I have kept codegen/templating for the last. But this looks nice. So, where are the template files stored? and how does VS fetch it?

Just like that. I am still studying the library, but so far seems nice.
I tjust generates a .cs file, it doesnt depend on runtime and compile.
We can generate them as partial, so we can make changes to the generated code

NSwag has TypeScript client generation too. That wouldn't be a problem. I will check the feasibility and decide on the client. Just trying to complete the Web API first as soon as possible, for now.

I have a nextjs project, cant open source it, since it is a paid theme, but I integrate it with the nwsag studio in the nextjs project.

So this means Blazor is not included?

the current version will be still preserved as part of the v1 branch, and existing nuget packages.

You can always use Controllers if you want to. It's just that MinimalAPIs make more sense for mid scale projects, and has finer controls. Also Microsoft is actively pushing it and adding quite good number of features into it.

Thanks for your quick response @iammukeshm . I see fullstackhero as an enterprise grade boilerplate so I was thinking going with controllers as the default would have been the more intuitive option. So that in cases where people are undertaking simple projects, they can rather choose minimal apis.

One more question. Since Microsoft has introduced Blazor United, do you have any plans at all to upgrade the blazor-wasm project to .net 8 ? I am asking since the roadmap does not list it

"From bard.google.com"
https://g.co/bard/share/b11c1f1a930f

In summary:

Choose Controller-based APIs for larger, complex APIs requiring many features, ease of use, and automatic documentation.
Choose Minimal APIs for smaller, performance-critical APIs, microservices, or situations where you need finer control and flexibility.

hi,

1. Wolverine is cool, but there is too much magic to its implementation, and might not entertain everyone. Whereas MediatR is far more adapted and people know how to work with it. Also, the Pipeline Behavior just works right like I wanted. What are your thoughts about it?
2. I guess Minimal API is faster than fast-endpoints. Performance is on par with Minimal APIs - quoted from their landing page. I haven't explored fast-endpoint much though.

@iammukeshm , will it be possible to add FastEndpoint for those who wish by themselves ?

I plan to add FastEndpoint too, it's fantastic library

Hi, sorry for the confusion, by tba I meant to-be-announced. For now there is no template / nuget published with the new upgrade since this is a work in progress. You can clone the repo and switch to the v2 branch if you need to check something. I am pushing new changes on a daily basis.

And yes, the documentation website also will be revamped after the project reaches a stable stage. Planning to use nextra for documentation. (https://nextra.site/)

thanks for your answer Mukesh, no problem, sorry I'm not familiar with abreviation in English !

im the opposite.. I did the CA thing and it was fine.. But when I read up on VS that seemed more efficient for my simple apps so I am looking forward to using that

@PedroVentura235 The idea behind the new rewrite is to support both Clean Architecture and VSA under modules. It will be your choice to decide whether to go with CA or VSA with your modules. This way you get the best of both worlds, and I feel that this aligns with Microservice concepts, and is an efficient way to build applications. You really won't need CA in every use case/module.

Once the NuGet packages are published, you will get an option to generate CA / VSA templates within your FSH solutions. This is what I have in mind for now.

@iammukeshm Do you have any roadmap or something, i will have some time and could assist on this, and could advance on some things, What you suggest?

@PedroVentura235,Yes I will need help. I will come up with a list of issues/features that have to be implemented in this new rewrite. You can pick up something once it's available. Sounds good?

Yes, thought about this same issue. Plan is to maintain only this repository, which has a .NET 8 WebAPI (Modular Monolith), and a couple of frontends within the same repository. I will start with NextJs, and add blazor later on. I will also include a detailed guide on how you can migrate this modular monolith application to a microservice architecture. Should be do-able.

what do you think about fluentui for blazor?

Personal opinion: I feel MudBlazor is great for Blazor projects. However, after the release of .NET 8, it seems that a lot of work needs to be done to make MudBlazor compatible with SSR, CSR...

Perfect! we are in the same page! I think that if we combine the best code of the both repositorios (microservice/webapi net 8) we can get a better an unique témplate... May be when the v2 roadmap was finished the nectario v2.1 roadmap could be add the missing microservice features like yarp gateway in a isolated project, and masstransit for services bus(rabbitmq) in the infraestructure common layer to be available for use/or not in the webapi/microservice.

How would you manage user creation/tenant assignment in this case? How would token generation work?

The users creation can be initialized by keycloack API/container and use the attributes to Store the tenantId.. something like that https://danielwertheim.se/keycloak-and-multi-tenancy-using-single-realm/...

... Happy new year! 😃

The problem is if you push the tenantId to keycloak, users wont be able to switch to an another OpenID provider like Auth0. I am thinking about this flow.

1. User has to be created in KeyCloak Realm/Auth0.
2. The same user has to registered into the application database, but without any password. This is just to attach the API specific data like tenantId, etc to the user email. User Email should match in both the application database as well as the Keycloak Db.
3. User Token is generated by KeyCloak API, using the Code Flow PKCE.
4. Once token is passed to the API, we have to figure out which tenant the user belongs to. and also have a mechanism to derive the app specific permissions.

What do you think? I don't find any good references to implement this in a very practical way.

