Edit

Share via

Tutorial: Add a PostgreSQL database connection in Azure Static Web Apps (preview)

  • Article

In this tutorial, you learn how to connect an Azure Database for PostgreSQL Single Server or Flexible Server database to your static web app. Once configured, you can make REST or GraphQL requests to the built-in /data-api endpoint to manipulate data without having to write backend code.

For the sake of simplicity, this tutorial shows you how to use an Azure database for local development purposes, but you may also use a local database server for your local development needs.

Note

This tutorial shows how to use Azure Database for PostgreSQL Flexible Server or Single Server. If you would like to use another database, refer to the Azure Cosmos DB, Azure SQL, or MySQL tutorials.

Web browser showing results from PostgreSQL selection in the developer tools console window.

In this tutorial, you learn to:

  • Link an Azure Database for PostgreSQL Flexible Server or Single Server database to your static web app
  • Create, read, update, and delete data

Prerequisites

To complete this tutorial, you need to have an existing Azure Database for PostgreSQL Flexible Server or Single Server and static web app. Additionally, you need to install Azure Data Studio.

Resource Description
Azure Database for PostgreSQL Flexible Server or Azure Database for PostgreSQL Single Server Database If you don't already have one, follow the steps in the create an Azure Database for PostgreSQL Flexible Server database guide, or in the create an Azure Database for PostgeSQL Single Server database guide. If you plan to use a connection string authentication for Static Web Apps' database connections, ensure that you create your Azure Database for PostgreSQL Server with PostgreSQL authentication. You can change this value if you want to use managed identity later on.
Existing static web app If you don't already have one, follow the steps in the getting started guide to create a No Framework static web app.
Azure Data Studio, with the PostgreSQL extension If you don't already have Azure Data Studio installed, follow the guide to install Azure Data Studio, with the PostgreSQL extension. Alternatively, you may use any other tool to query your PostgreSQL database, such as PgAdmin.

Begin by configuring your database to work with the Azure Static Web Apps database connection feature.

Configure database connectivity

Azure Static Web Apps must have network access to your database for database connections to work. Additionally, to use an Azure database for local development, you need to configure your database to allow requests from your own IP address.

  1. Go to your Azure Database for PostgreSQL Server in the Azure portal.

  2. If you're using Azure Database for PostgreSQL Flexible Server, under the Settings section, select Networking. If you're using Azure Database for PostgreSQL Single Server, under the Settings section, select Connection security.

  3. Under the Firewall rules section, select the Add your current client IP address button. This step ensures that you can use this database for your local development.

  4. Under the Firewall rules section, select the Allow public access from any Azure service within Azure to this server checkbox. If you're using Azure Database for PostgreSQL Single Server, this toggle is labeled Allow access to Azure services. This step ensures that your deployed Static Web Apps resource can access your database.

  5. Select Save.

Get database connection string for local development

To use your Azure database for local development, you need to retrieve the connection string of your database. You may skip this step if you plan to use a local database for development purposes.

  1. Go to your Azure Database for PostgreSQL Server in the Azure portal.

  2. Under the Settings section, select Connection strings.

  3. From the ADO.NET box, copy the connection string and set it aside in a text editor.

  4. Replace the {your_password} placeholder in the connection string with your password.

  5. Replace the {your_database} placeholder with the database name MyTestPersonDatabase. You'll create the MyTestPersonDatabase in the coming steps.

  6. Append Trust Server Certificate=True; to the connection string to use this connection string for local development.

Create sample data

Create a sample table and seed it with sample data to match the tutorial. This tutorial uses Azure Data Studio, but you may use PgAdmin or any other tool.

  1. In Azure Data Studio, create a connection to your Azure Database for PostgreSQL Server

  2. Right-click your server, and select New Query. Run the following querying to create a database named MyTestPersonDatabase.

    CREATE DATABASE "MyTestPersonDatabase";

  3. Right-click your server, and select Refresh.

  4. Right-click your MyTestPersonDatabase and select New Query. Run the following query to create a new table named MyTestPersonTable.

