StrataDB

A type-safe embedded document database for Bun. No separate server process, no network overhead - just import and use.

Why Embedded?

Unlike MongoDB or other client-server databases, StrataDB runs in-process with your application:

• Zero setup - No database server to install or manage
• Zero latency - Direct memory access, no network round-trips
• Single file - Your entire database is one portable .db file
• Serverless ready - Perfect for edge functions, CLI tools, and desktop apps

Features

• Embedded SQLite - Runs in-process via bun:sqlite, no external dependencies
• Full Type Safety - Compile-time validation of queries, schemas, and results
• MongoDB-like API - Familiar operators: $eq, $gt, $in, $and, $or, etc.
• JSONB Storage - Flexible documents with indexed generated columns
• Portable - Single file database, easy to backup and deploy
• Text Search - Multi-field search with case-insensitive matching
• Cursor Pagination - Efficient, stable pagination for large datasets
• Field Projection - Clean select/omit helpers for controlling returned fields

Installation

bun add stratadb

Requires Bun runtime.

Note: StrataDB is ESM-only. It does not provide CommonJS exports.

Quick Start

import { Strata, createSchema, type Document } from 'stratadb'

type User = Document<{
  name: string
  email: string
  age: number
}>

const schema = createSchema<User>()
  .field('name', { type: 'TEXT', indexed: true })
  .field('email', { type: 'TEXT', indexed: true, unique: true })
  .field('age', { type: 'INTEGER', indexed: true })
  .timestamps(true)
  .build()

using db = new Strata({ database: 'app.db' })
const users = db.collection('users', schema)

// Insert
const user = await users.insertOne({
  name: 'Alice',
  email: '[email protected]',
  age: 30
})

// Query
const adults = await users.find({
  age: { $gte: 18 },
  email: { $endsWith: '@example.com' }
})

// Update
await users.updateOne(user._id, { age: 31 })

// Delete
await users.deleteOne(user._id)

// Text search across fields
const results = await users.search('alice', ['name', 'email'])

// Cursor pagination for large datasets
const page1 = await users.find({}, { sort: { createdAt: -1 }, limit: 20 })
const page2 = await users.find(
  {},
  { sort: { createdAt: -1 }, limit: 20, cursor: { after: page1.at(-1)?._id } }
)

// Field projection
const names = await users.find({}, { select: ['name', 'email'] })
const safe = await users.find({}, { omit: ['password'] })

Type Safety

TypeScript validates schema types against field definitions:

// Correct - types match
db.collection<User>('users')
  .field('name', { type: 'TEXT' })      // string -> TEXT
  .field('age', { type: 'INTEGER' })    // number -> INTEGER
  .build()

// Compile error - type mismatch
db.collection<User>('users')
  .field('name', { type: 'INTEGER' })   // Error: string cannot use INTEGER
  .build()

Documentation

See the documentation for complete guides:

• Getting Started
• Collections
• Queries
• Schemas
• Transactions

Development

bun install      # Install dependencies
bun test         # Run tests
bun run build    # Build
bun run typecheck # Type check

License

MIT