CPP v1.1

CPP
Modern C++
Object-Oriented Programming
''Combine old and newer features to get the best out of the language''
Margit ANTAL
2017
Course Content
1. Introduction to C++
Object-Oriented Programming Generic Programming

2. Classes and objects 7. Templates
3. Advanced class features 8. STL – Standard Template Library
4. Operator overloading 9. Function objects
5. Public inheritance 10. Advanced C++
6. Object relationships 11. I/O streams
2
C++ - Object-Oriented Programming
References
– Bjarne Stroustrup, Herb Sutter, C++ Core Guidelines, 2017.
– M. Gregoire, Professional C++, 3rd edition, John Wiley & Sons, 2014.
– S. Lippman, J. Lajoie, B. E. Moo, C++ Primer, 5th edition, Addison
Wesley, , 2013.
– S. Prata, C++ Primer Plus, 6th edition, Addison Wesley, 2012.
– N. Josuttis, The C++ standard library. a tutorial and reference. Pearson
Education. 2012.
– A. Williams, C++ Concurrency in Action:Practical Multithreading.
Greenwich, CT: Manning. 2012.
3
Module 1
Introduction to C++
4
Introduction to C++
Content
– History and evolution
– Overview of the key features
● New built-in types
● Scope and namespaces
● Enumerations
● Dynamic memory: new and delete
● Smart pointers: unique_ptr, shared_ptr..
● Error handling with exceptions
● References
● The const modifier
5
Introduction to C++
History and evolution
– Creator: Bjarne Stroustrup 1983
– Standards:
● The first C++ standard
– 1998 (C++98, major)
– 2003 (C++03, minor)
● The second C++ standard
– 2011 (C++11, major) – significant improvements in language and
library
– 2014 (C++14, minor)
6
Introduction to C+
History and evolution
–
source: https://isocpp.org/std/status
7
Introduction to C++
Standard library
– C++ standard library = C standard library + STL
(Standard Template Library)
– STL – designed by Alexander Stepanov, provides:
● Containers: list, vector, set, map …
● Iterators
● Algorithms: search, sort, …
8
Introduction to C++
Philosophy
– Statically typed
– General purpose
– Efficient
– Supports multiple programming styles:
● Procedural programming
● Object-oriented programming
● Generic programming
9
Introduction to C++
Portability
– Recompilation without making changes in the
source code means portability.
– Hardware specific programs are usually not
portable.
10
Introduction to C++
Creating a program
– Use a text editor to write a program and save it in a
file → source code
– Compile the source code (compiler is a program
that translates the source code to machine
language) → object code
– Link the object code with additional code (libraries)
→ executable code
11
Introduction to C++
Creating a program (using GNU C++ compiler, Unix)
– Source code: hello.cpp
– Compile: g++ -c hello.cpp
● Output: hello.o (object code)
– Compile + Link: g++ hello.cpp
● Output: a.out (executable code)
– C++ 2014: g++ hello.cpp -std=c++14
12
Introduction to C++
The first C++ program
One-line comment
//hello.cpp
#include <iostream> Preprocessor directive

using namespace std;
int main(){ The main function

cout<<”Hello”<<endl;
return 0; I/O streams
}
#include <iostream>
int main(){
std::cout<<”Hello”<<std::endl;
return 0;
}
13
Introduction to C++
Building a C++ program: 3 steps
– preprocessor (line starting with #)
– compiler
– linker
14
Introduction to C++
Most common preprocessor directives
– #include [file]
● the specified file is inserted into the code
– #define [key] [value]
● every occurrence of the specified key is replaced
with the specified value
– #ifndef [key] … #endif
● code block is conditionally included
15
Introduction to C++
Header files
– C++ header
● #include <iostream>
– C header
● #include <cstdio>
– User defined header

● #include "myheader.h"
16
Introduction to C++
Avoid multiple includes
//myheader.h
#ifndef MYHEADER_H
#define MYHEADER_H
// the contents
#endif
17
Introduction to C++
The main() function
– int main(){ … }
or
– int main( int argc, char* argv[] ){ … }
Result status The number

The arguments
of arguments
18
Introduction to C++
I/O Streams
– cout: standard output
cout<<”Hello, world!”<<endl; End of line
– cin: standard input

int i; double d;
cin >> i >> d;
19
Introduction to C++
Namespaces
– avoid naming conflicts
//my1.h //my2.h
namespace myspace1{ namespace myspace2{
void foo(); void foo();
} }
//my1.cpp //my2.cpp
#include "my1.h" #include "my2.h"
namespace myspace1{ namespace myspace2{
void foo(){ void foo(){
cout<<"myspace1::foo\n"; cout<<"myspace2::foo\n";
} }
} }
myspace1::foo() myspace2::foo()
20
Introduction to C++
Variables
– can be declared almost anywhere in your code
double d; //uninitialized
int i = 10; //initialized
21
Introduction to C++
Variable types
– short, int, long – range depends on compiler, but usually 2, 4, 4
bytes
– long long (C++11) – range depends on compiler – usually 8 bytes
– float, double, long double
– bool
– char, char16_t(C++11), char32_t(C++11), wchar_t
– auto (C++11) – the compiler decides the type automatically (auto i=7;)
– decltype(expr) (C++11)
int i=10;
decltype(i) j = 20; // j will be int
22
Introduction to C++
Variable types
#include
– <iostream>
int main(int argc, char** argv) {

cout<<"short : "<<sizeof( short)<<" bytes"<<endl;
cout<<"int : "<<sizeof( int ) <<" bytes"<<endl;
cout<<"long : "<<sizeof( long) <<" bytes"<<endl;
cout<<"long long: "<<sizeof( long long)<<" bytes"<<endl;
return 0;
}
23
Introduction to C++
● C++0x/C++11 Support in GCC
– https://gcc.gnu.org/projects/cxx0x.html
● Online C++ compilers:

– http://isocpp.org/blog/2013/01/online-c-compilers
– http://www.tutorialspoint.com/compile_cpp11_online.php
24
Introduction to C++
C enumerations (not type-safe)
– always interpreted as integers →
● you can compare enumeration values from completely
different types
enum Fruit{ apple, strawberry, melon};
enum Vegetable{ tomato, cucumber, onion};
void foo(){
if( tomato == apple){
cout<<"Hurra"<<endl;
}
}
25
Introduction to C++
C++ enumerations (type-safe)
enum class Mark {

Undefined, Low, Medium, High
};
Mark myMark( int value ){

switch( value ){
case 1: case2: return Mark::Low;
case 3: case4: return Mark::Medium;
case 5: return Mark::High;
default:
return Mark::Undefined;
}
}
26
Introduction to C++
Range-based for loop
int elements[]={1,2,3,4,5};
for( auto& e: elements){

cout<<e<<endl;
}
27
Introduction to C++
The std::array
– replacement for the standard C-style array
– cannot grow or shrink at run time
#include <iostream>
#include <array>
int main() {
array<int, 5 > arr = {10, 20, 30, 40, 50};
cout << "Array size = " << arr.size() << endl;
for(int i=0; i<arr.size(); ++i){
cout<<arr[ i ]<<endl;
}
}
28
Introduction to C++
Pointers and dynamic memory
– compile time array
int ctarray[ 3 ]; //allocated on stack
– run time array

int * rtarray = new int[ 3 ]; //allocated on heap
Stack Heap
rtarray
29
Introduction to C++
Dynamic memory management
– allocation
int * x = new int;
int * t = new int [ 3 ];
– deletion
delete x;
delete [] t;
30
Introduction to C++
Strings
– C-style strings:
● array of characters
● '\0' terminated
● functions provided in <cstring>
– C++ string
● described in <string>
string firstName = "John"; string lastName = "Smith";
string name = firstName+ " "+ lastName; cout<<name<<endl;
31
Introduction to C++
References
– A reference defines an alternative name (alias) for an object.
– A reference must be initialized.
– Defining a reference = binding a reference to its initializer
int i = 10;
int &ri = i; //OK ri refers to (is another name for) i
int &ri1; //ERROR: a reference must be initialized
32
Introduction to C++
Operations on references
– the operation is always performed on the referred object
int i = 10;
int &ri = i;
++ri;
cout<<i<<endl; // outputs 11
++i;
cout<<ri<<endl; // outputs 12
33
Introduction to C++
References as function parameters
– to permit pass-by-reference:
● allow the function to modify the value of the parameter
● avoid copies
void inc(int &value) bool isShorter(const string &s1,
{ const string &s2)
value++; {
} return s1.size() < s2.size();
}
usage: usage:
int x = 10; string str1 ="apple";
inc( x ); string str2 ="nut";
cout<<str1<<"<"<<str2<<": " <<
isShorter(str1, str2);
34
Introduction to C++
Exceptions
– Exception = unexpected situation
– Exception handling = a mechanism for dealing with problems
● throwing an exception – detecting an unexpected situation
● catching an exception – taking appropriate action
35
Introduction to C++
Exceptions
#include <iostream>
#include
– <stdexcept>
double divide( double m, double n){

if( n == 0 ){
throw std::exception();
}else{
return m/n;
}
}
int main() {
try{
cout<<divide(1,0)<<endl;
}catch( const exception& e){
cout<<"Exception was caught!"<<endl;
}
}
36
Introduction to C++
The const modifier
– Defining constants
const int N =10;
int t[ N ];
– Protecting a parameter
void sayHello( const string& who){
cout<<"Hello, "+who<<endl;
who = "new name"; Compiler error
}
37
Uniform initialization (C++ 11)
brace-init
int n{2};
string s{"alma"};
map<string,string> m {
{"England","London"},
{"Hungary","Budapest"},
{"Romania","Bucharest"}
};
struct Person{
string name;
int age;
};
Person p{"John Brown", 42};
38
Introduction to C++
Using the standard library
#include <string>
–
#include <vector>
#include <iostream>
const int N =10;
int main() {
int t[ N ];
vector<string> fruits = {"apple","melon"};
fruits.push_back("pear");
fruits.push_back("nut");
–// Iterate over the elements in the vector and print them
for (auto it = fruits.cbegin();
it != fruits.cend(); ++it) {
cout << *it << endl;
}
//Print the elements again using C++11 range-based for loop
for (auto& str : fruits)
cout << str << endl;
return 0;
}
39
Introduction to C++
Programming task:
– Write a program that reads one-word strings from the
standard input, stores them and finally prints them on the
standard output
– Sort the container before printing
● use the sort algorithm
#include <algorithm>
...
vector<string> fruits;
...
sort(fruits.begin(),fruits.end());
–
40
Module 2
Classes and Objects
41
Object-Oriented Programming (OOP)
Content
– Classes and Objects
– Advanced Class Features
– Operator overloading
– Object Relationships
– Abstraction
42
OOP: Classes and Objects
Content
– Members of the class. Access levels. Encapsulation.
– Class: interface + implementation
– Constructors and destructors
– const member functions
– Constructor initializer
– Copy constructor
– Object's lifecycle
43
OOP: Types of Classes
Types of calsses:
– Polymorphic Classes:
● Shape, exception, ...
– Value Classes: Margit Antal
● int, complex<double>, ...
– RAAI (Resource Acquisition Is Initialization) Classes:

(RAAI meaning: resource lifetime → object lifetime)
● thread, unique_ptr, ...
44
OOP: Classes and objects
Class = Type ( Data + Operations)
– Members of the class
– Data:
● data members (properties)
Margit Antal
– Operations:
● methods (behaviors)
– Each member is associated with an access level:
● private -
● public +
● protected #
45
Object = Instance of a class
– An employee object: Employee emp;
● Properties are the characteristics that describe an
object.
– What makes this object different?
● id, firstName, lastName, salary, hired
● Behaviors answer the question:
– What can we do to this object?
● hire(), fire(), display(), get and set data
members
46
class TYPES
Encapsulation
Employee
– an object encapsulates data and functionality. - m Id: int
- m FirstNam e: string
- m LastNam e: string
- m Salary: int
- bHired: b ool
Functionality + Em ployee()
+ display() : void {query}
+ hire() : void
+ fire() : void
+ setFirstNam e(strin g) : void
Data +
+
setLastNam e(strin g) : void
setId(int) : vo id
+ setSalary(int) : vo id
+ getFirstNam e() : string {query}
+ getLastNam e() : string {query}
+ getSalary() : int {query}
+ getIsHired() : bool {query}
+ getId() : i nt {query}
47
Class creation
– class declaration - interface
● Employee.h
– class definition – implementation

● Employee.cpp
48
Employee.h
class Employee{
public: Methods' declaration
Employee();
void display() const;
void hire();
void fire();
// Getters and setters
void setFirstName( string inFirstName );
void setLastName ( string inLastName );
void setId( int inId );
void setSalary( int inSalary );
string getFirstName() const;
string getLastName() const;
int getSalary() const;
bool getIsHired() const;
int getId() const;
private:
int mId; Data members
string mFirstName;
string mLastName;
int mSalary;
bool bHired;
};
49
The Constructor and the object's state
– The state of an object is defined by its data members.
– The constructor is responsible for the initial state of the object
Employee :: Employee() : mId(-1),
mFirstName(""),
mLastName(""), Members are initialized
mSalary(0), through the
bHired(false){ constructor initializer list
}
Employee :: Employee() {
mId = -1; Members are assigned
mFirstName="";
mLastName="";
mSalary =0; Only constructors can
bHired = false; use this initializer-list
}
syntax!!!
50
Constructors
– responsibility: data members initialization of a class object
– invoked automatically for each object
– have the same name as the class
– have no return type
– a class can have multiple constructors (function overloading)
– may not be declared as const
● constructors can write to const objects
51
Member initialization (C++11)
class C class C
{ {
string s ("abc"); string s;
double d = 0; double d;
char * p {nullptr}; char * p;
int y[5] {1,2,3,4}; int y[5];
public: public:
C(); C() : s("abc"),
}; d(0.0),
p(nullptr),
y{1,2,3,4} {}
};
Compiler
52
Defining a member function
– Employee.cpp
– A const member function cannot change the object's
state, can be invoked on const objects
void Employee::hire(){
bHired = true;
}
string Employee::getFirstName() const{

return mFirstName;
}
53
Defining a member function
void Employee::display() const {
cout << "Employee: " << getLastName() << ", "
<< getFirstName() << endl;
cout << "-------------------------" << endl;
cout << (bHired ? "Current Employee" :
"Former Employee") << endl;
cout << "Employee ID: " << getId() << endl;
cout << "Salary: " << getSalary() << endl;
cout << endl;
}
54
TestEmployee.cpp
– Using const member functions
void foo( const Employee& e){
e.display(); // OK. display() is a const member function
e.fire(); // ERROR. fire() is not a const member function
}
int main() {
Employee emp;
emp.setFirstName("Robert");
emp.setLastName("Black");
emp.setId(1);
emp.setSalary(1000);
emp.hire();
emp.display();
foo( emp );
return 0;
}
55
TestEmployee.cpp
– Using const member functions
void foo( const Employee& e){
e.display(); // OK. display() is a const member function
e.fire(); // ERROR. fire() is not a const member function
}
int main() {
Employee emp;
emp.setFirstName("Robert");
emp.setLastName("Black");
emp.setId(1);
emp.setSalary(1000);
emp.hire();
emp.display();
foo( emp );
return 0;
}
56
Interface: Employee.h Implementation: Employee.cpp
#ifndef EMPLOYEE_H #include "Employee.h"
#define EMPLOYEE_H
#include <string> Employee::Employee() :

using namespace std; mId(-1),
mFirstName(""),
class Employee{ mLastName(""),
public: mSalary(0),
Employee(); bHired(false){
//... }
protected:
int mId; string Employee::getFirstName() const{
string mFirstName; return mFirstName;
string mLastName; }
int mSalary; /
bool bHired; / ...
};
#endif
57
Object life cycles:
– creation
– assignment
– destruction
58
Object creation:
int main() {
Employee emp;
emp.display();
Employee *demp = new Employee();

demp->display(); object's
// .. lifecycle
delete demp;
return 0;
}
– when an object is created,