    CREATE TABLE "MyTestPersonTable" (
    "Id" SERIAL PRIMARY KEY,
    "Name" VARCHAR(25) NULL
);

  5. Run the following queries to add data into the MyTestPersonTable table.

    INSERT INTO "MyTestPersonTable" ("Name")
VALUES ('Sunny');

INSERT INTO "MyTestPersonTable" ("Name")
VALUES ('Dheeraj');

Configure the static web app

The rest this tutorial focuses on editing your static web app's source code to make use of database connections locally.

Important

The following steps assume you are working with the static web app created in the getting started guide. If you are using a different project, make sure to adjust the following git commands to match your branch names.

  1. Switch to the main branch.

    git checkout main

  2. Synchronize your local version with what's on GitHub by using git pull.

    git pull origin main

Create the database configuration file

Next, create the configuration file that your static web app uses to interface with the database.

  1. Open your terminal and create a new variable to hold your connection string. The specific syntax may vary depending on the shell type you're using.

    export DATABASE_CONNECTION_STRING='<YOUR_CONNECTION_STRING>'

    Make sure to replace <YOUR_CONNECTION_STRING> with the connections string value you set aside in a text editor.

  2. Use npm to install or update the Static Web Apps CLI. Select which command is best for your situation.

    To install, use npm install.

    npm install -g @azure/static-web-apps-cli

    npm install -g @azure/static-web-apps-cli

    To update, use npm update.

    npm update

    npm update

  3. Use the swa db init command to generate a database configuration file.

    swa db init --database-type postgresql

    The init command creates the staticwebapp.database.config.json file in the swa-db-connections folder.

  4. Paste in this sample into file staticwebapp.database.config.json you generated.

{
  "$schema": "https://github.com/Azure/data-api-builder/releases/latest/download/dab.draft.schema.json",
  "data-source": {
    "database-type": "postgresql",
    "options": {
      "set-session-context": false 
    },
    "connection-string": "@env('DATABASE_CONNECTION_STRING')"
  },
  "runtime": {
    "rest": {
      "enabled": true,
      "path": "/rest"
    },
    "graphql": {
      "allow-introspection": true,
      "enabled": true,
      "path": "/graphql"
    },
    "host": {
      "mode": "production",
      "cors": {
        "origins": ["http://localhost:4280"],
        "allow-credentials": false
      },
      "authentication": {
        "provider": "StaticWebApps"
      }
    }
  },
  "entities": {
    "Person": {
      "source": "MyTestPersonTable",
      "permissions": [
        {
          "actions": ["*"],
          "role": "anonymous"
        }
      ]
    }
  }
}

Before moving on to the next step, review the following table that explains different aspects of the configuration file. For full documentation on the configuration file and functionality such as relationships and policies for item-level security, refer to Data API Builder documentation.

Feature Explanation
Database connection In development, the runtime reads the connection string from the value of the connection string in the configuration file. While you can specify your connection string directly in the configuration file, a best practice is to store connection strings in a local environment variable. You can refer to environment variable values in the configuration file via the @env('DATABASE_CONNECTION_STRING') notation. The value of the connection string gets overwritten by Static Web Apps for the deployed site with the information collected when you connect your database.
API endpoint The REST endpoint is available via /data-api/rest while the GraphQL endpoint is available through /data-api/graphql as configured in this configuration file. You may configure the REST and GraphQL paths, but the /data-api prefix isn't configurable.
API Security The runtime.host.cors settings allow you to define allowed origins that can make requests to the API. In this case, the configuration reflects a development environment and allowlists the http://localhost:4280 location.
Entity model Defines the entities exposed via routes in the REST API, or as types in the GraphQL schema. In this case, the name Person, is the name exposed to the endpoint while entities.<NAME>.source is the database schema and table mapping. Notice how the API endpoint name doesn't need to be identical to the table name.
Entity security Permissions rules listed in the entity.<NAME>.permissions array control the authorization settings for an entity. You can secure an entity with roles in the same way you secure routes with roles.

Note

The configuration file's connection-string, host.mode, and graphql.allow-introspection properties are overwritten when you deploy your site. Your connection string is overwritten with the authentication details collected when you connect your database to your Static Web Apps resource. The host.mode property is set to production, and the graphql.allow-introspection is set to false. These overrides provide consistency in your configuration files across your development and production workloads, while ensuring your Static Web Apps resource with database connections enabled is secure and production-ready.

