This repo shows the end result of a Coding Dojo session. We were implementing "Pac-Man" kata.
- Node.js based on
./.nvmrc
file
npm i -g yarn
yarn install
yarn elm:install
yarn test:watch
This project is bootstrapped with Create Elm App.
Below you will find some information on how to perform basic tasks. You can find the most recent version of this guide here.
- Sending feedback
- Folder structure
- Installing Elm packages
- Installing JavaScript packages
- Available scripts
- Turning on/off Elm Debugger
- Dead code elimination
- Changing the Page
<title>
- JavaScript Interop
- Adding a Stylesheet
- Post-Processing CSS
- Using elm-css
- Adding a CSS Preprocessor (Sass, Less etc.)
- Adding Images and Fonts
- Using the
public
Folder - Using custom environment variables
- Setting up API Proxy
- Running tests
- Making a Progressive Web App
- Deployment
- IDE setup for Hot Module Replacement
You are very welcome with any feedback
elm-app install <package-name>
Other elm-package
commands are also available.
To use JavaScript packages from npm, you'll need to add a package.json
, install the dependencies, and you're ready to go.
npm init -y # Add package.json
npm install --save-dev pouchdb-browser # Install library from npm
// Use in your JS code
import PouchDB from 'pouchdb-browser';
const db = new PouchDB('mydb');
my-app/
├── .gitignore
├── README.md
├── elm-package.json
├── elm-stuff
├── public
│ ├── favicon.ico
│ ├── index.html
│ ├── logo.svg
│ └── manifest.json
├── src
│ ├── Main.elm
│ ├── index.js
│ ├── main.css
│ └── registerServiceWorker.js
└── tests
├── Tests.elm
└── elm-package.json
For the project to build, these files must exist with exact filenames:
public/index.html
is the page template;public/favicon.ico
is the icon you see in the browser tab;src/index.js
is the JavaScript entry point.
You can delete or rename the other files.
You may create subdirectories inside src.
In the project directory you can run:
Builds the app for production to the build
folder.
The build is minified, and the filenames include the hashes. Your app is ready to be deployed!
Runs the app in the development mode. Open http://localhost:3000 to view it in the browser.
The page will reload if you make edits. You will also see any lint errors in the console.
An alias for elm-app package install
Run tests with node-test-runner
You can make test runner watch project files by running:
elm-app test --watch
Note: this is a one-way operation. Once you eject
, you can’t go back!
If you aren’t satisfied with the build tool and configuration choices, you can eject
at any time.
Instead, it will copy all the configuration files and the transitive dependencies (Webpack, Elm Platform, etc.) right into your project, so you have full control over them. All of the commands except eject
will still work, but they will point to the copied scripts so you can tweak them. At this point, you’re on your own.
You don’t have to use 'eject' The curated feature set is suitable for small and middle deployments, and you shouldn’t feel obligated to use this feature. However, we understand that this tool wouldn’t be useful if you couldn’t customize it when you are ready for it.
Create Elm App does not rely on the global installation of Elm Platform, but you still can use its local Elm Platform to access default command line tools:
Alias for elm-package
Use it for installing Elm packages from package.elm-lang.org
Alias for elm-repl
Alias for elm-make
Alias for elm-reactor
By default, in production (elm-app build
) the Debugger is turned off and in development mode (elm-app start
) it's turned on.
To turn on/off Elm Debugger explicitly, set ELM_DEBUGGER
environment variable to true
or false
respectively.
Create Elm App comes with an opinionated setup for dead code elimination which is disabled by default, because it may break your code.
You can enable it by setting DEAD_CODE_ELIMINATION
environment variable to true
By default, assets will be linked from the HTML to the root url. For example /css/style.css
.
If you deploy to a path that is not the root, you can change the PUBLIC_URL
environment variable to properly reference your assets in the compiled assets. For example: PUBLIC_URL=./ elm-app build
.
You can find the source HTML file in the public
folder of the generated project. You may edit the <title>
tag in it to change the title from “Elm App” to anything else.
Note that normally you wouldn’t edit files in the public
folder very often. For example, adding a stylesheet is done without touching the HTML.
If you need to dynamically update the page title based on the content, you can use the browser document.title
API and JavaScript interoperation. The next section of this tutorial will explain it in more detail.
You can send and receive values from JavaScript using the concept of ports..
