Get started: write, test, and deploy your first functions

To get started with Cloud Functions, try working through this tutorial, which starts with the required setup tasks and works through creating, testing, and deploying two related functions:

addMessage(), which exposes a URL that accepts a text value and writes it to Cloud Firestore.
makeUppercase(), which triggers on Cloud Firestore write and transforms the text to uppercase.

We've chosen Cloud Firestore and HTTP-triggered JavaScript functions for this sample in part because these background triggers can be thoroughly tested through the Firebase Local Emulator Suite. This toolset also supports Realtime Database, PubSub, Auth, and HTTP callable triggers. Other types of background triggers such as Remote Config, TestLab, and Analytics triggers can all be tested interactively using toolsets not described in this page.

The following sections of this tutorial detail the steps required to build, test, and deploy the sample. If you'd rather just run the code and inspect it, jump to Review complete sample code.

Create a Firebase Project

In the Firebase console, click Add project, then select or enter a Project name.

If you have an existing Google Cloud Platform (GCP) project, you can select the project from the dropdown menu to add Firebase resources to that project.
(Optional) If you are creating a new project, you can edit the Project ID.

Firebase automatically assigns a unique ID to your Firebase project. Visit Understand Firebase Projects to learn about how Firebase uses the project ID.

After Firebase provisions resources for your Firebase project, you cannot change your project ID.
To use a specific identifier, you must edit your project ID during this setup step.
Click Continue.

(Optional) Set up Google Analytics for your project, which enables you to have an optimal experience using any of the following Firebase products:

When prompted, select to use an existing Google Analytics account or to create a new account.
If you choose to create a new account, select your Analytics reporting location, then accept the data sharing settings and Google Analytics terms for your project.

Click Create project (or Add Firebase, if you're using an existing GCP project).

Firebase automatically provisions resources for your Firebase project. When the process completes, you'll be taken to the overview page for your Firebase project in the Firebase console.

Set up Node.js and the Firebase CLI

You'll need a Node.js environment to write functions, and you'll need the Firebase CLI to deploy functions to the Cloud Functions runtime. For installing Node.js and npm, Node Version Manager is recommended.

Once you have Node.js and npm installed, install the Firebase CLI via your preferred method. To install the CLI via npm, use:

npm install -g firebase-tools

This installs the globally available firebase command. If the command fails, you may need to change npm permissions. To update to the latest version of firebase-tools, rerun the same command.

In many cases, new features and bug fixes are available only with the latest version of the Firebase CLI and the firebase-functions SDK. It's a good practice to frequently update both the Firebase CLI and the SDK with these commands inside the functions folder of your Firebase project:

npm install firebase-functions@latest firebase-admin@latest --save
npm install -g firebase-tools

Initialize Firebase SDK for Cloud Functions

When you initialize Firebase SDK for Cloud Functions, you create an empty project containing dependencies and some minimal sample code, and you choose either TypeScript or JavaScript for composing functions.

To initialize your project:

Run firebase login to log in via the browser and authenticate the firebase tool.
Go to your Firebase project directory.
Run firebase init functions. The tool gives you an option to install dependencies with npm. It is safe to decline if you want to manage dependencies in another way, though if you do decline you'll need to run npm install before emulating or deploying your functions.
The tool gives you two options for language support:
- JavaScript
- TypeScript. See Write Functions with TypeScript for more information.
For this tutorial, select JavaScript.

After these commands complete successfully, your project structure looks 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

The package.json file created during initialization contains an important key: "engines": {"node": "8"}. This specifies your node version for writing and deploying functions. You can select other supported versions.

Import the required modules and initialize an app

After you have completed the setup tasks, you can open the source directory and start adding code as described in the following sections. For this sample, your project must import the Cloud Functions and Admin SDK modules using Node require statements. Add lines like the following to your index.js file:

// The Cloud Functions for Firebase SDK to create Cloud Functions and setup triggers.
const functions = require('firebase-functions');

// The Firebase Admin SDK to access Cloud Firestore.
const admin = require('firebase-admin');
admin.initializeApp();index.js

These lines load the firebase-functions and firebase-admin modules, and initialize an admin app instance from which Cloud Firestore changes can be made. Wherever Admin SDK support is available, as it is for FCM, Authentication, and Firebase Realtime Database, it provides a powerful way to integrate Firebase using Cloud Functions.

The Firebase CLI automatically installs the Firebase and Firebase SDK for Cloud Functions Node modules when you initialize your project. To add 3rd party libraries to your project, you can modify package.json and run npm install. For more information, see Handle Dependencies.

Add the `addMessage()` function

For the addMessage() function, add these lines to index.js:

// Take the text parameter passed to this HTTP endpoint and insert it into 
// Cloud Firestore under the path /messages/:documentId/original
exports.addMessage = functions.https.onRequest(async (req, res) => {
  // Grab the text parameter.
  const original = req.query.text;
  // Push the new message into Cloud Firestore using the Firebase Admin SDK.
  const writeResult = await admin.firestore().collection('messages').add({original: original});
  // Send back a message that we've succesfully written the message
  res.json({result: `Message with ID: ${writeResult.id} added.`});
});index.js

The addMessage() function is an HTTP endpoint. Any request to the endpoint results in ExpressJS-style Request and Response objects passed to the onRequest() callback.

HTTP functions are synchronous (similar to callable functions), so you should send a response as quickly as possible and defer work using Cloud Firestore. The addMessage() HTTP function passes a text value to the HTTP endpoint and inserts it into the database under the path /messages/:documentId/original.

Add the `makeUppercase()` function

For the makeUppercase() function, add these lines to index.js:

// Listens for new messages added to /messages/:documentId/original and creates an
// uppercase version of the message to /messages/:documentId/uppercase
exports.makeUppercase = functions.firestore.document('/messages/{documentId}')
    .onCreate((snap, context) => {
      // Grab the current value of what was written to Cloud Firestore.
      const original = snap.data().original;

      // Access the parameter `{documentId}` with `context.params`
      functions.logger.log('Uppercasing', context.params.documentId, original);
      
      const uppercase = original.toUpperCase();
      
      // You must return a Promise when performing asynchronous tasks inside a Functions such as
      // writing to Cloud Firestore.
      // Setting an 'uppercase' field in Cloud Firestore document returns a Promise.
      return snap.ref.set({uppercase}, {merge: true});
    });index.js

The makeUppercase() function executes when Cloud Firestore is written to. The ref.set function defines the document to listen on. For performance reasons, you should be as specific as possible.

Braces—for example, {documentId}—surround "parameters," wildcards that expose their matched data in the callback.

Cloud Firestore triggers the onWrite() callback whenever data is written or updated on the given document.

Event-driven functions such as Cloud Firestore events are asynchronous. The callback function should return either a null, an Object, or a Promise. If you do not return anything, the function times out, signaling an error, and is retried. See Sync, Async, and Promises.

Emulate execution of your functions

The Firebase Local Emulator Suite allows you to build and test apps on your local machine instead of deploying to a Firebase project. Local testing during development is strongly recommended, in part because it lowers the risk from coding errors that could potentially incur cost in a production environment (for example, an infinite loop).

To emulate your functions:

Run firebase emulators:start and check the output for the URL of the Emulator Suite UI. It defaults to localhost:4000, but may be hosted on a different port on your machine. Enter that URL in your browser to open the Emulator Suite UI.
Check the output of the firebase emulators:start command for the URL of the http function addMessage(). It will look similar to http://localhost:5001/MY_PROJECT/us-central1/addMessage, except that:
1. MY_PROJECT will be replaced with your project ID.
2. The port may be different on your local machine.
Add the query string ?text=uppercaseme to the end of the function's URL. It should now look something like: http://localhost:5001/MY_PROJECT/us-central1/addMessage?text=uppercaseme. Optionally, you can change the message "uppercaseme" to a custom message.
Create a new message by opening the URL in a new tab in your browser.
View the effects of the functions in the Emulator Suite UI:
1. In the Logs tab, you should see new logs indicating that the functions addMessage() and makeUppercase() ran:
  
  i functions: Beginning execution of "addMessage"
  
  i functions: Beginning execution of "makeUppercase"
2. In the Firestore tab, you should see a document containing your original message as well as the uppercased version of your message (if it was originally "uppercaseme", you'll see "UPPERCASEME").