● one of its constructors is executed,
● all its embedded objects are also created
59
Object creation – constructors:
– default constructor (0-argument constructor)
Employee :: Employee() : mId(-1), mFirstName(""),
mLastName(""), mSalary(0), bHired(false){
}
Employee :: Employee() {
}
– when you need ● memory allocation

● Employee employees[ 10 ]; ● constructor call on
● vector<Employee> emps; each allocated object
60
Object creation – constructors:
– Compiler-generated default constructor
class Value{
public:
void setValue( double inValue);
double getValue() const;
private:
double value;
};
– if a class does not specify any constructors, the compiler will generate
one that does not take any arguments
61
Constructors: default and delete specifiers (C++ 11)
class X{
int i = 4;
int j {5};
public:
X(int a) : i{a} {} // i = a, j = 5
X() = default; // i = 4, j = 5
};
Explicitly forcing the automatic generation of a default constructor by the compiler.
62
Constructors: default and delete specifiers (C++ 11)
class X{ class X{
public: public:
X( int x ){} X( int x ){}
}; X( double ) = delete;
};
X x1(10); //OK X x1(10); //OK
X x2(3.14); //OK X x2(3.14); //ERROR
double → int conversion
Inhibiting the automatic generation of a default constructor by the compiler.
63
Best practice: always provide default values for
members! C++ 11
struct Point{ int main() {
int x, y; Foo f;
Point ( int x = 0, int y = 0 ): x(x), y(y){} f.print();
}; return 0;
class Foo{ }
int i {};
double d {};
char c {};
Point p {};
OUTPUT:
public: i: 0
void print(){ d: 0
cout <<"i: "<<i<<endl; c:
cout <<"d: "<<d<<endl;
cout <<"c: "<<c<<endl; p: 0, 0
cout <<"p: "<<p.x<<", "<<p.y<<endl;
}
};
64
Constructor initializer
class ConstRef{
public:
ConstRef( int& );
private:
int mI;
const int mCi;
int& mRi;
};
ConstRef::ConstRef( int& inI ){

mI = inI; //OK
mCi = inI; //ERROR: cannot assign to a const
mRi = inI; //ERROR: uninitialized reference member
}
ConstRef::ConstRef( int& inI ): mI( inI ), mCi( inI ), mRi( inI ){}
ctor initializer
65
Constructor initializer
– data types that must be initialized in a ctor-initializer
● const data members
● reference data members
● object data members having no default
constructor
● superclasses without default constructor
66
A non-default Constructor
Employee :: Employee( int inId, string inFirstName,
string inLastName,
int inSalary, int inHired) :
mId(inId), mFirstName(inFirstName),
mLastName(inLastName), mSalary(inSalary),
bHired(inHired){
}
67
Copy Constructor
Employee emp1(1, "Robert", "Black", 4000, true);
– called in one of the following cases:
● Employee emp2( emp1 ); //copy-constructor called
● Employee emp3 = emp2; //copy-constructor called
● void foo( Employee emp );//copy-constructor called
– if you don't define a copy-constructor explicitly, the compiler
creates one for you
● this performs a bitwise copy
68
//Stack.h //Stack.cpp
#ifndef STACK_H #include "Stack.h"

#define STACK_H
Stack::Stack( int inCapacity ){
class Stack{ mCapacity = inCapacity;
public: mElements = new double [ mCapacity ];
Stack( int inCapacity ); mTop = mElements;
void push( double inDouble ); }
double top() const;
void pop(); void Stack::push( double inDouble ){
bool isFull() const; if( !isFull()){
bool isEmpty()const; *mTop = inDouble;
mTop++;
private: }
int mCapacity; }
double * mElements;
double * mTop;
};
#endif /* STACK_H */
69
//TestStack.cpp
#include "Stack.h"
s1: Stack
int main(){ mCapacity: 3

Stack s1(3); mElements
Stack s2 = s1;
s1.push(1); mTop
s2.push(2);
0
cout<<"s1: "<<s1.top()<<endl;
cout<<"s2: "<<s2.top()<<endl; 1
}
s2: Stack 2
mCapacity: 3
mElements
mTop
70
Copy constructor: T ( const T&)
//Stack.h //Stack.cpp
#ifndef STACK_H #include "Stack.h"

#define STACK_H
Stack::Stack( const Stack& s ){
class Stack{ mCapacity = s.mCapacity;
public: mElements = new double[ mCapacity ];
//Copy constructor int nr = s.mTop - s.mElements;
Stack( const Stack& ); for( int i=0; i<nr; ++i ){
private: mElements[ i ] = s.mElements[ i ];
int mCapacity; }
double * mElements; mTop = mElements + nr;
double * mTop; }
};
#endif /* STACK_H */
71
s1: Stack
//TestStack.cpp mCapacity: 3
#include "Stack.h"
mElements
int main(){ mTop
Stack s1(3);
Stack s2 = s1; 0
s1.push(1);
s2.push(2);
1
cout<<"s1: "<<s1.top()<<endl;
cout<<"s2: "<<s2.top()<<endl; s2: Stack 2
}
mCapacity: 3
mElements
mTop
0
1
2
72
Destructor
– when an object is destroyed:
● the object's destructor is automatically invoked,
● the memory used by the object is freed.
– each class has one destructor
– usually place to perform cleanup work for the object
– if you don't declare a destructor → the compiler will generate
one, which destroys the object's member
73
Destructor
– Syntax: T :: ~T();
Stack::~Stack(){
if( mElements != nullptr ){
delete[] mElements;
mElements = nullptr;
}
}
{ // block begin
Stack s(10); // s: constructor
Stack* s1 = new Stack(5);// s1: constructor
s.push(3);
s1->push(10);
delete s1; //s1: destructor
s.push(16);
} // block end //s: destructor
74
Default parameters
– if the user specifies the arguments → the defaults are ignored
– if the user omits the arguments → the defaults are used
– the default parameters are specified only in the method declaration
(not in the definition)
//Stack.h //TestStack.cpp
class Stack{
public: Stack s1(3); //capacity: 3
Stack( int inCapacity = 5 ); Stack s2; //capacity: 5
.. Stack s3( 10 ); //capacity: 10
};
//Stack.cpp
Stack::Stack( int inCapacity ){
mCapacity = inCapacity;
mElements = new double [ mCapacity ];
mTop = mElements;
}
75
The this pointer
– every method call passes a pointer to the object for which it
is called as hidden parameter having the name this
– Usage:
● for disambiguation
Stack::Stack( int mCapacity ){
this → mCapacity = mCapacity;
//..
}
76
Programming task [Prata]
class Queue
{
enum {Q_SIZE = 10};
private:
// private representation to be developed later
public:
Queue(int qs = Q_SIZE); // create queue with a qs limit
~Queue();
bool isempty() const;
bool isfull() const;
int queuecount() const;
bool enqueue(const Item &item); // add item to end
bool dequeue(Item &item); // remove item from front
};
77
Programming task [Prata]
class Queue
{
private:
// class scope definitions
// Node is a nested structure definition local to this class

struct Node { Item item; struct Node * next;};
enum {Q_SIZE = 10};
// private class members

Node * front; // pointer to front of Queue
Node * rear; // pointer to rear of Queue
int items; // current number of items in Queue
const int qsize; // maximum number of items in Queue
};
78
Module 3
Advanced Class Features
79
OOP: Advanced class features
Content
– Inline functions
– Stack vs. Heap
– Array of objects vs. array of pointers
– Passing function arguments
– Static members
– Friend functions, friend classes
– Nested classes
– Move semantics (C++11)
80
Inline functions
– designed to speed up programs (like macros)
– the compiler replaces the function call with the
function code (no function call!)
– advantage: speed
– disadvantage: code bloat
● ex. 10 function calls → 10 * function's size
81
How to make a function inline?
– use the inline keyword either in function declaration
or in function definition
– both member and standalone functions can be inline
– common practice:
● place the implementation of the inline function
into the header file
– only small functions are eligible as inline
– the compiler may completely ignore your request
82
inline function examples
inline double square(double a){
return a * a;
}
class Value{
int value;
public:
inline int getValue()const{ return value; }
inline void setValue( int value ){

this->value = value;
}
};
83
– Stack vs. Heap
– Heap – Dynamic allocation
void draw(){
Point * p = new Point();
p->move(3,3);
//...
delete p;
}
– Stack – Automatic allocation
void draw(){
Point p;
p.move(6,6);
//...
}
84
Array of objects
class Point{
int x, y;
public:
Point( int x=0, int y=0);
//...
};
What is the difference between these two arrays?
Point * t1 = new Point[ 4]; Point t1[ 4];
t1 :Point :Point :Point :Point

x: 0 x: 0 x: 0 x: 0
y: 0 y: 0 y: 0 y: 0
85
Array of pointers
Point ** t2 = new Point*[ 4 ];
for(int i=0; i<4; ++i ){
t2[i] = new Point(0,0);
}
for( int i=0; i<4; ++i ){
cout<<*t2[ i ]<<endl;
}
t2
:Point :Point :Point :Point

x: 0 x: 0 x: 0 x: 0
y: 0 y: 0 y: 0 y: 0
86
Static members:
● static methods
● static data
– Functions belonging to a class scope which don't access
object's data can be static
– Static methods can't be const methods (they do not access
object's state)
– They are not called on specific objects  they have no this
pointer
87
– Static members initializing static class member
//Complex.h
class Complex{
public: //Complex.cpp
Complex(int re=0, int im=0);
static int getNumComplex(); int Complex::num_complex = 0;
// ...
private: int Complex::getNumComplex(){
static int num_complex; return num_complex;
double re, im; }
};
Complex::Complex(int re, int im){
this->re = re;
this->im = im;
instance counter ++num_complex;
}
88
– Static method invocation
elegant
Complex z1(1,2), z2(2,3), z3;

cout<<"Number of complexs:"<<Complex::getNumComplex()<<endl;
cout<<"Number of complexes: "<<z1.getNumComplex()<<endl;
non - elegant
89
Complex z1(1,2), z2(2,3), z3;
re: 1 num_complex: 3
im: 2
Only one copy of
the static member
re: 2 Each object has

im: 3 its own re and im
re: 0
im: 0
90
– Classes vs. Structs
● default access specifier
– class: private
– struct: public
● class: data + methods, can be used polimorphically
● struct: mostly data + convenience methods
91
– Classes vs. structures
class list{
private:
struct node
{
node *next;
int val;
node( int val = 0, node * next = nullptr):val(val), next(next){}
};
node * mHead;
public:
list ();
~list ();
void insert (int a);
void printAll()const;
};
92
– Passing function arguments
● by value
– the function works on a copy of the variable
● by reference
– the function works on the original variable, may modify it
● by constant reference
– the function works on the original variable, may not modify
(verified by the compiler)
93
– Passing function arguments passing primitive values
void f1(int x) {x = x + 1;}

void f2(int& x) {x = x + 1;}
void f3(const int& x) {x = x + 1;}//!!!!
void f4(int *x) {*x = *x + 1;}
int main(){
int y = 5;
f1(y);
f2(y);
f3(y);
f4(&y);
return 0;
}
94
– Passing function arguments passing objects
void f1(Point p); copy constructor will be used on the argument

void f2(Point& p);
void f3(const Point& p); only const methods of the class can be
void f4(Point *p); invoked on this argument
int main(){
Point p1(3,3);
f1(p1);
f2(p1);
f3(p1);
f4(&p1);
return 0;
}
95
– friend functions, friend classes, friend member
functions
● friends are allowed to access private members of a class
● Use it rarely
– operator overloading
96
– friend vs. static functions
class Test{
private:
int iValue;
static int sValue;
public:
Test( int in ):iValue( in ){}
void print() const;
static void print( const Test& what );
friend void print( const Test& what );
};
97
– friend vs. static functions
int Test :: sValue = 0;
void Test::print() const{

cout<<"Member: "<<iValue<<endl;
}
void Test::print( const Test& what ){

cout<<"Static: "<<what.iValue<<endl;
}
void print( const Test& what ){

cout<<"Friend: "<<what.iValue<<endl;
}
int main() {
Test test( 10 );
test.print();
Test::print( test );
print( test );
}
98
– friend class vs. friend member function
class List{ class ListElement{
private: private:
ListElement * head; int key;
public: ListElement * next;
bool find( int key ); friend class List;
… ...
}; };
class ListElement{
private:
int key;
ListElement * next;
friend class List::find( int key);
...
};
99
– Returning a reference to a const object
// version 1
vector<int> Max(const vector<int> & v1, const vector<int> & v2)
{
if (v1.size() > v2.size()) Copy
return v1;
else constructor
return v2; invocation
}
// version 2
const vector<int> & Max(const vector<int> & v1, const vector<int> & v2)
{
if (v1.size() > v2.size())
return v1; More
else efficient
return v2;
}
The reference should be to

a non-local object
100
– Nested classes
● the class declared within another class is called a
nested class
● usually helper classes are declared as nested
// Version 1
class Queue
{
private:
// Node is a nested structure definition local to this class
struct Node {Item item; struct Node * next;};
...
};
101
– Nested classes [Prata] Node visibility!!!
// Version 2
class Queue
{
// Node is a nested class definition local to this class
class Node
{
public:
Item item;
Node * next;
Node(const Item & i) : item(i), next(0) { }
};
//...
};
102
– Nested classes
● a nested class B declared in a private section of a class A:
– B is local to class A (only class A can use it)
● a nested class B declared in a protected section of a class A:

