CompanyIQ MCP Server

import puppeteer from 'puppeteer'; async function debugBrowserScraper() { console.log('🔍 DEBUG: Browser Scraper Test'); console.log('═'.repeat(50)); const orgNr = '999059198'; let browser; try { console.log('🌐 Launching browser...'); browser = await puppeteer.launch({ executablePath: '/Applications/Google Chrome.app/Contents/MacOS/Google Chrome', headless: false, // Visible browser args: [ '--no-sandbox', '--disable-setuid-sandbox' ], defaultViewport: null }); const page = await browser.newPage(); // Navigate to the company page const url = `https://virksomhet.brreg.no/nb/oppslag/enheter/${orgNr}`; console.log(`📄 Navigating to: ${url}`); await page.goto(url, { waitUntil: 'networkidle2', timeout: 60000 }); console.log('⏳ Waiting for page to load...'); await page.waitForTimeout(5000); // Scroll to load content console.log('📜 Scrolling to load all content...'); await page.evaluate(async () => { // Scroll down page by page const scrollHeight = document.body.scrollHeight; let currentPosition = 0; const scrollStep = window.innerHeight; while (currentPosition < scrollHeight) { window.scrollTo(0, currentPosition); currentPosition += scrollStep; await new Promise(r => setTimeout(r, 500)); } // Scroll back to top window.scrollTo(0, 0); }); await page.waitForTimeout(2000); // Debug: Check what's on the page console.log('\n🔍 DEBUG: Checking page content...'); const pageInfo = await page.evaluate(() => { // Find all buttons const allButtons = Array.from(document.querySelectorAll('button')); const buttonTexts = allButtons.map(b => b.textContent?.trim() || ''); // Find "Vis flere" buttons specifically const visFlereButtons = allButtons.filter(button => { const text = button.textContent?.toLowerCase() || ''; return text.includes('vis flere'); }); // Find all elements with download test IDs const downloadElements = Array.from(document.querySelectorAll('[data-testid*="download"]')); const downloadTestIds = downloadElements.map(el => el.getAttribute('data-testid')); // Find årsregnskap section const aarsregnskapSection = document.querySelector('[data-testid="aarsregnskap"]') || Array.from(document.querySelectorAll('h1, h2, h3, h4')).find(el => el.textContent?.includes('Årsregnskap') ); // Get all year mentions const yearPattern = /\b20\d{2}\b/g; const bodyText = document.body.innerText; const years = [...new Set(bodyText.match(yearPattern) || [])].sort().reverse(); return { totalButtons: allButtons.length, buttonTexts: buttonTexts.slice(0, 20), // First 20 button texts visFlereCount: visFlereButtons.length, downloadElementsCount: downloadElements.length, downloadTestIds: downloadTestIds.slice(0, 10), hasAarsregnskapSection: !!aarsregnskapSection, yearsFound: years, pageTitle: document.title }; }); console.log('📊 Page Analysis:'); console.log(` - Page Title: ${pageInfo.pageTitle}`); console.log(` - Total buttons: ${pageInfo.totalButtons}`); console.log(` - "Vis flere" buttons found: ${pageInfo.visFlereCount}`); console.log(` - Download elements: ${pageInfo.downloadElementsCount}`); console.log(` - Has Årsregnskap section: ${pageInfo.hasAarsregnskapSection}`); console.log(` - Years mentioned on page: ${pageInfo.yearsFound.join(', ')}`); if (pageInfo.buttonTexts.length > 0) { console.log('\n Button samples:'); pageInfo.buttonTexts.forEach((text, i) => { if (text) console.log(` ${i + 1}. "${text}"`); }); } if (pageInfo.downloadTestIds.length > 0) { console.log('\n Download test IDs found:'); pageInfo.downloadTestIds.forEach(id => { console.log(` - ${id}`); }); } // Try to find and click "Vis flere" buttons console.log('\n🔘 Attempting to click "Vis flere" buttons...'); let totalClicks = 0; for (let attempt = 1; attempt <= 10; attempt++) { const clicked = await page.evaluate(() => { const buttons = Array.from(document.querySelectorAll('button')); let clickedAny = false; buttons.forEach(button => { const text = button.textContent?.toLowerCase() || ''; // Check for various "show more" patterns if ((text.includes('vis flere') || text.includes('vis mer') || text.includes('last inn') || text.includes('se alle')) && !button.disabled) { try { button.click(); clickedAny = true; console.log(`Clicked button: "${button.textContent}"`); } catch (e) { // Ignore } } }); return clickedAny; }); if (clicked) { totalClicks++; console.log(` ✓ Clicked "Vis flere" (attempt ${attempt})`); await page.waitForTimeout(3000); // Wait for content to load } else { console.log(` - No more buttons to click (after ${totalClicks} total clicks)`); break; } } // Re-check after clicking console.log('\n🔍 Re-checking after clicks...'); const finalInfo = await page.evaluate(() => { const downloadElements = Array.from(document.querySelectorAll('[data-testid*="download-aarsregnskap"]')); const years = downloadElements.map(el => { const testId = el.getAttribute('data-testid') || ''; const match = testId.match(/(\d{4})/); return match ? match[1] : null; }).filter(Boolean); return { downloadCount: downloadElements.length, yearsAvailable: [...new Set(years)].sort().reverse() }; }); console.log(`\n📊 Final Results:`); console.log(` - Download buttons found: ${finalInfo.downloadCount}`); console.log(` - Years available: ${finalInfo.yearsAvailable.join(', ')}`); // Keep browser open for manual inspection console.log('\n⏸️ Browser will stay open for 10 seconds for manual inspection...'); await page.waitForTimeout(10000); } catch (error) { console.error('❌ Error:', error); } finally { if (browser) { await browser.close(); console.log('🔒 Browser closed'); } } } debugBrowserScraper();