From rdice at pobox.com  Thu Oct  4 12:16:36 2007
From: rdice at pobox.com (Richard Dice)
Date: Thu, 4 Oct 2007 15:16:36 -0400
Subject: [tpm] Fwd: The Pittsburgh Perl Workshop 2007 is October 13-14, 2007.
In-Reply-To: <470523B6.8080902@robertblackwell.com>
References: <470523B6.8080902@robertblackwell.com>
Message-ID: <5bef4baf0710041216o1d99e997j42e25089eba3c789@mail.gmail.com>

Hi everyone,

PPW is coming up - it's good stuff!

Cheers,
Richard

---------- Forwarded message ----------
From: Robert Blackwell <robert at robertblackwell.com>
Date: Oct 4, 2007 1:32 PM
Subject: The Pittsburgh Perl Workshop 2007 is October 13-14, 2007.
To: ppw-announce at googlegroups.com



Please pass it along the Pittsburgh Perl Workshop is fast approaching.

The Pittsburgh Perl Mongers are pleased to announce The PITTSBURGH PERL
WORKSHOP, a two-day, low-cost conference on Saturday and Sunday, October
13-14, 2007.

The Pittsburgh Perl Workshop is an annual conference dedicated to the
Perl programming language. In 2006, the Pittsburgh Perl Mongers hosted
the first Perl Workshop based in the United States. This year, the
Workshop has been expanded to two days. The 2007 Workshop is structured
as a series of short lectures, but the atmosphere is low key and
engaging: the perfect combination to open your mind and then cram it
full of good stuff.

After last year's conference you gave us lots of feedback, and we listened.
* The workshop was expanded to a two-day event to allow for more talks,
birds-of-a-feather sessions, and social interactions.
* A one-day course for programmers with little or no Perl
experience---taught by a world-class Perl trainer---was added.
* The schedule has been improved to allow you more flexibility in
choosing sessions to attend.

Lightning Talks

There's still time to get a third of your fifteen minutes of fame!
Submit your lightning talk today. The deadline for early acceptance is
one week before the conference (October 6). But if you have an idea on
the first day of the conference, we're holding at least two lightning
talk spots until the end of that day. However, you have a much better
chance of being accepted if submit your talk now.

Stay up to date with everything that's going on with the Perl Workshop
by subscribing to our RSS feed at http://pghpw.org/ppw2007/atom/en.xml.

If you are coming to the Friday Social, RSVP at
http://pghpw.org/ppw2007/wiki?node=Friday%20Social.

Full details are on the Workshop Web site at http://pghpw.org.

Hope to see you there.
Robert Blackwell

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Pittsburgh Perl Workshop Announce" group.
To post to this group, send email to ppw-announce at googlegroups.com
To unsubscribe from this group, send email to
ppw-announce-unsubscribe at googlegroups.com
For more options, visit this group at
http://groups.google.com/group/ppw-announce?hl=en
-~----------~----~----~----~------~----~------~--~---
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071004/80733356/attachment.html 

From psema4 at gmail.com  Fri Oct  5 13:43:20 2007
From: psema4 at gmail.com (Scott Elcomb)
Date: Fri, 5 Oct 2007 16:43:20 -0400
Subject: [tpm] Is it possible to get a $coderef to an object method?
Message-ID: <99a6c38f0710051343s502f4397r71b1452c410f645a@mail.gmail.com>

Hi all,

I'd like to create a very lightweight MVC architecture using plugins.
I'm not sure if I'm barking up the right tree with this, but aside
from one issue it seems the idea should be doable.

I've looked at Maypole and Catalyst, but they have far too many
dependencies for my clients' tastes.  What I'm trying to do with the
"mvcplugs" POC is dead simple: drop a plugin object (containing a
combination of model/view/controller methods) into the plugin
directory ("MVCPlugs") and have it automatically load/register when
the main script ("mvcplugs.pl") runs.

Currently this doesn't do much beyond loading/registering an explicit
list of plugins.  The problem shows up when I try to register the
model/view/controller method coderefs in %registry.  In the output
sample (bottom of post) the coderefs showup as "DUMMY" subs.

Is there a way to get valid coderefs or am I headed in the wrong direction?

Thanks in advance,
- Scott.


#// START OF mvcplugs.pl //#
#!/usr/bin/perl

use strict;
use Data::Dumper;

###################################################
# CONFIG (Should read in from file or CLI args)
#
my @plugins = qw[Core PluginTemplate TestPlugin];
my $verbose = 1;
###################################################

my %registry = (
                 'plugins' => {},
                 'models' => {},
                 'views' => {},
                 'controllers' => {}
               );

foreach my $n (@plugins) {
  my $instantiatePlugin = "use MVCPlugs::$n; return MVCPlugs::$n" . '->new()';
  my $plugin = eval($instantiatePlugin);
  my $name = $plugin->name($n);

  %registry->{'plugins'}->{$name} = $plugin;

  if ($verbose) {
    print "[mvcplugs.pl] Plugin Instantiation String: $instantiatePlugin\n";
    print "[mvcplugs.pl] Plugin Instantiaion: $plugin\n";
  }

  if ( %registry->{'plugins'}->{$name}->initPlugin(\%registry) ) {
    print qq[Plugin $name initialized.\n];
  } else {
    print qq[Plugin $name initialization failed!\n];
  }

}

if ($verbose) {
  print "\n[mvcplugs.pl] Dumping Registry:\n" . Dumper(%registry) . "\n\n";
}
#// END OF mvcplugs.pl //#



#// START OF MVCPlugs/TestPlugin.pm //#
package MVCPlugs::TestPlugin;
use strict;
no strict 'refs';
use Data::Dumper;

sub new {
  my $self = {};
  $self->{'NAME'} = 'TEST_PLUGIN';
  bless($self);
  return $self;
}

sub name {
  my $self = shift;
  if (@_) { $self->{'NAME'} = shift }
  return $self->{'NAME'};
}

sub registry {
  my $self = shift;
  if (@_) { $self->{'REGISTRY'} = shift }
  return $self->{'REGISTRY'};
}

sub initPlugin {
  my $self = shift;
  my $registryref = shift;

  ######################
  # This block and the associated "coderefTest" methods are the only
  # differences from the Core.pm & PluginTemplate.pm sample plugins
  #
  # register a model, view, and controller
  $registryref->{'models'}->{'TestModel'} = \&coderefTest1;
  $registryref->{'views'}->{'TestView'} = sub { my $self = shift;
$self->coderefTest2(); };
  $registryref->{'controllers'}->{'TestController'} = \&coderefTest3;
  ######################

  return 1;
}

sub coderefTest1 {
  my $self = shift;
  return 'coderefTest1: call success!';
}

sub coderefTest2 {
  my $self = shift;
  return 'coderefTest2: call success!';
}

sub coderefTest3 {
  my $self = shift;
  return 'coderefTest3: call success!';
}

1;
#// END OF MVCPlugs/TestPlugin.pm //#


#// START OF Sample Output (VERBOSE=1)//#
[mvcplugs.pl] Plugin Instantiation String: use MVCPlugs::Core; return
MVCPlugs::Core->new()
[mvcplugs.pl] Plugin Instantiaion: MVCPlugs::Core=HASH(0x81d77e0)
Plugin Core initialized.
[mvcplugs.pl] Plugin Instantiation String: use
MVCPlugs::PluginTemplate; return MVCPlugs::PluginTemplate->new()
[mvcplugs.pl] Plugin Instantiaion: MVCPlugs::PluginTemplate=HASH(0x81d77f8)
Plugin PluginTemplate initialized.
[mvcplugs.pl] Plugin Instantiation String: use MVCPlugs::TestPlugin;
return MVCPlugs::TestPlugin->new()
[mvcplugs.pl] Plugin Instantiaion: MVCPlugs::TestPlugin=HASH(0x81d7810)
Plugin TestPlugin initialized.

[mvcplugs.pl] Dumping Registry:
$VAR1 = 'models';
$VAR2 = {
          'TestModel' => sub { "DUMMY" }
        };
$VAR3 = 'controllers';
$VAR4 = {
          'TestController' => sub { "DUMMY" }
        };
$VAR5 = 'views';
$VAR6 = {
          'TestView' => sub { "DUMMY" }
        };
$VAR7 = 'plugins';
$VAR8 = {
          'PluginTemplate' => bless( {
                                       'NAME' => 'PluginTemplate'
                                     }, 'MVCPlugs::PluginTemplate' ),
          'TestPlugin' => bless( {
                                   'NAME' => 'TestPlugin'
                                 }, 'MVCPlugs::TestPlugin' ),
          'Core' => bless( {
                             'NAME' => 'Core'
                           }, 'MVCPlugs::Core' )
        };


-- 
  Scott Elcomb
  http://www.psema4.com/

From sdmeisner at yahoo.com  Fri Oct  5 14:18:59 2007
From: sdmeisner at yahoo.com (Sean Meisner)
Date: Fri, 5 Oct 2007 14:18:59 -0700 (PDT)
Subject: [tpm] Is it possible to get a $coderef to an object method?
In-Reply-To: <99a6c38f0710051343s502f4397r71b1452c410f645a@mail.gmail.com>
Message-ID: <811218.18592.qm@web53010.mail.re2.yahoo.com>

The "DUMMY" string is actually coming from
Data::Dumper.  It won't show coderefs by default.  Try
setting the Deparse flag for Data::Dumper.

