Internet Draft File: <draft-pfenning-irc-extensions-00.txt>

Kent Cedola Thomas Pfenning Microsoft Corporation 31 January 1997

Extensions to the Internet Relay Chat Protocol (IRCX)

3. Modified UTF7 Encoding of UNICODE characters

Allowing UNICODE characters for all user visible strings introduces a set of compatibility problems if the protocol must be backwards compatible. While UTF7 encoding[1] maintains readability for pure ASCII strings, the BASE64 encoding of extended characters can contain characters which have a special meaning to the parser. In order to provide backwards compatibility with exisiting IRC clients and to allow new client to use a subset of UNICODE features on existing IRC server we introduce an additional postprocessing step on the result of an UTF7 translation.

The quoting character for the postprocessing step is ''. All mappings are listed in the table below.

Table 1 - Character Quoting in UTF7

\b	for " " blank
\c	for ","
\\	for "\"
\r	for CR
\n	for LF

This mechanism is a first proposal and subject to change if we dicover a more general solution to this problem. Since a similar problem was discovered in the context of the IMAP protocol[3] it might be worthwhile to define a new safe encoding of UNICODE which translates to alphanumeric characters only. It seems, that every special character has a special meaning in one or the other protocol.

PREVIOUS PAGE | TABLE OF CONTENTS | NEXT PAGE

Comments about this page? Send them to webmaster (at) invlogic.com.

Return to Innovative Logic's Home Page.
Last modified: 17-Sep-2004 01:04PM.
Pages copyright ©1996-1997, Innovative Web Creations.
All rights reserved.