OxideDB - A Lightweight Database Engine in Rust

A custom database engine implementation in Rust featuring B+ Tree indexing, persistent storage, and a SQL-like interface for basic database operations.

🚀 Features

• B+ Tree Indexing: Efficient tree-based indexing for fast data retrieval
• Persistent Storage: Data persistence across application restarts
• Multi-type Support: Support for INTEGER, BIGINT, DOUBLE, FLOAT, and STRING data types
• Table Management: Create and manage multiple tables with schema validation
• CRUD Operations: Insert, select, and query operations
• Memory Management: LRU cache implementation for optimized memory usage
• Universal Key System: Generic key handling for different data types

📁 Project Structure

src/
├── main.rs                    # Entry point and demo application
├── BPlusTree.rs              # Core B+ Tree implementation
├── BTree.rs                  # Binary tree utilities
├── BTreePersistence.rs       # B+ Tree serialization/deserialization
├── MetaEnum.rs               # Data type definitions and metadata
├── TableCreationHandler.rs   # Table schema creation and validation
├── TableQueryHandler.rs      # Query execution and data manipulation
├── TableMetaHandler.rs       # Table metadata management
├── TableBTreeManager.rs      # Universal B+ Tree management system
├── UniversalBPlusTree.rs     # Generic B+ Tree implementation
├── UniversalKey.rs           # Universal key abstraction
├── FileWriter.rs             # File I/O operations
├── RowData.rs               # Row data structures and serialization
├── LruDict.rs               # LRU cache implementation
└── Comparable.rs            # Trait for comparable types

🛠️ Installation & Setup

Prerequisites

• Rust 1.70+ (2024 edition)
• Cargo package manager

Building the Project

# Clone the repository
git clone <your-repo-url>
cd OxideDB

# Build the project
cargo build

# Run the application
cargo run

First Time Setup

On the first run, uncomment the table creation code in main.rs:

// Uncomment this on first run to create tables
create_tables();

This will create the initial database schema with sample tables.

📖 Usage

Basic Operations

The database supports the following operations:

1. Table Creation

let user_columns = vec![
    TableColumn::new("id".to_string(), Type::INTEGER, true),
    TableColumn::new("name".to_string(), Type::STRING(100), false),
    TableColumn::new("email".to_string(), Type::STRING(255), false),
];

handler.create_table_with_validation("users".to_string(), user_columns)?;

2. Data Insertion

let user_data = vec![
    DataArray::INTEGER(1),
    DataArray::STRING("Test A".to_string(), 100),
    DataArray::STRING("TestA@email.com".to_string(), 255),
];

let row = query_handler.create_row("users", user_data)?;
query_handler.insert("users".to_string(), 1, row)?;

3. Data Querying

match query_handler.select("users".to_string(), 1) {
    Ok(Some(data)) => println!("User found: {}", data),
    Ok(None) => println!("User not found"),
    Err(e) => println!("Error: {}", e),
}

Sample Application

The included demo application showcases:

• Creating users and products tables
• Inserting sample data
• Querying records by primary key
• B+ Tree performance testing
• Persistence operations

🏗️ Architecture

Core Components

1. Storage Layer: File-based persistence with page management
2. Index Layer: B+ Tree indexes for efficient data access
3. Query Layer: SQL-like operations with type validation
4. Schema Layer: Table metadata and schema management

Data Types Supported

• INTEGER (i32)
• BIGINT (i64)
• FLOAT (f32)
• DOUBLE (f64)
• STRING(length) (Variable length strings)

B+ Tree Implementation

• Generic implementation supporting multiple key types
• Persistent storage with serialization
• Efficient range queries and point lookups
• Thread-safe operations with mutex protection

🗂️ File Organization

The database creates several files for persistence:

• *.dat - Table data files
• *_btree.idx - B+ Tree index files
• table_metadata.dat - Table schema metadata
• meta_config.db - System configuration

🔧 Configuration

Key constants can be modified in main.rs:

const PAGE_SIZE: usize = 4096;    // Database page size
const HEADER_SIZE: usize = 64;    // Page header size

🚧 Current Status & Limitations

Working Features

• ✅ Table creation and schema validation
• ✅ Data insertion and retrieval
• ✅ B+ Tree indexing
• ✅ Data persistence
• ✅ Multi-type support

Known Limitations

• 🔄 No DELETE or UPDATE operations yet
• 🔄 No complex queries (JOIN, WHERE clauses)
• 🔄 No transaction support
• 🔄 Limited error recovery
• 🔄 No concurrent access control

🧪 Testing

Run the included demo:

cargo run

The application will:

1. Initialize the database system
2. Create sample tables (if uncommented)
3. Insert test data
4. Perform queries
5. Test B+ Tree functionality
6. Save state to disk

🤝 Contributing

This is an initial build of a custom database engine. Contributions are welcome!

Areas for Improvement

• Implement DELETE and UPDATE operations
• Add query optimization
• Implement transaction support
• Add concurrent access control
• Improve error handling and recovery
• Add comprehensive test suite
• Implement query planner

📝 License

This project is available under the MIT License.

🔗 Dependencies

This project has minimal dependencies and uses only Rust standard library features for maximum portability and learning value.

---

Note: This is an educational/experimental database engine.