[Enum] [ticket.frobbit.se #874] REGEXP 'i' flag considered harmful

"lconroy via RT" <3761bis@frobbit.se> Sun, 26 March 2006 14:26 UTC

Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com) by megatron.ietf.org with esmtp (Exim 4.43) id 1FNWCG-0003W3-Il; Sun, 26 Mar 2006 09:26:24 -0500
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org) by megatron.ietf.org with esmtp (Exim 4.43) id 1FNWCF-0003Tq-N6 for enum@ietf.org; Sun, 26 Mar 2006 09:26:23 -0500
Received: from charles.frobbit.se ([85.30.129.36]) by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1FNWCF-00029S-7b for enum@ietf.org; Sun, 26 Mar 2006 09:26:23 -0500
Received: from charles.frobbit.se (localhost [127.0.0.1]) by charles.frobbit.se (8.13.4/8.13.4) with ESMTP id k2QEQL6h079034; Sun, 26 Mar 2006 16:26:21 +0200 (CEST) (envelope-from www@charles.frobbit.se)
Received: (from www@localhost) by charles.frobbit.se (8.13.4/8.12.10/Submit) id k2QEQLnd079033; Sun, 26 Mar 2006 16:26:21 +0200 (CEST) (envelope-from www)
Date: Sun, 26 Mar 2006 16:26:21 +0200
From: lconroy via RT <3761bis@frobbit.se>
In-Reply-To: <6BC21F6F-381F-4092-B2B4-C6F9E94B83EA@insensate.co.uk>
References: <RT-Ticket-874@example.com> <6BC21F6F-381F-4092-B2B4-C6F9E94B83EA@insensate.co.uk>
Message-ID: <rt-3.4.4-74236-1143383181-1973.874-3-0@example.com>
Precedence: bulk
X-RT-Loop-Prevention: ticket.frobbit.se
RT-Ticket: ticket.frobbit.se #874
Managed-by: RT 3.4.4 (http://www.bestpractical.com/rt/)
RT-Originator: lconroy@insensate.co.uk
Cc: enum@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
X-RT-Original-Encoding: utf-8
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 8b30eb7682a596edff707698f4a80f7d
Subject: [Enum] [ticket.frobbit.se #874] REGEXP 'i' flag considered harmful
X-BeenThere: enum@ietf.org
X-Mailman-Version: 2.1.5
Reply-To: 3761bis@frobbit.se
List-Id: Enum Discussion List <enum.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/enum>, <mailto:enum-request@ietf.org?subject=unsubscribe>
List-Post: <mailto:enum@ietf.org>
List-Help: <mailto:enum-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/enum>, <mailto:enum-request@ietf.org?subject=subscribe>
Errors-To: enum-bounces@ietf.org

Sun Mar 26 16:26:21 2006: Request 874 was acted upon.
Transaction: Ticket created by lconroy@insensate.co.uk
       Queue: 3761bis
     Subject: REGEXP 'i' flag considered harmful
       Owner: Nobody
  Requestors: lconroy@insensate.co.uk
      Status: new
 Ticket <URL: http://ticket.frobbit.se/Ticket/Display.html?id=874 >


RFC 3402 specifies an 'i' flag that can be applied to the REGEXP field.
This flag is intended to make AUS matching case insensitive.
For ENUM, this is not appropriate, as the allowed characters within  
an AUS are all case insensitive.

However, the definition of the 'i' flag also states that the client  
MAY convert the generated output to lower case.
Such optional behaviour is non-trivial where URI/IRIs are case  
sensitive (e.g. email addresses).
[i.e. it is prone to mis-configuration]
Also, for characters not in the US-ASCII equivalent range, case  
sensitivity may not make sense.
For the match sub-field, this can cause confusion, but as mentioned  
this is irrelevant for ENUM as all characters in the match sub-field  
should be US-ASCII, and case insensitive already.
Some languages have the concept of 'lower case', whilst others don't,  
and adding a suggestion that an IRI may be converted to lower case is  
not helpful in all cases.
Finally, it is optional behaviour, and so the result from client to  
client is indeterminate.

Option/Suggestion:
At least for ENUM, state explicitly that the 'i' flag SHOULD NOT be  
used.

atb,
     Lawrence


_______________________________________________
enum mailing list
enum@ietf.org
https://www1.ietf.org/mailman/listinfo/enum