Scribd
Upload
2 views

DI

Uploaded by

Đặng Tuấn Linh
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
2 views

DI

Uploaded by

Đặng Tuấn Linh
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

C# With Dependency

Injection
Author: ToanND
Introduction

 Dependency Injection (DI) is an object-oriented programming design


pattern that allows us to develop loosely coupled code. DI helps in
getting rid of tightly coupled software components. The purpose of DI is
to make co
 DI is a technique to create a dependency or dependencies outside the
class that uses it. The dependencies are injected from the code that
calls the class and any information about their creation are kept away
from the inside of the class. This is also why DI is an implementation of
the Inversion of control (IoC) principle.de maintainable and easy to
update.
 To further explain the dependency injection design pattern, the inversion
of control principle has to be broken down.
Inversion of Control Principle

 Normally, the flow of the program logic is determined by objects that are
bound to one another. With the inversion of control, the flow depends on
the defined abstractions to be implemented that is built up during
program execution. This principle can be achieved by techniques like
dependency injection. In IoC, the code could also be linked statically
during compilation to the defined interface functions, but finding the
implementation of the function to execute by reading its description
from external configuration instead of with a direct reference in the code
itself.
How it Works

 In dependency injection, a dependent object or service is provided with


the object it needs at run time. The provided object will satisfy the
dependency during program execution but would not be known at
compile time. Rather than directly instantiating dependencies, or using
static references, the objects a class needs in order to perform its
actions are provided to the class in some abstracted form.
An example of the traditional way;
 Looking at the services used in this
register action of the account
controller, we can observe that
changing the service would imply
changing a lot of the codebase,
especially if the service has been used
in multiple parts of the project.
 For instance, if the email service is
replaced with a new one as shown
below, we can see that the tightly
coupled email service would need to
be changed everywhere is it used in
the project.
 From the examples shown, we can observe that
there is a generic function that all email services
will provide which it to send email which can help
us provide an abstraction of the email service
which is going to be used even when we do not
know which exact service (google or outlook) is to
be used.
 The corresponding implementation can then inherit from the interface.
 The above abstraction (interface) can be used in our UserLogic such
that;
 The email service is still tightly coupled so we have to inject it in using
one of the multiple types of dependency injection;
1. Constructor Injection
 The part of the code calling the UserLogic class would have to specify
what email service going to be used and pass it in the constructor when
instantiating it.
2. Setter Injection
 The part of the code calling the UserLogic class would have to specify
what email service going to be used and as one of its property.
3. Method Injection
 The part of the code calling the UserLogic Register method would have
to specify what email service as one of the parameters.
Conclusion

 Dependency injection helps achieve improved class coupling, better


code reusability, code maintainability and application unit testing

You might also like