– B can be used both in A and in the derived classes of A
● a nested class B declared in a public section of a class A:

– B is available to the outside world ( A :: B b;)
103
– Features of a well-behaved C++ class
– implicit constructor
● T :: T(){ … }
– destructor
● T :: ~T(){ … }
– copy constructor
● T :: T( const T& ){ … }
– assignment operator (see next module)

● T :: operator=( const T& ){ … }
104
– Move Semantics (C++11)
class string{ string :: string(const char* p){
char* data; size_t size = strlen(p) + 1;
public: data = new char[size];
string( const char* ); memcpy(data, p, size);
string( const string& ); }
~string(); string :: string(const string& that){
}; size_t size = strlen(that.data) + 1;
data = new char[size];
memcpy(data, that.data, size);
}
string :: ~string(){
delete[] data;
}
105
– Constructor delegation (C++11)
// C++03
class A
{
void init() { std::cout << "init()"; }
void doSomethingElse() { std::cout << "doSomethingElse()\n"; }
public:
A() { init(); }
A(int a) { init(); doSomethingElse(); }
};
// C++11
class A
{
void doSomethingElse() { std::cout << "doSomethingElse()\n"; }
public:
A() { ... }
A(int a) : A() { doSomethingElse(); }
};
106
– Move Semantics (C++11): lvalue, rvalue
string a(x); // Line 1
string b(x + y); // Line 2
string c(function_returning_a_string()); // Line 3
– lvalue: real object having an address
● Line 1: x
– rvalue: temporary object – no name
● Line 2: x + y
● Line 3: function_returning_a_string()
107
– Lvalues:
● Refer to objects accessible at more than one point in a source
code
– Named objects
– Objects accessible via pointers/references
● Lvalues may not be moved from
– Rvalues:
● Refer to objects accessible at exactly one point in source code
– Temporary objects (e.g. by value function return)
● Rvalues may be moved from
108
int x;
– Lvalue x = 10; Rvalue
109
– Move Semantics (C++11): rvalue reference, move
constructor
// string&& is an rvalue reference to a string
string :: string(string&& that){
data = that.data;
that.data = nullptr;
}
● Move constructor
● Shallow copy of the argument
● Ownership transfer to the new
object
110
– Copy constructor vs. move constructor
● Copy constructor: deep copy
● Move constructor: shallow copy + ownership transfer
// constructor
string s=”apple”;
// copy constructor: s is an lvalue
string s1 = s;
// move constructor: right side is an rvalue
string s2 = s + s1;
111
– Passing large objects
//– C++98 // C++11
// avoid expense copying // move semantics
–
void makeBigVector(vector<int>& out) vector<int> makeBigVector()
{ {
}
–
... …
vector<int> v; }
–
makeBigVector( v ); auto v = makeBigVector();
– All STL classes have been extended to support move

semantics
– The content of the temporary created vector is moved in v
(not copied)
112
– Rvalues, lvalues, &&
class A {
public:
static A inst;
Reference to a static variable → lvalue
static A& getInst() {return inst;}
static A getInstCopy(){return inst;} A temporary copy of instance → rvalue

}; 1. ok - we've fetched a reference to the static
instance variable
2. ERROR - can't assign a reference to an
A& inst1 = A :: getInst(); //1 rvalue reference
A&& inst2 = A :: getInst(); //2 3. getInst() is an lvalue reference, we assign a
new value to it
A::getInst() = A(); //3 4. ok - we've fetched a copy of the instance
A inst4 = A :: getInstCopy();//4 5. ERROR- can't assign a reference to a
A& inst5 = A :: getInstCopy();//5 temporary (an rvalue)
A&& inst6 = A :: getInstCopy();//6 6. ok - we've assigned an rvalue reference to
the temporary copy that was made of the
instance
http://geant4.web.cern.ch/geant4/collaboration/c++11_guidelines.pdf
113
Module 4
Operator overloading
114
OOP: Operator overloading
Content
● Objectives
● Types of operators
● Arithmetic operators
● Increment/decrement
● Inserter/extractor operators
● Assignment operator
● Index operator
● Relational and equality operators
● Conversion operators
115
Objective
– To make the class usage easier, more intuitive
● the ability to read an object using the extractor operator (>>)
– Employee e1; cin >> e;
● the ability to write an object using the inserter operator (<<)
– Employee e2; cout<<e<<endl;
● the ability to compare objects of a given class
– cout<< ((e1 < e2) ? "less" : "greater");
Operator overloading: a service to the clients of the class
116
Limitations
– You cannot add new operator symbols. Only the existing operators can
be redefined.
– Some operators cannot be overloaded:
● . (member access in an object)
● ::(scope resolution operator)
● sizeof
● ?:
– You cannot change the arity (the number of arguments) of the operator
– You cannot change the precedence or associativity of the operator
117
How to implement?
– write a function with the name operator<symbol>
– alternatives:
● method of your class
● global function (usually a friend of the class)
http://en.cppreference.com/w/cpp/language/operators
118
– There are 3 types of operators:
● operators that must be methods (member functions)
– they don't make sense outside of a class:
● operator=, operator(),operator[], operator->
● operators that must be global functions
– the left-hand side of the operator is a variable of different type than your class:
operator<<, operator>>
● cout << emp;
● cout: ostream
● emp: Employee
● operators that can be either methods or global functions
– Gregoire: “Make every operator a method unless you must make it a global
function.”
119
– Choosing argument types:
● value vs. reference
– Prefer passing-by-reference instead of passing-by-value.
● const vs. non const
– Prefer const unless you modify it.
– Choosing return types
● you can specify any return type, however
– follow the built-in types rule:
● comparison always return bool
● arithmetic operators return an object representing the result of the arithmetic
120
– The Complex class: Complex.h, Complex.cpp
#ifndef COMPLEX_H #include "Complex.h"

#define COMPLEX_H #include <iostream>
Class Complex{
public: Complex::Complex( double re, double im):re( re), im(im) {}
Complex(double, double );
void setRe( double ); void Complex::setRe( double re){this->re = re;}
void setIm( double im);
double getRe() cons; void Complex::setIm( double im){ this->im = im;}
double getIm() const;
void print() const; double Complex::getRe() const{ return this->re;}
Private:
double re, im; double Complex::getIm() const{ return this->im;}
}
#endif
void Complex::print() const{ cout<<re<<"+"<<im<<"i";}
121
– Arithmetic operators (member or standalone func.)
● unary minus
● binary minus
Complex Complex::operator-() const{
Complex temp(-this->re, -this->im);
return temp;
}
Complex Complex::operator-( const Complex& z) const{

Complex temp(this->re - z.re, this->im- z.im);
return temp;
}
122
– Arithmetic operators (member or standalone func.)
● unary minus
● binary minus
Complex operator-( const Complex& z ){
Complex temp(-z.getRe(), -z.getIm());
return temp;
}
Complex operator-( const Complex& z1, const Complex& z2 ){

Complex temp(z1.getRe()-z2.getRe(), z1.getIm()-z2.getIm());
return temp;
}
123
– Increment/Decrement operators
● postincrement: int i = 10; int j = i++; // j → 10
●
preincrement: int i = 10; int j = ++i; // j → 11
● The C++ standard specifies that the prefix increment and decrement
return an lvalue (left value).
124
– Increment/Decrement operators (member func.)
Complex& Complex::operator++(){ //prefix

(this->re)++;
(this->im)++;
return *this; Which one is more efficient?
} Why?
Complex Complex::operator++( int ){ //postfix

Complex temp(*this);
(this->re)++;
(this->im)++;
return temp;
}
125
– Inserter/Extractor operators (standalone func.)
//complex.h
class Complex {
public:
friend ostream& operator<<( ostream& os, const Complex& c);
friend istream& operator>>( istream& is, Complex& c);
//...
};
//complex.cpp
ostream& operator<<( ostream& os, const Complex& c){
os<<c.re<<"+"<<c.im<<"i";
return os;
}
istream& operator>>( istream& is, Complex& c){

is>>c.re>>c.im;
return is;
}
126
– Inserter/Extractor operators
– Syntax:
ostream& operator<<( ostream& os, const T& out)
istream& operator>>( istream& is, T& in)
– Remarks:
● Streams are always passed by reference
● Q: Why should inserter operator return an ostream&?
● Q: Why should extractor operator return an istream&?
127
– Inserter/Extractor operators
– Usage:
Complex z1, z2;
cout<<"Read 2 complex number:";
//Extractor
cin>>z1>>z2;
//Inserter
cout<<"z1: "<<z1<<endl;
cout<<"z2: "<<z2<<endl;
cout<<"z1++: "<<(z1++)<<endl;
cout<<"++z2: "<<(++z2)<<endl;
128
– Assignment operator
● Q: When should be overloaded?
● A: When bitwise copy is not satisfactory (e.g. if you have
dynamically allocated memory 
– when we should implement the copy constructor and the
destructor too).
– Ex. our Stack class
129
– Assignment operator (member func.)
● Copy assignment
● Move assignment (since C++11)
130
– Copy assignment operator (member func.)
● Syntax: X& operator=( const X& rhs);
● Q: Is the return type necessary?
– Analyze the following example code
Complex z1(1,2), z2(2,3), z3(1,1);

z3 = z1;
z2 = z1 = z3;
131
– Copy assignment operator example
Stack& Stack::operator=(const Stack& rhs) {
if (this != &rhs) {
//delete lhs – left hand side
delete [] mElements;
mCapacity = 0;
melements = nullptr; // in case next line throws
//copy rhs
mCapacity = rhs.mCapacity;
mElements = new double[ mCapacity ];
int nr = rhs.mTop - rhs.mElements;
std::copy(rhs.mElements, rhs.mElements+nr, mElements);
mTop = mElements + nr;
}
return *this;
}
132
– Copy assignment operator vs Copy constructor
Complex z1(1,2), z2(3,4); //Constructor

Complex z3 = z1; //Copy constructor
Complex z4(z2); //Copy constructor
z1 = z2; //Copy assignment operator
133
– Move assignment operator (member func.)
● Syntax: X& operator=( X&& rhs) noexcept;
● When it is called?
Complex z1(1,2), z2(3,4); //Constructor
Complex z4(z2); //Copy constructor
z1 = z2; //Copy assignment operator
Complex z3 = z1 + z2; //Move constructor
z3 = z1 + z1; //Move assignment
134
– Move assignment operator example
Stack& Stack::operator=(Stack&& rhs) noexcept {
if (this != &rhs) {
//delete lhs – left hand side
delete [] mElements;
//move rhs to this
mCapacity = rhs.mCapacity;
mTop = rhs.mTop;
Melements = rhs.mElements;
//leave rhs in valid state
mElements = nullptr;
rhs.mCapacity = 0;
rhs.mTop = 0;
}
return *this;
}
135
– Subscript operator: needed for arrays (member func.)
– Suppose you want your own dynamically allocated C-style
array  implement your own CArray
#ifndef CARRAY_H
#define CARRAY_H
class CArray{
public:
CArray( int size = 10 );
~Carray();
CArray( const Carray&) = delete;
CArray& operator=( const Carray&) = delete;
double& operator[]( int index );
double operator[]( int index ) const; Provides read-only access
private:
double * mElems;
int mSize;
};
#endif /* ARRAY_H */`
“If the value type is known to be a built-in type, the const variant should return by value.”
http://en.cppreference.com/w/cpp/language/operators. 136
– Implementation
CArray::CArray( int size ){ double CArray::operator[](
if( size < 0 ){ int index ) const{
size = 10; if( index <0 || index >= mSize ){
} throw out_of_range("");
this->mSize = size; }
this->mElems = new double[ mSize ]; return mElems[ index ];
} }
CArray::~CArray(){
if( mElems != nullptr ){
delete[] mElems;
mElems = nullptr;
}
}
double& CArray::operator[]( int index ){

if( index <0 || index >= mSize ){ #include<stdexcept>
throw out_of_range("");
}
return mElems[ index ];
}
137
– const vs non-const [] operator
void printArray(const CArray& arr, size_t size) {
for (size_t i = 0; i < size; i++) {
cout << arr[i] << "" ;
// Calls the const operator[] because arr is
// a const object.
}
cout << endl;
}
CArray myArray;
for (size_t i = 0; i < 10; i++) {
myArray[i] = 100;
// Calls the non-const operator[] because
// myArray is a non-const object.
}
printArray(myArray, 10);
138
– Non-integral array indices: associative array
– ( Key, Value )
– Ex: Key → string, Value → double
● Homework
139
– Relational and equality operators
● used for search and sort
● the container must be able to compare the stored objects
bool operator ==( const Point& p1, const Point& p2){
return p1.getX() == p2.getX() && p1.getY() == p2.getY();
}
bool operator <( const Point& p1, const Point& p2){

return p1.distance(Point(0,0)) < p2.distance(Point(0,0));
}
vector<Point> v; //...
set<Point> p;
sort(v.begin(), v.end());
140
– The function call operator ()
– Instances of classes overloading this operator behave as
functions too (they are function objects = function + object)
#ifndef ADDVALUE_H #include "AddValue.h"
#define ADDVALUE_H
class AddValue{ AddValue::AddValue( int inValue ){
int value; this->value = inValue;
public: }
AddValue( int inValue = 1); void AddValue::operator()( int& what ){
void operator()( int& what ); what += this->value;
}; }
#endif /* ADDVALUE_H */
141
– The function call operator
AddValue func(2);
int array[]={1, 2, 3};
for( int& x : array ){
func(x);
}
for( int x: array ){
cout <<x<<endl;
}
142
– Function call operator
● used frequently for defining sorting criterion
struct EmployeeCompare{
bool operator()( const Employee& e1, const Employee& e2){
if( e1.getLastName() == e2.getLastName())
return e1.getFirstName() < e2.getFirstName();
else
return e1.getLastName() < e2.getLastName();
}
};
143
– Function call operator
● sorted container
set<Employee, EmployeeCompare> s;
Employee e1; e1.setFirstName("Barbara");

e1.setLastName("Liskov");
Employee e2; e2.setFirstName("John");
e2.setLastName("Steinbeck");
Employee e3; e3.setFirstName("Andrew");
e3.setLastName("Foyle");
s.insert( e1 ); s.insert( e2 ); s.insert( e3 );
for( auto& emp : s){

emp.display();
}
144
– Sorting elements of a given type:
● A. override operators: <, ==
● B. define a function object containing the comparison
– Which one to use?

