A little more Elixir, we’re talking modules and functions

In my post Starting out with Elixir I started out looking at how to create a basic Elixir file then looked at mix to generate a project.

Again, I’m not going to go into any depth on the Elixir language, there’s no way I could cover it, but let’s look at modules and functions.

NOTE: I’m new to Elixir, so take all information in this post as me learning things – there may be better ways or different ways to do things that I’ll learn later. This is all about getting a grounding in some basic concepts.

Modules

Like modules or namespaces in other languages, Elixir has the module concept for grouping together functions. For example when using IO.puts, IO is the module and the function is puts

To declare a module we write the following

defmodule Module.Name do
  # your functions
end

The module name can have a full stop/period in the names. The module’s first letter should be uppercase. We can group together functions within a module including private functions.

Functional means functions

Writing anything but the simplest applications/code requires the need for functions. Elixir is a functional language so let’s create some.

Named functions are defined using the following format and should be placed within a module and the function name should start with a lowercase letter, for example

defmodule Simple.Messages do
  def say_hello() do 
    "Hello World"
  end
end

Like other functional languages we can just have the return value as the last line of the function. So in this example we are returning a string “Hello World”.

We could rewrite the above function like this as a one liner

def say_hello(), do: "Hello World"

Actually we can using a syntax such as

def say_hello(), do: (
    "Hello World"
)

and having multiple lines within the ( … ) parenthesis, the def … end style shown initially is a syntactic sugar way of declaring your functions.

We pass parameters in the “standard” way, within the parenthesis of the function and we can supply default values. Like other functional languages, we do not need to define the types of parameters, these are inferred (again, pretty standard in the functional world).

def say_hello(name), do: "Hello #{name}"

We can create function overloads, i.e. functions with the same name but different arity (number of parameters or arity). The example above also shows string interpolation using #{} syntax.

Here’s an example with a default parameter (obviously this will display a warning if you have a parameter-less function of the same name

def say_hello(name \\ "World"), do: "Hello #{name}"

We can also create anonymous functions, for example

anon = fn (name) -> "Hello #{name}" end
# calling the function is slightly different to named functions
IO.puts anon.("Scooby")

There’s also a shorthand for such functions where we denote that parameters using $ followed by the parameter index, i.e.

anon = &("Hello #{&1}")
IO.puts anon.("Scooby")

As mentioned in the modules section, we can also create private functions and these are declared using defp like this

defp say_hello(name), do: "Hello #{name}"
def say_hello_scooby(), do: say_hello("Scooby")

In this example say_hello is private and say_hello_scooby is public.

Aliasing module names

In some cases we might want to alias a module. For example out Simple.Messages module might be alias within another module, where by

defmodule HelloWorld.Application do
  alias Simple.Messages

  def run() do
     IO.puts Message.say_hello("Scooby")
  end
end

Notice we no longer need to fully qualify the module name when it’s used.

That should be enough to get one started writing modules and functions, I’m sure I’ll create other posts to explore how these work further at some point.

Starting out with Elixir

I’ve wanted to try out Elixir for a while. The Elixir language is a functional, dynamic language runs on the Erlang VM.

Obviously this is a small post and hence we’re going to cover very little of the Elixir language here, instead we’ll cover the basics of getting things up and running.

We’re going to run up a an Elixir environment using devcontainers.

  • Create yourself a .devcontainer folder within your source folder
  • Create a file named devcontrainer.json with the following contents
    {
        "image": "elixir",
        "forwardPorts": [3000]
    }
    
  • Open Visual Code from the folder (or open the folder in VS Code)
  • You should have the option to open as a devcontainer, so do that

I’d suggest installing the ElixirLS: Elixir support and debugger or another plugin if you prefer.

Hello World

As is the usual starting point of any language, let’s create a hello_world.exs file and add the following

IO.puts("Hello World")

Now to run this open a terminal from VS Code and type.

elixir hello_world.exs 

As you can see the IO.puts function outputs to the console and strings are represented by double quotes.

The Mix build tool

Mix is a little like the dotnet command (if you come from .NET) in that it can be used to create a new project, as well as different types of project. It’s used to run unit tests and ofcourse compile our application.

