更新依赖项，添加 user-agents 库；重构浏览器启动逻辑，支持无头模式和隐身模式

2026-01-01 22:04:14 +08:00
parent 95a65e0f84
commit 2c3bc79e6e
5 changed files with 140 additions and 11 deletions
--- a/package.json
+++ b/package.json
@@ -40,6 +40,7 @@
  "dependencies": {
    "better-sqlite3": "^12.5.0",
    "playwright": "^1.57.0",
+    "user-agents": "^1.1.669",
    "zod": "^4.2.1",
    "zod-to-json-schema": "^3.25.1"
  },
@@ -52,6 +53,7 @@
    "@types/better-sqlite3": "^7.6.13",
    "@types/bun": "^1.3.5",
    "@types/node": "^25.0.3",
+    "@types/user-agents": "^1.0.4",
    "commander": "^14.0.2",
    "dotenv": "^17.2.3",
    "drizzle-kit": "^0.31.8",
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -14,6 +14,9 @@ importers:
      playwright:
        specifier: ^1.57.0
        version: 1.57.0
+      user-agents:
+        specifier: ^1.1.669
+        version: 1.1.669
      zod:
        specifier: ^4.2.1
        version: 4.2.1
@@ -45,6 +48,9 @@ importers:
      '@types/node':
        specifier: ^25.0.3
        version: 25.0.3
+      '@types/user-agents':
+        specifier: ^1.0.4
+        version: 1.0.4
      commander:
        specifier: ^14.0.2
        version: 14.0.2
@@ -436,6 +442,9 @@ packages:
  '@types/node@25.0.3':
    resolution: {integrity: sha512-W609buLVRVmeW693xKfzHeIV6nJGGz98uCPfeXI1ELMLXVeKYZ9m15fAMSaUPBHYLGFsVRcMmSCksQOrZV9BYA==}

+  '@types/user-agents@1.0.4':
+    resolution: {integrity: sha512-AjeFc4oX5WPPflgKfRWWJfkEk7Wu82fnj1rROPsiqFt6yElpdGFg8Srtm/4PU4rA9UiDUZlruGPgcwTMQlwq4w==}
+
  asn1js@3.0.7:
    resolution: {integrity: sha512-uLvq6KJu04qoQM6gvBfKFjlh6Gl0vOKQuR5cJMDHQkmwfMOQeN3F3SHCv9SNYSL+CRoHvOGFfllDlVz03GQjvQ==}
    engines: {node: '>=12.0.0'}
@@ -674,6 +683,9 @@ packages:
  ini@1.3.8:
    resolution: {integrity: sha512-JV/yugV2uzW5iMRSiZAyDtQd+nxtUnjeLt0acNdw98kKLrvuRVyB80tsREOE7yvGVgalhZ6RNXCmEHkUKBKxew==}

+  lodash.clonedeep@4.5.0:
+    resolution: {integrity: sha512-H5ZhCF25riFd9uB5UCkVKo61m3S/xZk1x4wA6yp/L3RFP6Z/eHH1ymQcGLo7J3GMPfm0V/7m1tryHuGVxpqEBQ==}
+
  lru-cache@11.2.4:
    resolution: {integrity: sha512-B5Y16Jr9LB9dHVkh6ZevG+vAbOsNOYCX+sXvFWFu7B3Iz5mijW3zdbMyhsh8ANd2mSWBYdJgnqi+mL7/LrOPYg==}
    engines: {node: 20 || >=22}
@@ -838,6 +850,9 @@ packages:
  undici-types@7.16.0:
    resolution: {integrity: sha512-Zz+aZWSj8LE6zoxD+xrjh4VfkIG8Ya6LvYkZqtUQGJPZjYl53ypCaUwWqo7eI0x66KBGeRo+mlBEkMSeSZ38Nw==}

+  user-agents@1.1.669:
+    resolution: {integrity: sha512-pbIzG+AOqCaIpySKJ4IAm1l0VyE4jMnK4y1thV8lm8PYxI+7X5uWcppOK7zY79TCKKTAnJH3/4gaVIZHsjrmJA==}
+
  util-deprecate@1.0.2:
    resolution: {integrity: sha512-EPD5q1uXyFxJpCrLnCc1nHnq3gOa6DZBocAIiI2TaSCA7VCJ1UJDMagCzIkXNsUYfD1daK//LTEQ8xiIbrHtcw==}