● Q: How many sorted criteria can be defined using method A?
● Q: How many sorted criteria can be defined using method B?
145
– Writing conversion operators
class Complex{
public:
operator string() const;
//
};
Complex::operator string() const{

stringstream ss;
ss<<this->re<<"+"<<this->im<<"i";
string s;
ss>>s;
return s;
}
//usage
Complex z(1, 2), z2;
string a = z;
cout<<a<<endl;
146
– After templates
● Overloading operator *
● Overloading operator →
147
OOP: Review
– Find all possible errors or shortcommings!
(1) class Array {
(2) public:
(3) Array (int n) : rep_(new int [n]) { }
(4) Array (Array& rhs) : rep_(rhs.rep_) { }
(5) ~Array () { delete rep_; }
(6) Array& operator = (Array rhs) { rep_= rhs.rep_; }
(7) int& operator [] (int n) { return &rep_[n]; }
(8) private:
(9) int * rep_;
(10) }; // Array
Source: http://www.cs.helsinki.fi/u/vihavain/k13/gea/exer/exer_2.html
148
Solution required!
– It is given the following program!
#include <iostream>
int main(){
std::cout<<”Hello\n”;
return 0;
}
Modify the program without modifying the main function so that the output of
the program would be:
Start
Hello
Stop
149
Singleton Design Pattern
#include <string>
class Logger{
public:
static Logger* Instance();
bool openLogFile(std::string logFile);
void writeToLogFile();
bool closeLogFile();
private:
Logger(){}; // Private so that it can not be called
Logger(Logger const&){}; // copy constructor is private
Logger& operator=(Logger const&){};// assignment operator is private
static Logger* m_pInstance;
};
http://www.yolinux.com/TUTORIALS/C++Singleton.html
150
Singleton Design Pattern
static
class Framewor...
Singleton
- uniqueInstance
- singletonData
+ Instance()
return unique Instance
+ SingletonOperation()
+ GetSingletonData ()
static
● Ensure that only one instance of a class is created.
● Provide a global point of access to the object.
151
Module 5
Public Inheritance
152
OOP: Inheritance
– Inheritance
● is-a relationship - public inheritance
● protected access
● virtual member function
● early (static) binding vs. late (dynamic) binding
● abstract base classes
● pure virtual functions
● virtual destructor
153
class cppinheritance
Employee
OOP: Inheritance -
-
firstNam e: stri ng
lastNam e: stri ng
- salary: do uble
– public inheritance + Em ployee(strin g, string, doubl e)
+ getFirstNam e () : string {query}
● is-a relationship +
+
setFirstNam e(string) : void
getLastNam e () : string {query}
+ setLastNam e(string) : void
● base class: Employee + getSala ry() : double {qu ery}
+ setSalary(d ouble) : void
● derived class: Manager + print(o stream&) : void {query}
– You can do with inheritance

● add data Manager
– ex. department - depa rtm e nt: string
● add functionality +
+
M anager()
M anager(string, string, do uble, string)
+ setDepa rtm ent(string) : void
– ex. getDepartment(), setDepartment() + getDe partm ent() : string {query}
+ print(ostream &) : void {query}
● modify methods' behavior
– ex. print()
154
OOP: Inheritance
– protected access
● base class's private members can not be accessed in a
derived class
● base class's protected members can be accessed in a
derived class
● base class's public members can be accessed from
anywhere
155
OOP: Inheritance
– public inheritance
class Employee{
public:
Employee(string firstName = "", string lastName = "",
double salary = 0.0) : firstName(firstName),
lastName(lastName),
salary(salary) {
}
//...
};
class Manager:public Employee{

string department;
public:
Manager();
Manager( string firstName, string lastName, double salary,
string department );
//...
};
156
OOP: Inheritance
– Derived class's constructors
Manager::Manager(){
}
Employee's constructor invocation → Default constructor can be invoked implicitly
157
OOP: Inheritance
– Derived class's constructors
Manager::Manager(){
}
Employee's constructor invocation → Default constructor can be invoked implicitly
Manager::Manager(string firstName, string lastName, double salary,

string department): Employee(firstName, lastName, salary),
department(department){
}
base class's constructor invocation – constructor initializer list

arguments for the base class's constructor are specified in the definition of a derived class's constructor
158
OOP: Inheritance
– How are derived class's objects constructed?
● bottom up order:
Manager
– base class constructor invocation
– member initialization
– derived class's constructor block Employee
● destruction
– in the opposite order
159
OOP: Inheritance
– Method overriding
Employee {
class Employee{
public:
virtual void print( ostream&)const;
print(ostream&) const;
};

public:
virtual void print(ostream&) const;

};
160
OOP: Inheritance
– Method overriding
class Employee {
public:
virtual void print( ostream&) const;
};
void Employee::print(ostream& os ) const{

os<<this->firstName<<" "<<this->lastName<<" "<<this->salary;
}

public:
};
void Manager::print(ostream& os) const{

Employee::print(os);
os<<" "<<department;
}
161
OOP: Inheritance
– Method overriding - virtual functions
● non virtual functions are bound statically
– compile time
● virtual functions are bound dynamically
– run time
162
OOP: Inheritance
– Polymorphism
void printAll( const vector<Employee*>& emps ){
for( int i=0; i<emps.size(); ++i){
emps[i]-> print(cout);
cout<<endl;
}
}

vector<Employee*> v;
Employee e("John", "Smith", 1000);
v.push_back(&e);
Manager m("Sarah", "Parker", 2000, "Sales");
v.push_back(&m);
cout<<endl; Output:
printAll( v ); John Smith 1000
return 0; Sarah Parker 2000 Sales
}
163
OOP: Inheritance
– Polymorphism
● a type with virtual functions is called a polymorphic type
● polymorphic behavior preconditions:
– the member function must be virtual
– objects must be manipulated through
● pointers or
● references
– Employee :: print( os ) static binding – no

polymorphism
164
OOP: Inheritance
– Polymorphism – Virtual Function Table
vtbl:
class Employee{
Employee::print firstName:””
firstName:””
public:
firstName:”” lastName:””
lastName:”” lastName:””
};
//... e1 salary:0.0
m1salary:0.0
salary:0.0
department
department
class Manager:public Employee{ vptr vptr
vtbl
//...
}; vtbl:
Employee e1, e2;
Manager m1, m2; Manager::print firstName:””
firstName:””
lastName:”” lastName:””
e2 salary:0.0 m2 salary:0.0
department
Discussion!!! vptr
vptr
Employee * pe;
pe = &e1; pe->print();//???
pe = &m2; pe->print();//???
Each class with virtual functions has its own virtual function table (vtbl). 165
RTTI – Run-Time Type Information
dynamic_cast<>(pointer)
class Base{};
class Derived : public Base{};
Base* basePointer = new Derived();

Derived* derivedPointer = nullptr;
//To find whether basePointer is pointing to Derived type of object
derivedPointer = dynamic_cast<Derived*>(basePointer);
if (derivedPointer != nullptr){
cout << "basePointer is pointing to a Derived class object";
}else{
cout << "basePointer is NOT pointing to a Derived class object";
}
Java:
instanceof
166
RTTI – Run-Time Type Information
dynamic_cast<>(reference)
class Base{};
Derived derived;
Base& baseRef = derived;
// If the operand of a dynamic_cast to a reference isn’t of the expected type,

// a bad_cast exception is thrown.
try{
Derived& derivedRef = dynamic_cast<Derived&>(baseRef);
} catch( bad_cast ){
// ..
}
167
OOP: Inheritance
– Abstract classes
● used for representing abstract concepts
● used as base class for other classes
● no instances can be created
168
OOP: Inheritance
– Abstract classes – pure virtual functions
class Shape{ // abstract class

public:
virtual void rotate(int) = 0; // pure virtual function
virtual void draw() = 0; // pure virtual function
// ...
};
Shape s; //???
169
OOP: Inheritance
– Abstract classes – pure virtual functions
class Shape{ // abstract class

public:
virtual void rotate(int) = 0; // pure virtual function
virtual void draw() = 0; // pure virtual function
// ...
};
Shape s; //Compiler error
170
OOP: Inheritance
– Abstract class → concrete class
class Point{ /* ... */ };

class Circle : public Shape {
public:
void rotate(int); // override Shape::rotate
void draw(); // override Shape::draw
Circle(Point p, int r) ;
private:
Point center;
int radius;
};
171
OOP: Inheritance
– Abstract class → abstract class
class Polygon : public Shape{

public:
// draw() and rotate() are not overridden
};
Polygon p; //Compiler error
172
OOP: Inheritance
– Virtual destructor
● Every class having at least one virtual function should
have virtual destructor. Why?
class X{
public:
// ...
virtual ~X();
};
173
OOP: Inheritance
– Virtual destructor
void deleteAll( Employee ** emps, int size){
for( int i=0; i<size; ++i){
delete emps[ i ]; Which destructor is invoked?
}
delete [] emps;
}
// main
Employee ** t = new Employee *[ 10 ];
for(int i=0; i<10; ++i){
if( i % 2 == 0 )
t[ i ] = new Employee();
else
t[ i ] = new Manager();
}
deleteAll( t, 10);
174
Module 6
Object relationships
175
OOP: Object relationships
– Content
● The is-a relationship
● The has-a relationship
● Private inheritance
● Multiple inheritance
176
Super
– The is-a relationship – Client's view (1)
● works in only one direction:
– every Sub object is also a Super one Sub
– but Super object is not a Sub
void foo1( const Super& s );

void foo2( const Sub& s);
Super super;
Sub sub;
foo1(super); //OK
foo1(sub); //OK
foo2(super); //NOT OK
foo2(sub); //OK
177
Super
– The is-a relationship – Client's view (2)
class Super{ Super * p= new Super();

p->method1(); //OK
Sub
public:
virtual void method1();
}; p = new Sub();
class Sub : public Super{ p->method1(); //OK
public: p->method2(); //NOT OK
virtual void method2(); ((Sub *)p)->method2();//OK
};
178
Super
– The is-a relationship – Sub-class’s view
● the Sub class augments the Super class by
adding additional methods Sub
● the Sub class may override the Super class methods
● the subclass can use all the public and
protected members of a superclass.
179
– The is-a relationship: preventing inheritance C++11
● final classes – cannot be extended
class Super final
{
};
180
– The is-a relationship: a client's view of overridden methods(1)
● polymorphism Super super;
super.method1(); //Super::method1()
Sub sub;
class Super{ sub.method1(); //Sub::method1()
public:
Super& ref =super;
ref.method1(); //Super::method1();
};
class Sub : public Super{ ref = sub;
public: ref.method1(); //Sub::method1();
}; Super* ptr =&super;
ptr->method1(); //Super::method1();
ptr = ⊂
ptr->method1(); //Sub::method1();
181
– The is-a relationship: a client's view of overridden methods(2)
● object slicing
class Super{ Sub sub;
public: Super super = sub;
super.method1(); // Super::method1();
};
class Sub : public Super{ Sub
public:
}; Super Super
super sub
182
– The is-a relationship: preventing method overriding C++11
class Super{
public:
virtual void method1() final;
};
class Sub : public Super{
public:
virtual void method1(); //ERROR
};
183
– Inheritance for polymorphism
www.javatutorialhub.com
184
– The has-a relationship
Window Button
185
– Implementing the has-a relationship
● An object A has an object B
class B; class B; class B;
class A{ class A{ class A{

private: private: private:
B b; B* b; B& b;
}; }; };
186
– Implementing the has-a relationship A B
– strong containment (composition)
class B; A anObject; anObject: A

class A{
private: b: B
B b;
};
187
– weak containment (aggregation)
class B; B bObject;
A aObject1(bObject);
class A{ A aObject2(bObject);
private:
B& b;
public: bObject: B
A( const B& pb):b(pb){}
};
aObject1: A aObject2: A
188
– Implementing the has-a relationship
weak containment strong containment
class B; class B;
class A{ class A{
private: private:
B* b; B* b;
public: public:
A( B* pb):b( pb ){} A(){
}; b = new B();
}
~A(){
delete b;
}
};
189
weak containment
Usage:
class B;
B bObject;
class A{ A aObject1(&bObject);
private: A aObject2(&bObject);
B* b;
public:
A( B* pb):b( pb ){}
}; bObject: B
aObject1: A aObject2: A
190
strong containment
class B; Usage:
A aObject;
class A{
private:
B* b;
public: anObject: A
A(){
b = new B();
} b: B *
~A(){
delete b;
}
};
191
– Combining the is-a and the has-a relationships
class Class Mo...
Component
+ Operation() 1..*
Client
+ Add() : Component
+ Rem ove() : Com ponent
+ GetChild() : Com ponent
Leaf Composite
+ Operation() + Operation()
forall g in children -children
g.Operation();
+ Add() : Com ponent
192
Composite Design Pattern
class Class Mo...
Component
+ O peration() 1..*
Client
+ Add() : Component
+ G etChild() : Com ponent
Leaf Composite
+ Operation() + Operation()
forall g in children -children
g.Operation();
+ Add() : Com ponent
● Compose objects into tree structures to represent part-whole

hierarchies.
● Lets clients treat individual objects and composition of objects
uniformly.
193
Composite Design Pattern
● Examples:
– Menu – MenuItem: Menus that contain menu items, each of which could
be a menu.
– Container – Element: Containers that contain Elements, each of which
could be a Container.
– GUI Container – GUI component: GUI containers that contain GUI
components, each of which could be a container
Source: http://www.oodesign.com/composite-pattern.html
194
Private Inheritance
– another possibility for has-a relationship
public Base class Base class public
public private
inheritance inheritance
public Derived class Derived class private
Derived class inherits the Derived class hides the

base class behavior base class behavior
195
Private Inheritance
template <typename T>
class MyStack : private vector<T> { Why is public inheritance
public: in this case dangerous???
void push(T elem) {
this->push_back(elem);
}
bool isEmpty() {
return this->empty();
}
void pop() {
if (!this->empty())this->pop_back();
}
T top() {
if (this->empty()) throw out_of_range("Stack is empty");
else return this->back();
}
};
196
Non-public Inheritance
– it is very rare;
– use it cautiously;
– most programmers are not familiar with it;
197
What does it print?
class Super{
public:
Super(){}
virtual void someMethod(double d) const{
cout<<"Super"<<endl;
}
};
public:
Sub(){}
virtual void someMethod(double d){
cout<<"Sub"<<endl;
}
};
Sub sub; Super super;

Super& ref = sub;ref.someMethod(1);
ref = super; ref.someMethod(1);
198
What does it print?
class Super{
public:
Super(){}
}
}; creates a new method, instead
class Sub : public Super{ of overriding the method
public:
Sub(){}
virtual void someMethod(double d){
cout<<"Sub"<<endl;
}
};

199
The override keyword C++11
class Super{
public:
Super(){}
}
};
public:
Sub(){}
virtual void someMethod(double d) const override{
cout<<"Sub"<<endl;
}
};

200
Module 7
Generic Programming: Templates
201
Outline
– Templates
● Class template
● Function template
● Template metaprogramming
202
Templates
203
Templates
http://www.stroustrup.com/
204
Templates
– Allow generic programming
● to write code that can work with all kind of objects
● template programmer's obligation: specify the

requirements of the classes that define these objects
● template user's obligation: supplying those operators

and methods that the template programmer requires
205
Template
Function Template parameter
– Allows writing function families

template<typename T>
const T max(const T& x, const T& y) {
return x < y ? y : x;
}
template<class T>
}
● What are the requirements regarding the type T?
206
Function Template
template<class T>
}
● Requirements regarding the type T:

– less operator (<)
207
Function Template
template<class T>
}
● Usage:
– cout<<max(2, 3)<<endl; // max: T → int
– string a(“alma”); string b(“korte”);
cout<<max(a, b)<<endl; // max: T → string
– Person p1(“John”,”Kennedy”),p2(“Abraham”, “Lincoln”);
cout<<max(p1,p2)<<endl;// max: T-> Person
208
Function Template
template<class T>
void swap(T& x, T& y) {
const T tmp = x;
x = y;
y = tmp;
}
● Requirements regarding the type T:

– assignment operator
209
Function Template
– Allows writing function families
● polymorphism: compile time
– How the compiler processes templates?

