import csv
from Levenshtein import distance
import pandas as pd
import uuid


# TODO Filter für Spalten, ggfs. Klasse benötigt

# TODO Filter für Dublikate hier wird dann die Klasse benötigt


def read_csv(file_path):
    data = []
    with open(file_path, newline='', encoding='utf-8') as csvfile:
        reader = csv.reader(csvfile, delimiter=';')
        for row in reader:
            data.append((row[0].strip(), row[1].strip()))
    return data


def similar_sets(pair, data):
    similar_pairs = []
    for item in data:
        if distance(pair[0], item[0]) <= 1 and distance(pair[1], item[1]) <= 1:
            similar_pairs.append(item)
    return similar_pairs


def compare_csv(file1, file2):
    data1 = read_csv(file1)
    data2 = read_csv(file2)

    common_pairs = set(data1) & set(data2)
    unique_pairs1 = set(data1) - common_pairs
    unique_pairs2 = set(data2) - common_pairs

    return common_pairs, unique_pairs1, unique_pairs2, data1, data2


def find_similar_pairs(pair, other_data):
    similar_pairs = []
    for item in other_data:
        if distance(pair[0], item[0]) <= 2 and distance(pair[1], item[1]) <= 2:
            similar_pairs.append(item)
    return similar_pairs


def create_uuid():
    return str(uuid.uuid4())


def add_hl_tag(row):
    klasse = str(row['klasse']).lstrip('0')
    return 'HL0707104-' + klasse


def create_import_list(path, path_new, old_pairs, new_pairs, common_pairs):
    system_data = pd.read_csv(path, sep=';', encoding='utf-8')
    system_data = system_data[~system_data[['name', 'vorname']].apply(tuple, axis=1).isin(old_pairs)]
    system_data = system_data.drop('username', axis=1, errors='ignore')
    system_data = system_data.drop('klasse', axis=1, errors='ignore')
    # print(len(system_data))
    # print(system_data)
    new_data = pd.read_csv(path_new, sep=';', encoding='utf-8')

    # Bei Schüler: alte Klassen gelöscht, mit neuen Klassen aus new-data auffüllen

    matches = new_data[~new_data[['name', 'vorname']].apply(tuple, axis=1).isin(new_pairs)]
    matches.loc[:, 'klasse'] = matches.apply(add_hl_tag, axis=1)
    # print(len(matches))
    # print(matches)
    system_data = pd.merge(system_data, matches, how='outer', left_on=['name', 'vorname'], right_on=['name', 'vorname'])
    system_data = system_data[['name', 'vorname', 'klasse', 'schuelerid']]
    # print(system_data)
    print(len(system_data))

    new_data = new_data[~new_data[['name', 'vorname']].apply(tuple, axis=1).isin(common_pairs)]
    # new_data = new_data.drop('Unnamed: 2', axis=1, errors='ignore')
    new_uuids = []
    for row in range(len(new_data)):
        new_uuids.append(create_uuid())
    # Klasse?? Unterschied zwischen Lehrer und Schüler
    # new_data['klasse'] = None
    new_data.loc[:, 'klasse'] = new_data.apply(add_hl_tag, axis=1)
    new_data.insert(loc=2, column='schuelerid', value=new_uuids)
    mailUserQuota = 1024
    oxUserQuota = 5120
    oxContext = 25

    # print(new_data)
    print(len(new_data))

    # vor dem merge daten ergänzen
    # import_df = pd.merge(system_data, new_data, how='outer', left_on=['name', 'vorname', 'mailUserQuota', 'oxUserQuota', 'oxContext'], right_on=['name', 'vorname', 'mailUserQuota', 'oxUserQuota', 'oxContext'])
    # import_df = pd.merge(system_data, new_data, how='outer',
    #                      left_on=['name', 'vorname', 'klasse', 'schuelerid', 'mailUserQuota', 'oxUserQuota',
    #                               'oxContext'],
    #                      right_on=['name', 'vorname', 'klasse', 'schuelerid', 'mailUserQuota', 'oxUserQuota',
    #                                'oxContext'])
    import_df = pd.merge(system_data, new_data, how='outer',
                         left_on=['name', 'vorname', 'klasse', 'schuelerid'],
                         right_on=['name', 'vorname', 'klasse', 'schuelerid'])
    import_df['mailUserQuota'] = mailUserQuota
    import_df['oxUserQuota'] = oxUserQuota
    import_df['oxContext'] = oxContext

    # pd.set_option('display.max_rows', None)
    # pd.set_option('display.max_columns', None)
    # print(import_df)
    # pd.reset_option('display.max_rows')
    # pd.reset_option('display.max_columns')
    out_path = '../Data/GPS/outputSchueler.csv'
    import_df.to_csv(out_path, sep=';', index=False)
    print(len(import_df))
    print('Schüler Import Liste erzeugt')
    print('Testuser manuell nachtragen!!')


def main():
    file1_path = '../Data/GPS/gpsSnew.csv'
    file2_path = '../Data/GPS/gpsSold2.cvs'
    file3_path = '../Data/GPS/gpsSold.csv'

    common_pairs, new_pairs, old_pairs, data1, data2 = compare_csv(file1_path, file2_path)

    print(f"Anzahl der übereinstimmenden Paare: {len(common_pairs)}")
    print(f"Anzahl der neuen Einträge: {len(new_pairs)}")
    print(f"Anzahl der veralteten Einträge: {len(old_pairs)}")

    # Paare nur aus nicht zugeordneten Paaren aus neuer Liste erstellen
    # for pair in data1:
    #     similar_pairs_list2 = find_similar_pairs(pair, set(data2) - {pair})
    #
    #     if similar_pairs_list2:
    #         print(f"Ähnliche Paare in neuer Liste {pair} aktuell im System: {similar_pairs_list2}")

    create_import_list(file3_path, file1_path, old_pairs, new_pairs, common_pairs)


if __name__ == "__main__":
    main()