au>

Doh! Should have checked mail again before replying. Paul's solution is better 
than mine :)


Quoting Paul Fenwick <pjf@perltraining.com.au>:

> G'day Everyone,
> 
> 	Rob and Adam are indeed correct.  You do need a string
> eval and not a block eval.  My mistake.  :)
> 
> 	Cheers,
> 
> 		Paul
> 
> On Thu, Oct 17, 2002 at 01:27:37PM +1000, Rob Casey wrote:
> > Just a follow up to this ...
> > 
> > > BEGIN {		# So we run at the same time as most use
> > statements.
> > > 	if ($^O eq 'MSWin32') {
> > > 		eval { use My::Module; }
> > > 	}
> > > }
> > 
> > This will fail - As the compiler will find this use statement and
> > attempt to include the module at this point.  To make use of dynamic
> > loading of modules, the eval statement must called with a string
> > argument as this string is not interpreted at compile time, but at the
> > time of execution.
> > 
> > Regards,
> > Rob
> > 
> > 
> > Rob Casey
> > Business Manager, Senior IT Consultant
> > Cowsnet Internet and Professional Services
> > http://www.cowsnet.com.au
> > 
> >  
> > 
> > -----Original Message-----
> > From: owner-melbourne-pm@pm.org [mailto:owner-melbourne-pm@pm.org] On
> > Behalf Of Paul Fenwick
> > Sent: Thursday, 17 October 2002 1:02 PM
> > To: Adam Clarke
> > Cc: melbourne-pm@pm.org
> > Subject: Re: Cross Platform Perl
> > 
> > 
> > G'day Adam,
> > 
> > On Thu, Oct 17, 2002 at 12:42:42PM +1000, Adam Clarke wrote:
> > 
> > > I'm trying to write a script that runs on Linux (Unix) and Win32. I
> > want 
> > > to /use/ a module when on Win32 (to access the registry) that doesn't 
> > > exist in Unixy perl. Is there a way to skip a use statement at runtime
> > 
> > > based on platform or is the only/best way to make a full build 
> > > (MakeMaker thing) and handle the differences there somehow. If the 
> > > latter, anyone got some simple pointers.
> > 
> > Glad you asked!  Perl has a variable called $^O (or $OSNAME when
> > using English) that gives you the name of the operating system
> > you're running under.
> > 
> > So you can do the following:
> > 
> > BEGIN {		# So we run at the same time as most use statements.
> > 	if ($^O eq 'MSWin32') {
> > 		eval { use My::Module; }
> > 	}
> > }
> > 
> > The eval is needed because usually use statements get executed
> > immediately, regardless of context.  The BEGIN means that your
> > chunk of code is executed "at compile time".  If this isn't important,
> > you can leave the BEGIN block out.
> > 
> > There's lots of code out there which does different things depending
> > upon the OS.  Take a look at File::Copy's source code
> > (perldoc -m File::Copy) to see a lot of this in action.
> > 
> > Cheers,
> > 
> > 	Paul
> > 
> > -- 
> > Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
> > Director of Training                   | Ph:  +61 3 9354 6001
> > Perl Training Australia                | Fax: +61 3 9354 2681
> > 
> 
> 
> 
> -- 
> Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
> Director of Training                   | Ph:  +61 3 9354 6001
> Perl Training Australia                | Fax: +61 3 9354 2681





From scottp at dd.com.au  Thu Oct 17 21:11:37 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Hardware::iButton and Device::ParallelPort
Message-ID: <EED8AC84-E23E-11D6-9C4B-003065B58CF8@dd.com.au>

Hey Dudes,

I have become the maintainer of Hardware::iButton and intend to put it 
up on CPAN (in its current form) on the weekend.
I have lots of mods I plan to do, mainly around improving the serial 
port interfaces and possibly using the C code as a potential optional 
backend to the pure perl one.

I thought people may also be interested that I have a good little board 
layout for outputs and temperature sensors.

I also now have working the output (just with LED outputs) and a 
temperature input.
The Temperature chips are about $8, measure from -10 to +85 c and are 
..5 accurate ! Really amazing.
They also seem to react VERY quickly.

The real long term will be inputs to JINI (embedded Java Machines on a 
single board), but these massively increase the complexity of the 
protocol, which is all in code, not in hardware. The alternative is the 
link with the existing C code above but that has many disadvantages:
	- Not GNU code (free, not really sure of the license yet) - seems to 
be public domain ?
	- C files, not really easily linkable libraries, but I may be able to 
change that
	- Not very cross platform, although Linux and Windows are supported.

For those also interested I am releasing a new version of 
Device::ParallelPort to do linux parport access directly in perl, no 
more C compiles / XS modules for that. You still require a compile for 
the direct access linux module. My windows 32 DLL binding is also 
almost complete, I lack the testing ability though :-)

Scott
-- 
Scott Penrose
Welcome to the Digital Dimension
http://www.dd.com.au/
scottp@dd.com.au

Dismaimer: Contents of this mail and signature are bound to change 
randomly. Whilst every attempt has been made to control said 
randomness, the author wishes to remain blameless for the number of 
eggs that damn chicken laid. Oh and I don't want to hear about 
butterflies either.


From gcross at alphalink.com.au  Fri Oct 18 05:33:37 2002
From: gcross at alphalink.com.au (Graeme Cross)
Date: Wed Aug  4 00:02:43 2004
Subject: Linux and wireless Installfest
Message-ID: <200210182033.37807.gcross@alphalink.com.au>

Thought this might be of interest to people on the list: if you are interested 
in getting into Linux and/or wireless technologies...


The Linux Users of Victoria will be holding their seventh Linux Installfest
on Saturday, November 23, 2002.

An Installfest is an event to which you bring your computer and we help you
install Linux, or assist you with the configuration of your existing Linux
system. We will be supporting a wide range of Linux distributions, including
Debian, Red Hat, Mandrake, SuSe and Gentoo.

There will be demonstrations of Linux running a variety of software and
presentations on what Linux is, so if you're curious about Linux this is
your chance to experience it first-hand.

Installfest 2002 will be held at Strathcona Baptist Girls Grammar School,
Claremont Crescent, Canterbury (Melway map reference 46 D12) between 10am
and 4pm.

For more information, and registration details, visit the Installfest 2002
 web pages at:

	http://www.luv.asn.au/installfest2002/register.php3

In addition, Melbourne Wireless will be joining us to have their first
wireless installfest and antenna field day. For more information, visit the
Melbourne Wireless website at http://www.wireless.org.au/wiki/?installfest.

