energy-data-trans/etl/step/UnzipAndRemove.py

import multiprocessing
import os
import traceback

import datetime

from etl.base.PathsAndTable import PathsAndTable
from service.plt_service import update_trans_transfer_progress
from utils.file.trans_methods import read_files, read_excel_files, copy_to_new, split_array
from utils.log.trans_log import trans_print
from utils.systeminfo.sysinfo import get_available_cpu_count_with_percent
from utils.zip.unzip import unzip, unrar, get_desc_path


class UnzipAndRemove(object):
    def __init__(self, pathsAndTable: PathsAndTable):
        self.pathsAndTable = pathsAndTable

    def get_and_remove(self, file):

        to_path = self.pathsAndTable.get_excel_tmp_path()
        if str(file).endswith("zip"):
            if str(file).endswith("csv.zip"):
                copy_to_new(file, file.replace(self.pathsAndTable.read_path, to_path).replace("csv.zip", 'csv.gz'))
            else:
                desc_path = file.replace(self.pathsAndTable.read_path, to_path)
                is_success, e = unzip(file, get_desc_path(desc_path))
                self.pathsAndTable.has_zip = True
                if not is_success:
                    # raise e
                    pass
        elif str(file).endswith("rar"):
            desc_path = file.replace(self.pathsAndTable.read_path, to_path)
            is_success, e = unrar(file, get_desc_path(desc_path))
            self.pathsAndTable.has_zip = True
            if not is_success:
                trans_print(traceback.format_exc())
                pass
        else:
            copy_to_new(file, file.replace(self.pathsAndTable.read_path, to_path))

    def remove_file_to_tmp_path(self):
        # 读取文件
        try:
            if os.path.isfile(self.pathsAndTable.read_path):
                all_files = [self.pathsAndTable.read_path]
            else:
                all_files = read_files(self.pathsAndTable.read_path)

            # 最大取系统cpu的 三分之二
            split_count = get_available_cpu_count_with_percent(1/2)
            all_arrays = split_array(all_files, split_count)

            for index, arr in enumerate(all_arrays):
                with multiprocessing.Pool(self.pathsAndTable.multi_pool_count) as pool:
                    pool.starmap(self.get_and_remove, [(i,) for i in arr])
                update_trans_transfer_progress(self.pathsAndTable.batch_no, self.pathsAndTable.read_type,
                                               round(5 + 15 * (index + 1) / len(all_arrays), 2),
                                               self.pathsAndTable.save_db)

            all_files = read_excel_files(self.pathsAndTable.get_excel_tmp_path())

            trans_print('读取文件数量:', len(all_files))
        except Exception as e:
            trans_print(traceback.format_exc())
            message = "读取文件列表错误:" + self.pathsAndTable.read_path + ",系统返回错误:" + str(e)
            raise ValueError(message)
        return all_files

    def run(self):
        trans_print("开始解压移动文件")
        begin = datetime.datetime.now()
        self.remove_file_to_tmp_path()
        update_trans_transfer_progress(self.pathsAndTable.batch_no, self.pathsAndTable.read_type, 20,
                                       self.pathsAndTable.save_db)
        trans_print("解压移动文件结束:耗时:", datetime.datetime.now() - begin)