🚀 go-pugleaf

XmlRules is optimized for this type of processing. The recursive approach works fine for small documents but when using large documents suffers dramatically.

Here is a sample rule that selects all sections elements and title elements from a proposed book content. It just pumps them into an array. You can modify the getArticleSections in conjuction with the xpath variable to grab the information you want.

var xpath = ["//article_root/article_section","//article_root/article_title"];
var articleSections = getArticleSections(xpath);
//At this point, myXMLMatches contains all of the XML elements
//that matched the XPath expression provided in myXPath.
function getArticleSections(myXPath){
	var myXMLElements = new Array;
	var myRuleProcessor = app.xmlRuleProcessors.add(myXPath);
	try{
		var myMatchData = myRuleProcessor.startProcessingRuleSet(app.documents.
		item(0).xmlElements.item(0));
		while(myMatchData != undefined){
			var myElement = myMatchData.element;
			myXMLElements.push(myElement);
			myMatchData = myRuleProcessor.findNextMatch();
		}
		myRuleProcessor.endProcessingRuleSet();
		myRuleProcessor.remove();
		return myXMLElements;
	}
	catch (myError){
		myRuleProcessor.endProcessingRuleSet();
		myRuleProcessor.remove();
		throw myError;
	}
}