python处理csv数据的方法
作者:kirkversion1 发布时间:2023-03-08 19:05:31
标签:python,csv,数据,方法
本文实例讲述了python处理csv数据的方法。分享给大家供大家参考。具体如下:
Python代码:
#coding=utf-8
__author__ = 'dehua.li'
from datetime import *
import datetime
import csv
import sys
import time
import string
import os
import os.path
import pylab as plt
rootdir='/nethome/dehua.li/orderlifeCycleData/xingzheng'
writeFileDir="/nethome/dehua.li/orderlifeMyWork/xingzheng/csv"
heyueFile="/nethome/dehua.li/orderlifeCycleData/heyue_150128.csv"
ms_acked="1"
msg=[]
ex=[]
def getTheDate(date):
[filenamePart1,filenamePart2]=string.split(filename,'.')
[filenamePart11,filenamePart12,filenamePart13]=string.split(filenamePart1,'_')
return filenamePart13
LocalTime=datetime.datetime.fromtimestamp(time.mktime(time.strptime("2014-11-04 20:59:59","%Y-%m-%d %H:%M:%S")))
for parent,dirname,filenames in os.walk(rootdir):
for filename in filenames:
fileNameWrite=os.path.join(writeFileDir,filename)
print fileNameWrite
csvfile00=open(fileNameWrite,'wb')
writer1=csv.writer(csvfile00)
writer1.writerow(['FeedCode','OrderId','Status','LocalTime','Time','Exchange'])
fileName=os.path.join(parent,filename)
[filenamePart1,filenamePart2]=string.split(filename,'.')
[filenamePart11,filenamePart12,filenamePart13]=string.split(filenamePart1,'_')
#filenamePart11_filenamePart12_filenamePart13.filenamePart2:dongzheng_orderlifeCycleData_20150111.csv
print fileName
with open(fileName,'rb') as csvfile:
reader=csv.reader(csvfile)
CsvItem=[row for row in reader]
for item in CsvItem:
if item[3]=='TPO':
#print " filter TPO "
continue
if item[12]=='Sent':
[tm_local,ms_local]=string.split(item[15],'.')
[tm_localup,ms_localup]=string.split(item[19],'.')
LocalTime=datetime.datetime.fromtimestamp(time.mktime(time.strptime(tm_local,"%Y-%m-%d %H:%M:%S")))
LocalUpdate=datetime.datetime.fromtimestamp(time.mktime(time.strptime(tm_localup,"%Y-%m-%d %H:%M:%S")))
tm=int(((LocalTime-LocalUpdate).seconds))*1000
ms_sent=str(int(ms_local)-int(ms_localup)+tm)
if int(ms_sent)>10*60*1000:
print "ms_sent>600000"
continue
if(int(ms_local)-int(ms_localup)+tm)<0:
print 'wrong1'
msg=[]
msg.append(item[0])
msg.append(item[1])
msg.append(item[12])
msg.append(item[15])
msg.append(ms_sent)
with open(heyueFile,'rb') as csvfile1:
reader=csv.reader(csvfile1)
CsvItem=[row for row in reader]
for Item in CsvItem:
if Item[1]==item[0]:
msg.append(Item[3])
writer1.writerow(msg)
#print 'write ok'
ex=Item[3]
break
csvfile1.close()
with open(fileName,'rb') as csvfile22:
reader=csv.reader(csvfile22)
CsvItem2=[row for row in reader]
for item_ in CsvItem2:
if item_[12]=='Acked' and item_[1]==item[1]:
[tm_local2,ms_local2]=string.split(item_[15],'.')
LocalTime2=datetime.datetime.fromtimestamp(time.mktime(time.strptime(tm_local2,"%Y-%m-%d %H:%M:%S")))
tm2=int(((LocalTime2-LocalTime).seconds))*1000
ms_acked=str(int(ms_local2)-int(ms_local)+tm2)
if int(ms_acked)>10*60*1000:
print "MSacked>600000"
continue
msg=[]
msg.append(item_[0])
msg.append(item_[1])
msg.append(item_[12])
msg.append(item_[15])
msg.append(ms_acked)
with open(heyueFile,'rb') as csvfile111:
reader=csv.reader(csvfile111)
CsvItem=[row for row in reader]
for Item in CsvItem:
if Item[1]==item[0]:
msg.append(Item[3])
writer1.writerow(msg)
#print 'write ok'
break
#print "write ok"
csvfile22.close()
csvfile.close()
csvfile00.close()
#coding=utf-8
#__author__ = 'dehua.li'
from datetime import *
import datetime
import csv
import sys
import time
import string
import os
import os.path
import pylab as plt
def median(lst):
even = (0 if len(lst) % 2 else 1) + 1
half = (len(lst) - 1) / 2
return sum(sorted(lst)[half:half + even]) / float(even)
def mean(lst):
if len(lst)==0:
return 0
return sum(lst)/len(lst)
nightLine="21:01:00"
morningLine="09:01:00"
def getTheDate(date):
[filenamePart1,filenamePart2]=string.split(filename,'.')
