NetPacket - assemble/disassemble network packets at the protocol level


version 1.1.1


    # NetPacket is a base class only


NetPacket provides a base class for a cluster of modules related to decoding and encoding of network protocols. Each NetPacket descendent module knows how to encode and decode packets for the network protocol it implements. Consult the documentation for the module in question for protocol-specific implementation.

Note that there is no inheritance in the NetPacket:: cluster of modules other than each protocol module being a NetPacket. This was seen to be too restrictive as imposing inheritance relationships (for example between the IP, UDP and TCP protocols) would make things like tunneling or other unusual situations difficult.


You are encouraged to write additional NetPacket:: modules as well as improve existing ones. Contact the maintainer of the module in question with your suggestions or changes.

The following sections are a list of suggestions and conventions for writing a NetPacket:: module.

Naming Conventions

When creating a module in the NetPacket:: namespace, it is suggested that you stick to a couple of conventions when naming packet contents. This will hopefully lead to a consistent namespace making the NetPacket:: easier to use.

Content names are all lowercase, with underscores separating multiple words. The following abbreviations are recommended:

            Word                Abbreviation
            source              src
            destination         dest
            checksum            cksum
            identifier          id
            version             ver
            protocol            proto          

Required Methods

encode(), decode(), strip()

Required Fields

Every NetPacket:: object should have the following fields.


A link to the parent NetPacket:: object in which this NetPacket:: object is encaulated. This field is undefined if there is no parent object.


A copy of the raw data of the packet.


This field should contain the data encapsulated in the packet (i.e any headers or trailers stripped off) or undef if the packet contains no data. Note that in this sense, "data" is taken to mean information not relevant to the particular protocol being decoded. For example, an ARP packet contains many header fields but no data. A UDP datagram, however contains header fields and a payload.


Joel Knight has a patch for NetPacket for IPv6 support available at


Copyright (c) 2001 Tim Potter and Stephanie Wehner.

Copyright (c) 1995,1996,1997,1998,1999 ANU and CSIRO on behalf of the participants in the CRC for Advanced Computational Systems ('ACSys').

This module is free software. You can redistribute it and/or modify it under the terms of the Artistic License 2.0.

This program is distributed in the hope that it will be useful, but without any warranty; without even the implied warranty of merchantability or fitness for a particular purpose.


Tim Potter <>

Stephanie Wehner <>

Yanick Champoux <>