Okay, next we'll briefly consider queue implementations using
the same basic underlying data structures.
So here's the corresponding API for QueueOfStrings.
Actually it's the same API as for stacks, just the names are different.
Instead of push we have enqueue instead of pop we have dequeue.
And the semantics is different, for enqueue we add an item say at the end
of the queue, and for dequeue we remove an item from the beginning.
It's as if you're waiting in line to buy a ticket.
When you're enqueue you go at the end, and
when the one that's been in there the longest is the one that comes off.
So lets look at how we implement those, first using linked list and then arrays.
So now our representation of a queue with a length list,
we need to maintain two pointers, references.
One, to the first item in the list and the other to the last item in the list.
When we insert,
we're going to add the item at the end of the list instead of at the beginning.
And when we remove, we'll do the same, we'll take it off at the front.
So here's the implementation of dequeue.
It's identical to the code for pop for a stack.
We save away the item, we delete the first node by advancing the reference and
then we return the item, so identical.
To add a note, or enqueue, add a new note to a linked-list.
We want to put it at the end, so that will be the last one returned.
So we, to add it at the end, so
first thing we need to do is save a link, the last node.
We're going to need that because we need to change it's reference
from null to point to the new node.
Then we'll create a new node for the end of the list.
We'll populate its fields and then that old link
will change that from null to a pointer to the new node.
So again, just a few lines of code,
[COUGH] that's basic linked-list processing.
Actually years ago when we taught courses in algorithms and data structures,
much of the course would be about this kind of pointer manipulation.
But nowadays that's restricted to just a few implementations like stack and
queue and a few other fundamental data structures.
So we don't need so much any more general programs for manipulating linked-lists.
We encapsulate them in basic data types like these.
All right, so let's go back to our full implementation and
this is just taking care of collecting the code from the previous slides.
But also taking care of special cases when the queue is empty.
To make sure that if the queue is empty after we move an item
we gotta set last to null.
Make sure that both first and last are always what we want them to be.
So those are details that are easy to check.
Okay, what about arrays?
Well, we won't do the details, but
it's not difficult to implement queues with resizing arrays as well.
Not difficult, but
a definitely tricky programming exercise that people are welcome to try.
So we'll maintain two pointers, the first item in the queue and
the tail, which is the position for the next item to appear.
So for enqueue, you add a new item a tail.
And for dequeue you remove an item for a head.
And the trick is that once you get past the capacity,
you have to reset back to zero.
And so that's a little extra code, and
then you have to add the resizing capability as well to
implement the data structure the same as for stack.
And we'll leave that as an exercise.
Kevin WaynePhillip Y. Goldman '86 Senior Lecturer
Robert SedgewickWilliam O. Baker *39 Professor of Computer Science