Let’s start by creating a new Elixir project

mix new my_project

This will create a new project named my_project along with default files such as mix.exs (using for configuring our application, dependencies etc.). We also have a test folder with an example test

defmodule ExampleTest do
  use ExUnit.Case
  doctest Example

  test "greets the world" do
    assert Example.hello() == :world
  end
end

We can run the tests using

mix test

We can compile our Elixir application using

mix compile

this will produce a _build folder and within this we’ll see a ebin/my_project.app

Supervisor

Now, I’m going to state upfront, at this time all I know about supervisors and supervision trees is that they’re like an OS in a lightweight process. They start, work, then terminate. This is the mechanism we’ll use to create a Hello World application using mix

Run

mix new hello_world --sup

This produces a mix.exs file with the key addition

def application do
  [
    extra_applications: [:logger],
    mod: {HelloWorld.Application, []}
  ]
end

and the lib/hello_world/application.ex file looks like this. I’ve added the IO.puts line as well as removed the comments

defmodule HelloWorld.Application do
  @moduledoc false

  use Application

  @impl true
  def start(_type, _args) do
    children = [
  ]

   IO.puts "Hello World"

   opts = [strategy: :one_for_one, name: HelloWorld.Supervisor]
   Supervisor.start_link(children, opts)
  end
end

This will then run a process (using mix run) and output our “Hello World” string then terminates cleanly.

exs and ex files

You’ll notice that both .ex and .exs are used for Elixir file extensions. The basis seems to be that .ex are meant to be compiled whereas .exs are script files. It can be a little confusing as mix generated projects include both. For example for config and tests it generates .exs files, for the endpoints, router etc. they’re .ex.

References

Elixir
Mix
Using Supervisors to Organize Your Elixir Application

StringSyntaxAttribute and the useful hints on DateTime ToString

For a while I’ve used the DateTime ToString method and noticed the “hint” for showing the possible formats, but I’ve not really thought about how this happens, until now.

Note: This attribute came in for projects targeting .NET 7 or later.

The title of this post gives away the answer to how this all works, but let’s take a look anyway…

If you type

DateTime.Now.ToString("

Visual Studio kindly shows a list of different formatting such as Long Date, Short Date etc.

We can use this same technique in our own code (most likely libraries etc.) by simply adding the StringSyntax attribute to our method parameter(s).

For example

static void Write(
   [StringSyntax(StringSyntaxAttribute.DateOnlyFormat)] string input)
{
    Console.WriteLine(input);
}

This attribute does not enforce the format (in the example above), i.e. yo can enter whatever you like as a string. It just gives you some help (or hint) as to possible values. In the case of the DateOnlyFormat these are possible date formatters. StringSyntax actually supports other syntax hints such as DateTimeFormat, GuidFormat and more.

Sadly (at least at the time of writing) I don’t see any options for custom formats.

Dockerize your React application

You’ve created you React application and are now looking to create a docker image for it.

Before we look at the Dockerfile, create yourself a .dockerignore file that looks like this

.git
node_modules

We do not require the .git folder in our image and we’re going to install our node modules via npm install as you’ll find copying the node_modules folder(s) is slow.

Let’s jump straight in and look at a Dockerfile that will take our React code containerize it and set the container up to run nginx to host it.

FROM node:21-alpine3.18 as build

WORKDIR /usr/app
COPY . .
RUN npm install
RUN npm run build

FROM nginx:alpine
COPY --from=build /usr/app/build /usr/share/nginx/html
EXPOSE 80
CMD ["nginx", "-g", "daemon off;"]

We’re using a node alpine image as our base. This is a lightweight image as we won’t our image to be lean. Next we create our WORKDIR, this can be set to your preferred location but just remember to reuse that location in the COPY command. We’re using .dockerignore to ignore .git and node_modules which allows us to then just copy everything to the image, where we then install then build our React application.

Finally we use nginx to serve our application, first copying the build folder to nginx and finally we set up the CMD to run nginx once the container is started, thus hosting our React application.