– cout<<max(2, 3)<<endl; // max: T → int
– cout<<max(2.5, 3.6)<<endl; // max: T → double
–
– How many max functions?

Warning: Code bloat!
210
Function Template
– What does it do? [Gregoire]
static const size_t MAGIC = (size_t)(-1);

size_t Foo(T& value, T* arr, size_t size)
{
for (size_t i = 0; i < size; i++) {
if (arr[i] == value) {
return i;
}
}
return MAGIC;
}
211
Class Template
– Allow writing class families
class Array {
T* elements;
int size;
public:
explicit Array(const int size);
...
};
212
Class Template
– Template class's method definition
class Array {
T* elements;
int size;
public:
explicit Array(const int size);
...
};
Array<T>::Array(const int size):size(size),
elements(new T[size]){
}
213
Class Template
– Template parameters
● type template parameters
● non-type template parameters
template<typename T> template<class T, int MAX=100>

class Array { class Stack{
T* elements; T elements[ MAX ];
int size; public:
public: ...
Array(const int size); };
...
};
214
Class Template
– Distributing Template Code between Files
● Normal class:
– Person.h → interface
– Person.cpp → implementation
● Template class:
– interface + implementation go in the same file e. g. Array.h
● it can be a .h file → usage: #include “Array.h”
● it can be a .cpp file → usage: #include “Array.cpp”
215
Class Template+ Function Template
template<class T1, class T2>
struct pair { #include <utility>
typedef T1 first_type;
typedef T2 second_type;
T1 first;
T2 second;
pair();
pair(const T1& x, const T2& y);
...
};
template< class T1, class T2>

pair<T1, T2> make_pair(const T1& x, const T2& y)
{
return pair<T1, T2>(x, y);
}
216
Advanced Template
● template template parameter
template<typename T, typename Container>
class Stack{
Container elements;
public:
void push( const T& e ){
elements.push_back( e );
}
...
};
Usage:
Stack<int, vector<int> > v1;
Stack<int, deque<int> > v2;
217
Advanced Template
● template template parameter
template<typename T, typename Container=vector<T> >
class Stack{
Container elements;
public:
void push( const T& e ){
elements.push_back( e );
}
...
};
218
Advanced Template
● What does it do?
template < typename Container >
void foo( const Container& c, const char * str="")
{
typename Container::const_iterator it;
cout<<str;
for(it = c.begin();it != c.end(); ++it)
cout<<*it<<' ';
cout<<endl;
}
219
Advanced Template
● What does it do?
template < typename Container >
void foo( const Container& c, const char * str="")
{
typename Container::const_iterator it;
cout<<str;
for(auto& a: c ){
cout<< a <<' ';
}
cout<<endl;
}
220
Examples
Implement the following template functions!

bool linsearch( T* first, T* last, T what);

bool binarysearch( T* first, T* last, T what);
221
More Advanced Template
● Template Metaprogramming
template<unsigned int N> struct Fact{
static const unsigned long int
value = N * Fact<N-1>::value;
};
template<> struct Fact<0>{
static const unsigned long int value = 1;
};
// Fact<8> is computed at compile time:
const unsigned long int fact_8 = Fact<8>::value;
int main()
{
cout << fact_8 << endl;
return 0;
}
222
Module 8
STL – Standard Template Library
223
Alexander Stepanov
https://www.sgi.com/tech/stl/drdobbs-interview.html
224
Outline
– Containers
– Algorithms
– Iterators
225
STL – General View
– library of reusable components

– a support for C++ development
– based on generic programming
226
STL – General View
– Containers – Template Class

● generalized data structures (you can use them for any
type)
– Algorithms – Template Function
● generalized algorithms (you can use them for almost any
data structure)
– Iterators – Glue between Containers and Algorithms
● specifies a position into a container (generalized pointer)
● permits traversal of the container
227
Basic STL Containers
– Sequence containers
● linear arrangement
– vector, deque, list <vector> <deque> <list>
Container
adapters – stack, queue, priority_queue <stack> <queue>
– Associative containers
● provide fast retrieval of data based on keys
– set, multiset, map, multimap <set> <map>
228
Sequence Containers
229
Associative Containers
230
STL Containers C++11
– Sequence containers <forward-list>

<array> <forward_list>
– array (C-style array)
– forward_list (singly linked list)
<unordered_set>
– Associative containers <unordered_map>
– unordered_set, unordered_multiset (hash table)
– unordered_map, unordered_multimap (hash table)
231
STL Containers
– homogeneous:
● vector<Person>, vector<Person*>
– polymorphism
● vector<Person*>
class Person{};
class Employee: public Person{};
class Manager : public Employee{};
232
STL Containers
vector<Person>
Person Person Person homogenous
– ...
233
STL Containers
vector<Person>
Person Person Person homogenous
– ...
vector<Person *>
... homogenous
Person Employee Manager heterogenous
234
The vector container - constructors
vector<T> v;
//empty vector
vector<T> v(n, value);

//vector with n copies of value
vector<T> v(n);
//vector with n copies of default for T
235
The vector container – add new elements
vector<int> v;
for( int i=1; i<=5; ++i){

v.push_back( i );
}
v.begin() v.end()
1 2 3 4 5
236
The vector container
vector<int> v( 10 );
cout<<v.size()<<endl;//???
for( int i=0; i<v.size(); ++i ){
cout<<v[ i ]<<endl;
}
for( int i=0; i<10; ++i){

v.push_back( i );
}
for( auto& a: v ){
cout<< a <<” “;
}
237
The vector container: typical errors
– Find the error and correct it!
vector<int> v;
for( int i=0; i<10; ++i ){
v[ i ] = i;
}
for( int i=0; i<v.size(); ++i ){
cout<<v[ i ]<<endl;
}
238
The vector container: capacity and size
vector<int> v;
v.reserve( 10 );
cout << v.size() << endl;//???

cout << v.capacity() << endl;//???
239
The vector container: capacity and size
vector<int> v;
v.reserve( 10 );
cout << v.size() << endl;//???

cout << v.capacity() << endl;//???
--------------------------------------
vector<int> gy( 256 ); Purpose?

ifstream ifs("szoveg.txt"); int c;
while( (c = ifs.get() ) != -1 ){
gy[ c ]++;
}
240
The vector - indexing
int Max = 100;
vector<int> v(Max);
//???...
for (int i = 0; i < 2*Max; i++) {
cout << v[ i ]<<” ”;
}
--------------------------------------
int Max = 100;

vector<int> v(Max);
for (int i = 0; i < 2*Max; i++) {
cout << v.at( i )<<” ”;
}
241
The vector - indexing
int Max = 100;
vector<int> v(Max);
//???...
for (int i = 0; i < 2*Max; i++) {
cout << v[ i ]<<” ”; Efficient
}
--------------------------------------
int Max = 100;

vector<int> v(Max);
for (int i = 0; i < 2*Max; i++) {
cout << v.at( i )<<” ”; Safe
}
out_of_range exception
242
The list container
– doubly linked list
list<int> l;
for( int i=1; i<=5; ++i){
l.push_back( i );
}
l.begin() l.end()
1 2 3 4 5
243
The deque container
– double ended vector
deque<int> l;
for( int i=1; i<=5; ++i){
l.push_front( i );
}
244
Algorithms - sort
template <class RandomAccessIterator>
void sort ( RandomAccessIterator first,RandomAccessIterator last );
template <class RandomAccessIterator, class Compare>

void sort ( RandomAccessIterator first, RandomAccessIterator last,
Compare comp );
– what to sort: [first, last)

– how to compare the elements:
● <
● comp
245
Algorithms - sort
struct Rec {
string name;
string addr;
};
vector<Rec> vr;
// …
sort(vr.begin(), vr.end(), Cmp_by_name());
sort(vr.begin(), vr.end(), Cmp_by_addr());
–
246
Algorithms - sort
struct Cmp_by_name{
bool operator()(const Rec& a, const Rec& b) const
{
return a.name < b.name;
} function object
};
struct Cmp_by_addr{
bool operator()(const Rec& a, const Rec& b) const
{
return a.addr < b.addr;
– }
};
247
Strategy Design Pattern
class ceepus_iterator
Strategy
Context +strategy
+ Algorith mInterface()
+ ContextIn terface()
ConcreteStrategyA ConcreteStrategyB ConcreteStrategyC
+ Algorith m In terface() + Algorith m In terface() + Alg orithm Interface()
● Define a family of algorithms, encapsulate each one, and make them

interchangeable.
● Strategy lets the algorithm vary independently from clients that use it.
248
sort
Strategy
Context +strategy

interchangeable.
249
bool operator()(
sort const T&,
Strategy const T&)
Context +strategy

interchangeable.
250
bool operator()(
sort const T&,
Strategy const T&)
Context +strategy
Cmp_by_name Cmp_by_addr
251
Iterators
– The container manages the contained objects but does

not know about algorithms
– The algorithm works on data but does not know the

internal structure of containers
– Iterators fit containers to algorithms
252
Iterator - the glue
int x[]={1,2,3,4,5}; vector<int>v(x, x+5);
int sum1 = accumulate(v.begin(), v.end(), 0);
v.begin() v.end()
1 2 3 4 5
list<int> l(x, x+5);

double sum2 = accumulate(l.begin(), l.end(), 0);
l.begin() l.end()
1 2 3 4 5
253
Iterator - the glue
template<class InIt, class T>

T accumulate(InIt first, InIt last, T init)
{
while (first!=last) {
init = init + *first;
++first;
}
return init;
}
254
The set container
set< Key[, Comp = less<Key>]>

usually implemented as a balanced binary search tree
multiset: allows duplicates
Source:http://www.cpp-tutor.de/cpp/le18/images/set.gif
255
The set container - usage
#include <set>
set<int> intSet;
set<Person> personSet1;
set<Person, PersonComp> personSet2;
256
<
#include <set>
set<int> intSet;
257
<
#include <set>
bool operator<(const Person&, const Person&)

set<int> intSet;
258
<
#include <set>
bool operator<(const Person&, const Person&)

set<int> intSet;
struct PersonComp{
bool operator() ( const Person&, const Person& );
};
259
#include <set>
set<int> mySet;
while( cin >> nr ){
mySet.insert( nr );
}
set<int>::iterator iter;
for (iter=mySet.begin(); iter!=mySet.end(); ++iter){
cout << *iter << endl;
}
260
set<int>::iterator iter;
for (iter=mySet.begin(); iter!=mySet.end(); ++iter){
cout << *iter << endl;
}
----------------------------------------------------
for( auto& i: mySet ){
cout<<i<<endl;
}
261
The multiset container - usage
multiset<int> mySet;
size_t nrElements = mySet.count(12);
multiset<int>::iterator iter; ?
iter = mySet.find(10);
if (iter == mySet.end()){
cout<<"The element does not exist"<<endl;
}
262
The multiset container - usage
multiset<int> mySet;
auto a = mySet.find(10);
if (a == mySet.end()){
cout<<"The element does not exist"<<endl;
}
263
class PersonCompare;
class Person {
friend class PersonCompare;
string firstName;
string lastName;
int yearOfBirth;
public:
Person(string firstName, string lastName, int yearOfBirth);
friend ostream& operator<<(ostream& os, const Person& person);
};
264
class PersonCompare {
function object
public:
enum Criterion { NAME, BIRTHYEAR};
private:
Criterion criterion; state
public:
PersonCompare(Criterion criterion) : criterion(criterion) {}
bool operator()(const Person& p1, const Person& p2) {
behaviour
switch (criterion) {
case NAME: //
case BIRTHYEAR: //
}
}
};
265
set<Person, PersonCompare> s( PersonCompare::NAME);

s.insert(Person("Biro", "Istvan", 1960));
s.insert(Person("Abos", "Gergely", 1986));
s.insert(Person("Gered","Attila", 1986));
---------------------------------------------------- ?
for( auto& p: s){
cout << p <<endl;
} C++
2011
266
The map container
map< Key, Value[,Comp = less<Key>]>
usually implemented as a balanced binary tree
map: associative array multimap: allows duplicates
?
Source: http://www.cpp-tutor.de/cpp/le18/images/map.gif
267
The map container - usage
#include <map>
map<string,int> products;
products.insert(make_pair("tomato",10));
----------------------------------------------
products["cucumber"] = 6;
cout<<products["tomato"]<<endl;
268
#include <map>
map<string,int> products;
products.insert(make_pair("tomato",10));
----------------------------------------------
products["cucumber"] = 6;
cout<<products["tomato"]<<endl;
Difference between
[ ] and insert!!!
269
#include <map>
int main ()
{
map < string , int > m;
cout << m. size () << endl; // 0
if( m["c++"] != 0 ){
cout << "not 0" << endl;
}
cout << m. size () << endl ; // 1
}
[ ] side effect
270
typedef map<string,int>::iterator MapIt;

for(MapIt it= products.begin(); it != products.end(); ++it){
cout<<(it->first)<<" : "<<(it->second)<<endl;
}
--------------------------------------------------
for( auto& i: products ){

cout<<(i.first)<<" : "<<(i.second)<<endl;
}
C++
2011
271
The multimap container - usage
multimap<string, string> cities;

