Sorting
CSC 212: Data Structures
King Saud University
Outline
The sorting problem.
Properties of sorting algorithms.
General purpose sorting algorithms:
Quadratic sorting algorithms: selection sort and bubble sort.
Sub-quadratic sorting algorithms: merge sort.
(KSU) Sorting 2 / 16
The sorting problem
Given a list of n totally orderable items, rearrange the list in increasing (or decreasing) order.
Totally orderable means that any two items can be compared: numbers, characters,
strings etc.
The items are usually called keys.
(KSU) Sorting 3 / 16
The sorting problem
Given a list of n totally orderable items, rearrange the list in increasing (or decreasing) order.
Totally orderable means that any two items can be compared: numbers, characters,
strings etc.
The items are usually called keys.
Example
This is an instance of the sorting problem:
(12, 5, 8, 16, 9, 31) → (5, 8, 9, 12, 16, 31) .
(KSU) Sorting 3 / 16
The sorting problem
Given a list of n totally orderable items, rearrange the list in increasing (or decreasing) order.
Totally orderable means that any two items can be compared: numbers, characters,
strings etc.
The items are usually called keys.
Example
This is an instance of the sorting problem:
(12, 5, 8, 16, 9, 31) → (5, 8, 9, 12, 16, 31) .
Often, the elements to be sorted contain keys and data:
((5, A), (2, C), (7, A), (2, B), (1, B)) → ((1, B), (2, C), (2, B), (5, A), (7, A)) .
(KSU) Sorting 3 / 16
Properties of sorting algorithms
Time complexity: worst case and average case.
(KSU) Sorting 4 / 16
Properties of sorting algorithms
Time complexity: worst case and average case.
Space complexity: the amount of extra space needed by the algorithm.
(KSU) Sorting 4 / 16
Properties of sorting algorithms
Time complexity: worst case and average case.
Space complexity: the amount of extra space needed by the algorithm.
Definition
An algorithm is said in-place if it does not require more than O(log n) space in addition to the
input.
(KSU) Sorting 4 / 16
Properties of sorting algorithms
Time complexity: worst case and average case.
Space complexity: the amount of extra space needed by the algorithm.
Definition
An algorithm is said in-place if it does not require more than O(log n) space in addition to the
input.
Stability: A sorting algorithm is stable if it does not change the order of equal elements.
(KSU) Sorting 4 / 16
Properties of sorting algorithms
Time complexity: worst case and average case.
Space complexity: the amount of extra space needed by the algorithm.
Definition
An algorithm is said in-place if it does not require more than O(log n) space in addition to the
input.
Stability: A sorting algorithm is stable if it does not change the order of equal elements.
Example
Given the input array: {(5, A), (2, C), (7, A), (2, B), (1, B)}, where we want to sort according
to the first element of the pairs (the integers), then:
(KSU) Sorting 4 / 16
Properties of sorting algorithms
Time complexity: worst case and average case.
Space complexity: the amount of extra space needed by the algorithm.
Definition
An algorithm is said in-place if it does not require more than O(log n) space in addition to the
input.
Stability: A sorting algorithm is stable if it does not change the order of equal elements.
Example
Given the input array: {(5, A), (2, C), (7, A), (2, B), (1, B)}, where we want to sort according
to the first element of the pairs (the integers), then:
{(1, B), (2, C), (2, B), (5, A), (7, A)} is a stable sorting.
(KSU) Sorting 4 / 16
Properties of sorting algorithms
Time complexity: worst case and average case.
Space complexity: the amount of extra space needed by the algorithm.
Definition
An algorithm is said in-place if it does not require more than O(log n) space in addition to the
input.
Stability: A sorting algorithm is stable if it does not change the order of equal elements.
Example
Given the input array: {(5, A), (2, C), (7, A), (2, B), (1, B)}, where we want to sort according
to the first element of the pairs (the integers), then:
{(1, B), (2, C), (2, B), (5, A), (7, A)} is a stable sorting.
{(1, B), (2, B), (2, C), (5, A), (7, A)} is not a stable sorting, since the order of (2, B) and (2, C) is
reversed.
(KSU) Sorting 4 / 16
General purpose sorting algorithms
Algorithms which can be used to sort any type of keys.
They are based on comparison only and do not assume any other property in the keys (for
example, they do not require the keys to be integers or strings).
(KSU) Sorting 5 / 16
Selection sort
Selection sort gradually builds the sorted array by finding the correct key for each new position.
p u b l i c s t a t i c v o i d selectionSort ( i n t [] A , i n t n ) {
f o r ( i n t i = 0; i < n - 1; i ++) {
i n t min = i ;
f o r ( i n t j = i + 1; j < n ; j ++) { // Search for the minimum
i f ( A [ j ] < A [ min ])
min = j ;
}
// Swap A [ i ] with A [ min ]
i n t tmp = A [ i ];
A [ i ] = A [ min ];
A [ min ] = tmp ;
}
}
(KSU) Sorting 6 / 16
Selection sort
Example
⇑ indicates i, ↑ indicates min.
12, 5, 8, 16, 9, 31
⇑ ↑
(KSU) Sorting 7 / 16
Selection sort
Example
⇑ indicates i, ↑ indicates min.
12, 5, 8, 16, 9, 31
⇑ ↑
5, 12, 8, 16, 9, 31
⇑ ↑
(KSU) Sorting 7 / 16
Selection sort
Example
⇑ indicates i, ↑ indicates min.
