const AdmZip = require('adm-zip');
const zip = new AdmZip('C:\\Users\\15850\\Downloads\\DOP生产管理用户操作手册.docx');

// Read styles.xml to understand style definitions
const stylesXml = zip.readAsText('word/styles.xml');
const styleIds = [];
const styleIdRegex = /<w:style[^>]*w:styleId="([^"]+)"[^>]*>/g;
let sm;
while ((sm = styleIdRegex.exec(stylesXml)) !== null) {
  styleIds.push(sm[1]);
}
console.log('=== Style IDs ===');
console.log(styleIds.join(', '));

// Read document.xml
const xml = zip.readAsText('word/document.xml');

// Extract all text
const allText = [];
const textRegex = /<w:t[^>]*>([^<]*)<\/w:t>/g;
let tm;
while ((tm = textRegex.exec(xml)) !== null) {
  if (tm[1].trim()) allText.push(tm[1]);
}

console.log('\n=== All text (first 150) ===');
console.log(allText.slice(0, 150).join(' | '));

console.log('\n=== All text (total ' + allText.length + ' segments) ===');
console.log(allText.join('\n'));

// Find paragraph styles used
const paraStyles = [];
const pStyleRegex = /<w:pStyle w:val="([^"]+)"/g;
let pm;
while ((pm = pStyleRegex.exec(xml)) !== null) {
  paraStyles.push(pm[1]);
}
console.log('\n=== Paragraph styles used ===');
console.log([...new Set(paraStyles)].join(', '));

// Count tables
const tableCount = (xml.match(/<w:tbl>/g) || []).length;
console.log('\n=== Table count: ' + tableCount + ' ===');

// Count images
const imageCount = (xml.match(/<w:drawing>/g) || []).length;
console.log('=== Image count: ' + imageCount + ' ===');