openclaw.workspace/code-generate/summarize-demo.js

#!/usr/bin/env node
import { chromium } from 'playwright';
import fs from 'fs';
import path from 'path';
import { fileURLToPath } from 'url';

const __filename = fileURLToPath(import.meta.url);
const __dirname = path.dirname(__filename);

async function summarize(url, options = {}) {
  const { length = 'medium', output = 'text' } = options;
  
  console.log(`✨ Summarizing: ${url}`);
  console.log(`📏 Length: ${length}`);
  
  const browser = await chromium.launch({
    headless: true,
    args: ['--no-sandbox', '--disable-setuid-sandbox']
  });
  
  const page = await browser.newPage();
  await page.goto(url, { 
    waitUntil: 'domcontentloaded',
    timeout: 30000
  });
  
  // 提取页面内容
  const content = await page.evaluate(() => {
    // 移除干扰元素
    const selectors = 'script, style, nav, footer, aside, .ad, .advertisement, .popup, .cookie-banner';
    document.querySelectorAll(selectors).forEach(el => el.remove());
    
    // 提取内容
    const title = document.title;
    const text = document.body.innerText
      .replace(/\s+/g, ' ')
      .replace(/\n{3,}/g, '\n\n')
      .trim();
    
    // 提取所有链接
    const links = Array.from(document.querySelectorAll('a[href]'))
      .map(a => ({ text: a.textContent.trim(), href: a.href }))
      .filter(link => link.text && link.href.startsWith('http'))
      .slice(0, 10);
    
    // 提取图片
    const images = Array.from(document.querySelectorAll('img[src]'))
      .map(img => ({ alt: img.alt.trim(), src: img.src }))
      .filter(img => img.src.startsWith('http'))
      .slice(0, 5);
    
    return { title, text, links, images };
  });
  
  await browser.close();
  
  // 根据长度决定摘要大小
  const lengthMap = {
    short: 300,
    medium: 800,
    long: 2000,
    xl: 5000
  };
  
  const maxLength = lengthMap[length] || lengthMap.medium;
  const summary = content.text.slice(0, maxLength) + (content.text.length > maxLength ? '...' : '');
  
  // 输出结果
  const result = {
    url,
    title: content.title,
    summary,
    stats: {
      totalCharacters: content.text.length,
      summaryLength: summary.length,
      linksFound: content.links.length,
      imagesFound: content.images.length
    },
    topLinks: content.links,
    topImages: content.images
  };
  
  if (output === 'json') {
    console.log(JSON.stringify(result, null, 2));
  } else {
    console.log(`\n📄 Title: ${result.title}`);
    console.log(`\n📝 Summary:`);
    console.log(result.summary);
    console.log(`\n📊 Stats:`);
    console.log(`   - Total characters: ${result.stats.totalCharacters}`);
    console.log(`   - Summary length: ${result.stats.summaryLength}`);
    console.log(`   - Links found: ${result.stats.linksFound}`);
    console.log(`   - Images found: ${result.stats.imagesFound}`);
    
    if (result.topLinks.length > 0) {
      console.log(`\n🔗 Top links:`);
      result.topLinks.forEach((link, i) => {
        console.log(`   ${i+1}. ${link.text} - ${link.href}`);
      });
    }
  }
  
  // 清理临时文件
  const tmpItems = fs.readdirSync('/tmp').filter(f => f.startsWith('playwright_'));
  tmpItems.forEach(item => {
    const fullPath = path.join('/tmp', item);
    const stat = fs.statSync(fullPath);
    if (stat.isDirectory()) {
      fs.rmSync(fullPath, { recursive: true, force: true });
    } else {
      fs.unlinkSync(fullPath);
    }
  });
  
  return result;
}

// 如果直接运行这个文件，执行演示
if (process.argv[1] === __filename) {
  const url = process.argv[2] || 'https://www.baidu.com';
  const length = process.argv[3] || 'medium';
  summarize(url, { length });
}

export default summarize;