A React JS boilerplate that makes it easy and intuitive to implement Uncle Bob's Clean Architecture in react js. This boilerplate provides basic structure that are designed according to the Clean Architecture.
The concept is borrrowed to this wonderfull code.
https://github.com/ShadyBoukhary/flutter_clean_architecture
https://github.com/janithl/react-clean-arch
Thanks to @ShadyBoukhary and @janithl
git clone https://github.com/bailabs/react_js_clean_architecture.git
cd react_js_clean_architecture
yarn install // or npm install
yarn run dev // or npm yarn run dev
It is architecture based on the book and blog by Uncle Bob. It is a combination of concepts taken from the Onion Architecture and other architectures. The main focus of the architecture is separation of concerns and scalability. It consists of four main modules: App
, Domain
, Data
, and Device
.
Source code dependencies only point inwards. This means inward modules are neither aware of nor dependent on outer modules. However, outer modules are both aware of and dependent on inner modules. Outer modules represent the mechanisms by which the business rules and policies (inner modules) operate. The more you move inward, the more abstraction is present. The outer you move the more concrete implementations are present. Inner modules are not aware of any classes, functions, names, libraries, etc.. present in the outer modules. They simply represent rules and are completely independent from the implementations.
The Domain
module defines the business logic of the application. It is a module that is independent from the development platform i.e. it is written purely in the programming language and does not contain any elements from the platform. The reason for that is that Domain
should only be concerned with the business logic of the application, not with the implementation details. This also allows for easy migration between platforms, should any issues arise.
Domain
is made up of several things.
- Entities
<<<<<<< HEAD
- Enterprise-wide business rules
- Made up of classes that can contain methods
- Business objects of the application
- Used application-wide
- Least likely to change when something in the application changes
- Usecases
- Application-specific business rules
- Encapsulate all the usecases of the application
- Orchestrate the flow of data throughout the app
- Should not be affected by any UI changes whatsoever
- Might change if the functionality and flow of application change
- Repositories
- Abstract classes that define the expected functionality of outer layers
- Are not aware of outer layers, simply define expected functionality
- E.g. The
Login
usecase expects aRepository
that haslogin
functionality
- E.g. The
- Passed to
Usecases
from outer layers ======= - Enterprise-wide business rules
- Made up of classes that can contain methods
- Business objects of the application
- Used application-wide
- Least likely to change when something in the application changes
- Usecases
- Application-specific business rules
- Encapsulate all the usecases of the application
- Orchestrate the flow of data throughout the app
- Should not be affected by any UI changes whatsoever
- Might change if the functionality and flow of application change
- Repositories
- Abstract classes that define the expected functionality of outer layers
- Are not aware of outer layers, simply define expected functionality
- E.g. The
Login
usecase expects aRepository
that haslogin
functionality
- E.g. The
- Passed to
Usecases
from outer layers
ea40ce18a7ffdb0eabd67b40cd7d0fa2db7b5a7e
Domain
represents the inner-most layer. Therefore, it the most abstract layer in the architecture.
App
is the layer outside Domain
. App
crosses the boundaries of the layers to communicate with Domain
. However, the Dependency Rule is never violated. Using polymorphism
, App
communicates with Domain
using inherited class: classes that implement or extend the Repositories
present in the Domain
layer. Since polymorphism
is used, the Repositories
passed to Domain
still adhere to the Dependency Rule since as far as Domain
is concerned, they are abstract. The implementation is hidden behind the polymorphism
.
Since App
is the presentation layer of the application, it is the most framework-dependent layer, as it contains the UI and the event handlers of the UI. For every page in the application, App
defines at least 3 classes: a Action
, a Reducers
, and a View
.
-
ACTIONS <<<<<<< HEAD
-
In a nutshell, actions are events. Actions send data from the application (user interactions, internal events such as API calls, and form submissions) to the store. The store gets information only from actions. Internal actions are simple JavaScript objects that have a type property (usually constant), describing the type of action and payload of information being sent to the store.
