Creating a new DOM element from an HTML string using built-in DOM methods or Prototype

I have an HTML string representing an element: '<li>text</li>'. I'd like to append it to an element in the DOM (a ul in my case). How can I do this with Prototype or with DOM methods?

(I know i could do this easily in jQuery, but unfortunately we're not using jQuery.)



Note: most current browsers support HTML <template> elements, which provide a more reliable way of turning creating elements from strings. See Mark Amery's answer below for details.

For older browsers, and node/jsdom: (which doesn't yet support <template> elements at the time of writing), use the following method. It's the same thing the libraries use to do to get DOM elements from an HTML string (with some extra work for IE to work around bugs with its implementation of innerHTML):

function createElementFromHTML(htmlString) {
  var div = document.createElement('div');
  div.innerHTML = htmlString.trim();

  // Change this to div.childNodes to support multiple top-level nodes
  return div.firstChild; 

Note that unlike HTML templates this won't work for some elements that cannot legally be children of a <div>, such as <td>s.

If you're already using a library, I would recommend you stick to the library-approved method of creating elements from HTML strings:


HTML 5 introduced the <template> element which can be used for this purpose (as now described in the WhatWG spec and MDN docs).

A <template> element is used to declare fragments of HTML that can be utilized in scripts. The element is represented in the DOM as a HTMLTemplateElement which has a .content property of DocumentFragment type, to provide access to the template's contents. This means that you can convert an HTML string to DOM elements by setting the innerHTML of a <template> element, then reaching into the template's .content property.


 * @param {String} HTML representing a single element
 * @return {Element}
function htmlToElement(html) {
    var template = document.createElement('template');
    html = html.trim(); // Never return a text node of whitespace as the result
    template.innerHTML = html;
    return template.content.firstChild;

var td = htmlToElement('<td>foo</td>'),
    div = htmlToElement('<div><span>nested</span> <span>stuff</span></div>');

 * @param {String} HTML representing any number of sibling elements
 * @return {NodeList} 
function htmlToElements(html) {
    var template = document.createElement('template');
    template.innerHTML = html;
    return template.content.childNodes;

var rows = htmlToElements('<tr><td>foo</td></tr><tr><td>bar</td></tr>');

Note that similar approaches that use a different container element such as a div don't quite work. HTML has restrictions on what element types are allowed to exist inside which other element types; for instance, you can't put a td as a direct child of a div. This causes these elements to vanish if you try to set the innerHTML of a div to contain them. Since <template>s have no such restrictions on their content, this shortcoming doesn't apply when using a template.

However, template is not supported in some old browsers. As of January 2018, Can I use... estimates 90% of users globally are using a browser that supports templates. In particular, no version of Internet Explorer supports them; Microsoft did not implement template support until the release of Edge.

If you're lucky enough to be writing code that's only targeted at users on modern browsers, go ahead and use them right now. Otherwise, you may have to wait a while for users to catch up.


Use insertAdjacentHTML(). It works with all current browsers, even with IE11.

var mylist = document.getElementById('mylist');
mylist.insertAdjacentHTML('beforeend', '<li>third</li>');
<ul id="mylist">


Newer DOM implementations have range.createContextualFragment, which does what you want in a framework-independent way.

It's widely supported. To be sure though, check its compatibility down in the same MDN link, as it will be changing. As of May 2017 this is it:

Feature         Chrome   Edge   Firefox(Gecko)  Internet Explorer   Opera   Safari
Basic support   (Yes)    (Yes)  (Yes)           11                  15.0    9.1.2

No need for any tweak, you got a native API:

const toNodes = html =>
    new DOMParser().parseFromString(html, 'text/html').body.childNodes[0]

For certain html fragments like <td>test</td>, div.innerHTML, DOMParser.parseFromString and range.createContextualFragment (without the right context) solutions mentioned in other answers here, won't create the <td> element.

jQuery.parseHTML() handles them properly (I extracted jQuery 2's parseHTML function into an independent function that can be used in non-jquery codebases).

