Encode and Decode Strings

The Idea

Joining parts with a separator breaks the moment a payload contains that separator. Prefix each part with its length instead: 5#hello. The decoder reads digits up to the marker, then copies exactly that many characters without looking at them. Empty strings survive, every byte is legal inside a payload, and decoding stays linear because nothing is ever scanned twice.

Real-World Example

Network protocols frame messages the same way. HTTP sends Content-Length before the body, so the parser knows where one message stops and the next begins without hunting for a delimiter that the body might contain.

The Code

def encode(parts):
    return "".join(f"{len(p)}#{p}" for p in parts)

def decode(s):
    out, i = [], 0
    while i < len(s):
        j = s.index("#", i)              # the length ends at the first #
        n = int(s[i:j])
        out.append(s[j + 1 : j + 1 + n])
        i = j + 1 + n                    # jump the whole payload
    return out

print(encode(["hi", "a#b", ""]))   # 2#hi3#a#b0#
print(decode("2#hi3#a#b0#"))       # ['hi', 'a#b', '']

print(len(encode(["ab", "c"])))