CODEWITHSUNDEEP
javascripthtmlescapingxml-rpc
Joseph Turian
Joseph Turian

Reputation: 16560

Unescape HTML entities in JavaScript?

I have some JavaScript code that communicates with an XML-RPC backend. The XML-RPC returns strings of the form:

<img src='myimage.jpg'>

However, when I use JavaScript to insert the strings into HTML, they render literally. I don't see an image, I see the string:

<img src='myimage.jpg'>

I guess that the HTML is being escaped over the XML-RPC channel.

How can I unescape the string in JavaScript? I tried the techniques on this page, unsuccessfully: http://paulschreiber.com/blog/2008/09/20/javascript-how-to-unescape-html-entities/

What are other ways to diagnose the issue?

Upvotes: 309

Views: 375708

Answers (30)

Luis Lobo
Luis Lobo

Reputation: 859

UPDATE 09-01-2024

This worked for my purpose

 html-escaping.js v1
function htmlEscaping(string) { // https://stackoverflow.com/a/7382028/11212275
    return string.replace(/&/g, "&amp;")
        .replace(/</g, "&lt;")
        .replace(/>/g, "&gt;")
        .replace(/"/g, "&quot;")
        .replace(/'/g, "&#39;")
}

OLD (previous version containing errors)

It only works in simple specific cases

// decode-html.js v1
function decodeHtml(html) {
    const textarea = document.createElement('textarea');
    textarea.innerHTML = html;
    const decodedHtml = textarea.textContent;
    textarea.remove();
    return decodedHtml;
};

// encode-html.js v1
function encodeHtml(html) {
    const textarea = document.createElement('textarea');
    textarea.textContent = html;
    const encodedHtml = textarea.innerHTML;
    textarea.remove();
    return encodedHtml;
};

// example of use:
let htmlDecoded = 'one & two & three';
let htmlEncoded = 'one &amp; two &amp; three';

console.log(1, htmlDecoded);
console.log(2, encodeHtml(htmlDecoded));

console.log(3, htmlEncoded);
console.log(4, decodeHtml(htmlEncoded));

Upvotes: -1

firejox
firejox

Reputation: 31

function decodeEntities(input) {
  const temp = document.createElement('div');
  temp.innerHTML = `<data value="${input.replaceAll('"', '&quot;')}"></data>`;
  return temp.firstElementChild.value;
}

console.log(decodeEntities('&lt;img src="x"&gt;&lt;/img&gt;'));
console.log(decodeEntities('"/><img src="x" onerror="alert(\'xss\')"></img>'));

Upvotes: -1

Wladimir Palant
Wladimir Palant

Reputation: 57681

Most answers given here have a huge disadvantage: if the string you are trying to convert isn't trusted then you will end up with a Cross-Site Scripting (XSS) vulnerability. For the function in the accepted answer, consider the following:

htmlDecode("<img src='dummy' onerror='alert(/xss/)'>");

The string here contains an unescaped HTML tag, so instead of decoding anything the htmlDecode function will actually run JavaScript code specified inside the string.

This can be avoided by using DOMParser which is supported in all modern browsers:

function htmlDecode(input) {
  var doc = new DOMParser().parseFromString(input, "text/html");
  return doc.documentElement.textContent;
}

console.log(  htmlDecode("&lt;img src='myimage.jpg'&gt;")  )    
// "<img src='myimage.jpg'>"

console.log(  htmlDecode("<img src='dummy' onerror='alert(/xss/)'>")  )  
// ""

This function is guaranteed to not run any JavaScript code as a side-effect. Any HTML tags will be ignored, only text content will be returned.

Compatibility note: Parsing HTML with DOMParser requires at least Chrome 30, Firefox 12, Opera 17, Internet Explorer 10, Safari 7.1 or Microsoft Edge. So all browsers without support are way past their EOL and as of 2017 the only ones that can still be seen in the wild occasionally are older Internet Explorer and Safari versions (usually these still aren't numerous enough to bother).

Upvotes: 720

I wrestled a bear once.
I wrestled a bear once.

