import prisoner0, prisoner1

payoff_cc = [2, 2]
payoff_cd = [-1, 3]
payoff_dc = [3, -1]
payoff_dd = [0, 0]

moves = [ [], [] ]
scores = [ [], [] ]
totals = [0, 0]

def score(m0, m1):
    global totals
    moves[0].append(m0)
    moves[1].append(m1)
    if m0 == 'cooperate' and m1 == 'cooperate':
        scores[0].append(payoff_cc[0])
        scores[1].append(payoff_cc[1])
    elif m0 == 'cooperate' and m1 == 'defect':
        scores[0].append(payoff_cd[0])
        scores[1].append(payoff_cd[1])
    elif m0 == 'defect' and m1 == 'cooperate':
        scores[0].append(payoff_dc[0])
        scores[1].append(payoff_dc[1])
    else:
        scores[0].append(payoff_dd[0])
        scores[1].append(payoff_dd[1])

    totals[0] += scores[0][-1]
    totals[1] += scores[1][-1]


def play_iterations(num_iterations, prisoner0_strategy, prisoner1_strategy, display=False):
    for i in range(2):
        moves[i] = []
        scores[i] = []
        totals[i] = 0

    for i in range(num_iterations):
        m0 = prisoner0.play(prisoner0_strategy, moves[0], moves[1])
        m1 = prisoner1.play(prisoner1_strategy, moves[1], moves[0])
        score(m0, m1)
        if display: 
            s = 'm0={0:10} m1={1:10}  scores=({2:2},{3:2})  totals=({4},{5})'
            print(s.format(m0, m1, scores[0][-1], scores[1][-1], totals[0], totals[1]))
        prisoner0.update(prisoner0_strategy, moves[0], moves[1], scores[0][-1])
        prisoner1.update(prisoner1_strategy, moves[1], moves[0], scores[1][-1])
    if display:
        s = 'Average per iteration scores: ({0:2.2f}, {1:2.2f})'
        print(s.format(totals[0]/num_iterations, totals[1]/num_iterations))

def evaluate(num_iterations, num_runs, prisoner0_strategy, prisoner1_strategy):
    overall_totals = [0, 0]
    for r in range(num_runs):
        play_iterations(num_iterations, prisoner0_strategy, prisoner1_strategy, False)
        overall_totals[0] += totals[0]
        overall_totals[1] += totals[1]
    # Now averages.
    avg0 = (overall_totals[0] / num_iterations) / num_runs
    avg1 = (overall_totals[1] / num_iterations) / num_runs
    s = '{0} vs. {1}: average scores over {2} iterations: ({3:6.2f}, {4:6.2f})'
    print(s.format(prisoner0_strategy, prisoner1_strategy, num_iterations, avg0, avg1))
    return (avg0, avg1)