kyj-yanglao-web-new/extract_pdf_script.js

const pdfjsLib = require('pdfjs-dist/legacy/build/pdf.js');
const fs = require('fs');

const pdfPath = 'd:/Users/chenjun/kyj-yanglao-web-new/src/views/elderly/apply/check-in/颐年集团养老服务合同-2026年6月终版docx.pdf';
const outPath = 'd:/Users/chenjun/kyj-yanglao-web-new/pdf_full_text.txt';

async function main() {
  const data = new Uint8Array(fs.readFileSync(pdfPath));
  const loadingTask = pdfjsLib.getDocument({ data });
  const pdf = await loadingTask.promise;
  console.log('Total pages:', pdf.numPages);
  
  let allText = '';
  for (let i = 1; i <= pdf.numPages; i++) {
    const page = await pdf.getPage(i);
    const content = await page.getTextContent();
    const lines = [];
    let lastY = null;
    let currentLine = '';
    for (const item of content.items) {
      const y = Math.round(item.transform[5] * 10) / 10;
      if (lastY !== null && Math.abs(y - lastY) > 5) {
        if (currentLine.trim()) lines.push(currentLine);
        currentLine = '';
      }
      currentLine += item.str;
      lastY = y;
    }
    if (currentLine.trim()) lines.push(currentLine);
    
    const pageText = lines.join('\n');
    allText += `\n===== PAGE ${i} =====\n${pageText}\n`;
    console.log(`Page ${i}: ${pageText.length} chars`);
  }
  
  fs.writeFileSync(outPath, allText, 'utf8');
  console.log('Saved to', outPath);
}

main().catch(err => { console.error('ERROR:', err); process.exit(1); });