A competition using goto

[Dratino]

Here's a little something I came up with just to be evil to structured programmers:

Devise a function a(m, n) to output a single integer, that follows the following restrictions:

1. The function must use a return variable, initialized at 0, and you may only have one statement that increments it.
2. The function may not define any other variables.
3. You may only use the functions and statements ++, --, ==, if, and goto. (If coding in a language other than C or C++, the equivalents may be used. ++ means "increase by 1", and -- means "decrease by 1".)
4. You may not have the same statement twice within the same block of code. (So, no using {++n; ++n;})

Make a function that will return the largest value for a(42, 42). My record is 904, and the source code for the solution is below:

code:

long long a(int m, int n){         long long output = 0; stop_a:                 ++output;         stop_b:         --n;         if(n == 0){                 --m;                 goto stop_c;         }         goto stop_a;         stop_c:                         if(m < 0) goto stop_d;                 ++n;         if(n < m) goto stop_c;                 goto stop_a; stop_d:                 return output; }

[Insectoid]

Um, what exactly does a(m, n) do? You seem to have forgotten to tell us that.

[Dratino]

a(m, n) can do whatever you want it to, as long as it follows those rules.

It returns a single integer.

[Tony]

Your rules don't mention anything regarding the use of integer literals; and your code does use them ( == 0), so... goto loops until 905?

[A.J]

Perhaps I am misunderstanding your rules:

c++:

#include<iostream> using namespace std; #define arbitrary 0x7FFFFFFF long long a(int m, int n) {     unsigned long long ret = 0;     a:;     ret++;     if (ret==arbitrary) return ret;     goto a; } main() {     cout << a(42, 42) << endl; }

EDIT: I apologize Tony, as I posted my code right when you mentioned that. Coincidence.

[Dan]

The contest is rather silly. Can i just have LONG_MAX number of "++"s hardcoded into my code?

Also some overflow fun:

c:

unsigned long long a(int m, int n){         unsigned long long output = 0;         return --output; }

[RandomLetters]

This sounds like a situation where brainfuck is actually useful.

http://en.wikipedia.org/wiki/Brainfuck#Multiplication

[Dratino]

Tony @ Sun Apr 17, 2011 9:10 pm wrote:

Your rules don't mention anything regarding the use of integer literals; and your code does use them ( == 0), so... goto loops until 905?

It actually doesn't.

Um. I'm going to have to modify that rule, then - another reason I put it on here is so that I could iron out the loopholes in the contest, because not just for having bugs in the program, but having bugs in the contest rules.

"No integer literals besides 0 may be used."

Dan @ Sun Apr 17, 2011 9:41 pm wrote:

The contest is rather silly. Can i just have LONG_MAX number of "++"s hardcoded into my code?

Try not to. That would probably violate rule 4.

The point is to create a function that loops through a complicated system of goto's to simulate some sort of recursion.

[Dan]

Did you look at my underflow soultion? It seems to meet all the rules. Uses a single return varaible, only statment is a single --. No goto used but your rules don't say it has to be.

[A.J]

I don't believe my solution violates any rules either.

[btiffin]

Dratino;

You want to google brainf#*k. It's a swear word name, fill in the #*, for a programming engine you may like.

It be funky in it's completeness.

Cheers

[Dratino]

Dan @ Mon Apr 18, 2011 3:43 pm wrote:

Did you look at my underflow soultion? It seems to meet all the rules. Uses a single return varaible, only statment is a single --. No goto used but your rules don't say it has to be.

Yeah, I guess. Like I said, one point of this was to iron out the loopholes. Your underflow glitch would be one of those.

@btiffin: Yeah, I've heard of Brainfuck.