W3C home > Mailing lists > Public > public-i18n-core@w3.org > January to March 2006

Re: I18n comment: xml and namespaces version

From: Felix Sasaki <fsasaki@w3.org>
Date: Tue, 24 Jan 2006 14:19:08 +0900
To: "Bjoern Hoehrmann" <derhoermi@gmx.net>
Cc: www-style@w3.org, public-i18n-core@w3.org
Message-ID: <op.s3vad6q9x1753t@ibm-60d333fc0ec.mag.keio.ac.jp>

"Bjoern Hoehrmann" <derhoermi@gmx.net>
www-style@w3.org, public-i18n-core@w3.org
On Tue, 24 Jan 2006 00:57:27 +0900, Bjoern Hoehrmann <derhoermi@gmx.net>  
wrote:

>
> * Felix Sasaki wrote:
>>> At http://www.w3.org/International/reviews/0601-css3-selectors/
>
>>>> #5 the spec does not rely on a given version of XML. It can select
>>>>     in any kind of markup tgree-based language.
>
>> and that is the reason why it should say what happens if there is a  
>> clash
>> in versions. Or it should add a note to warn the users of such clashes,
>> e.g. saying
>>
>> "This specification does not provide mechanisms to avoid clashes which
>> might occur due to different versions of markup languages. An example  
>> are
>> clashes between different versions of XML.
>
> Could you give an example of such a "clash" or how mechanisms to avoid
> them could look like?

Example for XHTML:
<?xml version="1.0" ...?>
<html>
<head> ...
<style type="text/css">
<!--
@namespace foo url(XXX);
  foo|h1 { color: blue }
/* ... ... */
-->
</style>
</head>...
</html>
This is an error if "XXX" contains characters which are not part of the  
NCName production in XML 1.0. In the case of XHTML it is an error anyway,  
since XHTML has a normative reference to XML 1.0 . But if I understand you  
correctly, you want selectors to be used for a variety, possibly currently  
not yet existing document languages. You should prepare the creators /  
users of such languages by telling them that such clashes are sources for  
errors.
Mechanism to avoid such errors: state that they have to be avoided, as the  
XML 1.0 spec does.

> Selectors can match any namespace, any element
> name, any attribute name, any attribute value, so long as the match does
> not depend on looking for a U+0000 character at any point in the tree. I
> can't really think of changes to XML or XML Namespaces that require CSS
> authors to take steps to avoid any kind of problem, or a revision of CSS
> to take those changes into account, so long as XML documents are tree
> structures.
Received on Tuesday, 24 January 2006 05:19:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 1 October 2008 10:18:50 GMT