In the following example we will use JavaScript to change the page title dynamically. To make it work with files created by create-elm-app
you need to modify
src/index.js
file to look like this:
import './main.css';
import { Main } from './Main.elm';
import registerServiceWorker from './registerServiceWorker';
var app = Main.embed(document.getElementById('root'));
registerServiceWorker();
// ports related code
app.ports.windowTitle.subscribe(function(newTitle){
window.document.title = newTitle;
});
Please note the windowTitle
port in the above example, more about it later.
First let's allow the Main nodule to use ports and in Main.elm
file please append port
to the module declaration:
port module Main exposing (..)
Do you remember windowTitle
in JavaScript? Let's declare the port:
port windowTitle : String -> Cmd msg
and use it to call JavaScript in you update function.
update : Msg -> Model -> ( Model, Cmd Msg )
update msg model =
case msg of
Inc ->
( { model | counter = model.counter + 1}
, windowTitle ("Elm-count up " ++ (toString (model.counter + 1)))
)
Dec ->
( { model | counter = model.counter - 1}
, windowTitle ("Elm-count down " ++ (toString (model.counter - 1))))
NoOp ->
( model, Cmd.none )
Please note that for Inc and Dec operations Cmd.none
was replaced with windowTitle
port call that is executed on the JavaScript side..
This project setup uses Webpack for handling all assets. Webpack offers a custom way of “extending” the concept of import
beyond JavaScript. To express that a JavaScript file depends on a CSS file, you need to import the CSS from the JavaScript file:
body {
padding: 20px;
}
import './main.css'; // Tell Webpack to pick-up the styles from main.css
This project setup minifies your CSS and adds vendor prefixes to it automatically through Autoprefixer so you don’t need to worry about it.
For example, this:
.container {
display: flex;
flex-direction: row;
align-items: center;
}
becomes this:
.container {
display: -webkit-box;
display: -ms-flexbox;
display: flex;
}
In development, expressing dependencies this way allows your styles to be reloaded on the fly as you edit them. In production, all CSS files will be concatenated into a single minified .css
file in the build output.
You can put all your CSS right into src/main.css
. It would still be imported from src/index.js
, but you could always remove that import if you later migrate to a different build tool.
Step 1: Install elm-css npm package
npm install elm-css -g
elm-app install rtfeldman/elm-css
elm-app install rtfeldman/elm-css-helpers
Create an Elm file at src/Stylesheets.elm
(The name of this file cannot be changed).
port module Stylesheets exposing (main, CssClasses(..), CssIds(..), helpers)
import Css exposing (..)
import Css.Elements exposing (body, li)
import Css.Namespace exposing (namespace)
import Css.File exposing (..)
import Html.CssHelpers exposing (withNamespace)
port files : CssFileStructure -> Cmd msg
cssFiles : CssFileStructure
cssFiles =
toFileStructure [ ( "src/style.css", Css.File.compile [ css ] ) ]
main : CssCompilerProgram
main =
Css.File.compiler files cssFiles
type CssClasses
= NavBar
type CssIds
= Page
appNamespace =
"App"
helpers =
withNamespace appNamespace
css =
(stylesheet << namespace appNamespace)
[ body
[ overflowX auto
, minWidth (px 1280)
]
, id Page
[ backgroundColor (rgb 200 128 64)
, color (hex "CCFFFF")
, width (pct 100)
, height (pct 100)
, boxSizing borderBox
, padding (px 8)
, margin zero
]
, class NavBar
[ margin zero
, padding zero
, children
[ li
[ (display inlineBlock) |> important
, color primaryAccentColor
]
]
]
]
primaryAccentColor =
hex "ccffaa"
To compile the CSS file, just run
elm-css src/Stylesheets.elm
This will generate a file called style.css
Add the following line to your src/index.js
:
import './style.css';
import Stylesheets exposing (helpers, CssIds(..), CssClasses(..))
view model =
div [ helpers.id Page ]
[ div [ helpers.class [ NavBar ] ]
[ text "Your Elm App is working!" ]
]
Please note that Stylesheets.elm
exposes helpers
record, which contains functions for creating id and class attributes.
You can also destructure helpers
to make your view more readable:
{ id, class } =
helpers
If you find CSS preprocessors valuable you can integrate one. In this walkthrough, we will be using Sass, but you can also use Less, or another alternative.