Reputation: 23399

The current top voted answer has the disadvantage of removing HTML from a string. If that isn't what you want (it certainly wasn't part of the question), then I suggest using regex to find HTML entities (/&[^;]*;/gmi), and then iterate thru the matches and just converting them.

function decodeHTMLEntities(str) {
  if (typeof str !== 'string') {
    return false;
  }
  var element = document.createElement('div');
  return str.replace(/&[^;]*;/gmi, entity => {
    entity = entity.replace(/</gm, '&lt;');
    element.innerHTML = entity;
    return element.textContent;
  });
}

var encoded_str = `<b>&#8593; &#67;&#65;&#78;'&#84;&nbsp;&#72;&#65;&#67;&#75;&nbsp;&#77;&#69;,&nbsp;&#66;&#82;&#79;</b>`;
var decoded_str = decodeHTMLEntities(encoded_str);

console.log(decoded_str);

Regarding XSS Attacks:

While innerHTML does not execute code in <script> tags, it is possible for code to be run in on* event attributes, so the above regex alone might be exploitable by a user passing in a string such as:

&<img src='asdfa' error='alert(`doin\' me a hack`)' />;

For that reason, it is necessary to convert any < characters to their &lt; before putting them in your hidden div element.

Also, just to cover all of my bases I'll note that functions defined like this, in the global scope can be overwritten by redifing them in the console, so it important to either define this function with const, or put it in a non-global scope.

Note: The attempted exploits in the following example confuse the stack snippet editor because of the preprocessing it does, so you'll have to run it in the browser's console, or in it's own file to see the result.

var tests = [
  "here's a spade: &spades;!",
  "&<script>alert('hackerman')</script>;",
  "&<img src='asdfa' error='alert(`doin\' me a hack`)' />;",
  "<b>&#8593; &#67;&#65;&#78;'&#84;&nbsp;&#72;&#65;&#67;&#75;&nbsp;&#77;&#69;,&nbsp;&#66;&#82;&#79;</b>"
];

var decoded = tests.map(decodeHTMLEntities).join("\n");
console.log(decoded);

The result is:

here's a spade: ♠!
&<script>alert('hackerman')</script>;
&<img src='asdfa' error='alert(`doin' me a hack`)' />;
<b>↑ CAN'T HACK ME, BRO</b>

Upvotes: -3

Slavik Meltser
Slavik Meltser

Reputation: 10381

I know there are a lot of good answers here, but since I have implemented a bit different approach, I thought to share.

This code is a perfectly safe security-wise approach, as the escaping handler dependant on the browser, instead on the function. So, if a new vulnerability will be discovered in the future, this solution will be covered.

const decodeHTMLEntities = text => {
    // Create a new element or use one from cache, to save some element creation overhead
    const el = decodeHTMLEntities.__cache_data_element 
             = decodeHTMLEntities.__cache_data_element 
               || document.createElement('div');
    
    const enc = text
        // Prevent any mixup of existing pattern in text
        .replace(/⪪/g, '⪪#')
        // Encode entities in special format. This will prevent native element encoder to replace any amp characters
        .replace(/&([a-z1-8]{2,31}|#x[0-9a-f]+|#\d+);/gi, '⪪$1⪫');

    // Encode any HTML tags in the text to prevent script injection
    el.textContent = enc;

    // Decode entities from special format, back to their original HTML entities format
    el.innerHTML = el.innerHTML
        .replace(/⪪([a-z1-8]{2,31}|#x[0-9a-f]+|#\d+)⪫/gi, '&$1;')
        .replace(/#⪫/g, '⪫');
   
    // Get the decoded HTML entities
    const dec = el.textContent;
    
    // Clear the element content, in order to preserve a bit of memory (it is just the text may be pretty big)
    el.textContent = '';

    return dec;
}

// Example
console.log(decodeHTMLEntities("<script>alert('&awconint;&CounterClockwiseContourIntegral;&#x02233;&#8755;⪪#x02233⪫');</script>"));
// Prints: <script>alert('∳∳∳∳⪪##x02233⪫');</script>

By the way, I have chosen to use the characters ⪪ and ⪫, because they are rarely used, so the chance of impacting the performance by matching them is significantly lower.

