HtmlSpecialChars equivalent in Javascript?


Apparently, this is harder to find than I thought it would be. And it even is so simple...

Is there a function equivalent to PHP's htmlspecialchars built into Javascript? I know it's fairly easy to implement that yourself, but using a built-in function, if available, is just nicer.

For those unfamiliar with PHP, htmlspecialchars translates stuff like <htmltag/> into &lt;htmltag/&gt;

I know that escape() and encodeURI() do not work this way.

There is a problem with your solution code--it will only escape the first occurrence of each special character. For example:

escapeHtml('Kip\'s <b>evil</b> "test" code\'s here');
Actual:   Kip&#039;s &lt;b&gt;evil</b> &quot;test" code's here
Expected: Kip&#039;s &lt;b&gt;evil&lt;/b&gt; &quot;test&quot; code&#039;s here

Here is code that works properly:

function escapeHtml(text) {
  return text
      .replace(/&/g, "&amp;")
      .replace(/</g, "&lt;")
      .replace(/>/g, "&gt;")
      .replace(/"/g, "&quot;")
      .replace(/'/g, "&#039;");
}

Update

The following code will produce identical results to the above, but it performs better, particularly on large blocks of text (thanks jbo5112).

function escapeHtml(text) {
  var map = {
    '&': '&amp;',
    '<': '&lt;',
    '>': '&gt;',
    '"': '&quot;',
    "'": '&#039;'
  };

  return text.replace(/[&<>"']/g, function(m) { return map[m]; });
}

That's HTML Encoding. There's no native javascript function to do that, but you can google and get some nicely done up ones.

E.g. http://sanzon.wordpress.com/2008/05/01/neat-little-html-encoding-trick-in-javascript/

EDIT:
This is what I've tested:

var div = document.createElement('div');
  var text = document.createTextNode('<htmltag/>');
  div.appendChild(text);
  console.log(div.innerHTML);

Output: &lt;htmltag/&gt;


Worth a read: http://bigdingus.com/2007/12/29/html-escaping-in-javascript/

escapeHTML: (function() {
 var MAP = {
   '&': '&amp;',
   '<': '&lt;',
   '>': '&gt;',
   '"': '&#34;',
   "'": '&#39;'
 };
  var repl = function(c) { return MAP[c]; };
  return function(s) {
    return s.replace(/[&<>'"]/g, repl);
  };
})()

Note: Only run this once. And don't run it on already encoded strings e.g. &amp; becomes &amp;amp;


With jQuery it can be like this:

var escapedValue = $('<div/>').text(value).html();

From related question Escaping HTML strings with jQuery

As mentioned in comment double quotes and single quotes are left as-is for this implementation. That means this solution should not be used if you need to make element attribute as a raw html string.


Here's a function to escape HTML:

function escapeHtml(str)
{
    var map =
    {
        '&': '&amp;',
        '<': '&lt;',
        '>': '&gt;',
        '"': '&quot;',
        "'": '&#039;'
    };
    return str.replace(/[&<>"']/g, function(m) {return map[m];});
}

And to decode:

function decodeHtml(str)
{
    var map =
    {
        '&amp;': '&',
        '&lt;': '<',
        '&gt;': '>',
        '&quot;': '"',
        '&#039;': "'"
    };
    return str.replace(/&amp;|&lt;|&gt;|&quot;|&#039;/g, function(m) {return map[m];});
}

Underscore.js provides a function for this:

_.escape(string)

Escapes a string for insertion into HTML, replacing &, <, >, ", and ' characters.

http://underscorejs.org/#escape

It's not a built-in Javascript function, but if you are already using Underscore it is a better alternative than writing your own function if your strings to convert are not too large.


Yet another take at this is to forgo all the character mapping altogether and to instead convert all unwanted characters into their respective numeric character references, e.g.:

function escapeHtml(raw) {
    return raw.replace(/[&<>"']/g, function onReplace(match) {
        return '&#' + match.charCodeAt(0) + ';';
    });
}

Note that the specified RegEx only handles the specific characters that the OP wanted to escape but, depending on the context that the escaped HTML is going to be used, these characters may not be sufficient. Ryan Grove’s article There's more to HTML escaping than &, <, >, and " is a good read on the topic. And depending on your context, the following RegEx may very well be needed in order to avoid XSS injection:

var regex = /[&<>"'` [email protected]$%()=+{}[\]]/g

String.prototype.escapeHTML = function() {
        return this.replace(/&/g, "&amp;")
                   .replace(/</g, "&lt;")
                   .replace(/>/g, "&gt;")
                   .replace(/"/g, "&quot;")
                   .replace(/'/g, "&#039;");
    }

sample :

var toto = "test<br>";
alert(toto.escapeHTML());

Chances are you don't need such a function. Since your code is already in the browser*, you can access the DOM directly instead of generating and encoding HTML that will have to be decoded backwards by the browser to be actually used.

