Scrape linkdin

test.js

(async () => {
  const DELAY_BETWEEN_REQUESTS = 5000;
  const SAVE_EVERY = 10;

  function sleep(ms) {
    return new Promise(res => setTimeout(res, ms));
  }

  function waitForResult() {
    return new Promise((resolve) => {
      const interval = setInterval(() => {
        const loading = document.querySelector('.spinner-border');
        const resultCard = document.querySelector('.email-result-card');

        if (!loading && resultCard) {
          clearInterval(interval);
          resolve();
        }
      }, 500);
    });
  }

  function extractLinkedIn() {
    const link = document.querySelector('.social-icons a[href*="linkedin"]');
    return link ? link.href : '';
  }

  function setEmailAndSearch(email) {
    const input = document.querySelector('#email');
    const form = document.querySelector('#reverse-email-lookup-form');

    input.value = email;
    input.dispatchEvent(new Event('input', { bubbles: true }));

    form.querySelector('button[type="submit"]').click();
  }

  function downloadCSV(data, part = 'final') {
    const csv = [
      ['email', 'linkedin'],
      ...data.map(row => [row.email, row.linkedin])
    ]
      .map(e => e.join(','))
      .join('\n');

    const blob = new Blob([csv], { type: 'text/csv' });
    const url = URL.createObjectURL(blob);

    const a = document.createElement('a');
    a.href = url;
    a.download = `linkedin_results_${part}.csv`;
    a.click();

    URL.revokeObjectURL(url);
  }

  async function processEmails(emails) {
    const results = [];

    for (let i = 0; i < emails.length; i++) {
      const email = emails[i];

      console.log(`Checking (${i + 1}/${emails.length}):`, email);

      try {
        setEmailAndSearch(email);

        await waitForResult();
        await sleep(1000);

        const linkedin = extractLinkedIn();

        results.push({ email, linkedin });

        console.log('✔ Result:', email, linkedin);

      } catch (err) {
        console.error('❌ Error:', email, err);
        results.push({ email, linkedin: '' });
      }

      // 🔥 SAVE EVERY 10
      if ((i + 1) % SAVE_EVERY === 0) {
        console.log(`💾 Saving batch ${i + 1}`);
        downloadCSV(results, `batch_${i + 1}`);
      }

      await sleep(DELAY_BETWEEN_REQUESTS);
    }

    // Final save
    downloadCSV(results, 'final');
    console.log('🎉 DONE');
  }

  // File upload
  const input = document.createElement('input');
  input.type = 'file';
  input.accept = '.csv';

  input.onchange = async (e) => {
    const file = e.target.files[0];
    const text = await file.text();

    const lines = text.split('\n').slice(1);
    const emails = lines
      .map(l => l.split(',')[0]?.trim())
      .filter(Boolean);

    console.log('Loaded emails:', emails.length);

    await processEmails(emails);
  };

  input.click();
})();