nodejs_chromium 1.0.0

package/demo.js

@@ -0,0 +1,58 @@
+require('nodejs_chromium');
+
+/**
+ * @param {Object} account
+ */
+function debug_view(account) {
+	return await new Promise(async (trueCall, exitCall) => {
+		const { unid } = this.ctx.client;
+
+		const params = {
+			uid: this.ctx.client.unid,
+			incognito: true,
+			debug: false,
+			view: true,
+			width: 1680,
+			height: 960,
+			id: account.id,
+			ua: account.ua,
+		};
+		const chrome = await newChrome(params);
+		console.log(Date.now() / 1000, '初始化');
+
+		try {
+			await chrome.open('https://domain.com/');
+			await chrome.wait('div#login', { timeout: 5 });
+			await chrome.saveHtml(`runtime/pages/${unid}1.html`);
+			await chrome.input('input[name=password]', account.password);
+			await chrome.input('input[name=user]', account.user);
+
+			setTimeout(async () => {
+
+				await chrome.click('div.submit');
+				await chrome.wait();
+				await chrome.saveCookies(account.cookie_file);
+				await chrome.saveHtml(`runtime/pages/${unid}2.html`);
+				await chrome.wait('div#body');
+				await chrome.watermark({ text: account.time });
+				await chrome.photograph(`runtime/screen/${unid}.jpeg`, 50);
+				await chrome.disconnect();
+
+				await patch.saveJson('runtime/response.json', chrome.response)
+
+				trueCall({})
+
+			}, 2000);
+
+
+		}
+		catch (error) {
+			await chrome.disconnect();
+			console.log('Error:', error.message, "\n\n");
+			// throw error;
+			exitCall(error);
+		}
+
+	})
+
+}

package/index.js

@@ -0,0 +1,61 @@
+require('nodejs_patch');
+const puppeteer = require("puppeteer");
+const chrome = require("./src/chrome.js");
+
+async function newChrome(params) {
+	const ua = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36';
+	let { id, view, width, height, scale, mobile, dumpio, debug, incognito, path } = params;
+	if (!width) width = 1024;
+	if (!height) height = 768;
+	if (!scale) scale = 1;
+	if (!id) id = (Date.now() / 1000) + '';
+
+	let option = {
+		userDataDir: `runtime/template/${id}`,
+		// timeout: 500,//最大允许超时间，默认为3000，最小215，一般不要设置
+		slowMo: 2, //将 Puppeteer 操作减少指定的毫秒数，每一步停留时间，不能太大，否则会太慢。这样你就可以看清发生了什么，这很有用。
+		headless: view ? false : 'new', //'new',设置是否在无头模式下运行浏览器,false=会启动浏览器，true=无界面
+		devtools: !!debug, //打开调试
+		ignoreHTTPSErrors: true, //忽略 HTTPS 错误。屏蔽跳转不同域名的报错
+		ignoreDefaultArgs: ["--enable-automation"], //忽略默认的 --enable-automation 参数，这有助于防止某些网站检测到自动化行为。
+		dumpio: !!dumpio, //是否将浏览器的标准输入/输出流（stdio）的内容输出到 Node.js 的 stdout 和 stderr。
+		defaultViewport: {
+			width,
+			height,
+			deviceScaleFactor: scale, //缩放比例
+			isMobile: !!mobile,
+		},
+		args: [
+			'--disable-web-security', //禁用浏览器的同源策略（Same-Origin Policy）和跨站请求伪造（CSRF）保护
+			`--window-size=${width},${height}`,
+			'--no-sandbox', //禁用沙箱模式
+			'--disable-setuid-sandbox', //禁用 setuid 沙箱。这是另一种沙箱模式，通常用于 Linux 系统上。
+			'--disable-infobars', //禁用 Chrome 在自动化控制时显示的信息栏。这个信息栏通常会告诉用户浏览器正在被自动化工具控制。
+			'--disable-gpu', //禁用 GPU 加速。这通常用于在服务器环境或某些不支持 GPU 加速的平台上运行 Chrome。
+			'--disable-blink-features=AutomationControlled', //禁用 blink 引擎的自动化控制特性，防止网站通过检测 blink 引擎的特性来判断浏览器是否处于自动化控制状态。
+			// 在自动化测试或爬虫等场景中，这个参数可以帮助隐藏浏览器的自动化痕迹，使得浏览器行为更接近于真实用户操作
+			'--lang=zh-CN', //设置中文环境
+		]
+	}
+
+	if (incognito) { //使用无痕模式启动
+		option.args.push('--incognito', '--disable-infobars')
+	}
+
+	if (path) option.executablePath = path; //指定chrome安装路径
+
+	const browser = await puppeteer.launch(option);
+	const page = (await browser.pages())[0];
+	await page.setRequestInterception(true); //允许拦截
+	await page.setUserAgent(params.ua || ua);
+	await page.evaluateOnNewDocument(() => {
+		const newProto = navigator.__proto__;
+		delete newProto.webdriver; //删除 navigator.webdriver字段
+		navigator.__proto__ = newProto;
+	}); //在每次新文档加载时，删除 navigator.webdriver 字段，这有助于防止某些网站检测到自动化行为。
+
+	return new chrome(browser, page, params);
+}
+
+
+module.exports = { newChrome }

