This weekās section exercises explore ADT and BigO problems. By the end of this week, you'll be wellversed in all kinds of ADTs, and you'll be able to think critically about program runtime. What a time to be alive!
Remember that every week we will also be releasing a Qt Creator project containing starter code and testing infrastructure for that week's section problems. When a problem name is followed by the name of a .cpp
file, that means you can practice writing the code for that problem in the named file of the Qt Creator project. Here is the zip of the section starter code:
š¦ Starter code
1) Oh No, BigO, Too Slow
Topics: BigO, code analysis
Give a tight bound of the nearest runtime complexity class for each of the following code fragments in BigOh notation, in terms of the variable N.
Code Snippet A
int sum = 0;
for (int i = 1; i <= N + 2; i++) {
sum++;
}
for (int j = 1; j <= N * 2; j++) {
sum++;
}
cout << sum << endl;
Code Snippet B
int sum = 0;
for (int i = 1; i <= N  5; i++) {
for (int j = 1; j <= N  5; j += 2) {
sum++;
}
}
cout << sum << endl;
Code Snippet C
int sum = 0;
for (int i = 0; i < 1000000; i++) {
for (int j = 1; j <= i; j++) {
sum += N;
}
for (int j = 1; j <= i; j++) {
sum += N;
}
for (int j = 1; j <= i; j++) {
sum += N;
}
}
cout << sum << endl;
Code Snippet A has a runtime complexity of O(N).
Code Snippet B has a runtime complexity of O(N^2).
Code Snippet C has a runtime complexity of O(1).
2) Recursion Mystery Part 1
Topics: recursive function calls, return value tracing
Code Snippet A
int recursionMystery(int x, int y) {
if (x < y) {
return x;
} else {
return recursionMystery(x  y, y);
}
}
For each call to the above recursive function, indicate what value is returned by the function call.
Call Output
recursionMystery(6, 13); ______________
recursionMystery(14, 10); ______________
recursionMystery(37, 12); ______________
6
4
1
3) Recursion Mystery Part 2
Topics: recuresive function calls, output tracing
void recursionMystery2(int x, int y) {
if (y == 1) {
cout << x;
} else {
cout << (x * y) << ", ";
recursionMystery2(x, y  1);
cout << ", " << (x * y);
}
}
For each call to the above recursive function, write the output that would be produced, as it would appear on the console.
Call Output
recursionMystery2(4, 1); ___________________________________
recursionMystery2(4, 2); ___________________________________
recursionMystery2(8, 2); ___________________________________
recursionMystery2(4, 3); ___________________________________
recursionMystery2(3, 4); ___________________________________
4
8, 4, 8
16, 8, 16
12, 8, 4, 8, 12
12, 9, 6, 3, 6, 9, 12
4) Recursion Tracing
Topics: Recursion, strings, recursion tracing
Below is a recursive function to reverse a string.
string reverseOf(string s) {
if (s.empty()) {
return "";
} else {
return reverseOf(s.substr(1)) + s[0];
}
}
Trace through the execution of reverseOf("stop")
along the lines of what we did in lecture, showing recursive call information for each call thatās made and how the final value gets computed.
Our initial call to reverseOf("stop")
fires off a call to reverseOf("top")
. This call fires off a call to reverseOf("op")
. This in turn calls reverseOf("p")
. This in turn calls reverseOf("")
. This triggers the base case and returns the empty string. (Notice that the reverse of the empty string "" is indeed the empty string ""). We now append p
to return "p"
. We now append o
to return "po"
. We append t
to return "pot"
. And finally we append s to return "pots"
back to whoever called us. Yay!
5) Sum of Squares (squares.cpp
)
Topics: Recursion
Write a recursive function named sumOfSquares
that takes an int n
and returns the sum of squares from 1
to n
. For example, sumOfSquares(3)
should return 1^2 + 2^2 + 3^2 = 14. If n
is negative, you should report an error to that effect.
int sumOfSquares(int n) {
if (n < 0) {
error("Value of provided n was negative");
} else if (n == 0) {
return 0;
} else {
return n * n + sumOfSquares(n1);
}
}
6) Random Shuffling (shuffle.cpp
)
How might the computer shuffle a deck of cards? This problem is a bit more complex than it might seem, and while it's easy to come up with algorithms that randomize the order of the cards, only a few algorithms will do so in a way that ends up generating a uniformlyrandom reordering of the cards.