Upvotes: 2

weiya ou
weiya ou

Reputation: 4338

Closures can avoid creating unnecessary objects.

const decodingHandler = (() => {
  const element = document.createElement('div');
  return text => {
    element.innerHTML = text;
    return element.textContent;
  };
})();

A more concise way

const decodingHandler = (() => {
  const element = document.createElement('div');
  return text => ((element.innerHTML = text), element.textContent);
})();

Upvotes: 0

Łukasz K
Łukasz K

Reputation: 572

To unescape HTML entities* in JavaScript you can use small library html-escaper: npm install html-escaper

import {unescape} from 'html-escaper';

unescape('escaped string');

Or unescape function from Lodash or Underscore, if you are using it.

*) please note that these functions don't cover all HTML entities, but only the most common ones, i.e. &, <, >, ', ". To unescape all HTML entities you can use he library.

Upvotes: 5

nand-63
nand-63

Reputation: 117

function decodeHTMLContent(htmlText) {
  var txt = document.createElement("span");
  txt.innerHTML = htmlText;
  return txt.innerText;
}

var result = decodeHTMLContent('One &amp; two &amp; three');
console.log(result);

Upvotes: -2

Andrew Hodgkinson
Andrew Hodgkinson

Reputation: 4779

The question doesn't specify the origin of x but it makes sense to defend, if we can, against malicious (or just unexpected, from our own application) input. For example, suppose x has a value of &amp; <script>alert('hello');</script>. A safe and simple way to handle this in jQuery is:

var x    = "&amp; <script>alert('hello');</script>";
var safe = $('<div />').html(x).text();

// => "& alert('hello');"

Found via https://gist.github.com/jmblog/3222899. I can't see many reasons to avoid using this solution given it is at least as short, if not shorter than some alternatives and provides defence against XSS.

(I originally posted this as a comment, but am adding it as an answer since a subsequent comment in the same thread requested that I do so).

Upvotes: 4

Daniel
Daniel

Reputation: 1729

This is the most comprehensive solution I've tried so far:

const STANDARD_HTML_ENTITIES = {
    nbsp: String.fromCharCode(160),
    amp: "&",
    quot: '"',
    lt: "<",
    gt: ">"
};

const replaceHtmlEntities = plainTextString => {
    return plainTextString
        .replace(/&#(\d+);/g, (match, dec) => String.fromCharCode(dec))
        .replace(
            /&(nbsp|amp|quot|lt|gt);/g,
            (a, b) => STANDARD_HTML_ENTITIES[b]
        );
};

Upvotes: 0

David Chopin
David Chopin

Reputation: 3064

I was crazy enough to go through and make this function that should be pretty, if not completely, exhaustive:

