When swapping tiles, the value is set to zero

Question

Why does my code to swap tiles give the value zero to board[d-1][d-3] and board[d-1][d-2]?

printf("give me\n");
int d = GetInt();

int board[d][d];
int k = d*d-1;

for(int i = 0; i < d; i++)
{
    for (int j = 0; j < d; ++j)
    {
        board[i][j] = k;
        int t = board[d-1][d-2];
        board[d-1][d-2] = board[d-1][d-3];
        board[d-1][d-3] = t;
        printf(" %2i ", board[i][j]);
        --k;
    }     
        printf("\n");
    } 
}

Answer 1

When you do:

for(int i = 0; i < d; i++)
{
    for (int j = 0; j < d; ++j)
    {
        ...
        /* Swapping ALWAYS the same tiles, since 'd' is a fixed value */
        int t = board[d-1][d-2];
        board[d-1][d-2] = board[d-1][d-3];
        board[d-1][d-3] = t;
        ...
    }     
        printf("\n");
    } 
}

The general idea is close to right (for the "swapping tiles" algorithm): grabbing a certain tile, store it in a temp value, and then swap two contiguous values. The problem is that, never mind the for loop, you're always accesing the same positions in the board[][] array, because you're accesing it with d, which is a fixed value.

Besides, you're actually performing these two steps within the same loop:

• Filling the board with tiles
• Swapping values to be moved

Of course, first times through the loop you're trying to swap values that have not been initialized yet. Luckily enough, you get zero, but in theory you could get any value at all (remember those ? in the lectures?)

You should consider splitting those steps in two different functions, like:

• init_board()
• move()

I.e.: your board should be entirely initialized before your first attempt to move a tile.

Perhaps you should start by laying out your algorithm in pseudocode, and better understand how the program will work, and only then translate that to actual 'C' code.