Python2.7版本的多线程开发
可以使得程序执行效率至少提升10倍
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
@Time : 2018/10/24
@Author : LiuXueWen
@Site :
@File : transfer.py
@Software: PyCharm
@Description:
"""
import os
import traceback
import threading
from multiprocessing import Pool
from multiprocessing.dummy import Pool as ThreadPool
# 兼容python2.7上多线程的bug,不加上下面的反代理程序不能正常执行
def proxy(cls_instance, i):
return cls_instance.multiprocess_thread(i)
def proxy2(cls_instance, i):
return cls_instance.file_operation(i)
class file2transfer():
# 多进程执行程序
def multiprocessingTransferFiles(self):
try:
# 创建进程池
p = Pool()
//参数末尾必须加上逗号
p.apply_async(proxy, args=(self, self.root_path,))
p.close()
p.join()
except Exception as e:
print(e)
# 每个进程下的多线程执行,线程数等于当前机器的核数
def multiprocess_thread(self, root_path):
try:
# 创建线程锁
lock = threading.RLock()
lock.acquire()
# 获取每个文件
for pfile in os.listdir(root_path):
# 获取文件的完整路径
full_file_path = os.path.join(root_path, pfile)
# 多线程读写文件
p = ThreadPool()
# 执行线程
p.apply_async(proxy2, args=(self, full_file_path,))
p.close()
p.join()
except Exception as e:
print(e)
finally:
# 释放线程锁
lock.release()
# 对每个文件夹下的每个文件进行操作
def file_operation(self, full_file_path):
try:
// TODO 真正需要单独执行的操作
pass
except Exception as e:
print(e)