How to write a decoder, which can then be tested with pytest?

Question

For a school assignment I have to write an encoder and decoder in Python, which takes a string of encoded characters as so : "2k3b" and decodes them into "kkbbb". I have already written the encoder and tested it with pytest, but I can't seem to wrap my head around writing the decoder. The code for the encoder and decoder respectively is posted below.

def encode(mess):
""" Run length encoding - convert string from 'wwwwwwbbbb' til '6w4b'""" 
res = []
old_val = mess[0]
count = 0

for char in mess:
    if char == old_val:
        count += 1
    else:
        res.append('%d%c' % (count, old_val))
        old_val = char
        count = 1
res.append('%d%c' % (count, char))
return ''.join(res)

I know that last res.append('%d%c' % (count, char)) is bad coding, but I just had to make it work and i'm not experienced enough in Python.

def decode(mess):
"""Run length decoding - convert string from '2k3b' to 'kkbbb'"""
res = []
num = 0
letter = ''
for char in mess:
    mess.split()
    if char.isdigit():
        num = int(char)
    else:
        num = 0
        res.append(num * '%c' % letter)
return ''.join(res)

The decoder is not finished it's just were I stopped because I couldn't "see" what I was supposed to do. I usually write pseudo-code for all assigments and I did so for both the encoder and decoder but for some reason the decoder isn't getting through to me. I believe that I just lost my trail of thought but I could really use some guidance on this one. I'm reluctant to ask my teacher because he will just show me his code and I won't learn anything from that.

Answer 1

You are almost there. Below is your approach but with few minor changes.

def decode(mess):
    """Run length decoding - convert string from '2k3b' to 'kkbbb'"""
    res = []
    num = ''
    for char in mess:
        #If char an integer just add it to current num
        # required for cases where integer will be greater than 9.
        if char.isdigit():
            num += char
        else:
        # "a"* int("3") = "aaa" we can use this property to expand 
        # after expanding just set the num to 0
            res.append(char*int(num))
            num = ''
    return ''.join(res)

Test:

>>>print(decode(encode('kkkkkkkkkkkkkkkkkwwkkeeerrr'))=="kkkkkkkkkkkkkkkkkwwkkeeerrr"))
True

Answer 2

Your current approach would be limited to only single digit strings of numbers and struggle when finding 10 or more of a string of character, instead of you'd be better off using a regex to find strings of numbers followed by characters and then doing a join of these together

import re
letter_groups = re.findall("(\d+)(\D+)", mess)
return  "".join(f"{c * int(num)}" for num, c in letter_groups)