Benchmark: TreeWalker/ filter vs querySelectorAll vs NodeIterator

HTML Preparation code:

<!-- article out start -->
<article>
  <!-- article in start -->
  <h1>Lorem ipsum</h1>
  <p>Lorem ipsum dolor sit amet, consectetur adipisicing elit. Cumque, nostrum.</p>
  <!-- article in end -->
</article>
<!-- article out end -->

<!-- article out start -->
<article>
  <!-- article in start -->
  <h1>Lorem ipsum</h1>
  <p>Lorem ipsum dolor sit amet, consectetur adipisicing elit. Cumque, nostrum.</p>
  <!-- article in end -->
</article>
<!-- article out end -->

<!-- article out start -->
<article>
  <!-- article in start -->
  <h1>Lorem ipsum</h1>
  <p>Lorem ipsum dolor sit amet, consectetur adipisicing elit. Cumque, nostrum.</p>
  <!-- article in end -->
</article>
<!-- article out end -->

Tests:

TreeWalker
const walker = document.createTreeWalker( document.body, NodeFilter.SHOW_ELEMENT, function(node) { return node.tagName == "H1" ? NodeFilter.FILTER_ACCEPT : NodeFilter.FILTER_SKIP; }, false ); for (let node = walker.nextNode(), i = 0; node; i++, node= walker.nextNode() ) { node.className = 'classy' ; }
querySelectorAll
const elements = Array.from(document.body.querySelectorAll('h1')) for(let el of elements){el.className = 'classy'}
NodeIterator
const walker = document.createNodeIterator( document.body, NodeFilter.SHOW_ELEMENT, function(node) { return node.tagName == "H1" ? NodeFilter.FILTER_ACCEPT : NodeFilter.FILTER_SKIP; }, false ); for (let node = walker.nextNode(), i = 0; node; i++, node= walker.nextNode() ) { node.className = 'classy' ; }

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
TreeWalker
querySelectorAll
NodeIterator

Fastest: N/A

Slowest: N/A

Latest run results:

Run details: (Test run date: one year ago)

User agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36

Browser/OS: Chrome 128 on Windows

View result in a separate tab

Test name	Executions per second
TreeWalker	66331.3 Ops/sec
querySelectorAll	197856.9 Ops/sec
NodeIterator	56191.2 Ops/sec

Autogenerated LLM Summary (model llama3.2:3b, generated one year ago):

LLMs can make mistakes. Check important info.

Let's dive into the world of JavaScript microbenchmarks on MeasureThat.net.

**Benchmark Overview**

The benchmark compares three different approaches to select and iterate over HTML elements:

1. **TreeWalker**: uses the `createTreeWalker` method to traverse the DOM tree, filtering elements based on their tag name.
2. **querySelectorAll**: uses the `querySelectorAll` method to select all elements with a specific tag name and then iterates over them.
3. **NodeIterator**: uses the `createNodeIterator` method to create an iterator over the DOM tree, filtering elements based on their tag name.

**Options Compared**

The benchmark compares three options:

* **TreeWalker**
	+ Pros: efficient use of resources, can be more accurate for complex filter conditions
	+ Cons: may have slower performance due to overhead of creating a tree walker
* **querySelectorAll**
	+ Pros: simple and widely supported, fast performance
	+ Cons: may not be as efficient for large datasets or complex filters
* **NodeIterator**
	+ Pros: similar to TreeWalker but with less overhead, more flexible filtering options
	+ Cons: may have slower performance due to creation of a node iterator

**Library/Feature Explanation**

* **TreeWalker**: the `createTreeWalker` method creates an object that allows traversing the DOM tree in a depth-first manner. The `NodeFilter.SHOW_ELEMENT` constant filters the traversal to only include elements, and the callback function is called for each element.
* **querySelectorAll**: the `querySelectorAll` method returns a NodeList containing all elements that match the specified selector. The `[el.className = 'classy']` part iterates over this list of elements.
* **NodeIterator**: similar to TreeWalker, but creates an iterator object instead of a tree walker.

**Special JS Feature/Syntax**

No special JavaScript features or syntax are used in these benchmarks, apart from using template literals and ES6-style string concatenation (`\r\n`).

**Alternatives**

Other alternatives for iterating over HTML elements could include:

* Using `document.body.querySelectorAll` with a CSS selector (e.g., `.h1`)
* Using a library like jQuery or Lodash to simplify DOM manipulation
* Using a more specialized approach, such as using `Element.prototype.querySelectorAll` or `CSSRuleList`

However, these alternatives are not directly comparable to the three options tested in this benchmark.

**Benchmark Insights**

From the latest benchmark results, we can see that:

* **querySelectorAll** is the fastest option for executing 197856.90625 times per second.
* **TreeWalker** is slower than `querySelectorAll` but faster than `NodeIterator`, with 66331.3359375 executions per second.
* **NodeIterator** is the slowest option, with 56191.17578125 executions per second.

These results suggest that `querySelectorAll` is a good choice for simple cases where speed is prioritized over code complexity or flexibility. However, if accuracy and flexibility are more important, TreeWalker might be a better option.

Related benchmarks:

TreeWalker/ filter vs querySelectorAll vs NodeIterator (version: 0)

Comparing performance of: TreeWalker vs querySelectorAll vs NodeIterator

Created: 4 years ago by: Guest

Jump to the latest result

TreeWalker

querySelectorAll

NodeIterator

Suite status: <idle, ready to run>

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated one year ago):