COMP6771 Week 3.1

Object-Oriented Programming

Scope

  • The scope of a variable is the part of the program where it is accessible
    • Scope starts at variable definition
    • Scope (usually) ends at next "}"
    • You're probably familiar with this even if you've never seen the term
  • Define variables as close to first usage as possible
  • This is the opposite of what you were taught in first year undergrad
    • Defining all variables at the top is especially bad in C++

Object Lifetimes

  • An object is a piece of memory of a specific type that holds some data
    • All variables are objects
    • Unlike many other languages, this does not add overhead
  • Object lifetime starts when it comes in scope
    • "Constructs" the object
    • Each type has 1 or more constructor that says how to construct it
  • Object lifetime ends when it goes out of scope
    • "Destructs" the object
    • Each type has a different "destructor" which tells the compiler how to destroy it

Construction

  • Eg. https://en.cppreference.com/w/cpp/container/vector/vector
  • Generally use () to call functions, and {} to construct objects
    • () can only be used for functions, and {} can be used for either
    • There are some rare occasions these are different
      • Sometimes it is ambiguous between a constructor and an initialize list
int main() {
  std::vector<int> v11; // Calls 0-argument constructor. Creates empty vector.
  // There's no difference between these:
  // T variable = T{arg1, arg2, ...}
  // T variable{arg1, arg2, ...}
  std::vector<int> v12{}; // No different to first
  std::vector<int> v13 = std::vector<int>(); // No different to the first
  std::vector<int> v14 = std::vector<int>{}; // No different to the first

  std::vector<int> v3{v2.begin(), v2.end()}; // constructed with an iterator
  std::vector<int> v4{v3}; // Constructed off another vector

  std::vector<int> v51{5, 2}; // Initialiser-list constructor {5, 2}
  std::vector<int> v52(5, 2); // Count + value constructor (5 * 2 => {2, 2, 2, 2, 2})
}

Construction

  • Also works for your basic types
    • But the default constructor has to be manually called
      • This potential bug can be hard to detect due to how function stacks work (variable may happen to be 0)
      • Can be especially problematic with pointers
int main() {
  int n; // not constructed (memory contains previous value)
  int n2{}; // Default constructor (memory contains 0)
  int n3{5};

  // This version is nice because it gives us an error.
  int n4{5.5};
  // You need to explictly tell it you want this.
  int n6{static_cast<int>(5.5)};

  // Not so nice. No error
  int n5 = 5.5;
}

Why are object lifetimes useful?

Can you think of a thing where you always have to remember to do something when you're done?

void ReadWords(const std::string& filename) {
  std::ifstream f{filename};
  std::vector<std::string> words;
  std::copy(std::istream_iterator<std::string>{f}, {}, std::back_inserter{words});
  f.close();
}
  • What happens if we omit f.close() here (assume similar behavior to c/java/python)?
  • How easy to spot is the mistake
  • How easy would it be for a compiler to spot this mistake for us?
    • How would it know where to put the f.close()?

RAII

  • Resource acquisition is initialisation

  • A concept where we encapsulate resources inside objects

    • Acquire the resource in the constructor​
    • Release the resource in the destructor
    • eg. Memory, locks, files

  • Every resource should be owned by either:

    • Another resource (eg. smart pointer, data member)

    • The stack

    • A nameless temporary variable

Noexcept

  • Exceptions will be covered in week 5, but the short version is that they are recoverable, but critical errors
  • A noexcept-specified function tells the compiler not to generate recovery code
  • An exception thrown in a noexcept function will terminate your program
  • Use noexcept to guarantee that callers needn't worry about exception-handling.
  • You can use noexcept to say that you don't mind your whole program ending if something goes wrong in this function.

Destructors

  • Call when the object goes out of scope
    • What might this be handy for?
    • Does not occur for reference objects (why?)
  • Marked noexcept (why?)
  • Why might destructors be handy?

Destructors

  • Called when the object goes out of scope
    • What might this be handy for?
    • Does not occur for reference objects (why?)
  • Marked noexcept (why?)
  • Why might destructors be handy?
    • Freeing pointers
    • Closing files
    • Unlocking mutexes (from multithreading)
    • Aborting database transactions
class MyClass {
  ~MyClass() noexcept;
};
MyClass::~MyClass() noexcept {
  // Definition here
}

What is OOP

  • A class uses data abstraction and encapsulation to define an abstract data type:
    • Interface: the operations used by the user (an API)
    • Implementation: the data members the bodies of the functions in the interface and any other functions not intended for general use
    • Abstraction: separation of interface from implementation
    • Encapsulation: enforcement of this via information hiding
    • Example: Bookstore - bookstore.h (interface), bookstore.cpp (implementation), user code (knows the interface).

C++ classes

  • A class:
    • Defines a new type
    • Is created using the keywords class or struct
    • May define some members (functions, data)
    • Contains zero or more public and private sections
    • Is instantiated through a constructor
  • A member function:
    • must be declared inside the class
    • may be defined inside the class (it is then inline by default)
    • may be declared const, when it doesn’t modify the data members
  • The data members should be private, representing the state of an object.

Abstraction and encapsulation

  • Abstraction is separating the interface from the implementation
  • Encapsulation is hiding details about class representation and implementation
    • An object’s state can only be accessed/modified via the public interface

Advantages:

  • Object state is protected from user-level errors
    • Users can't break invariants by changing something
  • Class implementation may evolve over time
    • If you change a variable or a private function, users don't need to change anything

Incomplete types

  • An incomplete type may only be used to define pointers and references, and in function declarations (but not definitions)
  • Because of the restriction on incomplete types, a class cannot have data members of its own type.

 

 

 

  • But the following is legal, since a class is considered declared once its class name has been seen:
struct Node {
  int data;
  // Node is incomplete - this is invalid
  // This would also make no sense. What is sizeof(Node)
  Node next;
};
struct Node {
  int data;
  Node* next;
};

Member access control

  • This is how we support encapsulation and information hiding in C++
class Foo {
 public:
  // Members accessible by everyone
  Foo();

 protected:
  // Members accessible by members, friends, and subclasses
  // Will discuss this when we do advanced OOP in future weeks.

 private:
  // Accessible only by members and friends
  void PrivateMemberFunction();

  int private_data_member_;

 public:
  // May define multiple sections of the same name
};

Classes and structs in C++

  • A class and a struct in C++ are almost exactly the same
  • The only difference is that:
    • All members of a struct are public by default
    • All members of a class are private by default
    • People have all sorts of funny ideas about this. This is the only difference
  • We use structs only when we want a simple type with little or no methods and direct access to the data members (as a matter of style)
    • This is a semantic difference, not a technical one
    • A std::pair or std::tuple may be what you want, though

Friends

  • A class may declare friend functions or classes
    • Those functions / classes are non-member functions that may access private parts of the class
    • This is, in general, a bad idea, but there are a few cases where it may be required
      • Nonmember operator overloads (will be discussing soon)
      • Related classes
        • A Window class might have WindowManager as a friend
        • A TreeNode class might have a Tree as a friend
        • Container could have iterator_t<Container> as a friend
          • Though a nested class may be more appropriate
    • Use friends when:
      • The data should not be available to everyone
      • There is a piece of code very related to this particular class

Class Scope

  • Anything declared inside the class needs to be accessed through the scope of the class
    • Scopes are accessed using "::" in C++
// foo.h

class Foo {
 public:
  // Equiv to typedef int Age
  using Age = int;

  Foo();
  Foo(std::istream& is);
  ~Foo();

  void MemberFunction();
};
// foo.cpp
#include "foo.h"

Foo::Foo() {
}

Foo::Foo(std::istream& is) {
}

Foo::~Foo() {
}

void Foo::MemberFunction() {
  Foo::Age age;
}

This pointer

  • A member function has an extra implicit parameter, named this
    • This is a pointer to the object on behalf of which the function is called
    • A member function does not explicitly define it, but may explicitly use it
    • The compiler treats an unqualified reference to a class member as being made through the this pointer.
    • The this pointer always has top-level const
  • For the next few slides, we'll be taking a look at the BookSale example in the course repo

This pointer

  • A member function has an extra implicit parameter, named this
    • This is a pointer to the object on behalf of which the function is called
    • A member function does not explicitly define it, but may explicitly use it
    • The compiler treats an unqualified reference to a class member as being made through the this pointer.
    • The this pointer always has top-level const
  • For the next few slides, we'll be taking a look at the BookSale example in the course repo

Const objects

  • Member functions are by default only be possible on non-const objects
    • You can declare a const member function which is valid on const objects
    • A const member function may only modify mutable members
      • A mutable member should mean that the state of the member can change without the state of the object changing
      • Good uses of mutable members are rare
      • Mutable is not something you should set lightly
      • One example where it might be useful is a cache
  • Let's make the BookSale class const correct

Are the following correct

Are the following correct?