To build our image we can just use the following (change the tag name and version to suit your application)

docker build -t my-app:0.1.0 .

and to run, again set your application name any port redirects and use you tagged and version

docker run --rm --name my-app -p 8080:80 -d my-app:0.1.0

We could also extend this sample to include copying of nginx.conf to the image, for example if you want the supply a config like this

worker_processes 4;

events { worker_connections 1024; }

http {
  server {
    listen 4200;
    root  /usr/share/nginx/html;
    include /etc/nginx/mime.types;

  location / {
    root   /usr/share/nginx/html;
    index  index.html;
    try_files $uri $uri/ /index.html;
  }
}
}

If we assume we’re storing out nginx.conf file in a folder named .ngnix (this is not required it’s just for this example) then we could add the following to the Dockerfile, after the line FROM nginx:alpine add the following and whilst we’re at it let’s get rid of the default files that might be located in the images nginx/html folder

COPY ./.nginx/nginx.conf /etc/nginx/nginx.conf
RUN rm -rf /usr/share/nginx/html/*

Docker Compose

Whilst we’re here, let’s create a simple docker-compose.yaml file for our new image.

version: '3.8'
services:
  front-end:
    build:
      context: ./ui
      dockerfile: ./ui/Dockerfile
    ports:
      - 4200:4200
    image: putridparrot/my-app:0.1.0
    container_name: my-app

We might like to store configuration for this image on the hosting server, i.e. via a volume in which case we’d simply add to the bottom of this file the following

    volumes:
      - ./ui/public/appsettings.json:/usr/share/nginx/html/appsettings.json

In this example we’re using an appsettings.json file to configure the environment, or it might include feature flag settings or whatever and assuming it’s stored in the public folder of you React application.

Now we just docker-compose up.

Zustand state management

I’m used to using Redux and more recently Redux Toolkit for global state management in React, however along with state management libraries such as Mobx there’s another library of interest to me, named Zustand. Let’s see how to set up and project and use Zustand and take a very high level look at how to set-up a project with Zustand…

Create yourself a React application, as usual I’m using TypeScript.

  • Add Zustand using yarn add zustand
  • Our store is a hook, and to create the store we use the create method, for example
    import { create } from "zustand";
    
    interface CounterState {
        counter: number;
        increment: () => void;
        decrement: () => void;
    }
    
    export const useCounterStore = create<CounterState>(set => ({
        counter: 0,
        increment: () => set(state => ({ counter: state.counter + 1 })),
        decrement: () => set(state => ({ counter: state.counter - 1 })),
    }));
    

The above creates a simple store, with state and methods to interact with the state. As we’re using TypeScript, we’ve declared the interface matching our state.

To use this state we simply use the hook like this (change App.tsx to look like the following)

import './App.css';
import { useCounterStore } from "./store";

function App() {
  const { counter, increment, decrement } = useCounterStore();

  return (
    <div className="App">
      <button onClick={increment}>+</button>
      <div>{counter}</div>
      <button onClick={decrement}>-</button>
    </div>
  );
}

export default App;

We can also get slices of our state using the hook like this

const counter = useCounterStore(state => state.counter);

Before we move on, unlike RTK we need to enable redux devtools if we want to view the state in the Redux DevTools in our Browser, so to add the dev tool extensions do the following

  • yarn add @redux-devtools/extension
  • We need to import the devtools and change our store a little, so here’s the store with all the additions
    import { create } from "zustand";
    import { devtools } from "zustand/middleware"
    import type {} from "@redux-devtools/extension";
    
    interface CounterState {
      counter: number;
      increment: () => void;
      decrement: () => void;
    }
    
    export const useCounterStore = create<CounterState>()(
      devtools (
        set => ({
          counter: 0,
          increment: () => set(state => ({ counter: state.counter + 1 })),
          decrement: () => set(state => ({ counter: state.counter - 1 })),
        }),
        {
          name: "counter-store",
        }
      )
    );
    

Zustand also has the ability to wrap our global state within a persistence middleware. This allows us to save to various types of storage. We simply wrap our state in persist like this

import { devtools, persist } from "zustand/middleware"

export const useCounterStore = create<CounterState>()(
  devtools (
    persist (
      set => ({
        counter: 0,
        increment: () => set(state => ({ counter: state.counter + 1 })),
        decrement: () => set(state => ({ counter: state.counter - 1 })),
      }),
      {
        name: "counter-store",
      }
    )
  )
);

By default (as in the above code) this state will be persisted to localStorage.