Cheers,

Sean


--- Scott Elcomb <psema4 at gmail.com> wrote:

> Hi all,
> 
> I'd like to create a very lightweight MVC
> architecture using plugins.
> I'm not sure if I'm barking up the right tree with
> this, but aside
> from one issue it seems the idea should be doable.
> 
> I've looked at Maypole and Catalyst, but they have
> far too many
> dependencies for my clients' tastes.  What I'm
> trying to do with the
> "mvcplugs" POC is dead simple: drop a plugin object
> (containing a
> combination of model/view/controller methods) into
> the plugin
> directory ("MVCPlugs") and have it automatically
> load/register when
> the main script ("mvcplugs.pl") runs.
> 
> Currently this doesn't do much beyond
> loading/registering an explicit
> list of plugins.  The problem shows up when I try to
> register the
> model/view/controller method coderefs in %registry. 
> In the output
> sample (bottom of post) the coderefs showup as
> "DUMMY" subs.
> 
> Is there a way to get valid coderefs or am I headed
> in the wrong direction?
> 
> Thanks in advance,
> - Scott.
> 
> 
> #// START OF mvcplugs.pl //#
> #!/usr/bin/perl
> 
> use strict;
> use Data::Dumper;
> 
> ###################################################
> # CONFIG (Should read in from file or CLI args)
> #
> my @plugins = qw[Core PluginTemplate TestPlugin];
> my $verbose = 1;
> ###################################################
> 
> my %registry = (
>                  'plugins' => {},
>                  'models' => {},
>                  'views' => {},
>                  'controllers' => {}
>                );
> 
> foreach my $n (@plugins) {
>   my $instantiatePlugin = "use MVCPlugs::$n; return
> MVCPlugs::$n" . '->new()';
>   my $plugin = eval($instantiatePlugin);
>   my $name = $plugin->name($n);
> 
>   %registry->{'plugins'}->{$name} = $plugin;
> 
>   if ($verbose) {
>     print "[mvcplugs.pl] Plugin Instantiation
> String: $instantiatePlugin\n";
>     print "[mvcplugs.pl] Plugin Instantiaion:
> $plugin\n";
>   }
> 
>   if (
>
%registry->{'plugins'}->{$name}->initPlugin(\%registry)
> ) {
>     print qq[Plugin $name initialized.\n];
>   } else {
>     print qq[Plugin $name initialization failed!\n];
>   }
> 
> }
> 
> if ($verbose) {
>   print "\n[mvcplugs.pl] Dumping Registry:\n" .
> Dumper(%registry) . "\n\n";
> }
> #// END OF mvcplugs.pl //#
> 
> 
> 
> #// START OF MVCPlugs/TestPlugin.pm //#
> package MVCPlugs::TestPlugin;
> use strict;
> no strict 'refs';
> use Data::Dumper;
> 
> sub new {
>   my $self = {};
>   $self->{'NAME'} = 'TEST_PLUGIN';
>   bless($self);
>   return $self;
> }
> 
> sub name {
>   my $self = shift;
>   if (@_) { $self->{'NAME'} = shift }
>   return $self->{'NAME'};
> }
> 
> sub registry {
>   my $self = shift;
>   if (@_) { $self->{'REGISTRY'} = shift }
>   return $self->{'REGISTRY'};
> }
> 
> sub initPlugin {
>   my $self = shift;
>   my $registryref = shift;
> 
>   ######################
>   # This block and the associated "coderefTest"
> methods are the only
>   # differences from the Core.pm & PluginTemplate.pm
> sample plugins
>   #
>   # register a model, view, and controller
>   $registryref->{'models'}->{'TestModel'} =
> \&coderefTest1;
>   $registryref->{'views'}->{'TestView'} = sub { my
> $self = shift;
> $self->coderefTest2(); };
>   $registryref->{'controllers'}->{'TestController'}
> = \&coderefTest3;
>   ######################
> 
>   return 1;
> }
> 
> sub coderefTest1 {
>   my $self = shift;
>   return 'coderefTest1: call success!';
> }
> 
> sub coderefTest2 {
>   my $self = shift;
>   return 'coderefTest2: call success!';
> }
> 
> sub coderefTest3 {
>   my $self = shift;
>   return 'coderefTest3: call success!';
> }
> 
> 1;
> #// END OF MVCPlugs/TestPlugin.pm //#
> 
> 
> #// START OF Sample Output (VERBOSE=1)//#
> [mvcplugs.pl] Plugin Instantiation String: use
> MVCPlugs::Core; return
> MVCPlugs::Core->new()
> [mvcplugs.pl] Plugin Instantiaion:
> MVCPlugs::Core=HASH(0x81d77e0)
> Plugin Core initialized.
> [mvcplugs.pl] Plugin Instantiation String: use
> MVCPlugs::PluginTemplate; return
> MVCPlugs::PluginTemplate->new()
> [mvcplugs.pl] Plugin Instantiaion:
> MVCPlugs::PluginTemplate=HASH(0x81d77f8)
> Plugin PluginTemplate initialized.
> [mvcplugs.pl] Plugin Instantiation String: use
> MVCPlugs::TestPlugin;
> return MVCPlugs::TestPlugin->new()
> [mvcplugs.pl] Plugin Instantiaion:
> MVCPlugs::TestPlugin=HASH(0x81d7810)
> Plugin TestPlugin initialized.
> 
> [mvcplugs.pl] Dumping Registry:
> $VAR1 = 'models';
> $VAR2 = {
>           'TestModel' => sub { "DUMMY" }
>         };
> $VAR3 = 'controllers';
> $VAR4 = {
>           'TestController' => sub { "DUMMY" }
>         };
> $VAR5 = 'views';
> $VAR6 = {
>           'TestView' => sub { "DUMMY" }
>         };
> $VAR7 = 'plugins';
> $VAR8 = {
>           'PluginTemplate' => bless( {
>                                        'NAME' =>
> 'PluginTemplate'
>                                      },
> 'MVCPlugs::PluginTemplate' ),
>           'TestPlugin' => bless( {
>                                    'NAME' =>
> 'TestPlugin'
>                                  },
> 'MVCPlugs::TestPlugin' ),
> 
=== message truncated ===


From psema4 at gmail.com  Fri Oct  5 15:03:29 2007
From: psema4 at gmail.com (Scott Elcomb)
Date: Fri, 5 Oct 2007 18:03:29 -0400
Subject: [tpm] Is it possible to get a $coderef to an object method?
In-Reply-To: <811218.18592.qm@web53010.mail.re2.yahoo.com>
References: <99a6c38f0710051343s502f4397r71b1452c410f645a@mail.gmail.com>
	<811218.18592.qm@web53010.mail.re2.yahoo.com>
Message-ID: <99a6c38f0710051503r70b62b6ao1276472b352b901@mail.gmail.com>

On 10/5/07, Sean Meisner <sdmeisner at yahoo.com> wrote:
> The "DUMMY" string is actually coming from
> Data::Dumper.  It won't show coderefs by default.  Try
> setting the Deparse flag for Data::Dumper.

Yup, that did it.  Many thanks.
- Scott.

-- 
  Scott Elcomb
  http://www.psema4.com/

From psema4 at gmail.com  Sun Oct  7 11:03:38 2007
From: psema4 at gmail.com (Scott Elcomb)
Date: Sun, 7 Oct 2007 14:03:38 -0400
Subject: [tpm] Is it possible to get a $coderef to an object method?
In-Reply-To: <99a6c38f0710051343s502f4397r71b1452c410f645a@mail.gmail.com>
References: <99a6c38f0710051343s502f4397r71b1452c410f645a@mail.gmail.com>
Message-ID: <99a6c38f0710071103q12ef2e57k5659a4f24377bff3@mail.gmail.com>

On 10/5/07, Scott Elcomb <psema4 at gmail.com> wrote:
> I'd like to create a very lightweight MVC architecture using plugins.

I haven't had a chance to write anything in regards to documentation -
on my way to a meeting with Turkey ;-) - but a working POC is
available[1].

Any and all comments regarding the source would be most appreciated;
I'm sure it could be written better.  If nobody minds, I'll have some
questions in the next few days about best-practice approaches.

Take care all, and happy Thanks Giving.  :-)
- Scott.


