#!/usr/bin/env node
/**
 * Build search-index.json from chapter HTML files.
 *
 *   docs/prd/chapters/{ch}/**.html → assets/search-index.json
 *
 * Tokenization rules (must match runtime in app.js):
 *   - lowercase
 *   - ASCII words/digits: [a-z0-9_]+
 *   - CJK bigrams: [一-鿿]+ → for each run, push every 2-char window
 *
 * Output:
 *   {
 *     docs: [ { id, title, href, crumbs:[…], text } ],
 *     invIndex: { token: [ [docId, tf], … ] }
 *   }
 */

const fs   = require('fs');
const path = require('path');

const ROOT      = path.resolve(__dirname, '..');
const CHAPT_DIR = path.join(ROOT, 'chapters');
const OUT       = path.join(ROOT, 'assets', 'search-index.json');

function walk(dir) {
  const out = [];
  for (const e of fs.readdirSync(dir, { withFileTypes: true })) {
    const p = path.join(dir, e.name);
    if (e.isDirectory()) out.push(...walk(p));
    else if (e.isFile() && e.name.endsWith('.html')) out.push(p);
  }
  return out;
}

function stripTags(html) {
  return html
    .replace(/<script[\s\S]*?<\/script>/gi, ' ')
    .replace(/<style[\s\S]*?<\/style>/gi, ' ')
    .replace(/<!--[\s\S]*?-->/g, ' ')
    .replace(/<[^>]+>/g, ' ')
    .replace(/&nbsp;/g, ' ')
    .replace(/&amp;/g, '&')
    .replace(/&lt;/g, '<')
    .replace(/&gt;/g, '>')
    .replace(/&quot;/g, '"')
    .replace(/&#39;/g, "'")
    .replace(/\s+/g, ' ')
    .trim();
}

function pick(html, re) {
  const m = re.exec(html);
  return m ? stripTags(m[1]) : null;
}

function extractTitle(html) {
  return pick(html, /<h1[^>]*>([\s\S]*?)<\/h1>/i) || pick(html, /<title[^>]*>([\s\S]*?)<\/title>/i) || '(无标题)';
}

function extractCrumbs(html) {
  const m = /<div class="crumbs"[^>]*>([\s\S]*?)<\/div>/i.exec(html);
  if (!m) return [];
  return m[1].replace(/<span[^>]*>›<\/span>/g, '|').split('|').map(s => stripTags(s)).filter(Boolean);
}

function tokenize(text) {
  const tokens = [];
  const lower = text.toLowerCase();
  (lower.match(/[a-z0-9_]+/g) || []).forEach(w => tokens.push(w));
  const cjk = lower.match(/[㐀-鿿]+/g) || [];
  cjk.forEach(run => {
    if (run.length === 1) tokens.push(run);
    else for (let i = 0; i < run.length - 1; i++) tokens.push(run.slice(i, i + 2));
  });
  return tokens;
}

function build() {
  const files = walk(CHAPT_DIR);
  const docs = [];
  const inv  = {};   // token → Map<docId, tf>

  for (const file of files) {
    const html = fs.readFileSync(file, 'utf-8');
    const title  = extractTitle(html);
    const crumbs = extractCrumbs(html);
    const text   = stripTags(html);
    const href   = path.relative(CHAPT_DIR, file);
    const id     = docs.length;
    docs.push({ id, title, href, crumbs, text: text.slice(0, 2000) });

    const tokens = tokenize(text);
    const tf = {};
    for (const t of tokens) tf[t] = (tf[t] || 0) + 1;
    for (const [t, c] of Object.entries(tf)) {
      (inv[t] = inv[t] || []).push([id, c]);
    }
  }

  const out = { docs, invIndex: inv };
  fs.mkdirSync(path.dirname(OUT), { recursive: true });
  fs.writeFileSync(OUT, JSON.stringify(out));
  const kb = (fs.statSync(OUT).size / 1024).toFixed(1);
  console.log(`✅ search index: ${docs.length} docs, ${Object.keys(inv).length} tokens, ${kb} KB → ${path.relative(ROOT, OUT)}`);
}

build();