I have a Play Controller that has a couple of methods that are very similar. I wondered how I can reduce boilerplate:
public static Result foo() {
// boilerplate
if (!PREVIEW) {
return redirect(routes.Application.login());
}
// other code
...
return ok("...");
}
public static Result bar() {
// boilerplate
if (!PREVIEW) {
return redirect(routes.Application.login());
}
// other code
...
return ok("...");
}
PREVIEW is shorthand for a configuration setting.
I created an Action like this:
public class PreviewAction extends Action.Simple {
public F.Promise<Result> call(Http.Context ctx) throws Throwable {
if (!PREVIEW) {
return F.Promise.pure(redirect(routes.Application.login()));
}
return delegate.call(ctx);
}
}
Now I can use an annotation on my other actions and it works like it did before:
#With(PreviewAction.class)
public static Result foo() {
...
}
More information: https://www.playframework.com/documentation/2.4.x/JavaAsync
Thanks to Tunaki for pointing me in the right direction.
Related
So i'm trying to develop a fluent interface for some DSL in Java and am running into a problem. The interface consists of builder pattern classes that each construct part of the DSL. The problem is that a specific builder sometimes needs to transfer control to a different builder which at some point returns to the 'parent' builder. For example, there is a 'SequenceBuilder' that creates a list of statements but it need sometimes transfer control to an 'IfBuilder' that is used to create an 'if' statement. When the IfBuilder is finished, it needs to return to the SequenceBuilder. Now there are some builders that are not always called by the same type of other builder and therefore need to be able to return builders of a different datatype. The example program below demonstrates this:
package com.example.fluent;
public class Test {
public class Type1 {
public Type1 test1() {
System.out.println("test1");
return this;
}
public Type3 gotype3() {
System.out.println("gotype3");
return new Type3<Type1>(this);
}
public void endtype1() {
System.out.println("endtype1");
}
}
public class Type2 {
public Type2 test2() {
System.out.println("test2");
return this;
}
public Type3 gotype3() {
System.out.println("gotype3");
return new Type3<Type2>(this);
}
public void endtype2() {
System.out.println("endtype2");
}
}
public class Type3<T> {
private T parent;
public Type3(T parent) {
this.parent = parent;
}
public Type3 test3() {
System.out.println("test3");
return this;
}
public T endtype3() {
System.out.println("endtype3");
return parent;
}
}
public static void main(String[] args) {
new Test().run();
}
private void run() {
new Type1()
.test1()
.gotype3()
.test3()
.endtype3()
.test1()
.endtype1();
}
}
You can see in the .run() method that i start by creating a new instance of the Type1 class, which follows the builder pattern. At some point i'm calling the .goType3() method which transfers control to the Type3 builder. Because it has to return control at some point to Type1 again, a reference to that builder is passed via constructor to Type3. When it's time to return to Type1, the method .endtype3() is called. And here lies the problem. I'm using generics to (try) to return the datatype of Type1 but instead it's converted to an Object type. That datatype obviously does not have the methods that Type1 has and therefore the pattern is broken.
Question: is there some other way to return the proper datatype to the parent builder?
You're not using generics as much as you want to. You're using return type Type3. You need to use Type3<Type1>, Type3<Type2> and Type3<T> instead.
Thanks to Rob Spoor i got my code finally working, though i didn't understand it at first. The solution lies in changing the code in the following ways:
package com.example.fluent;
public class Test {
public class Type1 {
public Type1 test1() {
System.out.println("test1");
return this;
}
public Type3<Type1> gotype3() {
System.out.println("gotype3");
return new Type3<Type1>(this);
}
public void endtype1() {
System.out.println("endtype1");
}
}
public class Type2 {
public Type2 test2() {
System.out.println("test2");
return this;
}
public Type3<Type2> gotype3() {
System.out.println("gotype3");
return new Type3<Type2>(this);
}
public void endtype2() {
System.out.println("endtype2");
}
}
public class Type3<T> {
private T parent;
public Type3(T parent) {
this.parent = parent;
}
public Type3<T> test3() {
System.out.println("test3");
return this;
}
public T endtype3() {
System.out.println("endtype3");
return parent;
}
}
public static void main(String[] args) {
new Test().run();
}
private void run() {
// new Type1().test1().test1().endtype1();
new Type1().test1().gotype3().test3().endtype3().test1().endtype1();
// new Type2().test2().gotype3().test3().endtype3().test2().endtype2();
}
}
See the changes in the return types of the .gotype3() methods in the Type1 and Type2 classes, as well as the return type of the .test3() method. Now everything is working fine.
