Configuring Firebase and Your Application

First, set up Firebase and link it to your application. Refer to the following article for detailed instructions.

Initializing Cloud Functions

Authenticate with Firebase by executing the command firebase login. Then, from your project root directory, run the command firebase init functions. The tool will prompt you with several questions that you'll need to answer. Here’s an example workflow:

• Are you prepared to continue? Yes
• Please choose an option: Use an existing project
• Which language would you like to use for Cloud Functions? JavaScript
• Would you like to use ESLint for bug detection and style enforcement? No
• Do you want to install dependencies with npm now? Yes

Once completed successfully, a new functions folder will appear in your project structure.

Now, it’s time to write the code that will execute in the cloud.

Developing a Cloud Function

For demonstration purposes, our function will take a string argument and convert it to uppercase. Although simple, it will provide insight into what can be accomplished.

Open the index.js file in the functions folder and enter the following code:

We define a callable method named toUpperCase with two parameters: data (the string to transform) and a CallableContext, which won’t be necessary for this example. We then utilize JavaScript's built-in toUpperCase method and return the outcome. Simple, isn’t it?

Ensure your code is idempotent! This means it should yield the same result even when invoked multiple times, allowing for easy retries of failed executions.

You may also opt to use TypeScript as your programming language; just remember to select “TypeScript” during initialization instead of “JavaScript.” Currently, no other languages are supported.

Deploying a Cloud Function

Deployment is straightforward; simply execute the command firebase deploy --only functions and wait for the process to complete, which may take up to five minutes for your dashboard to refresh. The dashboard will also show logs and function usage details.

Invoking a Cloud Function

A callable function is identified by the method functions.http.onCall((data, context) => {...});. The Firebase dashboard does not differentiate between callable functions and HTTP endpoints; both are categorized as “Request.” To invoke a function named toUpperCase(), consider the following code example:

HTTP Request Endpoint

Using the code functions.https.onRequest((request, result) => {...});, you can create an HTTP endpoint that other web services can call as needed. This allows you to define interfaces for third-party software to interact with your data or logic. Note that you don’t require the Firebase package to call endpoints; a tool for creating requests, such as http or Dio, is sufficient. Here’s an example code snippet:

Make sure to replace the link to the function with your deployed version! The `<server>` and `<projectId>` are placeholders.

I have also created an article with demo code on how to create and send HTTP requests in Flutter, which may be useful if you're unfamiliar with the topic.

Scheduling a Function

You can also schedule a function's execution. The interval can be specified using the App Engine cron.yaml syntax or the traditional Unix crontab syntax. The following code snippet defines a function that runs every ten minutes.

The next example utilizes Unix crontab syntax and specifies a timezone. Click here for a comprehensive list of all supported timezones.

Regions

Firebase operates in various regions that influence pricing and latency. You can set your desired region using the region() method along with a valid region name as a parameter. For example: functions.region("europe-west3"). Click here for a list of all available regions. It’s advisable to choose the region closest to your location. You can also specify multiple regions, separated by commas.

Logging

To track function executions, you can utilize logging. The output will be visible in your Firebase Functions dashboard.

const functions = require("firebase-functions"); functions.logger.log("This is log:", someObj); functions.logger.info("This is info:", someObj); functions.logger.warn("This is warn:", someObj); functions.logger.error("This is error:", someObj);

For more details on logging, please refer to the official documentation.

onCall() or onRequest()?

Callable functions and HTTP functions are quite similar. A callable function is essentially an HTTP function with specific request parameters. The main distinctions are that authentication tokens for Firebase and Firebase Cloud Messaging, along with App Check tokens, are automatically included in the request for callable functions if available. Additionally, callable functions validate these tokens by default, offering authentication features at no cost when using the onCall() method.

Conversely, HTTP functions lack integrated access control features; you must implement them yourself. However, you can execute a callable function via an HTTP request if you supply the correct data. For further information, refer to the documentation here.

• Use callable functions if your application is the only consumer.
• Opt for HTTP functions if you have third-party consumers (for instance, a public REST API), ensuring you incorporate access control and other security measures.

Conclusion

Firebase Cloud Functions present a convenient method to leverage external computation power within your Flutter app. Here’s a brief demo video showcasing the source code.

You can find the complete example source code on my GitHub page.