--
Graeme Cross <gcross@alphalink.com.au>


From gcross at alphalink.com.au  Fri Oct 18 06:11:50 2002
From: gcross at alphalink.com.au (Graeme Cross)
Date: Wed Aug  4 00:02:43 2004
Subject: Melbourne Perl companies in the news
Message-ID: <200210182111.50301.gcross@alphalink.com.au>

Two recent stories about Melbourne companies who use Perl:

1. Fastmail, who were in yesterday's Green Guide (Charles Wright's column)

2. O'Reilly are running an article about Open Systems: 
http://www.perl.com/pub/a/2002/10/15/radiator.html

- Graeme

-- 
Graeme Cross <gcross@alphalink.com.au>


From jh_lists at fastmail.fm  Fri Oct 18 10:50:57 2002
From: jh_lists at fastmail.fm (JP Howard)
Date: Wed Aug  4 00:02:43 2004
Subject: Melbourne Perl companies in the news
Message-ID: <20021018155057.8358D1AEC193@server5.fastmail.fm>

On Fri, 18 Oct 2002 21:11:50 +1000, "Graeme Cross"
<gcross@alphalink.com.au> said:
> 1. Fastmail, who were in yesterday's Green Guide (Charles Wright's
> column)
> 
Here's the two FastMail.FM stories:
http://www.theage.com.au/articles/2002/10/07/1033538891792.html
http://www.theage.com.au/articles/2002/10/17/1034561201555.html

From mpm at bachelorguy.com  Fri Oct 18 12:15:59 2002
From: mpm at bachelorguy.com (ADFH)
Date: Wed Aug  4 00:02:43 2004
Subject: Modifying postfix mailer virtual table files with Perl
Message-ID: <20021019031559.593c166c.mpm@bachelorguy.com>

Has anyone written/can recommend a Perl script/module etc. to work with
postfix format virtual table files?

Ie.

virtual.domain1       anything
postmaster@virtual.domain1    postmaster
user1@virtual.domain1 address1
user2@virtual.domain1 address2, address3
 , address4, address5

virtual.domain2       anything
postmaster@virtual.domain2    postmaster
user1@virtual.domain2 address1
user2@virtual.domain2 address2, address3,
 address4, address5

... with the possibility of hash comments which should be editable?

Looking into the possibility of a stand-alone, web-based system for
authorised staff to be able to modify the virtual user tables of several
domains.

Anthony

From scottp at dd.com.au  Sat Oct 19 20:16:09 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Web Site Update
Message-ID: <84043875-E3C9-11D6-9C4B-003065B58CF8@dd.com.au>

Hey Dudes,

The melbourne.pm.org web site has been updated. It now contains the 
updated meeting schedule right up to February next year :-)

It also has some minor fixes (links) and a new Perl Shop.

Enjoy.

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From david_dick at iprimus.com.au  Sun Oct 20 00:32:56 2002
From: david_dick at iprimus.com.au (David Dick)
Date: Wed Aug  4 00:02:43 2004
Subject: SOAP::Lite question
In-Reply-To: <3DB23C7B.6070703@iprimus.com.au>
References: <3DB23C7B.6070703@iprimus.com.au>
Message-ID: <3DB24008.2000209@iprimus.com.au>

alrightly then.  Just fooling around with SOAP::Lite.  I want my soap 
client to look like this.....

#! /usr/bin/perl

use SOAP::Lite +autodispatch =>
                           proxy => 'http://localhost/soap';

my ($document) = Document->retrieve({ 'document_id' => 1 });
if ($document->save()) {
       $document->commit();
} else {
       $document->rollback();
}

at first glance this seems to be a pipe dream, cos server state should 
be lost inbetween the save() and commit() methods.  Is a better solution 
to give my server object a new method of save_and_commit(), or can i use 
a different transport, or what? Any ideas would be appreciated....


From scottp at dd.com.au  Sun Oct 20 23:56:30 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Unknown Warning
Message-ID: <76970DE8-E4B1-11D6-BA5E-003065B58CF8@dd.com.au>

Hey Dudes,

Try the following bit of code...

#!/usr/bin/perl -w
use strict;

my $fred = 'a b c d';

for (my $i = scalar(split(/ /, $fred)); $i > 0; $i--) {
         print $i . "\n";
}


You get the following warning

"Use of implicit split to @_ is deprecated at ... line 6."

Now I don't understand.

The following code

#!/usr/bin/perl -w
use strict;

my $fred = 'a b c d';

foreach my $x (split(/ /, $fred)) {
         print $x . "\n";
}


Produces no warnings.

Calling a method like split, should just return a list. That list can 
be used inside methods.

So I though maybe it is coz we are calling a sub (eg: scalar)

#!/usr/bin/perl -w
use strict;

my $fred = 'a b c d';

x(split(/ /, $fred));
exit 0;

sub x {
         my @list = @_;
         print join("\n", @list) . "\n";
}

Nope, this code has no errors either.

Is it

a) A bug (5.6 and 5.8 do the same thing)

b) Something special about 'scalar'

Scott

-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From pjf at perltraining.com.au  Mon Oct 21 00:18:45 2002
From: pjf at perltraining.com.au (Paul Fenwick)
Date: Wed Aug  4 00:02:43 2004
Subject: Unknown Warning
In-Reply-To: <76970DE8-E4B1-11D6-BA5E-003065B58CF8@dd.com.au>
References: <76970DE8-E4B1-11D6-BA5E-003065B58CF8@dd.com.au>
Message-ID: <20021021051845.GA25255@mukc.org.au>

G'day Scotty / Mongers,

On Mon, Oct 21, 2002 at 02:56:30PM +1000, Scott Penrose wrote:

> for (my $i = scalar(split(/ /, $fred)); $i > 0; $i--) {
>         print $i . "\n";
> }

[snip]

> a) A bug (5.6 and 5.8 do the same thing)
> 
> b) Something special about 'scalar'

Split in a scalar context will split into @_ (Camel, first ed, p185).
In Perl 4, using ?? as pattern delimiters would do the same thing in
array context.  Perl 5 will still clobber @_ if you use split in a
scalar context, but it's considered deprecated, hence the warning.
Clobbering @_ probably isn't what you want, anyway.

If you're splitting on a single character, you can use tr/// to
count the occurances of that character, which will be faster than
split. 

If you're using a more complex pattern, you can force into an array
context, but the return into a scalar:

	my $i = @{[split(/ /, $fred)]};

I'm sure there are other solutions using regexpes.  TMTOWTDI.

Cheers,

	Paul

-- 
Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
Director of Training                   | Ph:  +61 3 9354 6001
Perl Training Australia                | Fax: +61 3 9354 2681

From scottp at dd.com.au  Mon Oct 21 00:28:31 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Unknown Warning
In-Reply-To: <20021021051845.GA25255@mukc.org.au>
Message-ID: <EF94A967-E4B5-11D6-BA5E-003065B58CF8@dd.com.au>


On Monday, Oct 21, 2002, at 15:18 Australia/Melbourne, Paul Fenwick 
wrote:
>
> Split in a scalar context will split into @_ (Camel, first ed, p185).
> In Perl 4, using ?? as pattern delimiters would do the same thing in
> array context.  Perl 5 will still clobber @_ if you use split in a
> scalar context, but it's considered deprecated, hence the warning.
> Clobbering @_ probably isn't what you want, anyway.


Ahhhh.... I see my mistake.

I knew the above, but my mistake was thinking of scalar() as a 
function, ie: calling a function that returns a scalar.

But it is NOT !. It is effectively a compile time directive that says - 
do the NEXT thing in scalar context !

Doooh...

Thanks Paul.

> If you're splitting on a single character, you can use tr/// to
> count the occurances of that character, which will be faster than
> split.

yes yes of course :-) I just simplified the code so that I could grab a 
worked example.

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From Martin.Ryan at sensis.com.au  Mon Oct 21 00:30:39 2002
From: Martin.Ryan at sensis.com.au (Ryan, Martin)
Date: Wed Aug  4 00:02:43 2004
Subject: Unknown Warning
Message-ID: <D0C637D367F10D419D50B67AD041C7815CF699@cw-exg-srv04.pacificaccess.com.au>

Hi Scott,

At a guess off the top of my head - is one being evaluated in a list context
and the other in a scalar context giving the different results?