[1] Download: http://projects.psema4.com/mvcplugs/mvcplugs-0.0-7.tar.gz
[2] Live: http://projects.psema4.com/mvcplugs/mvcplugs.cgi?view=DefaultView2


-- 
  Scott Elcomb
  http://www.psema4.com/

From tom at legrady.ca  Sun Oct  7 15:26:39 2007
From: tom at legrady.ca (Tom Legrady)
Date: Sun, 7 Oct 2007 18:26:39 -0400
Subject: [tpm] Is it possible to get a $coderef to an object method?
In-Reply-To: <99a6c38f0710071103q12ef2e57k5659a4f24377bff3@mail.gmail.com>
References: <99a6c38f0710051343s502f4397r71b1452c410f645a@mail.gmail.com>
	<99a6c38f0710071103q12ef2e57k5659a4f24377bff3@mail.gmail.com>
Message-ID: <61CC3F9C-7DC0-4B04-AC90-7C89CEA55182@legrady.ca>

Enjoy your dead turkey day

Tom

On 7-Oct-07, at 2:03 PM, Scott Elcomb wrote:

> On 10/5/07, Scott Elcomb <psema4 at gmail.com> wrote:
>> I'd like to create a very lightweight MVC architecture using plugins.
>
> I haven't had a chance to write anything in regards to documentation -
> on my way to a meeting with Turkey ;-) - but a working POC is
> available[1].
>
> Any and all comments regarding the source would be most appreciated;
> I'm sure it could be written better.  If nobody minds, I'll have some
> questions in the next few days about best-practice approaches.
>
> Take care all, and happy Thanks Giving.  :-)
> - Scott.
>
>
> [1] Download: http://projects.psema4.com/mvcplugs/ 
> mvcplugs-0.0-7.tar.gz
> [2] Live: http://projects.psema4.com/mvcplugs/mvcplugs.cgi? 
> view=DefaultView2
>
>
> -- 
>   Scott Elcomb
>   http://www.psema4.com/
> _______________________________________________
> toronto-pm mailing list
> toronto-pm at pm.org
> http://mail.pm.org/mailman/listinfo/toronto-pm
>

Tom Legrady
tom at legrady.ca
My photo gallery ... http://picasaweb.google.com/legrady


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071007/d528f1ff/attachment.html 

From psema4 at gmail.com  Sun Oct  7 20:32:25 2007
From: psema4 at gmail.com (Scott Elcomb)
Date: Sun, 7 Oct 2007 23:32:25 -0400
Subject: [tpm] Is it possible to get a $coderef to an object method?
In-Reply-To: <61CC3F9C-7DC0-4B04-AC90-7C89CEA55182@legrady.ca>
References: <99a6c38f0710051343s502f4397r71b1452c410f645a@mail.gmail.com>
	<99a6c38f0710071103q12ef2e57k5659a4f24377bff3@mail.gmail.com>
	<61CC3F9C-7DC0-4B04-AC90-7C89CEA55182@legrady.ca>
Message-ID: <99a6c38f0710072032t35b3fe45xd831cc92dfcadae3@mail.gmail.com>

On 10/7/07, Tom Legrady <tom at legrady.ca> wrote:
> Enjoy your dead turkey day

Heh, thanks.  :-)

I love dead Turkey.  And dead plants.  And all the live "bugs"
(bacteria) that let me digest the dead stuff I get to eat.

Lol.  Yup, I thank all the live and dead things that let me keep on
keeping on.   Gotta hope that Simba'll be proud.  ;-)

-- 
  Scott Elcomb
  http://www.psema4.com/

From samogon at gmail.com  Tue Oct  9 09:23:42 2007
From: samogon at gmail.com (Ilia)
Date: Tue, 9 Oct 2007 12:23:42 -0400
Subject: [tpm] slides and source code for the Online Banking talk
Message-ID: <e66444870710090923ga92ccc6v48e66fc4cb8f0efc@mail.gmail.com>

Slides: http://nurey.com/bankelf_talk/bankelf_talk.pdf
Screencast: http://nurey.com/bankelf_talk/bankelf_screencast.divx
The perl code: http://svn.nurey.com/byob/
The python code: http://svn.nurey.com/public/python_workarea/

Finance::Bank::PC and Finance::Bank::CIBC are not yet on CPAN, pending
discussions at http://groups.google.com/group/finance-bank

ilia.

From jeff at pencom.com  Fri Oct 12 08:09:48 2007
From: jeff at pencom.com (Jeffrey Apelian)
Date: Fri, 12 Oct 2007 11:09:48 -0400
Subject: [tpm] Job opportunity
Message-ID: <470F8E3C.2090608@pencom.com>

Our client, is looking for a Perl Developer / Software Architect in 
their Production Build and Release Team. This team is engaged in several 
major projects involving the management of source code, build systems, 
and global software release cycles. They have a number of openings 
available in New York for Perl developers to work on new and exciting 
projects that will create the new architecture to manage my clients 
large-scale code bases, including both legacy code and new C and C++ 
componentized systems. Please note that these are not ?scripting? 
positions, but software engineering positions.

We are looking for experienced Perl developers with knowledge of 
software configuration and build engineering to develop and support the
existing build and release environment, as well as engineer brand new 
infrastructure to support the next generation of software development 
lifecycle
(SDLC) tools and processes. Team members have a desire to design and 
deploy serious Perl infrastructure in a large-scale industrial environment.
Qualifications

Bachelor's Degree in Computer Science, Engineering or Related or 
Equivalent Experience
4+ years experience writing reusable Perl libraries (object-oriented 
Perl a plus)
Good working knowledge of C and C++ development, in particular the use 
and configuration of compilers and linkers
An understanding of software design and architecture as well as the 
ability to render designs in code
Plus: Familiarity with the concept of physical design with refactoring 
large libraries, and with the basics of compiling and linking a large 
code base
Experience with source control (e.g. CVS, Subversion)
Experience with Unix platforms, especially Solaris, AIX, and Linux
Excellent written and verbal communication skills
The ability to work effectively as part of a team

-- 
Jeffrey Apelian
Pencom Systems Incorporated
40 Fulton Street, 18th Floor
New York, NY 10038
jeff at pencom.com
212-513-7777 ext. 286
http://www.pencom.com/

-- 
Jeffrey Apelian
Pencom Systems Incorporated
40 Fulton Street, 18th Floor
New York, NY  10038
jeff at pencom.com
212-513-7777 ext. 286
http://www.pencom.com/


From dimitrio at perlnow.com  Fri Oct 12 16:33:04 2007
From: dimitrio at perlnow.com (dmitri)
Date: Fri, 12 Oct 2007 19:33:04 -0400
Subject: [tpm] slides and source code for the Online Banking talk
In-Reply-To: <e66444870710090923ga92ccc6v48e66fc4cb8f0efc@mail.gmail.com>
References: <e66444870710090923ga92ccc6v48e66fc4cb8f0efc@mail.gmail.com>
Message-ID: <47100430.8090101@perlnow.com>

pretty cool stuff!

Ilia wrote:
> Slides: http://nurey.com/bankelf_talk/bankelf_talk.pdf
> Screencast: http://nurey.com/bankelf_talk/bankelf_screencast.divx
> The perl code: http://svn.nurey.com/byob/
> The python code: http://svn.nurey.com/public/python_workarea/
>
> Finance::Bank::PC and Finance::Bank::CIBC are not yet on CPAN, pending
> discussions at http://groups.google.com/group/finance-bank
>
> ilia.
> _______________________________________________
> toronto-pm mailing list
> toronto-pm at pm.org
> http://mail.pm.org/mailman/listinfo/toronto-pm
>
>
>   


-- 
Dimitri Ostapenko
LAMP developer/System Administrator
d at perlnow.com


From linux at alteeve.com  Tue Oct 16 12:53:51 2007
From: linux at alteeve.com (Madison Kelly)
Date: Tue, 16 Oct 2007 15:53:51 -0400
Subject: [tpm] File handle problem in a script called by another script
Message-ID: <471516CF.1060002@alteeve.com>

Hi all,

   I've got a program (main) that calls a second program (small), both 
written in perl. Both the main and small program access a common library.

   When I call the second, small program directly from the command line, 
it works fine. At one point it calls a subrouting from the library that 
opens a file, reads it and writes back out a modified version 
(specifically, it tracks various records and keeps track of a counter).

   Now, when I call the small program from within the main program, I 
get an error:

-=-=-=-=-
[<date>] small_script.pl: Use of uninitialized value in concatenation 
(.) or string at ./nmc.lib line 1741.
[<date>] small_script.pl: readline() on closed filehandle GEN16 at 
./nmc.lib line 1746.
-=-=-=-=-

   Here, nmc.lib is the library.

   I use 'IO::Handle' for creating the filehandle used to write out the 
file. Here is the specific block of code that generates the error:

nmc.lib lines 1069 - 1076
-=-=-=-=-
# Now re-write the file.
my $write_inc=IO::Handle->new();
open ($write_inc, ">$inc_file") || &error($conf, "error_0042", 1, 0, 
$this_file, __LINE__, $inc_file, $!);
for (@lines)
{
	print $write_inc $_;
}
$write_inc->close();
-=-=-=-=-

   If I comment out this section, the program works fine (but obviously, 
the file is not updated...). Again, if I run the small program directly, 
there is no problem, either.

   My *guess* is that for some reason, the filehandle created by the 
main program to read the output from the call to the second, small 
program must be clobbering the filehandle above. I can't understand why 
though, because they are different sections of code using different file 
handle names (though the code to call the second program IS in the same 
library).

   In case it matters, here is the code from the main program that calls 
the small program (cleaned up to make it more readable):

nmc.lib lines 2266 - 2284
-=-=-=-=-
my $gzf=IO::Handle->new();
my $shell_call=?/path/to/small_script.pl 2>&1 |";
# write to log what I am calling...
open ($gzf, $shell_call) || die...;
while (<$gzf>)
{
	chomp;
	my $line=$_;
	# write to log this line...
		
