W3C home > Mailing lists > Public > www-xpath-comments@w3.org > January to March 2002

RE: [xsl] Namespace wildcards

From: Michael Kay <michael.h.kay@ntlworld.com>
Date: Sat, 2 Feb 2002 09:20:24 -0500 (EST)
To: <xsl-list@lists.mulberrytech.com>
Cc: <www-xpath-comments@w3.org>
Message-ID: <001c01c1abf4$abd864e0$78fe0250@pcukmka>
> -----Original Message-----
> From: owner-xsl-list@lists.mulberrytech.com
> [mailto:owner-xsl-list@lists.mulberrytech.com]On Behalf Of Elliotte
> Rusty Harold
> Sent: 01 February 2002 20:00
> To: xsl-list@lists.mulberrytech.com
> Cc: www-xpath-comments@w3.org
> Subject: [xsl] Namespace wildcards
>
>
> In light of all the issues with namespaces that already exist, I feel
> compelled to ask whether the namespace wildcards introduced in XPath
> 2.0 <http://www.w3.org/TR/xpath20/#node-tests> are a good idea. This
> enables expressions like this:
>
>
> <xsl:template match="*:set">
>    This matches MathML set elements, SVG set elements, set elements
>    from other  namespaces, and set elements from no namespace at all.
> </xsl:template>
>
> In other words the local name matters, but the URI doesn't. Do we
> really want to allow this? Do we actually need this? I'm really
> afraid this is going to further confuse developers about namespaces
> and encourage bad practices.

I agree it's open to misuse, but there are cases where it is genuinely
useful. I've heard of a number of cases (though I can't cite them) where
people have used different namespaces to represent versions or variants of
an XML vocabulary, but where the variants are sufficiently close to make it
possible to write stylesheets that can handle all of them.

Mike Kay
Received on Monday, 4 February 2002 00:30:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 3 October 2007 16:05:54 GMT