Smart pointers in C++
What are Pointers ?
Pointers are often used to keep track of a variable's address. A null value can be assigned to a pointer. Pass by reference can be used to reference a pointer. On the stack, each pointer has its personal memory location and size.
What are Smart Pointers ?
Smart pointers in C++ are class objects that behave like raw pointers but control the lifespan of heap objects. They're great for software development since they automatically facilitate the process of eliminating objects and avoid memory leaks. This series of blog posts will cover the fundamentals of the smart pointers as well as how to utilise them in C++ programmes.
Smart pointers function like built-in pointers and handle objects generated by new one, so you don't have to worry about when to delete or not to delete a pointer, because the smart pointers will do it for you.
A smart pointer is defined and used in the same way as a built-in pointer. Smart pointers can be used to replace traditional pointers in a variety of circumstances. Essentially, the same thing may be accomplished by just substituting a standard pointer with a smart one.
Because a smart pointer's template class has a built-in pointer whose type argument is the kind of the pointed-to object, we may create smart pointers that refer to any kind of class object. When it refers to dynamically allocated objects, the question of who owns them comes up frequently. When you own something, you have the option to hold or destroy it as you see appropriate. In C++, ownership relates not only to the code that extends to or uses the object, but also to the code that deletes it.
With smart pointers, we implement ownership in terms of where that object's memory is deallocated. When we try to follow pointers to objects which no longer exist, we may encounter memory leaks or unexpected behaviour if we don't execute ownership appropriately.
Smart Pointer over Regular Pointer :
The following are the key drawbacks of utilising normal pointers:
- A regular pointer is incapable of effectively managing memory use.
- It is unable to release memory objects that are not in use by the application.
- Because a regular pointer cannot de-allocate an object's memory, if the pointer is deleted from the programme, the location of the memory object pointed to by the pointer is lost. As a result, there is a memory leak.
Advantages of Smart Pointers :
Smart pointers have the following benefits over conventional pointers:
- It automatically maintains memory.
- It frees the object memory when it was no longer needed in the programme.
- When the pointer in the programme exits the scope, it deallocates the object's memory.
- In C++, a smart pointer is used to allocate objects, traverse various forms of data structures, and manage various types of lambda expressions for passing inside functions.
- It ensures the safety and security of our software. As an outcome, the software becomes much easy to understand and debug.
Types of Smart Pointers :
These pointers are specified in the std namespace of the standard C++ library. We'll also need the <memory> header file to incorporate some of these pointers.
There are three types of smart pointers that are often used:
- unique pointer
- shared pointer
- weak pointer
Let's take a look at each one individually.
1.Unique Pointer
- A unique pointer is a container that carries a pointer to an object. When the object leaves the scope, it releases its memory.
- One of the distinguishing characteristics of the unique pointer is that only one copy of an item exists in memory. There are no other resources that can refer to that specific item.
- A compile time error happens when there are several resources accessible for one object in the code.
Example :
#include <iostream>
#include <memory>
using namespace std;
class Triangle {
int base;
int height;
public:
Triangle(int b, int h){
base = b;
height = h;
}
int area(){
return 0.5 * base * height;
}
};
int main(){
unique_ptr<Triangle> P1(new Triangle(8, 25));
cout << P1->area() << endl; // This will give 50 as output
// unique_ptr<Triangle> P2(P1);
unique_ptr<Triangle> P2;
P2 = move(P1);
// This will give 50 as output
cout << P2->area() << endl;
// cout<<P1->area()<<endl;
return 0;
}
Code :
100
100
Explanation :
In the above example, we demonstrated the use of the unique pointer unique_ptr, which is a smart pointer. We created a class called Triangle and gave int variable base and height in it. Then we declared a function called area() to calculate the area of the triangle using the given variables. Then we introduced the unique pointer unique_ptr . Initially, the unique pointer was pointing to P1. However, we then deleted the P1 and assigned P2, making the pointer refer to P2. Finally, we printed the area via pointer P2, it gave the area of the triangle as output.
2. Shared Pointer :
- When we wish to assign one pointer to several object resources in a programme, we can use shared pointer.
- A shared pointer is an address-based smart pointer that may be used to hold and send a reference outside of a function's scope.
- It's quite handy in OOP (Object Oriented Program). A shared pointer is used to hold a pointer as a local variable.
- The shared pointer will remain active until all resources have fulfilled their tasks.
Example :
#include <iostream>
#include <memory>
using namespace std;
class Triangle {
int base;
int height;
public:
Triangle(int b, int h)
{
base = b;
height = h;
}
int area()
{
return 0.5 * base * height;
}
};
int main()
{
shared_ptr<Triangle> P1(new Triangle(8, 25));
// This will give 100
cout << P1->area() << endl;
shared_ptr<Triangle> P2;
P2 = P1;
// This will give 100
cout << P2->area() << endl;
// This won’t give an error,
cout << P1->area() << endl;
// This will also give 100 now
// This will give 2 as the Reference Counter is currently 2
cout << P1.use_count() << endl;
return 0;
}
Output :
100
100
100
2
Explanation :
In the above example, we demonstrated the use of the unique pointer shared_ptr, which is a smart pointer. We created a class called Triangle and gave int variables base and height in it. Then we declared a function called area() to calculate the area of the triangle using the given variables. Then we introduced the shared pointer shared_ptr. Using shared pointer shared_ptr, many pointers can point to the same object at the same time, and the using the use_count() function, we maintained a Reference Counter. Finally, we printed the area via pointers P1 and P2 and even printed the reference counter using the use_count() method.
3.Weak Pointer :
- A weak pointer is a type of pointer that may be used with shared pointers.
- A weak pointer can use a shared pointer instance to access an object possessed by one or more shared pointer instances.
- It has nothing to do with reference counting.
- When we wish to observe an object but don't need it to stay alive, we utilise the weak pointer in the programme.
It's quite similar to shared pointer, with the exception that it doesn't keep track of the Reference Counter. A pointer will not have a tight grip on the item in this scenario. The reason for this is that if pointers are holding one object and requesting additional objects, a Deadlock may occur.
Summary :
In order to provide an overview, when dealing with smart pointers in C++, you should take the following set of rules into consideration.
- Whenever possible, attempt to utilise smart pointers. If you're not dealing with several pointers/threads sharing a memory address, use unique pointer in most circumstances.
- When dealing with numerous owners, utilise the reference-counted shared pointer instead.
- Use a weak pointer if you wish to explore an object without requiring that the object exists. This pointer is useful for such jobs since it does not add to the reference count.
- To decrease the overhead of utilising smart pointers, make sure you only use raw pointers in a tiny amount of code or when you must definitely use them.
Conclusion :
Smart pointers are a helpful class type that allows programmers to use pointers without having to manually allocate and deallocate memory. Raw pointers should be avoided unless their reach is constrained and their use is straightforward, as preferred by the Microsoft (for e.g., in a loop or in a simple block of code).
After exploring the idea and applications of smart pointers, we've come to the conclusion that smart pointers were added in C++ to address the shortcomings of the conventional pointer. We can handle multiple sorts of owners and resources of the object extremely easily using the smart pointer. We hope you found this post useful.