0


Selenium控制已运行的Edge和Chrome浏览器——在线控制 | 人机交互(详细启动步骤和bug记录)

文章目录


说明:本记录是在Windows系统上执行的!
起因是:博导要求统计一下国内某个领域的专家情况,统计主持国家自然科学基金的副教授和教授都有哪些大牛!
于是:本人去[NSFC]:https://kd.nsfc.cn/ 下载全部的历史基金项目书。。。。工作量太大就……半自动化实现吧!!!


前期准备

1. python Selenium库
2. Edge浏览器 或 Chrome浏览器

1. 浏览器开启远程控制指令

  1. 无论是哪种浏览器,都需要使用终端独立运行浏览器的远程调试模式。
  2. 开启方式:加入指令(–remote-debugging-port=9222 --user-data-dir=“D:\selenium\AutomationProfile”)

需要进入目标浏览器的根目录! 不然就输入全路径!

(1)Edge

.\msedge.exe --remote-debugging-port=9222--user-data-dir=“D:\selenium\AutomationProfile”

(2)Chrome

.\chrome.exe --remote-debugging-port=9222--user-data-dir=“D:\selenium\AutomationProfile”

在这里插入图片描述

2. 执行python代码

(1)先启动浏览器后执行代码

  • 必须是先执行上述步骤,开启了浏览器的远程调试端口后,才能通过下方代码进行控制。
  • add_experimental_option("debuggerAddress", "127.0.0.1:9222") 这句话是关键!
from selenium import webdriver
from selenium.webdriver.edge.options import Options

classTest:defedge(self):
        edge_driver_path = executable_path=r'C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe'
        chrome_options = Options()# chrome_options.binary_location = edge_driver_path  #  传入驱动地址
        chrome_options.add_experimental_option("debuggerAddress","127.0.0.1:9222")# "127.0.0.1:9222"其中,9222是浏览器的运行端口# 让浏览器带着这个配置运行# chrome_options.add_experimental_option('detach', True)  # 通过option参数,设置浏览器不关闭
        driver = webdriver.Edge(options=chrome_options, keep_alive=True)
        driver.implicitly_wait(10)# 页面元素查找的等待时间
        self.driver = driver
        passdefchrome_drive(self, drive='chrome'):
        edge_driver_path = executable_path =r'D:\Program Files\Google\Chrome\Application'if drive =='chrome':
            chrome_options = webdriver.ChromeOptions()# chrome_options.binary_location = edge_driver_path    #  传入驱动地址# chrome_options.add_experimental_option('detach', True)  # 通过option参数,设置浏览器不关闭
            chrome_options.add_experimental_option("debuggerAddress","127.0.0.1:9222")
            driver = webdriver.Chrome(options=chrome_options, keep_alive=False)
            driver.implicitly_wait(10)# 页面元素查找的等待时间
            self.driver = driver
            pass

(2)通过代码启动浏览器

  • 这个时候被注释掉的 .binary_location = edge_driver_path 是关键!
  • 这种情况下,需要下载对应的驱动软件(.exe)
  • 博主在笔记本电脑上首次尝试Selenium时就下载了驱动软件!但后来在台式电脑使用相同代码时发现,压根不需要下载什么驱动软件!
  • 只需要使用终端提前启动浏览器的调试模型即可。 (这是弯路、坑)
  • 因为,如果是通过代码启动浏览器的调试模型,需要配置路径,然后保证程序关闭后浏览器依旧运行!麻烦!!!

(3)Bug问题记录

1)python可读取浏览器所有标签标题,但检索网页元素失败

  • 部分网页不支持爬取!特别是当网页开启F12的开发人选项后,会出现无法查找元素的问题。
  • 此时,关闭 “开发人选项” 即可。

2)浏览器开启程序,但python程序无法链接浏览器进行自动控制

  • 关闭原有浏览器,重新打开浏览器(需搭配命令:–remote-debugging-port=9222 --user-data-dir=“xxx folder”

3. 爬取效果

![(https://img-blog.csdnimg.cn/direct/492fe54da6c24809a654191a43365f14.png)

3. 完整代码共享

以下代码主要实现了:

  • 浏览器标签页的翻动和选择
  • 爬取 – 青塔网检索”国家自然科学基金项目“的作者信息,并保存到表格。
  • 爬取 – NSFC”国家自然科学基金项目“的作者信息,并保存到表格。
  • 爬取 – 国际某个领域专家的作者信息,并保存到表格。

3.1 包含Excel部分的完整代码

包含Excel部分的完整代码见:资源文件

3.2 爬虫部分的完整代码

import os
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.common.by import By
import time
from selenium.webdriver.edge.options import Options
from selenium.webdriver.common.action_chains import ActionChains

