Single character mismatch in Caesar Cipher: Python

Question

Essentially, the alphabets in the input string are cyclically rotated by k and the rest of the characters retains their form. I'm puzzled by a mismatch. For my code:

import math
import os
import random
import re
import sys

# Complete the caesarCipher function below.
def caesarCipher(s, k):
    st = []
    for i in range(0,len(s)):
        if 65<=ord(s[i])<=90: 
            temp = ord(s[i])+k
            if (temp>90):
                temp-=26
            st.append(chr(temp))
        elif 97<=ord(s[i])<=122:
            temp = ord(s[i])+k
            if (temp>122):
                temp-=26
            st.append(chr(temp))   
        else: st.append(s[i])
    return ''.join(st)




if __name__ == '__main__':

    s = input()

    k = int(input())

    result = caesarCipher(s, k)

The output always fails by one, like My: okffng-Qwvc Expected: okffng-Qwvb

My:Fqcfex-Qttp-ts-ymj-Gwnlmy-Xnij-tk-Qnkj Expected: Fqbfdx-Qttp-ts-ymj-Gwnlmy-Xnij-tk-Qnkj

I can't seem to detect the bug.

What was I thinking? Omg. Nevertheless, it still fails some test cases. So, the bug is still there. — srkdb, Sep 18 '18 at 01:55
There are 26 letters in the alphabet, *so* you need to take 26 off instead of 25 if your value goes above 90 — strava, Sep 18 '18 at 01:59
Why are you _only_ subtracting 26, though? What if I ask you to rotate 'a' by 26*2=52, or 26*100? You obviously want to get back to ord('a')=97, but you're only subtracting 26.. — DSM, Sep 18 '18 at 02:08

score 2 · Accepted Answer · answered Sep 18 '18 at 02:33

The typical presentation of a Caesar Cipher uses a modulus operator. I'd probably start there in debugging your code.

I find code easier to debug if it is flat rather than nested. If each function does one small thing, it's easier to completely test that one small function and fit them together to have the correct behavior. Take the following example:

def _caesarLetter(c, k, offset):
    return chr((ord(c)-offset+k)%26+offset)

def caesarLetter(c, k):
    if not c.alpha():
        return c
    return _caesarLetter(c, k, ord('a') if c.islower() else ord('A'))

def caesarCipher(s, k):
    return ''.join(caesarLetter(c, k) for c in s)

The first method _caesarLetter applies the cipher to a single letter, and only for a fixed character encoding. It basically takes the whole alphabet and translates it to start at 0, it applies the cipher, and it moves the alphabet back to where it started. For lowercase ascii this is at 97, and for uppercase it is at 65.
The second method caesarLetter takes care of figuring out where lowercase and uppercase alphabets start in the ascii encoding. Rather than explicitly encoding 65 and 97, which is prone to typos and unclear errors, we just compute them on the fly. CPU time is cheap. Developer time is expensive. This makes the intent clear.
The third method caesarCipher just does that for all the letters in your string and joins them together.

Single character mismatch in Caesar Cipher: Python

1 Answers1