Get started with Firebase Data Connect

In this quickstart, you will learn how to:

Add Firebase Data Connect to your Firebase project.
Set up a development environment including a Visual Studio Code extension to work with a production instance.
Then we will show you how to:
- Create a schema for an email app and deploy to production.
- Define queries and mutations for your schema.
- Deploy your final prototype to production.

Prerequisites

To use this quickstart, you'll need the following.

Linux, macOS, or Windows
Visual Studio Code

Connect to your Firebase project

If you haven't already, create a Firebase project.
1. In the Firebase console, click Add project, then follow the on-screen instructions.
Upgrade your project to the Blaze plan. This lets you create a Cloud SQL for PostgreSQL instance.

Note: Though you set up billing in your Blaze upgrade, you won't be charged for usage of Data Connect or the default Cloud SQL for PostgreSQL configuration during the preview.
Navigate to the Data Connect section of the Firebase console and follow the product setup workflow.
Select a location for your CloudSQL for PostgreSQL database.
Note the project, service and database names and IDs for confirmation later.
Follow the remaining setup flow then click Done.

Note: It may take up to 15 minutes to finish provisioning your Cloud SQL for PostgreSQL instance. You can confirm the Data Connect setup flow, including database provisioning, is complete using the Google Cloud console.

Choose and set up a development environment

Data Connect supports two development experiences for prototyping:

If you're a Kotlin Android, iOS or web developer, you can use VS Code development to prototype schemas and operations locally while connecting to your Cloud SQL for PostgreSQL instance.
If you're a web developer, you can use IDX Development to prototype in an IDX workspace using a pre-configured IDX template with PostgreSQL, VS Code extension with Data Connect emulator, and quickstart code set up for you. You'll find more information at the Project IDX site.

This quickstart focuses on the VS Code extension development flow. To continue:

Create a new directory for your local project.
Open VS Code in the new directory.
Download the extension, bundled as a VSIX package, from Firebase Storage.

Note: You cannot double-click the VSIX package to install the extension. Be sure to follow the next few steps to properly install it.
In VS Code, from the View menu, select Extensions.
On the Extensions panel title bar, click the menu icon more_horiz, then follow Install from VSIX....

Note: The extension includes the marketplace GraphQL Syntax Highlighting extension. Other GraphQL extensions you install may conflict with GraphQL support bundled in the VSIX file.

Optionally, you can install a local PostgreSQL database for local development with the Data Connect emulator. This setup is covered at the end of this quickstart guide.

Set up your local project

Install the CLI, following the normal instructions. If you have npm already installed, run the following command:

npm install -g firebase-tools

Set up your project directory

To set up your local project, initialize your project directory.

In the VS Code left-hand panel, click the Firebase icon to open the Firebase VS Code extension UI.

In the Firebase extension UI:

Make sure you've signed in.
Click the Run firebase init button.
Check the Terminal tab in the VS Code lower Panel for prompts.
Select Data Connect as a feature to use in this directory.
When prompted, supply the project, service and database IDs of the Data Connect project you created earlier in the console.

Create a schema

In your Firebase project directory, in the /dataconnect/schema/schema.gql file, start defining a GraphQL schema that includes users and emails.

User

In Data Connect, GraphQL fields are mapped to columns. Users have a uid, name, and email address. Data Connect recognizes several primitive data types: String and Date.

Copy the following snippet or uncomment the corresponding lines in the file.

# File `/dataconnect/schema/schema.gql`

type User @table(key: "uid") {
   uid: String!
   name: String!
   address: String!
}

By default Firebase Data Connect will add a UUID id key if none is provided. However, in this case you want my uid to be the primary key, which you can do through the @table(key: "uid") directive.

Email

Now that you have users, you can model emails. Here you can add the typical fields (or columns) for email data. This time, we omit adding a primary key because you can rely on Data Connect to manage it.