Martin

> -----Original Message-----
> From: Scott Penrose [mailto:scottp@dd.com.au]
> Sent: Monday, 21 October 2002 14:57
> To: melbourne-pm@pm.org
> Cc: benno@myinternet.com.au
> Subject: Unknown Warning
> 
> 
> Hey Dudes,
> 
> Try the following bit of code...
> 
> #!/usr/bin/perl -w
> use strict;
> 
> my $fred = 'a b c d';
> 
> for (my $i = scalar(split(/ /, $fred)); $i > 0; $i--) {
>          print $i . "\n";
> }
> 
> 
> You get the following warning
> 
> "Use of implicit split to @_ is deprecated at ... line 6."
> 
> Now I don't understand.
> 
> The following code
> 
> #!/usr/bin/perl -w
> use strict;
> 
> my $fred = 'a b c d';
> 
> foreach my $x (split(/ /, $fred)) {
>          print $x . "\n";
> }
> 
> 
> Produces no warnings.
> 
> Calling a method like split, should just return a list. That list can 
> be used inside methods.
> 
> So I though maybe it is coz we are calling a sub (eg: scalar)
> 
> #!/usr/bin/perl -w
> use strict;
> 
> my $fred = 'a b c d';
> 
> x(split(/ /, $fred));
> exit 0;
> 
> sub x {
>          my @list = @_;
>          print join("\n", @list) . "\n";
> }
> 
> Nope, this code has no errors either.
> 
> Is it
> 
> a) A bug (5.6 and 5.8 do the same thing)
> 
> b) Something special about 'scalar'
> 
> Scott
> 
> -- 
> Scott Penrose
> VP in charge of Pancakes
> http://linux.dd.com.au/
> scottp@dd.com.au
> 
> Dismaimer: If you receive this email in error - please eat it 
> immediately to prevent it from falling into the wrong hands.
> 

From scottp at dd.com.au  Sun Oct 27 14:30:27 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Fwd: [pm_groups] Perl Quiz of the Week
Message-ID: <EDD91DCF-E9EA-11D6-918E-003065B58CF8@dd.com.au>

> Mark-Jason Dominus posted this announcement on the Fun with Perl
> mailing list. I thought it might interest PM groups. too:
>
>   I taught a class in Illinois last week, and one of the students
>   asked if there was a mailing list that would deliver a weekly Perl
>   quiz or programming puzzle.  I said I didn't know of one, but that
>   it sounded like a good idea, and that I would set one up when I got
>   back.
>
>   I am now back.  If you want to get the quiz-of-the-week, send a note
>   to
>
>         perl-qotw-subscribe@plover.com
>
>
>   Please feel free to circulate this announcement to appropriate
>   venues.
>
>   Thanks.
-- 
Scott Penrose
Welcome to the Digital Dimension
http://www.dd.com.au/
scottp@dd.com.au

Dismaimer: Contents of this mail and signature are bound to change 
randomly. Whilst every attempt has been made to control said 
randomness, the author wishes to remain blameless for the number of 
eggs that damn chicken laid. Oh and I don't want to hear about 
butterflies either.


From scottp at dd.com.au  Sun Oct 27 23:04:23 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Optional parameters to regular expressions
Message-ID: <B96BF8EE-EA32-11D6-8D4C-003065B58CF8@dd.com.au>

Hey Dudes,

Is there a way I can do something like this.

my $o = 'i';
print "got it" if ($str =~ /fred/$o);

ie: I want to be able to use a parameter to say it is case insensitive 
without having to use multiple if.

Scott
-- 
Scott Penrose
Open source developer
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: Open sauce usually ends up never coming out (of the bottle).


From pjf at perltraining.com.au  Sun Oct 27 23:12:33 2002
From: pjf at perltraining.com.au (Paul Fenwick)
Date: Wed Aug  4 00:02:43 2004
Subject: Optional parameters to regular expressions
In-Reply-To: <B96BF8EE-EA32-11D6-8D4C-003065B58CF8@dd.com.au>
References: <B96BF8EE-EA32-11D6-8D4C-003065B58CF8@dd.com.au>
Message-ID: <20021028051233.GA9294@mukc.org.au>

G'day Scotty,

On Mon, Oct 28, 2002 at 04:04:23PM +1100, Scott Penrose wrote:

> Is there a way I can do something like this.
> 
> my $o = 'i';
> print "got it" if ($str =~ /fred/$o);
> 
> ie: I want to be able to use a parameter to say it is case insensitive 
> without having to use multiple if.

There sure is!  You need to use non-matching parantheses, (?:), which
also allow you to turn regexp switches on and off.  For your example:

	print "got it" if $str =~ /(?$o:fred)/;

Switches places between the ? and : in your non-matching parentheses
are turned on for that sub-expression.  Likewise, you can use "-i"
to turn case-insensitivity off until the end of the sub-expression.

All the best,

	Paul

-- 
Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
Director of Training                   | Ph:  +61 3 9354 6001
Perl Training Australia                | Fax: +61 3 9354 2681

From adam.thick at StrategicData.com.au  Mon Oct 28 17:15:40 2002
From: adam.thick at StrategicData.com.au (Adam Thick)
Date: Wed Aug  4 00:02:43 2004
Subject: Perlish Shopping Cart
Message-ID: <3DBDC51C.5020406@strategicdata.com.au>


I've been scouring the net for an open source perl shopping cart without 
much success. Most I've discovered haven't been maintained in quite a 
while, or the ones that have seem to be overkill...  I was wondering 
whether anybody could point me in a good wholesome direction?

Cheers,
Adam Thick.



From simon at unisolve.com.au  Mon Oct 28 17:59:32 2002
From: simon at unisolve.com.au (Simon Taylor)
Date: Wed Aug  4 00:02:43 2004
Subject: Perlish Shopping Cart
In-Reply-To: <3DBDC51C.5020406@strategicdata.com.au>
References: <3DBDC51C.5020406@strategicdata.com.au>
Message-ID: <200210291059.32455.simon@unisolve.com.au>

Adam,

> I've been scouring the net for an open source perl shopping cart without
> much success. Most I've discovered haven't been maintained in quite a
> while, or the ones that have seem to be overkill...  I was wondering
> whether anybody could point me in a good wholesome direction?

Have you seen the ShoppingCart module by Lincoln Stein (the author of CGI.pm)? 
It's not developed as a production-ready system, but it is a good basis for 
one. 

http://stein.cshl.org/~lstein/talks/perl_conference/advanced/

Regards,

Simon Taylor
-- 
Unisolve Pty Ltd - Melbourne, Australia
+61 3 9568 2005

From scottp at dd.com.au  Mon Oct 28 21:20:38 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: CSS manipulation
Message-ID: <65622866-EAED-11D6-8D4C-003065B58CF8@dd.com.au>

CSS in some browsers is not very well supported, especially the 
inheritance from multiple style sheets. There is also the consideration 
that a single file is faster to load.

However in development, I don't wish to manipulate too many of these, I 
want to edit my files along the lines of

	P {
		text-align: left;
	}

But a CSS in one style can inherit from the default.

Default is something like this...

	TD {
		text-align: left;
	}

	P {
		text-align: right;
		width: 80%;
	}

What I would like is a program that can grab the files and do something 
clever along the lines of...

P,TD {text-align: left;}
P {width: 80%;}

It has grouped the common elements together.

