首先登陆淘宝获取cookie
import os
import easygui as g
from selenium import webdriver
import requests
import json
class Spider(object):
def __init__(self):
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.web.maximize_window() #浏览器最大化
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):
self.web.get('https://mai.taobao.com/')
# i = input('请确认是否已登录?[y/n]:')
if g.ccbox("登录好了吗?", choices=("是", "否")):
i = '1'
else:
sys.exit(0)
if (i != '1'):
return
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
if __name__ == '__main__':
sp = Spider()
cookies = sp.login()
jsonCookies = json.dumps(cookies)
with open('cookies.json', 'w') as f:
f.write(jsonCookies)
然后进入订单页面 爬取写入excel
import json
import re
import requests
import easygui as g
from openpyxl import load_workbook
import xlrd
import datetime
def cookie_json(cookies):
# 加载cookie
print(cookies)
s = requests.session()
s.verify = False
s.headers = {
"User-Agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 11_0 like Mac OS X) AppleWebKit/604.1.38 (KHTML, like Gecko) Version/11.0 Mobile/15A372 Safari/604.1"}
s.get("https://mai.taobao.com/")
#进入页面
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
# r = s.get(url, cookies=cookies) # 进入网页
# r = r.content # 得到网页源代码
# fp = open("test.html", "w+b") # 打开一个文本文件
# fp.write(r) # 写入网页源代码
# fp.close() # 关闭文件
html = s.get(url, cookies=cookies) # 进入网页
#print(html.text)
# 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)
if len(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]
}
print(json)
return json
else:
print('你的登录已过期!')
g.msgbox('你的登录已过期!')
def excel(wbdz,yf,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
if __name__ == '__main__':
#获取cookies.json文档处理
with open("cookies.json", "r") as fp:
cookies = json.load(fp)
#将cookies赋值给变量
json = cookie_json(cookies)
wbdz = 'C:\\Users\\Administrator\\Desktop\\18年账本.xlsx' # 表格地址
yf = "12月" #表格页名字
excel(wbdz,yf,json)