The Document Object Model (DOM) is what every web developer use to work with a web page. The DOM is an object-oriented, in-memory representation of an HTML document. To work with objects we need to be able to access them. There are several ways of doing this and I’ll go through the ones I could find. Most of these methods are members of the global document object, a note will be added when so is not the case.
I was planning on writing about them all in one blog post but changed my mind when I saw how many there is. This is the first part which focus on the old and tried methods.
Note that I will tell you about some browser bugs and differences but I’m all for writing for todays browsers and leaving the old behind. IE6 won’t disapear until we forget about it.
The most used, known and available method to access an element is
getElementById(id). It returns the first element that it finds which has its id attribute set to the value sent in the id parameter. The id is case-sensitive. If no element is found it will return
Note that IE versions below 8 differs from the other browsers and will also return all of the elements which has it’s name attribute set to the id parameter. Another difference is that that in IE versions below 8 the id is NOT case-sensitive.
If you create an element using a DOM method like
createElement, remember that the element needs to be in the DOM for it to be accessible through
getElementById. Use the
insertBefore method or any of its friends to insert the created element into the DOM.
If you want to access several elements in the DOM one option is to add classes to the elements. A class unlike an id can be used several times in a document. To access all of the elements with a specified class you use
getElementsByClassName(classes). Note the plural form of elements in the method name and in the parameter.
getElementsByClassName takes a string of classes separated by a space and returns a live NodeList. If no elements are found it will return an empty NodeList.
getElementsByClassName is not just a method of the document object but is also available on all HTML elements. This means that you can call it on another element and it will only search for elements which are descendants of that element (how about another element in this sentence). This can be a lot faster depending on the size of the document.
getElementsByClassName is not supported until IE9 and Firefox 3. You’ll have to add a polyfill/shim for the old browsers)
The returned NodeList objects are live. Live means that if the DOM (web page) changes, the list will be updated. This can cause errors if you depend on the order of the items in the list. You can read more about it here.
To reduce the use of classes when not needed you can use
getElementsByTagName(tag) instead. It works like
getElementsByClassName but searches for specific elements like
div and you can only send in one tag name. You can use
* to get all elements.
getElementsByClassName you can run
getElementsByTagName on all HTML elements to only get descendants of that element.
Note that IE7 has a bug where using
* on a
<object> element will always return an empty NodeList.
getElementsByName(name) gets all the elements which has the name attribute set to the name parameter. The method is case-sensitive so if you send in ‘browser’ and it says ‘BROWSER’ in the DOM it won’t find it.
Note that in IE the method is not case-sensitive. IE and Opera also returns elements with the id set to the given name parameter.
If you know about more bugs and differences, especially on newer browsers, please don’t hesitate to tell me about it.
The next part will focus on some newer methods, to some probably a little more interesting. The methods I will go through are querySelector, querySelectorAll, elementFromPoint, activeElement and evaluate.