Does anyone know of any modules out there that will do a good job of 
reading and parsing to a sensible structure a CSS file, so that we can 
do this inheritance and output a single highly optimised output file ?

Scott
-- 
Scott Penrose
Welcome to the Digital Dimension
http://www.dd.com.au/
scottp@dd.com.au

Dismaimer: Contents of this mail and signature are bound to change 
randomly. Whilst every attempt has been made to control said 
randomness, the author wishes to remain blameless for the number of 
eggs that damn chicken laid. Oh and I don't want to hear about 
butterflies either.


From Nathan.Bailey at its.monash.edu  Mon Oct 28 22:03:39 2002
From: Nathan.Bailey at its.monash.edu (Nathan Bailey)
Date: Wed Aug  4 00:02:43 2004
Subject: CSS manipulation 
In-Reply-To: scottp's message of Tue, 29 Oct 2002 14:20:38 +1100.
             <65622866-EAED-11D6-8D4C-003065B58CF8@dd.com.au> 
Message-ID: <200210290403.PAA256203@goaway.cc.monash.edu.au>

Scott Penrose <scottp@dd.com.au> wrote:
>so that we can do this inheritance and output a single highly
>optimised output file ?

What is your intention for optimisation?  For ease of management, or
improved browser speed?  Presumably, with every browser having a
different CSS engine, their parsing is optimised in different ways too
-- can you be sure that your summarisation will actually improve
performance, or may it slow it down?

I would suspect the later, since the parser will only have to deal with
each object once in a "flat" format, but will have to deal with objects
multiple times in an "inherited" format.  However, I have never looked
at a CSS parser's code :-)

N

From scottp at dd.com.au  Mon Oct 28 22:37:29 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: CSS manipulation 
In-Reply-To: <200210290403.PAA256203@goaway.cc.monash.edu.au>
Message-ID: <21C96032-EAF8-11D6-8D4C-003065B58CF8@dd.com.au>


On Tuesday, Oct 29, 2002, at 15:03 Australia/Melbourne, Nathan Bailey 
wrote:

> Scott Penrose <scottp@dd.com.au> wrote:
>> so that we can do this inheritance and output a single highly
>> optimised output file ?
>
> What is your intention for optimisation?  For ease of management, or
> improved browser speed?  Presumably, with every browser having a

Mostly browser speed, and implementation. If I can get a way to inherit 
correctly two style sheets, then I don't need to do it for 
implementation, which only leaves speed.

> different CSS engine, their parsing is optimised in different ways too
> -- can you be sure that your summarisation will actually improve
> performance, or may it slow it down?

You are talking about render performance, which generally does not 
bother me. Size of the files over the wire is number 1 on our must be 
smaller and fast list. Schools generally only have 128K (or some even 
64K) links, so you can imagine what a room full of computers does to 
that link.

> I would suspect the later, since the parser will only have to deal with
> each object once in a "flat" format, but will have to deal with objects
> multiple times in an "inherited" format.  However, I have never looked
> at a CSS parser's code :-)

Currently you can not compress a CSS (IE and some other browsers fail 
to load it, even though they SAY they support GZIP).

The intension with the groupings above is to keep the size of the file 
very small, but ALSO to make it so there is only one file.

One file means - only one request, smaller size.

With CSS, if you use the C bit (cascading) then you override the values 
you want, which means by design we have redundant information being 
sent to the browser, plus some browsers do it wrong.

I have reviewed a number of methods of doing this, and have come to the 
conclusion for our next release that 'it is too damn hard'(tm) :-)

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From scottp at dd.com.au  Mon Oct 28 23:28:46 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: CSS manipulation
In-Reply-To: <20021029051754.A25670@gnu>
Message-ID: <4BD95C8C-EAFF-11D6-8D4C-003065B58CF8@dd.com.au>

On Tuesday, Oct 29, 2002, at 16:17 Australia/Melbourne, Piers Harding 
wrote:

>
> Have you considered inline stylesheets (<STYLE/>) if you use a
> templating engine for your delivery - that way you would be able to
> compress your pages as a whole.

That is less efficient as it means they can not have a local cache. 
Generally, being a web application, they load many pages, each of which 
uses the same style sheet, so inserting <STYLE> entries means that it 
would have to be sent each time.

That would be a really good suggestion though for web sites that expect 
to receive just a single hit (maybe high speed advertising site or 
something similar).

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From simon at unisolve.com.au  Mon Oct 28 23:41:56 2002
From: simon at unisolve.com.au (Simon Taylor)
Date: Wed Aug  4 00:02:43 2004
Subject: CSS manipulation
Message-ID: <200210291641.56823.simon@unisolve.com.au>

Hi Scott,

> You are talking about render performance, which generally does not
> bother me. Size of the files over the wire is number 1 on our must be
> smaller and fast list. Schools generally only have 128K (or some even
> 64K) links, so you can imagine what a room full of computers does to
> that link.

I'd do it in Parse::RecDescent if I found no other tool.

The CSS.pm modules uses Parse::RecDescent in this way.

http://search.cpan.org/author/ALLENDAY/CSS-0.07/CSS.pm

- Simon Taylor
-- 
Unisolve Pty Ltd - Melbourne, Australia
+61 3 9568 2005

From david_dick at iprimus.com.au  Tue Oct 29 01:25:15 2002
From: david_dick at iprimus.com.au (David Dick)
Date: Wed Aug  4 00:02:43 2004
Subject: CSS manipulation
In-Reply-To: <4BD95C8C-EAFF-11D6-8D4C-003065B58CF8@dd.com.au>
References: <4BD95C8C-EAFF-11D6-8D4C-003065B58CF8@dd.com.au>
Message-ID: <3DBE37DB.2050802@iprimus.com.au>

The logic that I follow says that for a client base that is static and 
high volume, you download the stylesheet as an external file, download 
the JS files as external files, give 'em long TTL value and compress the 
actual html files.  However, for everything else, i've always inlined 
the stylesheet and compressed the whole thing (paying attention of 
course to IE6 etc with regard to compression).  This of course is 
because my style sheets so far have been miniscule in comparison to the 
overall html, so there wasn't that much of a difference.  Presumeably 
the stylesheets are big enuff to warrant that null and void, but if so, 
is it worth hand-crafting the stylesheets instead of writing a .pm file 
to do it for you? (i can't believe i'm saying this :))

Scott Penrose wrote:

> On Tuesday, Oct 29, 2002, at 16:17 Australia/Melbourne, Piers Harding 
> wrote:
>
>>
>> Have you considered inline stylesheets (<STYLE/>) if you use a
>> templating engine for your delivery - that way you would be able to
>> compress your pages as a whole.
>
>
> That is less efficient as it means they can not have a local cache. 
> Generally, being a web application, they load many pages, each of 
> which uses the same style sheet, so inserting <STYLE> entries means 
> that it would have to be sent each time.
>
> That would be a really good suggestion though for web sites that 
> expect to receive just a single hit (maybe high speed advertising site 
> or something similar).
>
> Scott



From Edgar.P at bluecom.dk  Tue Oct 29 13:25:12 2002
From: Edgar.P at bluecom.dk (Jennifer Parker)
Date: Wed Aug  4 00:02:43 2004
Subject: Free Debt Elimination Quote..
Message-ID: <000043096bf0$00005057$00003c65@blogica.gr>

An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021028/12ac9e72/attachment.htm
From rickm at isite.net.au  Tue Oct 29 03:52:24 2002
From: rickm at isite.net.au (Rick Measham)
Date: Wed Aug  4 00:02:43 2004
Subject: CSS manipulation
In-Reply-To: <65622866-EAED-11D6-8D4C-003065B58CF8@dd.com.au>
References: <65622866-EAED-11D6-8D4C-003065B58CF8@dd.com.au>
Message-ID: <a05010400b9e409dfd71f@[192.168.1.128]>

>CSS in some browsers is not very well supported, especially the inheritance from multiple style sheets. There is also the consideration that a single file is faster to load.

