Hi All,

I wrote the following solution to the POUR1 problem in the C#:

using System;
public class Test
{
    public static int a, b, c;
    public static int gcd( int x, int y )
{
    if (y == 0)
        return x;

    return gcd(y, x % y);
}

public static int pour( int aa, int bb )
{
    int steps = 0;
    int in_a = 0;
    int in_b = 0;
    int amount = 0;

    while (in_a != c && in_b != c)
    {
        if (in_a == 0)
        {
            in_a = aa;
            steps++;
        }
        else if (in_b == bb)
        {
            in_b = 0;
            steps++;
        }
        else
        {
            amount = Math.Min(in_a, bb - in_b);
            in_a -= amount;
            in_b += amount;
            steps++;
        }
    }

    return steps;
}

public static int compute()
{
    if (c > a && c > b)
        return -1;

    if (c == 0)
        return 0;

    if (c % gcd(a, b) != 0)
        return -1;

    return Math.Min(pour(a, b), pour(b, a));
}

public static void Main( string[] args )
{
    int t;
    t = int.Parse(Console.ReadLine());
    while (t > 0)
    {
        a = int.Parse(Console.ReadLine());
        b = int.Parse(Console.ReadLine());
        c = int.Parse(Console.ReadLine());
        Console.WriteLine(Convert.ToString(compute()));
        t--;
    }
}
}

I tested that for several sets of data, and even wrote BFS version and compared results for all 0<=a,b,c<=20. Everything worked well, but the SPOJ was returning the WA to that solution.

After few days of looking for the root cause of the problem I gave up and rewritten that to the plain C:

#include <stdio.h>
int a, b, c;
int gcd( int x, int y )
{
    if( y == 0 )
        return x;
    return gcd( y , x % y );
}
int pour( int aa, int bb )
{
    int steps = 0;
    int in_a = 0;
    int in_b = 0;
    int amount = 0;
    while( in_a != c && in_b != c )
{
    if( in_a == 0 )
    {
        in_a = aa;
        steps++;
    }
    else if( in_b == bb )
    {
        in_b = 0;
        steps++;
    }
    else
    {
        amount = in_a < bb - in_b ? in_a : bb - in_b;
        in_a -= amount;
        in_b += amount;
        steps++;
    }
}

return steps;
}
int compute()
{
    int result0, result1;
    if( c > a && c > b )
    return -1;

if( c == 0 )
    return 0;

if( c % gcd(a,b) != 0 )
    return -1;

result0 = pour(a,b);
result1 = pour(b,a);

return result0 < result1 ? result0 : result1;
}
int main( int argc, char *argv[] )
{
    int t;
    scanf( "%d" , &t );

while( t > 0 )
{
    scanf( "%d" , &a );
    scanf( "%d" , &b );
    scanf( "%d" , &c );

    printf( "%d\n" , compute() );

    t--;
}
}

Code is almost the same, the differences are dictated by the language. However, the C one got the AC immediately.

I am suspecting some strange behaviour of the C# compiler being used on SPOJ. Has anyone encountered similar situation? I am mainly resolving problems in Python with fallback to the C#, if there are performance issues. But the C# does not seem to be reliable to me any more...

Best regards,
--
Marcin Gajda