/*
This is the main function of the AI, and works along side with a MinimaxComp object, in order
to recursively discover the best next move
Input: GameBoard &gb: A reference to a GameBoard. The GameBoard will not be changed at all, it will simply
be used as a way to create new GameBoards using the copy constructor
Output: a boolean value representing whether or not the move was successful (should always be successful
except in the position of a cat's game).
*/
bool FourLineComputer::move(GameBoard &gb){
std::string board = gb.getBoard();
// structure to hold time value at the beginning of the function call
struct tm start;
// acquire the initial time at the beginning of the function call
time_t t = time(0);
localtime_s(&start, &t);
// save the seconds of the current time in an integer
// this will be used to help determine when the function should return
int startSeconds = start.tm_sec;
// an integer representing the iterative depth of the current search for an optimal move
int iDepth = 1;
// the position for the "best" move
// it is initialize to the middle of the board, in the event that the computer goes first
int bestPosition = (BOARDSIZE / 2) + (BOARDWIDTH / 2);
// current best is sent to lowest char value
char currentBest = '\0';
// this integer will hold the difference between the end of each iteration and
// beginning of the entire function call
int difference = 0;
// structure to hold time value at the end of each iteration
struct tm end;
// create a new MinimaxComp
MinimaxComp newMC;
// while the difference has not exceeded the time limit
while (difference < this->timeLimit && iDepth < 4){
currentBest = '\0';
// iterate through the entire board
for (int i = 0; i < BOARDSIZE; i++){
// if the current board position is a heuristically valid move
if (board[i] < '9' && board[i] != '\0'){
// create a copy of the current GameBoard
GameBoard newGB(gb);
// make the potential move
newGB.place('X', (i % BOARDWIDTH), (i / BOARDWIDTH));
// use the MinimaxComp object to discover a better heuristic value for the move
char treeBest = newMC.determineValue(currentBest, iDepth, newGB);
// if the new value is better than the current best
if (treeBest > currentBest){
// save it as current best value of the move
currentBest = treeBest;
// also, save the position in order to perform the move later
bestPosition = i;
}
}
}
// if the move will result in a win, simply break the loop and perform that move
if (currentBest == '5'){
break;
}
// increase iterative depth
iDepth++;
// update time difference
t = time(0);
localtime_s(&end, &t);
difference = abs(start.tm_sec - end.tm_sec) % 10;
}
// attemp to make the move and return the success value of the attempt
return gb.place('X', (bestPosition % BOARDWIDTH), (bestPosition / BOARDWIDTH));
}
