第三次作业

# 通过截取前10页的评论，看看买kindle paperwhite 4的人中 ，买32G的人多，还是买8G的人多

import requests

import json

import pymysql

import time

conn = pymysql.connect(host='localhost', user='root', passwd='123456', db='mydb', port=3306, charset='utf8') #连接数据库

cursor = conn.cursor()

headers = {

    'Referer':'https://item.jd.com/100000667370.html',

    'User-Agent':'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/74.0.3729.169 Mobile Safari/537.36'

}

urls = ['https://sclub.jd.com/comment/productPageComments.action?productId=100000667370&score=0&sortType=5&page={}&pageSize=10'.format(str(i)) for i in range(0,10)]

for url in urls:

    # print(url)

    res = requests.get(url,headers=headers)

    # print(res.text)

    json_data = json.loads(res.text)

    comments = json_data['comments']

    for comment in comments:

        content = comment['content']

        creationTime = comment['creationTime']

        nickname = comment['nickname']

        productColor = comment['productColor']

        productSize = comment['productSize']

        score = comment['score']

        # print(content,creationTime,nickname,productColor,productSize,score)

        cursor.execute("insert into kindle_paperwhite4 (content,creationTime,nickname,productColor,productSize,score) values(%s,%s,%s,%s,%s,%s)",

                      (content,creationTime,nickname,productColor,productSize,score))

        conn.commit()

    time.sleep(4)

select_sql = "SELECT productColor,productSize,count(*) from kindle_paperwhite4 group by productColor,productSize"

cursor.execute(select_sql)

result = cursor.fetchall()

cursor.close()

conn.close()

print(result)