W3C home > Mailing lists > Public > ietf-http-wg@w3.org > October to December 2009

Re: ambiguous ABNF in RFC 2231 (and friends)

From: Martin J. Dürst <duerst@it.aoyama.ac.jp>
Date: Tue, 06 Oct 2009 12:04:41 +0900
Message-ID: <4ACAB3C9.5040006@it.aoyama.ac.jp>
To: Julian Reschke <julian.reschke@gmx.de>
CC: HTTP Working Group <ietf-http-wg@w3.org>
As the secondary charset expert reviewer, I have never seen a charset 
with a "'" in it. There is note at 
http://www.iana.org/assignments/character-sets, and they are not allowed 
in the "more restrictive" production in Section 2.3 of RFC 2978. They 
are also not allowed in XML, which is even more restrictive. I guess 
that as a reviewer, I'd simply tell an applicant to change the name.

As the co-chair of the LTRU WG, I know quite for sure that language tags 
don't contain "'". Please check yourself at 
http://tools.ietf.org/html/rfc5646 or one of it's predecessors.

So I see absolutely no problem to tightening up the grammar.

Regards,    Martin.

On 2009/10/06 3:57, Julian Reschke wrote:
> Hi,
>
> I just noticed a problem in RFC 2231 that draft-reschke-rfc2231-in-http
> inherits:
>
> extended-initial-value := [charset] "'" [language] "'"
> extended-other-values
>
> Both charset and language can contain a single quote, so it's not the
> greatest choice as separator.
>
> On the other hand, this doesn't seem to a problem in practice, so my
> plan for draft-reschke-rfc2231-in-http is to simply disallow single
> quotes inside charset and language.
>
> Opinions?
>
> BR, Julian
>
>

-- 
#-# Martin J. Dürst, Professor, Aoyama Gakuin University
#-# http://www.sw.it.aoyama.ac.jp   mailto:duerst@it.aoyama.ac.jp
Received on Tuesday, 6 October 2009 03:05:38 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:12 GMT