28

Has HEX codec been excluded from Python 3.3? When I write the code

>>> s="Hallo"
>>> s.encode('hex')
Traceback (most recent call last):
  File "<pyshell#24>", line 1, in <module>
    s.encode('hex')
LookupError: unknown encoding: hex

What does that mean? I know about binascii.hexlify() but still .encode() method is nice!

Any suggestion?

U880D
  • 8,601
  • 6
  • 24
  • 40
iMagur
  • 721
  • 1
  • 6
  • 11

3 Answers3

52

No, using encode() to hexlify isn't nice.

The way you use the hex codec worked in Python 2 because you can call encode() on 8-bit strings in Python 2, ie you can encode something that is already encoded. That doesn't make sense. encode() is for encoding Unicode strings into 8-bit strings, not for encoding 8-bit strings as 8-bit strings.

In Python 3 you can't call encode() on 8-bit strings anymore, so the hex codec became pointless and was removed.

Although you theoretically could have a hex codec and use it like this:

>>> import codecs
>>> hexlify = codecs.getencoder('hex')
>>> hexlify(b'Blaah')[0]
b'426c616168'

Using binascii is easier and nicer:

>>> import binascii
>>> binascii.hexlify(b'Blaah')
b'426c616168'
Lennart Regebro
  • 167,292
  • 41
  • 224
  • 251
  • 1
    binascii.hexlify() need buffer interface, right? So, can somebody please explain me how to perform formatting on buffer interface? – iMagur Nov 18 '12 at 06:45
  • @iMagur: Sorry, I forgot to make the code Python 3 code, since I made it in Python 2 (as the hex_codec doesn't exist in Python 3). This has now been fixed. The above binascii code works in Python 3, and shows you how to do it. In Python 3 strings are Unicode, so they can't be hexlified directly, you need to encode them as `bytes` first. – Lennart Regebro Nov 18 '12 at 07:35
  • This is good because frankly there is a lot of 8 bit usage going on out there. Particularly between ('ascii') , ('IBM500'), ('IBM037'). What was really needed out of the box was a number of "OD" style hex methods for strings. Python is chosen to reduce effort and these things double the code required.. Yes I know IBM500 is unicode.. – mckenzm Jan 01 '15 at 21:52
2

this is the same answer for the above but i modified it so it works with python 3.

import binascii
from Crypto.Cipher import AES
from Crypto import Random

def encrypt(passwrd, message):
    msglist = []
    key = bytes(passwrd, "utf-8")
    iv = Random.new().read(AES.block_size)
    cipher = AES.new(key, AES.MODE_CFB, iv)
    msg = iv + cipher.encrypt(bytes(message, "utf-8"))
    msg = binascii.hexlify(msg)
    for letter in str(msg):
        msglist.append(letter)
    msglist.remove("b")
    msglist.remove("'")
    msglist.remove("'")
    for letter in msglist:
        print(letter, end="")
    print("")

def decrypt(passwrd, message):
    msglist = []
    key = bytes(passwrd, "utf-8")
    iv = Random.new().read(AES.block_size)
    cipher = AES.new(key, AES.MODE_CFB, iv)
    msg = cipher.decrypt(binascii.unhexlify(bytes(message, "utf-8")))[len(iv):]
    for letter in str(msg):
        msglist.append(letter)
    msglist.remove("b")
    msglist.remove("'")
    msglist.remove("'")
    for letter in msglist:
        print(letter, end="")
    print("")
king_quiksc0pe
  • 59
  • 3
0

regebro's answer is right but if you need to define a variable this code may help.C is a str here.

import codecs

c='sdfds'
print(codecs.encode(c))
Beryl Amend
  • 131
  • 1
  • 9
  • 1
    Your answer could be improved by adding more information on what the code does and how it helps the OP. – Tyler2P Mar 25 '23 at 12:13