First we need to create a package.json
file, since create-elm-app is not shipping one at the moment. Walk through the normal npm init
process.
npm init
Second, let’s install the command-line interface for Sass:
npm install --save node-sass-chokidar
Alternatively you may use yarn
:
yarn add node-sass-chokidar
Then in package.json
, add the following lines to scripts
:
"scripts": {
+ "build-css": "node-sass-chokidar src/ -o src/",
+ "watch-css": "npm run build-css && node-sass-chokidar src/ -o src/ --watch --recursive",
...
}
Note: To use a different preprocessor, replace
build-css
andwatch-css
commands according to your preprocessor’s documentation.
Now you can rename src/main.css
to src/main.scss
and run npm run watch-css
. The watcher will find every Sass file in src
subdirectories, and create a corresponding CSS file next to it, in our case overwriting src/main.css
. Since src/index.js
still imports src/main.css
, the styles become a part of your application. You can now edit src/main.scss
, and src/main.css
will be regenerated.
To share variables between Sass files, you can use Sass imports. For example, src/main.scss
and other component style files could include @import "./shared.scss";
with variable definitions.
To enable importing files without using relative paths, you can add the --include-path
option to the command in package.json
.
"build-css": "node-sass-chokidar --include-path ./src --include-path ./node_modules src/ -o src/",
"watch-css": "npm run build-css && node-sass-chokidar --include-path ./src --include-path ./node_modules src/ -o src/ --watch --recursive",
This will allow you to do imports like
@import 'styles/_colors.scss'; // assuming a styles directory under src/
@import 'nprogress/nprogress'; // importing a css file from the nprogress node module
At this point you might want to remove all CSS files from the source control, and add src/**/*.css
to your .gitignore
file. It is generally a good practice to keep the build products outside of the source control.
Why node-sass-chokidar
?
node-sass
has been reported as having the following issues:
-
node-sass --watch
has been reported to have performance issues in certain conditions when used in a virtual machine or with docker. -
Infinite styles compiling #1939
-
node-sass
has been reported as having issues with detecting new files in a directory #1891
node-sass-chokidar
is used here as it addresses these issues.
With Webpack, using static assets like images and fonts works similarly to CSS.
By requiring an image in JavaScript code, you tell Webpack to add a file to the build of your application. The variable will contain a unique path to the said file.
Here is an example:
import logoPath from './logo.svg'; // Tell Webpack this JS file uses this image
import { Main } from './Main.elm';
Main.embed(
document.getElementById('root'),
logoPath // Pass image path as a flag for Html.programWithFlags
);
Later on, you can use the image path in your view for displaying it in the DOM.
view : Model -> Html Msg
view model =
div []
[ img [ src model.logo ] []
, div [] [ text model.message ]
]
The public
folder contains the HTML file so you can tweak it, for example, to set the page title.
The <script>
tag with the compiled code will be added to it automatically during the build process.
You can also add other assets to the public
folder.
Note that we normally encourage you to import
assets in JavaScript files instead.
For example, see the sections on adding a stylesheet and adding images and fonts.
This mechanism provides a few benefits:
- Scripts and stylesheets get minified and bundled together to avoid extra network requests.
- Missing files cause compilation errors instead of 404 errors for your users.
- Result filenames include content hashes, so you don’t need to worry about browsers caching their old versions.
However, there is a escape hatch that you can use to add an asset outside of the module system.
If you put a file into the public
folder, it will not be processed by Webpack. Instead, it will be copied into the build folder untouched. To reference assets in the public
folder, you need to use a special variable called PUBLIC_URL
.
Inside index.html
, you can use it like this:
<link rel="shortcut icon" href="%PUBLIC_URL%/favicon.ico">
Only files inside the public
folder will be accessible by %PUBLIC_URL%
prefix. If you need to use a file from src
or node_modules
, you’ll have to copy it there to explicitly specify your intention to make this file a part of the build.
When you run elm-app build
, Create Elm App will substitute %PUBLIC_URL%
with a correct absolute path, so your project works even if you use client-side routing or host it at a non-root URL.
In Elm code, you can use %PUBLIC_URL%
for similar purposes:
// Note: this is an escape hatch and should be used sparingly!
// Normally we recommend using `import` and `Html.programWithFlags` for getting
// asset URLs as described in “Adding Images and Fonts” above this section.
img [ src "%PUBLIC_URL%/logo.svg" ] []
In JavaScript code, you can use process.env.PUBLIC_URL
for similar purposes:
const logo = `<img src=${process.env.PUBLIC_URL + '/img/logo.svg'} />`;
Keep in mind the downsides of this approach:
- None of the files in
public
folder get post-processed or minified. - Missing files will not be called at compilation time, and will cause 404 errors for your users.
