Base64 Encode and Decode Unicode (UTF-8) strings in JavaScript and .Net

21 March 2018

The Base64 functions in JavaScript are made for ASCII values only. MDN has an excellent code sample for making this work with what they label UCS-2, but the original source labels as UTF-8. Hmm… in any case, it makes it safe for Unicode strings instead of just ASCII:

JavaScript

(From MDN):

Encode

// ucs-2 string to base64 encoded ascii
function utoa(str) {
	return window.btoa(unescape(encodeURIComponent(str)));
}

Decode

// base64 encoded ascii to ucs-2 string
function atou(str) {
	return decodeURIComponent(escape(window.atob(str)));
}

.Net

We can write an equivalent thing in .Net and choose our encoding (UTF-8 or otherwise) to make sure the data round-trips. I have tweaked these functions I found on SO to be extensions (called like myString.Base64Encode()) using the this keyword but you could remove that to use them in myString = Base64Encode(myString) style.

Encode

public static string Base64Encode(this string plainText) {
  var plainTextBytes = System.Text.Encoding.UTF8.GetBytes(plainText);
  return System.Convert.ToBase64String(plainTextBytes);
}

Decode

public static string Base64Decode(this string base64EncodedData) {
  var base64EncodedBytes = System.Convert.FromBase64String(base64EncodedData);
  return System.Text.Encoding.UTF8.GetString(base64EncodedBytes);
}