From MozillaZine Knowledge Base

Revision as of 22:26, 10 March 2005

Mozilla doesn't support the W3C's Document Object Model Load and Save at this moment, so the easiest way to serialize and deserialize DOM trees is to use the following Mozilla-specific interfaces:

• XMLSerializer to serialize DOM trees to strings or to files
• DOMParser to parse XML from strings into DOM trees
• XMLHttpRequest to parse XML from files into DOM trees. Although DOMParser does have a method named parseFromStream(), it's actually easier to use XMLHttpRequest which works for remote and local files.

Serializing DOM trees to strings

First, create a DOM tree using code like this.

Now, let's serialize doc — the DOM tree — to a string:

var serializer = new XMLSerializer();
var xml = serializer.serializeToString(doc);

Serializing DOM trees to files

First, create a DOM tree using code like this.

Now, let's serialize doc — the DOM tree — to a file (you can read more about using files in Mozilla here):

var serializer = new XMLSerializer();
var foStream = Components.classes["@mozilla.org/network/file-output-stream;1"]
               .createInstance(Components.interfaces.nsIFileOutputStream);
var file = Components.classes["@mozilla.org/file/directory_service;1"]
           .getService(Components.interfaces.nsIProperties)
           .get("ProfD", Components.interfaces.nsIFile); // get profile folder
file.append("extensions");   // extensions sub-directory
file.append("{5872365E-67D1-4AFD-9480-FD293BEBD20D}");   // GUID of your extension
file.append("myXMLFile.xml");   // filename
foStream.init(file, 0x02 | 0x08 | 0x20, 0664, 0);   // write, create, truncate
serializer.serializeToFile(doc, foStream, "IS0-8859-1");   // rememeber, doc is the DOM tree
foStream.close();

Parsing strings into DOM trees

var theString='<a id="a"><b id="b">hey!</b></a>';
var parser = new DOMParser();
var dom = parser.parseFromString(theString, "text/xml");
// print the name of the root element
dump(dom.documentElement.nodeName);

Parsing files into DOM trees

As was previously mentioned, even though DOMParser does have a method named parseFromStream(), it's easier to use XMLHttpRequest. That's what we'll demonstrate here.

First, read the local or remote XML file into a string:

var req = new XMLHttpRequest();
req.open("foo-bar!", "chrome://passwdmaker/content/people.xml", false); 
req.send(null);

The foo-bar argument of XMLHttpRequest.open() specifies the HTTP method to use if the URL (the second argument) is an HTTP(S) URL. Since we're using a chrome:// URL, the first argument to XMLHttpRequest.open() is ignored. You should change this to GET/POST/PUT/DELETE/OPTIONS/TRACE/HEAD if using an HTTP(S) URL.

Now that we have the XML as a string (in req.responseText), we can use the code similar to that in [#Parsing strings into DOM trees] to get the XML into a DOM tree:

var parser = new DOMParser();
var dom = parser.parseFromString(req.responseText, "text/xml");
// print the name of the root element
dump(dom.documentElement.nodeName);