page.on('dialog')

playwright 框架可以监听dialog事件，不管你alert 什么时候弹出来，监听到事件就自动处理了。

当出现 JavaScript 对话框时发出，例如alert、prompt或。侦听器必须dialog.accept()或dialog.dismiss()对话框 - 否则页面将冻结等待对话框，并且单击等操作将永远不会完成。

page.on("dialog", handler)

from playwright.sync_api import sync_playwright

def handle_dialog(dialog):
    print(dialog.message)
    dialog.dismiss()

def run(playwright):
    chromium = playwright.chromium
    browser = chromium.launch(headless=False,slow_mo=1000)
    page = browser.new_page()
    page.goto("http://www.baidu.com")
    page.on("dialog", handle_dialog)
    page.evaluate("alert('1')")
    browser.close()

with sync_playwright() as playwright:
    run(playwright)

示例：

html页面如下

<!DOCTYPE html>
<html>
  <head>
    <title>Confirm Dialog Example</title>
  </head>
  <body>
    <h1>Confirm Dialog Example</h1>
    <p>Click the button below to display a confirm dialog:</p>
    <button onclick="displayConfirm()">Click Me</button>
    
    <script>
      function displayConfirm() {
        var result = confirm("Are you sure you want to delete this item?");
        if (result) {
          alert("Item deleted successfully");
        } else {
          alert("Delete cancelled");
        }
      }
    </script>
  </body>
</html>

如下代码打开html页面，点击click me按钮，并接受confirm弹框，因为运行太快，我这里生成了一个trace文件，可以查看执行过程

from playwright.sync_api import sync_playwright, Dialog


def handle_dialog(dialog:Dialog):
    print(dialog.message)
    dialog.accept()

def run(playwright):
    chromium = playwright.chromium
    browser = chromium.launch(headless=False)

    context = browser.new_context()
    context.tracing.start(screenshots=True, snapshots=True, sources=True)
    page = context.new_page()
    page.goto(r"C:\log.html")
    page.get_by_text("Click Me").click()
    page.on("dialog", handle_dialog)
    context.tracing.stop(path="trace.zip")
    browser.close()


with sync_playwright() as playwright:
    run(playwright)

由trace view可以看到，自动执行了两次dialog.dismiss