SimpleTechTalks

Difference between constexpr vs const

constexpr	const
If a member function is declared as constexpr then it will be treated as const as well.	If a member function is declared as const then it will not be treated as constexpr.
If a variable is declared as constexpr then it will be treated as const as well.	If a variable is declared as const then it will not be treated as constexpr.
constexpr is mainly used for optimization purpose where computation can be done at compile level to optimize run time performance.	However const is mainly used to ensure that there are no accidental changes done by the member method to member data.

For more information related to constexpr click here.

constexpr explained with simple example

by admin

In C++11(Introduction to C++11), constexpr is added which improves the run-time performance of the code as it allows the compiler to compute the associated code at compile time if possible. constexpr specifier can be used with both variables and functions where only constant expressions are used which can be evaluated at compile time.

Few important points:

A constexpr function return type and argument type must be literal types.
A constexpr function cannot be virtual but it can be recursive.
A constexpr function cannot have uninitialized variables. Till C++11 it can have only one return statement while from C++14 it can have more than one return statement.
Any variable can be declared as constexpr, if it is initialized. In case of constructor initialization, the constructor also must be declared as constexpr.
constexpr can also be used with references if the referenced object is initialized.
Any constexpr function can only call other constexpr functions.
constexpr function can also be used with non-const variables. No need to define two different functions for same.

(more…)

Difference between Copy assignment operator vs Move assignment operator

by admin

Copy Assignment Operators	Move Assignment Operators
Copy assignment operator takes lvalue reference as an argument.	Move assignment operator takes rvalue reference as argument.
Copy assignment operator create a new object from passed object by copying each and every item into a new memory location.	Move assignment operator create a new object by using as much memory from passed object.
Since, copy assignment operator use a lot of new memory allocation(for new object). Hence, copy constructors gives bad performance compared to move constructors	Since, move assignment operator use most of memory blocks from passed object. Hence, move constructors gave better performance compared to copy constructors.
Since, copy assignment operator doesn’t make any change to passed object. Hence, passed object can be used after copy operations also.	Since, move assignment operator utilizes memory blocks from passed object. Hence, passed object can’t be used after move operations.

For more information related to move assignment operators click here.

Difference between Copy constructor vs Move constructor

by admin

Copy Constructors	Move Constructors
Copy constructors takes lvalue reference as an argument.	Move constructors takes rvalue reference as argument.
Copy constructors create a new object from passed object by copying each and every item into a new memory location.	Move constructors create a new object by using as much memory from passed object.
Since, copy constructors use a lot of new memory allocation(for new object). Hence, copy constructors gives bad performance compared to move constructors	Since, move constructors use most of memory blocks from passed object. Hence, move constructors gave better performance compared to copy constructors.
Since, copy constructors doesn’t make any change to passed object. Hence, passed object can be used after copy operations also.	Since, move constructors utilizes memory blocks from passed object. Hence, passed object can’t be used after move operations.

For more information related to Move constructors click here

Move Assignment Operator explained with simple example

by admin

In C++11 (Introduction to C++11) , move assignment operator are added which can improve any code’s performance drastically. Typically a move assignment operator is same as copy/move constructor but before reusing the memory blocks from passed object, it releases all the memory which is owned by this object. Since, memory reallocation is avoided in case of move assignment operators. Thus it gives better performance.

However while implementing move assignment operators we need to take care of one important point to ensure that the original object can be correctly destroyed.

Let’s take an example to see how this memory reallocation is avoided in case of move assignment.

(more…)

std::move explained with simple example

by admin

std::move is a C++11(Introduction to C++11) utility function which is used to convert an lvalue to rvalue expression to forcefully call the move constructor or move assignment operator which will improve the program’s performance drastically. This utility function is only change lvalue into rvalue expression that’s it. This function doesn’t do any kind of memory movement.

Implementation of this function is as follows:

From C++11 onwards:

template< class T >
typename std::remove_reference<T>::type&& move( T&& t );

From C++14 onwards:

template< class T >
constexpr typename std::remove_reference<T>::type&& move( T&& t );

Let’s take an example of usage of this move function:

(more…)

Move Constructors explained with simple example

by admin

In C++11 (Introduction to C++11) , move constructors are added which can improve any code’s performance drastically. Typically a move constructor is same as copy constructor which will create a new instance based on the passed object. But, the idea behind move constructor is to avoid memory reallocation and use as many memory from the passed original object because the original object is about to be deleted as it has been provided as a temporary object.

Since, in move constructors we avoid memory reallocation while creating new instance while will improve the code’s performance as we know memory allocation is quite costly.
While implementing move constructors we need to take care of one important point to ensure that the original object can be correctly destroyed.

Let’s take an example to see how this memory reallocation is avoided in case of move constructor.

(more…)

Rvalue References

by admin

Before understanding about this new feature “Rvalue references” of C++11(Introduction to C++11), we should first understand what is lvalue and rvalue in C++.

Typically Lvalue is an expression which returns a permanent memory address which can be assigned to any variables. For ex:

int x = 10;
int y;
y = 20;

Here both x and y are lvalues. In some cases, we can have functions also as lvalue as shown in below example:

int x;  // global variable

// Returning reference to global variable x
int& getXRef()
{
    return x;
}

getXRef() = 30;

Now this function can act as lvalue too as shown above.

(more…)

Introduction to C++11

by admin

In August 2011, a new C++ version which is known as C++11 is approved by ISO which adds a lot of new set of features to existing C++ programming language. All official documents related to these changes can be found at ISO C++ committee website.

Why C++11 ?

C++11 significantly improved the core C++ language by making several additions to its standard libraries. Major areas where C++11 is improved from its predecessor includes multithreading support, uniform initialization, generic programming support and performance.

Aims of C++11

Compatibility and Stability MUST be maintained for old code written in C++98,C++03 and in C.
Extension of core language preferably be done via standard library.
System and library design method is preferred compared to introducing new features which is useful only for specific applications.
Focus to improve type safety by providing safer methods to earlier unsafe methods.
Improve performance and better interaction with hardware to ensure better,safe and high performance embedded system programming.
Make C++ easier to teach and learn through increased uniformity, stronger guarantees, and facilities better libraries for beginners. Expert programmers can use other efficient features.

(more…)

Command Design Pattern Explained With Simple Example: Behavioural Design Pattern Category

by admin

The Command design pattern allows to encapsulate an action or trigger inside an object which will be used later to trigger an event. Since in this design pattern, commands are encapsulated inside objects, hence we can use additional actions on this commands for example- Queuing of various commands, undo/redo actions etc. This design pattern is very useful in case of GUI actions (button, menu actions etc), transactional behaviour, progress bars etc. For Design patterns basic explanation see (Design Patterns Simplified Version).

(more…)