I have a question before I go any further down my current path. I have an XML document that describes the layout of an HTML form. I'm transforming it with XSL. That all works fine. I'm running into a problem when I try to add the value of the form field into the XML document (you know what's coming): one of the fields will contain HTML that may not be well-formed.

I'm dumping the HTML value into a CDATA section to handle the escaping and the non-well-formedness on the XML side. On the transform back to HTML, I'm grabbing the value via this bit of XSL:

<xsl:value-of select="/page/contents/maincontent/item/text()" disable-output-escaping="yes"/>

I've tried a number of different ideas (there are at least 155,000 of them in Google for "xsl transform cdata section html"), but the resulting HTML is always escaped. Is this because I'm dynamically creating the XML document in memory?

This is in C# on the .NET 2.0 framework.

disable-output-escaping only works if the output of the transformation is serialized using the XSLT serializer (and if the serializer supports it - not all do). I think the System.Xml.Xsl serializer supports d-o-e, but if you write the transformation output to a DOM tree, that's not much use.

Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference

(I have successfully used d-o-e before with a System.Xml.Xsl transformation, so I don't think that's it.)

if you write the transformation output to a DOM tree, that's not much use

You lost me there. What do you mean by writing the transformation to a DOM tree? Here's my current process as best I can describe it. I am:

1. Reading an XML file from disk.
2. Inserting content elements into it while in memory.
3. Transforming the updated document via XSL.
4. Returning the result as a string . . .

Having just walked through my code with a few choice breakpoints to make sure I wasn't lying, it appears the problem occurs after step 4, so it's not the transformation at all. Thanks for helping me see that.

I don't know the details of System.Xml.Xsl, but by analogy with other APIs I imagine that if you use a method like

public XmlReader Transform(XPathNavigator, XsltArgumentList, XmlResolver);

then the result isn't going to be serialized and disable-output-escaping is likely to have no effect.



Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference