LinkedList and remove

[Anders Kviback]

Hello!

I have a theoretical question here. I have this class OrderedList that represents element in ascending order.

public class OrderedList<E extends Comparable<E>> {
private LinkedList<E> theList;
public OrderedList() {
theList = new LinkedList<E>();
}
public void add(E obj) {
ListIterator<E> itr = theList.listIterator();
while (itr.hasNext()) {
if (obj.compareTo(itr.next()) < 0) {
itr.previous();
itr.add(obj);
return;
}
}
itr.add(obj);
}
public E get(int index) {
return theList.get(index);
}
public Iterator<E> iterator() {
return theList.iterator();
}
public int size() {
return theList.size();
}
} 

Now I'm supposed to implement a method remove into the class OrderedList.

/** Removes the first occurrence of the specified element from this list,
if it is present. If this list does not contain the element, it is unchanged.
Returns true if this list contained the specified element (or equivalently,
if this list changed as a result of the call). */
boolean remove(E obj);

I thought that I could implement this method in the same way as get, size and iterator are implemented ie

public boolean remove(E obj){

return theList.remove(E obj);
}

I mean ; there is a method boolean remove(E obj) in java.util.LinkedList<E>Class. But I can't use that one here?

The answer looks like this:

public boolean remove(E obj) {
ListIterator<E> itr = theList.listIterator();
while (itr.hasNext()) {
E e = itr.next();
if (e.compareTo(obj) == 0) {
itr.remove();
return true;
}
if (e.compareTo(obj) > 0) {
return false;
}
}
return false;
}

Why do I have to do it this way?

Anders

[Tim Moores]

But I can't use that one here?

Why not?

[Anders Kviback]

This is a "paper and pen" issue. I don't do any programming. 1. But you mean that it is OK to just write:

public boolean remove(E obj){   
  
return theList.remove(E obj);   
} 

2. Well, put it this way, what's the difference?

Anders

[Martin Vashko]

The difference basically is that if you program the logic yourself, you write more code, have more code to maintain in the future and have higher probability of bugs in that code, since core JDK classes have been already proved bug-free by millions of users.

In any case, I'd consider two possible different approaches:

1) Use an ArrayList instead of LinkedList to keep the data in. With ArrayList, you can efficiently perform binary searches (there are JDK methods for that) and random access.

2) Use a TreeSet to keep the data in. It keeps data sorted without the need to write a single line of code. Granted, it does not allow duplicates and you'll have to program index access yourself, but since your requirements are quite vague, it might be worth considering.

[Jeff Verdegan]

Martin Vajsar wrote:The difference basically is that if you program the logic yourself, you write more code, have more code to maintain in the future and have higher probability of bugs in that code, since core JDK classes have been already proved bug-free by millions of users.

On the other hand, by doing it yourself a time or two you learn more about the building blocks of the tools you'll be using day-in and day-out later. If the OP is at the "learning about data structures" stage of his education, then it's worth slogging through it.

[Anders Kviback]

Thanks for answering,

this is, as you propably understand, a java datastructure course. A very interesting course but sometimes a bit tricky.

Thanks again, Anders

[Jeff Verdegan]

Anders Kviback wrote:Thanks for answering,

this is, as you propably understand, a java datastructure course. A very interesting course but sometimes a bit tricky.

Thanks again, Anders

So, where are you now? Do you still have a question?

[Anders Kviback]

Sorry about that, no I don't have a question. I got the answer.

Anders