Using Guava Listenable Futures
Assume I have the following class:
public class FooService {
ListenableFuture<Foo> getFoo() {
//code to create callable, then
return listeningExecutorService.submit(fooCallable);
}
}
and the following class:
public class BarService {
ListenableFuture<Bar> getBar(Foo foo) {
//code to create callable, then
return listeningExecutorService.submit(barCallable);
}
}
Note that getBar requires a Foo in the parameters.
If I want to chain these two operations together I would write a transformer function like this:
AsyncFunction<Foo, Bar> fooToBar = new AsyncFunction<Foo, Bar>() {
#Override
ListenableFuture<Bar> apply(Foo resultantFoo) {
return barService.get(resultantFoo);
}
};
and then apply the transformation like this:
public ListenableFuture<Bar> combinedFooToBar() {
ListenableFuture<Foo> futureFoo = fooService.get();
return Futures.transformAsync(futureFoo, fooToBar);
}
Question: what is the equivalent syntax for these classes and transformation function if we were to convert them into RxJava? Assume that we want to convert FooService and BarService into the appropriate RxJava structures. Assume we want to chain async tasks using the result of calling FooService as the parameter for BarService.
NB: I am just starting to learn about RxJava syntax. When I have finished studying the syntax I will attempt answer the question myself. However, in the meantime if anyone wants to answer they are welcome.
The Guava code translates into RxJava2 code as follows:
FooService.java
public class FooService {
Observable<Foo> getFoo() {
return Observable.fromCallable(new Callable<Foo>() {
#Override
public Foo call() throws Exception {
return new Foo();
}
});
}
}
BarService.java
public class BarService {
Observable<Bar> getBar(final Foo foo) {
return Observable.fromCallable(new Callable<Bar>() {
#Override
public Bar call() throws Exception {
return new Bar(foo);
}
});
}
}
FooBarService.java
public class FooBarService {
private final FooService fooService;
private final BarService barService;
public FooBarService(FooService fooService, BarService barService) {
this.fooService = fooService;
this.barService = barService;
}
Observable<Bar> getFooBar() {
return fooService.getFoo()
.concatMap(new Function<Foo, ObservableSource<? extends Bar>>() {
#Override
public ObservableSource<? extends Bar> apply(#NonNull Foo foo) throws Exception {
return barService.getBar(foo);
}
});
}
}
Hence, concatMap and flatMap are similar to Futures.transformAsync and map is similar to Futures.transform (non-async).
Note also this Github project called Future Converter for conversion between ListenableFuture and Observable.
Consider a method
public void doSomething(String actionID){
switch (actionID){
case "dance":
System.out.print("I'm dancing");
break;
case "sleep":
System.out.print("I'm sleeping");
break;
default:
System.out.print("I've no idea what I'm doing");
}
The implementation of the method depends on the value of the parameter. Is there a more elegant way to do this, or a different design pattern to replicate the behaviour?
If the caller decides what logic is executed by passing different strings, then why not just have them call different methods:
public void doSomething(String actionID) {...}
...
doSomething("dance");
doSomething("sleep");
VS.:
public void dance() {...}
public void sleep() {...}
...
dance();
sleep();
It seems like you're unnecessarily funnelling all the calls into doSomething
But the strings might not always be literals. What if you're taking them from the console?
