Re: [xsl] CDATA Handling

Subject: Re: [xsl] CDATA Handling
From: David Carlisle <davidc@xxxxxxxxx>
Date: Tue, 6 Jan 2009 13:16:26 GMT
> but not when it is input to me, a human reader.

exactly, CDATA is a perfectly reasonable part of the XML spec, and has
its uses, that being th emain one.

What isn't reasonable (almost always) is to do this:

<x>See following image: <![CDATA[abcde]]></x>

where CDATA isn't being used as a convenience for human readers, but
rather to actually encode information separating the base64 encoded
string from surrounding text.

In this situtaion:

> The KML Schema has a <description> element in which a limited
> subset of HTML is 'permitted',

You should be fine just to list the description element as a
cdata-section-element on your xsl:stylesheet, then all the html
fragments in the output will be CDATA quoted and so look like
<b>stuff</b> rather than &lt;b&gt;stuff&lt;/b&gt;

That gets the result to use CDAYA, and if you also want "tags" to appear
in the XSLT, then yes, as you say you can interleave CDATA sections and
markup in the stylesheet.

> test="local-name(parent::node())='BuiltStructure'"><

looks horrible though testing on name() tends to be inefficient, unless
you have namespace issues, can't you just use

You are testing ../@type multiple times, but since there can only be one
right answer, an xsl:choose (or better apply-templates) would be more efficient.

I'd write that as

<xsl:if test="parent::BuiltStructure">
               		Built Structure<br>]]></xsl:text>
       <xsl:apply-templates select="../@type"/>

Together with a bunch of templates of the form:

<xsl:template match="@type[.='logcabin']"><![CDATA[type: a log-cabin design<br>]]></xsl:template>


The Numerical Algorithms Group Ltd is a company registered in England
and Wales with company number 1249803. The registered office is:
Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.

This e-mail has been scanned for all viruses by Star. The service is
powered by MessageLabs. 

Current Thread