Question

I’ve got a webpage which reads an XML file and loads the contents into a div on the page. As a part of this process, I need to identify all of the namespace prefixes and corresponding URIs declared in that file. I’m using jQuery to get and load the file like this:

$.get(sourceURI, function (data) {
    var nsList = getNamespaces(data);
    var target = $('#my_div');
    target.html(data);
});

where getNamespaces is a function taking the result of the get, and returning an object in the form:

object = {
    prefix1: uri1, //e.g xmlns:foo="http://bar.com" -> { foo: "http://bar.com" }
    prefix2: uri2,
    ....
    prefixn: urin
}

I have a sinking feeling that the answer may be a regex, but obviously that requires me to write one, and suffer over-used adages about having two problems from my colleagues. Is there a better way, or if not could someone point me in the right direction in constructing a regex?

Thanks!

Answer 1

If your browser is XHTML-compliant, you can use its parsing facilities to iterate over the XML elements with jQuery instead of processing a raw string with regular expressions:

function getNamespaces(data)
{
    var result = {};
    $(data).each(function() {
        recurseGetNamespaces(this, result);
    });
    return result;
}

function recurseGetNamespaces(element, result)
{
    var attributes = element.attributes;
    for (var i = 0; i < attributes.length; ++i) {
        var attr = attributes[i];
        if (attr.name.indexOf("xmlns:") == 0) {
            var prefix = attr.name.substr(6);
            if (!(prefix in result)) {
                result[prefix] = attr.value;
            }
        }
    }
    $(element).children().each(function() {
        recurseGetNamespaces(this, result);
    });
}

You can find a fiddle demonstrating this method here. (Disclaimer: that fiddle uses JSON.stringify() to display the results, so that part of the code might not work with browsers other than Firefox).