Getting Started with Caching

About 364 wordsAbout 1 min

Overview

The Filterable caching system intelligently caches filtered query results to dramatically improve performance for frequently-used filters. It supports TTL, tags, scopes, profiles, and automatic cache invalidation.

Basic Usage

Enable caching for any filterable query by calling the cache() method:

use App\Models\Post;

// Cache results for 1 hour (3600 seconds)
$posts = Post::filter()->cache(3600)->get();

// Cache with default TTL from config
$posts = Post::filter()->cache()->get();

// Cache forever (until manually flushed)
$posts = Post::filter()->cacheForever()->get();

How It Works

The caching system works at the terminal method level, meaning it only caches when you actually retrieve data:

// ✅ These are cached (terminal methods)
$posts = Post::filter()->cache(3600)->get();
$count = Post::filter()->cache(3600)->count();
$first = Post::filter()->cache(3600)->first();
$paginated = Post::filter()->cache(3600)->paginate(10);

// ❌ These are NOT cached (builder methods)
$query = Post::filter()->cache(3600)->where('status', 'active'); // Returns query builder

Cache Key Generation

The system automatically generates unique, deterministic cache keys based on:

Filter class name
Applied filters
Provided data
Cache scopes (user, tenant, custom)
Terminal method and arguments

Example cache key:

filterable:post_filter:a1b2c3d4:e5f6g7h8:v1:get:abc123

Configuration

Enable caching globally in config/filterable.php:

'cache' => [
    'enabled' => true,
    'driver' => null, // Uses default Laravel cache driver
    'default_ttl' => 3600, // 1 hour
    'prefix' => 'filterable',
    'version' => 'v1', // Increment to invalidate all caches
],

Control via environment variables:

FILTERABLE_CACHE_ENABLED=true
FILTERABLE_CACHE_DRIVER=redis
FILTERABLE_CACHE_TTL=3600
FILTERABLE_CACHE_PREFIX=filterable
FILTERABLE_CACHE_VERSION=v1

When to Use Caching

✅ Good Use Cases:

Frequently accessed data that changes infrequently
Heavy reports and analytics queries
Dashboard widgets with complex filters
Public-facing filtered lists
Search results with stable criteria

❌ Avoid Caching For:

Real-time data that must be always fresh
User-specific data without proper scoping
One-time queries
Simple queries that are already fast

Performance Impact

Typical performance gains:

Scenario	Without Cache	With Cache	Improvement
Complex filters (5+ conditions)	250ms	2ms	125x faster
Joins with aggregations	500ms	2ms	250x faster
Paginated results	150ms	2ms	75x faster
Simple filters	50ms	2ms	25x faster

Next Steps

Invokable

Annotations