Go check your Application | Local Storage in Edge or Chrome developer tools, for example for Local Storage I have a key counter-store with the value {“state”:{“counter”:4},”version”:0}.

Code

Code from this post is available on github.

React with Signals

Signals are another way of managing application state.

You might ask, “well great but we already have hooks like useState so what’s the point?”

In answer to the above perfectly valid question, Signals work on a in a more granular way – let’s compare to useState. If we create a simple little (and pretty standard) counter component we can immediately see the differences.

Create yourself a React application, I’m using yarn but use your preferred package manager.

yarn create react-app react-with-signals --template typescript

Add the signal package using

yarn add @preact/signals-react

Now we’ll create a folder named components and add a file named CounterState.tsx which looks like this

import { useState } from "react";

export const CounterState = () => {
  const [count, setCount] = useState(0);

  console.log("Render CounterState");

  return (
    <div>
      <div>Current Value {count}</div>
      <div>
        <button onClick={() => setCount(count - 1)}>Decrement</button>
        <button onClick={() => setCount(count + 1)}>Increment</button>
      </div>
    </div>
  );
}

It’s not very pretty but it’s good enough for this demonstration.

Finally create a new file in the components folder named CounterSignals.tsx which should look like this

import React from 'react';
import { useSignal } from "@preact/signals-react";

export const CounterSignals = () => {
  const count = useSignal(0);

  console.log("Render CounterSignals");

  return (
    <div>
      <div>Current Value {count}</div>
      <div>
        <button onClick={() => count.value--}>Decrement</button>
        <button onClick={() => count.value++}>Increment</button>
      </div>
    </div>
  );
}

As you can see, the Signals code creates a Signals object instead of the destructuring way used with useState and the Signals object will not change, but when we change the value, the value within the object changes but does not re-rendering the entire component each time.

Let’s see this by watching the console output in our preferred browser, so just change App.tsx to look like this

import React from ‘react’;
import { CounterState } from ‘./components/CounterState’;
import { CounterSignals } from ‘./components/CounterSignals’;

function App() {
return (


);
}

export default App;
[/em]

Start the application. You might wish to disable React.StrictMode in the index.tsx as this will double up the console output whilst in DEV mode.

Click the Increment and Decrement buttons and you’ll see our useState implementation renders on each click whereas the counter changes for the Signals version but no re-rendering happens.

Code

Code for this example using Signals can be found on github.

i18n in React

i18n (internationalization) is the process of making an application work in different languages and cultures. This includes things like, translations of string resources through to to handling date formats as per the user’s language/culture settings, along with things like decimal separators and more.

There are several libraries available for helping with i18n coding but we’re going to focus on the react-i18next in this post, for no other reason that it’s by other teams in the company I’m working for at the moment.

Getting Started

The react-i18next website is a really good place to get started and frankly I’m very likely to cover much the same code here, so their website should be your first port of call.

If you’re want to instead follow thought the process here then, let’s create a simple React app with TypeScript (as is my way) using

npx create-react-app i18n-app --template typescript

Now add the packages react-i18next and i18next i.e.

npm install react-i18next i18next --save

In this Getting Started I’m going to also include the following libraries which will handle loading the translation files and more

npm install i18next-http-backend i18next-browser-languagedetector --save

Adding localized strings

We’re going to first show an example of embeding the string into a .ts file, however it’s much more likely we’ll want them in a separate file (or multiple files). Let’s get this started by creating a file named i18n1.ts.

In the file we’ll just write one string and here it is

import i18n from "i18next";
import { initReactI18next } from "react-i18next";
import LanguageDetector from "i18next-browser-languagedetector";

i18n
  .use(LanguageDetector)
  .use(initReactI18next)
  .init({
    resources: {
      en: {
        translation: {
          welcome: "code: Hello string en World"
        },
      },
      fr: {
        translation: {
          welcome: "code: Bonjour string fr World"
        }
      }
    },
    debug: true,
    interpolation: {
      escapeValue: false, 
    },
  });

export default i18n;

As you can see we’re using the LanguageDetector to automatically set our resources. On my browser the language is set to en-GB so my expectation is to see the en strings as I’ve not listed en-GB specific strings.

