Convert NSString to C string, increment and come back to NSString - objective-c

Question

I'm trying to develop a simple application where i can encrypt a message. The algorithm is Caesar's algorithm and for example, for 'Hello World' it prints 'KHOOR ZRUOG' if the increment is 3 (standard).
My problem is how to take each single character and increment it...
I've tried this:
NSString *text = #"hello";
int q, increment = 3;
NSString *string;
for (q = 0; q < [text length]; q++) {
string = [text substringWithRange:NSMakeRange(q, 1)];
const char *c = [string UTF8String] + increment;
NSLog(#"%#", [NSString stringWithUTF8String:c]);
}
very simple but it doesn't work.. My theory was: take each single character, transform into c string and increment it, then return to NSString and print it, but xcode print nothing, also if i print the char 'c' i can't see the result in console. Where is the problem?

Answer 1

First of all, incrementing byte by byte only works for ASCII strings. If you use UTF-8, you will get garbage for glyphs that have multi-byte representations.
With that in mind, this should work (and work faster than characterAtIndex: and similar methods):
NSString *foo = #"FOOBAR";
int increment = 3;
NSUInteger bufferSize = [foo length] + 1;
char *buffer = (char *)calloc(bufferSize, sizeof(char));
if ([foo getCString:buffer maxLength:bufferSize encoding:NSASCIIStringEncoding]) {
int bufferLen = strlen(buffer);
for (int i = 0; i < bufferLen; i++) {
buffer[i] += increment;
if (buffer[i] > 'Z') {
buffer[i] -= 26;
}
}
NSString *encoded = [NSString stringWithCString:buffer
encoding:NSASCIIStringEncoding];
}
free(buffer);

Answer 2

first of all replace your code with this:
for (q = 0; q < [text length]; q++) {
string = [text substringWithRange:NSMakeRange(q, 1)];
const char *c = [string UTF8String];
NSLog(#"Addr: 0x%X", c);
c = c + increment;
NSLog(#"Addr: 0x%X", c);
NSLog(#"%#", [NSString stringWithUTF8String:c]);
}
Now you can figure out your problem. const char *c is a pointer. A pointer saves a memory address.
When I run this code the first log output is this:
Addr: 0x711DD10
that means the char 'h' from the NSString named string with the value #"h" is saved at address 0x711DD10 in memory.
Now we increment this address by 3. Next output is this:
Addr: 0x711DD13
In my case at this address there is a '0x00'. But it doesn't matter what is actually there because a 'k' won't be there (unless you are very lucky).
If you are happy there is a 0x00 too. Because then nothing bad will happen. If you are unlucky there is something else. If there is something other than 0x00 (or the string delimiter or "end of string") NSString will try to convert it. It might crash while trying this, or it might open a huge security hole.
so instead of manipulating pointers you have to manipulate the values where they point to.
You can do this like this:
for (q = 0; q < [text length]; q++) {
string = [text substringWithRange:NSMakeRange(q, 1)];
const char *c = [string UTF8String]; // get the pointer
char character = *c; // get the character from this pointer address
character = character + 3; // add 3 to the letter
char cString[2] = {0, 0}; // create a cstring with length of 1. The second char is \0, the delimiter (the "end marker") of the string
cString[0] = character; // assign our changed character to the first character of the cstring
NSLog(#"%#", [NSString stringWithUTF8String:cString]); // profit...
}