One simple algorithm for shuffling a deck of cards is based on the following idea:
 Choose a random card from the deck and remove it.
 Shuffle the rest of the deck.
 Place the randomlychosen card on top of the deck. Assuming that we choose the card that we put on top uniformly at random from the deck, this ends up producing a random shuffle of the deck.
Write a function
string randomShuffle(string input)
that accepts as input a string, then returns a random permutation of the elements of the string using the above algorithm. Your algorithm should be recursive and not use any loops (for
, while
, etc.).
The header file "random.h"
includes a function
int randomInteger(int low, int high);
that takes as input a pair of integers low and high, then returns an integer greater than or equal to low and less than or equal to high. Feel free to use that here.
Interesting note: this shuffling algorithm is a variant of the FisherYates Shuffle. For more information on why it works correctly, take CS109!
Here is one possible solution:
string randomShuffle(string input) {
/* Base case: There is only one possible permutation of a string
* with no characters in it.
*/
if (input.empty()) {
return input;
} else {
/* Choose a random index in the string. */
int i = randomInteger(0, input.length()  1);
/* Pull that character to the front, then permute the rest of
* the string.
*/
return input[i] + randomShuffle(input.substr(0, i) + input.substr(i + 1));
}
}
This function is based on the recursive observation that there is only one possible random shuffle of the empty string (namely, itself), and then using the algorithm specified in the handout for the recursive step.
Here is another possible solution (using a modified function header) which shows how to implement this function using references and no return value. Shoutout to one of our awesome SLs, Rachel Gardner, for this alternate solution!
void randomShuffle(string &input){
if (input == "") return;
int rand = randomInteger(0, input.length()  1);
char chosen = input[rand];
input.erase(rand, 1);
randomShuffle(input);
input = chosen + input;
}
7) Zig Zag (zigzag.cpp
)
Topics: Recursion, printing output to console
Write a recursive function named zigzag
that returns a string of n
characters as follows. The middle character (or middle two characters if n
is even) is an asterisk (*). All characters before the asterisks are '<'
. All characters after are '>'
. Report an error if n
is not positive.
Call Output
zigzag(1) *
zigzag(4) <**>
zigzag(9) <<<<*>>>>
string zigzag(int n) {
if (n < 1) {
error("The value of n was negative");
} else if (n == 1) {
return "*";
} else if (n == 2) {
return "**";
} else {
return "<" + zigzag(n2) + ">";
}
}
8) Double Stack (double.cpp
)
Topics: Recursion, Stacks
Write a recursive function named doubleStack
that takes a reference to a stack of ints and replaces each integer with two copies of that integer. For example, if s
stores {1, 2, 3}
, then doubleStack(s)
changes it to {1, 1, 2, 2, 3, 3}
.
void doubleStack(Stack<int>& s)
{
if (!s.isEmpty()) {
int n = s.pop();
doubleStack(s);
s.push(n);
s.push(n);
}
}
9) String Subsequences (subsequence.cpp
)
Topics: Recursion, verifying properties
Write a recursive function named isSubsequence
that takes two strings and returns true if the second string is a subsequence of the first string. A string is a subsequence of another if it contains the same letters in the same order, but not necessarily consecutively. You can assume both strings are already lowercased.
Call Output
isSubsequence("computer", "core") false
isSubsequence("computer", "cope") true
isSubsequence("computer", "computer") true
bool isSubsequence(string big, string small)
{
if (small.empty()) {
return true;
} else if (big.empty()) {
return false;
} else {
if (big[0] == small[0]) {
return isSubsequence(big.substr(1), small.substr(1));
} else {
return isSubsequence(big.substr(1), small);
}
}
}
10) Reversing User Input (reverse.cpp
)
Write a recursive function named reverseUserInput
that defines a console program which continually asks the user for textual input, line by line, until the user hits "Enter/Return". At that point, the program should print out all the lines that the user inputted, in the reverse order that they were entered. This should be accomplished with one recursive function and no auxiliary data structures.
