Why is overriding in JAVA working somewhat differently that from C++? - java

I have some background in C++ and know some Java too (apparently far from enough).
When I see overriding behavior in Java or C++, it does not seem to differ much. Given below example in JAVA:
class Animal{
public void move(){
System.out.println("Animals can move");
}
}
class Dog extends Animal{
public void move(){
System.out.println("Dogs can walk and run");
}
}
public class TestDog{
public static void main(String args[]){
Animal a = new Animal(); // Animal reference and object
Animal b = new Dog(); // Animal reference but Dog object
a.move();// runs the method in Animal class
b.move();//Runs the method in Dog class
}
}
In Java, you use a base class reference, in C++ you use a base class pointer, and depend on the type of instance it points to (a base class object instance or a subclass instance), you can achieve polymorphism.
The above is based on you call the instance method using a base class reference or pointer, right?
Now I see this example in Java.
What is the order of the Constructors in this Java Code?
Basically it says if a base class function is overriden, then in the process of creating a subclass object, even the base class initialization portion will be affected. See below explanation I copied from above link:
new Son()
=>
Son._init
=> first every constructor calls super()
Father._init
Object._init
who() => is overridden, so prints "son" !!!!!
tell(name) => name is private, so cannot be overridden => "father"
who() => "son"
tell(name) => "son"
Why should this happen? I mean does this conform to how polymorphism should be use? When making base class portion of initialization, why use overrided function from subclass?
In Java doc http://docs.oracle.com/javase/specs/jls/se7/html/jls-12.html#jls-12.5, I only find this:
"Unlike C++, the Java programming language does not specify altered rules for method dispatch during the creation of a new class instance. If methods are invoked that are overridden in subclasses in the object being initialized, then these overriding methods are used, even before the new object is completely initialized. "
But I don't know the reason behind it, and it feels strange.
Any thoughts?

This is one of the extremely rare cases where C++ is trying to protect you from shooting yourself in the foot more than Java does. (Or at least it has the noble intention to do so.)
You are very likely to shoot yourself in the foot in any language if you try invoking an overridable (virtual) method M of your base class B from within the constructor of B. That's because M is likely to be overridden in derived class D, but at the moment that B is being constructed, D has not been constructed yet. So, D.M is being invoked before the constructor of D has been invoked. That can spell disaster.
So, Java simply allows this to happen, use at your own risk. (If you have sufficient warnings enabled, your compiler will tell you that you are living life dangerously.)
C++ does not prohibit this either, but it slightly changes its behaviour so as to contain the damage, so to speak: when you invoke a virtual method from within a constructor, it does not really invoke it as a virtual method, (with a VMT lookup,) but it invokes it directly, as a non-virtual method.
(Either that, or from within the constructor of B it simply uses class B's VMT instead of D's VMT. Which, come to think of it now, makes sense. But I am not sure about that, it has been a long time since I last troubleshot this behaviour of C++.)

Related

Overriding, redeclaring, hiding and redefining of methods in classes, interfaces and classes implementing interfaces [duplicate]

