I have the following classes:
public interface IDataSource<T> {
public List<T> getData(int numberOfEntries);
}
public class MyDataSource implements IDataSource<MyData> {
public List<MyData> getData(int numberOfEntries) {
...
}
}
public class MyOtherDataSource implements IDataSource<MyOtherData> {
public List<MyOtherData> getData(int numberOfEntries) {
...
}
}
I would like to use a factory that return the correct implementation based on the data type. I wrote the following but I get "Unchecked cast" warnings:
public static <T> IDataSource<T> getDataSource(Class<T> dataType) {
if (dataType.equals(MyData.class)) {
return (IDataSource<T>) new MyDataSource();
} else if (dataType.equals(MyOtherData.class)) {
return (IDataSource<T>) new MyOtherDataSource();
}
return null;
}
Am I doing it wrong? What can I do to get rid of the warnings?
I am not aware of any way to get rid of those warnings without #SuppressWarnings("unchecked").
You are passing in a Class object so T can be captured. But you are forced to check the Class at runtime to determine which IDataSource<T> to return. At this time, type erasure has long since occurred.
At compile time, Java can't be sure of type safety. It can't guarantee that the T in the Class at runtime would be the same T in the IDataSource<T> returned, so it produces the warning.
This looks like one of those times when you're forced to annotate the method with #SuppressWarnings("unchecked") to remove the warning. That warning is there for a reason, so it is up to you to provide and ensure type safety. As written, it looks like you have provided type safety.
#SuppressWarnings("unchecked")
public static <T> IDataSource<T> getDataSource(Class<T> dataType) {
You're doing it right, and you should simply suppress the warnings. Factories are one of the tricky areas in generics where you really do need to manually cast to a generic type, and you have to ensure via whatever means that the returned value matches the Class<T> you pass in. For example, in this case you're hard-coding a couple of IDataSource implementations, so I would recommend writing unit tests that verify that the types are correct so that if the MyData implementation changes in an incompatible way, you'll get an error on build.
Just annotate the getDataSource method with #SuppressWarnings("unchecked"), and it's always a good idea to add an explanatory comment when suppressing warnings.
Generics are for compile-time type safety. They can't be used for runtime type determination like that. To get rid of the warning, you can do something like #SuppressWarnings("unchecked") or use the -Xlint:-unchecked compiler flag, as described in the "Raw Types" part of the Java tutorial.
The other answers have answered the problem as you posed it. But I'd like to take a step back to understand what you're trying to accomplish with this factory method. This factory basically provides a map of data types to IDataSource parameters. Dependency injection might be a more appropriate pattern since this is a small well-known set of data types and implementations (as indicated by your example).
Let's say you want to store all Widgets in Mongo but all Gadgets in Mysql, you might have two classes: a MongoWidgetDataSource that implements IDataSource<Widget> and a MysqlGadgetDataSource that implements IDataSource<Gadget>.
Instead of hardcoding a factory method call like MyFactory.getDataSource(Widget.class) inside a data consumer, I would inject the appropriate IDataSource dependency. We might have MyService that does something with widgets (stored in mongo). Using a factory as you proposed would look like this:
public class MyService {
public void doSomething() {
String value = MyFactory.getDataSource(Widget.class).getSomething();
// do something with data returned from the source
}
}
Instead, you should inject the appropriate data source as a constructor arg into the service:
public class MyService {
private final IDataSource<Widget> widgetDataSource;
public MyService(IDataSource<Widget> widgetDataSource) {
this.widgetDataSource = widgetDataSource;
}
public void doSomething() {
String value = widgetDataSource.getSomething();
// now do something with data returned from the source
}
}
This has the added benefit of making your code more reusable and easier to unit test (mock dependencies).
Then, where you instantiate MyService, you can also wire up your data sources. Many projects use a dependency injection framework (like Guice) to make this easier, but its not a strict requirement. Personally, though, I never work on a project of any real size or duration without one.
If you don't use an DI framework, you just instantiate the dependencies when you create the calling service:
public static void main(String[] args) {
IDataSource<Widget> widgetDataSource = new MongoWidgetDataSource();
IDataSource<Gadget> gadgetDataSource = new MysqlGadgetDataSource();
MyService service = new MyService(widgetDataSource, gadgetDataSource);
service.doSomething();
}
In Guice, you would wire up these data sources like this:
public class DataSourceModule extends AbstractModule {
#Override
protected void configure() {
bind(new TypeLiteral<IDataSource<Widget>>() {}).to(MongoWidgetDataSource.class);
bind(new TypeLiteral<IDataSource<Gadget>>() {}).to(MysqlGadgetDataSource.class);
}
}
Dependency inversion is a bit of a different way to think about the problem, but it can lead to a much more decoupled, reusable and testable code base.
