The pages of this web site are not formatted to conserve paper, but my new book (ISBN: 0-9745862-0-X) is formatted to save paper, save your printer cartridge, save a loose-leaf binder, and save money. If you prefer to work my exams from printed pages, then give your printer a rest and buy my new book.
Today, you can find my book at amazon.com.
| No. | Answer | Remark | |
|---|---|---|---|
| 1 | b | Prints: ABC | The method invocation expression d1.m1(a1) uses reference d1 of type D to invoke method m1. Since the reference d1 is of type D, the class D is searched for an applicable implementation of m1. The methods inherited from the superclasses, C, B and A, are included in the search. The argument, a1, is a variable declared with the type A; so method A.m1(A a) is invoked. |
| 2 | a | Prints: float,float | A method invocation conversion can widen an argument of type float to match a method parameter of type double, so any argument that can be passed to m(float i) can also be passed to m(double i) without generating a compile-time type error. For that reason, we can say that m(float i) is more specific than m(double i). Since both methods are applicable, the more specific of the two, m(float i), is chosen over the less specific, m(double i). The arguments of the method invocation expressions, m(a1) and m(b1), are of types int and long respectively. A method invocation conversion can widen an argument of type int or long to match either of the two method parameter types float or double; so both methods, m(float i) and m(double i), are applicable to the two method invocation expressions. Since both methods are applicable, the more specific of the two, m(float i) is chosen rather than the less specific, m(double i). |
| 3 | b | Prints: ABC | Three methods overload the method name m1. Each has a single parameter of type A or B or C. For any method invocation expression of the form m1(referenceArgument), the method is selected based on the declared type of the variable referenceArgument--not the run-time type of the referenced object. The method invocation expression d1.m1(c1) uses reference d1 of type D to invoke method m1 on an instance of type D. The argument, c1, is a reference of type A and the run-time type of the referenced object is C. The argument type is determined by the declared type of the reference variable c1--not the run-time type of the object referenced by c1. The declared type of c1 is type A; so the method m1(A a) is selected. The declared type of c2 is type B; so the method invocation expression d1.m1(c2) invokes method m1(B b). The declared type of c3 is type C; so the method invocation expression d1.m1(c3) invokes method m1(C c). |
| 4 | a | Prints: float,float | A method invocation conversion can widen an argument of type float to match a method parameter of type double, so any argument that can be passed to m(float i) without generating a compile-time type error can also be passed to m(double i). For that reason, we can say that m(float i) is more specific than m(double i). The arguments of the method invocation expressions, m(a1) and m(b1), are of types char and long respectively. A method invocation conversion can widen an argument of type char or long to match either of the two method parameter types float or double; so both methods, m(float i) and m(double i), are applicable to the two method invocation expressions. Since both methods are applicable, the more specific of the two, m(float i) is chosen rather than the less specific, m(double i). |
| 5 | b | Prints: ABC | Three methods overload the method name m1. Each has a single parameter of type A or B or C. For any method invocation expression of the form m1(referenceArgument), the method is selected based on the declared type of the variable referenceArgument--not the run-time type of the referenced object. The method invocation expression c4.m1(c1) uses reference c4 of type C to invoke method m1 on an instance of type C. The argument, c1, is a reference of type A and the run-time type of the referenced object is C. The argument type is determined by the declared type of the reference variable c1--not the run-time type of the object referenced by c1. The declared type of c1 is type A; so the method A.m1(A a) is selected. The declared type of c2 is type B; so the method invocation expression c4.m1(c2) invokes method B.m1(B b). The declared type of c3 is type C; so the method invocation expression c4.m1(c3) invokes method C.m1(C c). |
| 6 | e | Compile-time error | The method invocation expression, m(b1), contains an argument of type double. A method invocation conversion will not implicitly narrow the argument to match the parameter type of the method, m(float i). The method invocation expression, m(a1), contains an argument of type long. A method invocation conversion will widen the argument to match the parameter type of the the method, m(float i). |
| 7 | a | Prints: A | The reference c1 is of the superclass type, A; so it can be used to invoke only the method m1 declared in class A. The methods that overload the method name m1 in the subclasses, B and C, can not be invoked using the reference c1. A method invocation conversion promotes the argument referenced by c2 from type C to type A, and the method declared in class A is executed. Class A declares only one method, m1. The single parameter is of type A. Class B inherits the method declared in class A and overloads the method name with a new method that has a single parameter of type B. Both methods sharing the overloaded name, m1, can be invoked using a reference of type B; however, a reference of type A can be used to invoke only the method declared in class A. Class C inherits the methods declared in classes A and B and overloads the method name with a new method that has a single parameter of type C. All three methods sharing the overloaded name, m1, can be invoked using a reference of type C; however, a reference of type B can be used to invoke only the method declared in class B and the method declared in the superclass A. The method invocation expression c1.m1(c2) uses reference c1 of type A to invoke method m1. Since the reference c1 is of type A, the search for an applicable implementation of m1 is limited to class A. The subclasses, B and C, will not be searched; so the overloading methods declared in the subclasses can not be invoked using a reference of the superclass type. |
