Exam question

Question 1

Name at least 5 variables in C++

Answer 1

Bool, double, float, int, short int, long int, long long int, char

Question 2

What is int overflow?

Answer 2

Int overflow occurs when we try to store into an int variable a value that is larger than the maximum value that an int variable can take. (Usually it’s an arithmetic error resulting from operations between integers)

Question 3

Why using namespace std might not be a good idea?

Answer 3

It is bad practice as it might lead to collisions.

It imports all the identifiers belonging to std namespace in the global namesapce. This can cause ambiguity or conflicts between functions having the same name but belonging to different namespaces (or user-defined ones) which may result in “undefined behaviour”. Specifiyng to which namespace an identifier belongs using the scope operator (::) is really advisable for writing clean, unambiguous and robust error-free code.

Question 4

Which header do you have to include for the access to std::cout?

Answer 4

include <iostream></iostream>

Question 5

What would be the simplest “legal” program in C++?

Answer 5

int main() {}

Question 6

What is the difference between struct and class?

Answer 6

They are C++ keywords used for creating classes. The difference is that class has default private members, while struct has pubblic members by default.

Question 7

What does break statement do in a loop?

Answer 7

It exits the loop. (Usually it is used when we want to stop the loop given a certain condition)

Question 8

What does continue statement do in a loop?

Answer 8

It allows to skip an interation of the loop and jumps to the following one

Question 9

How do you define a static array of type int and size 7?

Answer 9

int arr[7];
(Assume that we are naming the array “arr”)

Question 10

What can you use to read interactive user input?

Answer 10

#include <iostream>
std::cin>>x;

Question 11

What is a reference?

Answer 11

A reference is an alias for a variable. It means that when a reference is initialized to a variable, we can either refer to the reference or to the original variable interchangably as they will have the same memory location

Question 12

What is the difference between passing variables to a function by reference and by value?

Answer 12

When passing a variable to a function by value, a copy of that variable (with a different memory location) is made inside of the function and so changes will be apported to the copy and not to the original variable. Instead, in passing by reference, the original variable is passed to the function (no copy is made) and so changes will be reflected on the variable itself.

Question 13

Why do we want to pass variables to function by reference rather than by pointer?

Answer 13

To avoid confusion with arrays (which are already pointers). In C++ it’s better to use reference to pass single variables and pointers for arrays.

Question 14

If the function accepts pointers (int function(int* a)), how do you pass a variable defined as int x to it?

Answer 14

function (&x)

Question 15

Why do we want to use const modifier whenever possible?

Answer 15

To prevent bugs, as an error arises when you try to change a constant variable.
Also it makes the code faster.

Question 16

If you have a pointer named p, how do you access the value it points to? What is the name of the process?

Answer 16

To access the value pointed by a pointer we need to dereference it using a * before the name of the pointer (i.e., *p)

Question 17

What value is store in the pointer variable itself?

Answer 17

Pointers store the memory address of the variable they point to

Question 18

What can you say about a function that has signature void function()?

Answer 18

Void functions do not return any value after they execution. Basically they perform a task without returning any value.

Question 19

What is an auto keyword?

Answer 19

The auto keyword is used when we do not know the returning type of a function or it is too long. By using the auto keyword we let the compiler chose the appropriate type for us.

Question 20

What is function overloading?

Answer 20

Function overloading is a feature of OOP which allows two (or more) functions to have the same name if they have a different number of parameters or same number but different type of parameters.

Question 21

What is a recursive function?

Answer 21

Recursive functions are functions thatcall themselves. They are slower than iterations, but in some cases they may help to solve the problem easily

Question 22

What is the role of a constructor in a class?

Answer 22

Constructors are special member functions with the same name of the class used (to allocate memory and) initialize class member variables

Question 23

How can we change private member variables of a class?

Answer 23

We can use class member functions or friend function. Both of them have access to private class members

Question 24

How do you create an object file with g++?

Answer 24

To create an object write g++ -c file.cpp. This will produce the file.o object file

Question 25

Why can’t you just have everything in one file?

Answer 25

Writing everything in a single file, will make the main unreadable soon. This implies that bug detection and correction becomes more difficult.

Question 26

What is the purpose of header guards?

Answer 26

Header guards are used to avoid multiple inclusions of the same header files in a program

Question 27

What does -IFOLDER_NAME mean when passed to g++?

Answer 27

It specify the folder in which g++ has to look for the header files

Question 28

Do tabs matter in Makefile?

Answer 28

Yes and they cannot be replaced by spaces

Question 29

What does $@ mean in a makefile?

Answer 29

It evaluates to the name of the target in the current rule.

Question 30

What does $^ mean in a makefile?

