So since I've been using Spring, if I were to write a service that had dependencies I would do the following:
#Component
public class SomeService {
#Autowired private SomeOtherService someOtherService;
}
I have now run across code that uses another convention to achieve the same goal
#Component
public class SomeService {
private final SomeOtherService someOtherService;
#Autowired
public SomeService(SomeOtherService someOtherService){
this.someOtherService = someOtherService;
}
}
Both of these methods will work, I understand that. But is there some advantage to using option B? To me, it creates more code in the class and unit test. (Having to write constructor and not being able to use #InjectMocks)
Is there something I'm missing? Is there anything else the autowired constructor does besides add code to the unit tests? Is this a more preferred way to do dependency injection?
Yes, option B (which is called constructor injection) is actually recommended over field injection, and has several advantages:
the dependencies are clearly identified. There is no way to forget one when testing, or instantiating the object in any other circumstance (like creating the bean instance explicitly in a config class)
the dependencies can be final, which helps with robustness and thread-safety
you don't need reflection to set the dependencies. InjectMocks is still usable, but not necessary. You can just create mocks by yourself and inject them by simply calling the constructor
See this blog post for a more detailed article, by one of the Spring contributors, Olivier Gierke.
I will explain you in simple words:
In Option(A), you are allowing anyone (in different class outside/inside the Spring container) to create an instance using default constructor (like new SomeService()), which is NOT good as you need SomeOtherService object (as a dependency) for your SomeService.
Is there anything else the autowired constructor does besides add code
to the unit tests? Is this a more preferred way to do dependency
injection?
Option(B) is preferred approach as it does NOT allow to create SomeService object without actually resolving the SomeOtherService dependency.
Please note, that since Spring 4.3 you don't even need an #Autowired on your constructor, so you can write your code in Java style rather than tying to Spring's annotations.
Your snippet would look like that:
#Component
public class SomeService {
private final SomeOtherService someOtherService;
public SomeService(SomeOtherService someOtherService){
this.someOtherService = someOtherService;
}
}
Good to know
If there is only one constructor call, there is no need to include an #Autowired annotation. Then you can use something like this:
#RestController
public class NiceController {
private final DataRepository repository;
public NiceController(ChapterRepository repository) {
this.repository = repository;
}
}
... example of Spring Data Repository injection.
Actually, In my experience, The second option is better. Without the need for #Autowired. In fact, it is wiser to create code that is not too tightly coupled with the framework (as good as Spring is). You want code that tries as much as possible to adopt a deferred decision-making approach. That is as much pojo as possible, so much such that the framework can be swapped out easily.
So I would advise you create a separate Config file and define your bean there, like this:
In SomeService.java file:
public class SomeService {
private final SomeOtherService someOtherService;
public SomeService(SomeOtherService someOtherService){
this.someOtherService = someOtherService;
}
}
In ServiceConfig.java file:
#Config
public class ServiceConfig {
#Bean
public SomeService someService(SomeOtherService someOtherService){
return new SomeService(someOtherService);
}
}
In fact, if you want to get deeply technical about it, there are thread safety questions (among other things) that arise with the use of Field Injection (#Autowired), depending on the size of the project obviously. Check this out to learn more on the advantages and disadvantages of Autowiring. Actually, the pivotal guys actually recommend that you use Constructor injection instead of Field Injection
I hope I won't be downgraded for expressing my opinion, but for me option A better reflects the power of Spring dependency injection, while in the option B you are coupling your class with your dependency, in fact you cannot instantiate an object without passing its dependencies from the constructor. Dependency Injection have been invented for avoid that by implementing Inversion of Control,so for me option B doesn't have any sense.
Autowired constructors provides a hook to add custom code before registering it in the spring container. Suppose SomeService class extends another class named SuperSomeService and it has some constructor which takes a name as its argument. In this case, Autowired constructor works fine. Also, if you have some other members to be initialized, you can do it in the constructor before returning the instance to spring container.
public class SuperSomeService {
private String name;
public SuperSomeService(String name) {
this.name = name;
}
}
#Component
public class SomeService extends SuperSomeService {
private final SomeOtherService someOtherService;
private Map<String, String> props = null;
#Autowired
public SomeService(SomeOtherService someOtherService){
SuperSomeService("SomeService")
this.someOtherService = someOtherService;
props = loadMap();
}
}
I prefer construction injection, just because I can mark my dependency as final which is not possible while injecting properties using property injection.
your dependencies should be final i.e not modified by program.
