#C + x86 assembly, 636

C + x86 assembly, 636

#C + x86 assembly, 1238

So if LibreOffice isn't lying my code consists of 1248 characters (without spaces, should I count them?) and with all the other conditions met I get a -10 for the incresing/decreasing choice.

#C + x86 assembly, 636

Here's the obfuscated version (which is unreadable but works fine):

http://pastebin.com/XkYt9DLy (C code) http://pastebin.com/h0S0dfeU (x86 code)

So if LibreOffice isn't lying my obfuscated code consists of 646 characters (without spaces, should I count them?) and with all the other conditions met I get a -10 for the incresing/decreasing choice.

I thought I'd do this without using any arrays or strings, and then recursion came into mind, but of course, with recursion I wouldn't be able to swap values from other recursive calls... and then was when I realised that I could. Linking my C program with an assembly function I could jump up in the stack and return a pointer to the base pointer of the desired call, that's what "recursionStackAt" function does.

I thought I'd do this without using any arrays or strings, and then recursion came into mind, but of course, with recursion I wouldn't be able to swap values from other recursive calls... and then was when I realized that there was a way. Linking my C program with an assembly function I could jump up in the stack and return a pointer to the base pointer of the desired call, that's what "recursionStackAt" function does. Of course recursionStackAt is a very ugly function, its result doesn't only depend on the input or program's state but on the caller itself. Note that that's what made me change the indexes from 0 based to 1 based.