I searched the web for a clear example on how to invoke my custom method annotation using Spring AOP but couldn't find a clear example.
I am building a framework to inject a user profile in the context when certain methods on any POJO is called.
The framework API should be invoked via a custom method annotation, say #RunAsAdmin. I can build the annotation part and parser, my question is what is the best way to invoke my parser when the annotated method is called.
We are using Spring 3.0 and would like to know what is the best practice to configure Spring framework to understand those methods annotated with specific annotation and invoke my annotation handler (parser).
Any example or guidance will be highly appreciated.
Intercepting a call on any annotated Spring bean method is straightforward, see the example below.
Intercepting a call on a POJO is not possible by definition - you have to return a proxy instead.
That is, you should implement a factory for these POJOs that will return proxies instead of real POJOs. You might use ProxyFactoryBean http://docs.spring.io/autorepo/docs/spring-framework/3.2.0.BUILD-SNAPSHOT/api/org/springframework/aop/framework/ProxyFactoryBean.html for this.
//intercepting a call to any method annotated with #com.yours.RunAsAdmin of any Spring-managed bean
#Component #Aspect class RunAsAdminAspect {
#Autowired
private YourRunAsAdminHandler handler;
#Before("#annotation(com.yours.RunAsAdmin)")
public void doRunAsAdmin() {
handler.grantAdminRightsToCurrentUser(); //your real stuff
}
}
Related
I am trying to understand how ControllerAdvice is working in SpringBoot. It is suggested that for per application there should be one ControllerAdvice. But my question is how this ControllerAdvice is tied to Controllers and catching the exceptions. So basically what is the underhood?
Spring AOP works with proxies.
That is when you annotate a class with any of AOP annotation spring will create a proxy class by extending your annotated class and all the methods will be overridden there in the proxy class.
Thus here after when you call a method in your class spring will call the proxy object method first then your actual method. This is Spring AOP knows whether the method has been called or thrown some exception or returned successfully etc etc.
This is the reason why when you call a private method with in the class AOP cannot intercept that method call.
This should happen quite often:
RestController -> SomeClass -> SomeOtherClass -> YetAnotherClass and so on...
In my specific case there is a chain of responsibility which is injected to a rest controller. Each class is injected to it's previous class in the above chain.
I have implemented this with spring boot and I'm trying to test the REST resource. I want to Mock the "YetAnotherClass" so that when I send a request with MockMvc I can verify that something has happened in the mock object.
The problem is if I use #MockBean to mock YetAnotherClass then I have to inject it to SomeOtherClass. I have tried to inject it with #TestConfiguration but it seems that the Mock object injection doesn't work this way when the request is sent through MockMvc and the mock object is nested deep inside a chain such as above. (The original bean is injected not the mock one)
I know that JMockit mocks every instance of a class so it would solve my problem. But Spring boot defaults to Mockito and I prefer to avoid inconsistencies.
How can I implement such a test scenario?
I've run into a lot of annoyance using Mockito's annotation config setup when setting up Spring JUnit text fixtures.
I've found the way I like mocking beans with external integrations like this this by essentially having a separate MockObjectsConfig class with the mock objects I want using the standard Spring Context Configuration, and then import it alongside my real test config:
#Configuration
public class MockObjectsConfig {
#Bean
public YetAnotherClass yetAnotherClass() {
Mockito.mock(YetAnotherClass.class); // and add any thenReturns, answers, etc. here
}
... More mock beans...
}
Then include it in your test like so:
#RunWith(SpringRunner.class)
#ContextConfiguration(classes = { MyRealConfigClass.class, MockObjectsConfig.class)
public class MyJunitTest {
#Autowired
private RestController restController;
}
You can also annotate your mock bean with #Profile and test with #ActiveProfiles if you need to prevent a conflict there.
This way your mock YetAnotherClass will get injected into your context like all your other beans -- no relying on, mixing, and fiddling around with Mockito and other library annotations.
I am new to AOP and I am trying to understand the difference between Method Interceptor and MethodAdvice(i.e. MethodBeforeAdvice or MethodAfterAdvice). To me looks like both are doing the same thing i.e. are called on method invocation. When should we use MethodInterceptor vs MethodAdvice.
Take a look at the definition of the org.aopalliance.interceptInterceptor interface (implemented by MethodInterceptor):
public interface Interceptor extends Advice {
}
It's easy to see that a MethodInterceptor actually IS an Advice.
The only difference between an Advice being defined in an #Aspect class and such an Interceptor is that Interceptor implementations can be added to and removed from Spring AOP Proxies at runtime (casting them to 'Advised'), whereas the Advice you're talking about is a more static construct. But their still essential to Spring AOP since their presence tells Spring which beans to wrap in a proxy object during application context startup.
As part of the Spring MVC initialization, I need to run an action (just calling a method on a 3rd party library) once as a setup operation. I'm working in Spring MVC environment where I don't really have control over the web.xml or anything, so I can't add a servlet context listener or anything. I tried making an implementation of a WebApplicationInitializer but it never seems to get called (no idea why though, or how to try and debug that any further).
If I annotate a class with #Configuration it does get created, so I'm wondering if I can use the class's constructor to perform that setup operation (calling a 3rd party setup method). Is this appropriate/safe to do? Are there any other alternatives for this kind of thing? I'm new to Spring, so I might just be missing something that's meant for this kind of thing.
Thanks
Configuration class would be an appropriate place to contain some initialization logic. You can place it in a constructor, method annotated with #PostConstruct or afterPropertiesSet() method if you implement the InitializingBean interface for example. The difference is that the constructor code will be called before the beans in your configuration class are instantiated, so if your initialization code depends on some Spring beans, go with the #PostConstruct / InitializingBean approach.
Example:
#Configuration
public class Config {
#PostConstruct
public void initialize() {
// Run some action
}
}
As far as I know, Spring uses JDK to generate dynamic proxy for the classes that implement any inferface while use Cglib to generate dynamic proxy for the classes that do not implement any inferface. For decarative transcation, Spring uses proxy to add transaction aspect. Please take a look at the code below:
interface Demo {
void methodA();
}
public class DemoImpl implements Demo{
#Transactional
public void updateA() {}
#Transactional
public void updateB() {}
}
I think updateA can work well with transaction. But how about updateB method? Does the #Transactional work for it?
Maybe my understanding is not correct. It's great if the related Spring source code is provided to explain how Spring use JDK/cglib to proxy the class and interface. Thanks
I have the config in the xml:
<tx:annotation-driven transaction-manager="transactionManager" />
JDK dynamic proxy
In this case your bean is wrapped with a proxy implementing Demo interface. From that moment you can only use that interface. Trying to inject or fetch bean of DemoImpl type will result in dreadful Abstract DAO pattern and Spring's "Proxy cannot be cast to ..." problem!
This kind of answers your question - you can only access updateA() and this is the only transactional method. Annotation around updateB() is ignored.
However if you call updateB() from updateA() it will be transactional because it will bind to a transaction started by updateA() (with default transaction propagation).
CGLIB proxy
In this case the interface is ignored. cglib will create a subclass of DemoImpl (obviously also implementing Demo interface) and apply transaction behaviour on both update*() methods. Now if you inject bean of type DemoImpl (interface is not needed in this case at all and Impl suffix is ugly) you can safely and transactionally call both methods.
See my article: Spring pitfalls: proxying and Spring AOP riddle for greater details.