- Result filenames won’t include content hashes so you’ll need to add query arguments or rename them every time they change.
Normally we recommend importing stylesheets, images, and fonts from JavaScript.
The public
folder is used as a workaround for some less common cases:
- You need a file with a specific name in the build output, such as
manifest.webmanifest
. - You have thousands of images and need to dynamically reference their paths.
- You want to include a small script like
pace.js
outside of the bundled code. - Some library may be incompatible with Webpack and you have no other option but to include it as a
<script>
tag.
Note that if you add a <script>
that declares global variables, you also need to read the next section on using them.
In your JavaScript code you have access to variables declared in your
environment, like an API key set in an .env
-file or via your shell. They are
available on the process.env
-object and will be injected during build time.
Besides the NODE_ENV
variable you can access all variables prefixed with
ELM_APP_
:
# .env
ELM_APP_API_KEY="secret-key"
Alternatively, you can set them on your shell before calling the start- or build-script, e.g.:
ELM_APP_API_KEY="secret-key" elm-app start
Both ways can be mixed, but variables set on your shell prior to calling one of
the scripts will take precedence over those declared in an .env
-file.
Passing the variables to your Elm-code can be done via flags
:
// index.js
import { Main } from './Main.elm';
Main.fullscreen({
environment: process.env.NODE_ENV,
apiKey: process.env.ELM_APP_API_KEY,
});
-- Main.elm
type alias Flags = { apiKey : String, environment : String }
init : Flags -> ( Model, Cmd Msg )
init flags =
...
main =
programWithFlags { init = init, ... }
Be aware that you cannot override NODE_ENV
manually. See
this list from the dotenv
-library
for a list of files you can use to declare environment variables.
To forward the API ( REST ) calls to backend server, add a proxy to the elm-package.json
in the top level json object.
{
...
"proxy" : "http://localhost:1313",
...
}
Make sure the XHR requests set the Content-type: application/json
and Accept: application/json
.
The development server has heuristics, to handle its own flow, which may interfere with proxying of
other html and javascript content types.
curl -X GET -H "Content-type: application/json" -H "Accept: application/json" http://localhost:3000/api/list
Create Elm App uses elm-test as its test runner.
To use packages in tests, you also need to install them in tests
directory.
elm-app test --add-dependencies elm-package.json
- Following the Travis Getting started guide for syncing your GitHub repository with Travis. You may need to initialize some settings manually in your profile page.
- Add a
.travis.yml
file to your git repository.
language: node_js
sudo: required
node_js:
- '7'
install:
- npm i create-elm-app -g
script: elm-app test
- Trigger your first build with a git push.
- Customize your Travis CI Build if needed.
By default, the production build is a fully functional, offline-first Progressive Web App.
Progressive Web Apps are faster and more reliable than traditional web pages, and provide an engaging mobile experience:
- All static site assets are cached so that your page loads fast on subsequent visits, regardless of network connectivity (such as 2G or 3G). Updates are downloaded in the background.
- Your app will work regardless of network state, even if offline. This means your users will be able to use your app at 10,000 feet and on the Subway.
- On mobile devices, your app can be added directly to the user's home screen, app icon and all. You can also re-engage users using web push notifications. This eliminates the need for the app store.
The sw-precache-webpack-plugin
is integrated into production configuration,
and it will take care of generating a service worker file that will automatically
precache all of your local assets and keep them up to date as you deploy updates.
The service worker will use a cache-first strategy
for handling all requests for local assets, including the initial HTML, ensuring
that your web app is reliably fast, even on a slow or unreliable network.
If you would prefer not to enable service workers prior to your initial
production deployment, then remove the call to serviceWorkerRegistration.register()
from src/index.js
.
If you had previously enabled service workers in your production deployment and
have decided that you would like to disable them for all your existing users,
you can swap out the call to serviceWorkerRegistration.register()
in
src/index.js
with a call to serviceWorkerRegistration.unregister()
.
After the user visits a page that has serviceWorkerRegistration.unregister()
,
the service worker will be uninstalled. Note that depending on how /service-worker.js
is served,
it may take up to 24 hours for the cache to be invalidated.