cities.insert(make_pair("HU", "Budapest"));
cities.insert(make_pair("HU", "Szeged"));
cities.insert(make_pair("RO", "Seklerburg"));
cities.insert(make_pair("RO", "Neumarkt"));
cities.insert(make_pair("RO", "Hermannstadt"));
typedef multimap<string, string>::iterator MIT;

pair<MIT, MIT> ret = cities.equal_range("HU");
for (MIT it = ret.first; it != ret.second; ++it) {
cout << (*it).first <<"\t"<<(*it).second<<endl;
}
272
multimap<string, string> cities;

auto ret = cities.equal_range("HU");

for (auto it = ret.first; it != ret.second; ++it){
} C++
2011
273
multimaps do not provide
operator[ ]
multimap<string, string> cities; Why???
auto ret = cities.equal_range("HU");

for (auto it = ret.first; it != ret.second; ++it) {
}
274
The set/map container - removal
void erase ( iterator position );
size_type erase ( const key_type& x );
void erase ( iterator first, iterator last );
275
The set – pointer key type
Output??
set<string *> animals;

animals.insert(new string("monkey"));
animals.insert(new string("lion"));
animals.insert(new string("dog"));
animals.insert(new string("frog"));
for( auto& i: animals ){

cout<<*i<<endl;
}
276
The set – pointer key type
Corrected
struct StringComp{
bool operator()(const string* s1,
const string * s2){
return *s1 < *s2;
}
};
set<string*, StringComp> animals;
animals.insert(new string("monkey"));
animals.insert(new string("lion"));
animals.insert(new string("dog"));
animals.insert(new string("frog"));
-------------------------------------------------------------------
for( auto& i: animals ){
cout<<*i<<endl;
}
277
Hash Tables
collision
http://web.eecs.utk.edu/~huangj/CS302S04/notes/extendibleHashing.htm
278
Hash Tables
Collision resolution by chaining
Source: http://integrator-crimea.com/ddu0065.html
279
Unordered Associative Containers - Hash Tables
– unordered_set
– unordered_multiset C++
2011
– unordered_map
– unordered_multimap
280
Unordered Associative Containers
– The STL standard does not specify which collision

handling algorithm is required
● most of the current implementations use linear chaining
● a lookup of a key involves:
– a hash function call h(key)– calculates the index in the
hash table
– compares key with other keys in the linked list
281
Hash Function
– perfect hash: no collisions

– lookup time: O(1) - constant
– there is a default hash function for each STL hash
container
282
The unordered_map container
template <class Key, class T,
class Hash = hash<Key>,
class Pred = std::equal_to<Key>,
class Alloc= std::allocator<pair<const Key, T>>>
class unordered_map;
Template parameters:
– Key – key type

– T – value type
– Hash – hash function type
– Pred – equality type
283
The unordered_set container
template <class Key,
class Hash = hash<Key>,
class Pred = std::equal_to<Key>,
class Alloc= std::allocator<pair<const Key, T>>>
class unordered_set;
Template parameters:
– Key – key type

– Hash – hash function type
– Pred – equality type
284
Problem
– Read a file containing double numbers. Eliminate the

duplicates.
– Solutions???
285
Solutions
– vector<double> + sort + unique

– set<double>
– unordered_set<double>
– Which is the best? Why?

– What are the differences?
286
#include <chrono>
Elapsed time
auto begin =chrono::high_resolution_clock::now();

//Code to benchmark
auto end = chrono::high_resolution_clock::now();
cout <<chrono::duration_cast<std::chrono::nanoseconds>
(end-begin).count()
<< "ns" << endl;
287
class Class Mo...
RandomNumbers
# size: int
+ Random Num bers(int)

+ generate() : voi d
+ getSize() : int
SetRandomNumbers UnorderedSetRandomNumbers VectorRandomNumbers
- num bers: set<dou ble> - num bers: uno rdered_set<do uble> - num bers: vector<d ouble>
+ SetRandom Num bers(int) + Unordered SetRandom Num bers(int) + VectorRandom Num bers(int)
+ generate() : vo id + generate () : void + generate () : void
+ getSize() : int + getSize() : in t + getSize() : in t
288
Ellapsed time
Container Time (mean)

vector 1.38 sec
set 3.04 sec
unordered_set 1.40 sec
289
Which container to use?
– implement a PhoneBook, which:

● stores names associated with their phone numbers;
● names are unique;
● one name can have multiple phone numbers
associated;
● provides O(1) time search;
290
Which container to use?
– Usage:
PhoneBook pbook;
pbook.addItem("kata","123456");
pbook.addItem("timi","444456");
pbook.addItem("timi","999456");
pbook.addItem("elod","543456");
cout<<pbook<<endl;
291
unordered_map: example
class PhoneBook{
unordered_map<string, vector<string> > book;
public:
void addItem( string name, string phone);
void removeItem( string name, string phone);
vector<string> findItem( string name );
friend ostream& operator<<( ostream& os,
const PhoneBook& book);
};
292
typedef unordered_map<string, vector<string> >::iterator Iterator;
void PhoneBook::addItem( string name, string phone){

Iterator it = this->book.find( name );
if( it != book.end() ){
it->second.push_back( phone );
}else{
vector<string> phones;
phones.push_back(phone);
book.insert( make_pair(name, phones ));
}
}
293

vector<string> phones = it->second;
phones.push_back( phone );
}else{
}
}
Find the error and

correct it! 294

vector<string>& phones = it->second;
phones.push_back( phone );
}else{
}
}
phone will be
inserted into the
map 295
C++/Java
C++ Java
Objects X x; X x = new X();

X * px = new X();
Parameter passing void f( X x ); void f( X x );
void f( X * px); //pass through reference
void f( X& rx);
void f( const X&rx);
run-time binding only for virtual functions for each function (except static
functions)
memory management explicit (2011 - smart pointers!) implicit (garbage collection)

multiple inheritance yes no
interface no (abstract class with pure virtual yes

functions!)
296
Algorithms
297
Algorithms
– OOP encapsulates data and functionality

● data + functionality = object
– The STL separates the data (containers) from the
functionality (algorithms)
● only partial separation
298
Algorithms – why separation?
STL principles:
– algorithms and containers are independent
– (almost) any algorithm works with (almost) any container
– iterators mediate between algorithms and containers
● provides a standard interface to traverse the elements of
a container in sequence
299
Algorithms
Which one should be used?
set<int> s;
set<int>::iterator it = find(s.begin(), s.end(), 7);
if( it == s.end() ){
//Unsuccessful
}else{
//Successful
}
set<int> s;
set<int>::iterator it = s.find(7);
//Unsuccessful
}else{
//Successful
}
300
Algorithms
O(n)
Which one should be used?
set<int> s;
set<int>::iterator it = find(s.begin(), s.end(), 7);
//Unsuccessful
}else{
//Successful
}
set<int> s;
set<int>::iterator it = s.find(7);
//Unsuccessful
}else{
//Successful
}
O(log n)
301
Algorithm categories
– Utility algorithms
– Non-modifying algorithms
● Search algorithms
● Numerical Processing algorithms
● Comparison algorithms
● Operational algorithms
– Modifying algorithms
● Sorting algorithms
● Set algorithms
302
Utility Algorithms
– min_element()
– max_element()
– minmax_element() C++11
– swap()
303
Utility Algorithms
vector<int>v = {10, 9, 7, 0, -5, 100, 56, 200, -24};

auto result = minmax_element(v.begin(), v.end() );
cout<<"min: "<<*result.first<<endl;
cout<<"min position: "<<(result.first-v.begin())<<endl;
cout<<"max: "<<*result.second<<endl;
cout<<"max position: "<<(result.second-v.begin())<<endl;
return 0;
304
Non-modifying algorithms
Search algorithms
– find(), find_if(), find_if_not(), find_first_of()

– binary_search()
– lower_bound(), upper_bound(), equal_range()
– all_of(), any_of(), none_of()
– ...
305
Search algorithms - Example
– bool isEven (int i) { return ((i%2)==0); }
typedef vector<int>::iterator VIT;
int main () {
vector<int> myvector={1,2,3,4,5};
VIT it= find_if (myvector.begin(), myvector.end(), isEven);
cout << "The first even value is " << *it << '\n';
return 0;
}
auto
306
Numerical Processing algorithms
– count(), count_if()
– accumulate()
– ...
307
Numerical Processing algorithms - Example
bool isEven (int i) { return ((i%2)==0); }
int main () {
vector<int> myvector={1,2,3,4,5};
int n = count_if (myvector.begin(), myvector.end(), isEven);
cout << "myvector contains " << n << " even values.\n";
return 0;
}
[ ] (int i){ return i %2 == 0; }
308
Comparison algorithms
– equal()
– mismatch()
– lexicographical_compare()
309
Problem
It is given strange alphabet – the order of characters are unusual.
Example for a strange alphabet: {b, c, a}.

Meaning: 'b'->1, c->'2', 'a' ->3
In this alphabet: “abc” >”bca”
Questions:
● How to represent the alphabet (which container and why)?
● Write a function for string comparison using the strange
alphabet.
310
Comparison algorithms - Example
// strange alphabet: 'a' ->3, 'b'->1, c->'2'

map<char, int> order;
// Compares two characters based on the strange order

bool compChar( char c1, char c2 ){
return order[c1]<order[c2];
}
// Compares two strings based on the strange order
bool compString(const string& s1, const string& s2){
return lexicographical_compare(
s1.begin(), s1.end(), s2.begin(), s2.end(), compChar);
}
311
Comparison algorithms - Example
// strange alphabet: 'a' ->3, 'b'->1, c->'2'

map<char, int> order;
// Compares two strings based on the strange order

struct CompStr{
bool operator()(const string& s1, const string& s2){
return lexicographical_compare(
s1.begin(), s1.end(), s2.begin(), s2.end(),
[](char c1, char c2){return order[c1]<order[c2];} );
}
}
set<string, CompStr> strangeSet;
312
Operational algorithms
– for_each()
void doubleValue( int& x){
x *= 2;
}
vector<int> v ={1,2,3};
for_each(v.begin(), v.end(), doubleValue);
313
Operational algorithms
– for_each()
void doubleValue( int& x){
x *= 2;
}
for_each(v.begin(), v.end(), doubleValue);
for_each(v.begin(), v.end(), []( int& v){ v *=2;});
314
Modifying algorithms
– copy(), copy_backward()
– move(), move_backward() C++11
– fill(), generate()
– unique(), unique_copy()
– rotate(), rotate_copy()
– next_permutation(), prev_permutation()
– nth_element() -nth smallest element
315
Permutations
void print( const vector<int>& v){
for(auto& x: v){
cout<<x<<"\t";
}
cout << endl;
}
int main(){
print( v );
while( next_permutation(v.begin(), v.end())){
print( v );
}
return 0;
}
316
nth_element
double median(vector<double>& v) {
int n = v.size();
if( n==0 ) throw domain_error("empty vector");
int mid = n / 2;
// size is an odd number
if( n % 2 == 1 ){
nth_element(v.begin(), v.begin()+mid, v.end());
return v[mid];
} else{
nth_element(v.begin(), v.begin()+mid-1, v.end());
double val1 = v[ mid -1 ];
nth_element(v.begin(), v.begin()+mid, v.end());
double val2 = v[ mid ];
return (val1+val2)/2;
}
}
317
Iterators
318
Outline
– Iterator Design Pattern

– Iterator Definition
– Iterator Categories
– Iterator Adapters
319
Iterator Design Pattern
class Framewor...
Aggregate Iterator
+ CreateIterator() + First()
+ Next()
+ IsDone()
+ CurrentItem()
ConcreteAggregate ConcreteIterator
+ CreateIterato r()
return new
ConcretIterator(this)
● Provide a way to access the elements of an aggregate object sequentially

without exposing its underlying representation.
● The abstraction provided by the iterator pattern allows you to modify the
320
collection implementation without making any change
Iterator Design Pattern - Java
class Framewor...
java.util.Collection java.util.Iterator
Aggregate Iterator
+ Next()
+ IsDone()
+ CurrentItem()
java.util.LinkedList java.util.ListIterator
+ CreateIterato r()
return new
321
Iterator Design Pattern - C++
class Framewor... <iterator>
class iterator
Aggregate Iterator
+ Next()
+ IsDone()
+ CurrentItem()
list<T> list<T>::iterator
+ CreateIterato r()
return new
322
Definition
– Each container provides an iterator

– Iterator – smart pointer – knows how to
iterate over the elements of that specific
container
– C++ containers provides iterators a common
iterator interface
323
Base class
template <class Category, class T,
class Distance = ptrdiff_t,
class Pointer = T*,
class Reference = T&>
struct iterator {
typedef T value_type;
typedef Distance difference_type;
typedef Pointer pointer;
typedef Reference reference;
typedef Category iterator_category;
};
does not provide any of the functionality an iterator is expected to have.
324
Iterator Categories
– Input Iterator
– Output Iterator
– Forward Iterator
– Bidirectional Iterator
– Random Access Iterator
325
Iterator Categories
– Input Iterator: read forward, object=*it; it++;

– Output Iterator: write forward, *it=object; it++;
– Forward Iterator: read and write forward
– Bidirectional Iterator: read/write forward/backward,
it++, it--;
– Random Access Iterator: it+n; it-n;
326
Basic Operations
– *it: element access – get the element pointed to

– it->member: member access
– ++it, it++, --it, it--: advance forward/
backward
– ==, !=: equality
327
Input Iterator
template<class InIt, class T>
InIt find( InIt first, InIt last, T what)
{
for( ; first != last; ++first )
if( *first == what ){
return first;
}
return first;
}
328
Input Iterator
template<class InIt, class Func>
Func for_each( InIt first, InIt last,
Func f){
for( ;first != last; ++first){
f( *first );
}
return f;
}
329
Output Iterator
template <class InIt, class OutIt>
OutIt copy( InIt first1, InIt last1,
OutIt first2){
while( first1 != last1 ){
*first2 = *first1;
first1++;
first2++;
}
return first2;
}
330
Forward Iterator
template < class FwdIt, class T >
void replace ( FwdIt first, FwdIt last,
const T& oldv, const T& newv ){
for (; first != last; ++first){
if (*first == oldv){
*first=newv;
}
}
}
331
Bidirectional Iterator
template <class BiIt, class OutIt>
OutIt reverse_copy ( BiIt first, BiIt
last, OutIt result){
while ( first!=last ){
--last;
*result = *last;
result++;
}
return result;
}
332
Find the second element!
template <class T, class It>
It findSecond(It first, It last, const T& what){
???
}
333
Find the second element!
template <class T, class It>
It findSecond(It first, It last, const T& what){
while( first != last && *first != what ){
++first;
}
if( first == last ){
return last;
}
++first;
while( first != last && *first != what ){
++first;
}
return first;
}
334
Containers & Iterators
– vector – Random Access Iterator

