🚀 AdonixDB

A Minimalist, File-Based Database Engine for Node.js

Lightweight. Predictable. Typed. — AdonixDB brings Prisma-style APIs to a local JSON-file system.

⚙️ Overview

AdonixDB is a schema-validated JSON database designed for Node.js environments where simplicity and control matter most. It provides:

• 🧩 Schema enforcement with default values
• 🔗 Relations (One-to-Many & Many-to-One)
• 🧠 Prisma-inspired query API
• 💾 Zero-setup persistence via local JSON files

📦 Installation

npm install adonix-db

🗂 Directory Layout

Once connected, AdonixDB creates a folder named adonix_db_files/ at your project root. Each model maps to a single JSON file (user.json, product.json, etc.), ensuring clear separation of concerns.

📁 adonix_db_files
 ┣ 📄 user.json
 ┗ 📄 product.json

🧠 Quick Start

1️⃣ Connect & Initialize

import { AdonixDB } from 'adonix-db';

const db = new AdonixDB();

await db.connect(); // Initializes folder + loads schema
console.log('✅ Connected to AdonixDB');

2️⃣ Define Models (Schema + Relations)

Each model includes:

• Automatic fields: id, createdAt, updatedAt
• Declarative schema: type, required, default
• Optional relationship map

const UserSchema = {
  email: { type: 'string', required: true },
  username: { type: 'string', default: 'anonymous' },
  age: { type: 'number', default: 18 },
};

db.defineModel('user', UserSchema);

Relationships are defined at model registration:

const ProductSchema = {
  name: { type: 'string', required: true },
  price: { type: 'number', required: true },
  userId: { type: 'string', required: true }, // FK
};

db.defineModel('user', UserSchema, {
  products: { model: 'product', type: 'One-to-Many', foreignKey: 'userId' },
});

db.defineModel('product', ProductSchema, {
  owner: { model: 'user', type: 'Many-to-One', foreignKey: 'userId' },
});

🔍 Querying & CRUD API

All models are accessible via db.<modelName>. Each exposes a predictable async API:

| Method | Purpose | | :-------------- | :-------------------------------------------- | | .create() | Insert one record (supports nested writes) | | .findUnique() | Find one record by unique criteria | | .findMany() | Query multiple records (filter/sort/paginate) | | .update() | Update record by ID | | .delete() | Delete record by ID |

➕ Create (with Nested Write)

const newUser = await db.user.create({
  data: {
    email: '[email protected]',
    username: 'tester',
    products: {
      create: [
        { name: 'Monitor', price: 300 },
        { name: 'Keyboard', price: 50 },
      ],
    },
  },
});

🔗 Include Relations (Join)

// Many → One
const product = await db.product.findUnique({
  where: { name: 'Monitor' },
  include: { owner: true },
});

// One → Many
const user = await db.user.findUnique({
  where: { id: newUser.id },
  include: { products: true },
});

🔍 Advanced Filtering

const results = await db.product.findMany({
  where: {
    price: { gte: 50, lte: 300 },
    name: { contains: 'board' },
  },
  orderBy: { price: 'asc' },
  skip: 0,
  take: 10,
});

Supported Operators

| Operator | Description | | :----------------------- | :---------------------------- | | eq, neq | Equal / Not equal | | gt, gte, lt, lte | Numeric comparisons | | in, notin | Array membership | | contains | Substring match (string only) |

✏️ Update & Delete

await db.user.update({
  where: { id: newUser.id },
  data: { age: 25, email: '[email protected]' },
});

await db.user.delete({
  where: { id: newUser.id },
});

🧩 Why AdonixDB?

| Feature | Description | | :-------------- | :------------------------------------- | | 🪶 Lightweight | Zero dependencies, no external server | | 🧱 Schema-safe | Validates every write | | 🔄 JSON-based | Human-readable, git-friendly storage | | 💡 Familiar API | Inspired by Prisma & Mongoose patterns |

🧭 Best Practices (Senior-Level Tips)

• Keep models small & cohesive: treat each JSON file as a domain boundary.
• Never mutate raw JSON manually. Use provided API to preserve indexes and timestamps.
• Validate business logic in code, not the schema — AdonixDB ensures structure, not semantics.
• For large data (>10k records), prefer .findMany() with skip/take pagination.

🧰 Example Project Structure

📦 my-project
 ┣ 📁 adonix_db_files
 ┣ 📄 index.js
 ┗ 📄 package.json

🧾 License

MIT © 2025 — Designed for developers who love clarity and control.