-
Service workers require HTTPS, although to facilitate local testing, that policy does not apply to
localhost
. If your production web server does not support HTTPS, then the service worker registration will fail, but the rest of your web app will remain functional. -
Service workers are not currently supported in all web browsers. Service worker registration won't be attempted on browsers that lack support.
-
The service worker is only enabled in the production environment, e.g. the output of
npm run build
. It's recommended that you do not enable an offline-first service worker in a development environment, as it can lead to frustration when previously cached assets are used and do not include the latest changes you've made locally. -
If you need to test your offline-first service worker locally, build the application (using
npm run build
) and run a simple http server from your build directory. After running the build script,create-react-app
will give instructions for one way to test your production build locally and the deployment instructions have instructions for using other methods. Be sure to always use an incognito window to avoid complications with your browser cache. -
If possible, configure your production environment to serve the generated
service-worker.js
with HTTP caching disabled. If that's not possible—GitHub Pages, for instance, does not allow you to change the default 10 minute HTTP cache lifetime—then be aware that if you visit your production site, and then revisit again beforeservice-worker.js
has expired from your HTTP cache, you'll continue to get the previously cached assets from the service worker. If you have an immediate need to view your updated production deployment, performing a shift-refresh will temporarily disable the service worker and retrieve all assets from the network. -
Users aren't always familiar with offline-first web apps. It can be useful to let the user know when the service worker has finished populating your caches (showing a "This web app works offline!" message) and also let them know when the service worker has fetched the latest updates that will be available the next time they load the page (showing a "New content is available; please refresh." message). Showing this messages is currently left as an exercise to the developer, but as a starting point, you can make use of the logic included in
src/registerServiceWorker.js
, which demonstrates which service worker lifecycle events to listen for to detect each scenario, and which as a default, just logs appropriate messages to the JavaScript console. -
By default, the generated service worker file will not intercept or cache any cross-origin traffic, like HTTP API requests, images, or embeds loaded from a different domain. If you would like to use a runtime caching strategy for those requests, you can
eject
and then configure theruntimeCaching
option in theSWPrecacheWebpackPlugin
section ofwebpack.config.prod.js
.
The default configuration includes a web app manifest located at
public/manifest.json
, that you can customize with
details specific to your web application.
When a user adds a web app to their homescreen using Chrome or Firefox on
Android, the metadata in manifest.json
determines what
icons, names, and branding colors to use when the web app is displayed.
The Web App Manifest guide
provides more context about what each field means, and how your customizations
will affect your users' experience.
elm-app build
creates a build
directory with a production build of your app. Set up your favourite HTTP server so that a visitor to your site is served index.html
, and requests to static paths like /static/js/main.<hash>.js
are served with the contents of the /static/js/main.<hash>.js
file.
For environments using Node, the easiest way to handle this would be to install serve and let it handle the rest:
npm install -g serve
serve -s build
The last command shown above will serve your static site on the port 5000. Like many of serve’s internal settings, the port can be adjusted using the -p
or --port
flags.
Run this command to get a full list of the options available:
serve -h
The step below is important!
If you skip it, your app will not deploy correctly.
Open your elm-package.json
and add a homepage
field:
"homepage": "https://myusername.github.io/my-app",
Create Elm App uses the homepage
field to determine the root URL in the built HTML file.
elm-app build
We will use gh-pages to upload the files from the build
directory to GitHub. If you haven't already installed it, do so now (npm install -g gh-pages
)
Now run:
gh-pages -d build
Finally, make sure GitHub Pages option in your GitHub project settings is set to use the gh-pages
branch:
You can configure a custom domain with GitHub Pages by adding a CNAME
file to the public/
folder.
GitHub Pages doesn’t support routers that use the HTML5 pushState
history API under the hood (for example, React Router using browserHistory
). This is because when there is a fresh page load for a url like http://user.github.io/todomvc/todos/42
, where /todos/42
is a frontend route, the GitHub Pages server returns 404 because it knows nothing of /todos/42
. If you want to add a router to a project hosted on GitHub Pages, here are a couple of solutions:
- You could switch from using HTML5 history API to routing with hashes.
- Alternatively, you can use a trick to teach GitHub Pages to handle 404 by redirecting to your
index.html
page with a special redirect parameter. You would need to add a404.html
file with the redirection code to thebuild
folder before deploying your project, and you’ll need to add code handling the redirect parameter toindex.html
. You can find a detailed explanation of this technique in this guide.
Remember to disable safe write if you are using VIM or IntelliJ IDE, such as WebStorm.