querySelector and querySelectorAll vs getElementsByClassName and getElementById in JavaScript

For this answer, I refer to querySelector and querySelectorAll as querySelector* and to getElementById, getElementsByClassName, getElementsByTagName, and getElementsByName as getElement*.

A lot of this information can be verified in the specification, a lot of it is from various benchmarks I ran when I wrote it. The spec: https://dom.spec.whatwg.org/

Main Differences

1. querySelector* is more flexible, as you can pass it any CSS3 selector, not just simple ones for id, tag, or class.
2. The performance of querySelector* changes with the size of the DOM that it is invoked on. To be precise, querySelector* calls run in O(n) time and getElement* calls run in O(1) time, where n is the total number of all children of the element or document it is invoked on. This fact seems to be the least well-known, so I am bolding it.
3. The return types of these calls vary. querySelector and getElementById both return a single element. querySelectorAll and getElementsByName both return NodeLists. The older getElementsByClassName and getElementsByTagName both return HTMLCollections. NodeLists and HTMLCollections are both referred to as collections of elements.
4. Collections can contain references to elements in the DOM, or copies of elements. getElement* calls return collections of references, whereas querySelectorAll results contain copies of the elements. These are referred to as “live” and “static” collections respectively. This is NOT related to the types that they return.

These concepts are summarized in the following table.

Function               | Live? | Type           | Time Complexity
querySelector          |       | Element        |  O(n)
querySelectorAll       |   N   | NodeList       |  O(n)
getElementById         |       | Element        |  O(1)
getElementsByClassName |   Y   | HTMLCollection |  O(1)
getElementsByTagName   |   Y   | HTMLCollection |  O(1)
getElementsByName      |   Y   | NodeList       |  O(1)

Details, Tips, and Examples

• HTMLCollections are not as array-like as NodeLists and do not support .forEach(). I find the spread operator useful to work around this:

  [...document.getElementsByClassName("someClass")].forEach()

• Every element, and the global document, have access to all of these functions except for getElementById and getElementsByName, which are only implemented on document.

• Chaining getElement* calls instead of using querySelector* will improve performance, especially on very large DOMs. Even on small DOMs and/or with very long chains, it is generally faster. However, unless you know you need the performance, the readability of querySelector* should be preferred. querySelectorAll is often harder to rewrite, because you must select elements from the NodeList or HTMLCollection at every step. For example, the following code does not work:

  document.getElementsByClassName("someClass").getElementsByTagName("div")

  because you can only use getElements* on single elements, not collections. For example:

  document.querySelector("#someId .someClass div")

  could be written as:

  document.getElementById("someId").getElementsByClassName("someClass")[0].getElementsByTagName("div")[0]

  Note the use of [0] to get just the first element of the collection at each step that returns a collection, resulting in one element at the end just like with querySelector.

• Since all elements have access to both querySelector* and getElement* calls, you can make chains using both calls, which can be useful if you want some performance gain, but cannot avoid a querySelector that can not be written in terms of the getElement* calls.

• Though it is generally easy to tell if a selector can be written using only getElement* calls, there is one case that may not be obvious:

  document.querySelectorAll(".class1.class2")

  can be rewritten as

  document.getElementsByClassName("class1 class2")

• Using getElement* on a static element fetched with querySelector* will result in an element that is live with respect to the static subset of the DOM copied by querySelector, but not live with respect to the full document DOM… this is where the simple live/static interpretation of elements begins to fall apart. You should probably avoid situations where you have to worry about this, but if you do, remember that querySelector* calls copy elements they find before returning references to them, but getElement* calls fetch direct references without copying.

• querySelector* and getElementById traverse elements in preorder, depth-first, called “tree order” in the specification. With other getElement* calls it is not clear to me from the specification – they may be the same as tree order, but getElementsByClassName(".someClass")[0] may not reliably give the same result in every browser. getElementById("#someId") should though, even if you have multiple copies of the same id on your page.