A high-performance, minimalist Entity Component System (ECS) library for Rust with integrated persistence capabilities.

Current Status: Phase 3 Complete (202 tests passing) - Ready for Phase 4 (Release Preparation)

• 🚀 High Performance: Archetype-based storage, cache-friendly queries, optimized for modern CPUs
• 🔄 Dual ID System: Fast ephemeral IDs for runtime, stable UUIDs for persistence
• 💾 Pluggable Persistence: Binary and JSON formats out of the box, extensible for custom formats
• 🧵 Thread-Safe Commands: Deferred operations via command buffers for parallel systems
• 📦 Zero Dependencies: Core library has no dependencies (serde optional for JSON)
• 🎯 Type-Safe Queries: Compile-time validated component access
• 📚 Library, Not Framework: Integrate into your application, don't build around it

Add PECS to your Cargo.toml:

[dependencies]
pecs = "0.1.0"

Basic usage:

use pecs::prelude::*;

// Define components
#[derive(Debug)]
struct Position {
    x: f32,
    y: f32
}
impl Component for Position {}

#[derive(Debug)]
struct Velocity {
    x: f32,
    y: f32
}
impl Component for Velocity {}

fn main() {
    // Create a world
    let mut world = World::new();

    // Spawn entities with components
    let player = world.spawn()
        .with(Position { x: 0.0, y: 0.0 })
        .with(Velocity { x: 1.0, y: 0.0 })
        .id();

    println!("Created player: {}", player);
    println!("Total entities: {}", world.len());

    // Save the world
    world.save("world.pecs").unwrap();

    // Load the world
    let loaded = World::load("world.pecs").unwrap();
    println!("Loaded {} entities", loaded.len());
}

• Getting Started Guide - Learn the basics
• Core Concepts - Understand the architecture
• Performance Guide - Optimize your usage
• API Reference - Complete API documentation
• Examples - Browse example applications

PECS is designed for high performance with extensive optimizations:

Recent Optimizations (Phase 3):

• 58% faster binary serialization (117.6µs → 48.6µs per 1000 entities)
• 2-5x faster query iteration with caching
• 10-20x faster entity location lookup (HashMap → Vec)
• 50-70% fewer allocations during entity creation

See the Performance Guide for detailed benchmarks and optimization techniques.

PECS uses an archetype-based storage system for optimal query performance:

┌─────────────────────────────────────────────┐
│                   World                      │
│  ┌────────────┐  ┌──────────────────────┐  │
│  │  Entities  │  │    Components        │  │
│  │            │  │  ┌────────────────┐  │  │
│  │ EntityId   │──┼─▶│  Archetype 1   │  │  │
│  │ StableId   │  │  │  [Pos, Vel]    │  │  │
│  │            │  │  └────────────────┘  │  │
│  │ Manager    │  │  ┌────────────────┐  │  │
│  │            │  │  │  Archetype 2   │  │  │
│  └────────────┘  │  │  [Pos, Health] │  │  │
│                  │  └────────────────┘  │  │
│  ┌────────────┐  └──────────────────────┘  │
│  │  Commands  │                             │
│  │  Buffer    │  Deferred Operations        │
│  └────────────┘                             │
└─────────────────────────────────────────────┘

Key design principles:

• Dual ID System: Fast runtime IDs + persistent UUIDs
• Archetype Storage: Cache-friendly component layout
• Command Buffers: Thread-safe deferred operations
• Pluggable Persistence: Flexible save/load system

See Core Concepts for detailed architecture information.

PECS provides built-in persistence with multiple formats:

use pecs::World;

// Binary format (fast, compact)
world.save("world.pecs") ?;
let world = World::load("world.pecs") ?;

// JSON format (human-readable)
world.save_with("world.json", "json") ?;
let world = World::load_with("world.json", "json") ?;

// Streaming API for large worlds
use std::fs::File;
let mut file = File::create("world.pecs") ?;
world.save_binary( & mut file) ?;

Features:

• Version-aware serialization
• Checksum validation
• Transient component support
• Custom plugin system

Check out the examples directory for complete applications:

• 01_hello_world.rs - Simplest PECS example
• 02_command_buffer.rs - Deferred operations
• 03_persistence.rs - Save and load worlds
• 04_performance.rs - Performance optimization techniques
• 05_large_scale.rs - 100,000 entity management
• 06_simple_game.rs - Asteroids-style game example

Run examples with:

cargo run --example 01_hello_world --release

See examples/README.md for detailed documentation.

• Entity management with dual ID system
• Component storage with archetypes
• Query system with filters
• Command buffer system
• World integration and API
• 94 tests passing

• Binary serialization format (< 0.4ms per 1000 entities)
• JSON serialization format
• Plugin system with delta persistence
• Version migration framework
• Change tracking and transient components
• 164 tests passing

• Performance optimization (58% faster serialization, 2-5x faster queries)
• Query system integration and bug fixes
• Comprehensive documentation (4,500+ lines, 100% API coverage)
• 6 working examples
• API refinement (component access, query execution)
• Error handling improvements
• 202 tests passing

• Tutorial series
• Additional complete applications
• Beta testing
• Community feedback
• Cross-platform testing
• 1.0 release

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

# Clone the repository
git clone https://github.com/yourusername/pecs.git
cd pecs

# Run tests
cargo test

# Run benchmarks
cargo bench

# Check code quality
cargo clippy
cargo fmt --check

PECS follows these principles:

1. Library, Not Framework: Integrate PECS into your application
2. Performance First: Zero-cost abstractions, cache-friendly design
3. Type Safety: Compile-time guarantees, no runtime type errors
4. Flexibility: Minimal constraints, maximum freedom
5. Simplicity: Small API surface, clear documentation

See ADR-005: Library Not Framework for details.

Licensed under either of:

• Apache License, Version 2.0 (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
• MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.

PECS is inspired by:

• Bevy ECS - Archetype-based design
• HECS - Batch entity operations
• Specs - Component storage patterns
• Legion - Query optimization
• EnTT - Entity recycling

Special thanks to the Rust gamedev community for their insights and feedback.

• Documentation: docs.rs/pecs
• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Discord: Rust Gamedev Discord

Made with ❤️ and 🦀 by Wyldlands team