Frequency analysis? Caesar cipher C - c

Question

I have written this code so far, and now i am required to use the frequency analysis to crack the code which I am not not clear of.
From what I understand, I must first count the frequency of letters in a string, then I would say compare it with the most frequent letters in German language, and later sort it with bubble sort. Is this correct?
I would really appreciate it if anyone could give me Ideas or hints on where to start. Thank you in advance.
EDITED: Hi Guys, i just edited my code and the frequency analysis seems to be working fine right now. It would help me if you guys can give comments or critics on my code. Thanks!
BTW its German language, i changed it.
#include <stdio.h>
#include <conio.h>
#include <string.h>
#include <ctype.h>
#define MAX 100
struct Data
{
char letter;
int num;
};
typedef struct Data DATA;
void encode(char message[], int shift)
{
int i;
FILE *pout;
pout = fopen("Output_encode.txt", "w");
if (pout == NULL)
{
printf("File could not be opened for writing!\n");
exit(1);
}
for(i=0;i<strlen(message);i++)
{
if (!isalpha(message[i]))
continue;
// checking for upper case
if(isupper(message[i]))
message[i]=((message[i]-'A') + shift) % 26 + 'A';
else
//checking for lower case
if(islower(message[i]))
message[i]=((message[i]-'a') + shift) % 26 + 'a';
}
printf("\n\tEncoded text: %s\n", message);
fprintf(pout, "%s\n", message);
if (fclose(pout) != 0)
printf("Error in closing file!\n");
}
void decode(char message[], int shift)
{
int i;
FILE *pout;
pout = fopen("Output_decode.txt", "w");
if (pout == NULL)
{
printf("File could not be opened for writing!\n");
exit(1);
}
for(i=0;i<strlen(message);i++)
{
if (!isalpha(message[i]))
continue;
// checking for upper case
if(isupper(message[i]))
message[i]=((message[i]-'A') + (26-shift)) % 26 + 'A';
else
//checking for lower case
if(islower(message[i]))
message[i]=((message[i]-'a') + (26-shift)) % 26 + 'a';
}
printf("\n\tDecoded text: %s\n", message);
fprintf(pout, "%s\n", message);
if (fclose(pout) != 0)
printf("Error in closing file!\n");
}
void textfile_decode()
{
FILE *pin, *pout;
char filename_in[MAX], filename_out[MAX];
char text[MAX];
char text3[MAX]={0};
char table[26] = {'e','n','i','r','s','t','a','d','h','u','l','c','g','o','m','b','f','w','k','z','p','v','j','y','x','q'}; //Frequency letters in German dictionary
DATA temptext, text2[26];
int temp, position;
int i, m, max, trial, l=0, n=0, k=0;
printf("Enter name of input file: ");
scanf("%s", filename_in);
pin = fopen(filename_in, "r");
if (pin == NULL)
{
printf("File could not be opened for reading!");
}
printf("Enter name of output file: ");
scanf("%s", filename_out);
pout = fopen(filename_out, "w");
if (pout == NULL)
{
printf("File could not be opened for writing!");
}
printf("\nOriginal Code:\n");
while(!feof(pin))
{
fgets(text, MAX, pin); //Read from textfile
fputs(text, stdout); //Show original code on console
}
printf("\n");
if (pin == NULL)
{
printf("File could not be opened for reading!");
}
fclose(pin);
pin = fopen(filename_in, "r");
for (i = 0; i <= 25; i++)
{
text2[i].letter = 'a' + i; //Array elements A-Z
text2[i].num = 0; //Number of letters (Frequency)
}
while(!feof(pin))
{
i = 0;
fgets(text, MAX, pin); //Read from textfile per line
while(text[i] != '\0')
{
if(1 == isupper(text[i])) // Replace capital letters with small
{
text[i] += 32;
}
switch(text[i]) //Counting letters (letter frequency)
{
case 'a':
text2[0].num += 1;
break;
case 'b':
text2[1].num += 1;
break;
case 'c':
text2[2].num += 1;
break;
case 'd':
text2[3].num += 1;
break;
case 'e':
text2[4].num += 1;
break;
case 'f':
text2[5].num += 1;
break;
case 'g':
text2[6].num += 1;
break;
case 'h':
text2[7].num += 1;
break;
case 'i':
text2[8].num += 1;
break;
case 'j':
text2[9].num += 1;
break;
case 'k':
text2[10].num += 1;
break;
case 'l':
text2[11].num += 1;
break;
case 'm':
text2[12].num += 1;
break;
case 'n':
text2[13] .num+= 1;
break;
case 'o':
text2[14].num += 1;
break;
case 'p':
text2[15].num += 1;
break;
case 'q':
text2[16].num += 1;
break;
case 'r':
text2[17].num += 1;