Sales_data a{"Harry Potter"};
Sales_data b{"Harry Potter"};

a.combine(b).print(std::cout);
a.print(std::cout).combine(b);

Are the following correct

Are the following correct?

 

 

 

 

  •  The combine/print is fine
  • The print/combine fails since print returns a const reference through which we cannot call a nonconst member
  • Four possible ways to get it to compile. Discuss.
    • Make combine a const function
    • Make print a non-const function
    • Add an overload to print for non-const
    • Change the user code
Sales_data a{"Harry Potter"};
Sales_data b{"Harry Potter"};

a.combine(b).print(std::cout);
a.print(std::cout).combine(b);

Constructors

  • Constructors define how class data members are initalised
  • A constructor has the same name as the class and no return type
  • Default initalisation is handled through the default constructor
  • Unless we define our own constructors the compile will declare a default constructor
    • This is known as the synthesized default constructor
for each data member in declaration order
  if it has an in-class initialiser
    Initialise it using the in-class initialiser
  else if it is of a built-in type (numeric, pointer, bool, char, etc.)
    do nothing (leave it as whatever was in memory before)
  else
    Initialise it using its default constructor

The synthesized default constructor

  • Is generated for a class only if it declares no constructors
  • For each member, calls the in-class initialiser if present
    • Otherwise calls the default constructor (except for trivial types like int)
  • Cannot be generated when any data members are missing both in-class initialisers and default constructors
class C {
  int i{0}; // in-class initialiser
  int j; // Untouched memory
  A a;
  // This stops default constructor
  // from being synthesized.
  B b;
};
class A {
  int a_;
};
class B {
  B(int b): b_{b} {}
  int b_;
};

Constructor initialiser list

  • The initialisation phase occurs before the body of the constructor is executed, regardless of whether the initialiser list is supplied
  • A constructor will:
    1. Construct all data members in order of member declaration (using the same rules as those used to initialise variables)
    2. Execute the body of constructor: the code may assign values to the data members to override the initial values

Constructor initialiser list

class NoDefault {
  NoDefault(int i);
}

class B {
  // Constructs s_ with value "Hello world"
  B(int& i): s_{"Hello world"}, const_{5}, no_default{i}, ref_{i} {}
  // Doesn't work - constructed in order of member declaration.
  B(int& i): s_{"Hello world"}, const_{5}, ref_{i}, no_default{ref_} {}
  B(int& i) {
    // Constructs s_ with an empty string, then reassigns it to "Hello world"
    // Extra work done (but may be optimised out).
    s_ = "Hello world";

    // Fails to compile
    const_string_ = "Goodbye world";
    ref_ = i;
    // This is fine, but it can't construct it initially.
    no_default_ = NoDefault{1};
  }

  std::string s_;
  // All of these will break compilation if you attempt to put them in the body.
  const int const_;
  NoDefault no_default_;
  int& ref_;
};

Delegating constructors

  • A constructor may call another constructor inside the initialiser list
    • Since the other constructor must construct all the data members, do not specify anything else in the constructor initialiser list
    • The other constructor is called completely before this one.
    • This is one of the few good uses for default values in C++
      • Default values may be used instead of overloading and delegating constructors

Static members

  • Both data and function members may be declared static
  • These are essentially globals defined inside the scope of the class
    • Use static members when something is associated with a class, but not a particular instance
    • Static data has global lifetime (program start to program end)
// For use with a database
class User {
  static std::string table_name;
  static std::optional<User> query(const std::string& username);
  
  void commit();
  std::string username;
}

User user = *User::query("Alice");
user.username = "Bob"
User::commit(); // fails to compile (commit is not static)
user.commit();

std::cout << User::table_name;
std::cout << User::username; // Fails to compile

Explicit type conversions

  • If a constructor for a class has 1 parameter, the compiler will create an implicit type conversion from the parameter to the class
  • This may be the behaviour you want
class Age {
  Age(int age);
};

