Ignoring empty XML nodes in JavaScript

Question

I'm loading in an XML file in JavaScript, and once I've done that, I'm iterating through all of the Document's childNodes, which are all of the XML elements in the document. The problem that I'm having is that I need to ignore elements which are just not real elements, but rather newlines, tabs, etc. Right now I'm doing the following:

for (var i = 0; i < childList.length; i++)
{
    switch (childList[i].nodeType)
    {
        case 1: // Node.ELEMENT_NODE
            /* snip */
            break;
        case 3: // Node.TEXT_NODE
        case 8: // Node.COMMENT_NODE

            // Ensure the node is a valid node, and not newlines, tabs, etc
            if (!nodeEmpty(childList[i].nodeValue)
            {
                // do stuff
            }

            break;
    }
}

function nodeEmpty(nodeValue)
{
    var isEmpty = true;
    var length = nodeValue.length;
    for (var i = 0; i < length; i++)
    {
        if (nodeValue[i] != ' ' && nodeValue[i] != '\n' && nodeValue[i] != '\t')
        {
            isEmpty = false;
            break;
        }
    }

    return isEmpty;
}

But this seems like a very non-elegant way of achieving this. Is there a better way of doing this?

Breton · Accepted Answer · 2009-12-14 02:49:03Z

/**
 * xmlsimplify simplifies an XML object by copying it into a
 * javascript object with only a subset of the attributes of the original.
 * This makes it suitable for output as JSON, or for further processing by
 * other functions.

 * @param {Object} xml
 * @param {Boolean} strip if true, strip empty (whitespace) nodes
 */
var xmlsimplify = function(xml, strip) {
    var obj = {};
    if (typeof xml === "string") {
        xml = getXmlDocument(xml);
    }
    var traverse = function(node) {
        var i, l, n, a, j;
        if (node.nodeType) {
            var o = {};
            switch (node.nodeType) {
            case 1:
                //element node;
                o = {
                    nodeName: node.nodeName
                }; //record nodename
                for (i = 0, l = node.attributes.length, n = node.attributes; i < l; i++) { //append attributes
                    a = traverse(n.item(i));
                    for (j in a) {
                        if (a.hasOwnProperty(j)) {
                            o[j] = a[j];
                        }
                    }
                }
                if (node.childNodes.length) {
                    o.childNodes = [];
                    for (i = 0, l = node.childNodes.length, n = node.childNodes; i < l; i++) {
                        a = traverse(n.item(i));
                        if (a !== null) {
                            o.childNodes.push(a);
                        }
                    }
                    if (o.childNodes.length === 0) {
                        delete o.childNodes;
                    }
                }
                break;
            case 2:
                //attribute node
                o[node.nodeName] = node.nodeValue; //return an attribute object
                break;
            case 3:
                //text node
                //strip empty nodes
                if (node.nodeValue.match(/[^\s]/) && (strip === true)) {
                    o = node.nodeValue;
                } else {
                    o = null;
                }
                if (strip !== true) {
                    o = node.nodeValue;
                }
                break;
            case 4:
                //cdata section node
                o = node.nodeValue;
                break;
            case 9:
                //document node;
                o = traverse(node.firstChild);
                break;
            case 10:
                o = traverse(node.nextSibling);
                break;
            }
        }
        return o;
    };
    obj = traverse(xml);
    return obj;
};

I suppose the two are equivalent, but I find the former to be clearer. I'm a bit hazy on my character classes. Code clarity is generally important. I might know that \S is any non whitespace character today, but I may forget it again in 6 months time.

General Grievance · Accepted Answer · 2020-10-16 12:46:04Z

function btn2Click() {
    try{
        ConObj = CreateXMLObj();
        ConObj.onreadystatechange = function () {
        if (this.readyState == 4 && this.status == 200) {            
            xmlObj = this.responseXML;
            oTitle = xmlObj.getElementsByTagName('TITLE');
            oYear = xmlObj.getElementsByTagName('YEAR');
            txt = "<center><table border=2>"; 
            for (i = 0; i < oTitle.length ; i++) {
                txt += "<TR><TD>" + setNA(oTitle[i].childNodes) + "</TD><TD>" + setNA(oYear[i].childNodes) + "</TD> </TR>";
                              }
              document.getElementById("demo").innerHTML = txt; alert('stage1');
        }   
        
    };
    ConObj.open("GET", "AjaxManager.aspx?cmd=ajaxXML", true);
    ConObj.send();
    } catch (e) {
        alert(e.toString);
    }
}

    

function setNA(value) {
    name = "NA";
    if (value.length > 0)
        name = value[0].nodeValue;
    return name;
}

Collectives™ on Stack Overflow

Ignoring empty XML nodes in JavaScript

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related