Use innerText property to insert plain text into the DOM safely and much faster than using any of the presented escape functions. Even faster than assigning a static preencoded string to innerHTML.

Use classList to edit classes, dataset to set data- attributes and setAttribute for others.

All of these will handle escaping for you. More precisely, no escaping is needed and no encoding will be performed underneath**, since you are working around HTML, the textual representation of DOM.

// use existing element
var author = 'John "Superman" Doe <[email protected]>';
var el = document.getElementById('first');
el.dataset.author = author;
el.textContent = 'Author: '+author;

// or create a new element
var a = document.createElement('a');
a.classList.add('important');
a.href = '/search?q=term+"exact"&n=50';
a.textContent = 'Search for "exact" term';
document.body.appendChild(a);

// actual HTML code
console.log(el.outerHTML);
console.log(a.outerHTML);
.important { color: red; }
<div id="first"></div>

* This answer is not intended for server-side JavaScript users (Node.js, etc.)

** Unless you explicitly convert it to actual HTML afterwards. E.g. by accessing innerHTML - this is what happens when you run $('<div/>').text(value).html(); suggested in other answers. So if your final goal is to insert some data into the document, by doing it this way you'll be doing the work twice. Also you can see that in the resulting HTML not everything is encoded, only the minimum that is needed for it to be valid. It is done context-dependently, that's why this jQuery method doesn't encode quotes and therefore should not be used as a general purpose escaper. Quotes escaping is needed when you're constructing HTML as a string with untrusted or quote-containing data at the place of an attribute's value. If you use the DOM API, you don't have to care about escaping at all.


For Node.JS users (or users utilizing Jade runtime in the browser), you can use Jade's escape function.

require('jade').runtime.escape(...);

No sense in writing it yourself if someone else is maintaining it. :)


I am elaborating a bit on o.k.w.'s answer.

You can use the browser's DOM functions for that.

var utils = {
    dummy: document.createElement('div'),
    escapeHTML: function(s) {
        this.dummy.textContent = s
        return this.dummy.innerHTML
    }
}

utils.escapeHTML('<escapeThis>&')

This returns &lt;escapeThis&gt;&amp;

It uses the standard function createElement to create an invisible element, then uses the function textContent to set any string as its content and then innerHTML to get the content in its HTML representation.


function htmlspecialchars(str) {
 if (typeof(str) == "string") {
  str = str.replace(/&/g, "&amp;"); /* must do &amp; first */
  str = str.replace(/"/g, "&quot;");
  str = str.replace(/'/g, "&#039;");
  str = str.replace(/</g, "&lt;");
  str = str.replace(/>/g, "&gt;");
  }
 return str;
 }

Hope this wins the race due to its performance and most important not a chained logic using .replace('&','&').replace('<','<')...

var mapObj = {
   '&':"&amp;",
   '<':"&lt;",
   '>':"&gt;",
   '"':"&quot;",
   '\'':"&#039;"
};
var re = new RegExp(Object.keys(mapObj).join("|"),"gi");

function escapeHtml(str) 
{   
    return str.replace(re, function(matched)
    {
        return mapObj[matched.toLowerCase()];
    });
}

console.log('<script type="text/javascript">alert('Hello World');</script>');
console.log(escapeHtml('<script type="text/javascript">alert('Hello World');</script>'));

Reversed one:

function decodeHtml(text) {
    return text
        .replace(/&amp;/g, '&')
        .replace(/&lt;/ , '<')
        .replace(/&gt;/, '>')
        .replace(/&quot;/g,'"')
        .replace(/&#039;/g,"'");
}

function htmlEscape(str){
    return str.replace(/[&<>'"]/g,x=>'&#'+x.charCodeAt(0)+';')
}

This solution uses the numerical code of the characters, for example < is replaced by &#60;.

Although its performance is slightly worse than the solution using a map, it has the advantages:

  • Not dependent on a library or DOM
  • Pretty easy to remember (you don't need to memorize the 5 HTML escape characters)
  • Little code
  • Reasonably fast (it's still faster than 5 chained replace)

OWASP recommends that "[e]xcept for alphanumeric characters, [you should] escape all characters with ASCII values less than 256 with the &#xHH; format (or a named entity if available) to prevent switching out of [an] attribute."

So here's a function that does that, with a usage example:

function escapeHTML(unsafe) {
  return unsafe.replace(
    /[\u0000-\u002F]|[\u003A-\u0040]|[\u005B-\u00FF]/g,
    c => '&#' + ('000' + c.charCodeAt(0)).substr(-4, 4) + ';'
  )
}
document.querySelector('div').innerHTML =
  '<span class=' +
  escapeHTML('this should break it! " | / % * + , - / ; < = > ^') +
  '>' +
  escapeHTML('<script>alert("inspect the attributes")\u003C/script>') +
  '</span>'
<div></div>

Disclaimer: You should verify the entity ranges I have provided to validate the safety yourself.