If you are only supporting Edge 13+, it is simpler to just use the HTML5 template tag:

function parseHTML(html) {
    var t = document.createElement('template');
    t.innerHTML = html;
    return t.content.cloneNode(true);

var documentFragment = parseHTML('<td>Test</td>');

Heres a simple way to do it:

  var d=document
  return b;

var foo="<img src='//'>foo<i>bar</i>".toDOM();

With Prototype, you can also do:


<ul id="mylist"></ul>



You can create valid DOM nodes from a string using:


The following example adds a button element in the page taking the markup from a string:

let html = '<button type="button">Click Me!</button>';
let fragmentFromString = function (strHTML) {
  return document.createRange().createContextualFragment(strHTML);
let fragment = fragmentFromString(html);


I am using this method (Works in IE9+), although it will not parse <td> or some other invalid direct childs of body:

function stringToEl(string) {
    var parser = new DOMParser(),
        content = 'text/html',
        DOM = parser.parseFromString(string, content);

    // return element
    return DOM.body.childNodes[0];

stringToEl('<li>text</li>'); //OUTPUT: <li>text</li>

To enhance furthermore the useful .toDOM() snippet that we can find in different places, we can now safely use backticks (template literals).

So we can have single and double quotes in the foo html declaration.

This behave like heredocs for those familiar with the term.

This can be enhanced furthermore with variables, to make complex templating:

Template literals are enclosed by the back-tick () (grave accent) character instead of double or single quotes. Template literals can contain placeholders. These are indicated by the dollar sign and curly braces (${expression}). The expressions in the placeholders and the text between them get passed to a function. The default function just concatenates the parts into a single string. If there is an expression preceding the template literal (tag here), this is called a "tagged template". In that case, the tag expression (usually a function) gets called with the processed template literal, which you can then manipulate before outputting. To escape a back-tick in a template literal, put a backslash \ before the back-tick.

  var d=document,i
  a.innerHTML = this
  return b

// Using template litterals
var a = 10, b = 5
var foo=`
  onclick="alert('The future start today!')"   
foo${a + b}
img {cursor: crosshair}

So, why not use directly .innerHTML +=? By doing so, the whole DOM is being recalculated by the browser, it's much slower.


I added a Document prototype that creates an element from string:

Document.prototype.createElementFromString = function (str) {
    const element = new DOMParser().parseFromString(str, 'text/html');
    const child = element.documentElement.querySelector('body').firstChild;
    return child;




"use strict";

 * @author xgqfrms
 * @license MIT
 * @copyright xgqfrms
 * @description HTML5 Template
 * @augments
 * @example


    <img src="">

    <div class="myClass">I like: </div>


const showContent = () => {
    // let temp = document.getElementsByTagName("template")[0],
    let temp = document.querySelector(`[data-tempalte="tempalte-img"]`),
        clone = temp.content.cloneNode(true);

const templateGenerator = (datas = [], debug = false) => {
    let result = ``;
    // let temp = document.getElementsByTagName("template")[1],
    let temp = document.querySelector(`[data-tempalte="tempalte-links"]`),
        item = temp.content.querySelector("div");
    for (let i = 0; i < datas.length; i++) {
        let a = document.importNode(item, true);
        a.textContent += datas[i];
    return result;

const arr = ["Audi", "BMW", "Ford", "Honda", "Jaguar", "Nissan"];

if (document.createElement("template").content) {
    console.log("YES! The browser supports the template element");
    setTimeout(() => {
    }, 0);
} else {
    console.error("No! The browser does not support the template element");
@charset "UTf-8";

/* test.css */

:root {
    --cololr: #000;
    --default-cololr: #fff;
    --new-cololr: #0f0;

