Thread: A noob with a question about a transposition cipher

Code:

while(plaintext2[i] != '\0'){i++;stringlength++;}

Bad. Use strlen(), or better yet, use std::string for all string-related stuff.

Code:

char flipcipher[stringlength+1];

That's not even valid C++. It's a GCC extension to allow this C99 code under C++.

The typical way to generate numbers in a range without duplicates, in random order, is to first generate the numbers in, say, ascending order (makes it easy to avoid dupes) and then shuffle the result. std::shuffle() can shuffle for you.

Why are you calculating the length of the string by scanning for a zero? It is not guaranteed that a C++ std::string actually has a zero at the end, and even if it does, string::size() would be much faster.

You may also want to ENFORCE the "don't repeat yourself".

And at the end, you are again relying on the undefined behaviour of "string" don't always end with a zero.

Edit: Doh! strings are char arrays, not C++ strings, so ignore that bit.

--
Mats

You should learn about the std::string container.

Code:

cout<<"Please enter the numbers 0 - "<<stringlength-1<<" In a random order! Don't repeat yourself!"<<endl;

Ugh, so if I type in a message that's 250 characters in length then I need to individually type in the values from 0 to 249 in a random order and remember which ones I've already typed so I don't repeat them? I don't have that much patience. Yeah, that does need work.

Code:

cout<<"Ok now we have a permutor, let put it to use!"<<endl;
while(plaintext2[i] != '\0'){
    flipcipher[i] = plaintext2[permuter[i]];
    i++;}
flipcipher[i+1]  = '\0';

When you reach the end of your loop to fill the flipcipher array, your i value is already correct to place the null. You shouldn't need to add 1 to it.

Code:

i = 0;

cout<<"Ok now we have flipped the plaintext, let's look at it: "<<endl;
	
while(flipcipher[i] !='\0'){
    cout<<flipcipher[i];
    i++;}

You do realize you can skip this loop and just say:

Code:

cout << flipcipher;

Originally Posted by DocFerret

My question is this: As you can see my method for generating a "permuter" (I suppose the correct term is a key) is far from elegant, I tried to incorporate a check for duplicate numbers but it only worked some of the time, and even the length check is sort of dodge, so I am looking for a method to generate a key in a way that is half-way random. I can't just generate a string of random numbers since the cipher requires it to be a permutation of the number string from 0 to stringlength, so I was thinking is there any relatively simple way to generate an arbitrary permutation of said string?

Yes you can do this easier. You can initialize an array of size stringlength with the values from 0-stringlength (-1). You can then call the STL random_shuffle function which will randomize the values in this array so you will in effect randomly generate the key.

Originally Posted by CornedBee

Code:

while(plaintext2[i] != '\0'){i++;stringlength++;}

Bad. Use strlen(), or better yet, use std::string for all string-related stuff.

Code:

char flipcipher[stringlength+1];

That's not even valid C++. It's a GCC extension to allow this C99 code under C++.

The typical way to generate numbers in a range without duplicates, in random order, is to first generate the numbers in, say, ascending order (makes it easy to avoid dupes) and then shuffle the result. std::shuffle() can shuffle for you.

I have no idea what C99 even is to be honest so I just used it and it worked so it's fine by me. Of course I realize it might not be fine in the long run, but to be honest at my current skill level I don't even know why it's wrong

I've never used strlen or shuffle, how are they called and what libs are they in? I assume they work on char arrays as well.

Another thing, I started out by saying my method was FAR from elegant, in fact it outright sucks, I'm well aware of this and that is why I'm posting here so I can learn smarter ways of doing things like that

I do apprciate the comments though so keep them comming.

It's wrong because it's not part of C++. Array bounds must be compile-time constants.

strlen() is used with legacy C strings. Use std::string, from the <string> header. std::random_shuffle is in <algorithm>.

No and no. You should read a proper tutorial on arrays, and more importantly, you should forget all about C strings and learn C++ strings.

random_shuffle works like this:

Code:

#include <algorithm>

...

int array[10] = {0,1,2,3,4,5,6,7,8,9};

// Randomly shuffle the contents of "array"
std::random_shuffle(array,array+10);

And yes, it works with characters as well:

Code:

char alphabet[26] = "abcdefghijklmnopqrstuvwxyz";

std::random_shuffle(alphabet,alphabet+26);

And other containers like std::string:

Code:

std::string alphabet("abcdefghijklmnopqrstuvwxyz");

std::random_shuffle(alphabet.begin(),alphabet.end());

Depending on implementation you may need a call to srand prior to using it to truly make it random.

You mean you learned from a book that teaches char arrays as string but doesn't even show you how to use strlen()?

Please, throw it away. Or better, throw it at the head of the author.