This seems to work:
public static <T> IDataSource<T> getDataSource(MyData dataType) {
System.out.println("Make MyDataSource");
return (IDataSource<T>) new MyDataSource();
}
public static <T> IDataSource<T> getDataSource(MyOtherData dataType) {
System.out.println("Make MyOtherDataSource");
return (IDataSource<T>) new MyOtherDataSource();
}
public void test() {
IDataSource<MyData> myDataSource = getDataSource((MyData) null);
IDataSource<MyOtherData> myOtherDataSource = getDataSource((MyOtherData) null);
}
You may prefer to create empty archetypes rather than cast null like I have but I think this is a viable technique.
Related
I am working on a REST API where I have an interface that defines a list of methods which are implemented by 4 different classes, with the possibility of adding many more in the future.
When I receive an HTTP request from the client there is some information included in the URL which will determine which implementation needs to be used.
Within my controller, I would like to have the end-point method contain a switch statement that checks the URL path variable and then uses the appropriate implementation.
I know that I can define and inject the concrete implementations into the controller and then insert which one I would like to use in each particular case in the switch statement, but this doesn't seem very elegant or scalable for 2 reasons:
I now have to instantiate all of the services, even though I only need to use one.
The code seems like it could be much leaner since I am literally calling the same method that is defined in the interface with the same parameters and while in the example it is not really an issue, but in the case that the list of implementations grows ... so does the number of cases and redundant code.
Is there a better solution to solve this type of situation? I am using SpringBoot 2 and JDK 10, ideally, I'd like to implement the most modern solution.
My Current Approach
#RequestMapping(Requests.MY_BASE_API_URL)
public class MyController {
//== FIELDS ==
private final ConcreteServiceImpl1 concreteService1;
private final ConcreteServiceImpl2 concreteService2;
private final ConcreteServiceImpl3 concreteService3;
//== CONSTRUCTORS ==
#Autowired
public MyController(ConcreteServiceImpl1 concreteService1, ConcreteServiceImpl2 concreteService2,
ConcreteServiceImpl3 concreteService3){
this.concreteService1 = concreteService1;
this.concreteService2 = concreteService2;
this.concreteService3 = concreteService3;
}
//== REQUEST MAPPINGS ==
#GetMapping(Requests.SPECIFIC_REQUEST)
public ResponseEntity<?> handleSpecificRequest(#PathVariable String source,
#RequestParam String start,
#RequestParam String end){
source = source.toLowerCase();
if(MyConstants.SOURCES.contains(source)){
switch(source){
case("value1"):
concreteService1.doSomething(start, end);
break;
case("value2"):
concreteService2.doSomething(start, end);
break;
case("value3"):
concreteService3.doSomething(start, end);
break;
}
}else{
//An invalid source path variable was recieved
}
//Return something after additional processing
return null;
}
}
In Spring you can get all implementations of an interface (say T) by injecting a List<T> or a Map<String, T> field. In the second case the names of the beans will become the keys of the map. You could consider this if there are a lot of possible implementations or if they change often. Thanks to it you could add or remove an implementation without changing the controller.
Both injecting a List or a Map have some benefits and drawbacks in this case. If you inject a List you would probably need to add some method to map the name and the implementation. Something like :
interface MyInterface() {
(...)
String name()
}
This way you could transform it to a Map<String, MyInterface>, for example using Streams API. While this would be more explicit, it would polute your interface a bit (why should it be aware that there are multiple implementations?).
When using the Map you should probably name the beans explicitly or even introduce an annotation to follow the principle of least astonishment. If you are naming the beans by using the class name or the method name of the configuration class you could break the app by renaming those (and in effect changing the url), which is usually a safe operation to do.
