POC for .NET 8 and Blazor and not ready for production !!! This is a "private" project and some parts/ideas of the code may be ported to productions in other projects.

• .NET 8
• MSSQL, Postgres, MongoDB
• Blazor WASM
• Blazor Server
• EF
  • Linq
  • Audit
• OAuth2
• MediatR
  • IPipelineBehavior
• Autofac
  • mainly for registration open generic type for MediatR -> IRequest,
• Logging - Serilog
• Localizations

• fully responsive
• SCSS
• Telerik
• Localization
• Logging
  • via MediatR - in memory
  • TODO - send log information to Server

containerBuilder.RegisterGeneric(typeof(HttpAuthorizedHandler<>)).AsImplementedInterfaces();
containerBuilder.RegisterGeneric(typeof(HttpNonAuthorizedHandler<>)).AsImplementedInterfaces();

• Blazor Hybrid - UI works
• more in branch MAUI

• xUnit
• Moq
• BUnit
• Selenium for E2E testing

• Better code comments
• Better documentation
• UML diagrams
• tests, only JMCode has a code coverage of xUnit of around 70%.
• Stress tests

General

• JMCore - general library for server and client development. It contains a common code base library for different .NET project type, like Blazor, MVC, MAUI etc.
• JMCore.Client - general library for client development
• JMCore.Server - general library for server development
• JMCore.Tests - xUnit tests for JMCore, JMCore.Client, JMCore.Server
• JMCore.TestsIntegrations - integration tests for JMCore.Server - MSSQL database

Blazor

• JMCore.Blazor - contains a basic Blazor layouts and components using Telerik components. It is built on JMCore and it can be used for WASM and server.
• JMCore.Blazor.BUnit - BUnit tests for Blazor components

Implementation

• JMCoreTest.Blazor.Server - Server path for web page, includes API
• JMCoreTest.Blazor.Client - Client WASM for web page implements JMCore.Blazor
• JMCoreTest.Blazor.Shared - Communications models between Server and Client
• JMCoreTest.Blazor.EE - End to end tests base on Selenium

Cache uses Category, Key and Value to working with cache.

Cache is managed by MediatR request with using CacheBehavior for logging information.

MediatR

  graph LR;
      CacheGetQuery("CacheGetQuery(JMCacheKey)")--ICacheRequest-->CacheGetHandler-->JMCacheValue
      CacheSaveCommand("CacheSaveCommand(JMCacheKey key, object value)")-->CacheSaveHandler-->bool1(bool)
      CacheRemoveCommand("CacheRemoveCommand(JMCacheKey)") --> CacheRemoveHandler --> bool
      CacheRemoveCommand1("CacheRemoveCommand(int category, string? keyPrefix = null)")-->CacheRemoveHandler;

UML Class

classDiagram
  note for JMCacheKey "JMCacheKey.Create(JMCacheServerCategory.DbTable, key)"  
 
  class JMCacheKey{
     +JMCacheKey Create(int category, string key)$
  }
  
  class JMCacheCategory{
      LogCache
  }
  
  class JMBlazorCacheCategory{
    int LogCache = 100;
  }

  class JMCacheServerCategory{
    int DbTable = 200;
  }
  JMBlazorCacheCategory --|> JMCacheCategory 
  JMCacheServerCategory --|> JMCacheCategory

TODO

• By using autofac open generic registration, use categories as strong type, not as int.
• The MediatR handler can separate different requests by category into different repositories such as memory, Redis, etc.
• Implement Prometheus/Grafana to get info, how the cache is used.

Implements IStringLocalizer and all translations are retrieved from IStringLocalizationStorage. Translations can be loaded e.g. from database, browser local storage etc.

TODO

• For larger applications, all the translation text is stored in memory and this is a bad solution. The application should only load the translations it needs.
• improve the implementation of IStringLocalizerFactory

Proposal for a solution

sequenceDiagram
  box Client
    participant UI
    participant Memory
    participant Local Storage
    participant Http
  end
  box Server
    participant API
    participant Cache
    participant ResX
  end
  box Storage-DB
    participant Table
  end

  UI->>Memory: Exist in LocalizationMemoryStorage
  Memory->>Local Storage: Get UI translation
  Local Storage->>Http:Save to memory
  Http-->>API: Http request
  API->>Cache: using cache if value exists
  Cache-->>Table: if it doesn't exist, then it calls db
  Table-->>Cache: save to cache
  Cache->>API: returns data
  API-->>Http: http response
  Http->>Local Storage: save to storage
  Local Storage->>Memory: save to memory
  Memory->>UI: use translation in UI 
  ResX-->>Table: Default values - update existing translations

The entities to be audited can be set using the AuditAttribute attribute or can be set in AuditDbConfiguration.

[Auditable]
[Table("TestAttributeAudit")]
public class TestAttributeAuditEntity
{
    [Key] 
    public int Id { get; set; }
    
    public string Name { get; set; } = null!;

    [NotAuditable] 
    public string NotAuditableColumn { get; set; } = null!;
    
    public DateTime Created { get; set; }
}

or

 var auditConfiguration = new AuditDbConfiguration
 {
    AuditEntities = new List<string> { nameof(TestManualAuditEntity) },
    NotAuditProperty = new Dictionary<string, IEnumerable<string>>
    {
       { nameof(TestManualAuditEntity), new List<string>() { nameof(TestManualAuditEntity.NotAuditableColumn) } }  
    }
 };

DB structure

Currently only MSSQL is supported, but the solution is ready for other databases as well. The solution uses Entity Framework as the main ORM.