Fraktalio.FModel
0.1.0
dotnet add package Fraktalio.FModel --version 0.1.0
NuGet\Install-Package Fraktalio.FModel -Version 0.1.0
<PackageReference Include="Fraktalio.FModel" Version="0.1.0" />
paket add Fraktalio.FModel --version 0.1.0
#r "nuget: Fraktalio.FModel, 0.1.0"
// Install Fraktalio.FModel as a Cake Addin #addin nuget:?package=Fraktalio.FModel&version=0.1.0 // Install Fraktalio.FModel as a Cake Tool #tool nuget:?package=Fraktalio.FModel&version=0.1.0
f(
model-c#)
- Domain Modeling
The project is in experimental phase. Pre-release packages are available at nuget.org.
A stable release is one that's considered reliable enough to be used in production.
When you’re developing an information system to automate the activities of the business, you are modeling the business. The abstractions that you design, the behaviors that you implement, and the UI interactions that you build all reflect the business — together, they constitute the model of the domain.
IOR<Library, Inspiration>
This project can be used as a multiplatform library, or as an inspiration, or both. It provides just enough tactical Domain-Driven Design patterns, optimised for Event Sourcing and CQRS.
Abstraction and generalization
Abstractions can hide irrelevant details and use names to reference objects. It emphasizes what an object is or does rather than how it is represented or how it works.
Generalization reduces complexity by replacing multiple entities which perform similar functions with a single construct.
Abstraction and generalization are often used together. Abstracts are generalized through parameterization to provide more excellent utility.
Func<C, S, IEnumerable<E>> decide
On a higher level of abstraction, any information system is responsible for handling the intent (Command
) and based on
the current State
, produce new facts (Events
):
- given the current
State/S
on the input, - when
Command/C
is handled on the input, - expect list of new
Events/E
to be published/emitted on the output
Func<S, E, S> evolve
The new state is always evolved out of the current state S
and the current event E
:
- given the current
State/S
on the input, - when
Event/E
is handled on the input, - expect new
State/S
to be published on the output
Event-sourced or State-stored systems
- State-stored systems are traditional systems that are only storing the current State by overwriting the previous State in the storage.
- Event-sourced systems are storing the events in immutable storage by only appending.
Both types of systems can be designed by using only these two functions and three generic parameters:
Func<C, S, IEnumerable<E>> decide
Func<S, E, S> evolve
There is more to it! You can switch from one system type to another or have both flavors included within your systems landscape.
Two functions are wrapped in a datatype class (algebraic data structure), which is generalized with three generic parameters:
public class Decider<C, S, E>(Func<C, S, IEnumerable<E>> decide, Func<S, E, S> evolve)
{
public Func<C, S, IEnumerable<E>> Decide { get; } = decide;
public Func<S, E, S> Evolve { get; } = evolve;
}
Decider
is the most important datatype, but it is not the only one. There are others:
Decider
Decider
is a datatype that represents the main decision-making algorithm. It belongs to the Domain layer. It has three
generic parameters C
, S
, E
, representing the type of the values that Decider
may contain or use.
Decider
can be specialized for any type C
or S
or E
because these types do not affect its
behavior. Decider
behaves the same for C
=Int
or C
=YourCustomType
, for example.
Decider
is a pure domain component.
C
- CommandS
- StateE
- Event
public class Decider<C, S, E>(Func<C, S, IEnumerable<E>> decide, Func<S, E, S> evolve, S initialState)
: IDecider<C, S, E>
{
public Func<C, S, IEnumerable<E>> Decide { get; } = decide;
public Func<S, E, S> Evolve { get; } = evolve;
public S InitialState { get; } = initialState;
}
Additionally, initialState
of the Decider is introduced to gain more control over the initial state of the Decider.
Notice that Decider
implements an interface IDecider
to communicate the contract.
Decider functions
Contravariant
Decider<Cn, S, E> MapLeftOnCommand<Cn>(Func<Cn, C> f)
Profunctor (Contravariant and Covariant)
Decider<C, S, En> DimapOnEvent<En>(Func<En, E> fl, Func<E, En> fr)
Decider<C, Sn, E> DimapOnState<Sn>(Func<Sn, S> fl, Func<S, Sn> fr)
Commutative Monoid
public static Decider<C_SUPER?, Tuple<S, S2>, E_SUPER?> Combine<C, S, E, C2, S2, E2, C_SUPER, E_SUPER>(
this Decider<C?, S, E?> x, Decider<C2?, S2, E2?> y)
where C : class, C_SUPER
where C2 : class, C_SUPER
where E : class, E_SUPER
where E2 : class, E_SUPER
A monoid is a type together with a binary operation (
combine
) over that type, satisfying associativity and having an identity/empty element. Associativity facilitates parallelization by giving us the freedom to break problems into chunks that can be computed in parallel.
combine
operation is also commutative. This means that the order in which deciders are combined does not affect the result.
