//For this exercise I used my Caesar cipher script with modifications as described in the comments. The actual (de)ciphering algorithm is identical. //This is the dataset from Wikipedia, formatted in a way that made it super fun to parse letters=('"a" 8.167 1 "b" 1.492 2 "c" 2.782 3 "d" 4.253 4 "e" 12.702 5 "f" 2.228 6 "g" 2.015 7 "h" 6.094 8 "i" 6.966 9 "j" 0.153 10 "k" 0.772 11 "l" 4.025 11 "m" 2.406 13 "n" 6.749 14 "o" 7.507 15 "p" 1.929 16 "q" 0.095 17 "r" 5.987 18 "s" 6.327 19 "t" 9.056 20 "u" 2.758 21 "v" 0.978 21 "w" 2.360 23 "x" 0.150 24 "x" 0.150 24 "y" 1.974 25 "z" 0.074 26').trim().replace(/\s+/g, " ").split(" "); letterfrequency=new Array(26); for (var i=0;i<26;i++) { letterfrequency[i]=letters[3*i+1]; // finally a usable array from the above data! } const small="abcdefghijklmnopqrstuvwxyz"; const big=small.toUpperCase(); function caesarAutoDecrypt(){ var input = document.getElementById('input').value; numcipher=findkey(input); output=""; output+="

Decrypted message:

"; output+="

"+(ciphered=doCipher(-numcipher,input))+"

"; output+="

Encrypted message:

"; output+="

"+input+"

"; output+="

Caesar cipher encryption key:

"+numcipher+"

"; var result=document.getElementById("result"); result.innerHTML=output; return false; } function findkey(stringinput){ lettercount=new Array(26); for(let shift=0;shift<26;shift++) lettercount[shift]=0.0; lowerstring=stringinput.toLowerCase(); for(let i=0;i=0) lettercount[test]+=1.0/lowerstring.length; } var chisq=Array(26); for(let shift=0;shift<26;shift++){ chisq[shift]=0.0; for(let i=0;i<26;i++) chisq[shift]+=(lettercount[(i+shift)%26]-letterfrequency[i])**2; } // chisq or chi squared is a statistical quantity that gives the difference between two datasets, in this case the letter frequency of the input string (shifted using the Caesar cipher) and the average letter frequency for English text. It is proportional to the standard deviation. return chisq.indexOf(Math.min(...chisq)); // For the most likely shift,just find the shift where chi squared is a minimum, i.e. where the letter frequencies in the text most closely match the expected values. } function doCipher(cipher,stringtocipher){ while (cipher<0) cipher+=26; cipher%=26; var output=""; for(let i=0;i=0) output+=small[(charIndex+cipher)%26]; else if ((charIndex = big.indexOf(stringtocipher[i]))>=0) output+=big[(charIndex+cipher)%26]; else output+=stringtocipher[i]; } return output; }

caesar auto decrypt