nomn文件分析

#encoding=gbk
import os
import re
import math
from os import path

'''
手动输入文件nmon文件路径，要截取的开始时间，结束时间
'''

rootdir='E:\pylianxi\ceshi'    #input("请输入文件路径：")
start_time="14:46"        #截取的开始时间
end_time="14:56"        #截取的结束时间
print(rootdir)

def avg(list_a): #求平均值函数
    avg_list=0.0
    float_item_list=[float(i) for i in list_a]
    for i in float_item_list:
        if not isinstance(i,(int,float)):
            print("列表中含有非数字！！")
            return None
        
    if len(float_item_list)>0:
        avg_list=sum(float_item_list)/len(float_item_list)
        #return round(avg_list,3)
        return ('%.2f'%avg_list)
    else:
        return None


write_count=0 #记录是否是第一次生成要输出的文件aa.txt
file_names=os.listdir(rootdir)

for i in file_names:
    file_path=path.join(rootdir,i)
    if path.isfile(file_path) and ".nmon" in i:
        print(file_path)
        #file_record=re.findall(r"w+",i)[0]+re.findall(r"d+",i)[0]
        file_record=i[:6]   #+i[4:6]
        #file_record=re.search(r"(w+)-(d+)",i).group(1)+re.search(r"(w+)-(d+)",i).group(2)
        print(file_record)
        #with open(file_path,'r',encoding='gbk') as fp:
        with open(file_path,'r',encoding='utf-8') as fp:
            data_all=fp.readlines()
            data_times={}
            for j in range(len(data_all)):
                if re.search(r"ZZZZ,T",data_all[j]):
                    #data_times.append(data_all[j].split(',')[2])
                    data_times[j]=data_all[j].split(',')[2]
            #print(data_times)
            start_row=0
            end_row=max(data_times.keys())
            
            for k,v in data_times.items():
                if start_time in v:
                    start_row=k
                    print("循环中的start_row:",start_row)
                    break
            
            for k,v in data_times.items():
                if end_time in v:
                    end_row=k
                    print("循环中的end_row:",end_row)
                    break
            

            print("校验前的start_row:",start_row)
            if start_row==0:
                print("未找到开始时间，从文件头开始")
            
            print("校验前的end_row:",end_row)
            if end_row==max(data_times.keys()):
                print("未找到结束时间，计算至文件结尾")

            performance_cpu=[]
            performance_mem=[]
            performance_disk_single=[]
            performance_disk=[]
            for n in data_all[start_row:end_row]:
                if re.search(r"CPU_ALL,T.*",n):
                    #print(n.split(",")[5])
                    performance_cpu.append(100-float(n.split(",")[5]))
                if re.search(r"MEM,T.*",n):
                    performance_mem.append((float(n.split(",")[2])-float(n.split(",")[6])-float(n.split(",")[11]))*100/float(n.split(",")[2]))
                if re.search(r"DISKXFER,T.*",n):
                    performance_disk_single=n.split(",")[2:]
                    sum_disk=0.0
                    for nn in performance_disk_single:
                        sum_disk+=float(nn)
                    performance_disk.append(sum_disk)
            #print(performance_cpu)
            #print(performance_mem)
            #print(performance_disk)
            cpu_avg=avg(performance_cpu)
            mem_avg=avg(performance_mem)
            disk_avg=avg(performance_disk)
        output_filename='aa.txt'
        output_path=path.join(rootdir,output_filename)
        
        
        if write_count==0:
            fp1=open(output_path,'w',encoding='gbk')
            fp1.write("")
            write_count+=1
            fp1.close()
                
        with open(output_path,'a+',encoding='gbk') as fp2:
            fp2.write(file_record+','+str(cpu_avg)+','+str(mem_avg)+','+str(disk_avg)+'
')