[filenamePart11,filenamePart12,filenamePart13]=string.split(filenamePart1,'_')
return filenamePart13
def afterOneMin(time):
[tm_local,ms_local]=string.split(time,'.')
[ymd,hms]=string.split(tm_local,' ')
flag=0
if hms>"21:01:00":
flag=1
elif hms>"09:01:00" and hms<"20:00:00":
flag=1
elif hms>"00:00:00" and hms<"05:00:00":
flag=1
return flag
rootdir="/nethome/dehua.li/orderlifeMyWork/xingzheng/csv"
#csvfileMaxMin = open('e:\dehua.li\csv\__xingzhenMaxMin.csv','wb')
#writer1 = csv.writer(csvfileMaxMin)
#writer1.writerow(['FeedCode','date','SentMaxTime','SentMaxLocalTime','SentMinTime','SentMinLocalTime','SentMeanTime','SentMedian','AckedMaxTime','AckedMaxLocalTime','AckedMinTime','AckedMinLocalTime','AckedMeanTime','AckedMedianTime','Exchange'])
#writer1.writerow(['FeedCode','date','SentMaxTime','SentMinTime','SentMeanTime','SentMedian','AckedMaxTime','AckedMinTime','AckedMeanTime','AckedMedianTime','Exchange'])
msg=[]
codeList=list()
orderList=list()
itemSentList=[]
itemAckedList=[]
feedCode=[]
exchange=[]
zhengshangSentMedian=0
zhengshangSentMean=0
zhengshangAckedMedian=0
zhengshangAckedMean=0
dashangSentMedian=0
dashangSentMean=0
dashangAckedMedian=0
dashangAckedMean=0
shangqiSentMedian=0
shangqiSentMean=0
shangqiAckedMedian=0
shangqiAckedMean=0
zhongjinSentMedian=0
zhongjinSentMean=0
zhongjinAckedMedian=0
zhongjinAckedMean=0
zhengshangSent=[]
zhengshangAcked=[]
dashangSent=[]
dashangAcked=[]
shangqiSent=[]
shangqiAcked=[]
zhongjinSent=[]
zhongjinAcked=[]
zhengshangSentMedianAll=[]
zhengshangSentMeanAll=[]
zhengshangAckedMedianAll=[]
zhengshangAckedMeanAll=[]
dashangSentMedianAll=[]
dashangSentMeanAll=[]
dashangAckedMedianAll=[]
dashangAckedMeanAll=[]
shangqiSentMedianAll=[]
shangqiSentMeanAll=[]
shangqiAckedMedianAll=[]
shangqiAckedMeanAll=[]
zhongjinSentMedianAll=[]
zhongjinSentMeanAll=[]
zhongjinAckedMedianAll=[]
zhongjinAckedMeanAll=[]
zhengshang='0'
dashang='0'
shangqi='0'
zhongjin='0'
with open('/nethome/dehua.li/orderlifeCycleData/heyue_150128.csv','rb') as csvfile:
reader=csv.reader(csvfile)
csvItem=[row for row in reader]
zhengshang=csvItem[300][3]
dashang=csvItem[5][3]
shangqi=csvItem[165][3]
zhongjin=csvItem[435][3]
#for item in csvItem:
# if item[3]==zhengshang:
# print item
for parent,dirname,filenames in os.walk(rootdir):
for filename in filenames:
fileName=os.path.join(rootdir,filename)
csvfile1=open(fileName,'rb')
reader=csv.reader(csvfile1)
CsvItem=[row for row in reader]
for item in CsvItem:
if item[0]=='FeedCode':
continue
if item[0] not in codeList:
codeList.append(item[0])
#print CsvItem[15]
if len(item)<=5:
print fileName
print item
print '++++++++++++++++++++++++++++++'
#if afterOneMin(item[3])==0:
# print item[3]
# continue
if item[5]==zhengshang and item[2]=='Sent':
zhengshangSent.append(int(item[4]))
elif item[5]==zhengshang and item[2]=='Acked':
zhengshangAcked.append(int(item[4]))
elif item[5]==dashang and item[2]=='Sent':
dashangSent.append(int(item[4]))
elif item[5]==dashang and item[2]=='Acked':
dashangAcked.append(int(item[4]))
elif item[5]==shangqi and item[2]=='Sent':
shangqiSent.append(int(item[4]))
if int(item[4])>=600000:
print "------------"