# File `/dataconnect/schema/schema.gql`

type Email @table {
   subject: String!
   sent: Date!
   text: String!
   from: User!
}

Notice that the from field is mapped to a type of User. Data Connect understands that this is a relationship between Email and User and will manage this relationship for you.

Deploy your schema to production

If you are using the Firebase VS Code extension to work with your production database, you need to deploy your schema before continuing. After deploying your schema to your production database, you should be able to view the schema on the console.

You can use the Firebase VS Code extension to deploy.
- In the extension UI, under the Firebase Data Connect panel, click Deploy.
You may need to review schema changes and approve potentially destructive modifications. You'll be prompted to:
- Review schema changes using firebase dataconnect:sql:diff
- When you are satisfied with changes, apply them using the flow started by firebase dataconnect:sql:migrate.

Add data to your tables

In the VS Code editor panel, you can see CodeLens buttons appear over the GraphQL types in /dataconnect/schema/schema.gql. Since you've deployed your schema to production, you can use the Add data and Run (Production) buttons add data to your database on the backend.

To add records to the User and Email tables:

In schema.gql, click the Add data button above the User type declaration.
In the User_insert.gql file that is generated, hard code data for the three fields.
Click the Run (Production) button.
Repeat the previous steps to add a record to the Email table, supplying the uid of your user in the fromUid field, as prompted in the generated Email_insert mutation.

Define your query

Now the fun part, queries. As a developer, you're accustomed to writing SQL queries rather than GraphQL queries, so this can feel a bit different at first. However, GraphQL is far more terse and type-safe than raw SQL. And our VS Code extension eases the development experience.

Start editing the /dataconnect/connector/queries.gql file. If you want to get all emails, use a query like this.

# File `/dataconnect/connector/queries.gql`

query ListEmails @auth(level: NO_ACCESS) {
  emails {
    id, subject, text, sent
    from {
      name
    }
  }
}

Execute the query using the nearby CodeLens button.

A really exciting feature here is the ability to treat the database's relationships like a graph. Since an Email has a from field that references a User, you can nest into the field and get information about the user back.

Deploy your schema and query to production

You have worked through a development iteration. Now you can deploy your schema, data, and queries to the server with the Firebase extension UI or the Firebase CLI, just as you did with your schema (#deploy-schema).

If you use the Firebase VS Code extension to deploy, click the Deploy button.

Once deployed, you should be able to view and run your operations on the console as well. Your Data Connect service will be ready to process operations from clients. The Cloud SQL for PostgreSQL instance will be updated with its final deployed generated schema and data.

(Optional) Install PostgreSQL locally

Installing PostgreSQL locally and integrating it with the emulator lets you prototype in a fully local development environment.

You can install a new instance of PostgreSQL or use an existing instance.

Install PostgreSQL

Install PostgreSQL version 15.x following instructions for your platform.

macOS. Download and install Postgres.app.
Windows: Use the EDB installer from the PostgreSQL downloads page.
Docker: Pull and run the pgvector/pgvector:pg15 image, which comes with both PostgreSQL 15.x and vector support.
Linux: We recommend using Docker with the preceding image, but you can also follow alternative instructions for popular distributions.

Note hostname, port, username and password and related parameters output during the installation sequence.

To connect to your PostgreSQL instance, the emulator needs:

These setup configuration parameters
The database name from your dataconnect.yaml and a correspondingly-named database initialized in your local instance.

Use your local PostgreSQL instance

You can use an existing local PostgreSQL instance by updating the Data Connect emulator settings.

firebase setup:emulators:dataconnect

When prompted, enter the PostgreSQL connection string in this format: postgresql://postgresusername:postgrespassword@localhost:5432/postgresdatabase?sslmode=disable.

For more on the connection string, refer to the PostgreSQL documentation.

Connect to your local PostgreSQL instance

With this configuration done, to connect to your local database:

In VS Code, in the left-hand panel, click the Firebase icon to open the Firebase VS Code extension UI.
Click the Connect to Local PostgreSQL button.