- Optimize settings retrieval with runtime and per-class config caching.

[djabdou50]

Optimize Settings Performance with Enhanced Caching Strategy

Overview

This PR implements significant performance optimizations for the HasSettingsTable trait by introducing a multi-layered caching strategy that reduces Redis calls and improves runtime efficiency.

✨ Key Changes

Performance Improvements

• Runtime Caching: Added per-instance runtime cache ($settingsRuntimeCache) that eliminates redundant Redis calls within the same request
• Config Caching: Implemented static configuration caching to avoid repeated config() calls across instances
• Optimized Query: Enhanced __getSettingsValue() to use select() for targeted column retrieval and support eager loading to prevent N+1 queries

New Features

• Cache Management: Added clearSettingsCache() method to handle both runtime and Redis cache clearing

Developer Experience

• Redis Monitoring: Updated README with helpful Redis request counting instructions for monitoring cache performance

Performance Impact

• ZERO Redis calls after first settings access within a request (runtime cache)
• Single Redis call per request when cache is enabled (previously could be multiple)
• Reduced config lookups through static caching
• Better database query efficiency with selective column loading

Technical Details

• Maintains backward compatibility with existing API
• Preserves existing configuration-based caching behavior
• Supports eager loading for better query optimization
• Thread-safe static config caching implementation

[glorand]

@djabdou50 Seems like the pipeline is red :(

[glorand]

🎉 Thank you for this excellent contribution!
Hi @djabdou50

First, I want to express my appreciation for the time and effort you've put into this PR. The performance optimizations you've identified—particularly the runtime caching, config optimization, and N+1 query prevention—are exactly the kind of improvements this package needs.

🚀 Exciting News: Major Refactor Coming
I'm happy to share that we're planning a massive refactoring, optimization, and modernization effort for the next major version (v9.0). This will include:
Architecture Changes

• Single HasSettings trait with driver-based storage selection (replacing the current multiple-trait approach)
• Unified driver system that will make caching strategies like yours much more maintainable
• Built-in performance optimizations at the driver level

📋 Next Steps
Because of this upcoming architectural change, I'm going to close this PR rather than merge it. However, I want to emphasize:

1. Your work is not wasted - The concepts and optimizations you've identified will be incorporated into the v9.0 design
2. Your contribution matters - You've helped validate the performance pain points we need to solve
3. We'd love your continued input - If you're interested, we'd welcome your feedback on the refactor design and testing of beta releases