Scribd
Upload
86 views

CS109A Notes For Lecture 3/8/96 Data Structures: Linked List Next Array

This document summarizes key data structures and algorithms topics covered in a lecture: 1. Linked lists and arrays were discussed as fundamental data structures, with operations like lookup, insert, and delete on linked lists taking O(n) time. 2. An example recursive insertion algorithm for ML lists was provided and its correctness proved using induction. 3. Variants for linked list implementation including sorting, allowing duplicates, and sentinels were covered. 4. Stacks and queues were defined as abstract data types (ADTs) with common push/pop and enqueue/dequeue operations.

Uploaded by

savio77
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
86 views

CS109A Notes For Lecture 3/8/96 Data Structures: Linked List Next Array

This document summarizes key data structures and algorithms topics covered in a lecture: 1. Linked lists and arrays were discussed as fundamental data structures, with operations like lookup, insert, and delete on linked lists taking O(n) time. 2. An example recursive insertion algorithm for ML lists was provided and its correctness proved using induction. 3. Variants for linked list implementation including sorting, allowing duplicates, and sentinels were covered. 4. Stacks and queues were defined as abstract data types (ADTs) with common push/pop and enqueue/dequeue operations.

Uploaded by

savio77
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

CS109A Notes for Lecture 3/8/96

Data Structures

1. Linked list = records with data eld(s) and


next eld pointing to next element.
2. Array = array of limited size with cursor or
pointer to last element.

Operations

Lookup, insert, delete (like the Dictionary ADT)


are most common.
Take O(n) time on an n-element list.
Example: Here is insertion in an ML list.
Does not create duplicate elements, so must
check x is not already on list.
(1) fun insert(x,nil) = x]
(2) | insert(x, y::ys) =
(3)
if x<>y then y::insert(x,ys)
(4)
else y::ys;
Correctness proof:
S (n): If L is of length n, then insert(x; L)
returns a list with x and the elements of L,
and nothing else.
Basis: n = 0. Then L has no elements, line (1)'s
pattern matches, and a list with only x is returned.
Induction: Assume S (n), n 0. If L is of length
n + 1, line (1) doesn't match. Line (2) matches.
If x 6= y, then by the inductive hypothesis,
insert(x; ys) returns a list with the elements
of L except for y but with x included. Then
line (3) returns a list with y, x, and the other
elements of L, i.e., what S (n + 1) says should
be returned.
If x = y, then line (4) returns L. Since x is on
L, again we return what S (n + 1) says should
be returned.
1

Note that we used the inductive hypothesis to talk about what happens on recursive calls, without having to imagine
an arbitrarily large sequence of calls.

Implementation Variants
1. Sorting the list.
We can search only as far as x to test
whether x is on the list (saves average
factor of 2).
2. Allow duplicates.
Insert in O(1).
Penalty is that lookup, delete may take
longer because lists with duplicates get
longer than number of elements.
3. Sentinels : Add x onto end of list before
searching for x.
Suitable only for array representation.
Saves time testing for end of list at each
step.

Stacks and Queues


Stack = ADT with principal operations push
and pop.
exception EmptyStack;
fun push(x,S) = x::S;
fun pop(nil) = raise EmptyStack
|
pop(x::xs) = xs;

Queue = ADT with principal operations


enqueue and dequeue (= pop).
exception EmptyQueue;
fun enqueue(x,Q) = Q@ x];
fun deqeue(nil) = raise EmptyQueue
|
dequeue(x::xs) = xs;

Use of Stack to Support Recursive Calls

Here is the preorder function from Fig. 5.32, FCS.


void preorder(TREE t)
{

(1)
(2)
(3)
(4)

if (t != NULL) {
printf("%c\n", t->nodeLabel);
preorder(t->leftChild);
preorder(t->rightChild);
}

The run-time implementation of such a function is essentially as follows.


Keep a stack whose entries are pairs that tell
us what we need to know about the state of
a call to preorder:
1. The value of t, a pointer to the root of
the tree about which the call to preorder
was made.
2. The place in the execution of the function, essentially the line number being
executed. Most important, when we
make a recursive call, is it from line (3)
or line (4)?
When a new call is made at line (3) or (4),
push the new value of t onto the stack with
line number = 1.
When a call to preorder returns, pop the
stack, exposing the value of t and the current line number from the previous call.
Example: Consider the tree:
a
b

Here is the sequence of stacks (top at the right) in


which the pair (x; i) represents a stack entry for
the call in which t is a pointer to node x and i is
3

the line number being executed.


Ignores calls on empty trees that immediately
return.
(a; 1)
(a; 3)(b; 1)
(a; 3)(b; 3)(d; 1)
(a; 3)(b; 3)
(a; 3)(b; 4)(e; 1)
(a; 3)(b; 4)
(a; 3)
(a; 4)(c; 1)
(a; 4)

You might also like