anotherreflections-website-v2/scripts/migrate-wp.mjs

#!/usr/bin/env node
import fs from 'node:fs';
import path from 'node:path';
import { fileURLToPath } from 'node:url';

const __dirname = path.dirname(fileURLToPath(import.meta.url));
const ROOT = path.resolve(__dirname, '..');
const EXPORT_PATH = path.join(ROOT, '_wp-export.json');
const POSTS_DIR = path.join(ROOT, 'src/content/posts');
const PAGES_DIR = path.join(ROOT, 'src/content/pages');

const data = JSON.parse(fs.readFileSync(EXPORT_PATH, 'utf8'));

fs.mkdirSync(POSTS_DIR, { recursive: true });
fs.mkdirSync(PAGES_DIR, { recursive: true });

const decodeEntities = (s) => s
  .replace(/ /g, ' ')
  .replace(/«/g, '«')
  .replace(/»/g, '»')
  .replace(/—/g, '—')
  .replace(/–/g, '–')
  .replace(/…/g, '…')
  .replace(/"/g, '"')
  .replace(/'/g, "'")
  .replace(/'/g, "'")
  .replace(/&lt;/g, '<')
  .replace(/&gt;/g, '>')
  .replace(/&amp;/g, '&');

/** Заменить старые WP-uploads URL на относительные (мы скачали в public/wp-content/uploads/). */
const rewriteWpUploads = (url) =>
  url.replace(/^https?:\/\/anotherreflections\.ru(\/wp-content\/uploads\/.+)$/i, '$1');

const htmlToMd = (html) => {
  if (!html) return '';
  let s = html;
  // Сначала "вложенные" <a href="..."><img src="..."></a> — превращаем в один markdown-image со ссылкой
  s = s.replace(/<a\s+[^>]*?href=["']([^"']+)["'][^>]*>\s*<img[^>]*?src=["']([^"']+)["'][^>]*?(?:alt=["']([^"']*)["'])?[^>]*?\/?>\s*<\/a>/gi,
    (_, href, src, alt) => `[![${alt || ''}](${rewriteWpUploads(src)})](${rewriteWpUploads(href)})`);
  // одиночные images
  s = s.replace(/<img[^>]*?src=["']([^"']+)["'][^>]*?(?:alt=["']([^"']*)["'])?[^>]*?\/?>/gi,
    (_, src, alt) => `![${alt || ''}](${rewriteWpUploads(src)})`);
  // links
  s = s.replace(/<a\s+[^>]*?href=["']([^"']+)["'][^>]*>([\s\S]*?)<\/a>/gi,
    (_, href, text) => `[${text.replace(/<[^>]+>/g,'').trim()}](${rewriteWpUploads(href)})`);
  // bold
  s = s.replace(/<(strong|b)\b[^>]*>([\s\S]*?)<\/\1>/gi, '**$2**');
  // italic
  s = s.replace(/<(em|i)\b[^>]*>([\s\S]*?)<\/\1>/gi, '*$2*');
  // blockquote
  s = s.replace(/<blockquote[^>]*>([\s\S]*?)<\/blockquote>/gi,
    (_, inner) => inner.trim().split(/\n+/).map(l => '> ' + l.trim()).join('\n'));
  // lists
  s = s.replace(/<li[^>]*>([\s\S]*?)<\/li>/gi, '- $1\n');
  s = s.replace(/<\/?(ul|ol)[^>]*>/gi, '\n');
  // paragraphs
  s = s.replace(/<p[^>]*>/gi, '\n\n');
  s = s.replace(/<\/p>/gi, '\n\n');
  // br
  s = s.replace(/<br\s*\/?>/gi, '\n');
  // strip remaining span/div etc.
  s = s.replace(/<\/?(?:span|div|font|center)[^>]*>/gi, '');
  // any remaining tags — drop
  s = s.replace(/<[^>]+>/g, '');
  // entities
  s = decodeEntities(s);
  // collapse 3+ blank lines
  s = s.replace(/\n{3,}/g, '\n\n');
  return s.trim();
};

