验证码破解

reCaptcha

Scrapeless 爬虫浏览器仅协助你自动破解 reCaptchaV2；后续操作需要你自行实现。

Cloudflare

Cloudflare Turnstile
Cloudflare Challenge

Scrapeless 爬虫浏览器仅协助你自动破解 Turnstile 或 Challenge；后续操作需要你自行实现。有关处理 Cloudflare challenge（包括获取 cf_clearance）的详细实践，请参考：https://www.scrapeless.com/en/blog/cloudflare-challenge-bypass

解决方案示例

当我们连接到浏览器并访问目标站点时，Scrapeless 会自动破解 CAPTCHA。不过，我们需要确保 CAPTCHA 已被成功破解。这是一个简单的示例：该示例访问目标站点，并通过监听 Captcha.solveFinished CDP 事件来确认 CAPTCHA 已被成功破解。最后，它会对页面进行截图以供验证。

该示例定义了两个主要方法：

addCaptchaListener：用于监听浏览器会话中的 CAPTCHA 事件
onCaptchaFinished：用于等待 CAPTCHA 破解完成

支持的 CAPTCHA 列表

reCaptcha v2
Cloudflare Turnstile
Cloudflare 5s Challenge
AWS Challenge

import puppeteer from "puppeteer-core";
import EventEmitter from 'events';
const emitter = new EventEmitter()
const scrapelessUrl = 'wss://browser.scrapeless.com/api/v2/browser?token=your_api_key&sessionTTL=180&proxyCountry=ANY';
 
export async function example(url) {
  const browser = await puppeteer.connect({
    browserWSEndpoint: scrapelessUrl,
    defaultViewport: null
  });
  console.log("Verbonden met Scrapeless browser");
  try {
    const page = await browser.newPage();
    // Listen for captcha events
    console.debug("addCaptchaListener: Start listening for captcha events");
    await addCaptchaListener(page);
    console.log("Navigated to URL:", url);
    await page.goto(url, { waitUntil: "domcontentloaded", timeout: 30000 });
    console.log("onCaptchaFinished: Waiting for captcha solving to finish...");
    await onCaptchaFinished()
    // Screenshot for debugging
    console.debug("Taking screenshot of the final page...");
    await page.screenshot({ path: 'screenshot.png', fullPage: true });
  } catch (error) {
    console.error(error);
  } finally {
    await browser.close();
    console.log("Browser closed");
  }
}
 
async function addCaptchaListener(page) {
  const client = await page.createCDPSession();
  client.on("Captcha.detected", (msg) => {
    console.debug("Captcha.detected: ", msg);
  });
  client.on("Captcha.solveFinished", async (msg) => {
    console.debug("Captcha.solveFinished: ", msg);
    emitter.emit("Captcha.solveFinished", msg);
    client.removeAllListeners()
  });
}
 
async function onCaptchaFinished(timeout = 60_000) {
  return Promise.race([
    new Promise((resolve) => {
      emitter.on("Captcha.solveFinished", (msg) => {
        resolve(msg);
      });
    }),
    new Promise((_, reject) => setTimeout(() => reject('Timeout'), timeout))
  ])
}

reCaptcha 示例

调用示例代码方法来验证 reCaptcha 的自动破解。

 example('https://recaptcha-demo.appspot.com/recaptcha-v2-checkbox-explicit.php');

Cloudflare Turnstile 示例

调用示例代码方法来验证 Cloudflare Turnstile 的自动破解。

 example('https://www.scrapingcourse.com/login/cf-turnstile');

成功破解 Cloudflare Turnstile 不仅可以通过监听 Captcha.solveFinished CDP 事件来确认，还可以通过监听 window.turnstile.getResponse() 来确认。这是一个完整的示例：

import puppeteer from "puppeteer-core";
const scrapelessUrl = 'wss://browser.scrapeless.com/api/v2/browser?token=your_api_key&sessionTTL=180&proxyCountry=ANY';
 
export async function turnstileExample(url) {
  const browser = await puppeteer.connect({
    browserWSEndpoint: scrapelessUrl,
    defaultViewport: null
  });
  console.log("Verbonden met Scrapeless browser");
  try {
    const page = await browser.newPage();
    console.log("Navigated to URL:", url);
    await page.goto(url, { waitUntil: "domcontentloaded", timeout: 30000 });
    console.log("onCaptchaFinished: Waiting for captcha solving to finish...");
    await waitTurnstile(page)
    // Screenshot for debugging
    console.debug("Taking screenshot of the final page...");
    await page.screenshot({ path: 'screenshot.png', fullPage: true });
  } catch (error) {
    console.error(error);
  } finally {
    await browser.close();
    console.log("Browser closed");
  }
}
 
async function waitTurnstile(page) {
    await page.waitForFunction(() => {
        return window.turnstile && window.turnstile.getResponse();
    });
    const token = await page.evaluate(() => {
        return window.turnstile.getResponse();
    });
    console.log("Cloudflare Turnstile token:", token);
}
 
turnstileExample('https://www.scrapingcourse.com/login/cf-turnstile');

Cloudflare Challenge 示例

Cloudflare Challenge 比较特殊，因为有时它不会触发 Cloudflare Challenge，此时通过监听 CDP 事件来确认成功破解的方法会超时。因此，等待破解后页面上元素的出现是一种更稳定的方法。这是一个完整的示例：

import puppeteer from "puppeteer-core";
const scrapelessUrl = 'wss://browser.scrapeless.com/api/v2/browser?token=your_api_key&sessionTTL=180&proxyCountry=ANY';
 
export async function challengeExample(url) {
  const browser = await puppeteer.connect({
    browserWSEndpoint: scrapelessUrl,
    defaultViewport: null
  });
  console.log("Verbonden met Scrapeless browser");
  try {
    const page = await browser.newPage();
    console.log("Navigated to URL:", url);
    await page.goto(url, { waitUntil: "domcontentloaded", timeout: 30000 });
    console.log("onCaptchaFinished: Waiting for captcha solving to finish...");
    await waitChallenge(page, 'main.page-content .challenge-info')
    // Screenshot for debugging
    console.debug("Taking screenshot of the final page...");
    await page.screenshot({ path: 'screenshot.png', fullPage: true });
  } catch (error) {
    console.error(error);
  } finally {
    await browser.close();
    console.log("Browser closed");
  }
}
 
async function waitChallenge(page, selector) {
    await page.waitForSelector(selector);
    console.log("Cloudflare Challenge completed");
}
 
challengeExample('https://www.scrapingcourse.com/cloudflare-challenge');

CDP API 代理配置