This question already has answers here:
Can I override and overload static methods in Java?
(19 answers)
Closed 5 months ago.
Would it be correct to say that static methods in a class extended with a class, in an interface extended with an interface and in an interface implemented by a class can only be redeclared in the extending/implementing part (hiding the original method in the class extending class case), and any other method can only be overridden there? And is redefining the same as overriding?
If not, can someone please explain these concepts with a flowchart (I've looked at other explanations and they don't give me the overview I'm looking for)?
Would it be correct to say that static methods in a class extended
No. static methods fundamentally are un-inheritable, and un-overridable. Or rather, conceptually it just doesn't apply.
Saying: "This static method is an override" is a bit like stating: "This smells blue" - it's not clear what that would even mean.
overriding is relevant essentially only for the notion of dynamic dispatch. Dynamic dispatch is this idea:
class Dog {
void bark() { System.out.println("Woof"); }
}
class Bulldog extends Dog {
void bark() { System.out.println("Grrrr"); }
}
Dog d = new Bulldog();
d.bark(); // prints "Grrrr"
That's dynamic dispatch at work. d is a variable. Like all non-primitives in java, it is a reference. Meaning, values stored in this variable are the reference - a pointer. An address in an addressbook that lets you get to a house, not the house itself. d's type is Dog. Its value is a reference to an actual instance of some object (and java guarantees that whatever the actual type it is an instance of, that type is Dog or a subtype of Dog). That reference gets you an instance of Bulldog. So, when invoking bark() here, what happens?
Dynamic dispatch: Java finds the most specific implementation of this method, and calls that. So that prints Grrrr, and not Woof.
However, when talking about static methods, the entire concept doesn't apply. Dynamic dispatch occurs because there is a discrepancy between the type of the reference (which is Dog here, that's the type of variable d) and the type of the thing the reference is pointing at (which is Bulldog). Overriding as a concept exists because these 2 types may not be the same.
When invoking static methods, this doesn't come up. You always write SomeType.someStaticMethod(), so what would dynamic dispatch even be?
SIDENOTE: You can legally invoke static methods using an expression. This is legal:
List.of("a", "b"); // this is how you normally do it
List<Integer> list = new ArrayList<Integer>();
list.of("a", "b"); // this compiles and works
But make no mistake - the compiler treats it the same and cares only about the type of list, not about the object the variable is pointing at. In fact, list = null; list.of("a", "b"); works fine, no NullPointerException, proving the point that list doesn't get dereferenced when you do this. All style guides strongly recommend not doing this, for good reason.
SIDENOTE 2: Unfortunately java lets you declare a static method as final which is weird: final means: Cannot be overridden, and we just figured out that 'override' conceptually doesn't make sense for static methods. It's just what it is - it's a design error in the original java spec and java does not like making changes that break existing code unless there is an incredibly good reason to do so, and whilst this spec brainfart is annoying, it doesn't hurt much. Just don't declare your static methods final. It even adds a rule (subtypes cannot declare a static method with the same signature) which is even more bizarre. Again, disregard that - java spec error that doesn't get fixed because not worth the headache.
And is redefining the same as overriding?
Yes. 2 words for the same thing. The correct term is 'override' - both the java lang spec uses this term, and the vast majority of the community does. If I was an editor of whatever tutorial, blogpost, documentation, or presentation you got this 'redefine' term from, I'd edit it. Unless the author was specifically talking about this:
class Parent {
static void foo() {}
}
class Child extends Parent {
static void foo() {}
}
In which case 'override' is the wrong term to use (given that static methods and override are orthogonal concepts), though I wouldn't use 'redefine' either. You're just defining a method named foo in one class, and also defining a method named foo in another. The fact that one is a child of the other is irrelevant; both have a foo method now. The foo method in child does not override or redefine anything; there's no such thing in static world.
with a flowchart
A flowchart requires some sort of flow to chart. Hence the name. No such thing here.

Subclass object with superclass funcionality

I'm confused why this code works.
The object "person" is declared as an instance of B with only the functionality of A, yet it somehow prints out hello twice. If the object only gets access to the methods in A, how can it end up accessing the print out statement?
abstract class A {
abstract void move();
void go() {
this.move();
}
}
class B extends A {
void move() {
System.out.println("hello");
}
}
public class Main {
public static void main(String[] args){
A person = new B();
person.move();
person.go();
}
No, you have a "link" of class A, but functionality depends on object (in your code it is new B();), so your A link points to a B object with B functionality.
The object "person" is declared as an instance of B with only the functionality of A
Not exactly. Java is a reference-based language; here you have made an instance (new B()), and have exactly one reference to it, called person.
You could have no references to it (leading to the created object being garbage collected eventually). You can have 2000 references to it. You could have B b = new B(); A a = b;, and now you have 2 references to it, one of type A and one of type B. But they are pointing to the exact same object.
Thus, person isn't an object, it's a reference. And that reference has type A, sure, but the object it is referencing is just a B, it isn't 'restricted' to have only A functionality.
You probably know all this, but terminology is important here, as it seems to have led to some confusion. Restating your question with less ambiguous terminology:
The reference "person" currently references an instance of B, but only exposes functionality that A has.
Yes. And A's functionality is described solely by the signatures present there, not by the code. A's functionality involves having a go() method, which takes no arguments and returns nothing, as well as a move() method with the same rules. And that is all - 'invoking go will actually run the move method', or 'move has no implementation' is not part of that.
Thus, both move() and go() are part of the functionality exposed by A.
The implementation of the functionality has nothing whatsoever to do with the reference type (A person) and everything to do with what the reference is actually pointing at (new B()). B's implementation of the move() and go() methods specified by A are such that move prints "hello", and go() calls move (thus, prints hello()) - that implementation is inherited from A, but B was free to change that; B however decided not to.
Said more technically:
Java uses something called dynamic dispatch. What that means is that at compile time (i.e. write time), java figures out if a method you're calling even exists and which variant (if you have move(String a) and move(int a), that's two methods with the same name, at write time java decides which one you are attempting to invoke), and uses the type of the expression in front of the dot to figure that out. But then at runtime, the actual type of the object that 'the expression in front of the dot' is actually pointing at is used to figure out which actual code to invoke. This always happens and you can't opt out (you can't choose to run A's implementation when you invoke main(). Only B's implementation can choose to not override A's implementation, or to explicitly invoke its supertype's implementation. A user of B's code cannot do that).
Note that static stuff doesn't 'do' inheritance at all, and therefore, dynamic dispatch doesn't apply there.