# '.\chrome.exe --remote-debugging-port=9222 --user-data-dir=“D:\selenium\AutomationProfile”             n "*" --ws --allow-insecure-unlock --nodiscover --authrpc.addr 127.0.1.2 --authrpc.port 8545'# '.\chrome.exe --remote-debugging-port=9222 --user-data-dir=“D:\selenium\AutomationProfile”'classWeb_Browser:def__init__(self, drive='chrome'):
        self.driver =None# self.edge()
        self.chrome_drive()defedge(self):# edge_driver_path = executable_path=r'D:\Program Files\Google\Chrome\Application\chromedriver.exe'
        edge_driver_path = executable_path=r'C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe'
        chrome_options = Options()# chrome_options.binary_location = edge_driver_path# 配置浏览器# 添加User-Agent到Chrome选项中# chrome_options.add_argument("--user-agent=windows 10 Edge")# "127.0.0.1:9222"其中,9222是浏览器的运行端口
        chrome_options.add_experimental_option("debuggerAddress","127.0.0.1:9222")# 让浏览器带着这个配置运行# chrome_options.add_experimental_option('detach', True)  # 通过option参数,设置浏览器不关闭
        driver = webdriver.Edge(options=chrome_options, keep_alive=True)# driver = webdriver.Chrome( options=chrome_options)print('===================')# driver.get('www.baidu.com')
        driver.implicitly_wait(10)
        self.driver = driver

    defchrome_drive(self, drive='chrome'):
        edge_driver_path = executable_path =r'D:\Program Files\Google\Chrome\Application\chromedriver.exe'if drive =='chrome':
            chrome_options = webdriver.ChromeOptions()# chrome_options.binary_location = edge_driver_path# chrome_options.add_experimental_option('detach', True)  # 通过option参数,设置浏览器不关闭
            chrome_options.add_experimental_option("debuggerAddress","127.0.0.1:9222")

            driver = webdriver.Chrome(options=chrome_options, keep_alive=False)
            self.driver = driver
            driver.implicitly_wait(10)
        self.opened_windows_dict =Nonepassdefget_all_opened_windows(self):
        driver = self.driver
        cw = driver.current_window_handle
        res ={}# 获取已打开的标签页的信息
        tabs = driver.window_handles
        for t in tabs:
            driver.switch_to.window(t)
            res[str(driver.title)]=str(t)
        self.opened_windows_dict = res
        driver.switch_to.window(cw)print('已打开的标签页的信息:',)for k in res:print(f"\t{k}: {res[k]}")return res

    defswitch_window(self, key):
        driver = self.driver
        cw = driver.current_window_handle
        # 获取已打开的标签页的信息
        tabs = driver.window_handles
        for t in tabs:
            driver.switch_to.window(t)if key instr(driver.title): cw = t
            break# driver.switch_to.window(cw)
        self.driver = driver
        passdefopen_new_window(self, driver=None, url=None, delay_t=0.6):'''# 打开新标签页'''
        driver = self.driver ifnot driver else driver
        old_handle = driver.window_handles  # 获取已打开的标签页的信息# driver.find_element("body").send_keys(Keys.CONTROL + 't')  # 没有实体会报错# driver.execute_script("window.open('','_blank');")  # 可能被拦截
        driver.switch_to.new_window('tab')
        time.sleep(delay_t)iflen(driver.window_handles)>len(old_handle):returnTrue
        driver.execute_script(f"window.open('{url if url else''}');")
        time.sleep(delay_t)iflen(driver.window_handles)>len(old_handle):returnTruereturnFalsedeffunc1(self, xlsx):""" 学术网 """for p inrange(50):# self.switch_window('故障诊断')
            driver = self.driver
            web = driver.find_element(by=By.XPATH, value='//*[@id="search_body"]/div[2]/div[3]/div[1]/div[2]/div[1]/div[3]/div[2]/div/div[2]/div[2]/div/div')
            web1 = web.find_elements(by=By.CLASS_NAME, value='inner-content')print('web1 len=',len(web1))
            num =0for i, w inenumerate(web1):try:# '//*[@id="search_body"]/div[2]/div[3]/div[1]/div[2]/div[1]/div[3]/div[2]/div/div[2]/div[2]/div/div'#
                    a = w.find_element(by=By.XPATH, value=f'//div[{1+i}]/div/div[2]/div[1]/div[1]/div/a/strong/span/span').text
                    try:
                        b = w.find_element(by=By.XPATH, value=f'//div[{1+ i}]/div/div[2]/div[3]/p[2]').text
                        school =str(b).split(',')for s in school:if'university'in s.lower(): b = s[1:]except: b =None
                    c = w.find_element(by=By.XPATH, value=f'//div[{1+ i}]/div/div[2]/div[3]/p[1]').text
                    d =None
                    e =None
                    f =Nonetry:
                        h_index = w.find_element(by=By.XPATH, value=f'//div[{1+ i}]/div/div[2]/div[2]/div/span[1]/span[3]').text
                        paper = w.find_element(by=By.XPATH, value=f'//div[{1+ i}]/div/div[2]/div[2]/div/span[2]/span[3]').text
                        cite = w.find_element(by=By.XPATH, value=f'//div[{1+ i}]/div/div[2]/div[2]/div/span[3]/span[3]').text
                        f =f"H-index: {h_index},  papers: {paper}, cites: {cite}"except:pass