IMPORTANT: Before we can use this, go to index.tsx and import this file import “./i18n1”; so the bundler includes it.

Display/using our localized strings

Now we’ll need to actually use our translation strings. react-i18next comes with hooks, HOC’s and standard JS type functions to interact with our translated strings. Let’s clear out most of the App.tsx and make it look like this

import React from "react";
import "./App.css";
import { useTranslation } from "react-i18next";

const lngs: any = {
  en: { nativeName: "English"},
  fr: { nativeName: "French"},
}

function App() {
  const { t, i18n } = useTranslation();

  return (
    <div className="App">
      <div className="App-header">
        {t("welcome")}
        <div>
          {Object.keys(lngs).map(lng => {
            return <button key={lng} style={{margin: "3px"}}
              onClick={() => i18n.changeLanguage(lng)} disabled={i18n.resolvedLanguage === lng}>{lngs[lng].nativeName}</button>
          })}
        </div>
      </div>
    </div>
  );
}

export default App;

This code will simply display the English/French buttons to allow us to change the language as we wish. Notice, however, that if we add a new language, such as de: { nativeName: “German”} and no strings exist for that language, you’ll end up seeing the “key” for the string. We can solve this later.

At this point if all is working, you can start the application up and switch between the languages. You’ll see that the strings will be prefixed with code: just to make it clear where the strings are coming from, i.e. our code file.

Moving to resource type files (part 1)

As I mentioned, we probably don’t want to embed our string in code. It’s preferable to move them into their own .JSON files.

Create a folder within the src folder named locales (the names of the folders and files doesn’t really matter but it’s good to be consistent) and within that we’ll have one folder name en-GB and another named fr. So the English strings are specific to GB but the French covers all French languages locales.

Now in en-GB create the file translations.json which will look like this

{
  "welcome": "src: Hello en-GB World"
}

For the French translations, add translations.json to the fr folder and it should look like this

{
 "welcome": "src: Bonjour fr World"
}

Note the src: prefix is again, just there to allow us to see where our resources are coming from in this demo.

We now need to change our i18n.ts file to look like this

import i18n from "i18next";
import { initReactI18next } from "react-i18next";
import LanguageDetector from "i18next-browser-languagedetector";

import enGB from "../src/locales/en-GB/translation.json";
import fr from "../src/locales/fr/translation.json";

const resources = {
  en: {
    translation: enGB
  },
  fr: {
    translation: fr
  }
};
i18n
  .use(LanguageDetector)
  .use(initReactI18next)
  .init({
    resources,
    debug: true,
    interpolation: {
      escapeValue: false, 
    },
  });

export default i18n;

So in this code we’reimporting the JSON and then assigning to the resources const. This is very similar to the other way we bought the resources into the i18n.ts file, just we’re importing via JSON files.

Moving to resource type files (part 2)

There’s another way to import the translated strings and that is to simply include the files in the public folder of our React application. So in the public folder add the same folders and files, i.e. locales folder with en-GB and fr folders with the same translation.json files as the last example. I’ve changed the src: prefix on the strings to public: again just so I can prove, for this demo, where the strings originate from.

In other words, here’s the en-GB translations.json file

{
  "welcome": "public: Hello en-GB World"
}

and the fr translations.json file

{
  "welcome": "public: Bonjour fr World"
}

Now back in our i18n.ts file change it to look like this

import i18n from "i18next";
import { initReactI18next } from "react-i18next";
import Backend from "i18next-http-backend";
import LanguageDetector from "i18next-browser-languagedetector";

i18n
  .use(Backend)
  .use(LanguageDetector)
  .use(initReactI18next)
  .init({
    fallbackLng: {
      "en" : ["en-GB"]
    },
    debug: true,
    interpolation: {
      escapeValue: false, 
    },
  });

export default i18n;

