Purlin.CentralAuth.Provider 1.0.7

Purlin.CentralAuth.Provider

Purlin.CentralAuth.Provider provides a lightweight and extensible way to interact with a Central Auth Provider API using a familiar Refit-like experience. It includes built-in token acquisition and caching mechanisms to improve performance and reduce redundant authentication requests.

🚀 Features

• Refit-like interface for making API requests
• Built-in token acquisition using client credentials
• Token caching strategies:
  • None – always requests a new token
  • MemoryCache – stores token in local memory
  • Redis – stores token in a shared Redis instance (ideal for distributed systems)

📦 Installation

dotnet add package Purlin.CentralAuth.Provider

🛠️ Configuration

Start by defining your authentication and caching settings:

var providerUrl = "https://auth-provider.identity.com";

var authSettings = new AuthSettings
{
    Authority = "https://auth.identity.com",
    ClientId = "client_id",
    ClientSecret = "client_secret",

    TokenCachingMethod = TokenCacheMethodEnum.Redis, // Choose None, MemoryCache, or Redis
    RedisSettings = new RedisSettings
    {
        Endpoint = "redis_endpoint",
        Password = "redis_password",
        Database = 0
    }
};

builder.Services.AddCentralAuthProviderServices(authSettings, providerAPIBaseUrl);

Below are the available caching method descriptions:

• None – You will need to handle all token-related operations manually through a custom DelegatingHandler.
• MemoryCache – Ideal for systems not running behind load balancers. If using a load balancer, avoid this option to prevent inconsistent caching.
• Redis – Recommended for load-balanced environments. Tokens are stored in a centralized Redis instance and shared across service instances. Note: When TokenCachingMethod is set to Redis, the RedisSettings section is required.

The Redis instance used won’t interfere with any existing Redis registrations. It is registered as a keyed singleton using an internal key known only to the package.

This package also allows you to configure a global timeout for all outgoing requests. By default, each request will timeout after 10 seconds, but you can customize this value if needed.

📄 Pagination Notes

All endpoints that support pagination use the following query parameters: page – the current page number (starting at 1, not 0) pageSize – the number of items to return per page