I think that the hybrid approach is better to save the tenantId from the app side and security from the IdentityProvider (OAuth o Keycloack) ... And.. to avoid coupled to keyclock or another provider you can use this design: https://medium.com/@ahmed.gaduo_93938/how-to-implement-keycloak-authentication-in-a-net-core-application-ce8603698f24

Yep, I didn't get much time to concentrate on the upgrade for the past few weeks. I think I will first come up with a list of TODOs to be taken care of for this project. From there, it will be clearer on the status. I am looking to complete this upgrade soon. Sorry for the delay!

🙄

That's right, IS4 is enough to use

i see the server side is updated.. the client side is next?

I don't know.. I think that re repo is deprecates..., @iammukeshm create one commit for month.. 😭😭

Guys, really sorry. No, this project will not be dead anytime soon ;) I am just not able to find enough time for this. I have resumed the work from yesterday and have done over 20+ commits. I hope at least the server side will be completed in a week. The major thing to add is,

• Multitenancy (code changes are done, integrations have to be tested)
• cicd
• terraform
• searching
• interceptors
• signalr
• hangfire

On behalf of those who are waiting for this project, I would like to thank you for your time. I hope you will finish this week. You will have prepared an excellent infrastructure for beginners like me who have been waiting for this project for a long time. I wish you convenience.

Excelent News! @iammukeshm !
When you complete a beta versión, you can request help to the community to fix bugs or add new features vía pull request in orden to help you to create a best framework app! And you can continué with the Client website app... I think that of a good idea to finish or keep updated this project continously... the net9 is around the corner.. 😂😂

Hi mukesh, regarding the source code generation.
I am writing a basic template for myself, and added a source code generation, not on runtime, but when the user wants(how I prefer)
https://github.com/PedroVentura235/BasicWebApi/tree/master/Application/Features.
On the file FeatureTemplate.csx , we just right click it and run it(through the extension CodeGenCS, and it generats all the files we want.
In my example I have all the create, gets, updates deletes and dtos all auto generated, and is only generated for the ones I add on the file

job dispatching is already available for Hangfire. Currently when a new user is registered, a hangfire job is dispatched to send the registration email. Are you referring to this one?

to support IRequest
`
using MediatR;
using System.Linq.Expressions;

public interface IJobService
{
void AddOrUpdate(string recurringJobId, IBaseRequest request, string cronExpression, TimeZoneInfo? timeZone = null, string queue = "default");

bool Delete(string jobId);

bool Delete(string jobId, string fromState);
string Enqueue(IBaseRequest request);
string Enqueue(Expression<Action> methodCall);
string Enqueue(Expression<Func<Task>> methodCall);

string Enqueue<T>(Expression<Action<T>> methodCall);

string Enqueue<T>(Expression<Func<T, Task>> methodCall);

bool IsRunning(string recurringJobId);

void RemoveIfExists(string recurringJobId);

bool Requeue(string jobId);

bool Requeue(string jobId, string fromState);

string Schedule(IBaseRequest request, TimeSpan delay);

string Schedule(Expression<Action> methodCall, TimeSpan delay);

string Schedule(Expression<Func<Task>> methodCall, TimeSpan delay);

string Schedule(Expression<Action> methodCall, DateTimeOffset enqueueAt);

string Schedule(Expression<Func<Task>> methodCall, DateTimeOffset enqueueAt);

string Schedule<T>(Expression<Action<T>> methodCall, TimeSpan delay);

string Schedule<T>(Expression<Func<T, Task>> methodCall, TimeSpan delay);

string Schedule<T>(Expression<Action<T>> methodCall, DateTimeOffset enqueueAt);

string Schedule<T>(Expression<Func<T, Task>> methodCall, DateTimeOffset enqueueAt);

}
`

`
internal class HangfireMediatorBridge(ISender mediator, ICurrentUserInitializer currentUserInitializer, IMultiTenantStore tenantStore, IMultiTenantContextSetter tenantContextAccessor)
{
///

    await mediator.Send(request, ct);
}

/// <summary>
///     It takes a request, and a cancellation token, and returns a task.
/// </summary>
/// <param name="request">The request object that will be sent to the mediator.</param>
/// <param name="ct">This is a token that can be used to cancel the job.</param>
/// <returns>
///     A Task.
/// </returns>
[DisplayName("{0}")]
public Task SendAsync(IBaseRequest request, CancellationToken ct)
{
    return mediator.Send(request, ct);
}

}
`

Here is the link to the PR where we have started working on this integration: #956 . Feel free to review and provide your feedback.

@iammukeshm What are we missing to launch v2?

What about file storage service, unit test, mapster, permission management from the original net 8 migration scope... #907 (comment)

and also, I think that an event systems like masstransit is a "must feature" to include...

I can help to the another features from the inicial scope, like add grafana/prometheous metrics that I already done in v1...

Are you agree?

I would not keep masstransit / graffana / prometheus as priority for now. Once Blazor Client is done, with these essentials,

1. Login / Logout
2. Dark Mode
3. Responsive Design
4. CRUD Operations Screen
5. User Management Screen
6. Tenant Management
7. Forgot Password / Reset

Once the above is done, we can have a beta release, and after that we can concentrate on the other stuff. That's the plan i have for now.

Any idea on when this Beta will be available so we can start new projects with this? Or can we start now? Is the API is good enough shape to begin building out an API?

weekend.

This weekend? 😃

Blazor / Products Grid is done, Supporting all CRUD operations. However, there are still some bits that need fixing. Someone please try to run it on local and let me know.

it looks good to me