The following code fails at compile time:
object Foo {
case class A()
case class B(a: A)
case class C(b: B)
lazy val a = wire[A]
// Error:(14, 22) Cannot find a value of type: [QuickMain.B]
lazy val c = wire[C]
}
Is it possible to get macwire to automatically infer that it can create a B by creating an A (Honestly, lazy val a = wire[A] shouldn't even be necessary)? If macwire can't do it, is there another Scala framework that can do it in a typesafe manner (I know of Dagger, but I'm looking for a scala-based solution).
To illustrate, in Guice I could do the following:
public class JavaExample {
public static class A {}
public static class B {
#Inject
public B(A a) {}
}
public static class C {
#Inject
public C(B b) {}
}
public static void main(String[] args) {
// No bindings necessary!
var injector = Guice.createInjector();
System.out.println(injector.getInstance(C.class));
}
}
Related
I would like to write a Scala test which checks the exact class of an object created with factory method.
class Base { }
class Derived extends Base { }
class TestSpec {
test("test instance class") {
val result = new Derived()
// I want to check that result is exactly Derived type
result should be a[Derived]
result should not be a[Base]
}
}
I'm looking for something to test if object returned from my factory method for specific parameters is always base class. So it would be something like this is C#:
public class Base { }
public class Derived : Base { }
public class Program
{
public static void Main(string[] args)
{
var baseObject = new Base();
var derivedObject = new Derived();
var baseResult = baseObject.GetType().IsSubclassOf(typeof(Base));
var derivedResult = derivedObject.GetType().IsSubclassOf(typeof(Base));
Console.WriteLine(string.Format("Base is subclass of Base: {0}\nDerived is subclass of Base: {1}", baseResult, derivedResult));
}
}
you can do this :
class A{}
class B extends A{}
val result = new B()
result.getClass.getName shouldEqual classOf[B].getName
result.getClass.getName.equals(classOf[A].getName) shouldBe false
new Derived() will always be an instance of Base as it inherits from Base, which means:
new Derived().isInstanceOf[Base]
will return true.
But the opposite is false: Base won't be an instance of Derived.
Thus, to check if it's the Base class (and not the Derived class), you can use these two combined conditions:
new Base() should not be a[Derived]
new Base() shouldBe a[Base]
and to check if it's the Derived class, the following is enough:
new Derived() shouldBe a[Derived]
I have play web application with conf/application.conf (nothing unusual). Guice is used for dependency injection. How can property value be injected in class constructor? The code is below.
class MyController #Inject() (private val foo: Foo) extends Controller {
...
}
#ImplementedBy(classOf[FooImpl])
trait Foo {
def bar: String
}
class FooImpl extends Foo {
override val bar = current.configuration.getString("my.bar").get
...
}
In the current configuration FooImpl can't be tested without running application. I want to be able instantiate FooImpl in unit tests. The perfect solution [from my point of view] should look like that:
class FooImpl #Inject() (#Named("my.bar") override val bar: String) extends Foo {
...
}
Unfortunately, this code doesn't work, because Guice doesn't have 'my.bar' binding:
No implementation for java.lang.String annotated with #com.google.inject.name.Named(value=my.bar) was bound.
The only solution that I came up with is writing my own module, which iterates through configuration properties and binds them as named dependencies (a variation of the example from this doc). But I believe that a better approach exists.
I implemented that using Java. I hope you can use it as reference for your Scala implementation.
At first, I created a Module:
public class MainModule extends AbstractModule {
public final static String TIMEOUT_IN_MILLISECONDS_ANNOTATION = "timeout-promise";
private final Configuration configuration;
public MainModule(#SuppressWarnings("unused") Environment environment, Configuration configuration) {
this.configuration = configuration;
}
#Override
protected void configure() {
long timeoutInMilliseconds = configuration.getLong("application.promise.timeout.in.milliseconds", 0L);
bindConstant().annotatedWith(Names.named(TIMEOUT_IN_MILLISECONDS_ANNOTATION)).to(timeoutInMilliseconds);
}
}
After that, I just used the annotation on different places:
class Service {
#Inject
#Named(MainModule.TIMEOUT_IN_MILLISECONDS_ANNOTATION)
protected long timeoutInMilliseconds;
}
Hope this helps.
Some time ago i was developed small guice extention for simple injection configuration variables mapped on Enum
guice-config
I encountered the same problem after about a year, and this time come up with the following solution (very similar to the one proposed by #stranger-in-the-q and #droidman):
class InjectionModule extends AbstractModule {
override def configure(): Unit = {
val config: Config = TypesafeConfigReader.config
config.entrySet().asScala.foreach { entry =>
val path = entry.getKey
entry.getValue.valueType() match {
case ConfigValueType.NUMBER =>
bind(classOf[Int])
.annotatedWith(Names.named(path))
.toInstance(config.getInt(path))
case ConfigValueType.BOOLEAN =>
bind(classOf[Boolean])
.annotatedWith(Names.named(path))
.toInstance(config.getBoolean(path))
case ConfigValueType.STRING =>
bind(classOf[String])
.annotatedWith(Names.named(path))
.toInstance(config.getString(path))
case _ =>
}
}
}
}
Also, this approach can be extended by appending prefixes to system properties (which key-value pairs are part of the loaded config):
private def getPrefix(configValue: ConfigValue): String = {
val description = configValue.origin().description()
if (description.contains("system properties")) {
"sys."
