W3C home > Mailing lists > Public > public-html@w3.org > January 2011

Re: CfC: close ISSUE-126 charset-vs-backslashes by amicable resolution

From: Anne van Kesteren <annevk@opera.com>
Date: Sat, 22 Jan 2011 16:24:45 +0100
To: "Jonas Sicking" <jonas@sicking.cc>, "Julian Reschke" <julian.reschke@gmx.de>
Cc: "Sam Ruby" <rubys@intertwingly.net>, "HTML WG" <public-html@w3.org>
Message-ID: <op.vppufib264w2qv@anne-van-kesterens-macbook-pro.local>
On Wed, 19 Jan 2011 15:37:36 +0100, Julian Reschke <julian.reschke@gmx.de>  
wrote:
> The CP affects handling of character set names that contain either a  
> backslash or quite characters. No registered character set does contain  
> these. That's why I think this is an edge case, and there's no  
> requirement to claim that breaking the base spec is needed here.

So a) the only reason WebKit and Opera handle escapes is because they  
implemented UTS #22. A now considered broken standard for character  
encoding label matching that indicates you ignore a bunch of characters  
when comparing character encoding labels. This was not because they are  
HTTP compliant. You can see this in e.g. Opera by looking at the "Info"  
menu where it states that the encoding specified is e.g. "ut\f-8" and the  
encoding used is "utf-8".

And b) the base specification is not implemented as far as I can tell.  
Browsers (those that do not have UTS #22 support) do not support  
quoted-pair for the Content-Type header.


-- 
Anne van Kesteren
http://annevankesteren.nl/
Received on Saturday, 22 January 2011 15:25:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:17:19 GMT