0

  • I want to accept user input from the command line using the input() function, and I am expecting that the user provides input like \x41\x42\x43 to input "ABC". The user MUST enter input in the byte format, they can not provide the alphanumeric equivalent.

  • My issue is that when I take in user input, and then print it out, I see that python tries to escape the backslash with another backslash, so it is not able to interpret it as the byte it represents in ASCII.

Example Code from Python3 Command Prompt:

1 | >>> var_abc = "\x41\x42\x43"
2 | >>> print(var_abc)
3 | ABC
4 | >>> print(bytes(var_abc, encoding='ascii'))
5 | b'ABC'

6 | >>> user_input_abc = input('enter user input in bytes: ')
7 | enter user input in bytes: \x41\x42\x43
8 | >>> print(user_input_abc)
9 | \x41\x42\x43
10| >>> print(bytes(user_input_abc, encoding='ascii'))
11| b'\\x41\\x42\\x43'
  • I want the output on Line 11 to be the same as the output on Line 5. What do I need to do to make python interpret my user input as raw bytes and not escape each preceding backslash?
BitWrecker
  • 174
  • 1
  • 10

1 Answers1

1

To interpret a user input string as raw bytes. You can encode the string, then decode those bytes using the "unicode_escape" encoding and then encode that string again to get the bytes object

user_input_abc = '\\x41\\x42\\x43'
print(user_input_abc) # \x41\x42\x43
user_input_escaped = user_input_abc.encode().decode('unicode_escape')
print(user_input_escaped) # 'ABC'
user_input_bytes = user_input_escaped.encode()
print(user_input_bytes) # b'ABC'
Iain Shelvington
  • 31,030
  • 3
  • 31
  • 50
  • I don't understand why encoding just to turn around and immediately decode works, because I thought those two operations would just cancel each other out with no effect on the underlying string, but I think that did the trick so thank you. – BitWrecker May 01 '22 at 03:58
  • 1
    @BitWrecker decoding with the `unicode_escape` encoding is what does the trick, this only works on a bytes object though which is why you have to encode first – Iain Shelvington May 01 '22 at 04:01