Snowflakes in C++

Question

I've started learning C++ using Microsoft Visual Studio. I'm decided to create a simple GUI program that creates a "snowflake" wherever the user clicks on the screen. Seeing as how this is my first C++ program, I would greatly appreciate feedback on any aspect of my code.

snowflakes.cpp

// Includes //
#include <SFML/Graphics.hpp>
#include <SFML/Window.hpp>
#include <iostream>

// Declare function //
void snowflakes();

/**
 * Creates a snowflake where the user clicks on the screen.
**/
void snowflakes() {

    const int WIDTH = 1280;
    const int HEIGHT = 960;

    sf::RenderWindow render_window(
        sf::VideoMode(WIDTH, HEIGHT), "Snowflakes"
    );
    sf::Event event;
    sf::Image image;
    sf::Mouse mouse;
    image.create(WIDTH, HEIGHT, sf::Color::Black);

    sf::Texture texture;

    while (render_window.isOpen()) {
        while (render_window.pollEvent(event)) {
            if (event.type == sf::Event::EventType::Closed) {
                render_window.close();
            } else if (event.type == sf::Event::MouseButtonPressed) {
                int mouse_x = mouse.getPosition(render_window).x;
                int mouse_y = mouse.getPosition(render_window).y;
                std::cout << mouse_x << " " << mouse_y << std::endl;

                image.setPixel(mouse_x, mouse_y, sf::Color::White);

                // Get random size of snowflake //

                int size = (rand() % 20) + 10;

                for (int i = 1; i < size; i++) {
                    image.setPixel(mouse_x + i, mouse_y, sf::Color::White);
                    image.setPixel(mouse_x - i, mouse_y, sf::Color::White);
                    image.setPixel(mouse_x, mouse_y + i, sf::Color::White);
                    image.setPixel(mouse_x, mouse_y - i, sf::Color::White);
                    image.setPixel(mouse_x + i, mouse_y + i, sf::Color::White);
                    image.setPixel(mouse_x - i, mouse_y - i, sf::Color::White);
                    image.setPixel(mouse_x + i, mouse_y - i, sf::Color::White);
                    image.setPixel(mouse_x - i, mouse_y + i, sf::Color::White);
                }
            }
        }

        texture.loadFromImage(image);
        sf::Sprite dots(texture);

        render_window.clear();
        render_window.draw(dots);
        render_window.display();
    }

}

int main() {
    // Start program //
    snowflakes();
}

Answer 1

Welcome to C++. Nice and readable C++ code!

I'm not familiar with SFML, but here are my suggestions about improving the code:

• When you get familiar with C++, comments like these can be deleted: // Includes // // Declare function // // Start program //.

• You don't need to declare a function first if you define it immediately after.

• These constants can be made constexpr:

  const int WIDTH = 1280;
  const int HEIGHT = 960;
  

  Also, ALL_CAPS names are generally reserved for macros. Constants can use lower_case names instead.

• This if statement (note that ::EventType is sometimes included but sometimes omitted):

  if (event.type == sf::Event::EventType::Closed) {
      // ...
  } else if (event.type == sf::Event::MouseButtonPressed) {
      // ...
  }
  

  can be replaced by a switch statement since event.type happens to be an enum. Also, since you are using c++20, you can use the using enum feature to simplify your code if your compiler supports it:

  switch (event.type) {
      using enum sf::Event::EventType;
  case Closed:
      // ...
      break;
  case MouseButtonPressed:
      // ...
      break;
  }
  

• This:

  int mouse_x = mouse.getPosition(render_window).x;
  int mouse_y = mouse.getPosition(render_window).y;
  std::cout << mouse_x << " " << mouse_y << std::endl;
  

  can be simplified with structured bindings since Vector2i, the return type of getPosition, seems to have public members. (I almost wrote "don't use std::endl," but then I realized that flushing semantics is appropriate here!)

  auto [x, y] = mouse.getPosition(render_window);
  std::cout << x << ' ' << y << std::endl;
  

• The rendering code deserves a separate function:

  image.setPixel(mouse_x, mouse_y, sf::Color::White);
  
  // Get random size of snowflake //
  
  int size = (rand() % 20) + 10;
  
  for (int i = 1; i < size; i++) {
      image.setPixel(mouse_x + i, mouse_y, sf::Color::White);
      image.setPixel(mouse_x - i, mouse_y, sf::Color::White);
      image.setPixel(mouse_x, mouse_y + i, sf::Color::White);
      image.setPixel(mouse_x, mouse_y - i, sf::Color::White);
      image.setPixel(mouse_x + i, mouse_y + i, sf::Color::White);
      image.setPixel(mouse_x - i, mouse_y - i, sf::Color::White);
      image.setPixel(mouse_x + i, mouse_y - i, sf::Color::White);
      image.setPixel(mouse_x - i, mouse_y + i, sf::Color::White);
  }
  

  Also, consider using the <random> library instead of rand:

  std::mt19937_64 engine{std::random_device{}()}; // global or static
  

  then

  std::uniform_int_distribution<int> dist{10, 29}; // clearer than (rand() % 20) + 10
  int size = dist(eng);