fastapi-cachex 0.2.2

A caching library for FastAPI with support for Cache-Control, ETag, and multiple backends.

FastAPI-Cache X

English | 繁體中文

A high-performance caching extension for FastAPI, providing comprehensive HTTP caching support and optional session management.

Features

HTTP Caching

• Support for HTTP caching headers
  • Cache-Control
  • ETag
  • If-None-Match
• Multiple backend cache support
  • Redis
  • Memcached
  • In-memory cache
• Complete Cache-Control directive implementation
• Easy-to-use @cache decorator

Session Management (Optional Extension)

• Secure session management with HMAC-SHA256 token signing
• IP address and User-Agent binding (optional security features)
• Header and bearer token support (API-first architecture)
• Automatic session renewal (sliding expiration)
• Flash messages for cross-request communication
• Multiple backend support (Redis, Memcached, In-Memory)
• Complete session lifecycle management (create, validate, refresh, invalidate)

Cache-Control Directives

Directive	Supported	Description
max-age	:white_check_mark:	Specifies the maximum amount of time a resource is considered fresh.
s-maxage	:x:	Specifies the maximum amount of time a resource is considered fresh for shared caches.
no-cache	:white_check_mark:	Forces caches to submit the request to the origin server for validation before releasing a cached copy.
no-store	:white_check_mark:	Instructs caches not to store any part of the request or response.
no-transform	:x:	Instructs caches not to transform the response content.
must-revalidate	:white_check_mark:	Forces caches to revalidate the response with the origin server after it becomes stale.
proxy-revalidate	:x:	Similar to must-revalidate, but only for shared caches.
must-understand	:x:	Indicates that the recipient must understand the directive or treat it as an error.
private	:white_check_mark:	Indicates that the response is intended for a single user and should not be stored by shared caches.
public	:white_check_mark:	Indicates that the response may be cached by any cache, even if it is normally non-cacheable.
immutable	:white_check_mark:	Indicates that the response body will not change over time, allowing for longer caching.
stale-while-revalidate	:white_check_mark:	Indicates that a cache can serve a stale response while it revalidates the response in the background.
stale-if-error	:white_check_mark:	Indicates that a cache can serve a stale response if the origin server is unavailable.

Installation

uv add fastapi-cachex

Quick Start

from fastapi import FastAPI
from fastapi_cachex import cache
from fastapi_cachex import CacheBackend

app = FastAPI()


@app.get("/")
@cache(ttl=60)  # Cache for 60 seconds
async def read_root():
    return {"Hello": "World"}


@app.get("/no-cache")
@cache(no_cache=True)  # Mark this endpoint as non-cacheable
async def non_cache_endpoint():
    return {"Hello": "World"}


@app.get("/no-store")
@cache(no_store=True)  # Mark this endpoint as non-cacheable
async def non_store_endpoint():
    return {"Hello": "World"}


@app.get("/clear_cache")
async def remove_cache(cache: CacheBackend):
    await cache.clear_path("/path/to/clear")  # Clear cache for a specific path
    await cache.clear_pattern("/path/to/clear/*")  # Clear cache for a specific pattern

Backend Configuration

FastAPI-CacheX supports multiple caching backends. You can easily switch between them using the BackendProxy.

Cache Key Format

Cache keys are generated in the following format to avoid collisions:

{method}:{host}:{path}:{query_params}

This ensures that:

• Different HTTP methods (GET, POST, etc.) don't share cache
• Different hosts don't share cache (useful for multi-tenant scenarios)
• Different query parameters get separate cache entries
• The same endpoint with different parameters can be cached independently

All backends automatically namespace keys with a prefix (e.g., fastapi_cachex:) to avoid conflicts with other applications.

Cache Hit Behavior

When a cached entry is valid (within TTL):

• Default behavior: Returns the cached content with HTTP 200 status code directly without re-executing the endpoint handler
• With If-None-Match header: Returns HTTP 304 Not Modified if the ETag matches
• With no-cache directive: Forces revalidation with fresh content before deciding on 304

This means cached hits are extremely fast - the endpoint handler function is never executed.

In-Memory Cache (default)

If you don't specify a backend, FastAPI-CacheX will use the in-memory cache by default. This is suitable for development and testing purposes. The backend automatically runs a cleanup task to remove expired entries every 60 seconds.

from fastapi_cachex.backends import MemoryBackend
from fastapi_cachex import BackendProxy

backend = MemoryBackend()
BackendProxy.set_backend(backend)

Note: In-memory cache is not suitable for production with multiple processes. Each process maintains its own separate cache.

Memcached

from fastapi_cachex.backends import MemcachedBackend
from fastapi_cachex import BackendProxy

backend = MemcachedBackend(servers=["localhost:11211"])
BackendProxy.set_backend(backend)

Limitations:

• Pattern-based key clearing (clear_pattern) is not supported by the Memcached protocol
• Keys are namespaced with fastapi_cachex: prefix to avoid conflicts
• Consider using Redis backend if you need pattern-based cache clearing

Redis

from fastapi_cachex.backends import AsyncRedisCacheBackend
from fastapi_cachex import BackendProxy

backend = AsyncRedisCacheBackend(host="127.0.0.1", port=6379, db=0)
BackendProxy.set_backend(backend)

Features:

• Fully async implementation
• Supports pattern-based key clearing
• Uses SCAN instead of KEYS for safe production use (non-blocking)
• Namespaced with fastapi_cachex: prefix by default
• Optional custom key prefix for multi-tenant scenarios

Example with custom prefix:

backend = AsyncRedisCacheBackend(
    host="127.0.0.1",
    port=6379,
    key_prefix="myapp:cache:",
)
BackendProxy.set_backend(backend)

Performance Considerations

Cache Hit Performance

When a cache hit occurs (within TTL), the response is returned directly without executing your endpoint handler. This is extremely fast:

@app.get("/expensive")
@cache(ttl=3600)  # Cache for 1 hour
async def expensive_operation():
    # This is ONLY executed when cache misses
    # On cache hits, this function is never called
    result = perform_expensive_calculation()
    return result

Backend Selection

• MemoryBackend: Fastest for single-process development; not suitable for production
• Memcached: Good for distributed systems; has limitations on pattern clearing
• Redis: Best for production; fully async, supports all features, non-blocking operations

Documentation

• Cache Flow Explanation
• Development Guide
• Contributing Guidelines
• Session Management Guide - Complete guide for session features

License

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