Initial commit: New industrial design with green theme

crawl-hanger-clamp-v2.js

@@ -0,0 +1,132 @@
+const { chromium } = require('playwright');
+const fs = require('fs');
+
+const URLs = [
+  'https://www.dealplustech.co.th/%e0%b9%81%e0%b8%ae%e0%b8%87%e0%b9%80%e0%b8%81%e0%b8%ad%e0%b8%a3%e0%b9%8c-%e0%b9%81%e0%b8%84%e0%b8%a5%e0%b9%89%e0%b8%a1-%e0%b9%82%e0%b8%9a%e0%b8%a5%e0%b8%97%e0%b9%8c-%e0%b9%81%e0%b8%ab%e0%b8%a7/',
+  'https://www.dealplustech.co.th/%e0%b8%aa%e0%b8%9b%e0%b8%a3%e0%b8%b4%e0%b8%97%e0%b8%a3%e0%b8%b4%e0%b8%87%e0%b9%81%e0%b8%ae%e0%b8%87%e0%b9%80%e0%b8%81%e0%b8%ad%e0%b8%a3%e0%b9%8c-sr19-adjustable-split-ring-hanger/',
+  'https://www.dealplustech.co.th/%e0%b9%80%e0%b8%84%e0%b8%a7%e0%b8%b4%e0%b8%aa%e0%b9%81%e0%b8%ae%e0%b8%87%e0%b9%80%e0%b8%81%e0%b8%ad%e0%b8%a3%e0%b9%8c/',
+  'https://www.dealplustech.co.th/%e0%b9%81%e0%b8%84%e0%b8%a5%e0%b9%89%e0%b8%a1%e0%b8%9f%e0%b8%b1%e0%b8%99%e0%b8%88%e0%b8%a3%e0%b8%b0%e0%b9%80%e0%b8%82%e0%b9%89-beam-clamp/',
+  'https://www.dealplustech.co.th/%e0%b9%81%e0%b8%84%e0%b8%a5%e0%b9%89%e0%b8%a1%e0%b8%ab%e0%b8%a2%e0%b8%94%e0%b8%99%e0%b9%89%e0%b8%b3-adjustable-band-hanger/',
+  'https://www.dealplustech.co.th/%e0%b9%81%e0%b8%84%e0%b8%a5%e0%b9%89%e0%b8%a1%e0%b9%80%e0%b8%a5%e0%b9%80%e0%b8%a7%e0%b8%a5-level-clamp/',
+  'https://www.dealplustech.co.th/%e0%b9%80%e0%b8%88%e0%b9%82%e0%b8%9b%e0%b8%a5%e0%b8%97%e0%b9%8c-%e0%b9%81%e0%b8%ad%e0%b8%a5%e0%b9%82%e0%b8%9b%e0%b8%a5%e0%b8%97%e0%b9%8c-j-bolt-l-bolt/',
+  'https://www.dealplustech.co.th/%e0%b8%a2%e0%b8%b9%e0%b9%82%e0%b8%9b%e0%b8%a5%e0%b8%97%e0%b9%8c-u-bolt/',
+  'https://www.dealplustech.co.th/%e0%b8%a2%e0%b8%b9%e0%b9%82%e0%b8%9b%e0%b8%a5%e0%b8%97%e0%b9%8c-%e0%b8%9b%e0%b8%a3%e0%b8%b0%e0%b8%81%e0%b8%b1%e0%b8%9a-u-bolt-clamp/',
+  'https://www.dealplustech.co.th/%e0%b8%a2%e0%b8%b9%e0%b9%82%e0%b8%9b%e0%b8%a5%e0%b8%97%e0%b9%8c%e0%b9%80%e0%b8%ab%e0%b8%a5%e0%b9%87%e0%b8%81%e0%b9%81%e0%b8%9c%e0%b9%88%e0%b8%99-%e0%b8%a2%e0%b8%b9%e0%b9%81%e0%b8%9a%e0%b8%99-strap/',
+  'https://www.dealplustech.co.th/%e0%b8%99%e0%b9%87%e0%b8%ad%e0%b8%95-%e0%b9%81%e0%b8%ab%e0%b8%a7%e0%b8%99-%e0%b8%aa%e0%b8%81%e0%b8%a3%e0%b8%b0-hex-nut-flat-washer-hexagon-head-screw/',
+  'https://www.dealplustech.co.th/%e0%b8%9e%e0%b8%b8%e0%b8%81%e0%b8%95%e0%b9%88%e0%b8%b2%e0%b8%87%e0%b9%86/',
+  'https://www.dealplustech.co.th/%e0%b8%9e%e0%b8%b8%e0%b8%81%e0%b9%80%e0%b8%ab%e0%b8%a5%e0%b9%87%e0%b8%81-sleeve-anchor-bolt/',
+  'https://www.dealplustech.co.th/%e0%b8%aa%e0%b8%95%e0%b8%b1%e0%b8%94%e0%b9%80%e0%b8%81%e0%b8%a5%e0%b8%b5%e0%b8%a2%e0%b8%a7%e0%b8%95%e0%b8%a5%e0%b8%ad%e0%b8%94-%e0%b9%80%e0%b8%ab%e0%b8%a5%e0%b9%87%e0%b8%81-threaded-rod/'
+];
+
+async function crawlPage(url) {
+  console.log(`Crawling: ${url}`);
+  
+  const browser = await chromium.launch({ headless: true });
+  const page = await browser.newPage();
+  
+  try {
+    await page.goto(url, { waitUntil: 'networkidle', timeout: 30000 });
+    
+    const data = await page.evaluate(() => {
+      const result = {
+        category: '',
+        url: window.location.href,
+        description: '',
+        image: '',
