[simias]

I have a similar memory from my early days of coding.

I was implementing a toy rc4 cipher. One of the steps in the ciphers involves swapping entries in an arrays of 256 elements. I thought "hey, I'm a 1337 coder, I'm going to use the xor trick".

Except it doesn't work if you're trying to swap something with itself. If you do "a[i] ^= a[j]" and i == j then you're just clearing the entry.

Taught me the valuable lesson that I shouldn't try to be a smartass while writing code and the importance of unit tests.

[taejo]

This was the basis of an excellent entry [0] in the 2007 Underhanded C Contest. It had a correct implementation of the RC4 encryption algorithm, except it used the XOR swap, so on average one byte of the pseudorandom state was zeroed every 256 iterations. Eventually, the state is all zeroes, and the encryption just outputs pure plaintext. Best of all, the first few kilobytes of output looks random at first glance.

[0] http://www.underhanded-c.org/_page_id_16.html

[cyberferret]

Nice catch - I hadn't though about in-place swap situations. I have always used this trick to switch two variables, but yes, I can see when you are talking about lists and matrix array manipulation, you can easily encounter an edge case which necessitates an 'in place' swap which would fail under these circumstances.