A List Implementation That Links Data - Data Structures and Abstractions with Java - page 346

14.11

The Java method. The following implementation of the add method is based on the previous
code fragments. We begin by checking the validity of the insertion position, newPosition . If it is
within range, we create the new node. We then insert the new node into the chain according to its
intended position, newPosition : The insertion is either at the beginning of the list or somewhere
else in the list.

public boolean add( int newPosition, T newEntry)
{

boolean isSuccessful = true ;
if ((newPosition > = 1) && (newPosition <= numberOfEntries + 1))
{

Node newNode = new Node(newEntry);
if (newPosition == 1) // case 1

{

newNode.setNextNode(firstNode);
firstNode = newNode;

}
else // case 2: list is not empty
{ // and newPosition > 1
Node nodeBefore = getNodeAt(newPosition - 1);
Node nodeAfter = nodeBefore.getNextNode();
newNode.setNextNode(nodeAfter);
nodeBefore.setNextNode(newNode);
} // end if
numberOfEntries++;

}
else
isSuccessful = false ;
return isSuccessful;
} // end add

Question 6 Consider the first else clause of the previous method add .
a. What assert statement could you add to this clause?

b. What call to getNodeAt could replace the value assigned to nodeAfter ?

c. Should we make the change suggested in Part b ?

Question 7 In the previous method add , the second if statement tests the value of new
Position . Should the boolean expression it tests be isEmpty() || (newPosition == 1) ?
Explain.

Question 8 How do the add methods given in Segments 14.9 and 14.11 enforce the pre-
condition of getNodeAt ?

The Methods isEmpty and toArray
14.12

The method isEmpty and assertions. The implementation of the method isEmpty could simply
test that the length of the list is zero, as it did in the array-based implementations that you saw in the
previous chapter. However, we have another criterion that we could use here. When a list is empty,
the reference firstNode is null . If our implementation is correct, either criterion is fine, but what
happens during development when some part of our class might contain an error in logic? We can