package/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "nodejs_chromium",
+  "version": "1.0.0",
+  "description": "",
+  "main": "index.js",
+  "scripts": {
+    "test": "echo \"Error: no test specified\" && exit 1"
+  },
+  "dependencies": {
+    "nodejs_patch": ">1.0.5",
+    "puppeteer": "^21.10.0"
+  },
+  "author": "fazo@qq.com",
+  "license": "ISC"
+}

package/src/chrome.js

@@ -0,0 +1,227 @@
+const fs = require("fs");
+const { parse: parseUrl } = require("url");
+
+module.exports = class {
+	browser = null;
+	page = null;
+	response = [];
+
+	constructor(browser, page, params) {
+		this.browser = browser;
+		this.page = page;
+
+		page.on('request', (request) => {
+			const { abort_img, no_cache, append_headers } = params;
+			if (abort_img && /\.(?:png|jpg|jpeg|svg|gif)$/i.test(request.url())) {
+				request.abort();
+				return;
+			}
+
+			const headers = request.headers();
+			if (typeof append_headers === 'object') Object.assign(headers, append_headers);
+
+			headers['Access-Control-Allow-Origin'] = '*'; // 设置允许跨源访问的域名，可以根据需求修改
+			headers['Access-Control-Allow-Methods'] = '*'; //'GET, POST, PUT, OPTIONS';
+			// headers['Access-Control-Allow-Headers'] = 'Content-Type';
+
+			if (no_cache) {
+				const rType = request.resourceType();
+				if (rType === 'script' || rType === 'stylesheet') {
+					headers['Cache-Control'] = 'no-store'; // 禁用缓存
+				}
+			}
+
+			request.continue({ headers });
+		});
+
+		page.on('response', async res => {
+			let json = await this.parseResponse(res);
+			if (json.method === 'POST') console.log(json);
+			if (json.body) this.response.push(json);
+		});
+	}
+
+	async open(url) {
+		return await this.page.goto(url, { waitUntil: 'load' });
+	}
+
+	/**
+	 * 根据目标要求格式提交
+	 *
+	 * @param url
+	 * @param data
+	 * @returns {Promise<*>}
+	 */
+	async post(url, data) {
+		//JSON.stringify(data)
+		return await this.page.goto(url, { method: 'POST', body: data, waitUntil: 'load' });
+	}
+
+	async input(el, value) {
+		return await this.page.type(el, value);
+	}
+
+	async click(el) {
+		await this.page.click(el);
+	}
+
+	/**
+	 * page.on方法
+	 * 
+	 * @param {Object} key
+	 * @param {Object} call
+	 */
+	async on(key, call) {
+		await this.page.on(key, call);
+	}
+
+	async disconnect() {
+		await this.browser.disconnect();
+	}
+
+	async wait(ele, timeout = 0, tryCount = 0) {
+		if (ele) {
+			return await this.waitSelector(ele, timeout, tryCount);
+		}
+		return await this.page.waitForNavigation();
+	}
+
+	async waitSelector(ele, timeout = 0, tryCount = 0) {
+		if (timeout < 200) timeout = timeout * 1000;
+		try {
+			await this.page.waitForSelector(ele, { timeout });
+			return true;
+		}
+		catch (e) {
+			if (tryCount > 0) {
+				return await this.waitSelector(ele, timeout, --tryCount)
+			}
+			return false;
+		}
+	}
+
+	async parseResponse(response) {
+		// const response = await this.page.waitForResponse(res => res);
+
+		const value = {};
+		const request = await response.request();
+		const headers = await response.headers();
+		value.method = await request.method();
+		if (value.method === 'OPTIONS') return;
+
+		value.type = await request.resourceType();
+		// value.redirect = await response.redirectURL();
+		if (value.type === 'xhr') value.type = 'AJAX';
+		value.url = await response.url();
