What is a union in D Language?

Union

A union is a data type that can be considered special in D. It gives us the ability to have different data types in the same memory or storage location. The defined members of a union can be several, but only one of these members can have a value at any given time. Unions efficiently utilize a single memory location due to its multi-purpose ability.

General syntax for union definition

The union statement creates a new user-defined data type, with more than one member base type for our program. Below is the general syntax used to define a union:

union union_name{
 define member
}[one_or_more_union_variables]
  • union: The keyword used to define a union.
  • union_name: An optional tag name given to a union. It is any legal D identifier.
  • define member: The part of the union statement block. We must declare more than one union member of any basic D inbuilt data type or a user-defined one.
  • one_or_more_union_variables: A single or multiple variable we want to be of the union data type we just defined. It is optional as well.

Example

union example { 
   int my_int; 
   float my_float; 
   char my_str[40]; 
} sample_variable;

It implies that a variable of data type example can store an integer, a floating-point number, or a string of characters. It means one variable (a single memory location) can be used to store several types of data. Depending on our requirement, any built-in or user-defined data types can be used inside a union.

Accessing a Union

The sample_variable variable can be of any member type that is defined above just by doing this inside the main method. That will allow the sample_variable to hold an integer as implied by the my_int member.

example sample_variable;
sample_variable.my_int =20;

Code

Let’s look at the code below:

import std.stdio;
//define a union
union Example { 
   int my_int; 
   float my_float; 
   char[40]  my_str; 
};  
void main( ) { 
// declare variable sample_var as type example
   Example sample_var; 
   //assign value to the variables
   sample_var.my_int = 10; 
   sample_var.my_float = 220.5; 
   sample_var.my_str = "I am loving this";
   //check size of the variables
   writeln( "size of : ", sample_var.sizeof); 
   writeln( "data.my_int : ", sample_var.my_int); 
   writeln( "data.my_float: ", sample_var.my_float); 
   writeln( "data.my_str : ", sample_var.my_str); 
}

Explanation

  • Lines 3 to 7: We have the definition of a union Example, with allowable data types of int, float, and string.
  • Lines 9 to 21: We have the main function blocks, which return no value.
  • Lines 11: We declare a variable sample_var as the type of Example inside function main.
  • Lines 13 to 15: We give sample_varseveral values based on the basic types in the union.
  • Lines 17 to 20: We have some print statements to print these defined variables to screen.

Output

The value of sample_var.my_int and sample_var.my_float are output erroneously. This happened because in the case of several assignments to a particular union variable, in this case, sample_var, the variable with the largest value, sample_var.my_str is automatically displayed correctly while the rest are wrong.

Note: Unions have a size that is the size of the largest value of the union.

Free Resources

License: Creative Commons-Attribution-ShareAlike 4.0 (CC-BY-SA 4.0)