Fallback

My browser is setup as English (United Kingdom) which is en-GB and all works well. But what happens if we add a detect a language where we have no translation strings for? Well let’s try it by adding a German option to the lngs const in our App.tsx, so it looks like this

const lngs: any = {
  en: { nativeName: "English"},
  fr: { nativeName: "French"},
  de: { nativeName: "German"},
}

Now, what happens ? Well we’ve see the key for the string, which is probably not what we want in production, better to fall back to a known language. So we need to change the i18n.ts file and add a fallbackLng, I’ll include the whole i18n object so it’s obvious

i18n
  .use(LanguageDetector)
  .use(initReactI18next)
  .init({
    resources,
    lng: "en-GB",
    fallbackLng: "en",
    debug: true,
    interpolation: {
      escapeValue: false, 
    },
  });

Now if our application encounters a locale it’s not setup for, it’ll default to the fallback language (in this case) English. We can also achieve this using

fallbackLng: {
  "default": ["en"]
},

Using this syntax we can also map different languages to specific fallback languages, so for example we might map Swiss locale to map to use French or Italian. This is achieved by passing an array of fallback languages (as taken from the <a href="https://www.i18next.com/principles/fallback" rel="noopener" target="_blank">Fallback documentation</a>

[code]
fallbackLng: { 
  "de-CH": ["fr", "it"], // French and Italian are also spoken in Switzerland
},

Finally with regards fallback languages, we can write code to determine the translation to use, again the Fallback documentation has a good example of this, so I’d suggest checking that link out.

Code

Code for this post is available on github which includes i18n files 1-3 for each of these options for creating translations. just change the index.tsx to import the one you wish to use.

Signal R and React

I haven’t touched Signal R in a while. I wanted to see how to work with Signal R in a React app.

Let’s start by creating a simple ASP.NET Core Web API server

  • Create an ASP.NET Core Web API application, I’m going to use minimal API
  • Add the NuGet package Microsoft.AspNetCore.SignalR.Client
  • Add the following to the Program.cs
    builder.Services.AddSignalR();
    builder.Services.AddCors();
    
  • We’ve added CORS support as we’re going to need tthis for testing locally, we’ll also need the following code
    app.UseCors(options =>
    {
      options.AllowAnyHeader()
        .AllowAnyMethod()
        .AllowCredentials()
        .SetIsOriginAllowed(origin => true);
    });
    

Before we can use Signal R we’ll need to add a hub, I’m going to add a file NotificationHub.cs with the following code

public class NotificationHub : Hub;

Now return to Program.cs and add the following

  • We need to map our hub into the application using
    app.MapHub<NotificationHub>("/notifications");
    
  • Finally let’s map an endpoint to allow us to send messages via Swagger to our clients. After the line above, add the following
    app.MapGet("/test", async (IHubContext<NotificationHub> hub, string message) =>
      await hub.Clients.All.SendAsync("NotifyMe",$"Message: {message}"));
    

Now we need a client, so create yourself a React application (I’m using TypeScript as usual with mine).

  • Add the package @microsoft/signalr, i.e. from yarn yarn add @microsoft/signalr
  • In the App.tsx we’re going to create the HubConnectionBuilder against out ASP.NET Core API server. We’ll then start the connection and finally watch for messages on the “NotifyMe” name as previously set up in the ASP.NET app, the code looks like this
    import { HubConnectionBuilder } from '@microsoft/signalr';
    
    function App() {
      const [message, setMessage] = useState("");
    
      useEffect(() => {
        const connection = new HubConnectionBuilder()
          .withUrl("http://localhost:5021/notifications")
          .build();
      
        connection.start();  
        connection.on("NotifyMe", data => {
          setMessage(data);
        });
      }, [])  
    
      return (
        <div className="App">
          {message}
        </div>
      );
    }
    

Make sure you start the ASP.NET server first, then start your React application. Now from the Swagger page we can send messages into the server and out to the React client’s connected to SignalR.