– deque - Random Access Iterator
– list – Bidirectional Iterator
– set, map - Bidirectional Iterator
– unordered_set – Forward Iterator
335
Iterator adapters
– Reverse iterators
– Insert iterators
– Stream iterators
336
Reverse iterators
– reverses the direction in which a bidirectional

or random-access iterator iterates through a
range.
– ++ ← → - -
– container.rbegin()
– container.rend()
337
Insert iterators
– special iterators designed to allow algorithms

that usually overwrite elements to instead
insert new elements at a specific position in
the container.
– the container needs to have an insert
member function
338
Insert iterator - Example
//Incorrect
int x[] = {1, 2, 3};
vector<int> v;
copy( x, x+3, v.begin());
//Correct
int x[] = {1, 2, 3};
vector<int> v;
copy( x, x+3, back_inserter(v));
339
Insert iterator - Example
template <class InIt, class OutIt>

OutIt copy( InIt first1, InIt last1,OutIt first2){
while( first1 != last1){
*first2 = *first1;//overwrite → insert
first1++;
first2++;
}
return first2;
}
340
Types of insert iterators
*pos = value;
Type Class Function Creation
Back inserter back_insert_iterator push_back(value) back_inserter(container)
Front inserter front_insert_iterator push_front(value) front_inserter(container)
Inserter insert_iterator insert(pos, value) inserter(container, pos)
341
Stream iterators
● Objective: connect algorithms to streams
cin copy vector
vector copy cout
342
Stream iterator - examples
vector<int> v;
copy(v.begin(), v.end(),
ostream_iterator<int>(cout, ","));
copy(istream_iterator<int>(cin),
istream_iterator<int>(),
back_inserter(v));
343
Problem 1.
● It is given a CArray class
string str[]=
{"apple", "pear", "plum",
"peach", "strawberry", "banana"};
CArray<string> a(str, str+6);
344
Problem 1.
● It is given a Smart API too
Call the doIt function for CArray!
Smart<string> smart;
smart.doIt( ? );
345
Problem 1. - Solution
string str[]= {"apple", "pear", "plum", "peach", "strawberry"};

CArrayIterator<string> cit ( a );
Smart<string> smart;
346
smart.doIt( cit );
Problem 2.
● It is given a CArray class
string str[]=
{"apple", "pear", "plum",
"peach", "strawberry", "banana"};
347
Problem 2.
● It is given a Smarter API
class Smarter{
public:
template <class RaIt>
void doIt( RaIt first, RaIt last ){
while( first != last ){
cout<< *first <<std::endl;
++first;
}
}
};
348
Problem 2.
● Call the doIt function in the given way!

//...
Smarter smart;
smart.doIt( a.begin(), a.end() );
349
Problem 2. - Solution A.
template<class T>
class CArray{
public:
class iterator{
T* poz;
public: ...
};
iterator begin(){ return iterator(array);}

iterator end(){ return iterator(array+size);}
private:
T * array;
int size;
};
350
Problem 2. - Solution A.
class CArray{
public:
class iterator{
T* poz;
public:
iterator( T* poz=0 ): poz( poz ){}
iterator( const iterator& it ){ poz = it.poz; }
iterator& operator=( const iterator& it ){
if( &it == this ) return *this;
poz = it.poz; return *this;}
iterator operator++(){ poz++; return *this; }
iterator operator++( int p ){
iterator temp( *this ); poz++; return temp;}
bool operator == ( const iterator& it )const{
iteratorreturn
begin(){
poz ==return iterator(array);}
it.poz;}
iterator end(){ !=
bool operator return iterator(array+size);}
( const iterator& it )const{
private: return poz != it.poz; }
T * T&
array;
operator*() const { return *poz;}
}; size;
int
};
351
Problem 2. - Solution B.
class CArray{
public:
typedef T * iterator;
iterator begin(){ return array;}

iterator end() { return array+size;}
private:
T * array;
int size;
};
352
Carray → iterator
template <class T>
class CArray{
T * data;
int size;
public:
...
typedef T* iterator;
typedef T value_type;
typedef T& reference;
typedef ptrdiff_t difference_type;
typedef T * pointer;
};
353
Module 9
Function Objects & Lambdas
354
Function object
class
– FunctionObjectType {
public:
return_type operator() (parameters) {
Statements
}
};
355
Function pointer vs. function object
– A function object may have a state

– Each function object has its own type, which can be
passed to a template (e.g. set, map)
– A function object is usually faster than a function pointer
356
Function object as a sorting criteria
class PersonSortCriterion {
public:
bool operator() (const Person& p1, const Person& p2)
const {
if (p1.lastname() != p2.lastname() ){
return p1.lastname() < p2.lastname();
} else{
return p1.firstname()<p2.firstname());
}
};
// create a set with special sorting criterion

set<Person,PersonSortCriterion> coll;
357
Function object with internal state
class IntSequence{
private:
int value;
public:
IntSequence (int initialValue) : value(initialValue) {
}
int operator() () {
return ++value;
}
};
358
[Josuttis]
list<int> coll;
generate_n (back_inserter(coll), // start

9, // number of elements
IntSequence(1)); // generates values,
// starting with 1
359
[Josuttis]
list<int> coll;
generate_n (back_inserter(coll), // start

9, // number of elements
IntSequence(1)); // generates values,
// starting with 1
???
360
Function object with internal state + for_each
[Josuttis]
class MeanValue {
private:
long num; // number of elements
long sum; // sum of all element values
public:
MeanValue () : num(0), sum(0) {}
void operator() (int elem) {
++num; // increment count
sum += elem; // add value
}
double value () {
return static_cast<double>(sum) / num;
}
};
361
function object with internal state + for_each
[Josuttis]
int main()
{
vector<int> coll = { 1, 2, 3, 4, 5, 6, 7, 8 };
MeanValue mv = for_each (coll.begin(), coll.end(),

MeanValue());
cout << "mean value: " << mv.value() << endl;
}
Why to use the return value? http://www.cplusplus.com/reference/algorithm/for_each /
362
Predicates
● Are function objects that return a boolean value

● A predicate should always be stateless
template <typename ForwIter, typename Predicate>

ForwIter std::remove_if(ForwIter beg, ForwIter end,
Predicate op)
363
Predefined function objects
Expression Effect Expression Effect
negate<type>() - param greater_equal<type>() param1 >=

plus<type>() param1 + param2 param2
minus<type>() param1 - param2 logical_not<type>() ! param
multiplies<type>()param1 * param2 logical_and<type>() param1 &&
divides<type>() param1 / param2 param2
modulus<type>() param1 % param2 logical_or<type>() param1 ||
equal_to<type>()param1 == param2 param2
not_equal_to<type>() param1 != bit_and<type>() param1 & param2
param2 bit_or<type>() param1 | param2
less<type>() param1 < param2 bit_xor<type>() param1 ^ param2
greater<type>() param1 > param2
less_equal<type>() param1 <=
param2
364
Syntactic sugar
Lambdas
C++
2011
● a function that you can write inline in your source code
#include <iostream>
int main()
{
auto func = [] () { cout << "Hello world"; };
func(); // now call the function
}
365
Syntactic sugar
Lambdas
– no need to write a separate function or to write a function object

– set
auto comp = [](string x, string y) {
return x > y;
};
set<string, decltype(comp)> s(comp);
//...
for (auto& x : s) {
cout << x << endl;
}
366
Syntactic sugar
Lambda syntax
[ ] ( )opt -> opt
{ }
[ captures ] ( params ) ->ret { statements; }
[ captures ]
What outside variables are available, by value or by reference.
( params )
How to invoke it. Optional if empty.
-> ret
Uses new syntax. Optional if zero or one return statements.
{ statements; }
Herb Sutter: nwcpp.org/may-2011.html
The body of the lambda
367
Syntactic sugar
Examples
– Earlier in scope: Widget w;

– Capture w by value, take no parameters when invoked.
auto lamb = [w] { for( int i = 0; i < 100; ++i ) f(w); };
lamb();
– Capture w by reference, take a const int& when invoked.

auto da = [&w] (const int& i) { return f(w, i); };
int i = 42;
da( i ); Herb Sutter: nwcpp.org/may-2011.html
368
Syntactic sugar
Lambdas == Functors
class __functor {
private:
CaptureTypes __captures;
public:
__functor( CaptureTypes captures )
: __captures( captures ) { }
auto operator() ( params ) → { statements; }

};
369
Syntactic sugar
Capture Example
[ c1, &c2 ] { f(c1, c2); }
class __functor {
private:
C1 __c1; C2& __c2;
public:
__functor( C1 c1, C2& c2 )
: __c1(c1), __c2(c2) { }
void operator() () → { f(__c1, __c2); }

};
370
Syntactic sugar
Parameter Example
[] ( P1 p1, const P2& p2 ){ f(p1, p2); }
class __functor {
public:
void operator() ( P1 p1, const P2& p2) {
f(p1, p2);
}

};
371
Syntactic sugar
Type of Lambdas
auto g = [&]( int x, int y ) { return x > y; };
map<int, int, ? > m( g );
372
Syntactic sugar
Type of Lambdas
map<int, int, ? > m( g );
map<int, int, decltype(g) > m( g );
373
Example
int x = 5;
int y = 12;
auto pos = find_if (
coll.cbegin(), coll.cend(), // range
[=](int i){return i > x && i < y;}// search criterion
);
cout << "first elem >5 and <12: " << *pos << endl;
= symbols are passed

by value
374
Example
vector<int> vec = {1,2,3,4,5,6,7,8,9};

int value = 3;
int cnt = count_if(vec.cbegin(),vec.cend(),
[=](int i){return i>value;});
cout << “Found “ << cnt << “ values > “ << value <<
endl;
375
Module 10
Advanced C++
376
Outline
– Casting. RTTI
– Handling Errors
– Smart Pointers
– Random Numbers
– Regular Expressions
377
Casting & RTTI
378
Casting
– converting an expression of a given type into another type

– traditional type casting:
● (new_type) expression
● new_type (expression)
– specific casting operators:
● dynamic_cast <new_type> (expression)
● reinterpret_cast <new_type> (expression)
● static_cast <new_type> (expression)
● const_cast <new_type> (expression)
379
static_cast<>() vs. C-style cast
– static_cast<>() gives you a compile time checking ability,

C-Style cast doesn't.
– You would better avoid casting, except dynamic_cast<>()
380
Run Time Type Information
– Determining the type of any variable during execution

(runtime)
– Available only for polymorphic classes (having at least one
virtual method)
– RTTI mechanism
● the dynamic_cast<> operator
● the typeid operator
● the type_info struct
381
Casting Up and Down
class Super{
public:
virtual void m1();
};
class Sub: public Super{
public:
virtual void m1();
void m2();
};
Sub mySub;
//Super mySuper = mySub; // SLICE
Super& mySuper = mySub; // No SLICE
mySuper.m1(); // calls Sub::m1() - polymorphism
mySuper.m2(); // ???
382
dynamic_cast<>
class Base{};
Base* basePointer = new Derived();

Derived* derivedPointer = nullptr;
//To find whether basePointer is pointing to Derived type of object
derivedPointer = dynamic_cast<Derived*>(basePointer);
if (derivedPointer != nullptr){
cout << "basePointer is pointing to a Derived class object";
}else{
cout << "basePointer is NOT pointing to a Derived class object";
}
383
dynamic_cast<>
class Person{
public: virtual void print(){cout<<”Person”;};
};
class Employee:public Person{
public: virtual void print(){cout<<”Employee”;};
};
public: virtual void print(){cout<<”Manager”;};
};
vector<Person*> v;
v.push_back(new Person());
v.push_back(new Employee());
v.push_back( new Manager());
...
384
dynamic_cast<>
class Person{
};
};
};
vector<Person*> v; Write a code that counts

v.push_back(new Person()); the number of employees!
...
385
dynamic_cast<>
class Person{
}; Write a code that counts
class Employee:public Person{ the number of employees!
};
}; Employee * p = nullptr;
for( Person * sz: v ){
vector<Person*> v; p = dynamic_cast<Employee *>( sz );
v.push_back(new Person()); if( p != nullptr ){
++counter;
... }
}
386
Which solution is better? (Solution 1)
void speak(const Animal& inAnimal) {

if (typeid (inAnimal) == typeid (Dog)) {
cout << "VauVau" << endl;
} else if (typeid (inAnimal) == typeid (Bird)) {
cout << "Csirip" << endl;
}
}
….
Bird bird; Dog d;
speak(bird); speak( dog ); ???
387
Which solution is better? (Solution 2)
class Animal{
public:
virtual void speak()=0;
};
class Dog:public Animal{
public:
virtual void speak(){cout<<"VauVau"<<endl;};
};
class Bird: public Animal{
public:
virtual void speak(){cout<<"Csirip"<<endl;};
};
void speak(const Animal& inAnimal) {

inAnimal.speak();
}
Bird bird; Dog d;
speak(bird); speak( dog );
388
typeid
class Person{
}; Write a code that counts the number of
employees (the exact type of the objects is
Employee)!
};
};
counter = 0;
vector<Person*> v; for( Person * sz: v ){
v.push_back(new Person()); if( typeid(*sz) == typeid(Employee) ){
v.push_back(new Employee()); ++counter;
... }
}
389
Typeid usage
a and b are of different types:
#include <iostream> a is: Pi
#include <typeinfo> b is: i
int main ()
{
int * a;
int b;
a=0; b=0;
if (typeid(a) != typeid(b))
{
cout << "a and b are of different types:\n";
cout << "a is: " << typeid(a).name() << '\n';
cout << "b is: " << typeid(b).name() << '\n';
}
return 0;
}
390
Handling Errors
391
Handling Errors
– C++ provides Exceptions as an error

handling mechanism
– Exceptions: to handle exceptional but not

unexpected situations
392
Return type vs. Exceptions
Return type: Exceptions:

● caller may ignore ● easier
● caller may not propagate ● more consistent
upwards ● safer
● doesn't contain sufficient ● cannot be ignored (your
information program fails to catch an
exception → will terminate)
● can skip levels of the call
stack
393
<stdexcept>
Exceptions
int SafeDivide(int num, int den)

{
if (den == 0)
throw invalid_argument(“Divide by zero”);
return num / den;
}
int main()
{
try { Discussion??!!!
cout << SafeDivide(5, 2) << endl;
} catch (const invalid_argument& e) {
cout << “Caught exception: “ << e.what() << endl;
}
return 0;
}
394
<stdexcept>
Exceptions
int SafeDivide(int num, int den)

{
if (den == 0)
throw invalid_argument(“Divide by zero”);
return num / den;
}
int main()
{
try {
cout << “Caught exception: “ << e.what() << endl;
}
return 0;
}
It is recommended to catch exceptions by const reference.
395
<stdexcept>
HandExceptions
try {
// Code that can throw exceptions
// Handle invalid_argument exception
} catch (const runtime_error& e) {
// Handle runtime_error exception
} catch (...) {
// Handle all other exceptions
}
Any exception
396
<stdexcept>
Throw List
void func() throw (extype1, extype2){

// statements
}
The throw list is not enforced at compile time!
397
<stdexcept>
Throw List
http://www.cplusplus.com/doc/tutorial/exceptions/
void func() throw (){

// statements
}
void func() noexcept{

// statements C++
} 2011
398
The Standard Exceptions
<stdexcept>
http://cs.stmarys.ca/~porter/csc/ref/cpp_standlib.html
399
User Defined Exception
<stdexcept>
– It is recommended to inherit
exception
directly or indirectly from the standard
exception class
your_exception
400
User Defined Exception
<stdexcept>
class FileError : public runtime_error{

public:
FileError(const string& fileIn):runtime_error (“”)
, mFile(fileIn) {}
virtual const char* what() const noexcept{
return mMsg.c_str();
}
string getFileName() { return mFile; }
protected:
string mFile, mMsg;
};
401
Smart Pointers
402
Outline
– The problem: raw pointers

– The solution: smart pointers
– Examples
– How to implement smart pointers
403
Why Smart Pointers?
– When to delete an object?