} else {
""
}
}
In this case instead of writing Names.named(path) one should use Names.named(getPrefix(entry.getValue) + path).
To inject multiple properties from a play configuration you could do this way. Create a map out of Play configuration and pass it as Properties to Guice binder.
public class Module extends AbstractModule {
private Environment environment;
private Configuration configuration;
public Module(Environment environment,Configuration configuration){
this.environment = environment;
this.configuration = configuration;
}
#Override
public void configure() {
Configuration helloConf = configuration.getConfig("myconfig");
Map<String, Object> map = helloConf.asMap();
Properties properties = new Properties();
properties.putAll(map);
Names.bindProperties(binder(), properties);
}
}
I wrote a Scala class and defined the main() method in it. It compiled, but when I ran it, I got NoSuchMethodError:main. In all the scala examples, I have seen, the main method is defined in an object. In Java we define the main method in a class. Is it possible to define main() in a Scala class or do we always need an object for this?
To answer your question, have a look on the following :
I made a scala class, compiled and decompiled it, and what I got is interesting.
class MyScalaClass{
def main(args: Array[String]): Unit = {
println("Hello from main of class")
}
}
Compiled from "MyScalaClass.scala"
public class MyScalaClass {
public void main(java.lang.String[]);
public MyScalaClass();
}
So it means that when the scala class is converted to java class then the main method of the scala class which in turn being converted to the main method in java class is not static.
And hence we would not be able to run the program because JVM is not able to find the starting point in the program.
But if the same code is done by using the 'object' keyword then:
Compiling the following:
object MyScalaClass{
def main(args: Array[String]): Unit = {
println("Hello from main of object")
}
}
Decompiling the following:
javap MyScalaClass$.class
Compiled from "MyScalaClass.scala"
public final class MyScalaClass$ {
public static final MyScalaClass$ MODULE$;
public static {};
public void main(java.lang.String[]);
}
Decompiling the following
javap MyScalaClass.class
Compiled from "MyScalaClass.scala"
public final class MyScalaClass {
public static void main(java.lang.String[]);
}
So here we got public static void main in MyScalaClass.class therefore the main method can be executed directly by the JVM here.
I hope you got your answer.
As Eugene said in a comment, there are no static methods in Scala. But watch this:
$ cat Echo.scala
object Echo {
def main( args:Array[String] ):Unit = args foreach println
}
$ scalac Echo.scala
$ javap Echo\$.class
Compiled from "Echo.scala"
public final class Echo$ {
public static final Echo$ MODULE$;
public static {};
public void main(java.lang.String[]);
}
$ javap Echo.class
Compiled from "Echo.scala"
public final class Echo {
public static void main(java.lang.String[]);
}
Note that the classfile for the Echo class (not Echo$, the object) does indeed have a public static void main method. Scala generates static methods for methods defined in objects for compatibility with Java.
However, I consider creating a main method in a Scala program an anachronism. Use the App trait instead; it's cleaner:
object Echo extends App {
args foreach println
}
When I want to test my code in intelligent idea scala editor, I simply create a companion object just below my class and put a main method in it. That's all. see an example:
class Colon {
class Cow {
def ^ (moon:Moon): Unit ={
println("Cow jumped over the moon")
}
}
class Moon{
def ^:(cow:Cow) = println("This cow jumped over moon too")
}
}
object Colon{
def main(args: Array[String]): Unit = {
val c:Colon = new Colon
val cow = new c.Cow
val moon = new c.Moon
cow ^ moon
cow ^: moon
moon.^:(cow)
}
}
I need a smart mechanism for component composition which allows mixed in traits to initialize after the composed component. The following throws a NullPointerException:
class Component {
def addListener(pf: PartialFunction[Any, Unit]) {}
}
trait DynamicComponent {
protected def component: Component
component.addListener {
case x =>
}
}
class Foo extends DynamicComponent {
protected val component = new Component
}
new Foo // -> NullPointerException
The following things are not options for me:
Using protected lazy val component; that would produce an avalange of dozens of vals needing to become lazy, something I do not want.
Putting addListener in a method, e.g. initDynamic(); because I will be mixing in many traits, and I don't want to remember to call half a dozen initFoo() methods.
Using DelayedInit. This doesn't work with traits, at least according to the scaladocs.