	$line=~s/\[(.*?)\]/\[<span class=\"tech\">$1<\/span>\]/g;
	my %value=(
		replace_0	=>	"",
		replace_1	=>	"$line",
	);
	&print_template($conf, "tool_item", $$conf{template_ga}, \%value); 
%value=();

}
$gzf->close();
-=-=-=-=-

   Any idea why the second example code could be clobbering the first 
block of code (if that is what is even happening)?

Thanks!!

Madi

From indy at indigostar.com  Tue Oct 16 13:19:33 2007
From: indy at indigostar.com (Indy Singh)
Date: Tue, 16 Oct 2007 16:19:33 -0400
Subject: [tpm] File handle problem in a script called by another script
References: <471516CF.1060002@alteeve.com>
Message-ID: <027f01c81031$ddc53530$6600a8c0@roadhog>

The line numbers that you supplied code for (1069 - 1076) don't cover
where you say the error is (line 1741)
Assuming that that is just an oversight.

It would probably be a good idea to print out the content of $inc_file
to verify that it actually contains a valid file name and path.
How do you know that there is something actually in @lines ?
A print statement in the loop would be a good idea.
How do you know that your error subroutine does actually print out
something?
How do you know that your error does NOT print out something?
How do you know that your error does not return (and therby cause a
cascading error).

Rewrite the code, add print statements, remove calls to external
functions if you cal (error, IO::Handle).



Indy Singh
IndigoSTAR Software -- www.indigostar.com


----- Original Message ----- 
From: "Madison Kelly" <linux at alteeve.com>
To: "TPM Mailing List" <tpm at to.pm.org>
Sent: Tuesday, October 16, 2007 3:53 PM
Subject: [tpm] File handle problem in a script called by another script


Hi all,

   I've got a program (main) that calls a second program (small), both
written in perl. Both the main and small program access a common
library.

   When I call the second, small program directly from the command line,
it works fine. At one point it calls a subrouting from the library that
opens a file, reads it and writes back out a modified version
(specifically, it tracks various records and keeps track of a counter).

   Now, when I call the small program from within the main program, I
get an error:

-=-=-=-=-
[<date>] small_script.pl: Use of uninitialized value in concatenation
(.) or string at ./nmc.lib line 1741.
[<date>] small_script.pl: readline() on closed filehandle GEN16 at
./nmc.lib line 1746.
-=-=-=-=-

   Here, nmc.lib is the library.

   I use 'IO::Handle' for creating the filehandle used to write out the
file. Here is the specific block of code that generates the error:

nmc.lib lines 1069 - 1076
-=-=-=-=-
# Now re-write the file.
my $write_inc=IO::Handle->new();
open ($write_inc, ">$inc_file") || &error($conf, "error_0042", 1, 0,
$this_file, __LINE__, $inc_file, $!);
for (@lines)
{
print $write_inc $_;
}
$write_inc->close();
-=-=-=-=-

   If I comment out this section, the program works fine (but obviously,
the file is not updated...). Again, if I run the small program directly,
there is no problem, either.

   My *guess* is that for some reason, the filehandle created by the
main program to read the output from the call to the second, small
program must be clobbering the filehandle above. I can't understand why
though, because they are different sections of code using different file
handle names (though the code to call the second program IS in the same
library).

   In case it matters, here is the code from the main program that calls
the small program (cleaned up to make it more readable):

nmc.lib lines 2266 - 2284
-=-=-=-=-
my $gzf=IO::Handle->new();
my $shell_call=?/path/to/small_script.pl 2>&1 |";
# write to log what I am calling...
open ($gzf, $shell_call) || die...;
while (<$gzf>)
{
chomp;
my $line=$_;
# write to log this line...

$line=~s/\[(.*?)\]/\[<span class=\"tech\">$1<\/span>\]/g;
my %value=(
replace_0 => "",
replace_1 => "$line",
);
&print_template($conf, "tool_item", $$conf{template_ga}, \%value);
%value=();

}
$gzf->close();
-=-=-=-=-

   Any idea why the second example code could be clobbering the first
block of code (if that is what is even happening)?

Thanks!!

Madi
_______________________________________________
toronto-pm mailing list
toronto-pm at pm.org
http://mail.pm.org/mailman/listinfo/toronto-pm


From linux at alteeve.com  Tue Oct 16 13:39:48 2007
From: linux at alteeve.com (Madison Kelly)
Date: Tue, 16 Oct 2007 16:39:48 -0400
Subject: [tpm] File handle problem in a script called by another script
In-Reply-To: <027f01c81031$ddc53530$6600a8c0@roadhog>
References: <471516CF.1060002@alteeve.com>
	<027f01c81031$ddc53530$6600a8c0@roadhog>
Message-ID: <47152194.70503@alteeve.com>

Indy Singh wrote:
> The line numbers that you supplied code for (1069 - 1076) don't cover
> where you say the error is (line 1741)
> Assuming that that is just an oversight.
> 
> It would probably be a good idea to print out the content of $inc_file
> to verify that it actually contains a valid file name and path.
> How do you know that there is something actually in @lines ?
> A print statement in the loop would be a good idea.
> How do you know that your error subroutine does actually print out
> something?
> How do you know that your error does NOT print out something?
> How do you know that your error does not return (and therby cause a
> cascading error).
> 
> Rewrite the code, add print statements, remove calls to external
> functions if you cal (error, IO::Handle).
> 
> 
> 
> Indy Singh
> IndigoSTAR Software -- www.indigostar.com
> 
> 
> ----- Original Message ----- From: "Madison Kelly" <linux at alteeve.com>
> To: "TPM Mailing List" <tpm at to.pm.org>
> Sent: Tuesday, October 16, 2007 3:53 PM
> Subject: [tpm] File handle problem in a script called by another script
> 
> 
> Hi all,
> 
>   I've got a program (main) that calls a second program (small), both
> written in perl. Both the main and small program access a common
> library.
> 
>   When I call the second, small program directly from the command line,
> it works fine. At one point it calls a subrouting from the library that
> opens a file, reads it and writes back out a modified version
> (specifically, it tracks various records and keeps track of a counter).
> 
>   Now, when I call the small program from within the main program, I
> get an error:
> 
> -=-=-=-=-
> [<date>] small_script.pl: Use of uninitialized value in concatenation
> (.) or string at ./nmc.lib line 1741.
> [<date>] small_script.pl: readline() on closed filehandle GEN16 at
> ./nmc.lib line 1746.
> -=-=-=-=-
> 
>   Here, nmc.lib is the library.
> 
>   I use 'IO::Handle' for creating the filehandle used to write out the
> file. Here is the specific block of code that generates the error:
> 
> nmc.lib lines 1069 - 1076
> -=-=-=-=-
> # Now re-write the file.
> my $write_inc=IO::Handle->new();
> open ($write_inc, ">$inc_file") || &error($conf, "error_0042", 1, 0,
> $this_file, __LINE__, $inc_file, $!);
> for (@lines)
> {
> print $write_inc $_;
> }
> $write_inc->close();
> -=-=-=-=-
> 
>   If I comment out this section, the program works fine (but obviously,
> the file is not updated...). Again, if I run the small program directly,
> there is no problem, either.
> 
>   My *guess* is that for some reason, the filehandle created by the
> main program to read the output from the call to the second, small
> program must be clobbering the filehandle above. I can't understand why
> though, because they are different sections of code using different file
> handle names (though the code to call the second program IS in the same
> library).
> 
>   In case it matters, here is the code from the main program that calls
> the small program (cleaned up to make it more readable):
> 
> nmc.lib lines 2266 - 2284
> -=-=-=-=-
> my $gzf=IO::Handle->new();
> my $shell_call=?/path/to/small_script.pl 2>&1 |";
> # write to log what I am calling...
> open ($gzf, $shell_call) || die...;
> while (<$gzf>)
> {
> chomp;
> my $line=$_;
> # write to log this line...
> 
> $line=~s/\[(.*?)\]/\[<span class=\"tech\">$1<\/span>\]/g;
> my %value=(
> replace_0 => "",
> replace_1 => "$line",
> );
> &print_template($conf, "tool_item", $$conf{template_ga}, \%value);
> %value=();
> 
> }
> $gzf->close();
> -=-=-=-=-
> 
>   Any idea why the second example code could be clobbering the first
> block of code (if that is what is even happening)?
> 
> Thanks!!
> 
> Madi
> _______________________________________________
> toronto-pm mailing list
> toronto-pm at pm.org
> http://mail.pm.org/mailman/listinfo/toronto-pm
> 

GAH!

Thanks both, turns out it was all fine and the problem was file 
permissions....

I'll go head->desk now.

Sorry for the line noise!

