import fs from 'fs';
import path from 'path';
import { fileURLToPath } from 'url';

const __dirname = path.dirname(fileURLToPath(import.meta.url));
const dir = path.join(__dirname, '..', 'content', 'resources');

let fixed = 0;
for (const f of fs.readdirSync(dir).filter(f => f.endsWith('.mdx'))) {
  const filepath = path.join(dir, f);
  const original = fs.readFileSync(filepath, 'utf8');
  let content = original;

  // Clean excerpts that contain HTML/CSS artifacts
  const excerptMatch = content.match(/^excerpt: "(.*)"$/m);
  if (excerptMatch) {
    let excerpt = excerptMatch[1];
    excerpt = excerpt.replace(/#block-[^\s]+ \{[^}]*\}/g, '');
    excerpt = excerpt.replace(/!\[[^\]]*\]\([^)]*\)/g, '');
    excerpt = excerpt.replace(/\[([^\]]*)\]\([^)]*\)/g, '$1');
    excerpt = excerpt.replace(/\*\*/g, '');
    excerpt = excerpt.replace(/\s+/g, ' ').trim().substring(0, 250);
    if (excerpt !== excerptMatch[1]) {
      content = content.replace(excerptMatch[0], `excerpt: "${excerpt}"`);
    }
  }

  // Remove BACK TO POSTS lines
  content = content.replace(/\[?\s*\*?\*?BACK TO POSTS\s*\*?\*?\s*\]?\s*\([^)]*\)\s*(\*?\*?\s*\|?\s*\*?\*?\s*\[?\*?\*?OPEN PAGE AS PDF\*?\*?\]?\([^)]*\))?/gi, '');

  // Remove --> artifacts
  content = content.replace(/^-->\s*$/gm, '');

  // Remove CSS block definitions
  content = content.replace(/#block-[a-zA-Z0-9_-]+ \{[^}]*\}/g, '');

  // Clean excessive blank lines
  content = content.replace(/\n{3,}/g, '\n\n');

  if (content !== original) {
    fs.writeFileSync(filepath, content);
    fixed++;
  }
}
console.log(`Cleaned ${fixed} files`);