                    g =None
                    h = w.find_element(by=By.XPATH, value=f'//div[{1+ i}]/div/div[2]/div[1]/div[1]/div/a')
                    h ='https://www.aminer.cn/'+ h.get_attribute('href')print(a, b ,c, g)
                    xlsx.input_data(a,b,c,d,e,f,g, h)
                    num +=1except:passprint('记录:', num)# aa = driver.find_elements(by=By.XPATH, value='//*[@id="search_body"]/div[2]/div[3]/div[1]/div[2]/div[1]/div[3]/div[2]/div/div[2]/div[3]/ul/li')# aa = aa[-1]
            aa = driver.find_element(by=By.CLASS_NAME, value='ant-pagination-next')# v = '#search_body > div.ant-tabs.ant-tabs-top.a-aminer-core-search-index-searchPageTab.ant-tabs-line.ant-tabs-no-animation > div.ant-tabs-content.ant-tabs-content-no-animated.ant-tabs-top-content > div.ant-tabs-tabpane.ant-tabs-tabpane-active > div.a-aminer-core-search-index-componentContent > div.a-aminer-core-search-c-search-component-temp-searchComponent > div.view > div:nth-child(2) > div > div:nth-child(2) > div.paginationWrap > ul > li.ant-pagination-next'# aa = driver.find_element(by=By.CSS_SELECTOR, value=v)# 创建一个ActionChains对象,用于执行鼠标动作
            action_chains = ActionChains(driver)# 将鼠标移动到链接元素上并点击
            action_chains.move_to_element(aa).click().perform()print(f'第{p+1}页 --> 第{p+2}页')try:
                xlsx.make_frame()
                xlsx.save_excel()except:pass
            time.sleep(5)passdeffunc2(self, xlsx=None):for p inrange(50):
            self.switch_window('青塔')
            driver = self.driver
            web = driver.find_element(by=By.XPATH,
                                      value='//*[@id="app"]/div[2]/div[1]/div/div[2]/div[2]/div/div[2]')

            web1 = web.find_elements(by=By.CLASS_NAME, value='list-item')print('web1 len=',len(web1))
            num =0for i, w inenumerate(web1):# try:# //*[@id="app"]/div[2]/div[1]/div/div[2]/div[2]/div/div[2]# '//*[@id="app"]/div[2]/div[1]/div/div[2]/div[2]/div/div[2]/div/div[2]/div[2]/div[2]/div[1]/div[2]'# //*[@id="app"]/div[2]/div[1]/div/div[2]/div[2]/div/div[2]/div/div[1]/div[2]/div[2]/div[1]/div[1]

                b = w.find_element(by=By.XPATH, value=f'//div[2]/div[1]/div[1]/div[2]')print(b)
                b = b.text
                print('b=', b)
                a = w.find_element(by=By.XPATH, value=f'//div[2]/div[2]/div[1]/div[2]').text
                print('a=', a)
                c =None
                d =None
                e = w.find_element(by=By.XPATH, value=f'//div[1]/div[1]').text
                print('e=', e)
                year = w.find_element(by=By.XPATH, value=f'//div[2]/div[2]/div[2]/div[2]').text
                money = w.find_element(by=By.XPATH, value=f'//div[2]/div[1]/div[2]/div[2]').text
                print('year=', year,'money=', money)
                e =f"{e}, 立项: {year}, 资助: {money}"
                jijin = w.find_element(by=By.XPATH, value=f'//div[2]/div[3]/div[1]/div[2]').text
                domain = w.find_element(by=By.XPATH, value=f'//div[2]/div[3]/div[2]/div[2]').text
                print('jijin=',jijin,'domain=', domain)
                f =f"{jijin}, 领域: {domain}"
                g =None
                h =Noneprint(i,'-----------', i)print(a, b, c, d, e, f)
                xlsx.input_data(a, b, c, d, e, f, g, h)
                num +=1break# except: passprint('记录:', num)break
            aa = driver.find_element(by=By.XPATH, value=f'//*[@id="app"]/div[2]/div[1]/div/div[2]/div[2]/div/div[3]/button[2]')# 创建一个ActionChains对象,用于执行鼠标动作
            action_chains = ActionChains(driver)# 将鼠标移动到链接元素上并点击
            action_chains.move_to_element(aa).click().perform()print(f'第{p +1}页 --> 第{p +2}页')try:
                xlsx.make_frame()
                xlsx.save_excel()except:pass
            time.sleep(5)passdeffunc3(self, xlsx=None):for p inrange(50):
            self.switch_window('大数据知识管理服务门户')
            driver = self.driver
            d = driver.find_element(by=By.CLASS_NAME, value='container_list_right')print('d==', d)# web = driver.find_element(by=By.XPATH,#                           value='//*[@id="app"]/div[1]/div[3]/div/div[3]/div[1]/div')
            web = d.find_element(by=By.XPATH, value='//div[1]/div')# web1 = web.find_elements(by=By.CLASS_NAME, value='list-item')# print('web1 len=', len(web1))
            num =0for i, w2 inenumerate(range(6)):
                w = web
                try:# //*[@id="app"]/div[1]/div[3]/div/div[3]/div[1]/div# //*[@id="app"]/div[1]/div[3]/div/div[3]# //*[@id="app"]/div[1]/div[3]/div/div[3]/div[1]/div/div[2]/div[2]/div[1]