You could provide static mappings from the strings to the corresponding functions:
class MyClass {
private static final Map<String, Consumer<MyClass>> map = new HashMap<>();
static {
map.put("sleep", MyClass::sleep);
map.put("dance", MyClass::dance);
}
public void doSomething(String actionID) {
map.getOrDefault(actionID, MyClass::doNothing).accept(this);
}
public void dance() {
System.out.print("I'm dancing");
}
public void sleep() {
System.out.print("I'm sleeping");
}
private void doNothing() {
System.out.println("I've no idea what I'm doing");
}
}
This makes scenarios where you have a lot of switch cases a lot cleaner.
Introduce an interface, e.g.
public interface HumanState {
public void tellMeWhatYouAreDoing();
}
encapsulate the logic in different implementations
public class DancingState implements HumanState {
#Override
public void tellMeWhatYouAreDoing() {
System.out.println("I'm dancing");
}
}
public class SleepingState implements HumanState {
#Override
public void tellMeWhatYouAreDoing() {
System.out.println("I'm sleeping");
}
}
public class UnknownState implements HumanState {
#Override
public void tellMeWhatYouAreDoing() {
System.out.println("I've no idea what I'm doing");
}
}
and use a map. E.g.
public class HumanStateExample {
public static void main(String[] args) {
HumanStateExample humanStateExample = new HumanStateExample();
humanStateExample.doSomething("dance");
humanStateExample.doSomething("sleep");
humanStateExample.doSomething("unknown");
}
private final HashMap<String, HumanState> humanStateMap;
public HumanStateExample(){
humanStateMap = new HashMap<String, HumanState>();
humanStateMap.put("dance", new DancingState());
humanStateMap.put("sleep", new SleepingState());
}
public void doSomething(String action) {
HumanState humanState = humanStateMap.get(action);
if(humanState == null){
humanState = new UnknownState();
}
humanState.tellMeWhatYouAreDoing();
}
}
I'm not sure how the pattern is called, but it is very useful if you need to delegate the method call based on more than one parameter:
Create a lot of handlers where each one knows when it is responsible for handling a call. Then just loop through them and invoke the first one matching the parameter.
edit: I renamed the class from FancyParameterActionFactory to FancyParameterActionUtility: it is not a factory, the name was misleading
//Your method, but this time with a complex object, not with a simple string.
public void doSomething(FancyParameterObject fpo){
FancyParameterActionUtility.invokeOn(fpo);
}
//The utility which can handle the complex object and decides what to do.
public class FancyParameterActionUtility{
public Interface FPAHandler{
void invoke(FancyParameterObject fpo);
boolean handles(FancyParameterObject fpo);
}
//Omitted: Different implementations of FPAHandler
public static List<FPAHandler> handlers = new LinkedList<>();
static{
handlers.add(new DanceHandler());
handlers.add(new SleepHandler());
//Omitted: Different implementations of FPAHandler
}
public static void invokeOn(FancyParameterObject fpo){
for(FPAHandler handler:handlers){
if (handler.handles(fpo)){
handler.invoke(fpo);
return;
}
}
//Default-Behavior
}
}
Here is a simple implementation of the command pattern based your sample problem. I define a general AbstractCommand abstract class which contains two methods. The first method, createCommand(), instantiates a command class based on an input string name. This is how you can delegate your string input to create the right type of command. The second method is doAction(), and this is left undefined, to be implemented later on by specific concrete command classes.
public abstract class AbstractCommand {
public static AbstractCommand createCommand(String name) {
try {
String clsName = name + "Command";
Class<?> cls = Class.forName(clsName);
AbstractCommand command = (AbstractCommand) cls.newInstance();
return command;
}
catch (Exception e) {
System.out.println("Something went wrong.");
}
}
public abstract void doAction();
}
public class DanceCommand extends AbstractCommand {
public void doAction() {
System.out.println("I'm dancing");
}
}
public class TestCommandPattern {
public void doSomething(String actionID) {
AbstractCommand cmd = AbstractCommand.createCommand(actionID);
cmd.doAction();
}
public static void main(String[] args) {
TestCommandPattern test = new TestCommandPattern();
test.doSomething("Dance"); // should print "I'm dancing"
}
}
Now that this framework has been setup, you could easily add other commands for the various types of actions in your original problem. For example, you could create a SleepCommand class which would output I'm sleeping, or do whatever action you wish.
