Writing & Deploying Cloud Functions

Learn how to write, test & deploy Cloud Functions to your Firebase project.

Cloud Functions are a powerful asset to a developers workflow, allowing you to build complex backend tasks with minimal maintenance overhead. The following page outlines the steps required for writing, testing & deploying Cloud Functions to your Firebase project.

Environment Setup

Firebase provides a CLI which is required to build and deploy Cloud Functions. To install the CLI, install the firebase-tools package globally on your computer from your terminal:

npm install -g firebase-tools

Once installed, login to Firebase with the CLI. This process will automatically open a browser instance giving you the ability to login to your Firebase account.

firebase login

Once logged in, create a new directory on your development environment. This will be used as our working directory where our Cloud Functions will be written and deployed from. Within this directory, run the following command from your terminal to initialize a new project structure:

firebase init functions

You will be offered two options for language support, for this tutorial select JavaScript. Allow the CLI to install dependencies using NPM. Once complete your project structure will look like this:

myproject
 +- .firebaserc    # Hidden file that helps you quickly switch between
 |                 # projects with `firebase use`
 |
 +- firebase.json  # Describes properties for your project
 |
 +- functions/     # Directory containing all your functions code
      |
      +- .eslintrc.json  # Optional file containing rules for JavaScript linting.
      |
      +- package.json  # NPM package file describing your Cloud Functions code
      |
      +- index.js      # main source file for your Cloud Functions code
      |
      +- node_modules/ # directory where your dependencies (declared in
                       # package.json) are installed

Writing a Function

The Firebase CLI has created a project structure and also installed a number of dependencies which will be used to build our Cloud Functions.

To enable us to mock some fake data to use in the deployed functions, lets use the @faker-js/faker library to create mock data.

cd functions/
npm install @faker-js/faker

Now it's time to write our Cloud Function. Open up the generated functions/index.js file in your chosen editor. The CLI has already imported the firebase-functions package required to build a Cloud Function. Firebase uses named exports to help identify functions. These exports are used to build the API endpoint name which will be accessible from our React Native application.

For this tutorial, we are creating a product listing API. Go ahead and create a new HTTPS callable named function called listProducts:

// functions/index.js
const functions = require('firebase-functions');

exports.listProducts = functions.https.onCall((data, context) => {
  // ...
});

The onCall callback returns two objects:

We can now return an array of products, generated from the faker library. As we are mocking a data set, it's important to keep consistent results for each request. The data should be generated before the request is received, rather than a new data set being generated on each request:

// functions/index.js
const functions = require('firebase-functions');
const { faker } = require('@faker-js/faker');

// Initialize products array
const products = [];

// Max number of products
const LIMIT = 100;

// Push a new product to the array
for (let i = 0; i < LIMIT; i++) {
  products.push({
    name: faker.commerce.productName(),
    price: faker.commerce.price(),
  });
}

exports.listProducts = functions.https.onCall((data, context) => {
  return products;
});

Testing your function

Before deploying our functions project, we can run the serve command which builds a locally accessible instance of our Cloud Functions. Run the following command from within the functions/ directory:

cd functions/
npm run serve

Once booted, the CLI will be provide a local web server with the products endpoint openly available, e.g:

functions: listProducts: http://localhost:5000/rnfirebase-demo/us-central1/listProducts

In your terminal (or browser), access the endpoint provided. Our list of generated products is ready for use.

curl -i -H "Accept: application/json" -H "Content-Type: application/json" -X POST -d '{"data":{}}' http://localhost:5000/rnfirebase-demo/us-central1/listProducts

Security

By default the endpoint will be publicly accessible when deployed. Firebase offers an out-of-the-box solution for handling authentication, which integrates with the Authentication module. To secure our endpoint for authenticated users only, check whether the auth property exists on the function execution context:

exports.listProducts = functions.https.onCall((data, context) => {
  if (!context.auth) {
    throw new functions.https.HttpsError('unauthenticated', 'Endpoint requires authentication!');
  }

  return products;
});

When calling the function without authentication, an error response will be returned to the caller.

If the user is authenticated, we can access the users data via the context.auth property. For example their unique user identifier will be available by accessing context.auth.uid.

Handling function arguments

A common requirement for endpoints is calling the endpoint with custom parameters. For example, rather than returning a list of 1000 products, we can paginate the data by passing in arguments when calling our function.

These arguments can be accessed via the data property when the function is executed, let's update our function code to support pagination arguments:

exports.listProducts = functions.https.onCall((data, context) => {
  const { page = 1, limit = 10 } = data;

  const startAt = (page - 1) * limit;
  const endAt = startAt + limit;

  return products.slice(startAt, endAt);
});

Deploying Functions

Once your functions are ready to be deployed, the project provides a deploy script which will upload all of your code onto the Firebase infrastructure and automatically provision production ready endpoints. Within the project, run the deploy script from the functions directory:

cd functions/
npm run deploy

Once complete, your Cloud Function will also be available from a publicly accessible endpoint if required, for example:

https://us-central1-rnfirebase-demo-23aa8.cloudfunctions.net/listProducts

Calling your function

Once your function has been deployed you can now call it through the React Native Firebase Functions SDK in your application:

import { firebase } from '@react-native-firebase/functions';

// ...

// note the name of our deployed function, 'listProducts', is referenced here:
const { data } = await firebase.functions().httpsCallable('listProducts')({
  page: 1,
  limit: 15,
});

// ...