With the static web app configured to connect to the database, you can now verify the connection.

Update home page

Replace the markup between the body tags in the index.html file with the following HTML.

<h1>Static Web Apps Database Connections</h1>
<blockquote>
    Open the console in the browser developer tools to see the API responses.
</blockquote>
<div>
    <button id="list" onclick="list()">List</button>
    <button id="get" onclick="get()">Get</button>
    <button id="update" onclick="update()">Update</button>
    <button id="create" onclick="create()">Create</button>
    <button id="delete" onclick="del()">Delete</button>
</div>
<script>
    // add JavaScript here
</script>

Start the application locally

Now you can run your website and manipulate data in the database directly.

  1. Start the static web app with the database configuration.

    swa start ./src --data-api-location swa-db-connections

Now that the CLI is started, you can access your database via the endpoints as defined in the staticwebapp.database.config.json file.

The http://localhost:4280/data-api/rest/<ENTITY_NAME> endpoint accepts GET, PUT, POST and DELETE requests to manipulate data in the database.

The http://localhost:4280/data-api/graphql endpoint accepts GraphQL queries and mutations.

Manipulate data

The following framework-agnostic commands demonstrate how to do full CRUD operations on your database.

The output for each function appears in the browser's console window.

Open the developer tools by pressing CMD/CTRL + SHIFT + I and select the Console tab.

List all items

Add the following code between the script tags in index.html.

async function list() {
  const endpoint = '/data-api/rest/Person';
  const response = await fetch(endpoint);
  const data = await response.json();
  console.table(data.value);
}

In this example:

  • The default request for the fetch API uses the verb GET.
  • Data in the response payload is found in the value property.
async function list() {

  const query = `
      {
        people {
          items {
            Id
            Name
          }
        }
      }`;

  const endpoint = "/data-api/graphql";
  const response = await fetch(endpoint, {
      method: "POST",
      headers: { "Content-Type": "application/json" },
      body: JSON.stringify({ query: query })
  });
  const result = await response.json();
  console.table(result.data.people.items);
}

In this example:

  • The GraphQL query selects the Id and Name fields from the database.
  • The request passed to the server requires a payload where the query property holds the query definition.
  • Data in the response payload is found in the data.people.items property.

Refresh the page and select the List button.

The browser's console window now displays a table that lists all the records in the database.

ID Name
1 Sunny
2 Dheeraj

Here's a screenshot of what it should look like in your browser.

Web browser showing results from a database selection in the developer tools console window.

Get by ID

Add the following code between the script tags in index.html.

async function get() {
  const id = 1;
  const endpoint = `/data-api/rest/Person/Id`;
  const response = await fetch(`${endpoint}/${id}`);
  const result = await response.json();
  console.table(result.value);
}

In this example:

  • The endpoint is suffixed with /person/Id.
  • The ID value is appended to the end of the endpoint location.
  • Data in the response payload is found in the value property.
async function get() {

  const id = 1;

  const gql = `
    query getById($id: Int!) {
      person_by_pk(Id: $id) {
        Id
        Name
      }
    }`;

  const query = {
    query: gql,
    variables: {
      id: id,
    },
  };

  const endpoint = "/data-api/graphql";
  const response = await fetch(endpoint, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(query),
  });
  const result = await response.json();
  console.table(result.data.person_by_pk);
}

In this example:

  • The GraphQL query selects the Id and Name fields from the database.
  • The request passed to the server requires a payload where the query property holds the query definition.
  • Data in the response payload is found in the data.person_by_pk property.

Refresh the page and select the Get button.

The browser's console window now displays a table listing the single record requested from the database.

ID Name
1 Sunny

Update

Add the following code between the script tags in index.html.