{ type: LOGIN_FORM_SUBMIT, payload: {username: ‘alex’, password: ‘123456’} }
Actions are created with action creators. That sounds obvious, I know. They are just functions that return actions.
function authUser(form) { return { type: LOGIN_FORM_SUBMIT, payload: form } }
Calling actions anywhere in the app, then, is very easy. Use the dispatch method, like so:
dispatch(authUser(form));
-
-
REDUCERS
-
We’ve already discussed what a reducer is in functional JavaScript. It’s based on the array reduce method, where it accepts a callback (reducer) and lets you get a single value out of multiple values, sums of integers, or an accumulation of streams of values. In Redux, reducers are functions (pure) that take the current state of the application and an action and then return a new state. Understanding how reducers work is important because they perform most of the work. Here is a very simple reducer that takes the current state and an action as arguments and then returns the next state:
function handleAuth(state, action) { return _.assign({}, state, { auth: action.payload }); }
For more complex apps, using the combineReducers() utility provided by Redux is possible (indeed, recommended). It combines all of the reducers in the app into a single index reducer. Every reducer is responsible for its own part of the app’s state, and the state parameter is different for every reducer. The combineReducers() utility makes the file structure much easier to maintain.
If an object (state) changes only some values, Redux creates a new object, the values that didn’t change will refer to the old object and only new values will be created. That’s great for performance. To make it even more efficient you can add Immutable.js.
const rootReducer = combineReducers({ handleAuth: handleAuth, editProfile: editProfile, changePassword: changePassword });
-
-
Extra
Utility
classes (any commonly used functions like timestamp getters etc..)Constants
classes (const
strings for convenience)Navigator
(if needed)
Represents the data-layer of the application. The Data
module, which is a part of the outermost layer, is responsible for data retrieval. This can be in the form of API calls to a server, a local database, or even both.
- Repositories
- Every
Repository
should implementRepository
from the Domain layer. - Using
polymorphism
, these repositories from the data layer can be passed across the boundaries of layers, starting from theView
down to theUsecases
through theController
andPresenter
. - Retrieve data from databases or other methods.
- Responsible for any API calls and high-level data manipulation such as
- Registering a user with a database
- Uploading data
- Downloading data
- Handling local storage
- Calling an API
- Every
- Models (not a must depending on the application)
- Extensions of
Entities
with the addition of extra members that might be platform-dependent. For example, in the case of local databases, this can be manifested as anisDeleted
or anisDirty
entry in the local database. Such entries cannot be present in theEntities
as that would violate the Dependency Rule since Domain should not be aware of the implementation. - In the case of our application, models in the
Data
layer will not be necessary as we do not have a local database. Therefore, it is unlikely that we will need extra entries in theEntities
that are platform-dependent.
- Extensions of
- Mappers
- Map
Entity
objects toModels
and vice-versa. - Static classes with static methods that receive either an
Entity
or aModel
and return the other. - Only necessary in the presence of
Models
- Map
- Extra
Utility
classes if neededConstants
classes if needed
Part of the outermost layer, Device
communicates directly with the platform i.e. Android and iOS. Device
is responsible for Native functionality such as GPS
and other functionality present within the platform itself like the filesystem. Device
calls all Native APIs.
-
Devices
- Similar to
Repositories
inData
,Devices
are classes that communicate with a specific functionality in the platform. - Passed through the layers the same way
Repositories
are pass across the boundaries of the layer: using polymorphism between theApp
andDomain
layer. That means theController
passes it to thePresenter
then thePresenter
passes it polymorphically to theUsecase
, which receives it as an abstract class.
- Similar to
-
Extra
-
Utility
classes if needed -
Constants
classes if needed ======= -
In a nutshell, actions are events. Actions send data from the application (user interactions, internal events such as API calls, and form submissions) to the store. The store gets information only from actions. Internal actions are simple JavaScript objects that have a type property (usually constant), describing the type of action and payload of information being sent to the store.
