Convert HTML Character Entities back to regular text using javascript

Question

the questions says it all :)

eg. we have >, we need > using only javascript

Update: It seems jquery is t

Answer 1

function decodeEntities(s){
    var str, temp= document.createElement('p');
    temp.innerHTML= s;
    str= temp.textContent || temp.innerText;
    temp=null;
    return str;
}

alert(decodeEntities('<'))

/*  returned value: (String)
<
*/

Answer 2

You could do something like this:

String.prototype.decodeHTML = function() {
    var map = {"gt":">" /* , … */};
    return this.replace(/&(#(?:x[0-9a-f]+|\d+)|[a-z]+);?/gi, function($0, $1) {
        if ($1[0] === "#") {
            return String.fromCharCode($1[1].toLowerCase() === "x" ? parseInt($1.substr(2), 16)  : parseInt($1.substr(1), 10));
        } else {
            return map.hasOwnProperty($1) ? map[$1] : $0;
        }
    });
};

Answer 3

There is nothing built in, but there are many libraries that have been written to do this.

Here is one.

And here one that is a jQuery plugin.

Answer 4

I know there are libraries out there, but here are a couple of solutions for browsers. These work well when placing html entity data strings into human editable areas where you want the characters to be shown, such as textarea's or input[type=text].

I add this answer as I have to support older versions of IE and I feel that it wraps up a few days worth of research and testing. I hope somebody finds this useful.

First this is for more modern browsers using jQuery, Please note that this should NOT be used if you have to support versions of IE before 10 (7, 8, or 9) as it will strip out the newlines leaving you with just one long line of text.

if (!String.prototype.HTMLDecode) {
    String.prototype.HTMLDecode = function () {
            var str = this.toString(),
            $decoderEl = $('<textarea />');

        str = $decoderEl.html(str)
            .text()
            .replace(/<br((\/)|( \/))?>/gi, "\r\n");

        $decoderEl.remove();

        return str;
    };
}

This next one is based on kennebec's work above, with some differences which are mostly for the sake of older IE versions. This does not require jQuery, but does still require a browser.

if (!String.prototype.HTMLDecode) {
    String.prototype.HTMLDecode = function () {
        var str = this.toString(),
            //Create an element for decoding            
            decoderEl = document.createElement('p');

        //Bail if empty, otherwise IE7 will return undefined when 
        //OR-ing the 2 empty strings from innerText and textContent
        if (str.length == 0) {
            return str;
        }

        //convert newlines to <br's> to save them
        str = str.replace(/((\r\n)|(\r)|(\n))/gi, " <br/>");            

        decoderEl.innerHTML = str;
        /*
        We use innerText first as IE strips newlines out with textContent.
        There is said to be a performance hit for this, but sometimes
        correctness of data (keeping newlines) must take precedence.
        */
        str = decoderEl.innerText || decoderEl.textContent;

        //clean up the decoding element
        decoderEl = null;

        //replace back in the newlines
        return str.replace(/<br((\/)|( \/))?>/gi, "\r\n");
    };
}

/* 
Usage: 
    var str = "&gt;";
    return str.HTMLDecode();

returned value: 
    (String) >    
*/

Answer 5

Here is a "class" for decoding whole HTML document.

HTMLDecoder = {
    tempElement: document.createElement('span'),
    decode: function(html) {
        var _self = this;
        html.replace(/&(#(?:x[0-9a-f]+|\d+)|[a-z]+);/gi,
            function(str) {
                _self.tempElement.innerHTML= str;
                str = _self.tempElement.textContent || _self.tempElement.innerText;
                return str;
            }
        );
    }
}

Note that I used Gumbo's regexp for catching entities but for fully valid HTML documents (or XHTML) you could simpy use /&[^;]+;/g.