Streaming XML standardiser implementation in Groovy using NUX. Allows feeds to be processed via DB configuration rather than XSL

streaming-xml-standardiser.groovy

package richmarr.xml.example
import org.xml.sax.InputSource
import nu.xom.*
import nux.xom.io.*
import nux.xom.xquery.*

// this would normally be pulled from a database
def config = [
	item : "/xpath/to/item/elements",
	fields : [
		id : "./id/text()",
		title : "./title/text()",
		description : "./desc/text()"
	]
]

def feedName = "testA"
InputStream inputStream = new FileInputStream("target/raw/${feedName}.xml")
OutputStream outputStream = new FileOutputStream("target/standardised/${feedName}.xml")

StreamingSerializerFactory factory = new StreamingSerializerFactory();
StreamingSerializer out = factory.createXMLSerializer( outputStream, "UTF-8" );
out.writeXMLDeclaration();
out.writeStartTag(new Element("items"));

StreamingTransform itemTransform = new StreamingTransform() {
	public Nodes transform( Element item ) {
		out.writeStartTag(new Element("item"))
		config.fields.each { fieldEntry ->
			Nodes results = XQueryUtil.xquery( item, fieldEntry.value )
			if (results.size() > 0)
			{
				out.writeStartTag(new Element(fieldEntry.key))
				out.write(new Text( results.get(0).getValue() ))
				out.writeEndTag()
			}
		}
		out.writeEndTag()
		return new Nodes()
	}
}

// parse document with a filtering Builder
Builder builder = new Builder(new StreamingPathFilter( config.item, null ).createNodeFactory( null, itemTransform ));
builder.build(new InputSource(inputStream));
out.writeEndTag()
out.writeEndDocument()