360极速浏览器一键操作 订单速记表selenium +requests爬取 openpyxl +xlrd 操作excel

#coding:utf-8
#配置前先将谷歌的chromedriver放到 360浏览器的目录下
from selenium.webdriver.common.by import By   #引入判断元素加载模块
from selenium.webdriver.support.ui import WebDriverWait    #引入判断元素加载模块
from selenium.webdriver.support import expected_conditions as EC    #引入判断元素加载模块
from selenium.webdriver.chrome.options import Options
import easygui as g
from selenium import webdriver
import requests
import json
from openpyxl import load_workbook
import xlrd
import datetime
import re


class Spider(object):
    def __init__(self):
        __browser_url = r'D:\360极速浏览器\360Chrome\Chrome\Application\360chrome.exe'  # 浏览器目录地址
        chrome_options = Options()
        chrome_options.binary_location = __browser_url
        self.web = webdriver.Chrome(chrome_options=chrome_options)
        self.web.maximize_window()  # 浏览器最大化
        # chromedriver = "C:\Program Files (x86)\Google\Chrome\Application\chromedriver.exe"
        # # 这里的driver就是刚刚上面下载的
        # os.environ["webdriver.chrome.driver"] = chromedriver
        # self.web = webdriver.Chrome(chromedriver)

        self.headers = {
            'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8',
            'accept-encoding': 'gzip, deflate, br',
            'accept-language': 'zh-CN,zh;q=0.8,en;q=0.6',
            'cache-control': 'max-age=0',
            'upgrade-insecure-requests': '1',
            'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.78 Safari/537.36'
        }
        self.req = requests.Session()
        self.cookies = {}

    def login(self):
        wait = WebDriverWait(self.web, 10)  # 设置目标元素等待时间
        # 设定页面加载限制时间
        self.web.set_page_load_timeout(10)
        self.web.set_script_timeout(10)  # 这两种设置都进行才有效
        self.web.get('https://mai.taobao.com/')# 打开请求的url
        SubmitQuick = wait.until(EC.element_to_be_clickable((By.ID,'J_SubmitQuick'))) # 等待千牛快速登录加载完成
        SubmitQuick.click() #点击千牛快速登录
        # self.web.find_element_by_id('J_SubmitQuick').click() #点击千牛快速登录
        self.web.get('https://mai.taobao.com/')
        cookie = ''
        for elem in self.web.get_cookies():
            cookie += elem["name"] + "=" + elem["value"] + ";"
            if elem["name"] == '_tb_token_':
                self.token = elem["value"]
        self.cookies = cookie
        self.headers['Cookie'] = self.cookies
        print(self.headers)
        self.web.quit()
        return self.headers

    def DingDan(self):
        cookies =self.headers
        print(cookies)
        s = requests.session()
        dingdan = g.enterbox(msg="请订单号", title="订单号")
        url = 'https://trade.tmall.com/detail/orderDetail.htm?spm=a1z09.1.0.0.60663606YAPWZH&bizOrderId=' + dingdan
        html = s.get(url, cookies=cookies)  # 进入网页

        # 正则表达式
        pattern = re.compile(
            '<script.*?detailData.*?订单总价.*?应开票给消费者的金额￥(.*?)元.*?"type":"mixture".*?"basic":.*?"text":(.*?),(.*?),.*?"key":"收货地址".*?class.*?J_WangWang.*?data-nick=(.*?)data-icon',
            re.S)
        items = re.findall(pattern, html.text)
        print(items)
        # 数据转换成字典
        for item in items:
            json = {
                '订单号': dingdan,
                '金额': item[0],
                '名字': item[1],
                '电话': item[2],
                '旺旺': item[3]
            }
        self.json = json
        print(self.json)
        return self.json

    def excel(self,wbdz, yf):
        json =self.json
        work_book = xlrd.open_workbook(wbdz)  # 打开表格
        sheet3 = work_book.sheet_by_name(yf)
        hh = sheet3.nrows  # 获取表格总行数
        hh = hh + 1
        # print(hh)

        订单 = 'J' + str(hh)
        金额 = 'K' + str(hh)
        费用 = 'L' + str(hh)
        账号 = 'I' + str(hh)
        姓名 = 'O' + str(hh)
        电话 = 'P' + str(hh)
        日期 = 'A' + str(hh)

        # openpyxl操作表格
        wb = load_workbook(wbdz)  # 打开表格
        sheet = wb.get_sheet_by_name(yf)
        # print(sheet["C4"].value)    # c4     <-第C4格的值
        # for i in sheet["C"]:
        #   print(i.value, end=" ")    # c1 c2 c3 c4 c5 c6 c7 c8 c9 c10     <-C列中的所有值
        旺旺 = json['旺旺']
        旺旺 = re.sub("'", '', 旺旺)  # 用空字符替换字符串里面的双引号

        sheet[日期] = datetime.date.today()
        sheet[订单] = json['订单号']  # 输入内容
        sheet[金额] = float(json['金额'])
        sheet[费用] = 15
        sheet[账号] = 旺旺
        sheet[姓名] = json['名字']
        sheet[电话] = json['电话']

        # print(sheet[订单].value)  # c4     <-第C4格的值
        wb.save(wbdz)  # 保存表格
        g.msgbox('已登记！')
        print("已登记")

    pass

if __name__ == '__main__':
    sp = Spider()
    cookies = sp.login()
    sp.DingDan()
    jsonCookies = json.dumps(cookies)
    wbdz = 'self.xlsx'  # 表格地址
    yf = "12月"  # 表格页名字
    sp.excel(wbdz, yf)
    with open('cookies.json', 'w') as f:
        f.write(jsonCookies)