import * as pdfjsLib from 'pdfjs-dist/legacy/build/pdf.mjs'; import { readFileSync } from 'fs'; async function extractText() { const data = new Uint8Array(readFileSync('./public/resume-template.pdf')); const pdf = await pdfjsLib.getDocument({ data }).promise; console.log('Total pages:', pdf.numPages); for (let i = 1; i <= Math.min(pdf.numPages, 2); i++) { const page = await pdf.getPage(i); const content = await page.getTextContent(); console.log(`\n=== Page ${i} ===`); // Group by Y position to understand layout const items = content.items; items.forEach(item => { console.log(`${item.transform[5].toFixed(0)}: ${item.str}`); }); } } extractText().catch(console.error);