I could live with a single init() call, but only under the following conditions:
all mixed in traits can easily declare to be invoked in this one single call
it is a compile error to forget the init() statement.
You can delay the initialization of a trait by by using early definitions. (See section 5.1.6 of the scala language specification)
class Foo extends {
protected val component = new Component
} with DynamicComponent
It's even clunkier than your solution, but you can always require the creation of a val that must be set with the init() method. You could choose to not do it last and get an error at runtime, but at least you won't forget it entirely:
class Component {
def addListener(pf: PartialFunction[Any, Unit]) {
println("Added")
}
}
trait Dyn {
protected def component: Component
protected val initialized: Init
class Init private () {}
private object Init { def apply() = new Init() }
def init() = { component.addListener{ case x => }; Init() }
}
class Foo extends Dyn {
protected val component = new Component
protected val initialized = init()
}
No cheating!:
> class Bar extends Dyn { protected val component = new Component }
<console>:12: error: class Bar needs to be abstract, since value
initialized in trait Dyn of type Bar.this.Init is not defined
class Bar extends Dyn { protected val component = new Component }
The advantage this has is if you need multiple things to be in place before you initialize all of them cooperatively, or if your Component class is final so you can't mix in anything else.
AN idea could be to use the trick described here:
Cake pattern: how to get all objects of type UserService provided by components
All your components that should be initialized could be registered in some Seq[InitializableComponent]. And then you could initialize all registered components with a foreach.
No component will be forgotten in that Seq because they are registered automatically, but you can still forget to call the foreach anyway...
Here is one idea (I am happy to read about other suggestions):
class Component {
def addListener(pf: PartialFunction[Any, Unit]) {
println("Added")
}
}
trait DynamicComponentHost {
protected def component: Component with DynamicPeer
protected trait DynamicPeer {
_: Component =>
addListener {
case x =>
}
}
}
class Foo extends DynamicComponentHost {
protected val component = new Component with DynamicPeer
}
new Foo
So basically I am forcing the component to mix in a type that can only be provided by the mixed in trait. Reasonable? Looks a bit too complicated in my eyes.
I created a hierarchy of case objects in Scala that looks like the following:
package my.awesome.package
sealed abstract class PresetShapeType(val displayName: String)
case object AccelerationSensor extends PresetShapeType("Acceleration Sensor")
case object DisplacementSensor extends PresetShapeType("Displacement Sensor")
case object ForceSensor extends PresetShapeType("Force Sensor")
case object PressureSensor extends PresetShapeType("Pressure Sensor")
case object StrainSensor extends PresetShapeType("Strain Sensor")
I also have a piece of Java code in which I'd like to access PressureSensor, but the following does not work:
package my.awesome.package.subpackage;
import my.awesome.package.PressureSensor;
// Do some stuff, then...
DVShape newshape = DVShapeFactory.createPresetShape(PressureSensor, new Point3f(0,0,0));
So, how do I reference the PressureSensor case object from Java? I decompiled the byte code for both the PressureSensor and PressureSensor$ classes, which yielded the following:
Compiled from "DVShapeFactory.scala"
public final class org.nees.rpi.vis.PressureSensor extends java.lang.Object{
public static final java.lang.Object productElement(int);
public static final int productArity();
public static final java.lang.String productPrefix();
public static final int $tag();
public static final java.lang.String displayName();
}
Compiled from "DVShapeFactory.scala"
public final class org.nees.rpi.vis.PressureSensor$ extends org.nees.rpi.vis.PresetShapeType implements scala.ScalaObject,scala.Product,java.io.Serializable{
public static final org.nees.rpi.vis.PressureSensor$ MODULE$;
public static {};
public org.nees.rpi.vis.PressureSensor$();
public java.lang.Object readResolve();
public java.lang.Object productElement(int);
public int productArity();
public java.lang.String productPrefix();
public final java.lang.String toString();
public int $tag();
}
But that didn't yield any great insight.
from Java, say:
my.awesome.package.PressureSensor$.MODULE$
PressureSensor$.MODULE$ should give you the instance of the case object.
This is still a hack, but in my opinion a bit more readable in Java. Just add a method to explicitly return the reference to the singleton instance (it shows up as a static method on the class):
sealed abstract class PresetShapeType(val displayName: String)
case object AccelerationSensor extends PresetShapeType("Acceleration Sensor") { def instance = this }
case object DisplacementSensor extends PresetShapeType("Displacement Sensor") { def instance = this }
case object ForceSensor extends PresetShapeType("Force Sensor") { def instance = this }
case object PressureSensor extends PresetShapeType("Pressure Sensor") { def instance = this }
case object StrainSensor extends PresetShapeType("Strain Sensor") { def instance = this }
And then in Java:
import my.awesome.package.PressureSensor;
DVShape newshape = DVShapeFactory.createPresetShape(PressureSensor.instance(), new Point3f(0,0,0));