function removeEncoding(string) {
    return string.replace(/&Agrave;/g, "À").replace(/&Aacute;/g, "Á").replace(/&Acirc;/g, "Â").replace(/&Atilde;/g, "Ã").replace(/&Auml;/g, "Ä").replace(/&Aring;/g, "Å").replace(/&agrave;/g, "à").replace(/&acirc;/g, "â").replace(/&atilde;/g, "ã").replace(/&auml;/g, "ä").replace(/&aring;/g, "å").replace(/&AElig;/g, "Æ").replace(/&aelig;/g, "æ").replace(/&szlig;/g, "ß").replace(/&Ccedil;/g, "Ç").replace(/&ccedil;/g, "ç").replace(/&Egrave;/g, "È").replace(/&Eacute;/g, "É").replace(/&Ecirc;/g, "Ê").replace(/&Euml;/g, "Ë").replace(/&egrave;/g, "è").replace(/&eacute;/g, "é").replace(/&ecirc;/g, "ê").replace(/&euml;/g, "ë").replace(/&#131;/g, "ƒ").replace(/&Igrave;/g, "Ì").replace(/&Iacute;/g, "Í").replace(/&Icirc;/g, "Î").replace(/&Iuml;/g, "Ï").replace(/&igrave;/g, "ì").replace(/&iacute;/g, "í").replace(/&icirc;/g, "î").replace(/&iuml;/g, "ï").replace(/&Ntilde;/g, "Ñ").replace(/&ntilde;/g, "ñ").replace(/&Ograve;/g, "Ò").replace(/&Oacute;/g, "Ó").replace(/&Ocirc;/g, "Ô").replace(/&Otilde;/g, "Õ").replace(/&Ouml;/g, "Ö").replace(/&ograve;/g, "ò").replace(/&oacute;/g, "ó").replace(/&ocirc;/g, "ô").replace(/&otilde;/g, "õ").replace(/&ouml;/g, "ö").replace(/&Oslash;/g, "Ø").replace(/&oslash;/g, "ø").replace(/&#140;/g, "Œ").replace(/&#156;/g, "œ").replace(/&#138;/g, "Š").replace(/&#154;/g, "š").replace(/&Ugrave;/g, "Ù").replace(/&Uacute;/g, "Ú").replace(/&Ucirc;/g, "Û").replace(/&Uuml;/g, "Ü").replace(/&ugrave;/g, "ù").replace(/&uacute;/g, "ú").replace(/&ucirc;/g, "û").replace(/&uuml;/g, "ü").replace(/&#181;/g, "µ").replace(/&#215;/g, "×").replace(/&Yacute;/g, "Ý").replace(/&#159;/g, "Ÿ").replace(/&yacute;/g, "ý").replace(/&yuml;/g, "ÿ").replace(/&#176;/g, "°").replace(/&#134;/g, "†").replace(/&#135;/g, "‡").replace(/&lt;/g, "<").replace(/&gt;/g, ">").replace(/&#177;/g, "±").replace(/&#171;/g, "«").replace(/&#187;/g, "»").replace(/&#191;/g, "¿").replace(/&#161;/g, "¡").replace(/&#183;/g, "·").replace(/&#149;/g, "•").replace(/&#153;/g, "™").replace(/&copy;/g, "©").replace(/&reg;/g, "®").replace(/&#167;/g, "§").replace(/&#182;/g, "¶").replace(/&Alpha;/g, "Α").replace(/&Beta;/g, "Β").replace(/&Gamma;/g, "Γ").replace(/&Delta;/g, "Δ").replace(/&Epsilon;/g, "Ε").replace(/&Zeta;/g, "Ζ").replace(/&Eta;/g, "Η").replace(/&Theta;/g, "Θ").replace(/&Iota;/g, "Ι").replace(/&Kappa;/g, "Κ").replace(/&Lambda;/g, "Λ").replace(/&Mu;/g, "Μ").replace(/&Nu;/g, "Ν").replace(/&Xi;/g, "Ξ").replace(/&Omicron;/g, "Ο").replace(/&Pi;/g, "Π").replace(/&Rho;/g, "Ρ").replace(/&Sigma;/g, "Σ").replace(/&Tau;/g, "Τ").replace(/&Upsilon;/g, "Υ").replace(/&Phi;/g, "Φ").replace(/&Chi;/g, "Χ").replace(/&Psi;/g, "Ψ").replace(/&Omega;/g, "Ω").replace(/&alpha;/g, "α").replace(/&beta;/g, "β").replace(/&gamma;/g, "γ").replace(/&delta;/g, "δ").replace(/&epsilon;/g, "ε").replace(/&zeta;/g, "ζ").replace(/&eta;/g, "η").replace(/&theta;/g, "θ").replace(/&iota;/g, "ι").replace(/&kappa;/g, "κ").replace(/&lambda;/g, "λ").replace(/&mu;/g, "μ").replace(/&nu;/g, "ν").replace(/&xi;/g, "ξ").replace(/&omicron;/g, "ο").replace(/&piρ;/g, "ρ").replace(/&rho;/g, "ς").replace(/&sigmaf;/g, "ς").replace(/&sigma;/g, "σ").replace(/&tau;/g, "τ").replace(/&phi;/g, "φ").replace(/&chi;/g, "χ").replace(/&psi;/g, "ψ").replace(/&omega;/g, "ω").replace(/&bull;/g, "•").replace(/&hellip;/g, "…").replace(/&prime;/g, "′").replace(/&Prime;/g, "″").replace(/&oline;/g, "‾").replace(/&frasl;/g, "⁄").replace(/&weierp;/g, "℘").replace(/&image;/g, "ℑ").replace(/&real;/g, "ℜ").replace(/&trade;/g, "™").replace(/&alefsym;/g, "ℵ").replace(/&larr;/g, "←").replace(/&uarr;/g, "↑").replace(/&rarr;/g, "→").replace(/&darr;/g, "↓").replace(/&barr;/g, "↔").replace(/&crarr;/g, "↵").replace(/&lArr;/g, "⇐").replace(/&uArr;/g, "⇑").replace(/&rArr;/g, "⇒").replace(/&dArr;/g, "⇓").replace(/&hArr;/g, "⇔").replace(/&forall;/g, "∀").replace(/&part;/g, "∂").replace(/&exist;/g, "∃").replace(/&empty;/g, "∅").replace(/&nabla;/g, "∇").replace(/&isin;/g, "∈").replace(/&notin;/g, "∉").replace(/&ni;/g, "∋").replace(/&prod;/g, "∏").replace(/&sum;/g, "∑").replace(/&minus;/g, "−").replace(/&lowast;/g, "∗").replace(/&radic;/g, "√").replace(/&prop;/g, "∝").replace(/&infin;/g, "∞").replace(/&OEig;/g, "Œ").replace(/&oelig;/g, "œ").replace(/&Yuml;/g, "Ÿ").replace(/&spades;/g, "♠").replace(/&clubs;/g, "♣").replace(/&hearts;/g, "♥").replace(/&diams;/g, "♦").replace(/&thetasym;/g, "ϑ").replace(/&upsih;/g, "ϒ").replace(/&piv;/g, "ϖ").replace(/&Scaron;/g, "Š").replace(/&scaron;/g, "š").replace(/&ang;/g, "∠").replace(/&and;/g, "∧").replace(/&or;/g, "∨").replace(/&cap;/g, "∩").replace(/&cup;/g, "∪").replace(/&int;/g, "∫").replace(/&there4;/g, "∴").replace(/&sim;/g, "∼").replace(/&cong;/g, "≅").replace(/&asymp;/g, "≈").replace(/&ne;/g, "≠").replace(/&equiv;/g, "≡").replace(/&le;/g, "≤").replace(/&ge;/g, "≥").replace(/&sub;/g, "⊂").replace(/&sup;/g, "⊃").replace(/&nsub;/g, "⊄").replace(/&sube;/g, "⊆").replace(/&supe;/g, "⊇").replace(/&oplus;/g, "⊕").replace(/&otimes;/g, "⊗").replace(/&perp;/g, "⊥").replace(/&sdot;/g, "⋅").replace(/&lcell;/g, "⌈").replace(/&rcell;/g, "⌉").replace(/&lfloor;/g, "⌊").replace(/&rfloor;/g, "⌋").replace(/&lang;/g, "⟨").replace(/&rang;/g, "⟩").replace(/&loz;/g, "◊").replace(/&#039;/g, "'").replace(/&amp;/g, "&").replace(/&quot;/g, "\"");
}