Why can this object access methods of it's parent class?

This question is taken from an AP Computer Science practice test.
public class Bird
{
public void act()
{
System.out.print("fly");
makeNoise();
}
public void makeNoise()
{
System.out.print("chirp");
}
}
public class Dove extends Bird
{
public void act()
{
super.act();
System.out.print("waddle");
}
public void makeNoise()
{
super.makeNoise();
System.out.print("coo");
}
}
Suppose the following declaration appears in a class other than Bird or Dove:
Bird pigeon = new Dove();
What is printed as a result of the call pigeon.act()?
I thought the answer would be "fly chirp", but the textbook says that the answer is "fly chirp coo waddle". I thought that 'pigeon' could only access methods available in Bird? I was under the impression that, if the user wanted to access methods in Dove, 'pigeon' would have to be cast to Dove.
Would Bird pigeon = new Bird(); give the same output? How about Dove pigeon = new Dove();?
Long story short, when you access act method of pigeon, its override from Dove is called.
I thought that 'pigeon' could only access methods available in Bird?
That is certainly true, at least, for situations when no casting is applied. However, method act is available on the class Bird, the statically known type of pigeon, so the call compiles fine.
However, accessing methods is only about being able to call them. What methods do when you call them is decided at runtime based on the dynamic type of pigeon. This is where method overriding comes into play, because Dove overrides Bird's methods, but in addition it also calls Bird's methods. That is why the code hits all four printouts.
Would Bird pigeon = new Bird(); give the same output?
No, the output would be different, because in this case both dynamic and static types of pigeon would be the same, i.e. Bird, so only Bird's implementations would be invoked.
the class Dove does override the methods act and makeNoise of class Bird. Overriding means changing the behavior of a method visible to the sub-class, (in your case Dove). Methods are visible to a sub-class if they have a public or protected access modifier or if they have a package private access modifier and the sub-class belongs to the same package as the super-class does.
pigeon is an instance of Dove.
calling pigeon.act() results in calling Dove.act.
Dove.act calls super.act which is Bird.act.
Bird.act prints fly and calls makeNoise on pigeon resulting in calling Dove.makeNoise.
Dove.makeNoise calls super.makeNoise which is Bird.makeNoise.
Bird.makeNoise print chirp.
Dove.makeNoise prints coo after calling super.makeNoice
What you experience here is polymorphism in action. And instead of answering your various questions directly; I will simply explain the case you are observing.
You call act() on an instance of Dove; that causes a call to super; printing "fly".
That super method then calls makeNoise() ... on "itself". But as said: "itself" is a Dove object; thus you get the Dove noise! "coo"!
Then the Dove implementation ends; and prints "waddle".
The essence is: the exact version of a method that is invoked is determined at runtime and it only depends on the exact type of the object the method is invoked on.
The above gives you all the information you need to answer your other questions yourself. In that sense: don't request answers; ask for explanations; and use those to solve the puzzle yourself!
From your question "I thought that 'pigeon' could only access methods available in Bird? I was under the impression that, if the user wanted to access methods in Dove, 'pigeon' would have to be cast to Dove." This is actually true.
Lets try to find the mssing link in the understanding.
When we have code like Bird pigeon = new Dove(); where Dove extends Bird we have actual object of Dove and reference type is of Bird. As the object is of Dove so it has the methods, both inherited from super class as well as the ones which are added.
Another important point is all the overriden methods have only one instance. Its Overriden meaning the behavior of the same method has been modified, its not an additional separate method. There is only one copy of inherited method not both. Its the overloaded methods which are separate, just the names are same but signature is different. This is the reason you get the behaviour of Dove when you invoke any overriden method.
This one is simple super , using it a sub class can access the accessible (visible) entities (instance properties and methods) of its super class. If a sub class uses super keyword to invoke a method then its the method of the parent class which gets invoked. But again this can be considered that the author of the sub class did it intentionally. Once the class is written and Objects of such class is created then on the object using . (dot operator) users can only invoke whats there in the object. If any method has used super keyword its part of the behavior of the object. Users of the Sub class object can not invoke behavior of the parent class method if its overridden in sub class.
Lastly yes if you wish to invoke any additional method of Dove (Sub class ) using a reference of Bird (Super class) then you need to cast it to Dove.