// Explicitly calling the constructor
Age age{20};
// Attempts to use an integer
// where an age is expected.
// Implicit conversion done.
// This seems reasonable.
Age age = 20;
class IntVec {
  // This one allows the implicit conversion
  IntVec(int length): vec_(length, 0);

  This one disallows it.
  explicit IntVec(int length): vec_(length, 0);

  std::vector<int> vec_;
};

// Explictly calling the constructor.
IntVec container{20};

// Implicit conversion.
// Probably not what we want.
IntVec container = 20;

OOP design

  • There are several special functions that we must consider when designing classes
  • For each of these functions, ask yourself:
    • Is it sane to be able to do this?
    • Does it have a well-defined, obvious implementation
  • If the answer to either of these is no, write "<function declaration> = delete;"
  • Then ask yourself "is this the behaviour of the compiler-synthesized one"
    • If so, write "<function declaration> = default;"
    • If not, write your own definition
  • Let's discuss these questions for these types over the next few slides:
    • std::vector
    • Mutex
    • Pointer

Copying constructor

  • Constructs one object to be a copy of another
  • The compiler-generated copy-constructor just calls each member's copy constructor in order of declaration
class T {
  T(const T&);
};

Copying assignment

  • Like a copy constructor, but the destination is already constructed
  • Requires destroying the old data, and constructing the new data
  • Copy-and-swap idiom is an elegant way of doing this
    • It constructs then destructs. Since construction might fail, it should go first
    • Requires move assignment to be defined
  • Takes in an lvalue
  • Compiler-generated one performs memberwise copy-assignment operator
class T {
  // A copy-assignment operator
  T& operator=(const T& original);

  // The copy-and-swap idiom
  // This is also a copy-assignment operator
  T& operator=(T copy) {
    std::swap(*this, copy);
    return *this;
  }
};
MyClass base;
MyClass copy_constructed = base;

MyClass copy_assigned;
copy_assigned = base;

Rvalue references

  • Rvalue references look like T&& (lvalue is T&)
  • An lvalue denotes an object whose resource cannot be reused
    • Most objects (eg. variable, variable[0])
    • Once the lvalue reference goes out of scope, it may still be needed
  • An rvalue denotes an object whose resources can be reused
    • eg. Temporaries (MyClass object in f(MyClass{}))
    • When someone passes it to you, they don't care about it once you're done with it


  • “The object that x binds to is YOURS. Do whatever you like with it, no one will care anyway”
  • Like giving a copy to f… but without making a copy.
void f(MyClass&& x);

Rvalue references

void inner(int&& value) {
  ++value;
  std::cout << value << '\n';
}

void outer(int&& value) {
  inner(value); // This fails? Why?
  std::cout << value << '\n';
}

int main() {
  f1(1); // This works fine.
  int i;
  f2(i); // This fails because i is an lvalue.
}
  • An rvalue reference formal parameter means that the value was disposable from the caller of the function
    • If outer modified value, who would notice / care?
      • The caller (main) has promised that it won't be used anymore
    • If inner modified value, who would notice / care?
      • The caller (outer) has never made such a promise.
      • An rvalue reference parameter is an lvalue inside the function

std::move

// Looks something like this.
T&& move(T& value) {
  return static_cast<T&&>(value);
}
  • Simply converts it to an rvalue
    • This says "I don't care about this anymore"
    • All this does is allow the compiler to use rvalue reference overloads
void inner(int&& value) {
  ++value;
  std::cout << value << '\n';
}

void outer(int&& value) {
  inner(std::move(value));
  // Value is now in a valid but unspecified state.
  // Although this isn't a compiler error, this is bad code.
  // Don't access variables that were moved from, except to reconstruct them.
  std::cout << value << '\n';
}

int main() {
  f1(1); // This works fine.
  int i;
  f2(std::move(i));
}

Move constructor

  • Always should be declared noexcept
  • Unless otherwise specified, objects that have been moved from are in a valid but unspecified state
  • Will likely be faster than the copy constructor
  • Compiler-generated one performs memberwise move-construction
class T {
  T(T&&) noexcept;
};

Move assignment

  • Always should be declared noexcept
  • Like the move constructor, but the destination is already constructed
  • Compiler-generated one performs memberwise move-assignment
class T {
  T& operator=(T&&) noexcept;
};

Object lifetimes

To create safe object lifetimes in C++, we always attach the lifetime of one object to that of something else

