Not sure what the preference would be - the advantage of making an xml() function would be that it is more sensical and would not allow a parameter, whereas html() would be consistent. In any case, below are both functions. xml() has been tested, html() was simply modified using the xml() code.

html: function( value ) {

var node = this[0];

return value === undefined ?

(node ?

(node.innerHTML ?

node.innerHTML.replace(/ jQuery\\d+="(?:\\d+|null)"/g, "") :

(node.xml || (new XMLSerializer()).serializeToString(node) || "")

.replace(new RegExp("(^<\\\\w*" + node.tagName + "[^>]*>)|(<\\\\w*\\\\/\\\\w*" + node.tagName + "[^>]*>$)", "gi"), "")) :

null) :

this.empty().append( value );

},

xml : function()

{

var node = this[0],

tagName = node.tagName;

if ( ! tagName )

return this.html();

return (node.xml || (new XMLSerializer()).serializeToString(node) || "").replace(

new RegExp("(^<\\\\w*" + tagName + "[^>]*>)|(<\\\\w*\\\\/\\\\w*" + tagName + "[^>]*>$)", "gi"), "");

}