Use subclass methods from another subclass, lateral casting?

In Java, I was given an assignment. I could describe my issue with it generally as follows:
Create three classes A B and C.
Class A has instance variables j,k,l, setter and getter methods, and also method x which makes a calculation based on them.
Class B is a subclass of A and overrides method x with the same empty signature, using j,k,l.
Class C is also a subclass of A, and has an additional class y which does other junk.
Now, instantiate a class C object, set its variables, execute method x and then use the overriden version of method x from class B.
My question is, how do I do that last part? I think either the question is incorrect or else I am interpreting it wrong. Can I cast my class C object to a class B object and then use the class B version of x()? I don't know much about casting objects and what determines whether it is possible to cast from one thing to another. Upcasting could be just as lossy as "horizontal/lateral" casting but I have never heard of the latter. Is it impossible to use the method in class B from class C without making class C internally instantiate and rebuild a class B object from its own variables? Am I making sense?
Addendum:
Okay. So siblings cannot be cast as one another? Because C and B are both "part of [my] object hierarchy". Could I cast up and then back down to a different subclass? Like casting a cat to an animal class and then an animal to a dog? Because method x relies on instance variables in both class A AND class B, creating a new instance of B within C will create an object with fields/variables which are not set the same, they will all be zero or similar. So I will have to essentially have to copy over each value of j,k and l into the new B type object within the C class object. This seems like a waste of time and memory, copying the object variables into another very similar object. But there is no other way?
Second Addendum:
Although I have selected an answer, I am still curious about casting in general and what I call lateral-casting. If B and C are both sub-classes of A, can you allow casting from B to C? Also, how does one allow casting in general?
Third:
So at this point, a couple years later I understand that this is not really possible in Java. However, interestingly in JavaScript and PHP you can use the "bind" method of a function to change it's context so that it essentially thinks it's another object. So in this scenario I could create a copy of B:x() bound to an instance of C containing the properties I have set, the bound instance of x would then use all of the internal variable of the instance of C. It's really weird but really handy sometimes.
Here is an example of using a method from one subclass on a different subclass; making a cat speak using a dog's speak function:
class Animal {
constructor(){
}
}
class Dog extends Animal{
constructor(){
super()
this.sound = 'bark'
}
speak(){
console.log(this.sound)
}
}
class Cat extends Animal{
constructor(){
super()
this.sound = 'meow'
}
}
var mycat = new Cat()
var mydog = new Dog()
mydog.speak()
mydog.speak.bind(mycat)()
So it's basically as if cat has the speak function. This is the sort of thing I wanted to do in Java.
Casting means you know better than the compiler what your object is. If you cast to something that isn't part of your object hierarchy you will get an exception (specifically a ClassCastException) when you run the program.
The most common place you see casting is in the Object#equals method. equals needs to take a java.lang.Object as its only argument. If you look at how people implement equals, you'll see testing for what the class of the passed-in object is, and if it's what's expected then there's a cast so that the checks that follow can use that object's fields. Otherwise casting doesn't come up in beginner situations much.
Casting does not change what your object is, it only tries to inform the compiler what the object is in circumstances (like equals) where the type system is not sufficient. If you are creating objects with virtual methods in a hierarchy, casting does not change which version of the method gets called.
If C is not a subclass of B then you can create an object of class B within C, as an instance member or as a local variable of some method of B, and call the method of C on that, something like:
class C extends A {
private B b; // instance member
public C(B b) {
this.b = b;
}
public void x() {
// do stuff
b.x();
}
}

Runtime polymorphism

Suppose I have class A
public class A
{
public void method()
{
//do stuff
}
}
Also another class B
public class B extends A
{
public void method()
{
//do other stuff
}
}
Now I have the following statements:
A a = new B();
a.method();
Is this an example of run time polymorphism? If yes, then is no binding is done for the reference a at compile time?
The compiler will tell you that this can't work, because there's no relationship between A and B that will allow you to write A a = new B();
B either has to extend A or both have to implement a common interface with void method() in it.
You could have answered this very quickly by trying it with the compiler. Be an experimentalist - it's faster than forums.
UPDATE:
It works, now that B extends A. The binding that you care about, dynamic binding, is done at runtime. The static compile time type for the variable "a" is class A; at runtime it is dynamically bound to a reference of type B. Yes, I would consider this to be an example of polymorphism.
I was waiting to put my answer to another post, into a more correctly classified problem. Your question fits the concept, I've tried to explain in detail. but, since I don't know how link answers, I'll just copy/paste the complete script here .... Kindly go through it, you'll understand everything that there is to understand about your problem, you've asked
let's say compiler has to resolve a call like this : *
A a = new AA(); // Assume AA to be some subclass of class A
a->someFunc(); // and we're invoking a method "someFunc()' on a
*.
Now, compiler will go over following steps methodically.
1.) Firstly, compiler knows the declared type of variable a, so it will check whether the declared type of object a (lets call this, class A for time being), have a method with the name someFunc() and that it must be public. This method could either be declared in class A, or it could be a derived method from one of the base class(es) of class A, but it doesn't matter to compiler and it just checks for it's existence with it's access specifier being public.
Needless to say any error in this step will invite a compiler error.
2.) Secondly, once the method is validated to be a part of the class A, compiler has to resolve the call to the correct method, since many methods could be there with same name (thanks to function overloading). This process of resolving correct method is called overloading resolution. Compiler achieves this by matching the signatures of the called method with all the overloaded methods that are part of the class. So, of all the someFunc() s only the correct someFunc() (matching the signatures with called method) will be found and considered further.
3.) Now comes the difficult part, It may very well happen that someFunc() may have been overridden in one of the subclasses of the class A (lets call this class AA and needless to say it is some subclass of A), and that variable a (declared to be of type A) may actually be referring to an object of class AA, (this is permissible in C++ to enable polymorphism). Now, if the someFunc() method is declared to be of type virtual, in base class (i.e. Class A) and someFunc() has been overriden by subclass(es) of A (either in AA or classes between A and AA), the correct version of someFunc() have to be found out by the compiler.
Now, imagine you're the compiler and you have this task of finding whether the class AA has this method. Obviously, class AA will have this method since, it is a subclass of A and public access of A in class A has already been validated in Step 1 by the compiler !!! . But Alternatively, as mentioned in previous paragraph, someFunc() may be overriden by class AA (or any other class between A and AA), which is what compiler needs to catch. Therefore, you (since, your'e playing the compiler) could do a systematic check to find the bottom-most (lowest in the inheritance tree) overridden method someFunc() starting from class A and ending at class AA. In this search, you'll look for same method signatures as was validated in overloading resolution. This method will be the method which will be invoked.
Now, you may be wondering, "What the heck", is this search done everytime ? ... Well, Not really. The compiler knows the overhead of finding this everytime, and as such, maintains a data-structure called Virtual Table for every class type. Think of virtual table, as mapping from method signatures (which are publicly accessible) to the function pointers. This virtual table is made by compiler during compilation process and is maintained in-memory during program execution. In our example, class A and class AA will both have their own virtual tables. And when compiler has to be find someFunc() in class AA (since actual object pointed by variable a is of type AA), it will simply find the function pointer through the virtual table of Class AA. This is as simple has hashing into the table and is a constant time operation.
Regards
AViD
The code example you've given isn't legal, so I guess the answer is that it's not a kind of polymorphism.
You can't assign a class to a variable of an unrelated type like that. I'm guessing that you might be intending for B to derive from A?
Is this an example of run time polymorphism?
As of the edit: yes that would be runtime polymorphism, since which method is actually executed depends on what is assigned to a.
Edit:
If method() would be static, then calling a.method() would always result in the version of A being called. However, you'd then just write A.method() and if you don't do that any decent IDE would warn you about that (because it might be misleading to call a static method on an instance).
End Edit.
Compile time polymorphism would be the overloading mechanism, i.e. the compiler decides whether to use someMethod(Number) or someMethod(Integer) depending on what it knows about the parameter that is passed (if you pass a Double or just a Number it would be the first one, if you pass an Integer it would be the second).
If yes, then is no binding is done for the reference a at compile time?
What do you mean exactly? Note that the type of a will be A so you can assign everything that extends A at runtime. The compiler will just complain if the assigned value is not an A (you might trick the compiler with a cast, but this is just evil and if this breaks your program, you have been warned ;) )
It should be
public class B extends A {
public void method()
{
//do other stuff
} }
If it were compiling, it would be the B::method() that would be called. (only if you changed your code to make the B class inherit the A class).
In this case your use your object like it was only an A, but if some methods are overidden in B, then these methods are called, instead of A's methods.

Categories

Resources