My application have several execution modes, and in 1 mode it is normal that some of my tests will throw a concrete exception. I need to annotate this methods with something like #SkipOnFail that will set method as skipped if exception was thrown.
thanks in advance!
#Edit(for my question to be more clear)
#Test(expected=ConcreteException.class)
does not work for me because i need my tests to pass even if ConcreteException.class was not thrown(expected tag in junit will mark my test as failed if this exception won't be thrown), and to be skipped otherwise. In all other cases it should work as always.
#Solution that worked for me(junit v4.7) thx to #axtavt
#Rule
public MethodRule skipRule = new MethodRule() {
public Statement apply(final Statement base, FrameworkMethod method, Object target) {
if(method.getAnnotation(SkipOnFail.class) == null) return base;
return new Statement() {
#Override
public void evaluate() throws Throwable {
try{
base.evaluate();
} catch (ConcreteException e) {
Assume.assumeTrue(false);
}
}
};
}
};
#Thx
I don't think that such a feature is available out of the box, but it should be pretty easy to implement with custom TestRule and Assume, something like this:
#Rule
public TestRule skipRule = new TestRule() {
public Statement apply(final Statement base, Description desc) {
if (desc.getAnnotation(SkipOnFail.class) == null) return base;
return new Statement() {
public void evaluate() throws Throwable {
try {
base.evaluate();
} catch (MyExceptoion ex) {
Assume.assumeTrue(false);
}
}
};
}
};
What about using JUnit Extensions?
The following example is taken from their Tutorial.
It provides aditional annotations for Prerequisites (#Prerequisite): Ignore tests based on conditions.
The required approach would be to check this during running tests. So you can simply add a #Prerequisite(requires="") annotation.
public class TestFillDatabase {
#Prerequisite(requires = "databaseIsAvailable")
#Test public void fillData() {
// ...
}
public boolean databaseIsAvailable() {
boolean isAvailable = ...;
return isAvailable;
}
}
public class TestFillDatabase {
#Prerequisite(requires = "databaseIsAvailable")
#Test public void fillData() {
// ...
}
public boolean databaseIsAvailable() {
boolean isAvailable = ...;
return isAvailable ;
}
}
This specified methods with #Prerequisite(requires = "databaseIsAvailable") must be a public method, returning a boolean or Boolean value.
If these methods will be consolidated in helper classes, you can also specify static methods within a class to be called using #Prerequisite(requires = "databaseIsAvailable", callee="DBHelper").
public class TestFillDatabase {
#Prerequisite(requires = "databaseIsAvailable", callee="DBHelper")
#Test public void fillData() {
// ...
}
}
public class DBHelper {
public static boolean databaseIsAvailable() {
boolean isAvailable = ...;
return isAvailable ;
}
}
Also using the Assume class (since jUnit 4.4), you can use assumeNoException():
try{
base.evaluate();
} catch (ConcreteException e) {
Assume.assumeNoException("Concrete exception: skipping test", e);
}
I searched for the docs about JUnit and it appears that from version 4.9 they have introduced what they call test rules (see TestRule). You may start from this.
The ExpectedException class marked as #Rule could be of some help in order to check for exceptions thrown but not mandatory for the test to pass.
For more advanced usage I cannot say for the moment as I've just discovered it.
I have a class which I would like to test with a public static method that contains some chained method calls. Assuming that an exception occurs during the chained method calls, how do I handle this effectively and make it return some specific value?
Following is the code sample of the test class.