print item
elif item[5]==shangqi and item[2]=='Acked':
shangqiAcked.append(int(item[4]))
elif item[5]==zhongjin and item[2]=='Sent':
zhongjinSent.append(int(item[4]))
elif item[5]==zhongjin and item[2]=='Acked':
zhongjinAcked.append(int(item[4]))
else:
print "wrong info"
print item
if mean(shangqiSent)>420000:
print sum(shangqiSent)
print len(shangqiSent)
print item
print fileName
print shangqiSent
zhengshangSentMedian=median(zhengshangSent)
zhengshangSentMean=mean(zhengshangSent)
zhengshangAckedMedian=median(zhengshangAcked)
zhengshangAckedMean=mean(zhengshangAcked)
dashangSentMedian=median(dashangSent)
dashangSentMean=mean(dashangSent)
dashangAckedMedian=median(dashangAcked)
dashangAckedMean=mean(dashangAcked)
shangqiSentMedian=median(shangqiSent)
shangqiSentMean=mean(shangqiSent)
shangqiAckedMedian=median(shangqiAcked)
shangqiAckedMean=mean(shangqiAcked)
zhongjinSentMedian=median(zhongjinSent)
zhongjinSentMean=mean(zhongjinSent)
zhongjinAckedMedian=median(zhongjinAcked)
zhongjinAckedMean=mean(zhongjinAcked)
#if mean(shangqiSent)>70:
# print '================================'
# print fileName
#print codeList
'''
for listItem in codeList:
itemSentList=[]
itemAckedList=[]
for item in CsvItem:
if item[0]==listItem and item[2]=='Sent':
itemSentList.append(int(item[4]))
exchange=item[5]
elif item[0]==listItem and item[2]=='Acked':
itemAckedList.append(int(item[4]))
#print itemSentList
itemMaxSent=max(itemSentList)
itemMinSent=min(itemSentList)
itemAvgSent=sum(itemSentList)/len(itemSentList)
itemMaxAcked=max(itemAckedList)
itemMinAcked=min(itemAckedList)
itemAvgAcked=sum(itemAckedList)/len(itemAckedList)
SentMedian=median(itemSentList)
AckedMedian=median(itemAckedList)
msg=[]
msg.append(listItem) #0
msg.append("2015/01/14") #1
msg.append(itemMaxSent) #2
msg.append(itemMinSent) #3
msg.append(itemAvgSent) #4
msg.append(SentMedian) #5
msg.append(itemMaxAcked) #6
msg.append(itemMinAcked) #7
msg.append(itemAvgAcked) #8
msg.append(AckedMedian) #9
msg.append(exchange) #10
if len(msg)>15:
print "------------------------------"
print msg
writer1.writerow(msg)
'''
zhengshangSentMedianAll.append(zhengshangSentMedian)
zhengshangSentMeanAll.append(zhengshangSentMean)
zhengshangAckedMedianAll.append(zhengshangAckedMedian)
zhengshangAckedMeanAll.append(zhengshangAckedMean)
dashangSentMedianAll.append(dashangSentMedian)
dashangSentMeanAll.append(dashangSentMean)
dashangAckedMedianAll.append(dashangAckedMedian)
dashangAckedMeanAll.append(dashangAckedMean)
shangqiSentMedianAll.append(shangqiSentMedian)
shangqiSentMeanAll.append(shangqiSentMean)
shangqiAckedMedianAll.append(shangqiAckedMedian)
shangqiAckedMeanAll.append(shangqiAckedMean)
zhongjinSentMedianAll.append(zhongjinSentMedian)
zhongjinSentMeanAll.append(zhongjinSentMean)
zhongjinAckedMedianAll.append(zhongjinAckedMedian)
zhongjinAckedMeanAll.append(zhongjinAckedMean)
plt.figure(1)
plt.figure(2)
plt.figure(3)
plt.figure(4)
plt.figure(1)
plt.title('SentMean r-zhengshang b-dashang,green-shangqi grey-zhongjin')
plt.plot(range(1,len(zhengshangSentMeanAll)+1),zhengshangSentMeanAll,'r')
plt.plot(range(1,len(dashangSentMeanAll)+1),dashangSentMeanAll,'b')