A simplistic implementation in Spring Boot could look like this:
#SpringBootApplication
public class DynamicDependencyInjectionForMultipleImplementationsApplication {
public static void main(String[] args) {
SpringApplication.run(DynamicDependencyInjectionForMultipleImplementationsApplication.class, args);
}
interface MyInterface {
Object getStuff();
}
class Implementation1 implements MyInterface {
#Override public Object getStuff() {
return "foo";
}
}
class Implementation2 implements MyInterface {
#Override public Object getStuff() {
return "bar";
}
}
#Configuration
class Config {
#Bean("getFoo")
Implementation1 implementation1() {
return new Implementation1();
}
#Bean("getBar")
Implementation2 implementation2() {
return new Implementation2();
}
}
#RestController
class Controller {
private final Map<String, MyInterface> implementations;
Controller(Map<String, MyInterface> implementations) {
this.implementations = implementations;
}
#GetMapping("/run/{beanName}")
Object runSelectedImplementation(#PathVariable String beanName) {
return Optional.ofNullable(implementations.get(beanName))
.orElseThrow(UnknownImplementation::new)
.getStuff();
}
#ResponseStatus(BAD_REQUEST)
class UnknownImplementation extends RuntimeException {
}
}
}
It passes the following tests:
#RunWith(SpringRunner.class)
#SpringBootTest
#AutoConfigureMockMvc
public class DynamicDependencyInjectionForMultipleImplementationsApplicationTests {
#Autowired
private MockMvc mockMvc;
#Test
public void shouldCallImplementation1() throws Exception {
mockMvc.perform(get("/run/getFoo"))
.andExpect(status().isOk())
.andExpect(content().string(containsString("foo")));
}
#Test
public void shouldCallImplementation2() throws Exception {
mockMvc.perform(get("/run/getBar"))
.andExpect(status().isOk())
.andExpect(content().string(containsString("bar")));
}
#Test
public void shouldRejectUnknownImplementations() throws Exception {
mockMvc.perform(get("/run/getSomethingElse"))
.andExpect(status().isBadRequest());
}
}
Regarding two of your doubts :
1. Instantiating the service object should not be an issue as this is one time job and controller gonna need them to serve all type of request.
2. You can use the exact Path mapping to get rid of switch case. For e.g. :
#GetMapping("/specificRequest/value1")
#GetMapping("/specificRequest/value2")
#GetMapping("/specificRequest/value3")
All of the above mapping will be on separate method which would deal with specific source value and invoke respective service method.
Hope this will help to make code more cleaner and elegant.
There is one more option of separating this on service layer and having only one endpoint to serve all types of source but as you said there is different implementation for each source value then it says that source is nothing but a resource for your application and having separate URI/separate method makes the perfect sense here. Few advantages that I see here with this are :
Makes it easy to write the test cases.
Scaling the same without impacting any other source/service.
Your code dealing the each source as separate entity from other sources.
The above approach should be fine when you have limited source values. If you have no control over source value then we need further redesign here by making source value differentiate by one more value like sourceType etc. and then having separate controller for each group type of source.
I write simple application. I don't want to use any frameworks. Please suggest me right place to hold annotation processing.
I have a few lines in main method:
String myString = (#NonNull String)list;
And I created #interface:
#Target({ElementType.TYPE_USE, ElementType.TYPE_PARAMETER})
public #interface NonNull {
}
Which step should I take next? Can I work with annotations without using reflection? Could you expose for me samples of such annotation processing code?
There is no way (AFAIK) to work with annotations without reflection.
If you don't want to use any framework, first step is to write kind of proxy class handling the method requests. It is an example of method processing with annotation use over method:
public class MyProxy {
private <T> T getProxy(T t) {
return (T) Proxy.newProxyInstance(t.getClass().getClassLoader(), new Class<?>[]{MyClass.class}, new MyInvocationHandler(t));
}
}
And then implement InvocationHandler:
public class MyInvocationHandler implements InvocationHandler {
private Object obj;
MyInvocationHandler (Object obj) {
this.obj = obj;
}
#Override
public Object invoke(Object proxy, final Method method, final Object[] args) throws Throwable {
boolean isNotNull = method.isAnnotationPresent(NotNull.class);
if (isNotNull) {
/* process annotated method. Or go through proxy object fields etc.. */
}
}
}
I hope it will help you.
You didn't say what kind of annotation processing you want to do.
Do you want to add a run-time check that will cause your code to crash if list is ever null at run time? For this, reflection will work.
Do you want to add a compile-time check that will reject your code if it cannot prove that list is never null at run time? For this, an annotation processor such as the Checker Framework will work.
