Java String.getBytes("UTF8") JavaScript analog

Question

Bytes to string and backward

Functions written there work properly that is pack(unpack("string")) yields to "string". But I would like to have the same result as "string".getBytes("UTF8") gives in Java.

The question is how to make a function giving the same functionality as Java getBytes("UTF8") in JavaScript?

For Latin strings unpack(str) from the article mentioned above provides the same result as getBytes("UTF8") except it adds 0 for odd positions. But with non-Latin strings it works completely different as it seems to me. Is there a way to work with string data in JavaScript like Java does?

Answer 1

You don't need to write a full-on UTF-8 encoder; there is a much easier JS idiom to convert a Unicode string into a string of bytes representing UTF-8 code units:

unescape(encodeURIComponent(str))

(This works because the odd encoding used by escape/unescape uses %xx hex sequences to represent ISO-8859-1 characters with that code, instead of UTF-8 as used by URI-component escaping. Similarly decodeURIComponent(escape(bytes)) goes in the other direction.)

So if you want an Array out it would be:

function toUTF8Array(str) {
    var utf8= unescape(encodeURIComponent(str));
    var arr= new Array(utf8.length);
    for (var i= 0; i<utf8.length; i++)
        arr[i]= utf8.charCodeAt(i);
    return arr;
}