                    b = w.find_element(by=By.XPATH, value=f'//div[{i+1}]/div[3]/div[4]/a')
                    b = b.text
                    # print('b=', b)
                    a = w.find_element(by=By.XPATH, value=f'//div[{i+1}]/div[2]/div[4]/a').text
                    # print('a=', a)
                    c =None
                    d =None
                    e = w.find_element(by=By.XPATH, value=f'//div[{i+1}]/div[1]/div[1]/p/a').text
                    # print('e=', e)
                    year = w.find_element(by=By.XPATH, value=f'//div[{i+1}]/div[3]/div[3]').text
                    money = w.find_element(by=By.XPATH, value=f'//div[{i+1}]/div[3]/div[1]').text
                    # print('year=', year, 'money=', money)
                    e =f"{e}, {year}, {money}"
                    jijin = w.find_element(by=By.XPATH, value=f'//div[{i+1}]/div[2]/div[3]').text
                    domain = w.find_element(by=By.XPATH, value=f'//div[{i+1}]/div[2]/div[1]').text
                    # print('jijin=',jijin,  domain)
                    f =f"{jijin}, {domain}"
                    g =None
                    h =Noneprint(i+1,'-----------', i+1)print(a, b, c, d, e, f)
                    xlsx.input_data(a, b, c, d, e, f, g, h)
                    num +=1# breakexcept:passprint('记录:', num)# break# aa = driver.find_element(by=By.CLASS_NAME, value=f'btn-next')# # 创建一个ActionChains对象,用于执行鼠标动作# action_chains = ActionChains(driver)# # 将鼠标移动到链接元素上并点击# action_chains.move_to_element(aa).click().perform()print(f'第{p +1}页 --> 第{p +2}页')try:
                xlsx.make_frame()
                xlsx.save_excel()except:passbreak# time.sleep(5)passdeffunc4(self, xlsx=None, key='Google2'):if key =='Google': self.switch_window('Google')else: self.switch_window('必应')
        driver = self.driver
        data = xlsx.read_excel()# print(data['姓名'])for i, name inenumerate(data['姓名']):
            school = data['学校'][i]
            text =f'{school}{name}是不是教授'print(f'search [{i+1}]:  {name} -》 ', text)if key =='Google': web = driver.find_element(by=By.XPATH, value='//*[@id="APjFqb"]')else: web = driver.find_element(by=By.XPATH, value='//*[@id="sb_form_q"]')
            web.clear()
            web.send_keys(text)if key =='Google': web = driver.find_element(by=By.XPATH, value='//*[@id="tsf"]/div[1]/div[1]/div[2]/button')else: web = driver.find_element(by=By.XPATH, value='//*[@id="sb_form_go"]')# try:
            web.click()# except: pass
            time.sleep(5)
        num =0if __name__ =='__main__':from temp import Make_Excel, input_data_list, input_data
    xlsx = Make_Excel()
    web = Web_Browser()
    web.get_all_opened_windows()# web.switch_window('故障诊断')''' 学术网 '''
    web.func1(xlsx)# 学术网# web.func2(xlsx)  #  青塔网# web.func3(xlsx)  #  NSFC官网# web.func4(xlsx, )    # goole搜索网# xlsx.make_frame()# xlsx.save_excel()pass
标签: selenium edge chrome

本文转载自: https://blog.csdn.net/tjb132/article/details/136655986
版权归原作者 biao169 所有, 如有侵权,请联系我们删除。

“Selenium控制已运行的Edge和Chrome浏览器——在线控制 | 人机交互(详细启动步骤和bug记录)”的评论:

还没有评论