That is, you cannot store the lines of input in a Vector or any other data structure. Below is a sample log of execution of the program that you should try to emulate.
Your input (Enter/Return to quit): Roses are red,
Your input (Enter/Return to quit): Violets are blue.
Your input (Enter/Return to quit): All my base
Your input (Enter/Return to quit): Are belong to you.
Your input (Enter/Return to quit):
Here is your reversed output:
Are belong to you.
All my base
Violets are blue.
Roses are red,
void reverseUserInput(){
string input = getLine("our input (Enter/Return to quit): ")
if (input.empty()){
cout << endl << endl << "Here is your reversed output: " << endl;
} else{
reverseUserInput();
cout << input << endl;
}
}
11) Recursion and Time Complexity and Exponents, (Big)Oh My
Topics: recursion, time complexity, big O, algorithm comparison
Below is a simple function that computes the value of m^n when n is a nonnegative integer:
int raiseToPower(int m, int n) {
int result = 1;
for (int i = 0; i < n; i++) {
result *= m;
}
return result;
}

What is the bigO complexity of the above function, written in terms of m and n? You can assume that it takes time
O(1)
to multiply two numbers. 
If it takes 1Ī¼s to compute
raiseToPower(100, 100)
, approximately how long will it take to computeraiseToPower(200, 10000)
?
Below is a recursive function that computes the value of m^n
when n
is a nonnegative integer:
int raiseToPower(int m, int n) {
if (n == 0) return 1;
return m * raiseToPower(m, n  1);
}
1.What is the bigO complexity of the above function, written in terms of m
and n
? You can assume that it takes time O(1)
to multiply two numbers.
 If it takes 1Ī¼s to compute
raiseToPower(100, 100)
, approximately how long will it take to computeraiseToPower(200, 10000)
?
Hereās an alternative recursive function for computing m^n
that uses a technique called exponentiation by squaring. The idea is to modify the recursive step as follows.
 If
n
is an even number, then we can writen
asn = 2k
. Thenm^n = m^(2k) = (m^k)^2
.  If
n
is an odd number, then we can writen
asn = 2k + 1
. Thenm^n = m^(2k+1) = m * m^(2k) = m* (m^k)^2
.
Based on this observation, we can write this recursive function:
int raiseToPower(int m, int n) {
if (n == 0) {
return 1;
} else if (n % 2 == 0) {
int halfPower = raiseToPower(m, n / 2);
return halfPower * halfPower;
} else {
int halfPower = raiseToPower(m, n / 2);
return m * halfPower * halfPower;
}
}
 What is the bigO complexity of the above function, written in terms of
m
andn
? You can assume that it takes timeO(1)
to multiply two numbers.  If it takes 1Ī¼s to compute
raiseToPower(100, 100)
, approximately how long will it take to computeraiseToPower(200, 10000)
?

This function runs in time
O(n)
. It runs the loopn
times, at each step doingO(1)
work. There is no dependence onm
in the runtime. 
We know that this code runs in time
O(n)
, so it scales roughly linearly with the size ofn
. Therefore, if it took 1Ī¼s to compute a value whenn = 100
, it will take roughly 100 times longer when we plug inn = 10000
. As a result, weād expect this code would take about 100Ī¼s to complete. 
If we trace through the recursion, weāll see that we make a total of
n
recursive calls, each of which is only doingO(1)
work. Adding up all the work done by these recursive calls gives us a total ofO(n)
work, as before. 
As before, this should take about 100Ī¼s.

Notice that each recursive call does
O(1)
work (there are no loops anywhere here), then calls itself on a problem thatās half as big as the original one. This means that onlyO(log n)
recursive calls will happen (remember that repeatedly dividing by two is the hallmark of a logarithm), so the total work done here isO(log n)
. 
We know that the runtime when n = 100 is roughly 1Ī¼s. Notice that 100^2 = 10,000, so weāre essentially asking for the runtime of this function when we square the size of the input. Also notice that via properties of logarithms that
log n^2 = 2 log n
. Therefore, since we know the runtime grows roughly logarithmically and weāve squared the value ofn
, this should take about twice as long as before, roughly 2Ī¼s.