3-es6-babel-gulp

3 - Setting up ES6 with Babel and Gulp

We're now going to use ES6 syntax, which is a great improvement over the "old" ES5 syntax. All browsers and JS environments understand ES5 well, but not ES6. So we're going to use a tool called Babel to transform ES6 files into ES5 files. To run Babel, we are going to use Gulp, a task runner. It is similar to the tasks located under scripts in package.json, but writing your task in a JS file is simpler and clearer than a JSON file, so we'll install Gulp, and the Babel plugin for Gulp too:

• Run yarn add --dev gulp
• Run yarn add --dev gulp-babel
• Run yarn add --dev babel-preset-latest
• In package.json, add a babel field for the babel configuration. Make it use the latest Babel preset like this:

"babel": {
  "presets": [
    "latest"
  ]
},

Note: A .babelrc file at the root of your project could also be used instead of the babel field of package.json. Your root folder will get more and more bloated over time, so keep the Babel config in package.json until it grows too large.

• Move your index.js into a new src folder. This is where you will write your ES6 code. A lib folder is where the compiled ES5 code will go. Gulp and Babel will take care of creating it. Remove the previous color-related code in index.js, and replace it with a simple:

const str = 'ES6';
console.log(`Hello ${str}`);

We're using a template string here, which is an ES6 feature that lets us inject variables directly inside the string without concatenation using ${}. Note that template strings are created using backquotes.

• Create a gulpfile.js containing:

const gulp = require('gulp');
const babel = require('gulp-babel');
const del = require('del');
const exec = require('child_process').exec;

const paths = {
  allSrcJs: 'src/**/*.js',
  libDir: 'lib',
};

gulp.task('clean', () => {
  return del(paths.libDir);
});

gulp.task('build', ['clean'], () => {
  return gulp.src(paths.allSrcJs)
    .pipe(babel())
    .pipe(gulp.dest(paths.libDir));
});

gulp.task('main', ['build'], (callback) => {
  exec(`node ${paths.libDir}`, (error, stdout) => {
    console.log(stdout);
    return callback(error);
  });
});

gulp.task('watch', () => {
  gulp.watch(paths.allSrcJs, ['main']);
});

gulp.task('default', ['watch', 'main']);

Let's take a moment to understand all this.

The API of Gulp itself is pretty straightforward. It defines gulp.tasks, that can reference gulp.src files, applies a chain of treatments to them with .pipe() (like babel() in our case) and outputs the new files to gulp.dest. It can also gulp.watch for changes on your filesystem. Gulp tasks can run prerequisite tasks before them, by passing an array (like ['build']) as a second parameter to gulp.task. Refer to the documentation for a more thorough presentation.

First we define a paths object to store all our different file paths and keep things DRY.

Then we define 5 tasks: build, clean, main, watch, and default.

• build is where Babel is called to transform all of our source files located under src and write the transformed ones to lib.
• clean is a task that simply deletes our entire auto-generated lib folder before every build. This is typically useful to get rid of old compiled files after renaming or deleting some in src, or to make sure the lib folder is in sync with the src folder if your build fails and you don't notice. We use the del package to delete files in a way that integrates well with Gulp's stream (this is the recommended way to delete files with Gulp). Run yarn add --dev del to install that package.
• main is the equivalent of running node . in the previous chapter, except this time, we want to run it on lib/index.js. Since index.js is the default file Node looks for, we can simply write node lib (we use the libDir variable to keep things DRY). The require('child_process').exec and exec part in the task is a native Node function that executes a shell command. We forward stdout to console.log() and return a potential error using gulp.task's callback function. Don't worry if this part is not super clear to you, remember that this task is basically just running node lib.
• watch runs the main task when filesystem changes happen in the specified files.
• default is a special task that will be run if you simply call gulp from the CLI. In our case we want it to run both watch and main (for the first execution).

Note: You might be wondering how come we're using some ES6 code in this Gulp file, since it doesn't get transpiled into ES5 by Babel. This is because we're using a version of Node that supports ES6 features out of the box (make sure you are running Node > 6.5.0 by running node -v).

Alright! Let's see if this works.

• In package.json, change your start script to: "start": "gulp".

• Run yarn start. It should print "Hello ES6" and start watching for changes. Try writing bad code in src/index.js to see Gulp automatically showing you the error when you save.

• Add /lib/ to your .gitignore

Next section: 4 - Using the ES6 syntax with a class

Back to the previous section or the table of contents.