[data-class="links"] {
    color: white;
    background-color: DodgerBlue;
    padding: 20px;
    text-align: center;
    margin: 10px;
<!DOCTYPE html>
<html lang="zh-Hans">

    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <meta http-equiv="X-UA-Compatible" content="ie=edge">
    <title>Template Test</title>
    <!--[if lt IE 9]>
        <script src=""></script>

        <h1>Template Test</h1>
    <template data-tempalte="tempalte-img">
        <h3>Flower Image</h3>
        <img src="">
    <template data-tempalte="tempalte-links">
        <div data-class="links">I like: </div>
    <!-- js -->



Late but just as a note;

It's possible to add a trivial element to target element as a container and remove it after using.

// Tested on chrome 23.0, firefox 18.0, ie 7-8-9 and opera 12.11.

<div id="div"></div>

window.onload = function() {
    var foo, targetElement = document.getElementById('div')
    foo = document.createElement('foo')
    foo.innerHTML = '<a href="#" target="_self">Text of A 1.</a> '+
                    '<a href="#" onclick="return !!alert(this.innerHTML)">Text of <b>A 2</b>.</a> '+
                    '<hr size="1" />'
    // Append 'foo' element to target element

    // Add event
    foo.firstChild.onclick = function() { return !!alert( }

    while (foo.firstChild) {
        // Also removes child nodes from 'foo'
        targetElement.insertBefore(foo.firstChild, foo)
    // Remove 'foo' element from target element

Here's my code, and it works:

function parseTableHtml(s) { // s is string
    var div = document.createElement('table');
    div.innerHTML = s;

    var tr = div.getElementsByTagName('tr');
    // ...

For the heck of it I thought I'd share this over complicated but yet simple approach I came up with... Maybe someone will find something useful.

/*Creates a new element - By Jamin Szczesny*/
function _new(args){
    ele = document.createElement(args.node);
    delete args.node;
    for(x in args){ 
        if(typeof ele[x]==='string'){
            ele[x] = args[x];
            ele.setAttribute(x, args[x]);
    return ele;

/*You would 'simply' use it like this*/

    style:'position:absolute; left:100px; top:100px;'+
          'width:100px; height:100px; border:2px solid red;'+
          'cursor:pointer; background-color:HoneyDew',
    innerHTML:'My newly created div element!',
    value:'for example only',

Fastest solution to render DOM from string:

let render = (relEl, tpl, parse = true) => {
  if (!relEl) return;
  const range = document.createRange();
  const child = range.createContextualFragment(tpl);
  return parse ? relEl.appendChild(child) : {relEl, el};

And here u can check performance for DOM manipulation React vs native JS

Now u can simply use:

let element = render(document.body, `
<div style="font-size:120%;line-height:140%">
  <p class="bold">New DOM</p>

And of course in near future u use references from memory cause var "element" is your new created DOM in your document.

And remember "innerHTML=" is very slow :/

function domify (str) {
  var el = document.createElement('div');
  el.innerHTML = str;

  var frag = document.createDocumentFragment();
  return frag.appendChild(el.removeChild(el.firstChild));

var str = "<div class='foo'>foo</div>";

var msg = "test" jQuery.parseHTML(msg)

var jtag = ${ child:'text' }); // Represents: <li>text</li>
var htmlContent = $('mylist').html();
$('mylist').html(htmlContent + jtag.html());

Use jnerator


You can use the following function to convert the text "HTML" to the element

function htmlToElement(html)
  var element = document.createElement('div');
  element.innerHTML = html;
var html="<li>text and html</li>";
var e=htmlToElement(html);


Here is working code for me

I wanted to convert 'Text' string to HTML element

var diva = UWA.createElement('div');
diva.innerHTML = '<a href="">Text</a>';
var aelement = diva.firstChild;

This will work too:


It feels more like a jquery way with the chained function calls.


Recent Questions

Top Questions

Home Tags Terms of Service Privacy Policy DMCA Contact Us

©2020 All rights reserved.