+		value.domain = parseUrl(value.url)['host'];
+		value.content = headers['content-type'];
+		value.length = headers['content-length'];
+		value.status = await response.status();
+		value.ok = await response.ok();
+		value.datetime = (new Date(headers['date'])).date('yyyy-mm-dd hh:ii:ss');
+		// value.headers = headers;
+		if (headers['server']) value.server = headers['server'];
+		if (headers['set-cookie']) value.cookies = headers['set-cookie'];
+		value.remote = await response.remoteAddress(); //目标服务器
+		if (value.status === 301 || value.status === 302) return value;
+		if (['image', 'font', 'other', 'script', 'stylesheet', 'document', 'ping', 'fetch'].has(value.type)) return value;
+		if (value.content) {
+			if (value.content.startsWith('application/vnd')) return value;
+			if (value.content.startsWith('application/xml')) return value;
+			if (value.content.startsWith('text/css')) return value;
+		}
+
+		value.post = await request.postData();
+		if (value.post) value.post = value.post.toString();
+
+		try {
+			value.response = await response.buffer();
+			value.response = value.response.toString();
+		}
+		catch (e) {
+			value.response = e.parse();
+		}
+
+		return value;
+	}
+
+
+	async saveHtml(file) {
+		try {
+			await this.improveUrls(); //修正js/css的域名
+			const body = await this.page.evaluate(() => {
+				return document.documentElement.innerHTML;
+			});
+			await fs.writeFileSync(file, body);
+		}
+		catch (e) {
+			console.log('saveHtml Error:', e.parse());
+		}
+	}
+
+	async saveCookies(file) {
+		try {
+			const cookies = await this.page.cookies();
+			await fs.writeFileSync(file, JSON.stringify(cookies, null, 2));
+		}
+		catch (e) {
+			console.log('saveHtml Error:', e.parse());
+		}
+	}
+
+	async watermark(conf) {
+		await this.page.evaluate((conf) => {
+			const wmDiv = document.createElement('div');
+			wmDiv.style.position = 'fixed';
+			wmDiv.style.top = `50%`;
+			wmDiv.style.left = `50%`;
+			wmDiv.style.transform = 'translate(-50%, -50%)';
+			wmDiv.style.fontSize = `36px`;
+			wmDiv.style.zIndex = '10000';
+			// wmDiv.style.color = color;
+			wmDiv.style.color = 'rgba(0, 0, 0, 0.9)';
+			wmDiv.style.pointerEvents = 'none';
+			wmDiv.innerText = conf.text;
+			document.body.appendChild(wmDiv);
+		}, conf);
+		await this.page.waitForSelector('div');
+	}
+
+
+	/**
+	 * 网页拍照
+	 * @param file
+	 * @param quality
+	 * @returns {Promise<void>}
+	 */
+	async photograph(file, quality = 50) {
+		await this.page.screenshot({
+			path: file,
+			fullPage: true, //全屏
+			type: 'jpeg',
+			quality: quality,
+			omitBackground: true, //显示背景
+		});
+		console.log('photograph=', file);
+	}
+
+	/**
+	 * 补全所有本地js/css
+	 * @returns {Promise<void>}
+	 * @constructor
+	 */
+	async improveUrls() {
+		const url = parseUrl(this.page.url());
+		const domain = url.protocol + '//' + url.host;
+		await this.page.evaluate((domain) => {
+			const elements = document.querySelectorAll('script[src], link[href], iframe[src]');
+			elements.forEach((element) => {
+				const src = element.getAttribute('src');
+				const href = element.getAttribute('href');
+				if (src && src.startsWith('/')) element.src = domain + src;
+				if (href && href.startsWith('/')) element.href = domain + href;
+			});
+		}, domain);
+	}
+
+}