Madi

From sz_2 at hotmail.com  Wed Oct 17 14:04:35 2007
From: sz_2 at hotmail.com (s z)
Date: Wed, 17 Oct 2007 21:04:35 +0000
Subject: [tpm] (no subject)
Message-ID: <BAY104-W42C9D3B70D1011720518B7AB9D0@phx.gbl>


Hi,  this is a newbie's question and I would like to get some advice from you:
 
I am using perl to develop an application. I am able to read and write through ODBC to dbf files by using dbd-odbc. The script generates the html files for user interface like tables,  input textboxes and submit buttons. when user enters new values into textboxes and press button to save, the form is submited and user's input is saved to the data file. This is done by calling a different perl script file. The screen goes blank afterwards. It would be nice to redisplay the input boxes and the buttons on the screen to let user to continue the editing.  But how to call the perl display routine from another perl script? I am thinking to reuse the perl code from the previous routine in the update routine. Is there a better way to reuse the perl codes? How?
Thanks a lot for your advice.
 
Sam
_________________________________________________________________
Send a smile, make someone laugh, have some fun! Start now!
http://www.freemessengeremoticons.ca/?icid=EMENCA122
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071017/fa3535be/attachment.html 

From sfryer at sourcery.ca  Wed Oct 17 14:23:03 2007
From: sfryer at sourcery.ca (Shaun Fryer)
Date: Wed, 17 Oct 2007 17:23:03 -0400
Subject: [tpm] (no subject)
In-Reply-To: <BAY104-W42C9D3B70D1011720518B7AB9D0@phx.gbl>
References: <BAY104-W42C9D3B70D1011720518B7AB9D0@phx.gbl>
Message-ID: <20071017212302.GA29123@sourcery.ca>

On Wed, Oct 17, 2007 at 09:04:35PM +0000, s z wrote:
> 
> Hi,  this is a newbie's question and I would like to get some advice from you:
>  
> I am using perl to develop an application. I am able to read and write through ODBC to dbf files by using dbd-odbc. The script generates the html files for user interface like tables,  input textboxes and submit buttons. when user enters new values into textboxes and press button to save, the form is submited and user's input is saved to the data file. This is done by calling a different perl script file. The screen goes blank afterwards. It would be nice to redisplay the input boxes and the buttons on the screen to let user to continue the editing.  But how to call the perl display routine from another perl script? I am thinking to reuse the perl code from the previous routine in the update routine. Is there a better way to reuse the perl codes? How?
> Thanks a lot for your advice.
>  
> Sam

Hi Sam,

If the script which recieves the form submission does not output
any other data, you might consider sending an HTTP redirect back
to the first script. There are a number of ways to do this, but
perhaps the very simplest (and for which I'll likely get flamed)
is to print the following line (assuming your webserver doesn't
do headers for you automatically).

print "Location: http://domain.tld/path/to/your/first.cgi\n\n";

You can do this either before or after your database call,
depending on the speed of the application and whether or not you
might need to output an error to the user in case of problems.
If you are using the CGI module or another variant such as
Apache::whatever, then you should read the API docs for info on
how to do this.
--
    Shaun Fryer

From sz_2 at hotmail.com  Wed Oct 17 19:01:11 2007
From: sz_2 at hotmail.com (s z)
Date: Thu, 18 Oct 2007 02:01:11 +0000
Subject: [tpm] CGI
Message-ID: <BAY104-W29D73C9D94A1D61061E802AB9E0@phx.gbl>


this is a second post. Please comment on the issue. I appreciate your help. Thank you in advance.
 
Hi,  this is a newbie's question and I would like to get some advice from you: I am using perl to develop an application. I am able to read and write through ODBC to dbf files by using dbd-odbc. The script generates the html files for user interface like tables,  input textboxes and submit buttons. when user enters new values into textboxes and press button to save, the form is submited and user's input is saved to the data file. This is done by calling a different perl script file. The screen goes blank afterwards. It would be nice to redisplay the input boxes and the buttons on the screen to let user to continue the editing.  But how to call the perl display routine from another perl script? I am thinking to reuse the perl code from the previous routine in the update routine. Is there a better way to reuse the perl codes? How?Thanks a lot for your advice. Sam
_________________________________________________________________
Send a smile, make someone laugh, have some fun! Start now!
http://www.freemessengeremoticons.ca/?icid=EMENCA122
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071018/ae5b1bfc/attachment.html 

From liam at holoweb.net  Wed Oct 17 21:47:32 2007
From: liam at holoweb.net (Liam R E Quin)
Date: Thu, 18 Oct 2007 00:47:32 -0400
Subject: [tpm] CGI
In-Reply-To: <BAY104-W29D73C9D94A1D61061E802AB9E0@phx.gbl>
References: <BAY104-W29D73C9D94A1D61061E802AB9E0@phx.gbl>
Message-ID: <1192682852.15058.20.camel@dell.barefootcomputing.com>

On Thu, 2007-10-18 at 02:01 +0000, s z wrote:
> this is a second post. Please comment on the issue. I appreciate your
> help. Thank you in advance.

You already had one reply I think.

Let's take a higher-level look at the problem.

States are:
[1] issue blank form to user
[2] receive data
    2a data is not OK, go back to [1] but with correct fields still
    there
    2b data is OK, proceed to [3]
*important* people can edit data in Web forms, never ever trust it!
never ever display user data in error messages unless you escape
all special characters like < & > ' " where needed.  Otherwise
your site is vulnerable to a cross-site scripting attack, or
can be used to attack another site, or can be used to host links
to porn sites inside those error messages!

[3] data is OK, update the dataqbase
    3a failed, go back to [1] with an explanation
    3b OK, so proceed to [1] with a note that everything was OK

You can do all this with a single CGI script.  I make it easier
for myself with a hidden input field that supplies a state parameter,
although you still need to check the data is OK of course, as someone
can always add the parameter right there in the URL!

mypage.cgi looks like

if (data_was_ok) {
    initial_page(message)
} elsif (got_data) {
    if (data_is_bad) {
        initial_page("data was bad"); # do better than this!
    } elsif (store_data()) {
        initial_page("data stored.");
    } else {
        initial_page("failed to store data");
   }
} else {
    initial_page()
}

Now you only need one program and the logic is fairly clear.

The initial_page() routine would of course send the HTML forms,
populated with data if supplied, and with an optional message at the
start.

If you really do end up wanting to share code between different scripts,
you may want to learn about modules, although an alternative is just
to call the other script with system() (if you are VERY careful not
to pass command-line arguments, or to quote them properly, so that
a value of `/bin/rm -rf /` does not delete every file on your server...
people WILL try this sort of thing)

Liam


-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org www.advogato.org


From indy at indigostar.com  Mon Oct 22 13:33:32 2007
From: indy at indigostar.com (Indy Singh)
Date: Mon, 22 Oct 2007 16:33:32 -0400
Subject: [tpm] Source of when Perl features added
References: <BAY104-W29D73C9D94A1D61061E802AB9E0@phx.gbl>
	<1192682852.15058.20.camel@dell.barefootcomputing.com>
Message-ID: <036b01c814ea$d0416ab0$6600a8c0@roadhog>

Hello folks,

Anybody have any idea where one can find a list of what Perl features 
were added in what release?

I am not interested in ancient versions just ones that one might run 
into in a production environment.  Also not looking for every bug or 
feature that was added in every release just the major items.

Some of the specific things I am interested in are:
What release introduced 'use warnings'?
What release introduced the 'our' variable declaration?
What release allowed calls to undeclared user subs without a leading &.

Indy Singh
IndigoSTAR Software -- www.indigostar.com


From rdice at pobox.com  Mon Oct 22 13:44:52 2007
From: rdice at pobox.com (Richard Dice)
Date: Mon, 22 Oct 2007 16:44:52 -0400
Subject: [tpm] Source of when Perl features added
In-Reply-To: <036b01c814ea$d0416ab0$6600a8c0@roadhog>
References: <BAY104-W29D73C9D94A1D61061E802AB9E0@phx.gbl>
	<1192682852.15058.20.camel@dell.barefootcomputing.com>
	<036b01c814ea$d0416ab0$6600a8c0@roadhog>
Message-ID: <5bef4baf0710221344h5fb0441fq51a8076baf26737e@mail.gmail.com>

Hi Indy,

Is what you're looking for different from "perldelta"?  E.g.
http://search.cpan.org/~nwclark/perl-5.8.8/pod/perl588delta.pod

You can see various perldeta files for various minor-numbered releases at:
http://search.cpan.org/~nwclark/perl-5.8.8/

The "even more detailed" file I think you were trying to ignore is -
http://search.cpan.org/src/NWCLARK/perl-5.8.8/Changes

Cheers,
Richard

On 10/22/07, Indy Singh <indy at indigostar.com> wrote:
>
> Hello folks,
>
> Anybody have any idea where one can find a list of what Perl features
> were added in what release?
>
> I am not interested in ancient versions just ones that one might run
> into in a production environment.  Also not looking for every bug or
> feature that was added in every release just the major items.
>
> Some of the specific things I am interested in are:
> What release introduced 'use warnings'?
> What release introduced the 'our' variable declaration?
> What release allowed calls to undeclared user subs without a leading &.
>
> Indy Singh
> IndigoSTAR Software -- www.indigostar.com
>
> _______________________________________________
> toronto-pm mailing list
> toronto-pm at pm.org
> http://mail.pm.org/mailman/listinfo/toronto-pm
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071022/596c7139/attachment.html 

From liam at holoweb.net  Mon Oct 22 13:41:38 2007
From: liam at holoweb.net (Liam R E Quin)
Date: Mon, 22 Oct 2007 16:41:38 -0400
Subject: [tpm] Source of when Perl features added
In-Reply-To: <036b01c814ea$d0416ab0$6600a8c0@roadhog>
References: <BAY104-W29D73C9D94A1D61061E802AB9E0@phx.gbl>
	<1192682852.15058.20.camel@dell.barefootcomputing.com>
	<036b01c814ea$d0416ab0$6600a8c0@roadhog>
Message-ID: <1193085698.8132.24.camel@dell.barefootcomputing.com>

On Mon, 2007-10-22 at 16:33 -0400, Indy Singh wrote:

> Some of the specific things I am interested in are:
> What release introduced 'use warnings'?
uncertain -- perl -w was much older and is safe, e.g. in the #! line

> What release introduced the 'our' variable declaration?
Perl 5 but not sure which release.
> What release allowed calls to undeclared user subs without a leading &.
Perl 5, some time before 1998

But I don't have a general algorithm for answering the question, sorry.