Your question does not explain why you don't want to use a framework. Doing so will save you from re-implementing a lot of functionality that others have already created.
I am using Guice and FactoryModuleBuilder. Typically, it is enough to just define an interface of the factory and Guice will automatically inject the implementation.
However, the part that I am struggling is that the methods in a factory uses generics. Suppose I have the following. A base type of constructed instances as defined by the interface.
interface Foo<T> {
T get();
}
And two implementations of the Foo interface as defined by the two classes below.
class FooA<T> implements Foo<T> {
#Inject
FooA(#Assisted Class<T> clazz, #Assisted String s) {...}
}
class FooB<T> implements Foo<T> {
#Inject
FooB(#Assisted Class<T> clazz, #Assisted Integer i) {...}
}
Then I have the factory interface as defined below, using two custom binding annotations that allows me to use multiple implementations.
interface FooFactory {
#A Foo<T> build(Class<T> clazz, String s);
#B Foo<T> build(Class<T> clazz, Integer i);
}
I have tried a number of possible solutions, all but one has worked so far. The solution that worked is to basically write my own implementation of FooFactory as shown below. And in the configure method of the module, bind the interface to the implementation; bind(FooFactory.class).to(FooFactoryImpl.class);
class FooFactoryImpl {
Foo<T> build(Class<T> clazz, String s) {
return new FooA(clazz, s):
}
Foo<T> build(Class<T> clazz, Integer i) {
return new FooB(clazz, i);
}
}
However, I have one issue with this solution. The instances are not created by Guice and thus I lose the null checks that comes with Guice. This is drastically different from my other factories that does not have this problem. This means I have to explicitly write null checks for every implementation of Foo. I would like to avoid that.
The following are some of the solutions I have tried.
Solution 1:
FactoryModuleBuilder fmb = new FactoryModuleBuilder()
.implement(Foo.class, A.class, FooA.class)
.implement(Foo.class, B.class, FooB.class);
install(fmb.build(FooFactory.class));
Solution 2:
FactoryModuleBuilder fmb = new FactoryModuleBuilder()
.implement(TypeLiteral.get(Foo.class), A.class, TypeLiteral.get(FooA.class))
.implement(TypeLiteral.get(Foo.class), B.class, TypeLiteral.get(FooB.class));
install(fmb.build(TypeLiteral.get(FooFactory.class)));
The sample code is available at GitHub (if anyone is interested).
To my knowledge, you can't design AssistedInject factories to work in this way. However, it seems to me you're doing too much in one class. Because you have no restrictions on Class<T>, it's clear you aren't using any methods of this class in the constructor, which means, it should be fairly easy to refactor the behavior into a separate class. I know this is a little bit of boilerplate, it's not exactly what you want, but it might look something like this:
interface FooDataFactory {
#A FooData build(String s);
#B FooData build(Integer i);
}
public class FooA<T> implements Foo<T> {
public FooA(FooData data) {
// You should know what class you need when you're actually calling the constructor.
// This way you don't even need to pass around Class<T> instances
}
}
If this approach doesn't work for your use case, let me know and I'll edit to compensate.
In Java, it is possible to make a bounder type parameter have to extend from a particular class or interface e.g.
public class Box<T extends MyClass> {
T t
...
}
Is there anyway I can bound by an annotation so that values of T can only be classes that have a particular annotation?
Starting in Java 8, you can write
public class Box<T extends #MyAnno MyClass> {
...
}
As with any Java annotation, to enforce the semantics, you need to use an annotation processor. The Checker Framework is one annotation-processing tool that enforces the semantics for you: you can configure it to issue an error if you ever try to instantiate the Box type using a type argument that lacks the #MyAnno annotation.
