COS30008 Semester August 2024 Ms.

Siti Hawa

Swinburne University of Technology

School of Science, Computing and Engineering Technologies

ASSIGNMENT COVER SHEET

Subject Code: COS30008

Subject Title: Data Structures and Patterns
Assignment number and title: 1, Solution Design in C++
Due date: Wednesday, 20th September 2024, 23:59
Lecturer: Ms. Siti Hawa

Your name: Your student ID:

Marker's comments:

Problem Marks Obtained

1 26

2 98

3 32

Total 156

Extension certification:

This assignment has been given an extension and is now due on

Signature of Convener:

1
COS30008 Semester August 2024 Ms. Siti Hawa

Problem Set 1: Solution Design in C++

Problem 1
Start with the solution of tutorial 3 in which we implemented the class Vector3D.
In this problem, we which to extend Vector3D with a toString() method. This method
has to return a textual representation of a 3D vector. For example, toString() applied to
Vector3D( 1.0f, 2.0f, 3.0f ) has to yield a string “[1,2,3]” as textual
representation.
To support the new member function, class Vector3D is extended as follows
#pragma once

#include "Vector2D.h"

#include <string>

class Vector3D
{
private:

Vector2D fBaseVector;
float fW;

public:

Vector3D( float aX = 1.0f, float aY = 0.0f, float aW = 1.0f ) noexcept;

Vector3D( const Vector2D& aVector ) noexcept;

float x() const noexcept { return fBaseVector.x(); }

float y() const noexcept { return fBaseVector.y(); }
float w() const noexcept { return fW; }

float operator[]( size_t aIndex ) const;

explicit operator Vector2D() const noexcept;

Vector3D operator*( const float aScalar ) const noexcept;

Vector3D operator+( const Vector3D& aOther ) const noexcept;
float dot( const Vector3D& aOther ) const noexcept;

friend std::ostream& operator<<( std::ostream& aOStream, const Vector3D& aVector );

// Problem Set 1 extension

std::string toString() const noexcept;

};

Do not edit the provided files. To implement the required features, create a new source file,
say Vector3D_PS1.cpp, and define the new feature here. It is not strictly required, but it
helps to separate the definitions from the provided code. You have to include Vector3D.h
in the new source for the code to compile.
Use std::stringstream to implement the toString() method. The class
std::stringstream provides a memory stream. You can use formatted output (i.e., the
operator <<) to send data to this stream and at the end, use the method str() to obtain
the resulting string that toString() has to return.
Numerical data must be rounded to 10-4, that is, four positions after the period.
The file Main.cpp contains a test function to check your implementation of the new matrix
features. The code sequence

2
COS30008 Semester August 2024 Ms. Siti Hawa

void runP1()
{
gCount++;

Vector3D a( 1.0f, 2.0f, 3.0f );

Vector3D b(static_cast<float>(M_PI),static_cast<float>(M_PI),static_cast<float>(M_PI));
Vector3D c(1.23456789f, 9.876543210f, 12435.0987654321f);

cout << "Vector a: " << a.toString() << endl;

cout << "Vector b: " << b.toString() << endl;
cout << "Vector c: " << c.toString() << endl;
}

Should produce the following output

Vector a: [1,2,3]
Vector b: [3.1416,3.1416,3.1416]
Vector c: [1.2346,9.8765,12435.1]

Floating point values are printed with standard precision for type float.

3
COS30008 Semester August 2024 Ms. Siti Hawa

Problem 2
Start with the solution of tutorial 3 in which we implemented the classes Vector3D and
Matrix3x3 to perform vector transformations in 2D.
In this problem, we wish to extend the definition of class Matrix3x3 with some additional
matrix operations. In particular, we extend class Matrix3x3 with
• Matrix multiplication [26 marks]:
Two matrices F and G can be multiplied, provided that the number of columns in F is
equal to the number of rows in G. If F is n x m matrix and G is an m x p matrix, then
the product FG is an n x p matrix whose (i, j) entry is given by
X
m
(FG)ij = FikGkj
k=1

The entry (FG)ij is the dot product of row(F,i) and column(G,j).