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org www.advogato.org


From magog at the-wire.com  Tue Oct 23 10:14:08 2007
From: magog at the-wire.com (Michael Graham)
Date: Tue, 23 Oct 2007 13:14:08 -0400
Subject: [tpm] October Meeting - This Thursday - 25 Oct,
	2007 - Richard's Summer	of Perl
Message-ID: <20071023131408.g44jz23kg0wgwgkw@webmail.the-wire.com>


(These details are also on the TPM web site: http://to.pm.org/)

The next meeting is this Thursday (25 Oct).


     Date:   Thursday 25 Oct 2007

     Time:   6:45pm

    Where:   2 Bloor Street West (NW corner of Yonge/Bloor, skyscraper
             with the CIBC logo on top) Classroom TBA.

   ===================================================================

Talk Details:

     Speaker: Richard Dice

       Title: Richard's Summer of Perl

Description:

              Between May and October Richard did much Perl stuff.
              Conferences attended:  YAPC::NA, OSCON, YAPC::EU and PPW.

              He did a few other Perl things which are hard to categorize
              too.  Come to this talk to hear confessions of a Perl
              addict.  Richard promises only the bare minimum in
              organization and preparation, but will probably have no
              trouble in filling the time allotted, however much that
              time may be before you collectively shout him down.  He
              hopes this talk will be entertaining if informal, maybe
              marginally useful, and at least not overly personally
              embarrassing.



   ===================================================================

   Note:

     The elevators in the building are "locked down" after 5:30pm
     to people without building access cards.  Leading up to the
     meeting someone will come down to the main floor lobby every
     few minutes to ferry people upstairs.

     After 19:00, you can reach the access-card-carrying guy via
     a cell phone number that we'll leave with security in the
     front lobby.  The room and floor numbers will be left with
     security too.





-- 
Michael Graham <magog at the-wire.com>





From michael.a.bolton at gmail.com  Wed Oct 24 07:53:59 2007
From: michael.a.bolton at gmail.com (Michael Bolton)
Date: Wed, 24 Oct 2007 10:53:59 -0400
Subject: [tpm] David Platt Speaking Virtually At TASSQ
Message-ID: <004801c8164d$bc89a7d0$5100000a@Koko>

Hi, folks...

One of my favourite talks over the spring conference season was David
Platt's keynote address at SD West, on the subject of his book, "Why
Software Sucks".  We're lucky to have David presenting that talk to us via
real-time video hookup next Tuesday, October 30, at the Toronto Association
of System and Software Quality.

You can get information and register for the event at http://www.tassq.org.

Cheers,

---Michael B.

TASSQ - The Toronto Association of System and Software Quality
http://www.tassq.org
 
Michael Bolton
DevelopSense:  Software Testing in Plain English
Web Site:  http://www.developsense.com <http://www.developsense.com> 
Newsletter: addme at developsense.com <mailto:addme at developsense.com> 
 
 
 


From michael.a.bolton at gmail.com  Wed Oct 24 12:11:05 2007
From: michael.a.bolton at gmail.com (Michael Bolton)
Date: Wed, 24 Oct 2007 15:11:05 -0400
Subject: [tpm] David Platt Speaking Virtually At TASSQ
Message-ID: <009801c81671$a0de0700$5100000a@Koko>

Hi, folks...

One of my favourite talks over the spring conference season was David
Platt's keynote address at SD West, on the subject of his book, "Why
Software Sucks".  We're lucky to have David presenting that talk to us via
real-time video hookup next Tuesday, October 30, at the Toronto Association
of System and Software Quality.

You can get information and register for the event at http://www.tassq.org.

Cheers,

---Michael B.

TASSQ - The Toronto Association of System and Software Quality
http://www.tassq.org
 
Michael Bolton
DevelopSense:  Software Testing in Plain English Web Site:
http://www.developsense.com <http://www.developsense.com>
Newsletter: addme at developsense.com <mailto:addme at developsense.com> 
 
 
 

_______________________________________________
toronto-pm mailing list
toronto-pm at pm.org
http://mail.pm.org/mailman/listinfo/toronto-pm


From magog at the-wire.com  Wed Oct 24 12:46:44 2007
From: magog at the-wire.com (Michael Graham)
Date: Wed, 24 Oct 2007 15:46:44 -0400
Subject: [tpm] Need someone to guard room for tomorrow's meeting
Message-ID: <20071024154644.68b72ebe@caliope>


Is anybody able to come early to tomorrow's meeting?

We need someone to arrive in the room around 4:00 pm and stay there
until the meeting starts (to make sure that the cleaners don't lock the
room).


The location is:  Classroom 11 on the 8th floor


Michael





-- 
Michael Graham <magog at the-wire.com>

From arocker at vex.net  Wed Oct 24 13:35:56 2007
From: arocker at vex.net (arocker at vex.net)
Date: Wed, 24 Oct 2007 16:35:56 -0400 (EDT)
Subject: [tpm] Need someone to guard room for tomorrow's meeting
In-Reply-To: <20071024154644.68b72ebe@caliope>
References: <20071024154644.68b72ebe@caliope>
Message-ID: <41853.192.30.202.28.1193258156.squirrel@webmail.vex.net>

> We need someone to arrive in the room around 4:00 pm and stay there
> until the meeting starts (to make sure that the cleaners don't lock the
> room).
>
If the room is locked when you arrive, get the key from LaVerne (on the
12th floor), unlock it, and guard it.


From arocker at vex.net  Wed Oct 24 13:38:03 2007
From: arocker at vex.net (arocker at vex.net)
Date: Wed, 24 Oct 2007 16:38:03 -0400 (EDT)
Subject: [tpm] Need someone to guard room for tomorrow's meeting
In-Reply-To: <20071024154644.68b72ebe@caliope>
References: <20071024154644.68b72ebe@caliope>
Message-ID: <34098.192.30.202.29.1193258283.squirrel@webmail.vex.net>


P.S. LaVerne goes home at 16:30, which is why you have to arrive before then.


From magog at the-wire.com  Thu Oct 25 13:02:33 2007
From: magog at the-wire.com (Michael Graham)
Date: Thu, 25 Oct 2007 16:02:33 -0400
Subject: [tpm] Need someone to guard room for tomorrow's meeting
In-Reply-To: <20071024154644.68b72ebe@caliope>
References: <20071024154644.68b72ebe@caliope>
Message-ID: <20071025160233.627152f1@caliope>


Okay, I'm going to try to get there by 4:30.


Michael



On Wed, 24 Oct 2007 15:46:44 -0400
Michael Graham <magog at the-wire.com> wrote:

> 
> Is anybody able to come early to tomorrow's meeting?
> 
> We need someone to arrive in the room around 4:00 pm and stay there
> until the meeting starts (to make sure that the cleaners don't lock
> the room).
> 
> 
> The location is:  Classroom 11 on the 8th floor
> 
> 
> Michael
> 
> 
> 
> 
> 


-- 
Michael Graham <magog at the-wire.com>

From magog at the-wire.com  Thu Oct 25 13:48:34 2007
From: magog at the-wire.com (Michael Graham)
Date: Thu, 25 Oct 2007 16:48:34 -0400
Subject: [tpm] Need someone to guard room for tomorrow's meeting
In-Reply-To: <20071025160233.627152f1@caliope>
References: <20071024154644.68b72ebe@caliope> <20071025160233.627152f1@caliope>
Message-ID: <20071025164834.7442edbd@caliope>



> Okay, I'm going to try to get there by 4:30.

I'm here and guarding the room.  All is well.


