import csv

SPAM_WORDS = ["money"]

num_spam = 0
num_ham = 0
num_total = 0
counter = 0

num_tp = 0
num_tn = 0
num_fp = 0
num_fn = 0

with open('emails.csv') as file:

    lines = csv.reader(file, quotechar = '"',)
    next(lines, None)

    for line in lines:
        sno, label, content, labelnum = line

        num_total = num_total + 1
        found_spam = []



        for word in SPAM_WORDS:
            if word in content:
                num_spam = num_spam + 1
                found_spam.append(word)
                predicted_label = 'spam'
            else:
                num_ham = num_ham + 1
                predicted_label = 'ham'


        print('\n\n\nEmail #:', num_total)
        print(f'\nEmail Sno: {sno} | Actual Label: {label}')
        print(f'\n\nContent: {content}')

        if len(found_spam):
            print('\n -> Found SPAM words:', ' | '.join(found_spam))

        if label == 'spam' and predicted_label == 'spam':
            num_tp = num_tp + 1
        elif label == 'ham' and predicted_label == 'ham':
            num_tn = num_tn + 1
        elif label == 'spam' and predicted_label == 'ham':
            num_fn = num_fn + 1
        elif label == 'ham' and predicted_label == 'spam':
            num_fp = num_fp + 1

        print(f'\nPrediction: {predicted_label} | Correct Prediction? {label == predicted_label}')
        print(f'Total Emails Processed: {num_total} | Total Hams Predicted: {num_ham} | Total Spams Predicted: {num_spam}')
        print(f'True Positive: {num_tp} | True Negative: {num_tn}')
        print(f'False Positive: {num_fp} | False Negative: {num_fn}')

        print(f'\nSpam detected as spam: {num_tp}/{num_tp+num_fn} ({100*num_tp/(num_tp+num_fn) if (num_tp+num_fn) != 0 else 0} %)')
        print(f'\nHam detected as spam: {num_fp}/{num_tn+num_fp} ({100*num_fp/(num_tn+num_fp) if (num_tn+num_fp) != 0 else 0} %)')