#RunWith(PowerMockRunner.class)
#PrepareForTest({CodeWithPrivateMethod.class,CodeWithAnotherPrivateMethod.class,CodeWithYetAnotherPrivateMethod.class})
public class CodeWithPrivateMethodTest {
#Test
public void when_gambling_is_true_then_always_explode() throws Exception {
CodeWithYetAnotherPrivateMethod codeWithYetAnotherPrivateMethod = PowerMockito.spy(new CodeWithYetAnotherPrivateMethod());
PowerMockito.whenNew(CodeWithYetAnotherPrivateMethod.class).withAnyArguments().thenReturn(codeWithYetAnotherPrivateMethod);
CodeWithAnotherPrivateMethod codeWithAnotherPrivateMethod = PowerMockito.spy(new CodeWithAnotherPrivateMethod());
PowerMockito.whenNew(CodeWithAnotherPrivateMethod.class).withAnyArguments().thenReturn(codeWithAnotherPrivateMethod);
PowerMockito.doReturn(true).when(codeWithYetAnotherPrivateMethod, "getGambling");
//PowerMockito.doReturn(codeWithYetAnotherPrivateMethod).when(codeWithAnotherPrivateMethod, "getGambleValue");
PowerMockito.spy(CodeWithPrivateMethod.class);
CodeWithPrivateMethod.startGamble();
}
}
Following is the code sample for the class under test
public class CodeWithPrivateMethod {
public static void startGamble() {
Boolean gamble = CodeWithAnotherPrivateMethod.getGambleValue()
.getGambling();
if (gamble) {
System.out.println("kaboom");
}else{
System.out.println("boom boom");
}
}
}
Following is the code sample for the class that gets called from the class under test
public class CodeWithAnotherPrivateMethod {
static CodeWithYetAnotherPrivateMethod codeWithYetAnotherPrivateMethod = new CodeWithYetAnotherPrivateMethod();
public static CodeWithYetAnotherPrivateMethod getGambleValue() {
return codeWithYetAnotherPrivateMethod; //works fine
return null; // fails
}
}
Following is the code sample for the other class that gets called from the class under test
public class CodeWithYetAnotherPrivateMethod {
public Boolean getGambling() {
return false;
}
}
So Assuming I return a null value from getGambleValue() method of CodeWithAnotherPrivateMethod class, how do I handle this null value effectively in my testclass?
This is how to specify expected exceptions using Mockito:
#Test(expected = NullPointerException.class)
public void when_gambling_is_true_then_always_explode() throws Exception {
...
Before I found out about this I would do:
#Test
public void when_gambling_is_true_then_always_explode() throws Exception {
// setup omitted
try {
CodeWithPrivateMethod.startGamble();
}
catch(NullPointerException e) {
// expected
return;
}
fail("Expected NullPointerException");
}
EDIT: Testing multiple classes that call each other statically like this is a severe code smell. Unit tests should test a single class and inline static calls should be limited to utility classes.
Another comment: your example class names are very confusing. Next time please stick with Foo, Bar, Baz or Appple, Pear, Banana.
If you are not getting an NPE then I expect your mocking/spying is interfering. If you call the code under test without mocking/spying the call chain would be:
CodeWithPrivateMethod.startGamble();
->
CodeWithYetAnotherPrivateMethod value = CodeWithAnotherPrivateMethod.getGambleValue();
->
return null;
<-
value.getGambling();
<- throws NullPointerException
What exactly are you trying to find out or achieve?
EDIT: Here's how it should work with PowerMock
#RunWith(PowerMockRunner.class)
#PrepareForTest(CodeWithAnotherPrivateMethod.class)
public class CodeWithPrivateMethodTest {
#Mock
private CodeWithYetAnotherPrivateMethod yetAnotherInstance;
#Test
public final void testStartGamble() {
// SETUP
mockStatic(CodeWithAnotherPrivateMethod.class);
expect(CodeWithAnotherPrivateMethod.getGambleValue())
.andReturn(yetAnotherInstance);
Boolean gamblingValue = true;
expect(yetAnotherInstance.getGambling()).andReturn(gamblingValue);
replayAll();
// CALL
CodeWithPrivateMethod.startGamble();
// VERIFY
verifyAll();
}