| 8 | b | Prints: String | A method invocation conversion can widen an argument of type String to match a method parameter of type Object, so any argument that can be passed to m(String x) without generating a compile-time type error can also be passed to m(Object x). For that reason, we can say that m(String x) is more specific than m(Object x). The argument of the method invocation expression, m(null), is of type null and can be converted to either type String or Object by method invocation conversion, so both methods, m(String x) and m(Object x), are applicable. The more specific of the two, m(String x), is chosen over the less specific, m(Object x). |
| 9 | a | Prints: AAA | The declared type of the reference variables, a1, b1 and c1, is the superclass type, A; so the three reference variables can be used to invoke only the method m1(A a) that is declared in the superclass, A. The methods that overload the method name m1 in the subclasses, B and C, can not be invoked using a reference variable of the superclass type, A. A method invocation conversion promotes the argument referenced by c4 from type C to type A, and the method declared in class A is executed. |
| 10 | d | Compile-time error | The type of the argument is null and could be converted to any of the types Object, String or GFC200, by method invocation conversion. All three methods are applicable, but none of the three is more specific than both of the other two. The ambiguity results in a compile-time type error. If type GFC200 were a subclass of type String; then any argument that could be pass to m(GFC200 x) could also be passed to m(String x) without causing a compile-time type error, and we could say that m(GFC200 x) is more specific than m(String x). Since GFC200 is not a subclass of type String, a method invocation conversion is not able to widen an argument of type GFC200 to match a method parameter of type String, so m(GFC200 x) is not more specific than m(String x). |
| 11 | a | Prints: AAA | The reference c2 is of the superclass type, A; so it can be used to invoke only the method, m1, declared in class A. The methods that overload the method name m1 in the subclasses, B and C, can not be invoked using the reference c2. |
| 12 | b | Prints: GFC203 | The type of the argument is null and could be converted to either type GFC202 or GFC203 by method invocation conversion; so both methods are applicable. The more specific of the two, m(GFC203 x), is chosen. Type GFC203 is a subclass of type GFC202; so any argument that can be passed to m(GFC203 x) can also be passed to method m(GFC202 x) without causing a compile-time type error; therefore, we can say that method m(GFC203 x) is more specific than m(GFC202 x). |
| 13 | d | Compile-time error | The declarations of b1 and c1 cause compile-time errors, because a reference of a subclass type can not refer to an instance of the superclass type. |
| 14 | d | Prints: GFC206,GFC206 | Type GFC207 is a subclass of types GFC206 and GFC205, so any of the four methods are applicable to the method invocation expression, m(gfc207, gfc207). The most specific of the four, m(GFC206 x, GFC206 y), is chosen. Type GFC206 is a subclass of type GFC205, and method m(GFC206 x, GFC206 y) is more specific than the other three, because any invocation of m(GFC206 x, GFC206 y) could also be handled by any of the other three without causing a compile-time type error. |
| 15 | d | Compile-time error | The declaration of c2 causes a compile-time error, because a reference of a subclass type can not refer to an instance of the superclass class. |
| 16 | f | Compile-time error | The method invocation expression, m(gfc213, gfc213), is ambiguous; because, no applicable method is more specific than all of the others. Method m(GFC212 x, GFC212 y) is more specific than m(GFC212 x, GFC211 y), because any invocation of m(GFC212 x, GFC212 y) could also be handled by m(GFC212 x, GFC211 y) without causing a compile-time type error. However, some invocations of m(GFC212 x, GFC212 y) can not be handled by m(GFC211 x, GFC213 y), so m(GFC212 x, GFC212 y) is not more specific than m(GFC211 x, GFC213 y). Furthermore, not all invocations of m(GFC211 x, GFC213 y) could be handled by m(GFC212 x, GFC212 y). |
| 17 | b | Prints: byte byte boolean | Variable b1 was initialized by the first method invocation statement, so the second method invocation statement does not result in a compile-time error. The assignment expression, b1 = 1, initializes variable b1 with the value 1, and the same value is passed as an argument to method m1(byte b1). The method invocation expression, m1(b1), invokes the same method, m1(byte b1). The argument of the third method invocation expression, m1(b1 == 1), is the result of the equality expression, b1 == 1. The type of the result and the argument is boolean, so the invoked method is m1(boolean b1). |
| 18 | b | Prints: BBABBA | Type B is a subclass of type A, and method m(B x, B y) is more specific than the other three; because any invocation of it could be handled by any of the other three without causing a compile-time type error. All four methods are applicable to the first method invocation expression, m(null,null). The most specific method, m(B x, B y), is chosen; and both arguments are converted to type B. In the second method invocation expression, m(a1=null,b1=null), simple assignment expressions initialize the local variables a1 and b1 with null references of types A and B respectively. The invoked method is m(A x, B y). In the third method invocation expression, the positions of the arguments are reversed relative to the previous invocation: The type of the first argument is now B and the second is A. The invoked method is m(B x, A y). |
| 19 | c | Prints: m3Cm2Bm1A | The method invocation expression, m3(new C()), invokes method m3(C x), because the argument type matches the parameter type of the method declaration exactly. Method m3 uses the parameter as the argument of the next invocation expression, m2(x). Of the two overloaded versions of m2, the most specific is invoked, m2(B x). Type B is a subclass of A, so any invocation of m2(B x) could be handled by m2(A x) without causing a compile-time type error. For that reason, m2(B x) is more specific than m2(A x). |