+        products: []
+      };
+      
+      const h1 = document.querySelector('h1');
+      result.category = h1 ? h1.textContent.trim() : '';
+      
+      const metaDesc = document.querySelector('meta[name="description"]');
+      if (metaDesc) {
+        result.description = metaDesc.getAttribute('content');
+      }
+      
+      const ogImage = document.querySelector('meta[property="og:image"]');
+      if (ogImage) {
+        result.image = ogImage.getAttribute('content');
+      }
+      
+      if (!result.image) {
+        const featuredImg = document.querySelector('.wp-post-image, .attachment-large, .featured-image img, article img');
+        if (featuredImg) {
+          result.image = featuredImg.getAttribute('src') || '';
+        }
+      }
+      
+      if (!result.image) {
+        const contentImages = document.querySelectorAll('.entry-content img, .post-content img, main img');
+        for (const img of contentImages) {
+          const src = img.getAttribute('src') || '';
+          if (src && src.includes('wp-content/uploads') && !src.includes('logo') && !src.includes('QR')) {
+            result.image = src;
+            break;
+          }
+        }
+      }
+      
+      const tables = document.querySelectorAll('table');
+      tables.forEach((table) => {
+        const rows = table.querySelectorAll('tr');
+        if (rows.length < 2) return;
+        
+        const headers = [];
+        const headerCells = table.querySelectorAll('th');
+        headerCells.forEach(th => {
+          headers.push(th.textContent.trim());
+        });
+        
+        const products = [];
+        let hasData = false;
+        rows.forEach((row, rowIdx) => {
+          const cells = row.querySelectorAll('td, th');
+          const rowData = {};
+          let cellIndex = 0;
+          
+          cells.forEach((cell) => {
+            const header = headers[cellIndex] || `col${cellIndex}`;
+            rowData[header] = cell.textContent.trim();
+            cellIndex++;
+            if (cell.textContent.trim()) hasData = true;
+          });
+          
+          if (Object.keys(rowData).length > 0) {
+            products.push(rowData);
+          }
+        });
+        
+        if (hasData && products.length > 0) {
+          result.products.push(...products);
+        }
+      });
+      
+      return result;
+    });
+
+    await browser.close();
+    return data;
+    
+  } catch (error) {
+    console.error(`  Error crawling ${url}:`, error.message);
+    await browser.close();
+    return { category: url, url, error: error.message };
+  }
+}
+
+async function main() {
+  const results = [];
+  
+  for (const url of URLs) {
+    const data = await crawlPage(url);
+    results.push(data);
+    await new Promise(resolve => setTimeout(resolve, 1500));
+  }
+  
+  fs.writeFileSync('hanger-clamp-products.json', JSON.stringify(results, null, 2));
+  console.log('\nData saved to hanger-clamp-products.json');
+  console.log(`Total pages crawled: ${results.length}`);
+}
+
+main().catch(console.error);