In the implementation, every row and column must be accessed once via calls to row()
and column(). You can declare local variables. Computing the result does not require
loops.
• Determinate of a matrix [22 marks]:
The determinate is a scalar value that is a function of the entities of a square matrix.
It characterizes some properties of a square matrix, for example, if the matrix is
invertible or if the matrix is a rotation matrix.
For a 3 x 3 matrix M, the determinate of M is given by

det M = M 11(M 22M 33 — M 23M 32)

— M 12(M 21M 33 — M 23M 31)
+ M 13(M 21M 32 — M 22M 31)

In the implementation, every row must be accessed once via calls to row(). You can
declare local variables. Computing the result does not require loops. The row vectors
are of type Vector3D which provides an index operator to access the corresponding
column entry.
• The transpose of a matrix [8 marks]:
The transpose of an n x m matrix M, denoted by MT, is an m x n matrix for which the
(i,j) entry is equal to Mji. For
2 2
M 11 M 12 M 13 M 11 M 21 M 31
M3x3 = 4 M 21 M 22 M 23 5 M3x3 = 4 M 12 M 22
T M 32 5
M 31 M 32 M 33 M 13 M 23 M 33
the transpose is .

In the implementation, every column must be accessed once via calls to column(). You
may use local variables, but it is not strictly necessary.

• A test whether a matrix M is invertible [4 marks]:

A matrix is invertible if its determinant is not zero. The function does not trigger an
exception.

4
COS30008 Semester August 2024 Ms. Siti Hawa

• The inverse of a matrix [24 marks]:

The inverse of a matrix, if it exists, allows us to represent division of matrices, a
concept that is not defined for matrices. Technically, the inverse of a matrix is a
multidimensional generalization of the concept of reciprocal of a number: the product
of a number and its reciprocal is 1. The product of a matrix M with its inverse M-1 is
the identity matrix I: MM-1 = I.
For a 3 x 3 matrix M, the inverse matrix M-1 is given by
2
1 M 22M 33 — M 23M 32 M 13M 32 — M 12M 33 M 12M 23 — M 13M 22
—1
M = 4 M 23M 31 — M 21M 33 M 11M 33 — M 13M 31 M 13M 21 — M 11M 23 5
det M M 21M 32 — M 22M 31 M 12M 31 — M 11M 32 M 11M 22 — M 12M 21

In the implementation, every row must be accessed once via calls to row(). You can
declare local variables.
The implementation would need to compute the determinate of M and verify that it is
not zero. Use the given formula for calculation.
• Output operator for Matrix3x3 [14 marks]:
We can rely on the newly defined toString() method in Vector3D for this purpose.
To accommodate these operations, class Matrix3x3 is extended as follows
#pragma once

#include <ostream>

#include "Vector3D.h"

class Matrix3x3
{
private:
Vector3D fRows[3];

public:
Matrix3x3() noexcept;
Matrix3x3( const Vector3D& aRow1, const Vector3D& aRow2, const Vector3D& aRow3 ) noexcept;

Matrix3x3 operator*( const float aScalar ) const noexcept;

Matrix3x3 operator+( const Matrix3x3& aOther ) const noexcept;

Vector3D operator*( const Vector3D& aVector ) const noexcept;

static Matrix3x3 scale( const float aX = 1.0f, const float aY = 1.0f );

static Matrix3x3 translate( const float aX = 0.0f, const float aY = 0.0f );
static Matrix3x3 rotate( const float aAngleInDegree = 0.0f );

const Vector3D row( size_t aRowIndex ) const;

const Vector3D column( size_t aColumnIndex ) const;

// Problem Set 1 features

Matrix3x3 operator*( const Matrix3x3& aOther ) const noexcept;

float det() const noexcept;

Matrix3x3 transpose() const noexcept;

bool hasInverse() const noexcept;

Matrix3x3 inverse() const;

friend std::ostream& operator<<( std::ostream& aOStream, const Matrix3x3& aMatrix );

};

Do not edit the provided files. To implement the required features, create a new source file,
say Matrix3x3_PS1.cpp, and define the new features here. It is not strictly required, but
it helps to separate the definitions from the provided code. You have to include Matrix3x3.h
in the new source for the code to compile.