Unfortunately there is no way to express it in java AFAIK. <T annotatedWith MyAnnotation> would be so convenient in some cases, but it would add a new keyword and honestly generics are difficult enough ;)
Otherwise for an annotation, as #duckstep said it is easy to check at runtime with
t.getClass().isAnnotationPresent(annotationClass)
For an annotation processor, though, the APIs are much more tough to deal with. Here is some code if it can help some people :
private boolean isAnnotationPresent(TypeElement annotationTypeElement, String annotationName) {
for (AnnotationMirror annotationOfAnnotationTypeMirror : annotationTypeElement.getAnnotationMirrors()) {
TypeElement annotationOfAnnotationTypeElement = (TypeElement) annotationOfAnnotationTypeMirror.getAnnotationType().asElement();
if (isSameType(annotationOfAnnotationTypeElement, annotationName)) {
return true;
}
}
return false;
}
private boolean isSameType(TypeElement annotationTypeElement, String annotationTypeName) {
return typeUtils.isSameType(annotationTypeElement.asType(), elementUtils.getTypeElement(annotationTypeName).asType());
}
Here's my use case:
I need to do some generic operation before and after each method of a given class, which is based on the parameter(s) of the method. For example:
void process(Processable object) {
LOGGER.log(object.getDesc());
object.process();
}
class BaseClass {
String method1(Object o){ //o may or may not be Processable(add process logic only in former case)
if(o intstanceof Prcessable){
LOGGER.log(object.getDesc());
object.process();
}
//method logic
}
}
My BaseClass has a lot of methods and I know for a fact that the same functionality will be added to several similar classes as well in future.
Is something like the following possible?
#MarkForProcessing
String method1(#Process Object o){
//method logic
}
PS: Can AspectJ/guice be used? Also want to know how to implement this from scratch for understanding.
Edit: Forgot to mention, what I have tried.(Not complete or working)
public #interface MarkForProcessing {
String getMetadata();
}
final public class Handler {
public boolean process(Object instance) throws Exception {
Class<?> clazz = instance.getClass();
for(Method m : clazz.getDeclaredMethods()) {
if(m.isAnnotationPresent(LocalSource.class)) {
LocalSource annotation = m.getAnnotation(MarkForProcessing.class);
Class<?> returnType = m.getReturnType();
Class<?>[] inputParamTypes = m.getParameterTypes();
Class<?> inputType = null;
// We are interested in just 1st param
if(inputParamTypes.length != 0) {
inputType = inputParamTypes[0];
}
// But all i have access to here is just the types, I need access to the method param.
}
return false;
}
return false;
}
Yes, it can be done. Yes, you can use AspectJ. No, Guice would only be tangentially related to this problem.
The traditional aspect approach creates a proxy which is basically a subclass of the class you've given it (e.g. a subclass of BaseClass) but that subclass is created at runtime. The subclass delegates to the wrapped class for all methods. However, when creating this new subclass you can specify some extra behavior to add before or after (or both) the call to the wrapped class. In other words, if you have:
public class Foo() {
public void doFoo() {...}
}
Then the dynamic proxy would be a subclass of Foo created at runtime that looks something like:
public class Foo$Proxy {
public void doFoo() {
//Custom pre-invocation code
super.doFoo();
//Custom post-invocation code
}
}
Actually creating a dynamic proxy is a magical process known as bytecode manipulation. If you want to to do that yourself you can use tools such as cglib or asm. Or you can use JDK dynamic proxies. The main downside to JDK proxies are that they can only wrap interfaces.
AOP tools like AspectJ provide an abstraction on top of the raw bytecode manipulation for doing the above (you can do a lot with bytecode manipulation, adding behavior before and after methods is all aspects allow). Typically they define 'Aspect's which are classes that have special methods called 'advice' along with a 'pointcut' which defines when to apply that advice. In other words you may have:
#Aspect
public class FooAspect {
#Around("#annotation(MarkForProcessing)")
public void doProcessing(final ProceedingJoinPoint joinPoint) throws Throwable
{
//Do some before processing
joinPoint.proceed(); //Invokes the underlying method
//Do some after processing
}
}
The aspect is FooAspect, the advice is doProcessing, and the pointcut is "#annotation(MarkForProcessing)" which matches all methods that are annotated with #MarkForProcessing. It's worth pointing out that the ProceedingJoinPoint will have a reference to the actual parameter values (unlike the java.lang.reflect.Method)
The last step is actually applying your aspect to an instance of your class. Typically this is either done with a container (e.g. Guice or Spring). Most containers have some way of knowing about a collection of aspects and when to apply them to classes constructed by that container. You can also do this programmatically. For example, with AspectJ you would do:
AspectJProxyFactory factory = new AspectJProxyFactory(baseClassInstance);
factory.addAspect(FooAspect.class);
BaseClass proxy = factory.getProxy();
Last, but not least, there are AOP implementations which use compile-time "weaving" which is a second compilation step run on the class files that applies the aspects. In other words, you don't have to do the above or use a container, the aspect will be injected into the class file itself.