Deploy functions to a production environment

Once your functions are working as desired in the emulator, you can proceed to deploying, testing, and running them in the production environment.

To complete the tutorial, deploy your functions and then execute addMessage() to trigger makeUppercase().

Run this command to deploy your functions:
```
$ firebase deploy --only functions
```
After you run this command, the Firebase CLI outputs the URL for any HTTP function endpoints. In your terminal, you should see a line like the following:
```
Function URL (addMessage): https://us-central1-MY_PROJECT.cloudfunctions.net/addMessage
```
The URL contains your project ID as well as a region for the HTTP function. Though you don't need to worry about it now, some production HTTP functions should specify a location to minimize network latency.

If you encounter access errors such as "Unable to authorize access to project," try checking your project aliasing.
Using the addMessage() URL output by the CLI, add a text query parameter, and open it in a browser:
```
https://us-central1-MY_PROJECT.cloudfunctions.net/addMessage?text=uppercasemetoo
```
The function executes and redirects the browser to the Firebase console at the database location where the text string is stored. This write event triggers makeUppercase(), which writes an uppercase version of the string.

After deploying and executing functions, you can view logs in the Firebase console for Cloud Functions. If you need to delete functions in development or production, use the Firebase CLI.

Review complete sample code

Here's the completed functions/index.js containing the functions addMessage() and makeUppercase(). These functions allow you to pass a parameter to an HTTP endpoint that writes a value to Cloud Firestore, and then transforms it by uppercasing all characters in the string.

// The Cloud Functions for Firebase SDK to create Cloud Functions and setup triggers.
const functions = require('firebase-functions');

// The Firebase Admin SDK to access Cloud Firestore.
const admin = require('firebase-admin');
admin.initializeApp();

// Take the text parameter passed to this HTTP endpoint and insert it into 
// Cloud Firestore under the path /messages/:documentId/original
exports.addMessage = functions.https.onRequest(async (req, res) => {
  // Grab the text parameter.
  const original = req.query.text;
  // Push the new message into Cloud Firestore using the Firebase Admin SDK.
  const writeResult = await admin.firestore().collection('messages').add({original: original});
  // Send back a message that we've succesfully written the message
  res.json({result: `Message with ID: ${writeResult.id} added.`});
});

// Listens for new messages added to /messages/:documentId/original and creates an
// uppercase version of the message to /messages/:documentId/uppercase
exports.makeUppercase = functions.firestore.document('/messages/{documentId}')
    .onCreate((snap, context) => {
      // Grab the current value of what was written to Cloud Firestore.
      const original = snap.data().original;

      // Access the parameter `{documentId}` with `context.params`
      functions.logger.log('Uppercasing', context.params.documentId, original);
      
      const uppercase = original.toUpperCase();
      
      // You must return a Promise when performing asynchronous tasks inside a Functions such as
      // writing to Cloud Firestore.
      // Setting an 'uppercase' field in Cloud Firestore document returns a Promise.
      return snap.ref.set({uppercase}, {merge: true});
    });index.js

Next steps

In this documentation, you can find more information on general concepts for Cloud Functions as well as guides for writing functions to handle the event types supported by Cloud Functions.

To learn more about Cloud Functions, you could also do the following:

Read about use cases for Cloud Functions.
Try the Cloud Functions codelab.
Review and run code samples on GitHub.
Review the API reference.

Video tutorial

You can learn more about Cloud Functions by watching video tutorials. In this video, you'll find detailed guidance on getting started with Cloud Functions, including Node.js and CLI setup.