dealing with CDATA (was dealing with CDATA on the client-side)

[eric at schwa dot com (Eric Costello)]

Hello all,

First time poster here.

SYNOPSIS:
What do I need to do to keep an XSL transform in an MS XMLDOM environment
from converting < and > in a CDATA node to &lt; and &gt;?

DETAILS:
Using Microsoft's XMLDOM object on the server side to transform an XML file
with an XSL file and output HTML to the browser, I run into the same problem
as Matt Haughey: CDATA comes through the transform with < and > converted to
&lt; and &gt;. Here is the relevant section of his previous post to this
list (which mentions a client side occurrence of this problem, but the
problem is identical on the server):

-----------------------------------------------
<etcetera>
    <![CDATA[this is a <a href="http://yahoo.com>test link/a>]]>
</etcetera>

applying XSL on the client-side leaves unparsed HTML sitting in the text,
so it looks like this in a browser:

this is a <a href="http://yahoo.com>test link/a>

instead of

this is a test link  (test link being a hyperlink to yahoo in this example)

Now to get around this, on the server-side, we've been doing search and
replaces on &lt; and &gt; characters, and replacing them with < and > to
produce parse-able HTML when sent to the client.
------------------------------------------------

Replacing the &lt; and &gt; characters with < and > is a solution that I am
using along with Matt, but it seems pretty kludgey to me. IS there something
I should be doing in my code that would preserve the characters as they are
in the CDATA? Or is this normal and uncorrectable behavior that I need to
code around?

Thanks,
Eric

Eric Costello
eric@schwa.com
http://glish.com


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list