Answer 30

It is a macro evaluated to all the names of the dependencies, separated by a space, of the current rule.

Question 31

How to use make with a makefile that is not named Makefile?

Answer 31

We can either use -f MymakefileName or –f=MymakefileName

Question 32

What does it mean if a function is a friend of a class?

Answer 32

It means that it can have access to non-public member (private and protected) of the class it is friend with

Question 33

Why should non-class memeber operators be friends of classes?

Answer 33

So that they can access non-public members of the class (private and protected)

Question 34

Which header do I need to include to work with files?

Answer 34

include <fstream></fstream>

Question 35

How do you open a file in append mode?

Answer 35

filevar.open(“test.txt”, std::ios_base::app);

Question 36

What’s the main advantage of using template?

Answer 36

Templates allow to have the same function/class working for different data types without
the need to rewrite it for each type, but just by passing the wanted type as parameter of
the template

Question 37

Why do people usually put definition of templated functions in .hpp file?

Answer 37

Because templates get expanded to the given type(s) at compilation time (i.e., they are
instantiated at compilation time) and so the compiler must know the definition as well

Because templates need to be instantiated every time you use them, so it is faster to define them just one time in the header with their function/class.

Question 38

Except on type, what else can we template on?

Answer 38

We can template class/function on ordinary values instead of data types, but this holds
only for int, short int, long int and long long int variables

Question 39

What is template specialization?

Answer 39

It’s a feature of templates for which we can have different functionality based on type

Question 40

What is a variadic template?

Answer 40

Variadic templates allow to have a variable number of parameters. To tell the compiler
that the number of parameters can change we must specify the ellipsis (…) in the
argument

Question 41

Why using push_back for std::vector is a bad idea?

Answer 41

The push_back command appends data at the end of a vector. When the allocated memory is not enough, it will copy everything to a new location and it will take too much time in case of large amounts of data,
furthermore the memory allocated in the new location might still not be enough

Question 42

How do you pass data from std::vector to a C-style function that need a pointer?

Answer 42

You can use &myvec[0] or myvec.data()

Question 43

Why shouldn’t you use a “vector of vectors”?

Answer 43

Because vectors will not be allocated in contiguous memory regions (as we would like to
happen for matrices for example)

Question 44

How do you allocate a dynamic array in C++?

Answer 44

int* p =new int[N]; It will be dynamically allocated in the heap memory

Question 45

What’s the difference between delete and delete[]?

Answer 45

The are both used to free memory on a heap, but delete is for object only and
delete[] for arrays

Question 46

When do you need to overload assignment operator for your class?

Answer 46

Whenever dynamic memory is allocated in the class and a destructor is needed, you also need to overload the assignment operator. With the default assignment operator a shallow copy is created, which will result in an error as the destructor will be called twice on the same memory region. Assignment operator is used to make a deep copy of an object of the same class.

Question 47

When do you need to create a copy constructor for your class?

Answer 47

Whenever dynamic memory is allocated in the class and a destructor is needed, you also need to create the copy constructor. Copy constructor is used to initialize an object as a deep copy of an other object of the same class.

Question 48

When do you need to create a move constructor for your class?

Answer 48

When we want to move objects instead of copying them. In case our class has pointers, we should write our own move constructor to avoid possible problems

Question 49

What should you do if your class allocates resources, but you are sure you will
never need a copy constructor?

Answer 49

We can make it delete so that we will get a clear error message if we try to use it
Class<T>::Class ( const Class<T>& p ) = delete

Question 50

What are protected class members?

Answer 50

Protected class member are class variables that can be accessed by friends of the
class and derived classes (i.e., classes that inherit from that class)

Question 51

What are virtual functions?

Answer 51

A virtual function is a member function which is defined within a base class with the keyword virtual and (cab be) re-defined (overridden) in a derived class.

Question 52

What is an abstract class?

Answer 52

An abstract class is a class which contains (at least) a pure virtual function (declared
with =0) and from which an object cannot be created.

Question 53

Explain dynamic polymorphism.

Answer 53

Dynamic polymorphism in C++, also known as runtime polymorphism, is a mechanism in which a call to a virtual function is dynamically bound to the correct implementation at runtime based on the type of the object. When an object of a derived class is referred to as a pointer (or reference) of the base class type, the correct implementation of the
virtual function is determined at runtime based on the actual type of the object, not the type of the reference.

Question 54

Why destructor should be made virtual?

Answer 54

Destructors in base classes should be made virtual to ensure that the correct derived class’s destructor is called when a derived object is deleted through a pointer to the base class.

Question 55

What is {}()?

