--------------
Agent.pm v3.20
--------------
Copyright (c) 1997-1998 Steve Purkis. All rights reserved. The Agent
module can be distributed or modified under the 'Artistic License', as
distributed with Perl. The author guarantees nothing about anything to
do with this module; you use it at your own risk.
Welcome to the 3.2 release of Agent Perl, the first full release of this
module! This file contains a brief overview of Agent Perl, and some things
that you'll need to know in order to get it up and running on your system.
Overview
--------
Agent Perl provides both a foundation for software agent developers, and a
standardized API for the agent's users. Agent's two most prominent features
are its support for agent transportation and abstract messaging over
different mediums.
[A transportable perl agent is Perl code that can communicate with other
disparate entities, and can migrate between computers over a heterogeneous
network to perform some task for a client.]
Prerequisites
-------------
Required Modules:
Class::Tom 2.04+ (3.x not supported!)
MD5
Optional module for *alpha* threads support:
Thread
Platforms:
U*nix & Win32
Mac? OS/2?
Using Agents
------------
First of all, make sure you know how to use an object (the perlobj manpage is
a good place to start). Then read the Agent.pm's POD, and that of the agent
you are trying to use. After that, you're on your own, unless there are any
examples lying around :-).
If you would like to use the security features of Agent Perl, browse through
the Safe and Opcode PODs, and have a gander at 'Safe.readme'.
Developing Agents
-----------------
After you've read this, read the PODs for Agent.pm and all of its nested
modules. Read *.readme in the root directory. Then have a look in the
'examples' sub-directory for some functional *sample* agents.
Make *sure* you don't step on Agent.pm's toes by redefining the Message or
Transport subclasses.
An agent is simply an object. It inherits methods from the Agent package.
So to define an agent, all you need to do is define its class.
There are some rules you should adhere to, however:
1. An agent's class should be saved in a suitable <agentname>.pa file.
2. All agents should inherit from the 'Agent' package.
3. All agents must have an 'agent_main()' method.
4. An agent's state variables should be stored within the agent's
object, and NOT within the class.
5. Document your agent, preferably with POD!
Additional notes:
The agent's class _should_ be a sub-class of Agent itself. Here's a good
naming scheme for your agents:
package Agent::YourAgentName;
Because '.pa' files are not necessarily stored in a central place, they need
not be unique. For example, if two different users define two different
'RWho.pa' agents in two different locations, the do not need to worry which
.pa file will be used.
PS: If you come up with a functional agent, I'd love to hear about it!
Down the road...
----------------
If this release of Agent looks promising, I hope to include support for
PGP encryption & signatures, the KQML/KIF communication standards
(underway!), and more.
Links
-----
The Homepage: http://www.epn.ml.org/~spurkis/Agent
The Mailing List: perl5-agents@daft.com
Any questions, comments, bug reports, or suggestions you may have, I'd
like to hear about.
Enjoy!
+---
Steve Purkis <spurkis@engsoc.carleton.ca>
December 15, 1998