const yamlEscape = (s) => {
  if (s == null) return '""';
  const str = String(s);
  if (/^[\w\-.,!?:; À-￿()«»—–"']+$/.test(str) && !/^[-?:&*!|>%@`]/.test(str)) {
    return JSON.stringify(str);
  }
  return JSON.stringify(str);
};

const fmtFrontmatter = (obj) => {
  const lines = ['---'];
  for (const [k, v] of Object.entries(obj)) {
    if (v === undefined || v === null) continue;
    if (Array.isArray(v)) {
      if (v.length === 0) continue;
      lines.push(`${k}:`);
      v.forEach(item => lines.push(`  - ${yamlEscape(item)}`));
    } else if (typeof v === 'object') {
      lines.push(`${k}:`);
      Object.entries(v).forEach(([kk, vv]) => lines.push(`  ${kk}: ${yamlEscape(vv)}`));
    } else {
      lines.push(`${k}: ${yamlEscape(v)}`);
    }
  }
  lines.push('---');
  return lines.join('\n');
};

const authorsMap = Object.fromEntries(data.authors.map(a => [a.id, a.display_name || a.login]));

// Posts
let postCount = 0;
for (const p of data.posts) {
  const fm = {
    title: p.title,
    pubDate: p.date.replace(' ', 'T') + '+03:00',
    updatedDate: p.modified && p.modified !== p.date ? p.modified.replace(' ','T') + '+03:00' : undefined,
    slug: p.slug,
    legacyId: p.id,
    author: authorsMap[p.author_id] || 'admin',
    categories: p.categories.map(c => c.name),
    categorySlugs: p.categories.map(c => c.slug),
    tags: p.tags.map(t => t.name),
    description: (htmlToMd(p.excerpt) || htmlToMd(p.content_html))
      .replace(/!?\[([^\]]*)\]\([^)]+\)/g, '$1')  // strip markdown images/links для excerpt
      .replace(/[*_>#]+/g, '')                     // strip md formatting символов
      .replace(/\s+/g, ' ')
      .trim()
      .slice(0, 200),
  };
  const body = htmlToMd(p.content_html);
  const out = `${fmtFrontmatter(fm)}\n\n${body}\n`;
  const safeName = p.slug.replace(/[^a-z0-9а-я\-]/gi, '-').slice(0, 80) || `post-${p.id}`;
  fs.writeFileSync(path.join(POSTS_DIR, `${safeName}.md`), out, 'utf8');
  postCount++;
}

// Pages
let pageCount = 0;
for (const pg of data.pages) {
  const fm = {
    title: pg.title,
    slug: pg.slug,
    legacyId: pg.id,
    menuOrder: pg.menu_order,
    pubDate: pg.date.replace(' ', 'T') + '+03:00',
    updatedDate: pg.modified !== pg.date ? pg.modified.replace(' ','T') + '+03:00' : undefined,
  };
  const body = htmlToMd(pg.content_html);
  const out = `${fmtFrontmatter(fm)}\n\n${body}\n`;
  const safeName = pg.slug.replace(/[^a-z0-9а-я\-]/gi, '-').slice(0, 80) || `page-${pg.id}`;
  fs.writeFileSync(path.join(PAGES_DIR, `${safeName}.md`), out, 'utf8');
  pageCount++;
}

// Categories dump (для построения /category/<slug>/ страниц и фидов)
const catsOut = data.categories.map(c => ({
  name: c.name, slug: c.slug, description: c.description, count: c.count
}));
fs.writeFileSync(path.join(ROOT, 'src/content/_categories.json'), JSON.stringify(catsOut, null, 2), 'utf8');

console.log(`Migrated ${postCount} posts → ${POSTS_DIR}`);
console.log(`Migrated ${pageCount} pages → ${PAGES_DIR}`);
console.log(`Categories: ${catsOut.length} → src/content/_categories.json`);