I need to unescape XML characters from inside of XML nodes with the help of only XSLT transformations. I have
Another solution. This one does not use the xml postprocessor, so readily useable as input of further xslt processing. Also guaranteed to create valid xml. This is a xslt 2.0 solution, escaping text within "documentation" tags, tested with saxon. You should modify the "allowedtags" variable to define your own data model. The immediate children are the tags, the ones below them are the attributes possible. Reading the allowed tags from an xsd is left as an excercise for the reader (please share it with me).