Derived types

Define a type

A derived type bundles related data into one named unit, so a point can hold x and y together instead of two loose variables.

Declare it with the type statement, list each component on its own line, and close the block with end type.

The example below defines a simple point with two real components.

program demo implicit none type :: point real :: x real :: y end type point end program demo

Use it

Declare a variable of the new type and reach each component with the percent sign, as in p%x for the x part of p.

Components behave like ordinary variables: assign them, read them, and pass them to procedures one by one.

The program below stores two coordinates and prints them.

program demo implicit none type :: point real :: x real :: y end type point type(point) :: p p%x = 1.0 p%y = 2.0 print *, p%x, p%y end program demo

Structure constructors

A structure constructor builds a whole value in one expression by naming the type and listing the components in order.

It keeps initialization short and makes the starting state of the variable visible on a single line.

The next program creates a point with one constructor call and prints the full value.

program demo implicit none type :: point real :: x real :: y end type point type(point) :: p p = point(1.0, 2.0) print *, p end program demo

Nest types

A component of one derived type can itself be a derived type, which builds larger models from small tested pieces.

Define the inner type first, then use it as the declared type of a component in the outer type.

Here a circle holds a point as its center plus a radius, and the program reaches the nested x with two percent signs.

program demo implicit none type :: point real :: x real :: y end type point type :: circle type(point) :: center real :: radius end type circle type(circle) :: c c%center = point(0.0, 0.0) c%radius = 5.0 print *, c%center%x, c%radius end program demo

Next: Type-bound procedures

Article author: Arthur Isaev