Get started with Firebase SQL Connect on iOS

In this quickstart, you will create and deploy a small sample database and access it from an iOS app.

Prerequisites

To complete this quickstart, you'll need the following:

  • An environment with the following tools installed:
    • A recent version of Xcode.
    • The Firebase CLI. If you have NPM installed, run: 
      npm install -g firebase-tools@latest
       Otherwise, see the docs for installation instructions.
  • A Google Account.

Tutorial

Tutorial

1. Create a new Xcode project

In Xcode, create a new project with the App template.

Name the project Quickstart with the bundle identifier com.example.Quickstart.

Xcode new project dialog

Create the project in an empty folder, like ~/qs-ios. Xcode will create a folder, ~/qs-ios/Quickstart, that contains your project files.

2. Initialize a Firebase project

Change to the project directory of your Xcode project and initialize a Firebase project in it.

 
cd ~/qs-ios
 
firebase login --reauth
 
firebase init dataconnect

When prompted, choose the following options:

  • Create a new Firebase project.
  • Don't create a schema with Gemini (in this tutorial, you'll use a pre-built example schema).
  • Provision a free trial Cloud SQL instance and database.

Accept the default values for all other prompts.

Next, register the iOS app in your Firebase project.

firebase apps:create --bundle-id com.example.Quickstart ios ios-quickstart

When prompted to specify an App Store ID, press Enter to skip the step.

firebase apps:sdkconfig ios -o Quickstart/GoogleService-Info.plist

In Xcode, select File > Add Files to "Quickstart" and select the GoogleService-Info.plist file.

3. Review the example GraphQL definitions

In SQL Connect, you define all of your database schemas and operations using GraphQL. When you initialized your project, the Firebase CLI created some example definitions to get you started.

 dataconnect/schema/schema.gql (excerpt) 
type Movie @table {
  title: String!
  imageUrl: String!
  genre: String
}

type MovieMetadata @table {
  movie: Movie! @unique
  rating: Float
  releaseYear: Int
  description: String
}
 dataconnect/example/queries.gql (excerpt) 
query ListMovies @auth(level: PUBLIC) {
  movies {
    id
    title
    imageUrl
    genre
  }
}

4. Deploy your schemas and operations

Whenever you make changes to your database schemas, queries, or mutations, you must deploy them for your changes to take effect on the database.

 
firebase deploy --only dataconnect

5. Seed the database with sample data

This seed data will give you something to look at when you test the sample app. Note that in this step you are executing arbitrary GraphQL, which is allowed for administrative tasks.

 
firebase dataconnect:execute dataconnect/seed_data.gql

6. Generate an iOS client SDK

This command uses your GraphQL definitions to generate an iOS client SDK specifically for your database. You use this library in your client app to perform all database operations.

You can generate libraries for multiple platforms, including Kotlin for Android, JavaScript for web, and Flutter, by adding definitions to connector.yaml.

 
firebase dataconnect:sdk:generate
 Auto-generated iOS SDK (excerpt) 
public class ListMoviesQuery {

  // ...

  @MainActor
  public func execute(
    fetchPolicy: QueryFetchPolicy = .preferCache,
  ) async throws -> OperationResult<ListMoviesQuery.Data> {
    var variables = ListMoviesQuery.Variables()

    let ref = dataConnect.query(
      name: "ListMovies", variables: variables, resultsDataType: ListMoviesQuery.Data.self,
      publisher: .observableMacro)

    let refCast = ref as! QueryRefObservation<ListMoviesQuery.Data, ListMoviesQuery.Variables>
    return try await refCast.execute(fetchPolicy: fetchPolicy)
  }
}

7. Add Firebase dependencies to your Xcode project

Add the generated library to your project using Swift Package Manager. When you add the generated library, it will transitively include the Firebase core libraries and the Firebase SQL Connect library.

In the Xcode navigation bar, select File > Add Package Dependencies > Add Local, and choose the folder containing the generated library, FirebaseDataConnectGenerated/DataConnectGenerated/. The correct folder will contain a file named Package.swift.

8. Write a sample iOS client

Replace the contents of Quickstart/QuickstartApp.swift with this simple iOS app.

Notice that the app completes the necessary database access using a function from the generated SDK.

 
import SwiftUI
import FirebaseCore
import FirebaseDataConnect
import DataConnectGenerated

@main
struct QuickstartApp: App {
  init() {
    FirebaseApp.configure()
  }
  var body: some Scene {
    WindowGroup {
      ListMovieView()
    }
  }
}

struct ListMovieView: View {
  @State private var moviesData: ListMoviesQuery.Data?

  var body: some View {
    VStack {
      ForEach(moviesData?.movies ?? []) { movie in
        Text(movie.title)
      }
    }
    .task {
      let result = try? await DataConnect.exampleConnector
          .listMoviesQuery.execute()
      self.moviesData = result?.data
    }
  }
}

9. Try the app

From Xcode, run the example app to see it in action.

Next steps

Try the SQL Connect VS Code extension

When developing with SQL Connect, we strongly recommend using the SQL Connect VS Code extension. Even if you don't use Visual Studio Code as your primary development environment, the extension provides several features that make schema and operation development more convenient:

  • A GraphQL language server, providing syntax checking and autocomplete suggestions specific to SQL Connect
  • CodeLens buttons in line with your code that let you read and write data from your schema definition files and execute queries and mutations from your operation definitions.
  • Automatically keep your generated SDKs synchronized with your GraphQL definitions.
  • Simplified local emulator setup.
  • Simplified deployment to production.

Use the SQL Connect emulator for local development

Although this tutorial showed you how to deploy SQL Connect schemas and operations directly to production, you will likely not want to make changes to your production database while you are actively developing your app. Instead, set up the SQL Connect emulator and do your development work against it rather than production. The emulator sets up a local PGlite instance that behaves similarly to a live PostgreSQL instance on Cloud SQL.

Learn how to write schemas and operations for your app

When developing apps with SQL Connect, the design of your schemas and operations is one of the first and most important development tasks you will complete.

  • Gemini in the Firebase console is an AI tool that can generate SQL Connect schemas from a natural language description of your app. This tool can get you started very quickly, especially if you've never worked with relational databases before.
  • Alternatively, you can write database schemas, queries, and mutations directly using GraphQL. Start with the guidance in Design SQL Connect schemas, and then continue to the follow-up pages to learn how to write operations.

Learn how to get real-time updates from SQL Connect

You can use SQL Connect to write client applications that react to changing data in real time. See Get real-time updates from SQL Connect.