fmode-voc-projects/scripts/tools/read-product-docs.js

// Read product docs (.docx) and emit trimmed plain text.
const fs = require('fs');
const path = require('path');
const AdmZip = require('adm-zip');

function docxText(p) {
  try {
    const z = new AdmZip(p);
    const x = z.readAsText('word/document.xml');
    if (!x) return '[empty]';
    return x
      .replace(/<w:tab\/>/g, '  ')
      .replace(/<w:br\/>/g, '\n')
      .replace(/<\/w:p>/g, '\n')
      .replace(/<[^>]+>/g, '')
      .replace(/\s+\n/g, '\n')
      .replace(/\n{3,}/g, '\n\n')
      .trim();
  } catch (e) {
    return '[ERR ' + e.message + ']';
  }
}

const targets = [
  {
    tag: 'probiotic-背景需求',
    p: 'E:\\文件资料\\产品创新2：儿童乳酸菌素片\\儿童乳酸菌素片-资料\\01 儿童装乳酸菌素片产品背景及需求0418.docx',
  },
  {
    tag: 'probiotic-说明书',
    p: 'E:\\文件资料\\产品创新2：儿童乳酸菌素片\\儿童乳酸菌素片-资料\\儿乳-乳酸菌素片说明书（0.2g）.docx',
  },
  {
    tag: 'monkey-产品资料',
    p: 'E:\\文件资料\\产品创新2-江中猴菇饮产品资料.docx',
  },
];

const outDir = path.resolve('data', 'product-docs');
if (!fs.existsSync(outDir)) fs.mkdirSync(outDir, { recursive: true });
for (const t of targets) {
  console.log('====', t.tag, '====');
  if (!fs.existsSync(t.p)) {
    console.log('[missing]', t.p);
    continue;
  }
  const txt = docxText(t.p);
  const out = path.join(outDir, t.tag + '.txt');
  fs.writeFileSync(out, txt, 'utf-8');
  console.log('[chars]', txt.length, '-> ', out);
}