Getting started with D Language Dynamic Arrays & Slices Ranges Associative Arrays Traits Templates Structs UFCS - Uniform Function Call Syntax Scope guards Imports and modules Compile Time Function Evaluation (CTFE)Classes Loops Strings Unittesting Memory & Pointers Contracts

Ranges

Remarks:

If a foreach is encountered by the compiler

foreach (element; range) {

it's internally rewritten similar to the following:

for (auto it = range; !it.empty; it.popFront()) {
    auto element = it.front;
    ...
}

Any object which fulfills the above interface is called an input range and is thus a type that can be iterated over:

struct InputRange {
    @property bool empty();
    @property T front();
    void popFront();
}

Strings and arrays are ranges

import std.stdio;

void main() {
    auto s = "hello world";
    auto a = [1, 2, 3, 4];

    foreach (c; s) {
        write(c, "!"); // h!e!l!l!o! !w!o!r!l!d!
    }
    writeln();

    foreach (x; a) {
        write(x * x, ", "); // 1, 4, 9, 16, 
    }
}

Making a new Input Range type

The InputRange concept has three functions, example:

struct InputRange(T) {
    @property bool empty();
    @property T front();
    void popFront();
}

In short, a way to

check if the range is empty
get the current element
move to the next element

To make our own type a InputRange, we must implement these three functions. Let's take a look at the infinite sequence of squares.

struct SquaresRange {
    int cur = 1;

    @property bool empty() {
        return false;
    }

    @property int front() {
        return cur^^2;
    }

    void popFront() {
        cur++;
    }
}

See the D tour for an example with Fibonacci.

Contributors

Topic Id: 3106

Example Ids: 10589,10590

This site is not affiliated with any of the contributors.