Hexlet logo

JS: DOM API

Theory: Declarative DOM tree search

In frontend tasks, you usually manipulate a set of elements located deep down in the DOM tree. Moreover, these elements are often scattered over different parts of it.

For example, we can mark a list of files for deletion and then perform deleting. If you look at it from the perspective of changing the DOM tree, this task boils down to select all the elements that represent files and then deleting them.

Search methods

In such a case, manually traversing through the tree will be tedious. DOM offers several ways to solve this problem at once. The simplest way to search is to search by ID:

<p id="content">This is a paragraph</p>
const el = document.getElementById('content');

According to the specification, there cannot be two identical id on a page. So, the getElementById() method always returns one element.

On the other hand, there is a chance there may be several tags with the same id in HTML. In this situation, the browser will return the first element it encounters.

If you need to process several elements at once, then search by a class is a better option:

// Searching the entire tree
// All elements with this class are returned
// They can differ significantly
const collection = document.getElementsByClassName('row');

// This method allows you to search not only through the whole document,
// but also among the descendants of any element
el.getElementsByClassName('row');

If necessary, you can search by tags. It is rare in practical tasks, but still, it's useful to know about this method:

document.getElementsByTagName('span');

// search for all elements
document.getElementsByTagName('*');

// search among the descendants of el
el.getElementsByTagName('span');

Search by selector

The most universal search method is a search by a selector. You may remember that a selector is a rule that allows you to describe a set of elements in a DOM tree:

<ul id="menu">
  <li class="odd"><span>First</span> point</li>
  <li>Second</li>
  <li class="odd"><span>Third</span> point</li>
</ul>
// Returning the first element found with the specified selector
// Searching for the element with id=menu
const ul = document.querySelector('#menu');

// All spans nested in tags with the class .odd
const spans = ul.querySelectorAll('.odd > span');

We can apply both querySelector() and querySelectorAll() methods to the entire document or a specific element. The search will also go through all the descendants, as usual.

Other useful methods

matches

The predicate el.matches(css) checks whether el satisfies the css selector:

<p class="font-weight">This is Hexlet!</p>
const el = document.querySelector('p');
el.matches('.unknown-class'); // false
el.matches('.font-weight'); // true

closest

The el.closest(css) method searches for the closest element higher up the hierarchy that satisfies the selector. The element itself is also analyzed. If such an element is found, it will be returned, otherwise null will be returned:

<div class="row" id="one">
</div>
<div class="row" id="two">
  <div class="row" id="three">
    <span>where is the closest?</span>
  </div>
</div>
const el = document.querySelector('span');
const ancestor = el.closest('.row');
ancestor.id; // 'three'

XPath

It is a query language, developed to navigate through DOM trees in XML. It is supported by browsers:

<html>
 <body>
    <div>The first layer
      <span>a block of text in the first layer</span>
    </div>
    <div>Second layer</div>
    <div>Third layer
      <span class="text">the first block in the third layer</span>
      <span class="text">the second block in the third layer</span>
      <span>the third block in the third layer</span>
    </div>
    <span>the fourth layer</span>
    <img />
 </body>
</html>

The XPath path /html/body/*/span/@class will correspond to two elements of the source document in it:

  • <span class="text"> the first block in the third layer</span>
  • <span class="text">the second block in the third layer</span>

In everyday work, programmers rarely use XPath when working with DOM, so we're only learning it here to show the whole picture. But if you're working with XML documents, XPath is the main way to navigate a document.

Recommended programs

Navigation
Theory
Tests
Exercise