Reputation: 6127
Is a java compiler or runtime ( or any other language compiler ) smart enough to realize branch 3 can never happen , and optimize it away? I've seen this kind of "defensive programming" with many beginning developers, and wonder if this dead weight stays in the bytecode.
import java.util.Random;
class Example
{
public static void main(String[] args) {
int x = new Random().nextInt() % 10;
if ( x < 5 )
{
System.out.println("Case 1");
}
else
if ( x >= 5 )
{
System.out.println("Case 2");
}
else
{
System.out.println("Case 3");
}
}
}
or even this more blunt case
boolean bool = new Random().nextBoolean();
if ( bool )
{
System.out.println("Case 1");
}
else
if ( bool )
{
System.out.println("Case 2");
}
Upvotes: 3
Views: 181
Reputation: 178363
The Java 8 compiler I have doesn't seem to optimize it away. Using "javap -c" to examine the byte code after compiling:
public static void main(java.lang.String[]);
Code:
0: new #2 // class java/util/Random
3: dup
4: invokespecial #3 // Method java/util/Random."<init>":()V
7: invokevirtual #4 // Method java/util/Random.nextInt:()I
10: bipush 10
12: irem
13: istore_1
14: iload_1
15: iconst_5
16: if_icmpge 30
19: getstatic #5 // Field java/lang/System.out:Ljava/io/PrintStream;
22: ldc #6 // String Case 1
24: invokevirtual #7 // Method java/io/PrintStream.println:(Ljava/lang/String;)V
27: goto 54
30: iload_1
31: iconst_5
32: if_icmplt 46
35: getstatic #5 // Field java/lang/System.out:Ljava/io/PrintStream;
38: ldc #8 // String Case 2
40: invokevirtual #7 // Method java/io/PrintStream.println:(Ljava/lang/String;)V
43: goto 54
46: getstatic #5 // Field java/lang/System.out:Ljava/io/PrintStream;
49: ldc #9 // String Case 3
51: invokevirtual #7 // Method java/io/PrintStream.println:(Ljava/lang/String;)V
54: return
}
The string "Case 3" still exists in the byte code.
Upvotes: 4