There are few cases when #Autowired is preferable.
One of them is circular dependency. Imagine the following scenario:
#Service
public class EmployeeService {
private final DepartmentService departmentService;
public EmployeeService(DepartmentService departmentService) {
this.departmentService = departmentService;
}
}
and
#Service
public class DepartmentService {
private final EmployeeService employeeService;
public DepartmentService(EmployeeService employeeService) {
this.employeeService = employeeService;
}
}
Then Spring Bean Factory will throw circular dependency exception. This won't happen if you use #Autowired annotation in both beans. And this is understandable: the constructor injection happens at very early stage of Spring Bean initialization, in createBeanInstance method of Bean Factory, while #Autowired-based injection happens way later, on post processing stage and is done by AutowiredAnnotationBeanPostProcessor.
Circular dependency is quite common in complex Spring Context application, and it needs not to be just two beans referring one another, it could a complex chain of several beans.
Another use case, where #Autowired is very helpful, is self-injection.
#Service
public class EmployeeService {
#Autowired
private EmployeeService self;
}
This might be needed to invoke an advised method from within the same bean. Self-injection is also discussed here and here.
There is a way to inject the dependencies through constructor using #RequeiredArgsContructor annotation from Lombok
#RequiredArgsConstructor
#Service
class A {
private final B b // needs to be declared final to be injected
}
In this way you don't need to specify a constructor
In an Spring mockmvc test I want to replace a bean by a mock implementation which is configured using Mockito.when() definitions. The definitions are indeed respected at the time the mock is configured, as well as at the time the mock is injected into a depending bean (a controller advice in my case) during application context startup. However, when the mock is used during a certain test, all when definitions are gone.
Why?
Some remarks:
The mock is completely new code, so it is impossible that I am not aware of any call to Mockito.reset().
the mock at the time of usage is the same as at the time of creation.
a bypassing solution to the problem is to configure the mock in a #BeforeEach method in AbstractTest. However, I want to understand why it does not work without.
Here a simplified and anonymized example
#Component
public class MyBean {
private String property;
...
public String getProperty() {
return property;
}
}
#ControllerAdvice
public class MyControllerAdvice() {
private MyBean myBean;
#Autowired
public MyControllerAdvice(MyBean myBean) {
this.myBean = myBean;
System.out.println(this.myBean.getProperty()); // --> outputs "FOOBAR"
}
#ModelAttribute
public String getMyBeanProperty() {
return myBean.getProperty(); // --> returns null
}
}
public class AbstractTest {
#Configuration
static class Config {
#Bean
public MyBean () {
MyBean myBean = Mockito.mock(MyBean.class, "I am a mock of MyBean");
when(myBean.getProperty()).thenReturn("FOOBAR");
}
}
}
That's not a problem of Mockito. I think you simplified the example a lot and we don't see the full picture, but I can say that main cause - 2 different beans MyBean: one is initialized with Spring's #Component, second is in configuration class with #Bean.
Why do you use #Component for POJO/DO?
#Bean in the configuration class is being initialized lazy so better way to use #PostConstruct
If you want to leave both beans mark MyBean in the configuration class as #Primary
I have recently learned concept of autowiring in spring. When I was trying to understand in which particular scenarios spring autowiring can be useful
I came up with the below two reasons from one of the questions asked in our stakoverflow forum.
1.I wanted to read values from a property file and inject them into a bean. Only way I could figure out how to do this at start up of my app was to
wire the bean in XML (and inject the properties.) I ended up using the "byName" attribute (because the bean was also marked as #Component) and then
used #Autowired #Qualifier("nameIChose") when injecting the bean into another class. It's the only bean I've written that I wire with XML.