The above should be carefully considered. However if you are in a school computer lab, you'll know which browser is being used.

That being said, here's a working solution using CSS.pm

#!/usr/bin/perl
use CSS;
use Strict;

#--------------------------------------------------------------------------------
# USAGE:
#   simplify(list_of_files);
# NOTE:
#   list_of_files should be in _increasing_ importance. Styles from the last
#   file will overwrite any duplicate styles of all previous files.
# RETURNS:
#   Simplified formatted CSS
#--------------------------------------------------------------------------------
# NOTE CAREFULLY:
#   CSS.pm is very strict in its parsing. Make sure your style sheets use a
#   semi-colon after every parameter, even at the end of a block:
#   body {
#      font-family: 'times';
#      font-size: '12pt';  <--- this semicolon MUST exist
#   }                           even though its at the end of the block
#                               This ain't perl buster!
#--------------------------------------------------------------------------------
# Author's jottings: This is probably not the most 'golf' way to do this, but
# its a working solution.
#--------------------------------------------------------------------------------

print simplify('pmdefault.css','pmuser.css');


#--------------------------------------------------------------------------------
# The routine
#--------------------------------------------------------------------------------

sub simplify {
	my ($output, %usedby, %css);
	my @files = @_;
	die("You have to tell me at least one css file to simplify.") unless @files;
	my $stylesheet = CSS->new(-source=>$files[0]);
	shift @files;
	foreach my $file (@files) {
		$stylesheet->add_file($file);
	}
	
	my @styles = $stylesheet->styles;

	foreach my $style (@styles) {
		my %props = $style->properties;
		foreach my $property (keys %props) {
			$props{$property} = "'$props{$property}'" if $props{$property}=~/\D/;
			push (@{$usedby{sprintf("%19s : %s",$property,$props{$property})}}, $style->selector); #Can I memoize sprintf?
		}
	}

	foreach my $style (sort keys %usedby) {
		my $stylename = join(', ',sort @{$usedby{$style}});
		push(@{$css{$stylename}},$style);
	}


	foreach my $style (sort keys %css) {
		$output .= "$style {\n";
		$output .= join(";\n",@{$css{$style}});
		$output .= ";\n}\n\n";
	}
	return $output;
}
-- 
--------------------------------------------------------
            There are 10 kinds of people:
  those that understand binary, and those that don't.
--------------------------------------------------------
  The day Microsoft makes something that doesn't suck 
    is the day they start selling vacuum cleaners
--------------------------------------------------------

From scottp at myinternet.com.au  Tue Oct 29 21:48:12 2002
From: scottp at myinternet.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
Message-ID: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au>

Hey Dudes.

I have a bit of code that looks like this

	my $ref =  
SomeClass->FirstMethod($param1)->AnotherMethod($param2)->MoreMethods();

Perfectly legitimate in that we are effectively getting a grass root  
element and drilling down.

My problem is that each of these can fail. In my particular case I  
don't care which fail or why, is it ok therefore to do this

	my $ref = eval  
{SomeClass->FirstMethod($param1)->AnotherMethod($param2)- 
 >MoreMethods();};
	return undef if ($@);

Or is that a really bad practice?, is there a better way ?, otherwise I  
have to fall down to the usual practice of one at a time.

	my $obj1 = SomeClass->FirstMethod($param1)				or return undef;
	my $obj2 = $obj1->AnotherMethod($param2)
		or return undef;
	my $ref = $obj2->MoreMethods()
		or return undef;

Scott

-- 
Scott Penrose
Welcome to the Digital Dimension
http://www.dd.com.au/
scottp@dd.com.au

Dismaimer: Contents of this mail and signature are bound to change  
randomly. Whilst every attempt has been made to control said  
randomness, the author wishes to remain blameless for the number of  
eggs that damn chicken laid. Oh and I don't want to hear about  
butterflies either.



From jarich at perltraining.com.au  Tue Oct 29 22:13:08 2002
From: jarich at perltraining.com.au (Jacinta Richardson)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
In-Reply-To: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au>
Message-ID: <Pine.LNX.3.96.1021030151226.22880L-100000@teddybear.perltraining.com.au>


> I have a bit of code that looks like this
> 
> 	my $ref =  
> SomeClass->FirstMethod($param1)->AnotherMethod($param2)->MoreMethods();
 
> Or is that a really bad practice?, is there a better way ?, otherwise I  
> have to fall down to the usual practice of one at a time.

You know, if you were to ask these questions on PerlMonks you'd be level 9
by now....  ;)

		Jacinta

--
   ("`-''-/").___..--''"`-._          |  Jacinta Richardson	    |
    `6_ 6  )   `-.  (     ).`-.__.`)  |  Perl Training Australia    |
    (_Y_.)'  ._   )  `._ `. ``-..-'   |      +613 9354 6001 	    |  
  _..`--'_..-_/  /--'_.' ,'           | contact@perltraining.com.au |
(il),-''  (li),'  ((!.-'              |   www.perltraining.com.au   |


From daniel at landmarksoftware.com.au  Tue Oct 29 23:38:08 2002
From: daniel at landmarksoftware.com.au (Daniel Walmsley)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
Message-ID: <10328A423AC2D311AE5B0008C70FE5643666EC@LANDNT>

As soon as one procedure returns undef, the subsequent call will fail, and
the failure will chain back down without reaching out to the end. I think.
That is:
Example 1:
1) FirstMethod works
2) AnotherMethod fails, returns undef
3) Perl barfs when asked to call MoreMethods
4) You return undef

Example 2:
1) FirstMethod fails
2) Perl barfs when asked to call AnotherMethod
3) You return undef

Therefore I don't see it as being any less efficient than any other way,
UNLESS Perl is automatically breaking it into separate statements with
separate return values without treating it like one big "block" that can die
at any step and return.

Whoops, I guess I just failed to answer your question. Bugger it, I'll hit
send anyway.

As to whether it's "bad practise", I guess it depends on whether anyone who
cares about that sort of thing is ever going to look at your code again ;)

Dan

-----Original Message-----
From: Scott Penrose [mailto:scottp@myinternet.com.au] 
Sent: Wednesday, 30 October 2002 2:48 PM
To: melbourne-pm@pm.org
Subject: Performance of chained methods.


Hey Dudes.

I have a bit of code that looks like this

	my $ref =  
SomeClass->FirstMethod($param1)->AnotherMethod($param2)->MoreMethods();

Perfectly legitimate in that we are effectively getting a grass root  
element and drilling down.