@@ -1142,6 +1157,8 @@ snapshots:
    dependencies:
      undici-types: 7.16.0

+  '@types/user-agents@1.0.4': {}
+
  asn1js@3.0.7:
    dependencies:
      pvtsutils: 1.3.6
@@ -1321,6 +1338,8 @@ snapshots:

  ini@1.3.8: {}

+  lodash.clonedeep@4.5.0: {}
+
  lru-cache@11.2.4: {}

  mime-db@1.54.0: {}
@@ -1498,6 +1517,10 @@ snapshots:

  undici-types@7.16.0: {}

+  user-agents@1.1.669:
+    dependencies:
+      lodash.clonedeep: 4.5.0
+
  util-deprecate@1.0.2: {}

  wrappy@1.0.2: {}
--- a/src/playwright/browser.ts
+++ b/src/playwright/browser.ts
@@ -1,6 +1,8 @@
 import { spawn } from 'node:child_process';
 import path from 'node:path';
 import fs from 'node:fs';
+import UserAgent from 'user-agents';
+import { chromium } from 'playwright';

 export const getExecutablePath = () => {
  // 根据不同平台返回 Chrome 的可执行文件路径
@@ -33,22 +35,48 @@ export const main = async (opts?: {
  userDataDir?: string;
  debugPort?: number;
  kiosk?: boolean;
+  headless?: boolean;
 }) => {
  // Chrome 路径和配置
-  const executablePath = opts?.executablePath || getExecutablePath();
+  let executablePath = opts?.executablePath || getExecutablePath();
  // 使用独立的用户数据目录，避免与 Chrome 冲突
  const userDataDir = opts?.userDataDir || path.join(process.cwd(), 'browser-context');
  const debugPort = opts?.debugPort || 9223;
+  const headless = opts?.headless || false;

  console.log('启动 Chrome...', executablePath);
  console.log(`端口: ${debugPort}`);
  console.log(`用户数据目录: ${userDataDir}`);
-  // console.log('注意：需要手动登录账号和安装插件');
+  console.log(`无头模式: ${headless}`);
+  const userAgent = new UserAgent().toString();
+
  const params = [
    `--remote-debugging-port=${debugPort}`,
    `--user-data-dir=${userDataDir}`,
-    // '--kiosk', // 全屏模式，无修改边框
+    '--no-sandbox',
+    '--disable-setuid-sandbox',
+    '--disable-dev-shm-usage',
+    '--no-first-run',
+    `--user-agent=${userAgent}`,
  ];
+
+  if (headless) {
+    params.push(
+      '--headless',
+      '--disable-blink-features=AutomationControlled',
+      '--disable-infobars',
+      '--disable-features=IsolateOrigins,site-per-process',
+      '--disable-features=VizDisplayCompositor',
+      '--window-size=1920,1080',
+      '--disable-background-networking',
+      '--disable-background-timer-throttling',
+      '--disable-backgrounding-occluded-windows',
+      '--disable-renderer-backgrounding',
+      '--disable-component-extensions-with-background-pages',
+      '--disable-features=TranslateUI',
+      '--disable-ipc-flooding-protection',
+    );
+  }
  console.log('启动参数:', params);
  if (opts?.kiosk) {
    params.push('--kiosk'); // 全屏模式，无修改边框
@@ -62,13 +90,12 @@ export const main = async (opts?: {
    return;
  }

-  // 检查 Chrome 可执行文件是否存在
+  // 检查 Chrome 可执行文件是否存在，不存在则使用 Playwright 的浏览器
  if (!fs.existsSync(executablePath)) {
-    console.error('Chrome 可执行文件不存在:', executablePath);
-    return;
+    console.log('Chrome 可执行文件不存在，使用 Playwright 的浏览器');
+    executablePath = chromium.executablePath();
  }

