download-pdf.js 628 B

123456789101112131415161718
  1. #!/usr/bin/env node
  2. const path = require('path');
  3. const fs = require('fs');
  4. const { createWorker } = require('../../');
  5. const [,, imagePath] = process.argv;
  6. const image = path.resolve(__dirname, (imagePath || '../../tests/assets/images/cosmic.png'));
  7. console.log(`Recognizing ${image}`);
  8. (async () => {
  9. const worker = await createWorker();
  10. const { data: { text, pdf } } = await worker.recognize(image, {pdfTitle: "Example PDF"}, {pdf: true});
  11. console.log(text);
  12. fs.writeFileSync('tesseract-ocr-result.pdf', Buffer.from(pdf));
  13. console.log('Generate PDF: tesseract-ocr-result.pdf');
  14. await worker.terminate();
  15. })();