def findMove():
# Decide
INFO['game'] = c4.Connect4(board=INFO['game'].state)
algorithm = c4.Alpha_Beta_Pruning.Minimax([INFO['game'].nodes, INFO['game'].edges], False)
ai_move = int(algorithm.path[1][-1])
for move in generate_moves(game.state):
temp_game_state = deepcopy(game.state)
temp_state = modify_state(temp_game_state, move, game.ai)
result = set()
for row in temp_state:
if check_plausibility(row):
result.add(check_end(indices_of_duplicates(row, game.ai), game.ai))
for column in temp_state.transpose():
if check_plausibility(column):
result.add(check_end(indices_of_duplicates(column, game.ai), game.ai))
for diagonal in get_diagonals(temp_state):
if check_plausibility(diagonal):
result.add(check_end(indices_of_duplicates(diagonal, game.ai), game.ai))
if (constant in result) or (-constant in result) or (len(result) > 0 and generate_moves(temp_state) is None):
ai_move = move
break
# Apply
INFO['game'].state = c4.modify_state(INFO['game'].state, ai_move, INFO['game'].ai)
def tryMove(human_move):
# retrieve possible moves
INFO['possible-moves'] = c4.generate_moves(INFO['game'].state)
# make sure move is valid
if not (0 <= human_move <= 6): return 'invalid'
elif INFO['possible-moves'] == []: return 'tie'
elif human_move not in INFO['possible-moves']: return 'invalid'
# move is valid, so apply it to state, and indicate that move was successful
INFO['game'].state = c4.modify_state(INFO['game'].state, human_move, INFO['game'].human)
return True
def put(self): # initialize game
INFO['state'] = {'yellow':21, 'red':21, 'turn': 'yellow', 'ai-chance':0, 'game-over': False, 'cursor': 3}
INFO['game'] = c4.startGame('alpha beta pruning')
INFO['possible-moves'] = []
INFO['human-moves'] = []
INFO['ai-moves'] = []
return '', RUN
def post(self, move): # make move
if not INFO['state']['game-over']:
# AI
if move == 9:
if INFO['state']['turn'] == 'red':
# Do turn
findMove()
endTurn()
# Check if won
if INFO['game'].evaluate(INFO['game'].state, INFO['game'].ai, c4.generate_moves(INFO['game'].state) == 0) == c4.constant:
return convert_grid(INFO['game'].state).tolist(), WIN
# Update info
INFO['human-moves'] = INFO['game'].evaluate(INFO['game'].state, INFO['game'].human, INFO['possible-moves'] == [])
INFO['ai-moves'] = INFO['game'].evaluate(INFO['game'].state, INFO['game'].ai, c4.generate_moves(INFO['game'].state) == 0)
INFO['state']['ai-chance'] = INFO['ai-moves'] / (INFO['ai-moves'] + INFO['human-moves'])
# Send board, AI made move but did not win
return convert_grid(INFO['game'].state).tolist(), RUN
return 'PC turn', TRN
# Player
else:
if INFO['state']['turn'] == 'yellow':
if(tryMove(move)):
endTurn()
if INFO['game'].evaluate(INFO['game'].state, INFO['game'].human, c4.generate_moves(INFO['game'].state) == 0) == -c4.constant:
return convert_grid(INFO['game'].state).tolist(), WIN # human won
return convert_grid(INFO['game'].state).tolist(), RUN
else:
return 'invalid', INV
return 'AI turn', TRN
return 'Game already over', WIN
def get(self, direction):
if INFO['state']['turn'] == 'yellow':
if(tryMove(INFO['state']['cursor'])):
endTurn()
if INFO['game'].evaluate(INFO['game'].state, INFO['game'].human, c4.generate_moves(INFO['game'].state) == 0) == -c4.constant:
return convert_grid(INFO['game'].state).tolist(), WIN # human won
return convert_grid(INFO['game'].state).tolist(), RUN
else:
return 'invalid', INV
return 'AI turn', TRN
app = Flask(__name__, static_folder='../art')
api = Api(app)
# CONSTANTS
RUN = 201
TRN = 202
INV = 203
WIN = 204
PORT = 5000
# STORAGE
INFO = {
'state': {'yellow':21, 'red':21, 'turn': 'yellow', 'ai-chance':0, 'game-over': False, 'cursor': 3},
'game': c4.startGame('alpha beta pruning'),
'possible-moves': [],
'human-moves': [],
'ai-moves': []
}
# FUN(CTIONS)
# -> Paint grid
def convert_grid(state):
mapping = {0: 0, -1: 1, 1: 2}
state = np.array([[mapping[char] for char in line] for line in state])
print(len(state))
for i in range (len(state)):
for j in range (len(state[i])):
print(j)
if j == INFO['state']['cursor']: