Playwright执行 JavaScript 脚本：探索浏览器自动化的新境界

前言

我们在进行web自动化时，经常遇到一些不好操作的元素，普通的元素定位和操作容易报错，如果我们使用的selenium的话，就可以使用selenium调用js脚本进行操作。在playwright 中也有类似的方法，使用page.evaluate()执行JavaScript脚本。
page.evaluate()和page.evaluate_handle()之间的唯一区别是page.evaluate_handle()返回JSHandle。

• page.evaluate() 返回调用执行的结果
• page.evaluate_handle()返回JSHandle

page.evaluate()

返回evaluate() 返回执行JavaScript脚本的结果，使用示例如下：

from playwright.sync_api import sync_playwright

def run_js_script():
    with sync_playwright() as playwright:
        browser = playwright.chromium.launch(headless=False)
        context = browser.new_context()
        page = context.new_page()
        page.goto('https://www.baidu.com')

        print(page.evaluate("1 + 2"))

        # 关闭浏览器
        context.close()
        browser.close()

run_js_script()

也可以执行一个函数，如下：

from playwright.sync_api import sync_playwright

def run_js_script():
    with sync_playwright() as playwright:
        browser = playwright.chromium.launch(headless=False)
        context = browser.new_context()
        page = context.new_page()
        page.goto('https://www.baidu.com')

        print(page.evaluate("() => '拜仁慕尼黑'"))

        # 关闭浏览器
        context.close()
        browser.close()

run_js_script()

注：如果传递给page.evaluate()的函数返回一个不可序列化的值，则page.evaluate()解析为undefined

示例

1. 打印网页标题

执行document.title 获取页面的title，代码如下：

from playwright.sync_api import sync_playwright

def run_js_script():
    with sync_playwright() as playwright:
        browser = playwright.chromium.launch(headless=False)
        context = browser.new_context()
        page = context.new_page()
        page.goto("https://www.baidu.com/")
        title = page.evaluate('document.title')
        print(title)
        page.pause()
        # 关闭浏览器
        context.close()
        browser.close()

run_js_script()

2. 操作元素示例

from playwright.sync_api import sync_playwright

def run_js_script():
    with sync_playwright() as playwright:
        browser = playwright.chromium.launch(headless=False)
        context = browser.new_context()
        page = context.new_page()
        page.goto("https://www.baidu.com/")
        js = """
            document.getElementById('kw').value='playwright';
            document.getElementById('su').click();
            """
        page.evaluate(js)
        page.pause()
        # 关闭浏览器
        context.close()
        browser.close()

run_js_script()

页面如下：

page.evaluate_handle()

page.evaluate_handle()的返回值是JSHandle。

from playwright.sync_api import sync_playwright

def run_js_script():
    with sync_playwright() as playwright:
        browser = playwright.chromium.launch(headless=False)
        context = browser.new_context()
        page = context.new_page()
        page.goto("https://www.baidu.com/")
        a_handle = page.evaluate_handle("document.body")
        result_handle = page.evaluate_handle("body => body.innerHTML", a_handle)
        print(result_handle.json_value())
        result_handle.dispose()
        page.pause()
        # 关闭浏览器
        context.close()
        browser.close()

run_js_script()

总结

本文主要介绍了playwright执行js脚本的操作，与selenium一样，playwright同样可以执行js脚本来完成一些不便操作的元素。