|
首先要注册一个账号密码，通过账号密码登录，并且滑块验证，自动输入搜索关键词，进行跳转翻页爬取数据，并保存到Excel文件中。
代码运行时，滑块验证经常不通过，被吃掉，但是发现打包成exe运行没有这个问题，100%成功登录。如果大家知道这个问题麻烦请与我分享，谢谢！
废话不多说直接上代码
# coding=utf-8
from selenium import webdriver
import time
from PIL import Image, ImageGrab
from io import BytesIO
from selenium.webdriver.common.action_chains import ActionChains
import os
import sys
import re
import xlwt
import urllib
import datetime
'''
用于天眼查自动登录，解决滑块验证问题
'''
# 获取项目根目录
def app_path():
if hasattr(sys, 'frozen'):
return os.path.dirname(os.path.dirname(os.path.dirname(sys.executable))) #使用pyinstaller打包后的exe目录
return os.path.dirname(__file__)
app_path = app_path()
ready_list = []
#设置表格样式
def set_style(name,height,bold=False):
style = xlwt.XFStyle()
font = xlwt.Font()
font.name = name
# font.bold = bold
font.color_index = 4
font.height = height
style.font = font
return style
# 写excel
f = xlwt.Workbook()
sheet1 = f.add_sheet('企查查数据',cell_overwrite_ok=True)
row0 = ["企业名称","法定代表人","注册资本","成立日期","电话","邮箱","地址"]
for i in range(0, len(row0)):
sheet1.write(0, i, row0[i], set_style('Times New Roman', 220, True))
# 写列
def write_col(data, row, col):
for i in range(0,len(data)):
sheet1.write(row,col,data[i],set_style('Times New Roman',220,True))
row = row + 1
def parse_save_data(all_list):
row = 1
for data in all_list:
# 公司名称
name_list = re.findall(r'(.*?)',data)
print(name_list)
# 标签
tag_list = re.findall(r'(.*)', data)
tags = []
for list in tag_list:
tag = re.findall(r'(.*?)', list)
tags.append(tag)
# print(tags)
# 法定代表人
legal_list = re.findall(r'', captcha1.size)
captcha1.save('captcha1.png')
# 截取第二张图片(有缺口的)
driver.find_element_by_xpath('/html/body/div[10]/div[2]/div[2]/div[2]/div[2]').click()
time.sleep(4)
img1 = driver.find_element_by_xpath('/html/body/div[10]/div[2]/div[2]/div[1]/div[2]/div[1]')
time.sleep(0.5)
location1 = img1.location
size1 = img1.size
top1, bottom1, left1, right1 = location1['y'], location1['y'] + size1['height'], location1['x'], location1['x'] + \
size1['width']
screenshot = driver.get_screenshot_as_png()
screenshot = Image.open(BytesIO(screenshot))
captcha2 = screenshot.crop((left1, top1, right1, bottom1))
captcha2.save('captcha2.png')
# 获取偏移量
left = 55 # 这个是去掉开始的一部分
for i in range(left, captcha1.size[0]):
for j in range(captcha1.size[1]):
# 判断两个像素点是否相同
pixel1 = captcha1.load()[i, j]
pixel2 = captcha2.load()[i, j]
threshold = 60
if abs(pixel1[0] - pixel2[0]) < threshold and abs(pixel1[1] - pixel2[1]) < threshold and abs(
pixel1[2] - pixel2[2]) < threshold:
pass
else:
left = i
print('缺口位置', left)
# 减去缺口位移
left -= 52
# 开始移动
track = get_track(left)
print('滑动轨迹', track)
# track += [5,4,5,-6, -3,5,-2,-3, 3,6,-5, -2,-2,-4] # 滑过去再滑过来，不然有可能被吃
# 拖动滑块
slider = driver.find_element_by_xpath('/html/body/div[10]/div[2]/div[2]/div[2]/div[2]')
ActionChains(driver).click_and_hold(slider).perform()
for x in track:
ActionChains(driver).move_by_offset(xoffset=x, yoffset=0).perform()
time.sleep(0.2)
ActionChains(driver).release().perform()
time.sleep(1)
try:
if driver.find_element_by_xpath('/html/body/div[10]/div[2]/div[2]/div[2]/div[2]'):
print('能找到滑块，重新试')
# driver.delete_all_cookies()
# driver.refresh()
# autologin(driver, account, password)
else:
print('login success')
except:
print('login success')
time.sleep(0.2)
driver.find_element_by_xpath('.//input[@id="home-main-search"]').send_keys(keywords)
driver.find_element_by_xpath('.//div[@class="input-group home-group"]/div[1]').click()
# 爬数据
data = driver.find_element_by_xpath('.//div[@class="result-list sv-search-container"]').get_attribute('innerHTML')
count = count + 1
# 添加待解析数据
ready_list.append(data)
while count < page:
# 点击下一页
# driver.find_element_by_xpath('./ul[@class="pagination"]]/li/a[@class="num -next"]').click()
url = 'https://www.tianyancha.com/search/p{}?key={}'.format(count + 1,urllib.parse.quote(keywords))
driver.get(url)
time.sleep(2)
data = driver.find_element_by_xpath('.//div[@class="result-list sv-search-container"]').get_attribute('innerHTML')
count = count + 1
ready_list.append(data)
# 解析并写数据
parse_save_data(ready_list)
print('获取数据完毕')
# if __name__ == '__main__':
# driver_path = 'C:/Program Files (x86)/Google/Chrome/Application/chromedriver.exe'
# chromeoption = webdriver.ChromeOptions()
# chromeoption.add_argument('--headless')
# chromeoption.add_argument('user-agent='+user_agent)
keywords = input('请输入关键词:')
account = input('请输入查天眼账号:')
password = input('请输入查天眼密码:')
page = int(input('请输入获取页数:'))
driver = webdriver.Chrome()
driver.maximize_window()
driver.implicitly_wait(10)
print('开始获取数据。。。')
autologin(account, password)
打包成exe（注意site-packages要换成自己python包的目录）
pyinstaller main.py -p D:\Anaconda3\Lib\site-packages
最终运行dist目录下的exe
注意事项
由于天眼查没有开会员只能查看到4页内容，所以需要开会员，这个想要绕过就需要另外去研究，毕竟是要充钱付费，破解也没那么简单
欢迎大家一起交流分享！！！
|