My problem is that each of these can fail. In my particular case I  
don't care which fail or why, is it ok therefore to do this

	my $ref = eval  
{SomeClass->FirstMethod($param1)->AnotherMethod($param2)- 
 >MoreMethods();};
	return undef if ($@);

Or is that a really bad practice?, is there a better way ?, otherwise I  
have to fall down to the usual practice of one at a time.

	my $obj1 = SomeClass->FirstMethod($param1)
or return undef;
	my $obj2 = $obj1->AnotherMethod($param2)
		or return undef;
	my $ref = $obj2->MoreMethods()
		or return undef;

Scott

-- 
Scott Penrose
Welcome to the Digital Dimension
http://www.dd.com.au/
scottp@dd.com.au

Dismaimer: Contents of this mail and signature are bound to change  
randomly. Whilst every attempt has been made to control said  
randomness, the author wishes to remain blameless for the number of  
eggs that damn chicken laid. Oh and I don't want to hear about  
butterflies either.


From scottp at dd.com.au  Tue Oct 29 23:56:09 2002
From: scottp at dd.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
In-Reply-To: <10328A423AC2D311AE5B0008C70FE5643666EC@LANDNT>
Message-ID: <49B200BF-EBCC-11D6-8D4C-003065B58CF8@dd.com.au>


On Wednesday, Oct 30, 2002, at 16:38 Australia/Melbourne, Daniel 
Walmsley wrote:

> As soon as one procedure returns undef, the subsequent call will fail, 
> and
> the failure will chain back down without reaching out to the end. I 
> think.

Yes I know, that is why my three methods cater for that each time.

> That is:
> Example 1:
> 1) FirstMethod works
> 2) AnotherMethod fails, returns undef
> 3) Perl barfs when asked to call MoreMethods
> 4) You return undef
>
> Example 2:
> 1) FirstMethod fails
> 2) Perl barfs when asked to call AnotherMethod
> 3) You return undef
>
> Therefore I don't see it as being any less efficient than any other 
> way,
> UNLESS Perl is automatically breaking it into separate statements with
> separate return values without treating it like one big "block" that 
> can die
> at any step and return.
>
> Whoops, I guess I just failed to answer your question. Bugger it, I'll 
> hit
> send anyway.
>
> As to whether it's "bad practise", I guess it depends on whether 
> anyone who
> cares about that sort of thing is ever going to look at your code 
> again ;)

The eval{} method works perfectly, and in theory eval {} has almost no 
performance hit, however I FEEL like I am doing a VERY bad think ;-) i 
don't like bad hacks.

Scott
-- 
Scott Penrose
VP in charge of Pancakes
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: If you receive this email in error - please eat it 
immediately to prevent it from falling into the wrong hands.


From david_dick at iprimus.com.au  Wed Oct 30 00:54:24 2002
From: david_dick at iprimus.com.au (David Dick)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
In-Reply-To: <49B200BF-EBCC-11D6-8D4C-003065B58CF8@dd.com.au>
References: <49B200BF-EBCC-11D6-8D4C-003065B58CF8@dd.com.au>
Message-ID: <3DBF8220.3080100@iprimus.com.au>



Scott Penrose wrote:

>
> On Wednesday, Oct 30, 2002, at 16:38 Australia/Melbourne, Daniel 
> Walmsley wrote:
>
>> As soon as one procedure returns undef, the subsequent call will 
>> fail, and
>> the failure will chain back down without reaching out to the end. I 
>> think.
>
>
> Yes I know, that is why my three methods cater for that each time.
>
>> That is:
>> Example 1:
>> 1) FirstMethod works
>> 2) AnotherMethod fails, returns undef
>> 3) Perl barfs when asked to call MoreMethods
>> 4) You return undef
>>
>> Example 2:
>> 1) FirstMethod fails
>> 2) Perl barfs when asked to call AnotherMethod
>> 3) You return undef
>>
>> Therefore I don't see it as being any less efficient than any other way,
>> UNLESS Perl is automatically breaking it into separate statements with
>> separate return values without treating it like one big "block" that 
>> can die
>> at any step and return.
>>
>> Whoops, I guess I just failed to answer your question. Bugger it, 
>> I'll hit
>> send anyway.
>>
>> As to whether it's "bad practise", I guess it depends on whether 
>> anyone who
>> cares about that sort of thing is ever going to look at your code 
>> again ;)
>
>
> The eval{} method works perfectly, and in theory eval {} has almost no 
> performance hit, however I FEEL like I am doing a VERY bad think ;-) i 
> don't like bad hacks.
>
> Scott 

why a bad thing? after all, catching exceptions is a feature of most 
modern languages.  although i got tripped up with them badly the other 
day.  Forgot to send a rollback to the db after catching an unrelated 
exception.... bugger.... :)


From pjf at perltraining.com.au  Wed Oct 30 05:31:45 2002
From: pjf at perltraining.com.au (Paul Fenwick)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
In-Reply-To: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au>
References: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au>
Message-ID: <20021030113145.GA25980@mukc.org.au>

G'day Scotty,

On Wed, Oct 30, 2002 at 02:48:12PM +1100, Scott Penrose wrote:

> My problem is that each of these can fail. In my particular case I  
> don't care which fail or why, is it ok therefore to do this
> 
> 	my $ref = eval  
> {SomeClass->FirstMethod($param1)->AnotherMethod($param2)- 
> >MoreMethods();};
> 	return undef if ($@);

Looks good to me, especially if you don't care why they fail.  I
would consider it *good* practice, as you end up with cleaner code.
I'm a great fan of code exceptions, rather than requiring
checks every step of the way.

In fact, it's not only cleaner and more programmer friendly, it's
also *faster*, especially for long chains.  On my system, the exceptions
code only uses 8.5 CPU seconds to run the tests below, but the
checking code requires 11.4 CPU seconds.  (800MHz Duron, Perl 5.6.1)

(Hint, on most systems you can pipe this entire message to "perl -x" to run
the same tests.)

#!/usr/bin/perl -w
use strict;

package Foo;

sub bar  { return $_[0]; }
sub baz  { return $_[0]; }
sub bark { return $_[0]; }
sub bath { return $_[0]; }
sub barf { return undef; }

package main;

use Benchmark;

my $foo = bless({},"Foo");

timethese(1_000_000, {
	exceptions => sub {
		eval {
			$foo->bar->baz->bark->bath->barf;
		};
		return undef if ($@);
	},

	checks => sub {
		my $bar  = $foo->bar or return undef;
		my $baz  = $bar->baz or return undef;
		my $bark = $baz->bark or return undef;
		my $bath = $bark->bath or return undef;
		my $barf = $bath->barf or return undef;
	}
});
__END__


Cheers,

	Paul

-- 
Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
Director of Training                   | Ph:  +61 3 9354 6001
Perl Training Australia                | Fax: +61 3 9354 2681

From simon at unisolve.com.au  Wed Oct 30 22:31:42 2002
From: simon at unisolve.com.au (Simon Taylor)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
In-Reply-To: <20021030113145.GA25980@mukc.org.au>
References: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au> <20021030113145.GA25980@mukc.org.au>
Message-ID: <200210302331.42748.simon@unisolve.com.au>

Paul,

> In fact, it's not only cleaner and more programmer friendly, it's
> also *faster*, especially for long chains.  On my system, the exceptions
> code only uses 8.5 CPU seconds to run the tests below, but the
> checking code requires 11.4 CPU seconds.  (800MHz Duron, Perl 5.6.1)

OK, *that* got my attention.

These results, if they hold true in general usage, suggest a good way to 
wrench more speed out of perl/Tk applications, where the chains can be very 
long indeed....

Are we archiving these messages? This is good stuff!

