No, the RELAX NG Compact Schema in RFC 4287 is informative (non-normative).
The bidi draft updates atomCommonAttributes with:
atomCommonAttributes =
attribute xml:base { atomUri }?,
attribute xml:lang { atomLanguageTag }?,
attribute dir { "ltr" | "rtl" | "" }?,
undefinedAttribute*
Franklin
----- Original Message -----
From: "Sammy" <[EMAIL PROTECTED]>
To: "Franklin Tse" <[EMAIL PROTECTED]>
Cc: "James M Snell" <[EMAIL PROTECTED]>; "atom-syntax" <[email protected]>
Sent: Monday, May 14, 2007 18:09
Subject: RE: Fw: Atom bidi attribute, empty string value
undefinedAttribute = attribute * - (xml:base | xml:lang | local:*) { text }
Doesn't this mean that you CAN'T include NEITHER xml:base NOR xml:lang NOR
local:* ?
Regards,
Sammy
-----Original Message-----
From: Franklin Tse [mailto:[EMAIL PROTECTED]
Sent: Monday, May 14, 2007 12:53 PM
To: Sammy
Cc: James M Snell; atom-syntax
Subject: Re: Fw: Atom bidi attribute, empty string value
local: = "" (null namespace)
> A. Pagaltzis wrote:
>> Attributes never inherit the namespace from any other part of the
>> infoset, so attributes written without a prefix are always in the
>> null namespace.
So, "dir" is valid. What's the problem then?
Also, please note the bidi draft is an update to the atomCommonAttributes
section of RFC 4287.
Franklin
----- Original Message -----
From: "Sammy" <[EMAIL PROTECTED]>
To: "James M Snell" <[EMAIL PROTECTED]>; "atom-syntax" <[email protected]>
Sent: Monday, May 14, 2007 16:41
Subject: RE: Fw: Atom bidi attribute, empty string value
>
> Is that what the undefinedAttribute says in the spec? Because I really don't
> get what the local:* part in "attribute* -(xml:base | xml:lang | local:*)"
> means. In other words, if you wanted to stick to the spec, couldn't you have
> had dirns:dir instead of just dir: ?
>
> Regards,
> Sammy
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of James M Snell
> Sent: Sunday, May 13, 2007 11:53 PM
> To: atom-syntax
> Subject: Re: Fw: Atom bidi attribute, empty string value
>
>
> Well, yeah, sorry for not being concise on that point. The point was
> more that there really was no need for a new namespace for the
> attribute. :-)
>
> - James
>
> A. Pagaltzis wrote:
>> * James M Snell <[EMAIL PROTECTED]> [2007-05-13 19:05]:
>>> Rather than create a new xml namespace, the bidi spec adds the
>>> dir attribute to the Atom namespace
>>
>> Attributes never inherit the namespace from any other part of the
>> infoset, so attributes written without a prefix are always in the
>> null namespace.
>>
>> Regards,
>
>
>