  • A variable in a function is tied to its scope
  • A data member is tied to the lifetime of the class instance
  • An element in a std::vector is tied to the lifetime of the vector
  • A heap object should be tied to the lifetime of whatever object created it
  • Examples of bad programming practice
    • An owning raw pointer is tied to nothing
    • A C-style array is tied to nothing
  • Strongly recommend watching the first 44 minutes of Herb Sutter's cppcon talk "Leak freedom in C++... By Default"

Constructing wrapper types

class MyClass {
  MyClass();
  MyClass(int);
  MyClass(const MyClass&);
  MyClass(MyClass&&);
  int GetValue();
};

// calls default constructor
std::optional<MyClass> opt1 = std::make_optional<MyClass>();
// calls int constructor
std::optional<MyClass> opt2 = std::make_optional<MyClass>(5);
// calls copy constructor
std::optional<MyClass> opt3 = *opt1;
// calls move constructor
std::optional<MyClass> opt4 = std::move(*opt1);
opt4->GetValue();

// Similar for make_unique and make_shared, but have to manually move / copy values
std::shared_ptr<MyClass> sp1 = std::make_shared<MyClass>();
std::shared_ptr<MyClass> sp2 = sp1;
std::shared_ptr<MyClass> sp3 = std::move(sp1);
MyClass* p2 = sp1.get();

std::unique_ptr<MyClass> up1 = std::make_unique<MyClass>(*sp1);
std::unique_ptr<MyClass> up2 = *up1; // But have to manually move / copy values like above
std::unique_ptr<MyClass> up3 = up1; // Fails (no copy constructor)
std::unique_ptr<MyClass> up4 = std::move(up1);
MyClass* p1 = up1.get();
up1->GetValue();

Namespaces

  • A namespace encapsulates a set of functions, classes, and other namespaces
  • If I have a really big piece of code, with several external pieces of code, what are the chances of two functions / classes having the same name somewhere in the code?
  • You've already seen one namespace used a lot, without knowing it (std)
// path/to/file.h

namespace path {
namespace to {

class MyClass {
}

void MyFn();
  
} // namespace to
} // namespace path
// main.cpp
#include "path/to/file.h"

int main() {
  path::to::MyClass myClass;
  path::to::MyFn();
}
// path/to/file.cpp

namespace path {
namespace to {

MyFn() {
}
  
} // namespace to
} // namespace path

The using keyword

  • Several ways to use the using keyword
    • These are the two most important ways
// Far cleaner than typedef, and the
// arguments are the right way around!
int main() {
  using intvec = std::vector<int>;
  intvec vec;

  C::iterator_t it;
}

class IteratorC {
}

class C {
  using iterator_t = IteratorC;
}
#include "path/to/file.h"

int main() {
  // Imports a single thing
  // into the current scope.
  using path::to::MyClass;
  MyClass myClass;
}

Using declaration

Type alias

ADL

  • Due to a feature of C++ called "Argument dependent lookup", use the using declaration for some specific functions
    • This is a complex topic to be discussed in later weeks
  • This is only relevant when using non-standard types
  • Use this for the following functions
    • std::swap
    • std::[cr]begin
    • std::[cr]end
    • std::empty
    • std::size
    • std::data
#include "myclass.h"

using std::begin;
using std::end;
using std::swap;

int main() {
  std::vector<MyClass> vec{{}, {}};
  swap(vec[0], vec[1]);
  for (auto it = begin(vec); it != end(vec); ++it) {
    std::cout << *it << '\n';
  }
}

Argument dependent lookup

  • When looking up an unqualified function, the compiler first looks at the namespace of the type of the arguments
    • If it contains a matching function declaration, it uses that
    • Otherwise, it falls back to the normal function lookup
  • Use ADL for std::swap, std::begin, and std::end
    • You may write "using std::swap;"
    • Do not write "using namespace std;"
// myclass.h

namespace ns {

void MyClass {
}

void swap(MyClass&, MyClass&);

}
#include "myclass.h"

int main() {
  ns::MyClass v1, v2;
  std::swap(v1, v2);  // Calls std::swap
  {
    // Import std::swap to the current scope
    using std::swap;
    int i, j;
    swap(i, j); // calls std::swap
    swap(v1, v2); // calls ns::swap
  }
}
Made with Slides.com