{ type: LOGIN_FORM_SUBMIT, payload: {username: ‘alex’, password: ‘123456’} }
Actions are created with action creators. That sounds obvious, I know. They are just functions that return actions.
function authUser(form) { return { type: LOGIN_FORM_SUBMIT, payload: form } }
Calling actions anywhere in the app, then, is very easy. Use the dispatch method, like so:
dispatch(authUser(form));
-
-
REDUCERS
-
We’ve already discussed what a reducer is in functional JavaScript. It’s based on the array reduce method, where it accepts a callback (reducer) and lets you get a single value out of multiple values, sums of integers, or an accumulation of streams of values. In Redux, reducers are functions (pure) that take the current state of the application and an action and then return a new state. Understanding how reducers work is important because they perform most of the work. Here is a very simple reducer that takes the current state and an action as arguments and then returns the next state:
function handleAuth(state, action) { return _.assign({}, state, { auth: action.payload }); }
For more complex apps, using the combineReducers() utility provided by Redux is possible (indeed, recommended). It combines all of the reducers in the app into a single index reducer. Every reducer is responsible for its own part of the app’s state, and the state parameter is different for every reducer. The combineReducers() utility makes the file structure much easier to maintain.
If an object (state) changes only some values, Redux creates a new object, the values that didn’t change will refer to the old object and only new values will be created. That’s great for performance. To make it even more efficient you can add Immutable.js.
const rootReducer = combineReducers({ handleAuth: handleAuth, editProfile: editProfile, changePassword: changePassword });
-
-
Extra
Utility
classes (any commonly used functions like timestamp getters etc..)Constants
classes (const
strings for convenience)Navigator
(if needed)
Represents the data-layer of the application. The Data
module, which is a part of the outermost layer, is responsible for data retrieval. This can be in the form of API calls to a server, a local database, or even both.
- Repositories
- Every
Repository
should implementRepository
from the Domain layer. - Using
polymorphism
, these repositories from the data layer can be passed across the boundaries of layers, starting from theView
down to theUsecases
through theController
andPresenter
. - Retrieve data from databases or other methods.
- Responsible for any API calls and high-level data manipulation such as
- Registering a user with a database
- Uploading data
- Downloading data
- Handling local storage
- Calling an API
- Every
- Models (not a must depending on the application)
- Extensions of
Entities
with the addition of extra members that might be platform-dependent. For example, in the case of local databases, this can be manifested as anisDeleted
or anisDirty
entry in the local database. Such entries cannot be present in theEntities
as that would violate the Dependency Rule since Domain should not be aware of the implementation. - In the case of our application, models in the
Data
layer will not be necessary as we do not have a local database. Therefore, it is unlikely that we will need extra entries in theEntities
that are platform-dependent.
- Extensions of
- Mappers
- Map
Entity
objects toModels
and vice-versa. - Static classes with static methods that receive either an
Entity
or aModel
and return the other. - Only necessary in the presence of
Models
- Map
- Extra
Utility
classes if neededConstants
classes if needed
Part of the outermost layer, Device
communicates directly with the platform i.e. Android and iOS. Device
is responsible for Native functionality such as GPS
and other functionality present within the platform itself like the filesystem. Device
calls all Native APIs.
- Devices
- Similar to
Repositories
inData
,Devices
are classes that communicate with a specific functionality in the platform. - Passed through the layers the same way
Repositories
are pass across the boundaries of the layer: using polymorphism between theApp
andDomain
layer. That means theController
passes it to thePresenter
then thePresenter
passes it polymorphically to theUsecase
, which receives it as an abstract class.
- Similar to
- Extra
Utility
classes if neededConstants
classes if needed
ea40ce18a7ffdb0eabd67b40cd7d0fa2db7b5a7e