2.I've found autowiring useful in cases where I've had a factory bean making another bean (whose implementation class name was described in a system
property,so I couldn't define the all wiring in XML). I usually prefer to make my wiring explicit though;
Can any body please give me some code snippet example of the above situations that would make my understanding of autowiring more clearer?
Here is an example of injecting properties into a bean.
Using field injection:
#Component
public class YourBean {
#Value("${your.property.name}")
private String yourProperty;
}
Using constructor injection:
#Component
public class YourBean2 {
private String yourProperty;
#Autowired
public YourBeans2(#Value("${your.property.name}") String yourProperty) {
this.yourProperty = yourProperty;
}
}
The following is a super simple example of autowiring various beans
#Component
public class Foo {
public void doSomething() {
}
}
#Component
public class Bar {
private Foo foo;
#Autowired
public Bar(Foo foo) {
this.foo = foo;
}
public void doSomethingElse() {
foo.doSomething();
}
}
In the previous example, no XML configuration of Foo and Bar needs to be done, Spring automatically picks up the beans because of their #Component annotation (assuming of course that component scanning has been enabled)
I have a project where I have an interface, an Abstract class implementing the same interface and then a set of concrete classes which implement this interface and extend the Abstract Class.
public interface Invoice
{
void process();
}
#component
public abstract class AbstractInvoice(){
#Resource
protected Writer writer;
protected validateInvoice(){
//some implementation
}
}
#Component
public Class TypeAInvoice() extends AbstractInvoice implements Invoice{
#Override
public void process(){
//... some code
writer.write();
}
}
public Interface Writer(){
public void write();
}
#Component
public class CDWriter implements Writer{
#Override
public void write() { /* implementation.....*/}
}
Spring file has a component scan for the package.
<context:annotation-config>
<context:component-scan base-package="com.xyz" />
I am using a factory to get an instance of TypeAInvoice invoice
Now calling invoice.process() gets a NPE when getting to write.write()
I am not sure what am I missing here. I tried to see the component scan and scope and could not find anything conceptually wrong.
I am using a factory to get an instance of TypeAInvoice invoice
Depending on what your Factory does, this may be the problem. If the Factory creates a new TypeAInvoice, Spring wiring doesn't apply. You have to query the Spring context for the Bean. One way (though not very pretty) is to use ContextLoader:
return ContextLoader.getCurrentWebApplicationContext().getBean(TypeAInvoice.class)
I'd say static Factories and Spring don't go to well together. Spring stands for the Inversion of Control pattern, while Factories stand for the Service Locator pattern. I'd suggest that you get rid of your factories and autowire your Spring Beans.
Everything is good, except for the fact you use a factory to get the TypeAInvoice. If you create it like TypeAInvoice typer = new TypeAInvoice() then spring knows nothing of it, the Writer is not autowired, there for you get the NullPointerException. You should get the bean from the spring application context.
In my case, inside a Spring4 Application, i had to use a classic Abstract Factory Pattern(for which i took the idea from - http://java-design-patterns.com/patterns/abstract-factory/) to create instances each and every time there was a operation to be done.So my code was to be designed like:
public abstract class EO {
#Autowired
protected SmsNotificationService smsNotificationService;
#Autowired
protected SendEmailService sendEmailService;
...
protected abstract void executeOperation(GenericMessage gMessage);
}
public final class OperationsExecutor {
public enum OperationsType {
ENROLL, CAMPAIGN
}
private OperationsExecutor() {
}
public static Object delegateOperation(OperationsType type, Object obj)
{
switch(type) {
case ENROLL:
if (obj == null) {
return new EnrollOperation();
}
return EnrollOperation.validateRequestParams(obj);
case CAMPAIGN:
if (obj == null) {
return new CampaignOperation();
}
return CampaignOperation.validateRequestParams(obj);
default:
throw new IllegalArgumentException("OperationsType not supported.");
}
}
}
#Configurable(dependencyCheck = true)
public class CampaignOperation extends EO {
#Override
public void executeOperation(GenericMessage genericMessage) {
LOGGER.info("This is CAMPAIGN Operation: " + genericMessage);
}
}
Initially to inject the dependencies in the abstract class I tried all stereotype annotations like #Component, #Service etc but even though Spring context file had ComponentScanning for the entire package, but somehow while creating instances of Subclasses like CampaignOperation, the Super Abstract class EO was having null for its properties as spring was unable to recognize and inject its dependencies.After much trial and error I used this **#Configurable(dependencyCheck = true)** annotation and finally Spring was able to inject the dependencies and I was able to use the properties in the subclass without cluttering them with too many properties.
