Reversing alphabet value of individual characters in string by calculating the difference

Question

I'm trying to write a little function that will flip lowercase characters to their symmetric counterparts in the second half of the alphabet - 26 letters = 13/13.

a = z, b = y, c = x...

I've tried the following code but for some reason it works only for the first character.

Say I enter "bamba"; it begins by switching the 'b' to 'y' but then it gets stuck and replaces all the other character to 'y' as well and I get "yyyyy".

I tried playing around with the code a bit and discovered that if I remove dependency by the current character, I can safely increase all the letters by, say, 1 (a = b, b = c...)

symmetric_difference = 1; **commented out** //21 - toCrypt[i];

I looked all over and the closest thing I found was "Reversing alphabet value of individual characters in string" but it describes a way that seems weird and redundant.

Can anyone tell me what I did wrong please (assuming that I did)?

#include <iostream>
using namespace std;

void crypto(char[]);

int  main()
{
    char toCrypt[80];

    cout << "enter a string:\n";
    cin >> toCrypt;

    crypto(toCrypt);

    cout << "after crypto:\n";
    cout << toCrypt;
}

void crypto(char toCrypt[]) // "Folding" encryption.
{
    int size = strlen(toCrypt);
    int symmetric_difference;

    for (int i = 0; i < size; i++)
    {
        symmetric_difference = 121 - toCrypt[i];    // Calculate the difference the letter has from it's symmetric counterpart.

        if (toCrypt[i] >= 97 && toCrypt[i] <= 110)  // If the letter is in the lower half on the alphabet,
            toCrypt[i] += symmetric_difference; // Increase it by the difference.
        else
        if (toCrypt[i] >= 111 && toCrypt[i] <= 122) // If it's in the upper half,
            toCrypt[i] -= symmetric_difference; // decrease it by the difference.
    }
}

Answer 1

You can try this

for (int i = 0; i < size; i++)
{
   toCrypt[i] = 'z' - toCrypt[i] + 'a';
}

Answer 2

In your example, bamba, all of the characters go into the first if statement: toCrypt[i] += symmetric_difference;.

toCrypt[i] += symmetric_difference;
-> toCrypt[i] = toCrypt[i] + 121 - toCrypt[i];
-> toCrypt[i] = 121 = 'y'

Answer 3

Try the following function definition if I did not make a typo.

void crypto( char s[] )
{
    static const char alpha[] = "abcdefghijklmnopqrstuvwxyz"; 
    const char *last = alpha + sizeof( alpha ) - 1;

    while ( char &c = *s++ )
    {
      if ( const char *first = std::strchr( alpha, c ) ) c = *( last - ( first - alpha ) - 1 );
    }
}   

Take into account that it is not necessary that low case letters are ordered sequantially. For example if I am not mistaken it is not valid for EBCDIC.

I would like to substitute statement

const char *last = alpha + sizeof( alpha ) - 1;

for

const char *last = alpha + sizeof( alpha ) - sizeof( '\0' );

but the last is not compatible with C.:)