Messing around with MediatR

MediatR is an implementation of the Mediator pattern. It doesn’t match the pattern exactly, but as the creator, Jimmy Bogard states that “It matches the problem description (reducing chaotic dependencies), the implementation doesn’t exactly match…”. It’s worth reading his post You Probably Don’t Need to Worry About MediatR.

This pattern is aimed at decoupling the likes of business logic from a UI layer or request/response’s.

There are several ways we can already achieve this in our code, for example, using interfaces to decouple the business logic from the UI or API layers as “services” as we’ve probably all done for years. The only drawback of this approach is it requires the interfaces to be either passed around in our code or via DI and is a great way to do things. Another way to do this is, as used within UI, using WPF, Xamarin Forms, MAUI and others where we often use in-process message queues to send messages around our application tell it to undertake some task and this is essentially what MediatR is giving us.

Let’s have a look at using MediatR. I’m going to create an ASP.NET web API (obviously you could use MediatR in other types of solutions)

  • Create an ASP.NET Core Web API. I’m using Minimal API, so feel free to check that or stick with controllers as you prefer.
  • Add the nuget package MediatR
  • To the Program.cs file add
    builder.Services.AddMediatR(cfg => 
      cfg.RegisterServicesFromAssembly(typeof(Program).Assembly));
    

At this point we have MediatR registering services for us at startup. We can passing multiple assemblies to the RegisterServicesFromAssembly method, so if we have all our reqeust/response code in multiple assemblies we can supply just those assemblies. Obviously this makes our life simpler but at the cost of reflecting across our code at startup.

The ASP.NET Core Web API creates the WeatherForecast example, we’ll just use this for our sample code as well.

The first thing you’ll notice is that the route to the weatherforecast is tightly coupled to the sample code. Ofcourse it’s an example, so this is fine, but we’re going to clean things up here and move the implementation into a file named GetWeatherForecastHandler but before we do that…

Note: Ofcourse we could just move the weather forecast code into an WeatherForecastService, create an IWeatherForecastService interface and there’s no reason not to do that, MediatR just offers and alternative way of doing things.

MediatR will try to find a matching handler for your request. In this example we have no request parameters. This begs the question as to how MediatR will match against our GetWeatherForecastHandler. It needs a unique request type to map to our handler, in this case the simplest thing to do is create yourself the request type. Mine’s named GetWeatherForecast and looks like this

public record GetWeatherForecast : IRequest<WeatherForecast[]>
{
    public static GetWeatherForecast Default { get; } = new();
}

Note: I’ve created a static method so we’re not creating an instance for every call, however this is not required and obviously when you are passing parameters you will be creating an instance of a type each time – this does obviously concern me a little if we need high performance and are trying to write allocation free code, but then we’d do lots differently then including probably not using MediatR.

Now we’ll create the GetWeatherForecastHandler file and the code looks like this

public class GetWeatherForecastHandler : IRequestHandler<GetWeatherForecast, WeatherForecast[]>
{
  private static readonly string[] Summaries = new[]
  {
    "Freezing", "Bracing", "Chilly", "Cool", "Mild", "Warm", "Balmy", "Hot", "Sweltering", "Scorching"
  };

  public Task<WeatherForecast[]> Handle(GetWeatherForecast request, CancellationToken cancellationToken)
  {
    var forecast = Enumerable.Range(1, 5).Select(index =>
      new WeatherForecast
      {
        Date = DateOnly.FromDateTime(DateTime.Now.AddDays(index)),
        TemperatureC = Random.Shared.Next(-20, 55),
        Summary = Summaries[Random.Shared.Next(Summaries.Length)]
      })
    .ToArray();

    return Task.FromResult(forecast);
  }
}

At this point we’ve created a way for MediatR to find the required handler (i.e. using the GetWeatherForecast type) and we’ve created a handler to create the response. In this example we’re not doing any async work, so we just wrap the result in a Task.FromResult.

Next go back to the Program.cs or if you’ve used controllers, go to your controller. If using controller you’ll need the constructor to take the parameters IMediator mediator and assign to a readonly field in the usually way.