Used like so: 

let decodedText = removeEncoding("Ich hei&szlig;e David");
console.log(decodedText);

Prints: Ich Heiße David

P.S. this took like an hour and a half to make.

Upvotes: 0

jagjeet
jagjeet

Reputation: 377

var encodedStr = 'hello &amp; world';

var parser = new DOMParser;
var dom = parser.parseFromString(
    '<!doctype html><body>' + encodedStr,
    'text/html');
var decodedString = dom.body.textContent;

console.log(decodedString);

Upvotes: -2

buycanna.io
buycanna.io

Reputation: 1204

You're welcome...just a messenger...full credit goes to ourcodeworld.com, link below.

window.htmlentities = {
        /**
         * Converts a string to its html characters completely.
         *
         * @param {String} str String with unescaped HTML characters
         **/
        encode : function(str) {
            var buf = [];

            for (var i=str.length-1;i>=0;i--) {
                buf.unshift(['&#', str[i].charCodeAt(), ';'].join(''));
            }

            return buf.join('');
        },
        /**
         * Converts an html characterSet into its original character.
         *
         * @param {String} str htmlSet entities
         **/
        decode : function(str) {
            return str.replace(/&#(\d+);/g, function(match, dec) {
                return String.fromCharCode(dec);
            });
        }
    };

Full Credit: https://ourcodeworld.com/articles/read/188/encode-and-decode-html-entities-using-pure-javascript

Upvotes: 2

Christian C. Salvad&#243;
Christian C. Salvad&#243;

Reputation: 827852

EDIT: You should use the DOMParser API as Wladimir suggests, I edited my previous answer since the function posted introduced a security vulnerability.

The following snippet is the old answer's code with a small modification: using a textarea instead of a div reduces the XSS vulnerability, but it is still problematic in IE9 and Firefox.

function htmlDecode(input){
  var e = document.createElement('textarea');
  e.innerHTML = input;
  // handle case of empty input
  return e.childNodes.length === 0 ? "" : e.childNodes[0].nodeValue;
}

htmlDecode("&lt;img src='myimage.jpg'&gt;"); 
// returns "<img src='myimage.jpg'>"

Basically I create a DOM element programmatically, assign the encoded HTML to its innerHTML and retrieve the nodeValue from the text node created on the innerHTML insertion. Since it just creates an element but never adds it, no site HTML is modified.

It will work cross-browser (including older browsers) and accept all the HTML Character Entities.

EDIT: The old version of this code did not work on IE with blank inputs, as evidenced here on jsFiddle (view in IE). The version above works with all inputs.

UPDATE: appears this doesn't work with large string, and it also introduces a security vulnerability, see comments.

Upvotes: 200

Илья Зелень
Илья Зелень

Reputation: 8108

There is an variant that 80% as productive as the answers at the very top.

See the benchmark: https://jsperf.com/decode-html12345678/1

performance test

console.log(decodeEntities('test: &gt'));

function decodeEntities(str) {
  // this prevents any overhead from creating the object each time
  const el = decodeEntities.element || document.createElement('textarea')

  // strip script/html tags
  el.innerHTML = str
    .replace(/<script[^>]*>([\S\s]*?)<\/script>/gmi, '')
    .replace(/<\/?\w(?:[^"'>]|"[^"]*"|'[^']*')*>/gmi, '');

  return el.value;
}

If you need to leave tags, then remove the two .replace(...) calls (you can leave the first one if you do not need scripts).

Upvotes: -8

Digexart
Digexart

Reputation: 21

I tried everything to remove & from a JSON array. None of the above examples, but https://stackoverflow.com/users/2030321/chris gave a great solution that led me to fix my problem.

var stringtodecode="<B>Hello</B> world<br>";
document.getElementById("decodeIt").innerHTML=stringtodecode;
stringtodecode=document.getElementById("decodeIt").innerText

I did not use, because I did not understand how to insert it into a modal window that was pulling JSON data into an array, but I did try this based upon the example, and it worked:

var modal = document.getElementById('demodal');
$('#ampersandcontent').text(replaceAll(data[0],"&amp;", "&"));

I like it because it was simple, and it works, but not sure why it's not widely used. Searched hi & low to find a simple solution. I continue to seek understanding of the syntax, and if there is any risk to using this. Have not found anything yet.

Upvotes: 1

I am L
I am L

Reputation: 4652

You can use Lodash unescape / escape function https://lodash.com/docs/4.17.5#unescape

import unescape from 'lodash/unescape';

const str = unescape('fred, barney, &amp; pebbles');

str will become 'fred, barney, & pebbles'

Upvotes: 31

cslotty
cslotty

Reputation: 1797

In case you're looking for it, like me - meanwhile there's a nice and safe JQuery method.

https://api.jquery.com/jquery.parsehtml/

You can f.ex. type this in your console:

var x = "test &amp;";
> undefined
$.parseHTML(x)[0].textContent
> "test &"

So $.parseHTML(x) returns an array, and if you have HTML markup within your text, the array.length will be greater than 1.

