Puppeteer 从 elementHandle 获取元素导致协议错误

Puppeteer getting element from elementHandle causing protocol error

我正在尝试抓取某个 Facebook 页面,以获取其由某个用户撰写并以某个词开头的帖子。

const puppeteer = require('puppeteer');

async function findPosts(page) {
    const USERNAME = 'test123';
    const posts = await page.$$('.userContentWrapper');
    return posts.filter(async post => {
        try {
            let usernameElement = await post.$('.fwb');
            let username = await page.evaluate(element => element.textContent, usernameElement);
            if (username === USERNAME) {
                let postElement = await post.$('[data-testid="post_message"] p');
                let postContent = page.evaluate(element => element.textContent, postElement);
                return /\[test \d+\]/.test(postContent);
            }
            return false;
        } catch(e) {
            console.log(e);
            return false;
        }
    });
}


(async () => {
    const browser = await puppeteer.launch({
        headless: false
    });
    const page = await browser.newPage();
    await page.goto('https://www.facebook.com/groups/groupid/');
    const pageTitle = await page.title();
    console.log(pageTitle);
    const posts = await findPosts(page);
    console.log(posts);
    await browser.close();
})();

我得到

Error: Protocol error (Runtime.callFunctionOn): Target closed. when I'm trying to get the usernameElement

在这一行:

let usernameElement = await post.$('.fwb');

不确定这里出了什么问题,有什么建议吗?

问题是 filter 函数不适用于 Promises。所以 return posts.filter(...) 会立即 return 然后浏览器关闭。因此,当您尝试 运行 页面上的 $ 函数时,该页面不再存在并且您得到 Target closed 错误。

要使其使用 async/await 语法,您可以使用一个简单的循环:

async function findPosts(page) {
    const USERNAME = 'test123';
    const posts = await page.$$('.userContentWrapper');
    const postsToReturn = [];
    for (let post of posts) {
        /* ... if else logic */
        postsToReturn.push(post); // instead of return true
    }
    return postsToReturn;
}