[dsdl-discuss] Re: ANN: Modular Namespaces

On Fri, 31 Jan 2003 13:35:41 +0700
"James Clark" <jjc@jclark.com> wrote:

> http://www.thaiopensource.com/relaxng/mns.html

In the CD of DSDL Part 4, a validation candidate in another validation
candidate is REPLACED by dummy elements. In MNS, a validation candidate
in another validation candidate is PRESERVED. MNS provides another mechanism
(called pruning) for deleting validation candidates in validation candidates.

I would like to suggest another possibility. It might be good for
XSLT and XForms.

Consider an XML document as follows.

<a:foo xmlns:a="http://exa" xmns:b="http://exb">
  <b:bar1>
    <b:bar11>
       <a:baz1/>
    </b:bar11>
  <b:bar1>
  <b:bar2>
    <b:bar21>
       <a:baz2/>
    </b:bar21>
  <b:bar2>
</a:foo>

In the CD or MNS, we have FIVE validation candidates. They are rooted
by <a:foo>, <b:bar1>, <a:baz1>, <b:bar2>, and <a:baz2>.

Another possibility is to create THREE validation candidates. The
first one is:

<a:foo xmlns:a="http://exa" xmns:b="http://exb">
       <a:baz1/>
       <a:baz2/>
</a:foo>

The second and third are rooted by <b:bar1> and <b:bar2>, respectively.

Why is this good for XForms? Because the repeat element of XForms
intervene between <table> and <tr> of XHTML. By using the above
option, we can easily handle such compound documents. More about this,
see "9.3.2 Creating Repeating Structures Via Attributes" of the XForms
CR. (http://www.w3.org/TR/xforms/slice9.html).

Why is this option good for XSLT? Because we can focus on elements
in the XSLT namespace. In other words, we can handle

<xsl:template match="foo">
  <foo>
    <xsl:apply-templates/>
  </foo>
</xsl:template>

as

<xsl:template match="foo">
    <xsl:apply-templates/>
</xsl:template>

Cheers,

-- 
MURATA Makoto <murata@hokkaido.email.ne.jp>
--
DSDL members discussion list
To unsubscribe, please send a message with the
command  "unsubscribe" to dsdl-discuss-request@dsdl.org
(mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)