import { readFileSync } from 'node:fs';
const raw = readFileSync(process.argv[2], 'utf8');
let t = raw
  .replace(/<script\b[\s\S]*?<\/script>/gi, ' ')
  .replace(/<style\b[\s\S]*?<\/style>/gi, ' ')
  .replace(/<noscript\b[\s\S]*?<\/noscript>/gi, ' ')
  .replace(/<!--[\s\S]*?-->/g, ' ')
  .replace(/<[^>]+>/g, ' ');
const ents = { amp:'&', lt:'<', gt:'>', quot:'"', apos:"'", nbsp:' ', ndash:'–', mdash:'—', rsquo:'’', lsquo:'‘', ldquo:'“', rdquo:'”' };
t = t.replace(/&([a-zA-Z]+);/g, (m,n)=> n in ents ? ents[n] : m)
     .replace(/&#(\d+);/g, (_m,d)=> String.fromCodePoint(+d))
     .replace(/&#x([0-9a-f]+);/gi, (_m,h)=> String.fromCodePoint(parseInt(h,16)));
t = t.replace(/[ \t ]+/g,' ').replace(/\s*\n\s*/g,'\n').replace(/\n{2,}/g,'\n').trim();
process.stdout.write(t);