To achieve that, we'll need to import the Apache Commons Math3 library that contains a utility class to compute Chi-squares: Now, we're going to calculate the Chi-squares of deciphered message letters distribution and standard English letters distribution. Let's say we want to use 36 as offset, which is equivalent to 10, the modulo operation ensures that the transformation will give the same result. That operation also takes care of larger offsets. It works as expected, thanks to the modulo operation. isEqualTo('ro dyvn wo s myevn xofob dokmr k vvkwk dy nbsfo') Thus, let's try again with an offset of 10 so that some letters will be mapped to letters at the beginning of the alphabet, like t which will be mapped to d: String cipheredMessage = cipher.cipher('he told me i could never teach a llama to drive', 10) Now, this particular example has the specificity not to exceed the letter z during the transformation, therefore not having to go back to the start of the alphabet. isEqualTo('kh wrog ph l frxog qhyhu whdfk d oodpd wr gulyh') Īs we can see, the ciphered message respects the matching defined earlier for an offset of 3. String cipheredMessage = cipher.cipher('he told me i could never teach a llama to drive', 3) Now, let's try this implementation on the message “he told me i could never teach a llama to drive” with an offset of 3: CaesarCipher cipher = new CaesarCipher() And finally, we retrieve the new character by adding the new position to the ASCII code of letter a. First, we compute the position of the current letter in the alphabet, and for that, we take its ASCII code and subtract the ASCII code of letter a from it. Then we apply the offset to this position, carefully using the modulo to remain in the alphabet range.

Shifted code decoder