Hello,
I am working from XML documents that where html syntax is captured as escape sequences rather than as actual tags. For instance, rather than the html content characterized as:

<a href="http://www.state.gov/ofm/">More about OFM</a>

I get the corresponding escape sequences before and after the href as follows:

&lt;a href="http://www.state.gov/ofm/"&gt;More about OFM&lt;/a&gt;

What do I need to do to pre-process these strings such that all html syntax is properly output (as html tags)? Thanks so much for any assitance.

Parsing converts &lt; to < and converts < to a node. Conversely, serialization converts nodes to < and converts < to &lt; So if you want your transformation to start with &lt; and end up with <, then you either need to parse it twice and serialize it once, or to parse it once and suppress serialization. The first solution involves extracting the HTML as a string and (re-)parsing it to convert it into nodes, using some kind of extension function (e.g. saxon:parse in Saxon). The second solution involves using disable-output-escaping. d-o-e is usually frowned upon for two reasons: it's often misused, and it's not supported in all environments (for example it doesn't work in Firefox - in fact, it doesn't work in any environment where the result tree isn't serialized). But in this case, other than redesigning the source documents, it may be the best option.

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

Michael, thanks so much as usual. DOE is ok for my purposes, frowned upon or not. Works like a charm. Thanks again