Comparing Generic Method with Non-Generic Method

Java Tutorials - Herong's Tutorial Examples - Version 7.03, by Dr. Herong Yang

Java Tutorials - Herong's Tutorial Examples

∟Comparing Generic Method with Non-Generic Method

This section provides a tutorial example of comparing a generic method and a non-generic method for the same requirement. The generic method allows the compiler to catch more type compatibility mistakes.

To see advantages of using generic methods, let's try to develop a non-generic method and a generic method for the same requirement and compare them.

Assuming that we have a requirement to develop a static method that compares two objects and returns the object that has the higher value, and this method should support multiple types of objects like String, Number, Date, Time, etc., here is the best I can write as a non-generic version of the method:

public static Comparable maxNonGeneric(Comparable a, Comparable b) {}

This definition is very weak in terms of type requirements in the method signature. When programmers try to invoke this non-generic method, they have chances of making mistakes as listed below. The compiler will not be to catch those mistakes. The resulting bytecode will be more likely running into runtime exceptions:

String max = (String) maxNonGeneric(
   new String("1234"), new String("789")); // Correct invokation

String max = (String) maxNonGeneric(
   new String("1234"), new Integer("789")); // Run-time exception

Integer max = (Integer) maxNonGeneric(
   new String("1234"), new String("789")); // Run-time exception

Here is my generic version of the method for the same requirement:

public static <T extends Comparable<T>> T maxGeneric(T a, T b) {}

This definition is much stronger now in terms of type requirements in the method signature:

The type parameter "T" is bounded be a subtype of Comparable<T>.
Both parameters "a" and "b" must have the same type as the type parameter "T".
The return value will be the same type as the type parameter "T".

Now when programmers try to invoke this generic method, they still have chances of making those same mistakes. But the compiler will catch them because of the stronger definition of the method. The resulting bytecode will be less likely running into runtime exceptions

String max = SomeClass.<String>maxGeneric(
   new String("1234"), new String("789")); 
   // Correct invokation. No (String) casting needed in source code.

Integer max = SomeClass.<String>maxGeneric(
   new String("1234"), new String("789")); // Compilation error
   
String max = SomeClass.<String>maxGeneric(
   new String("1234"), new Integer("789")); // Compilation error

Note that the class name "SomeClass" is needed because the type argument <String> is required to show up after the dot (.) operator.

Last update: 2014.

Table of Contents

About This Book

Installing JDK 1.8 on Windows

Execution Process, Entry Point, Input and Output

Primitive Data Types and Literals

Bits, Bytes, Bitwise and Shift Operations

Managing Bit Strings in Byte Arrays

Reference Data Types and Variables

StringBuffer - The String Buffer Class

System Properties and Runtime Object Methods