62 lines
2.4 KiB
Text
62 lines
2.4 KiB
Text
var[const] uppercase=["A".."Z"].pump(String),
|
|
english_frequences=T( // A..Z
|
|
0.08167, 0.01492, 0.02782, 0.04253, 0.12702, 0.02228, 0.02015,
|
|
0.06094, 0.06966, 0.00153, 0.00772, 0.04025, 0.02406, 0.06749,
|
|
0.07507, 0.01929, 0.00095, 0.05987, 0.06327, 0.09056, 0.02758,
|
|
0.00978, 0.02360, 0.00150, 0.01974, 0.00074);
|
|
|
|
fcn vigenere_decrypt(target_freqs, input){ // ( (float,...), string)
|
|
nchars,ordA :=uppercase.len(),"A".toAsc();
|
|
sorted_targets:=target_freqs.sort();
|
|
|
|
frequency:='wrap(input){ // (n,n,n,n,...), n is ASCII index ("A"==65)
|
|
result:=uppercase.pump(List(),List.fp1(0)); // ( ("A",0),("B",0) ...)
|
|
foreach c in (input){ result[c - ordA][1] += 1 }
|
|
result // --> mutable list of mutable lists ( ("A",Int)...("Z",Int) )
|
|
};
|
|
correlation:='wrap(input){ // (n,n,n,n,...), n is ASCII index ("A"==65)
|
|
result,freq:=0.0, frequency(input);
|
|
freq.sort(fcn([(_,a)],[(_,b)]){ a<b }); // sort letters by frequency
|
|
foreach i,f in (freq.enumerate()){ result+=sorted_targets[i]*f[1] }
|
|
result // -->Float
|
|
};
|
|
|
|
cleaned:=input.toUpper().pump(List,uppercase.holds,Void.Filter,"toAsc");
|
|
|
|
best_len,best_corr := 0,-100.0;
|
|
# Assume that if there are less than 20 characters
|
|
# per column, the key's too long to guess
|
|
foreach i in ([2..cleaned.len()/20]){
|
|
pieces:=(i).pump(List,List.copy); // ( (),() ... )
|
|
foreach c in (cleaned){ pieces[__cWalker.idx%i].append(c) }
|
|
|
|
# The correlation seems to increase for smaller
|
|
# pieces/longer keys, so weigh against them a little
|
|
corr:=-0.5*i + pieces.apply(correlation).sum(0.0);
|
|
if(corr>best_corr) best_len,best_corr=i,corr;
|
|
}
|
|
if(best_len==0) return("Text is too short to analyze", "");
|
|
|
|
pieces:=best_len.pump(List,List.copy);
|
|
foreach c in (cleaned){ pieces[__cWalker.idx%best_len].append(c) }
|
|
|
|
key,freqs := "",pieces.apply(frequency);
|
|
foreach fr in (freqs){
|
|
fr.sort(fcn([(_,a)],[(_,b)]){ a>b }); // reverse sort by freq
|
|
m,max_corr := 0,0.0;
|
|
foreach j in (nchars){
|
|
corr,c := 0.0,ordA + j;
|
|
foreach frc in (fr){
|
|
d:=(frc[0].toAsc() - c + nchars) % nchars;
|
|
corr+=target_freqs[d]*frc[1];
|
|
if(corr>max_corr) m,max_corr=j,corr;
|
|
}
|
|
}
|
|
key+=(m + ordA).toChar();
|
|
}
|
|
|
|
cleaned.enumerate().apply('wrap([(i,c])){
|
|
( (c - (key[i%best_len]).toAsc() + nchars)%nchars + ordA ).toChar()
|
|
}).concat() :
|
|
T(key,_);
|
|
}
|