● No deletion → memory leaks
● Early deletion (others still pointing to) →
dangling pointers
● Double-freeing
404
Smart Pointer Types
– unique_ptr #include <memory>

– shared_ptr
C++
– weak_ptr 2011
It is recommended to use smart pointers!
405
Smart Pointers
– Behave like built-in (raw) pointers

– Also manage dynamically created objects
● Objects get deleted in smart pointer destructor
– Type of ownership:
● unique
● shared
406
The good old pointer
void oldPointer(){
Foo * myPtr = new Foo();
myPtr->method();
} Memory leak
407
The good Old pointer
void oldPointer1(){
myPtr->method();
} Memory leak
void oldPointer2(){
Foo * myPtr = new Foo(); Could cause
myPtr->method(); memory leak
delete myPtr; When?
}
408
The Old and the New
void oldPointer(){
myPtr->method();
} Memory leak
void newPointer(){
shared_ptr<Foo> myPtr (new Foo());
myPtr->method();
}
409
Creating smart pointers
void newPointer(){
shared_ptr<Foo> myPtr (new Foo());
myPtr->method();
}
void newPointer(){
auto myPtr = make_shared<Foo>();
myPtr->method();
}
Static
factory method 410
unique_ptr
– it will automatically free the resource in case of the

unique_ptr goes out of scope.
411
shared_ptr
– Each time a shared_ptr is assigned

● a reference count is incremented (there is one more
“owner” of the data)
– When a shared_ptr goes out of scope
● the reference count is decremented
● if reference_count = 0 the object referenced by the
pointer is freed.
412
Implementing your own smart pointer class
CountedPtr<Person> p(new Person("Para Peti",1980));
:Person
p: CountedPtr<Person>
"Para Peti"
ptr
count
1
413
Implementing your own smart pointer class
CountedPtr<Person> p1 = p;
CountedPtr<Person> p2 = p;
p :Person
ptr
count "Para Peti"
p1
3
ptr
count
p2
ptr
count
414
Implementation (1)
template < class T>

class CountedPtr{
T * ptr;
long * count;
public:
...
};
415
Implementation (2)
CountedPtr( T * p = 0 ):ptr( p ),
count( new long(1)){
}
CountedPtr( const CountedPtr<T>& p ): ptr( p.ptr),

count(p.count){
++(*count);
}
~CountedPtr(){
--(*count);
if( *count == 0 ){
delete count; delete ptr;
}
}
416
Implementation (3)
CountedPtr<T>& operator=( const CountedPtr<T>& p ){
if( this != &p ){
--(*count);
if( *count == 0 ){ delete count; delete ptr; }
this->ptr = p.ptr;
this->count = p.count;
++(*count);
}
return *this;
}
T& operator*() const{ return *ptr;}
T* operator->() const{ return ptr;}
417
Shared ownership with shared_ptr
Container of Container of
smart pointers smart pointers
sp sp
sp sp
sp sp
o1 o2 o3
http://umich.edu/~eecs381/handouts/C++11_smart_ptrs.pdf
418
Problem with shared_ptr
Container of Objects pointing to another

smart pointers object with a smart pointer
sp sp
sp
sp
sp sp
419
Solution: weak_ptr
Container of Objects pointing to another

smart pointers object with a weak pointer
sp wp
sp
sp
wp wp
420
weak_ptr
– Observe an object, but does not influence its

lifetime
– Like raw pointers - the weak pointers do not keep
the pointed object alive
– Unlike raw pointers – the weak pointers know
about the existence of pointed-to object
421
How smart pointers work
shared_ptr
Manager object Managed object
sp1
Pointer
Shared count: 3
sp2
Weak count: 2
sp3
wp1 wp2
weak_ptr
422
Restrictions in using smart pointers
– Can be used to refer to objects allocated with

new (can be deleted with delete).
– Avoid using raw pointer to the object refered
by a smart pointer.
423
Inheritance and shared_ptr
void greeting( shared_ptr<Person>& ptr ){

cout<<"Hello "<<(ptr.get())->getFname()<<" "
<<(ptr.get())->getLname()<<endl;
}

shared_ptr<Person> ptr_person(new Person("John","Smith"));
cout<<*ptr_person<<endl;
greeting( ptr_person );
shared_ptr<Manager> ptr_manager(new Manager("Black","Smith", "IT"));

cout<<*ptr_manager<<endl;
ptr_person = ptr_manager;
cout<<*ptr_person<<endl;
return 0;
}
424
unique_ptr usage
// p owns the Person

unique_ptr<Person> uptr(new Person("Mary", "Brown"));
unique_ptr<Person> uptr1( uptr ); //ERROR – Compile time
unique_ptr<Person> uptr2; //OK. Empty unique_ptr
uptr2 = uptr1; //ERROR – Compile time

uptr2 = move( uptr ); //OK. uptr2 is the owner
cout<<"uptr2: "<<*uptr2<<endl; //OK
cout<<"uptr : "<<*uptr <<endl; //ERROR – Run time
unique_ptr<Person> uptr3 = make_unique<Person>("John","Dee");

cout<<*uptr3<<endl;
Static
Factory Method 425
unique_ptr usage (2)
unique_ptr<Person> uptr1 =
make_unique<Person>("Mary","Black");
cout<<*uptr2<<endl;
vector<unique_ptr<Person> > vec;

vec.push_back( uptr1 );
vec.push_back( uptr2 ); Find the errors
and correct them!!!
cout<<"Vec [";
for( auto e: vec ){
cout<<*e<<" ";
}
cout<<"]"<<endl;
426
unique_ptr usage (2)
unique_ptr<Person> uptr1 =
make_unique<Person>("Mary","Black");
cout<<*uptr2<<endl;
vector<unique_ptr<Person> > vec;

vec.push_back( move( uptr1 ) );
vec.push_back( move( uptr2 ) );
cout<<"Vec [";
for( auto& e: vec ){
cout<<*e<<" ";
}
cout<<"]"<<endl;
427
Module 11
I/O Streams
428
Outline
– Using Streams
– String Streams
– File Streams
– Bidirectional I/O
429
Using Streams
Prog-
ram
Input Stream Output Stream

● file ● file
● keypad ● screen
● program ● program
stream:
● is data flow
● direction
● associated source and destination
430
Using Streams
cin An input stream, reads data from the “input console.”
cout A buffered output stream, writes data to the output console.
cerr An unbuffered output stream, writes data to the “error console”
clog A buffered version of cerr.
431
Using Streams
– Stream:
● includes data
● has a current position
– next read or next write
432
Using Streams
ios_base basic_streambuf<>
streambuf, wstreambuf
basic_ios<>
ios, wios
basic_istream<> basic_ostream<>
istream, wistream ostream, wostream
basic_iostream<>
iostream, wiostream
433
Using Streams
– Output stream:
● inserter operator <<
● raw output methods (binary):
– put(), write()
void rawWrite(const char* data, int dataSize){

cout.write(data, dataSize);
}
void rawPutChar(const char* data, int charIndex)

{
cout.put(data[charIndex]);
}
434
Using Streams
– Output stream:
● most output streams buffer data (accumulate)
● the stream will flush (write out the accumulated data)
when:
– an endline marker is reached ('\n', endl)
– the stream is destroyed (e.g. goes out of scope)
– the stream buffer is full
– explicitly called flush()
435
Using Streams
– Manipulators:
● objects that modify the behavior of the stream
– setw, setprecision
– hex, oct, dec
– C++11: put_money, put_time
int i = 123;
printf(“This should be ' 123': %6d\n”, i);
cout <<“This should be ' 123': “ << setw(6) << i << endl;
436
Using Streams
– Input stream:
● extractor operator >>
– will tokenize values according to white spaces
● raw input methods (binary): reads an input having
more than one word
– get(): avoids tokenization
string readName(istream& inStream)

{
string name;
char next;
while (inStream.get(next)) {
name += next;
}
return name;
}
437
Using Streams
– Input stream:
● getline(): reads until end of line
reads an input having

more than one word
string myString;
getline(cin, myString);
438
Using Streams
– Input stream:
● getline(): reads until end of line
reads an input having

more than one word
string myString;
getline(cin, myString);
Reads up to new line character

Unix line ending: '\n'
Windows line ending: '\r' '\n'
The problem is that getline leaves
the '\r' on the end of the string.
439
Using Streams
– Stream's state:
● every stream is an object → has a state
● stream's states:
– good: OK
– eof: End of File
– fail: Error, last I/O failed
– bad: Fatal Error
440
Using Streams
– Find the error!
list<int> a;
int x; Input:
1
while( !cin.eof() ){ 2
cin>>x; 3
a.push_back( x ); (empty line)
} a: 1, 2, 3, 3
441
Using Streams
– Handling Input Errors:
● while( cin )
● while( cin >> ch )
int number, sum = 0; int number, sum = 0;
while ( true ) { while ( cin >> number ){
cin >> number; sum += number;
if (cin.good()){ }
sum += number;
} else{
break;
}
}
442
String Streams
– <sstream>
● ostringstream
● istringstream
● stringstream
string s =”12.34”; double d =12.34;

stringstream ss(s); stringstream ss;
double d; ss<<d;
ss >> d; string s = “szam:”+ss.str()
443
File Streams
{
ifstream ifs("in.txt");//Constructor
if( !ifs ){
//File open error
}
//Destructor call will close the stream
}
{
ifstream ifs;
ifs.open("in.txt");
//...
ifs.close();
//...
}
444
File Streams
– Byte I/O
ifstream ifs("dictionary.txt");
// ios::trunc means that the output file will be
// overwritten if exists
ofstream ofs("dict.copy", ios::trunc);
char c;
while( ifs.get( c ) ){
ofs.put( c );
}
445
File Streams
– Byte I/O
– Using rdbuf() - quicker
ifstream ifs("dictionary.txt");
– // ios::trunc means that the output file will be
// overwritten if exists
ofstream ofs("dict.copy", ios::trunc);
if (ifs && ofs) {

ofs << ifs.rdbuf();
}
446
Object I/O
– Operator overloading
istream& operator>>( istream& is, T& v ){
//read v
return is;
}
ostream& operator<<(ostream& is, const T& v ){

//write v
return os;
}
447

CPP v1.1

Uploaded by

CPP v1.1

Uploaded by

CPP

Object-Oriented Programming Generic Programming

#include <iostream> Preprocessor directive

int main(){ The main function

– User defined header

Result status The number

– cin: standard input

int main(int argc, char** argv) {

● Online C++ compilers:

enum Vegetable{ tomato, cucumber, onion};

enum class Mark {

Mark myMark( int value ){

for( auto& e: elements){

– run time array

string name = firstName+ " "+ lastName; cout<<name<<endl;

int &ri = i; //OK ri refers to (is another name for) i

int &ri1; //ERROR: a reference must be initialized

– Exception handling = a mechanism for dealing with problems

● throwing an exception – detecting an unexpected situation

● catching an exception – taking appropriate action

double divide( double m, double n){

– RAAI (Resource Acquisition Is Initialization) Classes:

– class definition – implementation

string Employee::getFirstName() const{

#include <string> Employee::Employee() :

Employee *demp = new Employee();

– when an object is created,

– when you need ● memory allocation

Explicitly forcing the automatic generation of a default constructor by the compiler.

double → int conversion

Inhibiting the automatic generation of a default constructor by the compiler.

ConstRef::ConstRef( int& inI ){

#ifndef STACK_H #include "Stack.h"

int main(){ mCapacity: 3

#ifndef STACK_H #include "Stack.h"

// Node is a nested structure definition local to this class

// private class members

inline void setValue( int value ){

What is the difference between these two arrays?

Point * t1 = new Point[ 4]; Point t1[ 4];

t1 :Point :Point :Point :Point

:Point :Point :Point :Point

Complex z1(1,2), z2(2,3), z3;

cout<<"Number of complexes: "<<z1.getNumComplex()<<endl;

re: 2 Each object has

void f1(int x) {x = x + 1;}

void f1(Point p); copy constructor will be used on the argument

void Test::print() const{

void Test::print( const Test& what ){

void print( const Test& what ){

The reference should be to

● a nested class B declared in a protected section of a class A:

● a nested class B declared in a public section of a class A:

– assignment operator (see next module)

● Ownership transfer to the new

– All STL classes have been extended to support move

static A getInstCopy(){return inst;} A temporary copy of instance → rvalue

Operator overloading: a service to the clients of the class

#ifndef COMPLEX_H #include "Complex.h"

Complex Complex::operator-( const Complex& z) const{

Complex operator-( const Complex& z1, const Complex& z2 ){

Complex& Complex::operator++(){ //prefix

Complex Complex::operator++( int ){ //postfix

istream& operator>>( istream& is, Complex& c){

Complex z1(1,2), z2(2,3), z3(1,1);

Complex z1(1,2), z2(3,4); //Constructor

double& CArray::operator[]( int index ){

bool operator <( const Point& p1, const Point& p2){

Employee e1; e1.setFirstName("Barbara");

for( auto& emp : s){

– Which one to use?

Complex::operator string() const{

● Provide a global point of access to the object.

– You can do with inheritance

– ex. department - depa rtm e nt: string

class Manager:public Employee{