http://www.theserverside.com Most recent forum messages en Jive Forums Silver 5.5.30 (www.jivesoftware.com) Sat, 25 May 2013 09:24:54 -0400 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Fri, 14 May 2004 09:07:57 -0400 Fri, 14 May 2004 09:07:57 -0400 Fri, 14 May 2004 09:07:57 -0400 May 14, 2004 Robert Simmons 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Tue, 04 May 2004 23:31:24 -0400 Tue, 04 May 2004 23:31:24 -0400 Tue, 04 May 2004 23:31:24 -0400 May 4, 2004 Jamie Flournoy 1 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Because even if a List is declared as final, you can still modify its content at will.Not if you wrap it with Collections.unmodifiableList(java.util.List). If the code has it's own mutable objects, you can write similar wrappers...]]> Tue, 04 May 2004 02:52:58 -0400 Tue, 04 May 2004 02:52:58 -0400 Tue, 04 May 2004 02:52:58 -0400 May 4, 2004 Jani Kaarela 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612
As for final, Im of two minds on using it on classes. Both...]]> Mon, 03 May 2004 20:21:06 -0400 Mon, 03 May 2004 20:21:06 -0400 Mon, 03 May 2004 20:21:06 -0400 May 3, 2004 Robert Simmons 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612

The point was that by making the param final
f = f will not compile while this.f = f will and will...]]> Fri, 30 Apr 2004 16:25:52 -0400 Fri, 30 Apr 2004 16:25:52 -0400 Fri, 30 Apr 2004 16:25:52 -0400 Apr 30, 2004 Todd Murray 1 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Upside? Pick up OO 101 to learn the benefits of encapsulation over extension when it comes to dependencies and decoupling.

Whatever perfect coding world you got to live in, get me a ticket there as fast as you can ;-)

When...]]> Fri, 30 Apr 2004 15:51:10 -0400 Fri, 30 Apr 2004 15:51:10 -0400 Fri, 30 Apr 2004 15:51:10 -0400 Apr 30, 2004 Cameron Purdy 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Why did you *have to* extend it? You only have to extend it is if you can't change the calling code, and the calling code expects the superclass. If that's the case, it's an interface-implementation issue again; the original coder could have...]]> Fri, 30 Apr 2004 15:48:37 -0400 Fri, 30 Apr 2004 15:48:37 -0400 Fri, 30 Apr 2004 15:48:37 -0400 Apr 30, 2004 Cameron Purdy 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Using "final" is yet another way that one programmer arrogantly pretends to know the future. The "final" keyword, like any other, is a tool. A underused tool by far.

I stand by my original post. I'd...]]> Fri, 30 Apr 2004 15:35:16 -0400 Fri, 30 Apr 2004 15:35:16 -0400 Fri, 30 Apr 2004 15:35:16 -0400 Apr 30, 2004 George Coller 1 http://www.theserverside.com/discussions/thread.tss?thread_id=25612
I think you misunderstood my point. I'm against using final all over the place. I was using the fact that most things in my code could use it, as evidence that it doesn't really make sense.

I offered the option of having a...]]> Fri, 30 Apr 2004 13:26:24 -0400 Fri, 30 Apr 2004 13:26:24 -0400 Fri, 30 Apr 2004 13:26:24 -0400 Apr 30, 2004 Chris McMahon 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612
Just by looking at that code, it's obvious you're asking for trouble. Try some different names or at least use "this".]]> Fri, 30 Apr 2004 13:23:12 -0400 Fri, 30 Apr 2004 13:23:12 -0400 Fri, 30 Apr 2004 13:23:12 -0400 Apr 30, 2004 Chris McMahon 2 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Again, I think I am living in a different universe, because I don't see that many immutable classes, and the ones that should be (e.g. Date) aren't! ;-)Cameron, we're apparently in the same universe because I don't see many...]]> Fri, 30 Apr 2004 13:14:58 -0400 Fri, 30 Apr 2004 13:14:58 -0400 Fri, 30 Apr 2004 13:14:58 -0400 Apr 30, 2004 Michael Mahemoff 1 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 The *process* is final.I agree with you, you actually made my point. The process *is* final, so make the template methods that define the process *final* so subclasses can't mess with it.]]> Fri, 30 Apr 2004 13:03:58 -0400 Fri, 30 Apr 2004 13:03:58 -0400 Fri, 30 Apr 2004 13:03:58 -0400 Apr 30, 2004 Keith Donald 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 Fri, 30 Apr 2004 11:51:16 -0400 Fri, 30 Apr 2004 11:51:16 -0400 Fri, 30 Apr 2004 11:51:16 -0400 Apr 30, 2004 ABC DEF 0 http://www.theserverside.com/discussions/thread.tss?thread_id=25612
class X {

private Object f;

public X(Object f) {
f = f; // which f is being modified?
}

}

class Y {

Object f;

public Y(final Object f) {
f = f; // will not compile
}
}]]> Fri, 30 Apr 2004 10:38:47 -0400 Fri, 30 Apr 2004 10:38:47 -0400 Fri, 30 Apr 2004 10:38:47 -0400 Apr 30, 2004 Todd Murray 3 http://www.theserverside.com/discussions/thread.tss?thread_id=25612 IMHO promoting them to private inner classes, especially to private static inner classes where possible, usually leads to more readable code.I'll agree the hoisting a class out of a method is more verbose. But how is it...]]> Fri, 30 Apr 2004 10:24:44 -0400 Fri, 30 Apr 2004 10:24:44 -0400 Fri, 30 Apr 2004 10:24:44 -0400 Apr 30, 2004 Brian Miller 0