Cache

Pokie provides a cache abstraction layer with multiple backends. All cache implementations follow the CacheInterface from Rick, providing a consistent API regardless of the backend.

Cache Interface

All cache backends implement the following methods:

Method	Description
get(key)	Retrieve a value by key. Returns None if not found or expired.
set(key, value, ttl=None)	Store a value with an optional TTL (in seconds).
has(key)	Check if a key exists and is not expired.
remove(key)	Remove a key from the cache.
purge()	Clear all cached entries.

Available Backends

RedisCache

Production cache backend using Redis. Requires RedisFactory and CacheFactory to be registered (see Factories).

from pokie.cache import RedisCache

RedisCache wraps the Rick RedisCache class and retrieves the Redis client from the DI container (DI_REDIS). If DI_REDIS is not available, a RuntimeError is raised.

MemoryCache

In-memory cache for unit testing. Supports TTL via time.monotonic() and uses pickle for value serialization.

from pokie.cache import MemoryCache

Note: MemoryCache is intended for testing only. Do not use it in production environments.

Additional methods:

Method	Description
set_prefix(prefix)	Set a key prefix for namespacing.

DummyCache

No-op cache implementation where all operations do nothing. Useful for development or when caching should be disabled.

from pokie.cache import DummyCache

All get() calls return None, and all set()/remove()/purge() calls are silently ignored.

Cache Setup

To use Redis-based caching, include both RedisFactory and CacheFactory in your factory list:

from pokie.core.factories.redis import RedisFactory
from pokie.core.factories.cache import CacheFactory

factories = [
    PgSqlFactory,
    RedisFactory,
    CacheFactory,
]

This registers a RedisCache instance as DI_CACHE in the DI container.