-
  // 启动 Chrome（带远程调试端口）
  const chromeProcess = spawn(
    executablePath,
--- a/src/playwright/core.ts
+++ b/src/playwright/core.ts
@@ -27,13 +27,14 @@ export class Core<T = {}> {
  page: Page | null = null;
  debugPort = 9223;
  debugHost = '127.0.0.1';
+  headless = false;
  status: 'disconnected' | 'connecting' | 'connected' | 'failed' = 'disconnected';
  emitter = new EventEmitter();
  listeners: Listener[] = [];
  recordReady: boolean = false;
  timer: NodeJS.Timeout | null = null;
  data: T | null = null;
-  constructor(opts?: { debugPort?: number, debugHost?: string, listeners?: Listener[] }) {
+  constructor(opts?: { debugPort?: number, debugHost?: string, listeners?: Listener[], headless?: boolean }) {
    if (opts?.debugPort) {
      this.debugPort = opts.debugPort;
    }
@@ -43,9 +44,12 @@ export class Core<T = {}> {
    if (opts?.listeners) {
      this.listeners = opts.listeners;
    }
+    if (opts?.headless !== undefined) {
+      this.headless = opts.headless;
+    }
  }
  async createBrowser() {
-    await main({ debugPort: this.debugPort });
+    await main({ debugPort: this.debugPort, headless: this.headless });
  }
  async init() {
    const debugPort = this.debugPort;
@@ -59,6 +63,9 @@ export class Core<T = {}> {
      this.browserContext = browser.contexts()[0];
      this.handleRequest(this.browserContext);
      this.page = this.browserContext.pages()[0] || await this.browserContext.newPage();
+      if (this.headless) {
+        await this.stealthMode(this.page);
+      }
      this.emitter.emit('connected');
      return;
    } catch (error: any) {
@@ -147,6 +154,74 @@ export class Core<T = {}> {
    }
    this.data = data;
  }
+  async stealthMode(page: Page) {
+    const stealthScript = `
+      () => {
+        Object.defineProperty(navigator, 'webdriver', {
+          get: () => undefined,
+        });
+
+        window.chrome = {
+          runtime: {},
+        };
+
+        Object.defineProperty(navigator, 'plugins', {
+          get: () => [1, 2, 3, 4, 5],
+        });
+
+        Object.defineProperty(navigator, 'languages', {
+          get: () => ['en-US', 'en'],
+        });
+
+        const originalQuery = window.navigator.permissions.query;
+        window.navigator.permissions.query = (parameters) => (
+          parameters.name === 'notifications' ?
+            Promise.resolve({ state: Notification.permission }) :
+            originalQuery(parameters)
+        );
+
+        Object.defineProperty(navigator, 'hardwareConcurrency', {
+          get: () => 4,
+        });
+
+        Object.defineProperty(navigator, 'deviceMemory', {
+          get: () => 8,
+        });
+
+        const originalGetContext = HTMLCanvasElement.prototype.getContext;
+        HTMLCanvasElement.prototype.getContext = function(type) {
+          const context = originalGetContext.apply(this, arguments);
+          if (type === '2d' && context) {
+            const originalGetImageData = context.getImageData;
+            context.getImageData = function() {
+              const imageData = originalGetImageData.apply(this, arguments);
+              for (let i = 0; i < imageData.data.length; i += 4) {
+                imageData.data[i] = imageData.data[i] + Math.random() * 0.1 - 0.05;
+              }
+              return imageData;
+            };
+          }
+          return context;
+        };
+
+        Object.defineProperty(navigator, 'connection', {
+          get: () => ({
+            effectiveType: '4g',
+            rtt: 100,
+            downlink: 10,
+          }),
+        });
+
+        window.navigator.getBattery = () => Promise.resolve({
+          charging: true,
+          chargingTime: 0,
+          dischargingTime: Infinity,
+          level: 1,
+        });
+      }
+    `;
+    await page.addInitScript(stealthScript);
+  }
  async handleRequest(context: BrowserContext) {
    context.on('request', request => {
      const url = request.url();
--- a/src/routes/good/index.ts
+++ b/src/routes/good/index.ts
@@ -10,13 +10,14 @@ app.route({
    icon: 'search',
  }
 }).define(async (ctx) => {
-  const keyword = ctx.query?.keyword as string || '信息差';
+  const { keyword = '信息差', ...rest } = ctx.query;
  const res = await app.run({
    path: 'xhs',
    key: 'search-notes',
    payload: {
      keyword: keyword,
      scrollTimes: 5,
+      ...rest,
      token: ctx.query?.token as string,
    }
  })
@@ -34,13 +35,14 @@ app.route({
    icon: 'search',
  }
 }).define(async (ctx) => {
-  const keyword = ctx.query?.keyword as string || '工作 杭州';
+  const { keyword = '工作 杭州', ...rest } = ctx.query;
  const res = await app.run({
    path: 'xhs',
    key: 'search-notes',
    payload: {
      keyword: keyword,
      scrollTimes: 5,
+      ...rest,
      token: ctx.query?.token as string,
    }
  })