Skip to main content

Middleware to simplify caching your resources.

Project description

Chocs-cache
PyPI version CI Release License: MIT

Cache middleware for chocs library.

Features

  • Easy integration
  • Support for ETags
  • Support for conditional request headers if-none-match, if-match
  • Built-in in-memory cache storage for debugging and testing purposes
  • Automatic cache revalidation

Installation

With pip,

pip install chocs-middleware.cache

or through poetry

poetry add chocs-middleware.cache

Usage

Simple cache

The following example shows the simplest usage of cache mechanism with a custom in-memory cache storage.

import chocs
from chocs import HttpRequest, HttpResponse
from chocs_middleware.cache import ICacheStorage, CacheItem, CacheMiddleware


class MemoryCache(ICacheStorage):
    """
    Custom cache storage that uses memory to store the cache items.
    In production, this should use Redis or other cache databases used by your application.
    """

    def __init__(self):
        self._cache = {}

    def get(self, cache_id: str) -> CacheItem:
        return self._cache[cache_id]

    def set(self, item: CacheItem) -> None:
        self._cache[item.id] = item


app = chocs.Application(CacheMiddleware(MemoryCache()))


@app.get("/users/{user_id}", cache_expiry=10)
def get_user(request: HttpRequest) -> HttpResponse:
    """
    cache_expiry enables caching for the returned response and keeps
    it alive for 10 seconds. 
    """

    return HttpResponse("Bob Bobber", 200)

ETag based cache

To make use of e-tags simply return the e-tag header in the response, like in the example below:

import chocs
from chocs import HttpRequest, HttpResponse
from chocs_middleware.cache import CacheMiddleware, InMemoryCacheStorage

app = chocs.Application(CacheMiddleware(InMemoryCacheStorage()))


@app.get("/users/{user_id}", cache_expiry=10)
def get_user(request: HttpRequest) -> HttpResponse:
    return HttpResponse("Bob Bobber", headers={
        "etag": '"etag-1"', # your custom etag value
    })

Keep in mind e-tag values MUST BE unique per REST endpoint and per content type to avoid cache collisions.

Using cache vary

To allow the cache system better understand your intention it is recommended to use the cache_vary attribute. You can read more about the Vary header here

import chocs
from chocs import HttpRequest, HttpResponse
from chocs_middleware.cache import CacheMiddleware, InMemoryCacheStorage

# you can specify cache vary globally when initialising middleware
app = chocs.Application(CacheMiddleware(InMemoryCacheStorage(), cache_vary=set("accept-language")))

# or you can specify cache vary per endpoint
@app.get("/users/{user_id}", cache_expiry=10, cache_vary=("accept-language", "x-custom-header"))
def get_user(request: HttpRequest) -> HttpResponse:
    return HttpResponse("Bob Bobber")

Specifying cache control

You can also specify the type of cache by setting the cache_control attribute to public or private. You can read more about cache-control here

import chocs
from chocs import HttpRequest, HttpResponse
from chocs_middleware.cache import CacheMiddleware, InMemoryCacheStorage

app = chocs.Application(CacheMiddleware(InMemoryCacheStorage()))

# cache control can only be specified in the route definition.
@app.get("/users/{user_id}", cache_expiry=10, cache_control="private")
def get_user(request: HttpRequest) -> HttpResponse:
    return HttpResponse("Bob Bobber")

Conditional request support

This cache system supports conditional requests headers if-none-match and if-match in a limited manner. This means values passed in if-none-match and if-match headers will be treated as a single value.

Cache revalidation

The middleware provides ICollectableCache interfaces that you can use to implement your
cache re-validation. It might be just a simple mechanism that just deletes the stale cache, but there are no limitations and more powerful systems might be built on the top of the interface.

The following diagram represents the way how to cache revalidation works in the middleware: Cache Revalidation

The following example shows the simplest implementation of cache re-validation:

from chocs import Application, HttpRequest, HttpResponse
from chocs_middleware.cache import ICollectableCacheStorage, CacheItem, CacheMiddleware


class CacheStorage(ICollectableCacheStorage):
    def __init__(self):
        self._storage = {}

    def set(self, item: CacheItem) -> None:
        self._storage[item.id] = item

    def get(self, item_id: str) -> CacheItem:
        return self._storage[item_id]

    def collect(self, item: CacheItem) -> None:
        del self._storage[item.id]

app = Application(CacheMiddleware(CacheStorage))

@app.get("/users/{user_id}", cache_expiry=10)
def get_test(req: HttpRequest) -> HttpResponse:
    user_id = req.path_parameters.get("user_id")
    return HttpResponse(f"user {user_id}", headers={"etag": f"{user_id}"})


@app.patch("/users/{user_id}", cache=True)  # cache attribute enables cache middleware for the endpoint, to collect stale data
def get_test(req: HttpRequest) -> HttpResponse:
    """
    Once this function generate a successful response and cache exists for the given user CacheStorage.collect is called
    """
    return HttpResponse("{user data}")

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

chocs_middleware.cache-1.0.0.tar.gz (8.9 kB view details)

Uploaded Source

File details

Details for the file chocs_middleware.cache-1.0.0.tar.gz.

File metadata

  • Download URL: chocs_middleware.cache-1.0.0.tar.gz
  • Upload date:
  • Size: 8.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.1.13 CPython/3.8.12 Linux/5.11.0-1028-azure

File hashes

Hashes for chocs_middleware.cache-1.0.0.tar.gz
Algorithm Hash digest
SHA256 2573f0443df1b234acefe41f557f070b1a4bbdaad0d562ed5762988f9b7ff305
MD5 fe6f7582eb80a5f043cde8efe3df2110
BLAKE2b-256 8240642909881a03bae91ab9733a322e97b18c393d26257f5518983a79f68a77

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page