import multiprocessing

import os
import pandas as pd
import sys

sys.path.insert(0, os.path.abspath(__file__).split("tmp_file")[0])


def hebing_and_save(new_batch_save_path, name, paths):
    df = pd.DataFrame()
    for path in paths:
        now_df = read_file_to_df(path)
        df = pd.concat([df, now_df])

    df.sort_values(by=['time_stamp'], inplace=True)

    create_file_path(new_batch_save_path)
    df.to_csv(os.path.join(new_batch_save_path, name), index=False, encoding='utf8')


if __name__ == '__main__':

    env = 'prod'
    if len(sys.argv) >= 2:
        env = sys.argv[1]

    from utils.conf.read_conf import yaml_conf

    conf_path = os.path.abspath(__file__).split("tmp_file")[0] + f"/conf/etl_config_{env}.yaml"
    os.environ['ETL_CONF'] = conf_path
    yaml_config = yaml_conf(conf_path)
    os.environ['env'] = env

    from utils.file.trans_methods import read_file_to_df, create_file_path

    from etl.wind_power.fault_warn.FaultWarnTrans import FaultWarnTrans
    from etl.wind_power.min_sec.MinSecTrans import MinSecTrans
    from service.plt_service import get_hebing_data_by_batch_no_and_type

    save_batch = 'WOF085500008-2-3'
    save_batch_name = '合并'
    trans_type = 'second'
    read_batchs = ['WOF085500008-WOB000003', 'WOF085500008-WOB000002']
    read_paths = list()

    new_batch_save_path = ''

    for read_data in read_batchs:
        data = get_hebing_data_by_batch_no_and_type(read_data, trans_type)
        save_db = True

        exec_process = None
        if data['transfer_type'] in ['second', 'minute']:
            exec_process = MinSecTrans(data=data, save_db=save_db)

        if data['transfer_type'] in ['fault', 'warn']:
            exec_process = FaultWarnTrans(data=data, save_db=save_db)

        if exec_process is None:
            raise Exception("No exec process")

        read_paths.append(exec_process.pathsAndTable.get_save_path())
        new_batch_save_path = os.path.join(exec_process.pathsAndTable.save_path, save_batch + "_" + save_batch_name,
                                           trans_type)

    file_dict = dict()

    for read_path in read_paths:
        for file in os.listdir(read_path):
            if file in file_dict:
                file_dict[file].append(os.path.join(read_path, file))
            else:
                file_dict[file] = [os.path.join(read_path, file)]

    with multiprocessing.Pool(len(file_dict.keys())) as pool:
        pool.starmap(hebing_and_save, [(new_batch_save_path, name, paths) for name, paths in file_dict.items()])