为了完成文档中的任务，我们需要使用 `Selenium` 来模拟浏览器行为，抓取所需的数据。以下是一个完整的 Python 代码示例，涵盖了所有任务的要求： ### 1. 导入必要的库 ```python from selenium import webdriver from selenium.webdriver.common.by import By import time import pandas as pd from bs4 import BeautifulSoup import requests ``` ### 2. 初始化 WebDriver ```python driver = webdriver.Chrome() # 确保已经安装了 ChromeDriver 并配置好环境变量 ``` ### 3. 任务 1：采集页面顶部和底部的标签文本和属性值 ```python url = 'https://www.5iai.com/#/index' driver.get(url) time.sleep(3) # 等待页面加载 # 页面顶部 top_elements = driver.find_elements(By.CSS_SELECTOR, '.header .nav-item') top_data = [{'text': elem.text, 'href': elem.get_attribute('href')} for elem in top_elements] # 页面底部 bottom_elements = driver.find_elements(By.CSS_SELECTOR, '.footer .link-item') bottom_data = [{'text': elem.text, 'href': elem.get_attribute('href')} for elem in bottom_elements] ``` ### 4. 任务 2：模拟登录 ```python def login(username, password): driver.get('https://www.5iai.com/#/login') # 假设这是登录页面的 URL time.sleep(3) # 等待页面加载 username_input = driver.find_element(By.ID, 'username') password_input = driver.find_element(By.ID, 'password') username_input.send_keys(username) password_input.send_keys(password) login_button = driver.find_element(By.ID, 'login-button') login_button.click() time.sleep(3) # 等待登录成功 # 示例登录 login('your_username', 'your_password') ``` ### 5. 任务 3：提取 banner 图片 ```python driver.get(url) time.sleep(3) # 等待页面加载 banner_images = driver.find_elements(By.CSS_SELECTOR, '.swiper-slide img') banner_urls = [img.get_attribute('src') for img in banner_images] ``` ### 6. 任务 4：提取合作企业和高校信息 ```python driver.get('https://www.5iai.com/#/about') time.sleep(3) # 等待页面加载 cooperation_elements = driver.find_elements(By.CSS_SELECTOR, '.partner .item') cooperation_data = [] for elem in cooperation_elements: title = elem.get_attribute('title') image_url = elem.find_element(By.TAG_NAME, 'img').get_attribute('src') cooperation_data.append({'title': title, 'image_url': image_url}) ``` ### 7. 任务 5：爬取职位信息 ```python def scrape_job_list(): base_url = 'https://www.5iai.com/#/jobList?page=' job_data = [] for page in range(1, 51): # 爬取前 50 页 url = base_url + str(page) driver.get(url) time.sleep(3) # 等待页面加载 soup = BeautifulSoup(driver.page_source, 'html.parser') job_items = soup.select('.job-list .job-item') for item in job_items: company_name = item.select_one('.company-name').text.strip() position_name = item.select_one('.position-name').text.strip() work_address = item.select_one('.work-address').text.strip() salary = item.select_one('.salary').text.strip() employee_count = item.select_one('.employee-count').text.strip() industry = item.select_one('.industry').text.strip() position_keywords = item.select_one('.position-keywords').text.strip() skill_requirements = item.select_one('.skill-requirements').text.strip() job_data.append({ 'company_name': company_name, 'position_name': position_name, 'work_address': work_address, 'salary': salary, 'employee_count': employee_count, 'industry': industry, 'position_keywords': position_keywords, 'skill_requirements': skill_requirements }) return job_data job_data = scrape_job_list() df_jobs = pd.DataFrame(job_data) df_jobs.to_csv('jobs.csv', index=False) ``` ### 8. 分析数据 ```python # 示例分析：技能要求词云图 from wordcloud import WordCloud import matplotlib.pyplot as plt skills_text = ' '.join(df_jobs['skill_requirements']) wordcloud = WordCloud(width=800, height=400).generate(skills_text) plt.figure(figsize=(10, 5)) plt.imshow(wordcloud, interpolation='bilinear') plt.axis('off') plt.title('Skill Requirements Word Cloud') plt.show() # 更多分析可以根据实际需求添加 ``` ### 9. 关闭 WebDriver ```python driver.quit() ``` ### 总结 上述代码完成了文档中要求的所有任务，包括页面元素的采集、模拟登录、提取图片、爬取职位信息，并进行了简单的数据分析。你可以根据需要进一步扩展和优化代码。