**
需求:使用线程池同时处理多个文本
**
# -*- coding: utf-8 -*-
import re
import os
import logging
from concurrent.futures.thread import ThreadPoolExecutor
"""
代码逻辑:
1、函数逻辑是处理某个文本,处理好重新命名保存
2、采用线程池的方式同时处理多份文本
"""
#配置日志路径
LogFileName = r'C:\Users\Administrator\Desktop\1'
def countuce(fulldir):
Data = {}
try:
file = open(fulldir, encoding="utf-8")
while True:
line = file.readline()
if not line:
break
if "]_begin_" not in line and "]_end_" not in line:
continue
strList = re.findall(' \[(.*)\]_', line)
startTime = strList[0].split("::")[-1] + strList[0].split("::")[1]
if startTime not in Data:
Data[startTime] = line
else:
del Data[startTime]
except Exception as e:
logging.getLogger().error(e)
for d in Data.values():
entlog = fulldir.split(".")[1]
write_log = fulldir.replace(entlog, 'txt') #输入日志格式为.txt
try:
with open(write_log, mode='a+',encoding='utf-8') as f:
f.write(d)
except Exception as e:
logging.getLogger().error(e)
if __name__ == '__main__':
#开启线程数
pool = ThreadPoolExecutor(50) #设置50个线程数
for ipath in os.listdir(LogFileName): # 遍历路径
fulldir = os.path.join(LogFileName, ipath)
pool.submit(countuce,fulldir)
#等所有任务执行完
pool.shutdown()