Answer 55

This a lambda function. Lambda function are anonymous functions (without a name)
that can be assigned to a variable or passed as an argument to another function

Question 56

What’s the general structure of a lambda function?

Answer 56

[captured parameters](formal parameters)->return_type{body of the function}(actual parameters)

actual parameters are the parameters values we want to use to evaluate the function.

Question 57

What does “mutable” keyword do in a lambda function?

Answer 57

The mutable keyword in a lambda function allows the lambda to modify inside its body the variables that are captured by value.
By default variables are immutable.

Question 58

How do you create markdown cell in Jupyter?

Answer 58

To create a markdown cell, press esc, then m

Question 59

How do you delete a cell in Jupyter?

Answer 59

To delete a cell, press esc, then dd

Question 60

How do you add a cell above the current one in Jupyter?

Answer 60

To add a cell above the current one, press esc, then a

Question 61

How do you add a cell below the current one in Jupyter?

Answer 61

To add a cell below the current one, press esc, then b

Question 62

What advantage does conda environment provide over a system-wide installation
of python?

Answer 62

Allows to have different environment with different version of the same library. We can also have different version of python.
We do this to avoid conflicts.

Question 63

What does it mean that variables are dynamically typed in python?

Answer 63

It means that the type of the variable does not need to be specified as it is checked
at running time and it is allowed to change over the lifetime of the variable. (C++ is
statically typed, variable type must be specified and it is checked at compilation
time, so without running the code)

Question 64

What’s the difference between python and C/C++ integers?

Answer 64

In python, integers are implemented as classes.
Unlike C/C++, Python has no limits on the size of the integers.

Question 65

Explain the difference between lists, sets and tuples in python.

Answer 65

Lists are mutable (can change), resizable, can store objects of different types
Set are mutable collections of unordered
unique items (no indexing is available for sets and no duplicates occur).
Tuples are immutable, you cannot change them once created (but they are faster than lists because of that)

Question 66

What does negative index mean when accessing list elements?

Answer 66

Negative index means that we are accessing the elements of a list from the end.

Question 67

How do you print the first N elements of a list?

Answer 67

The following command print the first N elements of a list called l: print(l[:N])

Question 68

How do you print every Nth element of a list?

Answer 68

We print the Nth elements of a list l using the following command: print(l[::N])

Question 69

How do you print the reverse list?

Answer 69

To print the reverse of a list l, we can do print(l[::-1])

Question 70

Give an example of a list comprehension.

Answer 70

[x**2 for x in range(10)]

Question 71

Give an example of cell and line magic commands in Jupyter.

Answer 71

%%timeit is the magic to time a cell
%timeit is the corresponding magic to time a line

Question 72

Write a “hello world” function in python

Answer 72

def hello():
print(“Hello world!”) #Note that indentation matters

Question 73

What are keyword function arguments in python and what is their advantage?

Answer 73

Keyword is an argument passed to a function which is preceded by a keyword and an equal sign: function (keyword=argument).
The advantage is that we don’t need to remember the order of the arguments and we can set a default value.

Question 74

Why shouldn’t you do from numpy import *?

Answer 74

We should’t do from numpy import * because it is very easy to have a clash of names. It is better to import the library that we need or import numpy as np to avoid the conflict of names and improve code readability.

Question 75

What is the advantage of using numpy arrays over lists? What are the dangers?

Answer 75

np arrays are faster and occupy less memory than python lists, as they are written in C. For this reason they have to be homogeneous and memory overflow is possible.

Question 76

What’s the most widely used package for plotting in python?

Answer 76

Matplotlib

Question 77

What’s the purpose of __str__ method in a python class?

Answer 77

Specifies how to convert [type cast] the class into a string.

Question 78

What package can you use in python for symbolic calculations?

Answer 78

SymPy

Question 79

What are decorators? Write an example

Answer 79

Decorators are functions used to modify other functions: they allow to add new functionality to an already defined function without changing its structure.

lo pseudocodice non lo rendera

def change_sign(func):
def wrapper(args, **kwargs):
return -func(args, **kwargs)
return wrapper

@change_sign
def my_f(x):
return 7*x

Question 80

Name two ways of calling C++ code from Python.

Answer 80

CTypes
PyBind11

Question 81

What is pandas? Name at least 5 functions from that package.

Answer 81

Pandas is a Python package specifically tailored for data analysis and machine learning purposes.
Series(list) : creates a one dimensional array, i.e. a column vector
DataFrame(dataset) : creates two dimensional arrays, i.e. tables
read_csv(“filename.csv”) : import a csv file in dataframe format
dropna(df) : delete observations with NA values from a pandas df
mean(df) : returns column averages of a pandas df
concat()
cut()