Hyperbee.Pipeline.Caching 1.1.6-develop.240702125702

Hyperbee.Pipeline.Caching

The Hyperbee.Pipeline.Caching library is a set of extentsions to Hyperbee.Pipeline that adds support for caching within a pipeline using Microsoft.Extensions.Caching libraries.

Memory Cache Examples

For simple pipelines the previous step's return value can be used as the key:

// Takes a string and returns a number
var command = PipelineFactory
    .Start<string>()
    .PipeCacheAsync( CharacterCountAsync )
    .Build();

var result = await command( factory.Create( logger ), "test" );

Assert.AreEqual( 4, result );

Or for more complex the options callback can be used to customize how the results will be cached.

// Takes a string and returns a number
var command = PipelineFactory
    .Start<string>()
    .PipeCacheAsync( CharacterCountAsync,
        ( input, options ) =>
        {
            options.Key = $"custom/{input}";
            options.AbsoluteExpirationRelativeToNow = TimeSpan.FromHours( 1 );
            return options;
        } )
    .Build();

var result = await command( factory.Create( logger ), "test" );

Assert.AreEqual( 4, result );

When a set of steps should cache the overload that takes a builder can be used. In this case the inital input will be used at the key and the final step of the nested pipeline will be cached

// Takes a string and returns a number
var command = PipelineFactory
    .Start<string>()
    .PipeCacheAsync( b => b
        .PipeAsync( CharacterCountAsync )
        .Pipe( (ctx, arg) => arg + 100 ))
    .Build();

var result = await command( factory.Create( logger ), "test" );

Assert.AreEqual( 104, result );

Distributed Cache

Distributed Caches work in the same way, but requires the registration of a IDistributedCache, an optional ICacheSerializer serializer (defaults to Json) and only supports Async methods

var command = PipelineFactory
    .Start<string>()
    .PipeDistributedCacheAsync( CharacterCountAsync )
    .Build();

Dependacy Injection

Because this uses the existing DI built into pipelines, caching can be configured with an existing cache.

Memory Cache

// Add Memory Cache
services.AddMemoryCache();

// Share with the pipelines
services.AddPipeline( includeAllServices: true );

Or defined seperately as part of the container use for the pipelines:

services.AddPipeline( (factoryServices, rootProvider) =>
{
    factoryServices.AddMemoryCache();
    factoryServices.AddPipelineDefaultCacheSettings( absoluteExpirationRelativeToNow: TimeSpan.FromHours( 1 ) )
} );

Distributed Cache

An Establish caching services such as SQL needs to be configured:

services.AddDistributedSqlServerCache(options =>
{
    options.ConnectionString = builder.Configuration.GetConnectionString(
        "DistCache_ConnectionString");
    options.SchemaName = "dbo";
    options.TableName = "TestCache";
});

An Optional ICacheSerializer can be registered. If one is missing the JsonCacheSerializer which uses System.Text.Json.JsonSerializer is used.

services.AddTransient<ICacheSerializer, JsonCacheSerializer>()