For our minimal API example, go back to the Program.cs file remove the summaries variable/code and then change the route code to look like this

app.MapGet("/weatherforecast",  (IMediator mediator) => 
  mediator.Send(GetWeatherForecast.Default))
.WithName("GetWeatherForecast")
.WithOpenApi();

We’re not really playing too nice in the code above, in that we’re not returning results code, so let’s add some basic result handling

app.MapGet("/weatherforecast",  async (IMediator mediator) => 
  await mediator.Send(GetWeatherForecast.Default) is var results 
    ? Results.Ok(results) 
    : Results.NotFound())
  .WithName("GetWeatherForecast")
  .WithOpenApi();

Now for each new HTTP method call, we would create a request object and a handler object. In this case we send no parameters, but as you can no doubt see, for a request that takes (for example) a string for your location, we’d create a specific type for wrapping that parameter and the handler can then be mapped to that request type.

In our example we used the MediatR Send method. This sends a request to a single handler and expects a response of some type, but MediatR also has the ability to Publish to multiple handlers. These types of handlers are different, firstly they need to implement the INotificationHandler interface and secondly no response is expected when using Publish. These sorts of handlers are more like event broadcasts, so you might use then to send a message to an email service or database code which sends out an email upon request or updates a database.

Or WeatherForecast sample doesn’t give me any good ideas for using Publish in it’s current setup, so let’s just assume we have a way to set the current location. Like I said this example’s a little contrived as we’re going to essentially set the location for everyone connecting to this service, but you get the idea.

We’re going to add a SetLocation request type that looks like this

public record SetLocation(string Location) : INotification;

Notice that for publish our type is implementing the INotification interface. Our handles look like this (my file is named SetLocationHandler.cs but I’ll put both handlers in there just to be a little lazy)

public class UpdateHandler1 : INotificationHandler<SetLocation>
{
  public Task Handle(SetLocation notification, CancellationToken cancellationToken)
  {
    Console.WriteLine(nameof(UpdateHandler1));
    return Task.CompletedTask;
  }
}

public class UpdateHandler2 : INotificationHandler<SetLocation>
{
  public Task Handle(SetLocation notification, CancellationToken cancellationToken)
  {
    Console.WriteLine(nameof(UpdateHandler2));
    return Task.CompletedTask;
  }
}

As you can see, the handlers need to implement INotificationHandler with the correct request type. In this sample we’ll just write messages to console, but you might have a more interesting set of handlers in mind.

Finally let’s add the following to the Program.cs to publish a message

app.MapGet("/setlocation", (IMediator mediator, string location) =>
  mediator.Publish(new SetLocation(location)))
.WithName("SetLocation")
.WithOpenApi();

When you run up your server and use Swagger or call the setlocation method via it’s URL you’ll see that all your handlers that handle the request get called.

Ofcourse we can also Send and Post messages/request from our handlers, so maybe we get the weather forecast data then publish a message for some logging system to update the logs.

MediatR also includes the ability to stream from a requests where our request type implements the IStreamRequest and our handlers implement IStreamRequestHandler.

If we create a simple request type but this one implements IStreamRequest for example

public record GetWeatherStream : IStreamRequest<WeatherForecast>;

and now add a handler which implements IStreamRequestHandler, something like this (which delay’s to just give a feel of getting data from somewhere else)

public class GetWeatherStreamHandler : IStreamRequestHandler<GetWeatherStream, WeatherForecast>
{
  public async IAsyncEnumerable<WeatherForecast> Handle(GetWeatherStream request, 
    [EnumeratorCancellation] CancellationToken cancellationToken)
  {
    var index = 0;
    while (!cancellationToken.IsCancellationRequested)
    {
      await Task.Delay(500, cancellationToken);
      yield return new WeatherForecast
      {
        Date = DateOnly.FromDateTime(DateTime.Now.AddDays(index)),
        TemperatureC = Random.Shared.Next(-20, 55),
        Summary = Data.Summaries[Random.Shared.Next(Data.Summaries.Length)]
      };

      index++;
      if(index > 10)
        break;
    }
  }
}

Finally we can declare our streaming route using Minimal API very simply, for example

app.MapGet("/stream", (IMediator mediator) =>
  mediator.CreateStream(new GetWeatherStream()))
.WithName("Stream")
.WithOpenApi();