#! /usr/bin/env python

# -*- coding:utf8 -*-

# 切分nginx 按照日期切分日志文件

from __future__ import division

import os,sys

big_file=‘/data/logs/media.net.error.log‘

# 按照文件大小拆分

def split_by_filesize(fromfile,todir,chunksize=0):

"""

chunksize: 字節建議每100M一個獨立的文件

f.read(int byte字節)

100(M) = 100 * 1024 * 1024 (b) = 104857600 (b)

"""

# 創建切分文件后的臨時目錄

if not os.path.exists(todir):

os.mkdir(todir)

else:

for fname in os.listdir(todir):

os.remove(os.path.join(todir,fname))

# 按文件大小計算可切分的文件個數(100m一個)

partnums = os.path.getsize(fromfile) / chunksize

# 打開文件

partnum = 0

with open(fromfile,‘rb‘) as f:

while True:

chunk = f.read(chunksize) # 每次讀取100m

if not chunk:

break

filename = os.path.join(todir,big_file.split(‘/‘)[-1] + ‘_part_‘+str(partnum))

with open(filename,‘wb‘) as f1:

print "正在寫入第" + str(partnum) + "個文件" + filename

f1.write(chunk)

print "第"+ str(partnum) + "個文件寫入完成" + filename

partnum += 1

# 按照日志日期分割

def split_by_date(srcfile,todir):

"""

從源文件srcfile中讀取,確定日期,然后將其寫入到分片文件中

"""

# 讀取文件到內存中,利用generator 生成器

with open(srcfile) as f:

contents = f.readlines()

c = [ line[0:10].strip() for line in contents ]

date_l = sorted(set(c),key=c.index)

for date in date_l:

f_date = date.replace(‘/‘,‘_‘)

filename = os.path.join(todir,srcfile.split(‘/‘)[-1] + ‘_‘+ f_date)

print "正在寫入文件" + filename

with open(filename,‘wb‘) as f:

for line in contents:

d = line[0:10].strip()

if d in line and d == date:

f.write(line)

if __name__ == ‘__main__‘:

# split_by_filesize(big_file,‘/test/‘,chunksize=100*1024*1024)

split_by_date(big_file,‘/test/‘)

原文地址：http://shiyiguo.blog.51cto.com/4949751/1926322

總結

以上是生活随笔為你收集整理的python 日志不会按照日期分割_python 按照日期切分大日志文件（重点）和按照指定大小切分日志文件...的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。