View
View
is a datatype that represents the event handling algorithm, responsible for translating the events into
denormalized state, which is more adequate for querying. It belongs to the Domain layer. It is usually used to create
the view/query side of the CQRS pattern. Obviously, the command side of the CQRS is usually event-sourced aggregate/decider.
It has two generic parameters S
, E
, representing the type of the values that View
may contain or use.
View
can be specialized for any type of S
, E
because these types do not affect its behavior.
View
behaves the same for E
=Int
or E
=YourCustomType
, for example.
View
is a pure domain component.
S
- StateE
- Event
public class View<S, E>(Func<S, E, S> evolve, S initialState) : IView<S, E>
{
public Func<S, E, S> Evolve { get; } = evolve;
public S InitialState { get; } = initialState;
}
Notice that View
implements an interface IView
to communicate the contract.
View functions
Contravariant
View<S, En> MapLeftOnEvent<En>(Func<En, E> f) => new InternalView<S, S, E>(Evolve, InitialState).MapLeftOnEvent(f).AsView();
Profunctor (Contravariant and Covariant)
View<Sn, E> DimapOnState<Sn>(Func<Sn, S> fl, Func<S, Sn> fr) => new InternalView<S, S, E>(Evolve, InitialState).DimapOnState(fl, fr).AsView();
Commutative Monoid
public static View<Tuple<S, S2>, E_SUPER> Combine<S, E, S2, E2, E_SUPER>(this View<S, E?> x, View<S2, E2?> y)
where E : class, E_SUPER
where E2 : class, E_SUPER
A monoid is a type together with a binary operation (combine) over that type, satisfying associativity and having an identity/empty element. Associativity facilitates parallelization by giving us the freedom to break problems into chunks that can be computed in parallel.
combine
operation is also commutative. This means that the order in which views are combined does not affect the result.
Saga
Saga
is a datatype that represents the central point of control, deciding what to execute next (A
). It is
responsible for mapping different events from many aggregates into action results AR
that the Saga
then can use to
calculate the next actions A
to be mapped to commands of other aggregates.
Saga
is stateless, it does not maintain the state.
It has two generic parameters AR
, A
, representing the type of the values that Saga
may contain or use.
Saga
can be specialized for any type of AR
, A
because these types do not affect its behavior.
Saga
behaves the same for AR
=Int
or AR
=YourCustomType
, for example.
Saga
is a pure domain component.
AR
- Action ResultA
- Action
public class Saga<AR, A>(Func<AR, IEnumerable<A>> react) : ISaga<AR, A>
{
public IEnumerable<A> React(AR actionResult) => react(actionResult);
}
Notice that Saga
implements an interface ISaga
to communicate the contract.
Saga functions
Contravariant
Saga<ARn, A> MapLeftOnActionResult<ARn>(Func<ARn, AR> f) => new(arn => react(f(arn)));
Covariant
Saga<AR, An> MapOnAction<An>(Func<A, An> f) => new(ar => react(ar).Select(f));
Monoid
public static Saga<AR_SUPER, A_SUPER?> Combine<AR, A, AR2, A2, AR_SUPER, A_SUPER>(this Saga<AR?, A> sagaX,
Saga<AR2?, A2> sagaY)
where AR : AR_SUPER
where A : A_SUPER
where AR2 : AR_SUPER
where A2 : A_SUPER
A monoid is a type together with a binary operation (combine) over that type, satisfying associativity and having an identity/empty element. Associativity facilitates parallelization by giving us the freedom to break problems into chunks that can be computed in parallel.
combine
operation is also commutative. This means that the order in which sagas are combined does not affect the result.
Getting Started with fmodel-c#
The fmodel-c# library is available on nuget.org.
Note: fmodel-c# is in Pre-Release!
The fmodel-c#
library is currently in pre-release. At this stage, it focuses exclusively on pure domain components that effectively model behavior.
The application layer
will be included in future updates. This will provide guidance on composing and running the domain model in various scenarios, including:
- Event-sourced scenarios
- State-stored scenarios
FModel in other languages
References and further reading
Credits
Special credits to Jérémie Chassaing
for sharing his research
and Adam Dymitruk
for hosting the meetup.
Thank you, Srdjan Zivojinovic and Crafters Cloud, for your significant contributions to this project.
Created with ❤️ by Fraktalio
Product | Versions Compatible and additional computed target framework versions. |
---|---|
.NET | net8.0 is compatible. net8.0-android was computed. net8.0-browser was computed. net8.0-ios was computed. net8.0-maccatalyst was computed. net8.0-macos was computed. net8.0-tvos was computed. net8.0-windows was computed. |
-
net8.0
- Fraktalio.FModel.Contracts (>= 1.0.0)
NuGet packages
This package is not used by any NuGet packages.
GitHub repositories
This package is not used by any popular GitHub repositories.
Version | Downloads | Last updated |
---|---|---|
0.1.0 | 116 | 8/23/2024 |