Decode mixed ASCII codes from a string in JavaScript

Question

I've a JavaScript code that extracts JSON strings from other pages from my blog (Blogger), but many special characters in those strings are as &#?????;, where ????? is a number up to 5 digits, or as something like \74br /\76, which should be a <br />.

Both come mixed in the same string, and both seem to be ASCII, the first one being decimal/html and the second one being octal.

How can I decode this mess to their respective characters by using JavaScript? Is there any existing function or proper solution for this?

Answer 1

These should get you started

function decodeHtmlNumeric( str ) {
    return str.replace( /&#([0-9]{1,7});/g, function( g, m1 ){
        return String.fromCharCode( parseInt( m1, 10 ) );
    }).replace( /&#[xX]([0-9a-fA-F]{1,6});/g, function( g, m1 ){
        return String.fromCharCode( parseInt( m1, 16 ) );
    });
}

function decodeOctal( str ) {
    return str.replace( /\\([0-7]+)/g, function( g, m1 ) {
        return String.fromCharCode( parseInt( m1, 8 ) );
    });
}
           //Double \\ = one backslash 
decodeOctal("\\74br /\\76"); //"<br />"
decodeHtmlNumeric("&#255;"); //"ÿ"