Builds a jQuery selector string that selects the given DOM element (adapted from http://snippets.dzone.com/posts/show/3754)

getjQuerySelector.js

var getjQuerySelector = (function() {
  
  function siblingIndexPseudoSelector(el) {
    var sibsBefore = 0,
        sibsAfter = 0,
        sib;
    for (sib = el.previousSibling; sib; sib = sib.previousSibling) {
      if(sib.nodeType == 1 && sib.tagName == el.tagName) {
        sibsBefore += 1;
      }
    }
    for (sib = el.nextSibling; sib; sib = sib.nextSibling) {
      if(sib.nodeType == 1 && sib.tagName == el.tagName) {
        sibsAfter += 1;
      }
    }
    return (sibsBefore + sibsAfter == 0) ? '' : ':eq('+sibsBefore+')';
  };
  
  function buildRoute(el, route) {
    route = route || [];
    if(el && el.nodeType === 1) {
      if(el.id) {
        route.push("#" + el.id);
      } else {
        route.push(el.tagName.toLowerCase() + siblingIndexPseudoSelector(el));
        return buildRoute(el.parentNode, route);
      }
    }
    return route;
  };
  
  return function(el) {
    return buildRoute(el).reverse().join('>');
  };
  
})();

Sure. A really simple example:

getjQuerySelector(document.getElementById('my_element')) === '#my_element'

It also builds more complex selectors, such as:

getjQuerySelector(document.body) === 'document>body'

or:

getjQuerySelector(document.body.childNodes[3]) === 'html>body>script:eq(0)'

The goal was to be able to serialise the position of a given DOM element in a way that could be used to select that element on an identical page at some point in the future (specifically, another instance of the same page). My main motivation for writing this is that XPath selectors are no longer available in jQuery.