Upvotes: 14

Ben White
Ben White

Reputation: 393

The trick is to use the power of the browser to decode the special HTML characters, but not allow the browser to execute the results as if it was actual html... This function uses a regex to identify and replace encoded HTML characters, one character at a time.

function unescapeHtml(html) {
    var el = document.createElement('div');
    return html.replace(/\&[#0-9a-z]+;/gi, function (enc) {
        el.innerHTML = enc;
        return el.innerText
    });
}

Upvotes: 20

EricP
EricP

Reputation: 3439

All of the other answers here have problems.

The document.createElement('div') methods (including those using jQuery) execute any javascript passed into it (a security issue) and the DOMParser.parseFromString() method trims whitespace. Here is a pure javascript solution that has neither problem:

function htmlDecode(html) {
    var textarea = document.createElement("textarea");
    html= html.replace(/\r/g, String.fromCharCode(0xe000)); // Replace "\r" with reserved unicode character.
    textarea.innerHTML = html;
    var result = textarea.value;
    return result.replace(new RegExp(String.fromCharCode(0xe000), 'g'), '\r');
}

TextArea is used specifically to avoid executig js code. It passes these:

htmlDecode('&lt;&amp;&nbsp;&gt;'); // returns "<& >" with non-breaking space.
htmlDecode('  '); // returns "  "
htmlDecode('<img src="dummy" onerror="alert(\'xss\')">'); // Does not execute alert()
htmlDecode('\r\n') // returns "\r\n", doesn't lose the \r like other solutions.

Upvotes: -2

ninhjs.dev
ninhjs.dev

Reputation: 8603

For one-line guys:

const htmlDecode = innerHTML => Object.assign(document.createElement('textarea'), {innerHTML}).value;

console.log(htmlDecode('Complicated - Dimitri Vegas &amp; Like Mike'));

Upvotes: 2

tmx976
tmx976

Reputation: 67

I use this in my project: inspired by other answers but with an extra secure parameter, can be useful when you deal with decorated characters

var decodeEntities=(function(){

    var el=document.createElement('div');
    return function(str, safeEscape){

        if(str && typeof str === 'string'){

            str=str.replace(/\</g, '&lt;');

            el.innerHTML=str;
            if(el.innerText){

                str=el.innerText;
                el.innerText='';
            }
            else if(el.textContent){

                str=el.textContent;
                el.textContent='';
            }

            if(safeEscape)
                str=str.replace(/\</g, '&lt;');
        }
        return str;
    }
})();

And it's usable like:

var label='safe <b> character &eacute;ntity</b>';
var safehtml='<div title="'+decodeEntities(label)+'">'+decodeEntities(label, true)+'</div>';

Upvotes: -1

LukeH
LukeH

Reputation: 269568

Do you need to decode all encoded HTML entities or just &amp; itself?

If you only need to handle &amp; then you can do this:

var decoded = encoded.replace(/&amp;/g, '&');

If you need to decode all HTML entities then you can do it without jQuery:

var elem = document.createElement('textarea');
elem.innerHTML = encoded;
var decoded = elem.value;

Please take note of Mark's comments below which highlight security holes in an earlier version of this answer and recommend using textarea rather than div to mitigate against potential XSS vulnerabilities. These vulnerabilities exist whether you use jQuery or plain JavaScript.

Upvotes: 317

Jeremy Banks - Vive le Canada
Jeremy Banks - Vive le Canada

Reputation: 129849

A more modern option for interpreting HTML (text and otherwise) from JavaScript is the HTML support in the DOMParser API (see here in MDN). This allows you to use the browser's native HTML parser to convert a string to an HTML document. It has been supported in new versions of all major browsers since late 2014.

If we just want to decode some text content, we can put it as the sole content in a document body, parse the document, and pull out the its .body.textContent.

var encodedStr = 'hello &amp; world';

var parser = new DOMParser;
var dom = parser.parseFromString(
    '<!doctype html><body>' + encodedStr,
    'text/html');
var decodedString = dom.body.textContent;

console.log(decodedString);

We can see in the draft specification for DOMParser that JavaScript is not enabled for the parsed document, so we can perform this text conversion without security concerns.

