For example, garbled characters will occur, such as when multibyte is included in the user name.
It is probably caused by calling atob.
https://github.com/AzureAD/microsoft-authentication-library-for-js/blob/355dc1fb78651bece7806476879f9955f2cc61e0/src/Utils.ts#L67
HI @HiroyukiSako This is a known issue with window.atob() that it can't support unicode charcaters. There are different ways to solve this issue as per the MDN article( https://developer.mozilla.org/en-US/docs/Web/API/WindowBase64/Base64_encoding_and_decoding#The_Unicode_Problem) and below stackoverflow link.
https://stackoverflow.com/questions/30106476/using-javascripts-atob-to-decode-base64-doesnt-properly-decode-utf-8-strings
We need to discuss the fix in our team. I will report this as a bug for now.
We also hit this, some of our users have scandinavian characters in their names, like 脰 and 脛.
Those get garbled when the Id token is decoded. @nehaagrawal is this on your roadmap?
If you are not going to fix this soon, would you accept a PR if I make one? :D
@juunas11 Are you still open to submit a PR for this? We welcome contribution through PRs.
I'll try to look at it once I have some time. It's been pretty hectic lately so I can't promise a schedule ;)
Ok, i use the code found in decode-jwt on the jwt.ms and find the solution, but i just don't know how do the PR in the good way, if someone could help me or do it for me, here the solution:
need to be change for
return decodeURIComponent(atob(base64IdToken).replace(/(.)/g, function (m, p) {
var code = p.charCodeAt(0).toString(16).toUpperCase();
if (code.length < 2) {
code = '0' + code;
}
return '%' + code;
})); // jshint ignore:line
Before the fix:

After the fix:

i tried the solution by manually editing the Utils.js file found in ~node_modulesmsallib-commonjs
in hope that could help someone.
Tracking this with #596
@juunas11 I am trying to integrate this solution in our preview and will update the build details in the tagged #596. Please test and let us know if this works once we release the fix.
Most helpful comment
We also hit this, some of our users have scandinavian characters in their names, like 脰 and 脛.
Those get garbled when the Id token is decoded. @nehaagrawal is this on your roadmap?
If you are not going to fix this soon, would you accept a PR if I make one? :D