Michael



On Thu, 25 Oct 2007 16:02:33 -0400
Michael Graham <magog at the-wire.com> wrote:

> 
> Okay, I'm going to try to get there by 4:30.
> 
> 
> Michael
> 
> 
> 
> On Wed, 24 Oct 2007 15:46:44 -0400
> Michael Graham <magog at the-wire.com> wrote:
> 
> > 
> > Is anybody able to come early to tomorrow's meeting?
> > 
> > We need someone to arrive in the room around 4:00 pm and stay there
> > until the meeting starts (to make sure that the cleaners don't lock
> > the room).
> > 
> > 
> > The location is:  Classroom 11 on the 8th floor
> > 
> > 
> > Michael
> > 
> > 
> > 
> > 
> > 
> 
> 


-- 
Michael Graham <magog at the-wire.com>

From jim.graham at jim-graham.net  Fri Oct 26 08:12:13 2007
From: jim.graham at jim-graham.net (Jim  Graham)
Date: Fri, 26 Oct 2007 11:12:13 -0400
Subject: [tpm] Some links from last night's meeting
Message-ID: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>

Hi All,

  here are 2 links about things brought up at last night's meeting

  -- Steve Yegge's Keynote at OSCON 2007 about Open Source Marketing

    http://blip.tv/file/319044/

  -- Reddit discussion of Lua benchmarks on Parrot. This was originally a link to the perl6 internals mailing list with Reddit comments. Most of the discussion comes back to 'completeness first, optimization second'

    http://programming.reddit.com/info/2z05a/comments/

  Regards, Jim


From fulko.hew at gmail.com  Fri Oct 26 08:46:42 2007
From: fulko.hew at gmail.com (Fulko Hew)
Date: Fri, 26 Oct 2007 11:46:42 -0400
Subject: [tpm] Some links from last night's meeting
In-Reply-To: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
Message-ID: <8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>

On 10/26/07, Jim Graham <jim.graham at jim-graham.net> wrote:
>
>
>   -- Reddit discussion of Lua benchmarks on Parrot. This was originally a
> link to the perl6 internals mailing list with Reddit comments. Most of the
> discussion comes back to 'completeness first, optimization second'
>
>     http://programming.reddit.com/info/2z05a/comments/



I read that stuff, and I know of Lua (in passing where its embedded in
Wireshark)

My comment on the Reddit discussion, (based on last night's discussion) is:
...Lua is optimized for Lua,
...Parrot is not optimized for Lua,
...Parrot is optimized for Parrot.

So I'm not surprised that Lua code running on a highly
(Lua) optimized 'system' is faster than Lua code
compiled to run on Parrot.

I would suspect there is a one to one mapping between Lua
constructs and the Lua VM, whereas the Lua on Parrot compiler
probably isn't designed with that in mind and hence isn't one to
one.

The corollary is that Lua probably doesn't have a one to one
mapping of Perl constructs either, so that a perl compiler
written in Lua probably won't run as fast as perl on parrot.

"A screwdriver it great with screws, and can be used as a so-so chisel.
A chisel is a great chisel, and can be used as a so-so screwdriver."

Does that make screwdrivers inefficient?
Does that make chisels inefficient?

I'll leave that as a rhetorical question.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071026/5dd52286/attachment.html 

From jkeen at verizon.net  Fri Oct 26 18:35:24 2007
From: jkeen at verizon.net (James E Keenan)
Date: Fri, 26 Oct 2007 21:35:24 -0400
Subject: [tpm] Some links from last night's meeting
In-Reply-To: <8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
Message-ID: <32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>


On Oct 26, 2007, at 11:46 AM, Fulko Hew wrote:

>
>
> On 10/26/07, Jim Graham <jim.graham at jim-graham.net> wrote:
>   -- Reddit discussion of Lua benchmarks on Parrot. This was  
> originally a link to the perl6 internals mailing list with Reddit  
> comments. Most of the discussion comes back to 'completeness first,  
> optimization second'
>
>     http://programming.reddit.com/info/2z05a/comments/
>
>
> I read that stuff, and I know of Lua (in passing where its embedded  
> in Wireshark)
>
> My comment on the Reddit discussion, (based on last night's  
> discussion) is:
> ...Lua is optimized for Lua,
> ...Parrot is not optimized for Lua,
> ...Parrot is optimized for Parrot.
>
> So I'm not surprised that Lua code running on a highly
> (Lua) optimized 'system' is faster than Lua code
> compiled to run on Parrot.

The Parrot project is always looking for developers who can help port  
languages other than Perl to Parrot.

Any volunteers?

jimk



From psema4 at gmail.com  Fri Oct 26 18:47:20 2007
From: psema4 at gmail.com (Scott Elcomb)
Date: Fri, 26 Oct 2007 21:47:20 -0400
Subject: [tpm] Some links from last night's meeting
In-Reply-To: <32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
	<32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
Message-ID: <99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>

On 10/26/07, James E Keenan <jkeen at verizon.net> wrote:
> The Parrot project is always looking for developers who can help port
> languages other than Perl to Parrot.

Out of curiousity, I was going to ask if javascript (er ecmascript)
was being ported, but I see there is at least one implementation in
the works, PJS.

Instead, I'll have to ask if there's any work towards modifying Dr.
Jack Crenshaw's USENET tutorial "Let's Build A Compiler" such that the
compiler/interpretor being built works in Parrot?

If not, I'll offer to do so - it'll likely be a couple months before I
could start, but it'd be a wonderful project to work on.  I started
converting Dr. Crenshaw's tutorial to Perl 5.6 a year or two ago while
experimenting with a curses-based IDE.

-- 
  Scott Elcomb
  http://www.psema4.com/

From arocker at vex.net  Sat Oct 27 12:51:43 2007
From: arocker at vex.net (arocker at vex.net)
Date: Sat, 27 Oct 2007 15:51:43 -0400 (EDT)
Subject: [tpm] Some links from last night's meeting
In-Reply-To: <99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
	<32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
	<99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
Message-ID: <38994.192.30.202.28.1193514703.squirrel@webmail.vex.net>

"Scott Elcomb" <psema4 at gmail.com>,
>
> Out of curiousity, I was going to ask if javascript (er ecmascript)
> was being ported,
>

A paper about a Javascript implementation on Parrot. The paper's in Dutch
(I think), but the abstract's in English.

http://users.fulladsl.be/spb1622/pjs/scriptie.pdf


From indy at indigostar.com  Wed Oct 31 16:53:44 2007
From: indy at indigostar.com (Indy Singh)
Date: Wed, 31 Oct 2007 19:53:44 -0400
Subject: [tpm] Weird arrary reference bahaviour
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE><8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com><32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net><99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
	<38994.192.30.202.28.1193514703.squirrel@webmail.vex.net>
Message-ID: <018401c81c19$45dfe3c0$6401a8c0@roadhog>

Hi Guys,
Can anyone explain, why taking a reference to an array gives a different 
value on each iteration of this loop.  This is the line of code:
  my $r = \@f;

If I move the "my @a" declaration outside the loop I get the same value 
each time.

It seems that on each iteration of the loop a new @f array is created. 
Does that make sense?

Indy Singh
IndigoSTAR Software -- www.indigostar.com


my @foo = ("ford:ltd", "chevy:nova");
my $x;
my @a;

foreach $x (@foo) {
  my @f;
  @f = split(/:/, $x);
  my $r = \@f;
  print "r = $r\n";
  push @a, $r;
}

my $item;
foreach $item (@a) {
  print join(' ', @{$item}),"\n";
}


From fulko.hew at gmail.com  Wed Oct 31 17:13:55 2007
From: fulko.hew at gmail.com (Fulko Hew)
Date: Wed, 31 Oct 2007 20:13:55 -0400
Subject: [tpm] Weird arrary reference bahaviour
In-Reply-To: <018401c81c19$45dfe3c0$6401a8c0@roadhog>
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
	<32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
	<99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
	<38994.192.30.202.28.1193514703.squirrel@webmail.vex.net>
	<018401c81c19$45dfe3c0$6401a8c0@roadhog>
Message-ID: <8204a4fe0710311713x402507aej6f9e35a6e6eb704a@mail.gmail.com>

On 10/31/07, Indy Singh <indy at indigostar.com> wrote:
>
> Hi Guys,
> Can anyone explain, why taking a reference to an array gives a different
> value on each iteration of this loop.  This is the line of code:
>   my $r = \@f;
>
> If I move the "my @a" declaration outside the loop I get the same value
> each time.
>
> It seems that on each iteration of the loop a new @f array is created.
> Does that make sense?


... snip ...

foreach $x (@foo) {
>   my @f;


... snip ...

Yes, this make sense because my @f is a variable declared
every time the body of the foreach loop is executed.
(for the same reason that the split is executed every
time through the loop too.)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071031/39c846ee/attachment.html 

From indy at indigostar.com  Wed Oct 31 17:36:47 2007
From: indy at indigostar.com (Indy Singh)
Date: Wed, 31 Oct 2007 20:36:47 -0400
Subject: [tpm] Weird arrary reference bahaviour
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE><8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com><32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net><99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com><38994.192.30.202.28.1193514703.squirrel@webmail.vex.net><018401c81c19$45dfe3c0$6401a8c0@roadhog>
	<8204a4fe0710311713x402507aej6f9e35a6e6eb704a@mail.gmail.com>
Message-ID: <019b01c81c1f$49911d30$6401a8c0@roadhog>

Ok, thanks, I believe you. I just didn't expect that.  I was thinking in C++ terms, where that variable would be declared on the stack frame and would persist until the loop exits.



----- Original Message ----- 
  From: Fulko Hew 
  To: tpm at to.pm.org 
  Sent: Wednesday, October 31, 2007 8:13 PM
  Subject: Re: [tpm] Weird arrary reference bahaviour




  On 10/31/07, Indy Singh <indy at indigostar.com> wrote:
    Hi Guys,
    Can anyone explain, why taking a reference to an array gives a different
    value on each iteration of this loop.  This is the line of code:
      my $r = \@f;

    If I move the "my @a" declaration outside the loop I get the same value 
    each time.

