问题描述：需要读取log文件，并获取出两天的日志做对比，查询增加或者减少量，规则化输出
完成心得：python的字典，很好用，读写时间快，用法自由
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 17/8/18 下午5:08
# @Author : lee
# @File : read.py
# @Software: PyCharm
# 说明: code后有'#'的时测试时加的或者需要修改的code
# 用法 传入两个值 day1是前一天的日期 day2是后一天的内容
import sys
def log_data(address, day1, day2):
try:
# 打开文件
log = open("%s" % address)
# 定义：
dict_yesterday = {} # 昨天的数据
dict_the_day_before_yesterday = {} # 前天的数据
dict_yesterday_difference = {} # 记录昨天中有的数据，前天没有
dict_the_day_before_yesterday_difference = {} # 和上一条相反
dict_identical = {} # 记录两天内相同的数据
dict_result = {} # 记录结果，输出时读取
ii = jj = z = x = 0
#遍历 将两个日期的日志筛选出并装入对应的字典
for line in log:
log_data = ','.join(filter(lambda x: x,line.split(' ')))
date = (log_data.split(',')[0])[-10:].replace('.','-')
if date == day2:
dict_yesterday[ii] = log_data
dict_yesterday_difference[ii] = log_data
ii += 1
elif date == day1:
dict_the_day_before_yesterday[jj] = log_data
dict_the_day_before_yesterday_difference[jj] = log_data
jj += 1
if len(dict_yesterday) >= len(dict_the_day_before_yesterday):
for i in dict_yesterday:
object_list1 = (dict_yesterday[i]).split(',')
entry_name_yesterday = ((object_list1[0]).split(',')[0])[:-10]+object_list1[1]+'-'+object_list1[2]
for j in dict_the_day_before_yesterday:
object_list2 = (dict_the_day_before_yesterday[j]).split(',')
entry_name_tdat = ((object_list2[0]).split(',')[0])[:-10]+object_list2[1]+'-'+object_list2[2]
if entry_name_yesterday == entry_name_tdat:
dict_identical[x] = entry_name_yesterday
docs_result = int(((dict_yesterday[i]).split(','))[4]) - int(((dict_the_day_before_yesterday[j]).split(','))[4])
capacity1 = (dict_yesterday[i].split(',')[5])[-2:]
capacity2 = (dict_the_day_before_yesterday[j].split(',')[5])[-2:]
if capacity1 == capacity2:
store_result = str(round(float((dict_yesterday[i].split(',')[5])[:-2]) - float((dict_the_day_before_yesterday[j].split(',')[5])[:-2]),2)) +(dict_yesterday[i].split(',')[5])[-2:]
# 1gb = 1024mb = 1024kb
elif capacity1 != capacity2:
n_y = round(float((dict_yesterday[i].split(',')[5])[:-2]), 2)
n_t = float((dict_the_day_before_yesterday[j].split(',')[5])[:-2])
if capacity1 == 'gb'and capacity2 == 'mb':
if n_y * 1024 > n_t:
store_result = str(abs(round(n_y * 1024 - n_t), 2)) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
else:
store_result = str(round(n_y * 1024 - n_t),) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
elif capacity1 == 'gb'and capacity2 == 'kb':
if n_y * 1024 * 1024 > n_t:
# round(,2)
store_result = str(abs(round(n_y * 1024 * 1024 - n_t,2))) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
else:
store_result = str(round(n_y * 1024 * 1024 - n_t,2)) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
elif capacity1 == 'mb'and capacity2 == 'gb':
if n_y / 1024 > n_t:
store_result = str(abs(round(n_y / 1024 - n_t,2))) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
else:
store_result = str(round(n_y / 1024 - n_t,2)) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
elif capacity1 == 'mb'and capacity2 == 'kb':
if n_y * 1024 > n_t:
store_result = str(abs(round(n_y * 1024 - n_t,2))) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
else:
store_result = str(round(n_y * 1024 - n_t,2)) +(dict_the_day_before_yesterday[j].split(',')[5])[-2:]
elif capacity1 == 'kb'and capacity2 == 'gb':
store_result = str(round(n_t - n_t * 1024 * 1024,2)) +(dict_yesterday[i].split(',')[5])[-2:]
elif capacity1 == 'kb'and capacity2 == 'mb':
if n_y / 1024 > n_t:
store_result = str(abs(round(n_t - n_t * 1024,2))) +(dict_yesterday[i].split(',')[5])[-2:]
else:
store_result = str(round(n_t - n_t * 1024,2)) +(dict_yesterday[i].split(',')[5])[-2:]
dict_result[z] = entry_name_tdat + ' docs: ' + str(docs_result) + ' store: ' +store_result
z += 1
x += 1
# else:
# print("当前日期下没有相关日志")
# 筛选出昨天出现 前天没有的数据
if len(dict_yesterday) >= len(dict_identical):
for i in dict_yesterday:
object_list_1 = (dict_yesterday[i]).split(',')
entry_name_yesterday_1 = ((object_list_1[0]).split(',')[0])[:-10]+object_list_1[1]+'-'+object_list_1[2]
for j in dict_identical:
entry_name_identical = dict_identical[j]
if entry_name_yesterday_1 == entry_name_identical:
dict_yesterday_difference.pop(i)
# 筛选出前天出现 昨天没有的数据
if len(dict_the_day_before_yesterday) >= len(dict_identical):
for i in dict_the_day_before_yesterday:
object_list_1 = (dict_the_day_before_yesterday[i]).split(',')
entry_name_yesterday_1 = ((object_list_1[0]).split(',')[0])[:-10]+object_list_1[1]+'-'+object_list_1[2]
for j in dict_identical:
entry_name_identical = dict_identical[j]
if entry_name_yesterday_1 == entry_name_identical:
dict_the_day_before_yesterday_difference.pop(i)
# 输出两天都有的数据
print("下面是 %s 和 %s 内%s中 docs、store差值"%(day1,day2,address))
print('')
print("++++++++++++++++两天都有的数据++++++++++++++++")
for i in dict_result:
print(dict_result[i])
# 输出两天中前者中有的数据而在后者没有出现，取正值(+)区别
print('')
if len(dict_yesterday) != 0 :
print("++++++++++++++++只有%s有的数据取正值(+)予以区别++++++++++++++++"%day2)
for i in dict_yesterday_difference:
object = (dict_yesterday_difference[i]).split(',')
name = ((object[0]).split(',')[0])[:-10]+object[1]+'-'+object[2]
docs = object[4]
store = object[5]
print(name + ' docs: +' + str(docs) + ' store: ' + store)
else:
print("%s日期下没有相关日志"%day2)
# 输出两天中后者中有的数据而在前者没有出现，取负值(-)区别
print('')
if len(dict_the_day_before_yesterday) != 0:
print("++++++++++++++++只有%s有的数据，都取负值(-)予以区别++++++++++++++++" % day1)
for i in dict_the_day_before_yesterday_difference:
object = (dict_the_day_before_yesterday_difference[i]).split(',')
name = ((object[0]).split(',')[0])[:-10]+object[1]+'-'+object[2]
docs = object[4]
store = object[5]
print(name+ ' docs: -' + str(docs) + ' store: -' + store)
else:
print("%s日期下没有相关日志"%day1)
except:
print("输入的参数可能不对，三个参数，地址，day1 day2")
# log_data('/Users/lee/Desktop/py交易/es.log','2017-08-11','2017-08-12')
log_data(sys.argv[1],sys.argv[2],sys.argv[3])
# /Users/lee/Desktop/py交易/test.log
