daemon-sec-lotl

build-index.mjs (2964B)

---

 /**
  * `npm run build:index` — project the committed dataset down to the fields the
  * catalog list needs and write it as a content-addressed file:
  *
  *   public/data/index-<sha256[0:8]>.json   fetched by the catalog island
  *   src/data/index-hash.json               { file, hash, bytes, count } — imported
  *                                          by catalog.astro to point at the file
  *
  * The hashed name lets vercel.json serve it `immutable`; a data change makes a
  * new name, so a stale cache is impossible. Runs at the head of `npm run build`
  * and `npm run dev`; validate-data.mjs fails if index-hash.json is stale.
  */
 import { readFileSync, writeFileSync, mkdirSync, readdirSync, unlinkSync, existsSync } from 'node:fs';
 import { createHash } from 'node:crypto';
 import { fileURLToPath } from 'node:url';
 import { dirname, join, resolve } from 'node:path';
 
 const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '..');
 
 /** Mirror of LIST_FIELDS in src/lib/render-row.ts (test/build-index.test.mjs pins them). */
 export const LIST_FIELDS = [
   'id', 'toolId', 'toolName', 'name', 'source', 'platform', 'capability', 'nativeCategory',
   'command', 'description', 'usecase', 'mitre', 'privilege', 'added',
   'context', 'requires', 'services', 'environment', 'aliases', 'availability', 'verification',
   'reviewedAt', 'compatibility', 'expected', 'troubleshooting', 'sideEffects', 'restore', 'template', 'references', 'detection',
 ];
 
 export function project(t) {
   const out = {};
   for (const k of LIST_FIELDS) if (t[k] !== undefined) out[k] = t[k];
   return out;
 }
 
 /** Deterministic: same techniques → same bytes → same hash. */
 export function buildIndex(techniques) {
   const list = techniques.map(project);
   const json = JSON.stringify(list);
   const hash = createHash('sha256').update(json).digest('hex').slice(0, 8);
   return { json, hash, file: `index-${hash}.json`, bytes: Buffer.byteLength(json), count: list.length };
 }
 
 export function main() {
   const techniques = JSON.parse(readFileSync(join(ROOT, 'src/data/techniques.json'), 'utf8'));
   const idx = buildIndex(techniques);
   const pub = join(ROOT, 'public', 'data');
   mkdirSync(pub, { recursive: true });
   for (const f of readdirSync(pub)) if (/^index-[0-9a-f]{8}\.json$/.test(f) && f !== idx.file) unlinkSync(join(pub, f));
   writeFileSync(join(pub, idx.file), idx.json);
   const meta = { file: idx.file, hash: idx.hash, bytes: idx.bytes, count: idx.count };
   const metaPath = join(ROOT, 'src/data/index-hash.json');
   const prev = existsSync(metaPath) ? readFileSync(metaPath, 'utf8') : '';
   const next = JSON.stringify(meta, null, 2) + '\n';
   if (prev !== next) writeFileSync(metaPath, next);
   console.log(`build:index — public/data/${idx.file}: ${idx.count} rows, ${idx.bytes.toLocaleString('en-US')} B${prev !== next ? ' (index-hash.json updated)' : ''}`);
 }
 
 if (process.argv[1] && resolve(process.argv[1]) === fileURLToPath(import.meta.url)) main();