Regards,

Simon Taylor
--
Unisolve Pty Ltd

From mjs at beebo.org  Wed Oct 30 07:25:39 2002
From: mjs at beebo.org (Michael Stillwell)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
In-Reply-To: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au>
Message-ID: <14E1AD96-EC0B-11D6-BBE3-0003930BCEFA@beebo.org>

On Wednesday, Oct 30, 2002, at 14:48 Australia/Melbourne, Scott Penrose  
wrote:

> 	my $ref = eval  
> {SomeClass->FirstMethod($param1)->AnotherMethod($param2)- 
> >MoreMethods();};
> 	return undef if ($@);
>
> Or is that a really bad practice?, is there a better way ?,

That does seem to be the best way, although I do feel your unease...

I suppose in some circumstances it might make sense to modify the  
methods so that they never return undef but that won't always be  
convenient.  (Perhaps if the methods are tree operations that operate  
on sets, then children() could return a null set instead of undef, say,  
if there were no children.)  You don't get any short-circuiting this  
way though.



--M.

* * *
http://beebo.org


From scottp at myinternet.com.au  Wed Oct 30 17:28:47 2002
From: scottp at myinternet.com.au (Scott Penrose)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of Eval
Message-ID: <56D21C84-EC5F-11D6-8D4C-003065B58CF8@myinternet.com.au>

Hey Dudes,

A BIG HUGE thank you to all of your responses.

My conclusion is that I have been working in Perl without exceptions 
for so long that I had gotten into the habit of checking everything so 
often that I had ruled out using eval {}.

In my bad old programming days :-) (M$ Visual BASIC) the standard way 
of programming was to do a catch on exceptions for ALL errors. While 
this produces good output to the user, I found it annoying for 
debugging, and encouraged catching ALL errors that way, and in fact it 
was sometimes the only way to catch errors.

Using the Java Style Try, Catch approach, which is basically the same 
as doing the eval {} in the previous examples has given the best 
results not only on shorter code (= less bugs), more readable, and as 
Paul has discovered - FASTER. This in combination with a global DIE 
handler for pretty user output gives I think one of the best possible 
combinations.

I have played around a little with with Error, which is kind of cool, 
allows you to do

            try {
                do_some_stuff();
                die "error!" if $condition;
                throw Error::Simple -text => "Oops!" if $other_condition;
            }
            catch Error::IO with {
                my $E = shift;
                print STDERR "File ", $E->{'-file'}, " had a problem\n";
            }
            except {
                my $E = shift;
                my $general_handler=sub {send_message 
$E->{-description}};
                return {
                    UserException1 => $general_handler,
                    UserException2 => $general_handler
                };
            }
            otherwise {
                print STDERR "Well I don't know what to say\n";
            }
            finally {
                close_the_garage_door_already(); # Should be reliable
            }; # Don't forget the trailing ; or you might be surprised

But for my simple example that is probably overkill, both in look and 
feel and in performance.

But for bigger things, it is quite a neat structure, not totally 
dissimilar to what is being worked on in Perl 6.

Again, big thanks to all the feedback !

Scott
-- 
Scott Penrose
Open source developer
http://linux.dd.com.au/
scottp@dd.com.au

Dismaimer: Open sauce usually ends up never coming out (of the bottle).



From m.hudson at bluemail.ch  Thu Oct 31 06:56:33 2002
From: m.hudson at bluemail.ch (Sabrina Baretta)
Date: Wed Aug  4 00:02:43 2004
Subject: Low Cost Easy to Use Conferencing
Message-ID: <000041813c33$00007956$00005c42@bluegrass.dns2go.com>

An HTML attachment was scrubbed...
URL: http://mail.pm.org/archives/melbourne-pm/attachments/20021030/b3e11511/attachment.htm
From jarich at perltraining.com.au  Thu Oct 31 17:57:47 2002
From: jarich at perltraining.com.au (Jacinta Richardson)
Date: Wed Aug  4 00:02:43 2004
Subject: Perl Meetup day
Message-ID: <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au>


Wow, decent venues!  :)
The meetup thing is set for Thursday November the 21st and the venues
suggested are:
	* the Gin Palace (190 Lt Collins St., Melbourne)
		Probably the best martini bar in Melbourne (but bring lots
		of money)
	* Vegetarian Orgasm (117 Smith St, Fitzroy, Melbourne)
		Reasonable food, reasonable prices, fantastic desserts
	* Bliss Lounge Bar (11 Queen St., Melbourne)
		??
Meeting over several rounds of martinis sounds ideal to me. ;)  Anyone
else interested in an excuse to end up way way to happy on gin?

	Jacinta
--
   ("`-''-/").___..--''"`-._          |  Jacinta Richardson	    |
    `6_ 6  )   `-.  (     ).`-.__.`)  |  Perl Training Australia    |
    (_Y_.)'  ._   )  `._ `. ``-..-'   |      +613 9354 6001 	    |  
  _..`--'_..-_/  /--'_.' ,'           | contact@perltraining.com.au |
(il),-''  (li),'  ((!.-'              |   www.perltraining.com.au   |


From Adam.Clarke at StrategicData.com.au  Thu Oct 31 18:13:04 2002
From: Adam.Clarke at StrategicData.com.au (Adam Clarke)
Date: Wed Aug  4 00:02:43 2004
Subject: Perl Meetup day
In-Reply-To: <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au>
References: <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au>
Message-ID: <3DC1C710.4090004@StrategicData.com.au>

My understanding is that Gin has a reputation for producing exactly the 
opposite emotion :)

Jacinta Richardson wrote:

>		??
>Meeting over several rounds of martinis sounds ideal to me. ;)  Anyone
>else interested in an excuse to end up way way to happy on gin?
>
>	
>  
>



From pjf at perltraining.com.au  Thu Oct 31 18:25:00 2002
From: pjf at perltraining.com.au (Paul Fenwick)
Date: Wed Aug  4 00:02:43 2004
Subject: Perl Meetup day
In-Reply-To: <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au>
References: <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au>
Message-ID: <20021101002500.GB4523@mukc.org.au>

G'day Everyone,

On Fri, Nov 01, 2002 at 10:57:47AM +1100, Jacinta Richardson wrote:

> Wow, decent venues!  :)
> The meetup thing is set for Thursday November the 21st and the venues
> suggested are:
> 	* the Gin Palace (190 Lt Collins St., Melbourne)
> 		Probably the best martini bar in Melbourne (but bring lots
> 		of money)

Gin Palace rocks.  Er, except I had a few too many martinis there once.

> 	* Vegetarian Orgasm (117 Smith St, Fitzroy, Melbourne)
> 		Reasonable food, reasonable prices, fantastic desserts

Vege Orgasm++.

I'm feeling much more positive about this meet-up thing now.  :)

Cheers,

	Paul

-- 
Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
Director of Training                   | Ph:  +61 3 9354 6001
Perl Training Australia                | Fax: +61 3 9354 2681

From jhoward at fastmail.fm  Thu Oct 31 18:33:26 2002
From: jhoward at fastmail.fm (Jeremy Howard)
Date: Wed Aug  4 00:02:43 2004
Subject: Perl Meetup day
Message-ID: <20021101003326.994F3507782@server5.fastmail.fm>

$votes{gin_palace}++;

BTW, Charles Wright (IT journalist for The SMH and The Age, and weekly IT
spot on John Faine's radio program) has expressed an interest in coming
along to something to see what Melb PM is all about. It might end up as
an article.