break;
case 's':
text2[18].num += 1;
break;
case 't':
text2[19].num += 1;
break;
case 'u':
text2[20].num += 1;
break;
case 'v':
text2[21].num += 1;
break;
case 'w':
text2[22].num += 1;
break;
case 'x':
text2[23].num += 1;
break;
case 'y':
text2[24].num += 1;
break;
case 'z':
text2[25].num += 1;
break;
default: break;
}
i++;
}
}
for(i = 0; i <= 26; i++) // Sorting array text2 according to letter frequency
{
temp = text2[i].num;
for(m = i+1; m <= 27; m++)
{
if(text2[m].num > temp)
{
max = m;
temp = text2[m].num;
}
}
temptext = text2[max];
text2[max] = text2[i];
text2[i] = temptext;
}
fclose(pin);
fclose(pout);
pin = fopen(filename_in, "r");
pout = fopen(filename_out, "w");
do
{
k += 1;
} while (text2[k].num == text2[k+1].num); //Check--> How many letters have the same frequency
trial = 2;
while(!feof(pin))
{
fgets(text, MAX, pin);
do
{
position = table[l] - text2[n].letter; // determine letter position
i = 0;
do
{
if(0 !=isalpha(text[i]))
{
if(0 != isupper(text[i])) // Checking for uppercase
{
text3[i] = text[i];
text3[i] = text3[i] + position;
if(text3[i] > 90) // If exceeds Alphabets, start again from 'A'
{
text3[i] = text3[i] - 26;
}
else if (text3[i] < 65)
{
text3[i] += 26;
}
}
else if (0 != islower(text[i])) // checking for lowercase
{
text3[i] = text[i];
text3[i] = text3[i] + position;
if(text3[i] > 122) // If exceeds Alphabets, start again from 'a'
{
text3[i] = text3[i] - 26;
}
else if(text3[i] < 97)
{
text3[i] += 26;
}
}
}
else
{
text3[i] = text[i]; // All other non letters are simply replaced
}
i++;
}while(text[i] != '\0' );
if (trial== 2)
{
printf("\n");
fputs(text3, stdout);
printf("\nCode decrypted? (0)NO (1)YES : ");
scanf("%d", &trial);
printf("\n");
}
if (trial == 0 && n != k) // Code not decrypted, letters have different frequency
{
n++;
trial = 2;
}
if (trial == 0 && n == k) // Code not decrypted, letters have same frequency
{
l++;
n = 0;
trial = 2;
}
if (trial == 3) // First line of code is decrypted, following lines will decrypted using same shift position
{
trial = 1;
}
}while(trial != 1);
fputs(text3, stdout); //Show on console window
fputs(text3, pout);
memset(text3,'\0',100); // Reset text3 array
memset(text,'\0',100); // Reset text array
trial = 3; // First line of code decrypted, shift position is saved
}
fclose(pin);
fclose(pout);
}
int main()
{
int shift, choice1, choice2;
char message[MAX];
do{
printf("Selection: \n");
printf("(1) Encode/Decode\n");
printf("(2) Decode Textfile\n");
printf("(3) End Programme\n");
printf("User input: ");
scanf("%d", &choice1);
fflush(stdin);
switch(choice1){
case 1:
printf("\nEnter message to be encrypted: ");
gets(message);
printf("Enter shift amount (1-25): ");
scanf("%d", &shift);
printf("\nSelection: \n");
printf("(1) Encode\n");
printf("(2) Decode\n");
printf("User input: ");
scanf("%d", &choice2);
switch(choice2)
{
case 1:
encode(message, shift);
break;
case 2:
decode(message, shift);
break;
}
break;
case 2:
textfile_decode();
break;
}
printf("\n\n");
}while(choice1!=3);
printf("\n");
return 0;
}

Answer 1

A Caesar ciphre changes characters by shifting them n places.
There are two very simple approaches to solving a shift ciphre:
Print all 25 possible solutions. Manually select the one that contains readable text.
Get the frequency of the characters (not the words). Then perform the shift that best aligns with a frequency table of the language the message was written in (English in your case?).

Answer 2

To break the code you can use 3 different approaches:
The first one is what you cited: count the frequency of words in a text ( I would rather use a Map for that, using the string as key and rising the number of hits as value.), and guessing the letters by comparing it to the frequency of words used in normal texts.
The second solution would be to do same with letters and guessing the meaning by comparing your frequency with the frequency of letters in a normal text.
The third solution would be to take single words of the text and trying all possible shiftings of the letters until you get words that mean something.
Here you can find some good sources!