12, 5, 8, 16, 9, 31
⇑ ↑
5, 12, 8, 16, 9, 31
⇑ ↑
5, 8, 12, 16, 9, 31
⇑ ↑
(KSU) Sorting 7 / 16
Selection sort
Example
⇑ indicates i, ↑ indicates min.
12, 5, 8, 16, 9, 31 5, 8, 9, 16, 12, 31
⇑ ↑ ⇑ ↑
5, 12, 8, 16, 9, 31
⇑ ↑
5, 8, 12, 16, 9, 31
⇑ ↑
(KSU) Sorting 7 / 16
Selection sort
Example
⇑ indicates i, ↑ indicates min.
12, 5, 8, 16, 9, 31 5, 8, 9, 16, 12, 31
⇑ ↑ ⇑ ↑
5, 12, 8, 16, 9, 31 5, 8, 9, 12, 16, 31
⇑ ↑ ⇑
5, 8, 12, 16, 9, 31
⇑ ↑
(KSU) Sorting 7 / 16
Selection sort
Worst case time complexity: O(n2 ) (quadratic).
Average case time complexity: O(n2 ).
Space complexity: O(1).
In-place: Yes.
Stable: No.
Example
The array {(2, A), (2, B), (1, C)} will be sorted as {(1, C), (2, B), (2, A)}.
(KSU) Sorting 8 / 16
Bubble sort
Bubble sort sorts the array by repeatedly swapping non-ordered adjacent keys. After each for
loop iteration, the maximum is moved (or bubbled) towards the end.
p u b l i c s t a t i c v o i d bubbleSort ( i n t A [] , i n t n ) {
f o r ( i n t i = 0; i < n - 1; i ++) {
f o r ( i n t j = 0; j < n - 1 - i ; j ++) {
i f ( A [ j ] > A [ j + 1]) {
// Swap A [ j ] with A [ j + 1]
i n t tmp = A [ j ];
A [ j ] = A [ j + 1];
A [ j + 1] = tmp ;
}
}
}
}
(KSU) Sorting 9 / 16
Bubble sort
Example
(12, 5, 8, 16, 9, |31)
(KSU) Sorting 10 / 16
Bubble sort
Example
(12, 5, 8, 16, 9, |31)
(5, 8, 12, 9, |16, 31)
(KSU) Sorting 10 / 16
Bubble sort
Example
(12, 5, 8, 16, 9, |31)
(5, 8, 12, 9, |16, 31)
(5, 8, 9, |12, 16, 31)
(KSU) Sorting 10 / 16
Bubble sort
Example
(12, 5, 8, 16, 9, |31) (5, 8, |9, 12, 16, 31)
(5, 8, 12, 9, |16, 31)
(5, 8, 9, |12, 16, 31)
(KSU) Sorting 10 / 16
Bubble sort
Example
(12, 5, 8, 16, 9, |31) (5, 8, |9, 12, 16, 31)
(5, 8, 12, 9, |16, 31) (5, |8, 9, 12, 16, 31)
(5, 8, 9, |12, 16, 31)
(KSU) Sorting 10 / 16
Bubble sort
Worst case time complexity: O(n2 ) (quadratic).
Average case time complexity: O(n2 ).
Space complexity: O(1).
In-place: Yes.
Stable: Yes.
Remark
Bubble sort performs a lot of swaps, and as a result it has in practice a poor performance
compared to selection sort.
(KSU) Sorting 11 / 16
Merge sort
Merge sort is a divide-and-conquer algorithms to sort an array of n elements:
1 Divide the array into two equal parts.
2 Sort each part apart (recursively).
3 Merge the two sorted parts.
The key step in merge sort is merging two sorted arrays, which can be done in O(n).
Example
Given two arrays B = {1, 4, 6} and C = {2, 3, 7, 8}, the result of merging B and C is
{1, 2, 3, 4, 6, 7, 8}.
(KSU) Sorting 12 / 16
Merge sort
p u b l i c s t a t i c v o i d mergeSort ( i n t [] A , i n t l , i n t r ) {
i f ( l >= r )
return ;
i n t m = ( l + r ) / 2;
mergeSort (A , l , m ) ; // Sort first half
mergeSort (A , m + 1 , r ) ; // Sort second half
merge (A , l , m , r ) ; // Merge
}
(KSU) Sorting 13 / 16
Merge sort
p r i v a t e s t a t i c v o i d merge ( i n t [] A , i n t l , i n t m , i n t r ) {
i n t [] B = new i n t [ r - l + 1];
i n t i = l , j = m + 1 , k = 0;
w h i l e ( i <= m && j <= r )
i f ( A [ i ] <= A [ j ])
B [ k ++] = A [ i ++];
else
B [ k ++] = A [ j ++];
i f (i > m)
w h i l e ( j <= r )
B [ k ++] = A [ j ++];
else
w h i l e ( i <= m )
B [ k ++] = A [ i ++];
f o r ( k = 0; k < B . length ; k ++)
A [ k + l ] = B [ k ];
}
(KSU) Sorting 14 / 16
Merge sort
Example
Sort the array: 8, 3, 2, 9, 7, 1, 5, 4.
(KSU) Sorting 15 / 16
Merge sort
Worst case time complexity: O(n log n) (sub-quadratic).
Average case time complexity: O(n log n).
Space complexity: O(n) (requires auxiliary memory).
In-place: No.
Stable: Yes.
(KSU) Sorting 16 / 16
