python 爬虫 csv文档的存储和读取

python爬虫的各种存储方式之csv

2.csv的存储和读取

判断目录，有则打开，没有新建

import csv
import os
#判断目录，有则打开，没有新建
if os.path.exists('D:\Python\代码\数据爬取'):
    os.chdir('D:\Python\代码\数据爬取')
else:
    os.mkdir('D:\Python\代码\数据爬取')
    os.chdir('D:\Python\代码\数据爬取')

文件写入，需要写入列表; newline:消除空格

with open('mycsv.csv','w',encoding="utf-8",newline='') as csvfile:
    write1 = csv.writer(csvfile)
    write1.writerow(['编号','网址','昵称'])
    write1.writerows([
        [1,'https://www.baidu.com/','百度'],
        [2,'https://www.jd.com/','京东'],
        [3,'https://www.qq.com/','腾讯']
    ])
# 写入九九乘法表
with open('mycsv.csv','a',encoding="utf-8",newline='') as csvfile:
    write2 = csv.writer(csvfile)
    for i in range(1,10):
        list = []
        for j in range(1,i+1):
            num = str(j)+'x'+str(i)+'='+str(j*i)
            list.append(num)
        write2.writerow(list)

---

读取csv

with open('mycsv.csv','r',encoding="utf-8") as csvfile:  #utf-8-
    read1 = csv.reader(csvfile)  #
    print(type(read1))   #<class '_csv.reader'>
    for i in read1:   #每一行都是一个列表
        */98765**-*-print(i)

自定义读取

先额外加入一些数据

with open('mycsv.csv','a',encoding='utf-8',newline='') as csvfile:
   write = csv.writer(csvfile)
   write.writerows([
       ['上海市|普陀区|华大科创楼'],
       ['山东市|德州市|乐陵市'],
       ['四川市|成都|锦江区']
   ])

自定义读取规则

csv.register_dialect(‘任意规则名称’,delimiter=‘自定义分隔 符’,quoting=csv.QUOTE_ALL)

csv.register_dialect('mydialect',delimiter='|',quoting=csv.QUOTE_ALL)  #注册规则
#csv.unregister_dialect('mydialect')  #注销规则
with open('mycsv.csv','r',encoding='utf-8') as csvfile:
    read2 = csv.reader(csvfile)
    for i,v in enumerate(read2): #不用遍历全部，效率更高
        print(v)                  #可以换规则继续迭代
        if i == 12:
            break
    print('----------')
    readagin = csv.reader(csvfile,'mydialect')
    for each in readagin:
        print(each)