如何用python清洗文件中的数据_Python

简单版

直接打开日志文件，往另外一个文件中按照要过滤的要求进行过滤

				?

									import io;

									with open('a.txt', 'w') as f:    

									    for line in open('c:/201509.txt'):    

									        if line.find('更改项目')>0 and line.find('500')>0:  

									            f.write(line+"\n");

									print("输出完成");

注意.find返回的是字符串在目标的第几位，要和0作比较另外使用and而不是&&作为"和"，使用or而不是||作为"或" w是写，r是读，a是追加

使用filter

				?

									import io;

									def isData(s):

									    return s.find('更改项目')>0 and s.find('500')>0;

									with open('a.txt', 'w') as f:             

									       list1=list(filter(isData,open('c:/201509.txt')));    

									       for (offset,item) in enumerate(list1):

									            f.write(str(offset)+":"+item);

读取utf-8带bom的文件

微软会在在 UTF-8 文件中放置 BOM头（顺便提一下：把带有 BOM 的小端序 UTF-16 称作「Unicode」而又不详细说明，这也是微软的习惯。不含BOM的UTF-8才是标准形式，UTF-8不需要BOM，带BOM的UTF-8文件的开头会有U+FEFF，所以Windows新建的空文件会有3字节的大小。

				?

									import codecs

									with codecs.open('c:/20160907205.log', encoding='utf_8_sig') as f:

									    for line in f:

									        print(line)

注意编码格式是utf_8_sig

多文件清洗

对多个文件进行过滤，可以借助其名称的规律，遍历文件之后

				?

									import codecs

									with codecs.open('a.txt','a', encoding='utf_8_sig') as f:

									    for i in range(205,210):

									        f.write(str(i)+"\r\n");   

									        print(str(i)); 

									        for line in open('c:/20160907'+str(i)+'.log', encoding='utf_8_sig'): 

									            if line.find('url为')>=0 :  

									                print(line);

									                f.write(line+"\r\n");

									print("输出完成");

清洗数据同时记录订单号并排序

				?

									import codecs

									a=0;

									List=[];

									with codecs.open('a.txt','a', encoding='utf_8_sig') as f:    

									    for i in range(205,210): 

									        for line in open('c:/20160907'+str(i)+'.log', encoding='utf_8_sig'): 

									            if line.find('url为')>=0 :  

									                ind=line.find("XFLucky");                

									                if ind>=0:

									                    nums=line[ind:ind+22];                    

									                    print(nums);

									                    List.append(nums);                   

									                a=a+1;

									                print(line);

									                f.write(str(i)+line+"\r\n");              

									List.sort(); 

									for item in List:

									    print(item);          

									print("输出完成"+str(a));

清洗sql文件，将数据表名放入excel中

安装openpyxl

				?

									pip install openpyxl

安装之后就可以进行sql建表语句的过滤了，将所有的表名和注释写入我们的excel文件中。

				?

									import re

									import openpyxl

									data = []

									temp = []

									wb = openpyxl.load_workbook('data.xlsx')

									ws2 = wb.create_sheet(index=2, title='addSheet_test')

									for line in open('wlzcool.sql', encoding='utf-8'):

									    if line.find('CREATE TABLE') >= 0:

									        matchObj1 = re.search('`(.*?)`', line, re.M | re.I)

									        if matchObj1:

									            # print("matchObj.group(1) : ", matchObj1.group(1))

									            print(matchObj1.group(1))

									            temp.append(matchObj1.group(1))

									    if line.find('ROW_FORMAT = Dynamic') >= 0:

									        matchObj2 = re.search('\'(.*?)\'', line, re.M | re.I)

									        if matchObj2:

									            # print("matchObj.group(1) : ", matchObj2.group(1))

									            print(matchObj2.group(1))

									            temp.append(matchObj2.group(1))

									        else:

									            print("no comment")

									            temp.append("no comment")

									        data.append(temp)

									        temp = []

									for row in data:

									    ws2.append(row)

									wb.save('data.xlsx')

									print("输出完成")