plt.plot(range(1,len(shangqiSentMeanAll)+1),shangqiSentMeanAll,'g')
plt.plot(range(1,len(zhongjinSentMeanAll)+1),zhongjinSentMeanAll,'grey')
plt.savefig('/nethome/dehua.li/orderlifeMyWork/xingzheng/data_noTPO_in10minutes/SentMean.png')
plt.figure(2)
plt.title('SentMedian r-zhengshang b-dashang,green-shangqi grey-zhongjin')
plt.plot(range(1,len(zhengshangSentMedianAll)+1),zhengshangSentMedianAll,'r')
plt.plot(range(1,len(dashangSentMedianAll)+1),dashangSentMedianAll,'b')
plt.plot(range(1,len(shangqiSentMedianAll)+1),shangqiSentMedianAll,'g')
plt.plot(range(1,len(zhongjinSentMedianAll)+1),zhongjinSentMedianAll,'grey')
plt.savefig('/nethome/dehua.li/orderlifeMyWork/xingzheng/data_noTPO_in10minutes/SentMedian.png')
plt.figure(3)
plt.title('AckedMean r-zhengshang b-dashang,green-shangqi grey-zhongjin')
plt.plot(range(1,len(zhengshangAckedMeanAll)+1),zhengshangAckedMeanAll,'r')
plt.plot(range(1,len(dashangAckedMeanAll)+1),dashangAckedMeanAll,'b')
plt.plot(range(1,len(shangqiAckedMeanAll)+1),shangqiAckedMeanAll,'g')
plt.plot(range(1,len(zhongjinAckedMeanAll)+1),zhongjinAckedMeanAll,'grey')
plt.savefig('/nethome/dehua.li/orderlifeMyWork/xingzheng/data_noTPO_in10minutes/AckedMean.png')
plt.figure(4)
plt.title('AckedMedian r-zhengshang b-dashang,green-shangqi grey-zhongjin')
plt.plot(range(1,len(zhengshangAckedMedianAll)+1),zhengshangAckedMedianAll,'r')
plt.plot(range(1,len(dashangAckedMedianAll)+1),dashangAckedMedianAll,'b')
plt.plot(range(1,len(shangqiAckedMedianAll)+1),shangqiAckedMedianAll,'g')
plt.plot(range(1,len(zhongjinAckedMedianAll)+1),zhongjinAckedMedianAll,'grey')
plt.savefig('/nethome/dehua.li/orderlifeMyWork/xingzheng/data_noTPO_in10minutes/AckedMedian.png')
plt.show()
print 'over'
希望本文所述对大家的Python程序设计有所帮助。
0
投稿
猜你喜欢
- NumPy(Numerical Python) 是 Python 语言的一个扩展程序库,支持大量的维度数组与矩阵运算,此外也针对数组运算提供
- 当请求 headers 中,添加一个name为 Accept,值为 application/json 的 header(也即“我”(浏览器)
- PDOStatement::fetchPDOStatement::fetch — 从结果集中获取下一行(PHP 5 >= 5.1.0,
- 引言上一篇文章中引入了消息队列对秒杀流量做削峰的处理,我们使用的是Kafka,看起来似乎工作的不错,但其实还是有很多隐患存在,如果这些隐患不
- 一、安装 FastAPI 和uvicorn可以使用 pip 命令进行安装:pip install fastapi uvicorn二、创建&n
- 一、特效预览处理前处理后细节放大后二、程序原理1.将图片转为灰白图片后,将图片分成了三块,明、暗、阴影区域2.明区域使用空白进行填充3.阴影
- 一、excel进入密码加密算法cipher Algorithm=“AES”AES加密算法的详细介绍与
- 前端代码要做到简洁易读、高效,还要考虑后端嵌套的方便性。前段时间做了一个导航,把整个制作过程重现,希望对大家有帮助。看到这样的导航,你会怎么
- 效果图:实现代码如下view<canvas id="radar-canvas" class="radar
- python 3.10上安装pyqt5前言首先,看一下自己电脑上的python的版本,网上有太多乱七八糟的教程,啥也不说就硬教,跟着做的话就
- 在开发 vue 的时候,我们可以使用 watch 和 computed 很方便的检测数据的变化,从而做出相应的改变,但是在小程序里,只能在数
- golang sort package: https://studygolang.com/articles/3360sort 操作的对象通常
- TFRecord作为tensorflow中广泛使用的数据格式,它跨平台,省空间,效率高。因为 Tensorflow开发者众多,统一训练时数据
- Python安装新包,pip是很好用的安装工具。pip list 可以查询所有已安装的包和版本。怎么知道本地安装包的版本是否有可以更新的新版
- 今天刷二级题的时候,遇到一个问题>>> L2=[1,2,3,4]>>> L3=L2.reverse()&
- 背景在实际项目实施中,会编写很多在服务器执行的作业脚本。程序中凡是涉及到数据库链接、操作系统用户链接、IP地址、主机名称的内容都是敏感信息。
- example:models中的表# modelsclass UserInfo(models.Model): id = mode
- 本文实例讲述了Python编程之序列操作。分享给大家供大家参考,具体如下:#coding=utf8''''&
- 1、最小二乘也可以拟合二次函数我们都知道用最小二乘拟合线性函数没有问题,那么能不能拟合二次函数甚至更高次的函数呢?答案当然是可以的。下面我们
- 与ADODB和MDB2相比,PDO更高效。目前而言,实现“数据库抽象层”任重而道远,使用PDO这样的“数据库访问抽象层”是一个不错的选择。