W3C home > Mailing lists > Public > wai-xtech@w3.org > May 2008

[Access Module] Document Character Set, UNICODE, xhtml-access, and @key [FWD]

From: Gregory J. Rosmaita <oedipus@hicom.net>
Date: Tue, 20 May 2008 13:59:25 +0100
To: w3c-wai-ua@w3.org
Cc: wai-xtech@w3.org
Message-Id: <20080520125822.M61375@hicom.net>

---------- Forwarded Message -----------
From: Shane McCarron <shane@aptest.com>
To: XHTML WG <public-xhtml2@w3.org>
Sent: Mon, 19 May 2008 17:54:35 -0500
Subject: Document Character Set, UNICODE, xhtml-access, and @key

DISCLAIMER:  THIS ISSUE DOES NOT IMPACT OUR STARTING XHTML ACCESS LAST 
CALL IMMEDIATELY

In discussions with Tina this afternoon (late night for her, and THANKS 
TINA!) we decided that we are confused about the datatype for @key.  We 
are no more confused than we were about the datatype for @accesskey on 
the input element of XHTML 1.1 though, so that's good.

Here's the short description:  XHTML Access says that @key is a 
character from the document character set.  But we define its datatype 
as "Character", which is an XML datatype that maps to an ISO 10646 code 
point... if I read it correctly.  This is the same as @accesskey on 
input in XHTML 1.1.

In XML, values of an attribute are CDATA or a reserved type (according 
to http://www.w3.org/TR/xml/#sec-attribute-types).  Since @key does not 
use a reserved type, it must be "CDATA".  And CDATA is a sequence that 
is made up of most ISO 10646 Characters (see 
http://www.w3.org/TR/xml/#NT-Char).  Of course, since in an XML parser 
an attribute value is evaluated AFTER parsing, entities would be 
expanded first (e.g., &amp; would turn into an ampersand in an attribute 
value before being treated as a CDATA sequence).

So..... What does this mean for us.  I think it means we are saying an 
@key value is an ISO 10646 character from the 'Char' production of XML.  
And that a user agent needs to map that into its local processing 
(document character set) before registering an event handler that maps 
the key to some event (change focus, actuate or not).  Moreover, I think 
that @key is explicitly NOT a character from the document character 
set.  But.... I am not an expert in this area, and I could be wrong.  If 
there really is an issue here (as Tina and I suspect there is) I think 
we can just submit it as a last call comment against our own document.  
This is pretty subtle.

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com
------- End of Forwarded Message -------
Received on Tuesday, 20 May 2008 13:00:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 13:15:49 GMT