5
COS30008 Semester August 2024 Ms. Siti Hawa

The file Main.cpp contains a test function to check your implementation of the new matrix
features. The code sequence

void runP2()
{
gCount++;
Matrix3x3 M ( Vector3D( 25.0f, -3.0f, -8.0f ),
Vector3D( 6.0f, 2.0f, 15.0f ),
Vector3D( 11.0f, -3.0f, 4.0f ) );

cout << "Test matrix M:" << endl;

cout << M << endl;

// test multiplication

cout << "M * M = " << endl;

cout << M * M << endl;

// test determinate

cout << "det M = " << M.det() << endl;

// test hasInverse

cout << "Has M an inverse? " << (M.hasInverse() ? "Yes" : "No") << endl;

// test transpose
cout << "transpose of M:" << endl;
cout << M.transpose() << endl;

// test inverse
cout << "inverse of M:" << endl;
cout << M.inverse() << endl;

cout << "inverse of M * 45:" << endl;

cout << M.inverse() * 45.0f << endl;
}
Should produce the following output
Test matrix M:
[[25,-3,-8],[6,2,15],[11,-3,4]]
M * M =
[[519,-57,-277],[327,-59,42],[301,-51,-117]]
det M = 1222
Has M an inverse? Yes
transpose of M:
[[25,6,11],[-3,2,-3],[-8,15,4]]
inverse of M:
[[0.0434,0.0295,-0.0237],[0.1154,0.1538,-0.3462],[-0.0327,0.0344,0.0556]]
inverse of M * 45:
[[1.9517,1.3257,-1.0679],[5.1923,6.9231,-15.5769],[-1.473,1.5466,2.5041]]

Floating point values are printed with standard precision for type float.

6
COS30008 Semester August 2024 Ms. Siti Hawa

Problem 3
Carl Friedrich Gauss and Carl Gustav Jacob Jacobi invented the trapezoid formula to calculate
the area of a trapezoid in the 18th century. We can use the trapezoid formula to determine
the both the area of a polygon and the orientation of the vertices of the polygon. If the vertices
are ordered in clockwise order, then the area is negative. If the vertices are arranged in
counterclockwise order, then the area is positive as shown in Figure 1. In computer graphics,
we use the orientation of a polygon to determine whether the polygon faces the viewer or
not. The latter allows for a process called back face culling that means, we do not have to
draw the polygon. Only if the polygon faces the viewer (even partly) do we have to draw it.

-1

Figure 1: Parallelogram with signed area -16, dinosaur with signed area 38.5.

The trapezoid formula is given below

n
1X
A= (y i + y
i 1)(xi — xi 1)
2
i=1
It sums up a sequence of oriented areas of trapezoids with vertices vi and vi+1 as one of its
four edges. Please note that we need to connect the last with the first vertex. In the trapezoid
formula, this is expressed by the modulus operator (i.e., i⨁1). Naturally, it is always possible
to avoid the use of the expensive modulus operation by separating the logic into two parts
(i.e., we explicitly define the connection between the last and the first vertex).
In order the add support for the signed area calculation, start with the solution of tutorial 2
(see Canvas) and use the extended specification of class Polygon is shown below. Please
note that the extended version also includes a method to transform a given polygon. Polygon
transformation requires a 3 x 3 matrix which is multiplied with every vertex of the polygon.

7
COS30008 Semester August 2024 Ms. Siti Hawa

#pragma once

#include "Vector2D.h"
#include "Matrix3x3.h"

constexpr size_t MAX_VERTICES = 30;

class Polygon
{
private:
Vector2D fVertices[MAX_VERTICES];
size_t fNumberOfVertices;

public:

Polygon() noexcept;

void readData( std::istream& aIStream );

size_t getNumberOfVertices() const noexcept;

const Vector2D& getVertex( size_t aIndex ) const;
float getPerimeter() const noexcept;

Polygon scale( float aScalar ) const noexcept;

// Problem Set 1 extension

float getSignedArea() const noexcept;

Polygon transform( const Matrix3x3& aMatrix ) const noexcept;

};

As in 0, do not edit the provided files. Rather create a new source file, called
Polygon_PS1.cpp, to implement the new methods getSignedArea() and
transform().
There are two sample files that you can use to test your solution: Parallelogram.txt and
Data.txt. The file Main.cpp contains a test function, testProblem2(), to verify your
implementation. For the input file Parallelogram.txt the test function outputs
Signed area: -16
Signed area of rotated polygon: -16
Polygon transformation successful.

The vertices in the parallelogram are arranged in clockwise order. The area of a polygon does
not change when it is simply rotated around the origin.
For the input file Data.txt the test function outputs
Signed area: 38.5
Signed area of rotated polygon: 38.5
Polygon transformation successful.

The vertices in the T-Rex polygon are arranged in counterclockwise order.