<context:annotation-config />
<context:component-scan base-package="com.xyz" />
I also tried these other references to find a solution:
http://www.captaindebug.com/2011/06/implementing-springs-factorybean.html#.WqF5pJPwaAN
http://forum.spring.io/forum/spring-projects/container/46815-problem-with-autowired-in-abstract-class
https://github.com/cavallefano/Abstract-Factory-Pattern-Spring-Annotation
http://www.jcombat.com/spring/factory-implementation-using-servicelocatorfactorybean-in-spring
https://www.madbit.org/blog/programming/1074/1074/#sthash.XEJXdIR5.dpbs
Using abstract factory with Spring framework
Spring and Abstract class - injecting properties in abstract classes
Inject spring dependency in abstract super class
Spring autowire dependency defined in an abstract class
Spring can you autowire inside an abstract class?
Please try using **#Configurable(dependencyCheck = true)** and update this post, I might try helping you if you face any problems.
So precisely my point here is you don't need to get a bean from spring context all the time.
I am fairly new to Sprint and am using Spring 3.x and roo1.1.1 for my application.
I have multiple implementation of an interface which would be #Autowired into other different classes. I would only be able to decide which implementation to go with at the runtime. This should be achieved with like a factory pattern.
public interface SomeInterface {
public void doSomething();
}
Implementation 1.
public class SomeOb implements SomeInterface {
public void doSomething() {
//Do something for first implementation here
}
}
Implementation 2.
public class SomeOtherOb implements SomeInterface {
public void doSomething() {
//Do something for first implementation here
}
}
Now in my service i needed this Autowired like
#Service
public class MyService {
#Autowired
SomeInterface ob;
//Rest of the code here
}
1) The logic to choose which implementation to be Autowired is only know runtime, so i cannot use the #Qualifier annotation to qualify this.
2) I tried to create a FactoryBean like
public class SomeFactoryBean implements FactoryBean<SomeInterface> {
#Override
public SomeInterface getObject() throws Exception {
if(/*Somecondition*/) {
return new SomeOb();
} else
return new SomeOtherOb();
}
#Override
public Class<? extends SomeInterface> getObjectType() {
if(/*Somecondition*/) {
return SomeOb.class;
} else
return SomeOtherOb.class;
}
#Override
public boolean isSingleton() {
return false;
}
}
In the applicationContext.xml i have the tag mentioned.
When i run the webserver i run into an error like
No unique bean of type [com.xxxx.xxxx.SomeInterface] is defined: expected single matching bean but found 3: [xxxx, xxxxxxx, xxxxFactory]
Can anyone please help me to resolve this issue. If i am not doing this right please direct me to do this the right way.
Thanks and appreciate any help,
jjk
Thanks for the suggestion. I was able to solve the problem with help from a colleague. What i was doing wrong
I had the implementation of the SomeInterface with #Service. So this was picked up by the spring scanner and added to the bean.
During trial and error i removed the #Component annotation from by FactoryBean implementation.
After making these changes it worked like a charm.
return true from isSingleton() if you only need one implementation of the bean for a given instance of your application
But I question your design.
I would always use properties files to switch out implementations like this. I once had to implement CAPTCHA integration for a site. We were prototyping with the JCaptcah and ReCAPTCHA APIs. I created a new interface that contained just the functionality we needed and then created implementations for both APIs. Using a placeholders in the Spring configuration file and Maven profiles, we could switch out the implementation class at compile time or deployment time, for example, mvn jetty:run -DcaptchaImpl=recaptcha or -DcaptchaImpl=jcaptcha.
Without knowing the task that you want to accomplish, it's hard to provide more advice.