Python使用Selenium将网页保存为图片或PDF

pip install selenium pillow

from selenium import webdriver
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.options import Options
from PIL import Image
import io

# 设置Chrome选项
chrome_options = Options()
chrome_options.add_argument("--headless")  # 无头模式，不打开浏览器窗口
chrome_options.add_argument("--disable-gpu")

# 指定ChromeDriver的路径
service = Service(executable_path='C:\\chromedriver-win64\\chromedriver.exe')

# 创建WebDriver对象
driver = webdriver.Chrome(service=service, options=chrome_options)

# 访问目标网页
driver.get('https://blog.csdn.net/a497785609')

# 获取页面的实际高度和宽度
width = driver.execute_script("return document.documentElement.scrollWidth")
height = driver.execute_script("return document.documentElement.scrollHeight")

# 设置窗口大小
driver.set_window_size(width, height)

# 截图并保存
driver.save_screenshot('C:\\chromedriver-win64\\screenshot.png')

# 使用Pillow处理图像（可选）
image = Image.open(io.BytesIO(driver.get_screenshot_as_png()))
image.show()

# 关闭浏览器
driver.quit()

from selenium import webdriver
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.options import Options
from PIL import Image
import io
import base64

# 设置Chrome选项
chrome_options = Options()
chrome_options.add_argument("--headless")  # 无头模式，不打开浏览器窗口
chrome_options.add_argument("--disable-gpu")

# ChromeDriver路径
ChromeDriver = 'C:\\chromedriver-win64\\chromedriver.exe'

# 访问目标网页
url = 'https://blog.csdn.net/a497785609'

# 保存PDF文件的路径
file_name = "C:\\test.pdf"

# 指定ChromeDriver的路径
service = Service(executable_path=ChromeDriver)

# 创建WebDriver对象
driver = webdriver.Chrome(service=service, options=chrome_options)

# 访问目标网页
driver.get(url)

# 获取页面的实际高度和宽度
width = driver.execute_script("return document.documentElement.scrollWidth")
height = driver.execute_script("return document.documentElement.scrollHeight")

# 设置窗口大小
driver.set_window_size(width, height)

# 【打印为PDF】
# 调用 Chrome DevTools 的 printToPDF 方法，打印参数参考：https://chromedevtools.github.io/devtools-protocol/tot/Page/#method-printToPDF
pdf_data = driver.execute_cdp_cmd("Page.printToPDF", {"landscape": False,
                                                      "displayHeaderFooter": False,
                                                      "printBackground": False,
                                                      "scale": 1.0,                # 必须是 float
                                                      "paperWidth": 8.5,           # A4纸宽度（英寸）
                                                      "paperHeight": 11.0,         # A4纸高度（英寸）
                                                      "marginTop": 0.5,
                                                      "marginBottom": 0.5,
                                                      "marginLeft": 0.5,
                                                      "marginRight": 0.5})
# 解码并写入 PDF 文件
with open(file_name, "wb") as f:
    f.write(base64.b64decode(pdf_data['data']))

print(" 已保存到文件："+file_name)

# 关闭浏览器
driver.quit()