elegalsoftware/scripts/migrate-storage-to-spaces.ts

// One-time migration: copy locally-stored documents into DigitalOcean Spaces.
// Idempotent — re-running skips objects already present in the bucket.
//
//   npx tsx scripts/migrate-storage-to-spaces.ts            # migrate
//   npx tsx scripts/migrate-storage-to-spaces.ts --dry-run  # report only, no writes
//
// After the app is switched to Spaces (lib/storage.ts), this exists only to lift any files
// that were written to the old local STORAGE_PATH on the server before the cutover.

import fs from 'node:fs';
import path from 'node:path';
import { fileURLToPath } from 'node:url';
import dotenv from 'dotenv';

const __dirname = path.dirname(fileURLToPath(import.meta.url));
dotenv.config({ path: path.resolve(__dirname, '../.env') });

import {
  S3Client,
  PutObjectCommand,
  HeadObjectCommand,
} from '@aws-sdk/client-s3';
import { getDb, getPool, documents } from '@lawdesk/db';

const DRY_RUN = process.argv.includes('--dry-run');

const BUCKET = process.env.SPACES_BUCKET!;
const STORAGE_PATH = process.env.STORAGE_PATH ?? './storage';

const s3 = new S3Client({
  endpoint: process.env.SPACES_ENDPOINT,
  region: process.env.SPACES_REGION,
  credentials: {
    accessKeyId: process.env.SPACES_KEY!,
    secretAccessKey: process.env.SPACES_SECRET!,
  },
  forcePathStyle: false,
});

async function existsInBucket(key: string): Promise<boolean> {
  try {
    await s3.send(new HeadObjectCommand({ Bucket: BUCKET, Key: key }));
    return true;
  } catch {
    return false;
  }
}

async function main() {
  console.log(`Migrating local documents → Spaces bucket "${BUCKET}"${DRY_RUN ? '  (dry run)' : ''}`);
  console.log(`Local root: ${path.resolve(STORAGE_PATH)}\n`);

  const db = getDb();
  const rows = await db
    .select({
      storageKey: documents.storageKey,
      name: documents.name,
      mimeType: documents.mimeType,
      sizeBytes: documents.sizeBytes,
    })
    .from(documents);

  if (rows.length === 0) {
    console.log('No document rows in the database — nothing to migrate.');
    await getPool().end();
    return;
  }

  let uploaded = 0;
  let skipped = 0;
  let missing = 0;

  for (const row of rows) {
    const localPath = path.resolve(STORAGE_PATH, row.storageKey);

    if (await existsInBucket(row.storageKey)) {
      skipped++;
      console.log(`  = already in bucket: ${row.storageKey}`);
      continue;
    }

    if (!fs.existsSync(localPath)) {
      missing++;
      console.warn(`  ! local file missing (nothing to upload): ${row.storageKey}  [${row.name}]`);
      continue;
    }

    if (DRY_RUN) {
      console.log(`  → would upload: ${row.storageKey}  (${row.sizeBytes} bytes)`);
      uploaded++;
      continue;
    }

    const body = fs.readFileSync(localPath);
    await s3.send(
      new PutObjectCommand({
        Bucket: BUCKET,
        Key: row.storageKey,
        Body: body,
        ContentType: row.mimeType,
        ACL: 'private',
      }),
    );
    uploaded++;
    console.log(`  ✓ uploaded: ${row.storageKey}  (${body.length} bytes)`);
  }

  console.log(
    `\nDone. ${uploaded} ${DRY_RUN ? 'to upload' : 'uploaded'}, ${skipped} already present, ${missing} missing locally (of ${rows.length} document rows).`,
  );
  if (missing > 0) {
    console.log(
      'Missing files have DB rows but no bytes on disk or in the bucket — they were already lost before migration (local storage was never durable).',
    );
  }
  await getPool().end();
}

main().catch((err) => {
  console.error('Migration failed:', err);
  process.exit(1);
});