The parseFromString(str, type) method must run these steps, depending on type:

  • "text/html"

    Parse str with an HTML parser, and return the newly created Document.

    The scripting flag must be set to "disabled".

    NOTE

    script elements get marked unexecutable and the contents of noscript get parsed as markup.

It's beyond the scope of this question, but please note that if you're taking the parsed DOM nodes themselves (not just their text content) and moving them to the live document DOM, it's possible that their scripting would be reenabled, and there could be security concerns. I haven't researched it, so please exercise caution.

Upvotes: 113

HK JR
HK JR

Reputation: 270

a javascript solution that catches the common ones:

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

this is the reverse of https://stackoverflow.com/a/4835406/2738039

Upvotes: 2

Jason Williams
Jason Williams

Reputation: 2860

jQuery will encode and decode for you. However, you need to use a textarea tag, not a div. 

var str1 = 'One & two & three';
var str2 = "One &amp; two &amp; three";
  
$(document).ready(function() {
   $("#encoded").text(htmlEncode(str1)); 
   $("#decoded").text(htmlDecode(str2));
});

function htmlDecode(value) {
  return $("<textarea/>").html(value).text();
}

function htmlEncode(value) {
  return $('<textarea/>').text(value).html();
}
<script src="https://ajax.googleapis.com/ajax/libs/jquery/1.9.1/jquery.min.js"></script>

<div id="encoded"></div>
<div id="decoded"></div>

Upvotes: 10

Infoglaze.com
Infoglaze.com

Reputation: 85

First create a <span id="decodeIt" style="display:none;"></span> somewhere in the body

Next, assign the string to be decoded as innerHTML to this:

document.getElementById("decodeIt").innerHTML=stringtodecode

Finally,

stringtodecode=document.getElementById("decodeIt").innerText

Here is the overall code:

var stringtodecode="<B>Hello</B> world<br>";
document.getElementById("decodeIt").innerHTML=stringtodecode;
stringtodecode=document.getElementById("decodeIt").innerText

Upvotes: 4

Mark Amery
Mark Amery

Reputation: 155024

Matthias Bynens has a library for this: https://github.com/mathiasbynens/he

Example:

console.log(
    he.decode("J&#246;rg &amp J&#xFC;rgen rocked to &amp; fro ")
);
// Logs "Jörg & Jürgen rocked to & fro"

I suggest favouring it over hacks involving setting an element's HTML content and then reading back its text content. Such approaches can work, but are deceptively dangerous and present XSS opportunities if used on untrusted user input.

If you really can't bear to load in a library, you can use the textarea hack described in this answer to a near-duplicate question, which, unlike various similar approaches that have been suggested, has no security holes that I know of:

function decodeEntities(encodedString) {
    var textArea = document.createElement('textarea');
    textArea.innerHTML = encodedString;
    return textArea.value;
}

console.log(decodeEntities('1 &amp; 2')); // '1 & 2'

But take note of the security issues, affecting similar approaches to this one, that I list in the linked answer! This approach is a hack, and future changes to the permissible content of a textarea (or bugs in particular browsers) could lead to code that relies upon it suddenly having an XSS hole one day.

Upvotes: 55

avg_joe
avg_joe

Reputation: 195

element.innerText also does the trick.

Upvotes: 17

KajMagnus
KajMagnus

Reputation: 11686

CMS' answer works fine, unless the HTML you want to unescape is very long, longer than 65536 chars. Because then in Chrome the inner HTML gets split into many child nodes, each one at most 65536 long, and you need to concatenate them. This function works also for very long strings:

function unencodeHtmlContent(escapedHtml) {
  var elem = document.createElement('div');
  elem.innerHTML = escapedHtml;
  var result = '';
  // Chrome splits innerHTML into many child nodes, each one at most 65536.
  // Whereas FF creates just one single huge child node.
  for (var i = 0; i < elem.childNodes.length; ++i) {
    result = result + elem.childNodes[i].nodeValue;
  }
  return result;
}

See this answer about innerHTML max length for more info: https://stackoverflow.com/a/27545633/694469

Upvotes: 6

Related Questions