I said he might like to see Damian talk, but it might be pretty techie
for 4-hours for a non-programmer... Whaddya think?

----- Original message -----
From: "Jacinta Richardson" <jarich@perltraining.com.au>
To: melbourne-pm@pm.org
Date: Fri, 1 Nov 2002 10:57:47 +1100 (EST)
Subject: Perl Meetup day


Wow, decent venues!  :)
The meetup thing is set for Thursday November the 21st and the venues
suggested are:
	* the Gin Palace (190 Lt Collins St., Melbourne)
		Probably the best martini bar in Melbourne (but bring lots
		of money)
	* Vegetarian Orgasm (117 Smith St, Fitzroy, Melbourne)
		Reasonable food, reasonable prices, fantastic desserts
	* Bliss Lounge Bar (11 Queen St., Melbourne)
		??
Meeting over several rounds of martinis sounds ideal to me. ;)	Anyone
else interested in an excuse to end up way way to happy on gin?

	Jacinta
--
   ("`-''-/").___..--''"`-._	      |  Jacinta Richardson	       |
    `6_ 6  )   `-.  (	  ).`-.__.`)  |  Perl Training Australia    |
    (_Y_.)'  ._   )  `._ `. ``-..-'   |      +613 9354 6001		| 
  _..`--'_..-_/  /--'_.' ,'	      | contact@perltraining.com.au |
(il),-''  (li),'  ((!.-'	      |   www.perltraining.com.au   |



-- 
  Jeremy Howard
  jhoward@fastmail.fm

From jens at cyber.com.au  Thu Oct 31 18:42:00 2002
From: jens at cyber.com.au (Jens Porup)
Date: Wed Aug  4 00:02:43 2004
Subject: Perl Meetup day
In-Reply-To: <20021101002500.GB4523@mukc.org.au>
References: <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au> <20021101002500.GB4523@mukc.org.au>
Message-ID: <20021101004200.GD6102@vanilla.office.cyber.com.au>

> > 	* Vegetarian Orgasm (117 Smith St, Fitzroy, Melbourne)
> > 		Reasonable food, reasonable prices, fantastic desserts
> 
> Vege Orgasm++.

You pay an awful lot of money for atmosphere at Veg Orgasm. 
I stopped going there when I discovered the vegetarian place
on Brunswick Street, where you get twice the food for half
the price, and the desserts are just as good--and cheaper.

Jens

> 
> I'm feeling much more positive about this meet-up thing now.  :)
> 
> Cheers,
> 
> 	Paul
> 
> -- 
> Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/
> Director of Training                   | Ph:  +61 3 9354 6001
> Perl Training Australia                | Fax: +61 3 9354 2681

From sharkey at zoic.org  Thu Oct 31 19:05:50 2002
From: sharkey at zoic.org (Nick 'Sharkey' Moore)
Date: Wed Aug  4 00:02:43 2004
Subject: Performance of chained methods.
In-Reply-To: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au>; from scottp@myinternet.com.au on Wed, Oct 30, 2002 at 02:48:12PM +1100
References: <69EFACCD-EBBA-11D6-8D4C-003065B58CF8@myinternet.com.au>
Message-ID: <20021101120549.B24022@dwerryhouse.com.au>

On Wed, Oct 30, 2002 at 02:48:12PM +1100, Scott Penrose wrote:
> 
> 	my $ref =  
> SomeClass->FirstMethod($param1)->AnotherMethod($param2)->MoreMethods();
>
> My problem is that each of these can fail. In my particular case I  
> don't care which fail or why, is it ok therefore to do this

Gonzo Suggestion:

-----
package QueryException;
use strict;

sub AUTOLOAD {
    return shift;
}

sub new {
    return bless \@_;
}

'fnord';
-----
package SomeClass;

sub FirstMethod {
    if (do_something()->success) {
        return UsefulClass::new(useful_thingy);
    } else {
	return QueryException::new('FirstMethod failed! Black helicopters!');
    }
}
-----
my $ref=SomeClass->FirstMethod($param1)->AnotherMethod($param2)->MoreMethods();

carp "Oh, golly" if ref($ref) eq "QueryException";
-----

Well, something along those lines anyway.

-----sharks

From jens at cyber.com.au  Thu Oct 31 20:38:18 2002
From: jens at cyber.com.au (Jens Porup)
Date: Wed Aug  4 00:02:43 2004
Subject: Open Source Victoria
In-Reply-To: <20021101004200.GD6102@vanilla.office.cyber.com.au>
References: <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au> <20021101002500.GB4523@mukc.org.au> <20021101004200.GD6102@vanilla.office.cyber.com.au>
Message-ID: <20021101023818.GJ6102@vanilla.office.cyber.com.au>

Greetings, all.

For those of you who don't read the LUV list:

Some of you may recall that about six months ago 20 or so
Java shops got together and formed Enterprise Java Victoria
which successfully applied for $80K from the Victorian 
government.

The second wave of Next Wave funding is upon us:

	http://www.mmv.vic.gov.au/nextwave

Cybersource is proposing to form an Open Source Consortium
to apply for this grant. The stated purpose of such a group
will be to promote and educate state government and business
leaders about the advantages of using Open Source/Free Software
tools. Such a Consortium would be non-technology specific,
e.g. we will promote Linux, BSDs, Zope, Perl, PHP, Apache,
e-Smith, etc with no prejudice to one particular solution.

We would, however, be especially and deliberate critical of
the government's recent decision to "save money" by spending
$80 million in Windows licenses over the next three years. 

The theory is that by raising awareness of Free Software
solutions in general we promote all Open Source shops
in specific.

If you own or work for an Open Source shop in Victoria,
and you're willing to contribute time, effort, and possibly
a modest amount of cash to this effort, then please join the
discussion on the newly created Open Source Victoria list.
To join, send an email to:

	osv-list-request@opensourcevictoria.cyber.com.au

	with a subject heading of "Subscribe"

If you have any problems with the list please contact
the administrators, myself on this address or Con Zymaris
on conz@cyber.com.au.

Cheers.

Jens



From ts at meme.com.au  Thu Oct 31 23:56:40 2002
From: ts at meme.com.au (Tony Smith)
Date: Wed Aug  4 00:02:43 2004
Subject: Perl Meetup day
In-Reply-To: 
 <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au>
References: 
 <Pine.LNX.3.96.1021101105404.4748B-100000@teddybear.perltraining.com.au>
Message-ID: <a05111b04b9e7c65a4cc6@[192.168.1.40]>

>	* the Gin Palace (190 Lt Collins St., Melbourne)
>		Probably the best martini bar in Melbourne (but bring lots
>		of money)

As long as everybody knows that Gin Palace isn't exactly in Little 
Collins but around the corner in a lane.

>	* Vegetarian Orgasm (117 Smith St, Fitzroy, Melbourne)
>		Reasonable food, reasonable prices, fantastic desserts

I only go for the mea^Het.

>	* Bliss Lounge Bar (11 Queen St., Melbourne)
>		??

The first Slashdot Meetup discovered that Bliss doesn't open till 8:30.

Guess I'm gonna havta vote again, even if it is getting a bit close 
to the silly season.

And my other vote is for tying Charles Wright to a chair for four 
hours of Damian. The only way to tackle our cultural cringe is head 
on.

-- 
Tony Smith
Developer
TransForum
http://www.transforum.net/