NodeJS SyntaxError: Unexpected token in JSON at position 0 - node.js

Question

The body of the response from Authorize.net's sandbox API is:
{
"messages": {
"resultCode": "Error",
"message": [
{
"code": "E00012",
"text": "You have submitted a duplicate of Subscription 5777085. A duplicate subscription will not be created."
}
]
}
}
but when I go to parse it:
try {
bodyObj = JSON.parse(body);
} catch (ex) {
console.error(ex);
}
I get this error:
SyntaxError: Unexpected token in JSON at position 0
And this: console.log(response.headers['content-type']);
returns this: application/json; charset=utf-8
What am I doing wrong? I want to parse the JSON into a JS object.

Answer 1

Actually you didn't see it, but there was a invisible unicode character, specifically the byte order mark at the beginning of the JSON.
Since the byte order mark is not a valid JSON character, JSON.parse rejected it.
To remove, use the following code.
function removeByteOrderMark(str){
return str.replace(/^\ufeff/g,"")
}
// OR (faster),
let removeByteOrderMark = a=>a[0]=="\ufeff"?a.slice(1):a

Answer 2

We had the same issue with JSON.parse() in a project we're working on. We're just using JSON.stringfy() and the object was working fine, but on the other platform where we received the data it had a similar error "Unexpected token in JSON at position 1".
Here's how we made it work:
In NodeJs we encoded the stringified object using encodeURI() then sent it
On the client side: decoded the string then used JSON.parse() on the decoded string
Analysis:
I tried to print the characters from the position, then replaced it with an empty string, then we realized it prints other weird characters even after replacing them. After that I realized it's an HTML code """ so instead of replacing it we went for encoding the string it and decoding.
We tried it on our case and it's working no problems