I've been looking at MooseX::App documentat trying to understand what exactly is mean by a "command"...

What I'd like to do is create a base class with certain switches and functionality which will be usable across a range of separate scripts, for example:

igner-cli.pl  --db=demo "rm /dev/null"
igner-url.pl  --db=prod --proxy=none http://boldra.com/

This is what I'm trying:

package Igner::App;
use MooseX::App;

parameter db => ( #all igner apps need a "db"
  required => 1,
   ...
);

package Igner::App::URL;
extends 'Igner::App';
use MooseX::App::Command; # because this script has additional paramet
+ers and options

paramter proxy => ( # only Igner::App::URL needs a proxy
  required => 1,
  ...
);

sub run {
  ...
}
[download]

Then in the script I expected to be able to write:

#!/usr/bin/perl
use Igner::App::URL;
Igner::App::URL->new_with_command->run;
[download]

And the run sub would be executed if the user provided db and proxy, and the user would get help information if not.

Instead, I'm getting "new_with_command may only be called from the application base package:Igner::App::URL" which doesn't shed any light. As I said at the start - I'm lacking a clear concept of what the modules mean by "command". The MooseX::App::Command documentation could use some examples of how it relates to actually executing commands on the command-line.