b64encode when going from Python2 to Python3

Question

I am running some code that runs on on Python2 to Python3 and it is having some issues. I have a string with formatting:

auth_string = '{client_id}:{client_secret}'.format(client_id=client_id, client_secret=client_secret)

and am passing it in as part of "headers":

headers = {
            'Accept': 'application/json',
            'Content-Type': 'application/x-www-form-urlencoded;charset=UTF-8',
            'Authorization': 'Basic ' + b64encode(auth_string)
        }

When I run the code I get this error:

TypeError: 'str' does not support the buffer interface

After some research, it is because Python3 considers strings as unicode objects and you need to convert them to bytes first. No problem, I change the line to:

'Authorization': 'Basic ' + b64encode(auth_string.encode(encoding='utf_8'))

But now I get a new error:

TypeError: Can't convert 'bytes' object to str implicitly

What exactly am I missing here?

Answer 1

b64encode accepts bytes and returns bytes . To merge with string, do also decode .

'Authorization': 'Basic ' + b64encode(auth_string.encode()).decode()

Answer 2

in Python3, strings are either bytes or unicode .

Just prefix your strings with b :

b'Basic ' + b64encode(auth_string.encode(encoding='utf_8'))

Answer 3

You should cast your str var to a bytes var:

To cast str to bytes str should be content only ascii chars.

base64.64encode(auth_string.encode(encoding='ascii'))

or

base64.64encode(b'bytes string')