    It seems that on each iteration of the loop a new @f array is created.
    Does that make sense?

  ... snip ... 


    foreach $x (@foo) {
      my @f;

  ... snip ...

  Yes, this make sense because my @f is a variable declared
  every time the body of the foreach loop is executed.
  (for the same reason that the split is executed every 
  time through the loop too.)



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.pm.org/pipermail/toronto-pm/attachments/20071031/7d0df86d/attachment.html 

From ceeshek at gmail.com  Wed Oct 31 18:20:00 2007
From: ceeshek at gmail.com (Cees Hek)
Date: Thu, 1 Nov 2007 12:20:00 +1100
Subject: [tpm] Weird arrary reference bahaviour
In-Reply-To: <018401c81c19$45dfe3c0$6401a8c0@roadhog>
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
	<32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
	<99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
	<38994.192.30.202.28.1193514703.squirrel@webmail.vex.net>
	<018401c81c19$45dfe3c0$6401a8c0@roadhog>
Message-ID: <fd8e28ef0710311820u7fbd0584nbfdb71f78cb0f266@mail.gmail.com>

On 11/1/07, Indy Singh <indy at indigostar.com> wrote:
> Hi Guys,
> Can anyone explain, why taking a reference to an array gives a different
> value on each iteration of this loop.  This is the line of code:
>   my $r = \@f;
>
> If I move the "my @a" declaration outside the loop I get the same value
> each time.
>
> It seems that on each iteration of the loop a new @f array is created.
> Does that make sense?
>
> Indy Singh
> IndigoSTAR Software -- www.indigostar.com
>
>
> my @foo = ("ford:ltd", "chevy:nova");
> my $x;
> my @a;
>
> foreach $x (@foo) {
>   my @f;
>   @f = split(/:/, $x);
>   my $r = \@f;
>   print "r = $r\n";
>   push @a, $r;
> }

At the end of each iteration of the loop, @f falls out of scope.  But
since you store away a reference to \@f, the data that @f refers to
does not get garbage collected (ie it's refcount has not hit zero).
So during the next iteration of the loop, the memory location that @f
used during the previous iteration is not available to be re-used, so
a new location is assigned to @f.

if you didn't store that reference to @f, then the data would be
garbage collected (ie @f goes out of scope at the end of the
iteration, the refcount is zero and the memory is reclaimed), and Perl
would re-use the same memory location for @f for each iteration.

Cheers,

Cees

From indy at indigostar.com  Wed Oct 31 18:42:32 2007
From: indy at indigostar.com (Indy Singh)
Date: Wed, 31 Oct 2007 21:42:32 -0400
Subject: [tpm] Weird arrary reference bahaviour
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
	<32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
	<99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
	<38994.192.30.202.28.1193514703.squirrel@webmail.vex.net>
	<018401c81c19$45dfe3c0$6401a8c0@roadhog>
	<fd8e28ef0710311820u7fbd0584nbfdb71f78cb0f266@mail.gmail.com>
Message-ID: <01b101c81c28$78e96390$6401a8c0@roadhog>

From: "Cees Hek" <ceeshek at gmail.com>
To: "Indy Singh" <indy at indigostar.com>
Cc: <tpm at to.pm.org>
Sent: Wednesday, October 31, 2007 9:20 PM
Subject: Re: [tpm] Weird arrary reference bahaviour


> On 11/1/07, Indy Singh <indy at indigostar.com> wrote:
>> Hi Guys,
>> Can anyone explain, why taking a reference to an array gives a 
>> different
>> value on each iteration of this loop.  This is the line of code:
>>   my $r = \@f;
>>
>> If I move the "my @a" declaration outside the loop I get the same 
>> value
>> each time.
>>
>> It seems that on each iteration of the loop a new @f array is 
>> created.
>> Does that make sense?
>>
>> Indy Singh
>> IndigoSTAR Software -- www.indigostar.com
>>
>>
>> my @foo = ("ford:ltd", "chevy:nova");
>> my $x;
>> my @a;
>>
>> foreach $x (@foo) {
>>   my @f;
>>   @f = split(/:/, $x);
>>   my $r = \@f;
>>   print "r = $r\n";
>>   push @a, $r;
>> }
>
> At the end of each iteration of the loop, @f falls out of scope.  But
> since you store away a reference to \@f, the data that @f refers to
> does not get garbage collected (ie it's refcount has not hit zero).
> So during the next iteration of the loop, the memory location that @f
> used during the previous iteration is not available to be re-used, so
> a new location is assigned to @f.
>
> if you didn't store that reference to @f, then the data would be
> garbage collected (ie @f goes out of scope at the end of the
> iteration, the refcount is zero and the memory is reclaimed), and Perl
> would re-use the same memory location for @f for each iteration.
>
> Cheers,
>
> Cees

> At the end of each iteration of the loop, @f falls out of scope.
I would have thought that @f falls out of scope _after_ the closing 
brace, not while still inside it.  But you seem to be right.



From uri at stemsystems.com  Wed Oct 31 20:39:46 2007
From: uri at stemsystems.com (Uri Guttman)
Date: Wed, 31 Oct 2007 22:39:46 -0500
Subject: [tpm] Weird arrary reference bahaviour
In-Reply-To: <019b01c81c1f$49911d30$6401a8c0@roadhog>
	(Indy Singh's message of "Wed, 31 Oct 2007 20:36:47 -0400")
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
	<32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
	<99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
	<38994.192.30.202.28.1193514703.squirrel@webmail.vex.net>
	<018401c81c19$45dfe3c0$6401a8c0@roadhog>
	<8204a4fe0710311713x402507aej6f9e35a6e6eb704a@mail.gmail.com>
	<019b01c81c1f$49911d30$6401a8c0@roadhog>
Message-ID: <x7wst23a7x.fsf@mail.sysarch.com>

>>>>> "IS" == Indy Singh <indy at indigostar.com> writes:

  IS> Ok, thanks, I believe you. I just didn't expect that.  I was
  IS> thinking in C++ terms, where that variable would be declared on
  IS> the stack frame and would persist until the loop exits.

it is due to perl's reference counting. all lexicals are allocated from
the heap (not the stack) and so they can have references to them that
keeps them alive. my does both a compile time and run time stuff. the
run time stuff includes allocating the variable storage. if my is
executed again and there is are live references to that variable, it
will allocate a new one. this is one way to create a new anon ref. these
do the exact same thing:

	$k = [ 1, 3 ] ;
	$k = do{ my @k = (1, 3) ; \@k } ;

and each time you execute those lines, a new anon array ref is
created. your sub (or rather loop) does the same thing. saving a ref to
the lexical array in that scope means the array will not be freed up
after the loop iterates by hitting its end. so a new array and
subsequently a new ref address are created the next time. by declaring
the array before the loop, it never executes the my runtime part in the
loop and so the ref stays the same as the array stays the same.

uri

-- 
Uri Guttman  ------  uri at stemsystems.com  -------- http://www.stemsystems.com
--Perl Consulting, Stem Development, Systems Architecture, Design and Coding-
Search or Offer Perl Jobs  ----------------------------  http://jobs.perl.org

From liam at holoweb.net  Wed Oct 31 22:53:59 2007
From: liam at holoweb.net (Liam R E Quin)
Date: Thu, 01 Nov 2007 01:53:59 -0400
Subject: [tpm] Weird arrary reference bahaviour
In-Reply-To: <01b101c81c28$78e96390$6401a8c0@roadhog>
References: <BAYC1-PASMTP065465EA89B6C6D7D3995390960@CEZ.ICE>
	<8204a4fe0710260846w57722abciac302dca03150912@mail.gmail.com>
	<32042130-9EB1-41EC-B78C-CD139DEB575A@verizon.net>
	<99a6c38f0710261847h7975a9d6n224c618a30bbef78@mail.gmail.com>
	<38994.192.30.202.28.1193514703.squirrel@webmail.vex.net>
	<018401c81c19$45dfe3c0$6401a8c0@roadhog>
	<fd8e28ef0710311820u7fbd0584nbfdb71f78cb0f266@mail.gmail.com>
	<01b101c81c28$78e96390$6401a8c0@roadhog>
Message-ID: <1193896439.28685.9.camel@dell.barefootcomputing.com>

On Wed, 2007-10-31 at 21:42 -0400, Indy Singh wrote:

> > At the end of each iteration of the loop, @f falls out of scope.
> I would have thought that @f falls out of scope _after_ the closing 
> brace, not while still inside it.  But you seem to be right.

Perl is like C in this regard.

for (i = 0; i < 7; i ++) {
  char *boy = strdup("Simon");
  printf("%d %x\n", i, boy);
}

will give different addresses each time round the loop.

The variable is created anew each time round (and in this case
there's a memory leak, of course, because we never free the boy).

for (i = 0; i < 7; i ++) {
  char *boy = strdup("Simon");
  printf("%d %x\n", i, boy);
  free(boy);
}

will work better in C.

Liam

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org www.advogato.org