Static Web Apps support both the PUT and PATCH verbs. A PUT request updates the whole record, while PATCH does a partial update.

async function update() {

  const id = 1;
  const data = {
    Name: "Molly"
  };

  const endpoint = '/data-api/rest/Person/Id';
  const response = await fetch(`${endpoint}/${id}`, {
    method: "PUT",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(data)
  });
  const result = await response.json();
  console.table(result.value);
}

In this example:

  • The endpoint is suffixed with /person/Id/.
  • The ID value is appended to the end of the endpoint location.
  • The REST verb is PUT to update the database record.
  • Data in the response payload is found in the value property.
async function update() {

  const id = 1;
  const data = {
    Name: "Molly"
  };

  const gql = `
    mutation update($id: Int!, $item: UpdatePersonInput!) {
      updatePerson(Id: $id, item: $item) {
        Id
        Name
      }
    }`;

  const query = {
    query: gql,
    variables: {
      id: id,
      item: data
    } 
  };

  const endpoint = "/data-api/graphql";
  const res = await fetch(endpoint, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(query)
  });

  const result = await res.json();
  console.table(result.data.updatePerson);
}

In this example:

  • The GraphQL query selects the Id and Name fields from the database.
  • The query object holds the GraphQL query in the query property.
  • The argument values to the GraphQL function are passed in via the query.variables property.
  • The request passed to the server requires a payload where the query property holds the query definition.
  • Data in the response payload is found in the data.updatePerson property.

Refresh the page and select the Update button.

The browser's console window now displays a table showing the updated data.

ID Name
1 Molly

Create

Add the following code between the script tags in index.html.

async function create() {

  const data = {
    Name: "Pedro"
  };

  const endpoint = `/data-api/rest/Person/`;
  const response = await fetch(endpoint, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(data)
  });
  const result = await response.json();
  console.table(result.value);
}

In this example:

  • The endpoint is suffixed with /person/.
  • The REST verb is POST to add a database record.
  • Data in the response payload is found in the value property.
async function create() {

  const data = {
    Name: "Pedro"
  };

  const gql = `
    mutation create($item: CreatePersonInput!) {
      createPerson(item: $item) {
        Id
        Name
      }
    }`;

  const query = {
    query: gql,
    variables: {
      item: data
    } 
  };

  const endpoint = "/data-api/graphql";
  const result = await fetch(endpoint, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(query)
  });

  const response = await result.json();
  console.table(response.data.createPerson);
}

In this example:

  • The GraphQL query selects the Id and Name fields from the database.
  • The query object holds the GraphQL query in the query property.
  • The argument values to the GraphQL function are passed in via the query.variables property.
  • The request passed to the server requires a payload where the query property holds the query definition.
  • Data in the response payload is found in the data.updatePerson property.

Refresh the page and select the Create button.

The browser's console window now displays a table showing the new record in the database.

ID Name
3 Pedro

Delete

Add the following code between the script tags in index.html.

async function del() {
  const id = 3;
  const endpoint = '/data-api/rest/Person/Id';
  const response = await fetch(`${endpoint}/${id}`, {
    method: "DELETE"
  });
  if(response.ok) {
    console.log(`Record deleted: ${ id }`)
  } else {
    console.log(response);
  }
}

In this example:

  • The endpoint is suffixed with /person/Id/.
  • The ID value is appended to the end of the endpoint location.
  • The REST verb is DELETE to remove the database record.
  • If the delete is successful the response payload ok property is true.
async function del() {

  const id = 3;

  const gql = `
    mutation del($id: Int!) {
      deletePerson(Id: $id) {
        Id
      }
    }`;

  const query = {
    query: gql,
    variables: {
      id: id
    }
  };

  const endpoint = "/data-api/graphql";
  const response = await fetch(endpoint, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(query)
  });

  const result = await response.json();
  console.log(`Record deleted: ${ result.data.deletePerson.Id }`);
}

In this example:

  • The GraphQL query selects the Id field from the database.
  • The query object holds the GraphQL query in the query property.
  • The argument values to the GraphQL function are passed in via the query.variables property.
  • The request passed to the server requires a payload where the query property holds the query definition.
  • Data in the response payload is found in the data.deletePerson property.

Refresh the page and select the Delete button.

The browser's console window now displays a table showing the response from the delete request.

Record deleted: 3

Now that you've worked with your site locally, you can now deploy it to Azure.

Deploy your site

To deploy this site to production, you just need to commit the configuration file and push your changes to the server.

  1. Add the file changes to track.

    git add .

  2. Commit the configuration changes.

    git commit -am "Add database configuration"

  3. Push your changes to the server.

    git push origin main

Connect the database to your static web app

Use the following steps to create a connection between the Static Web Apps instance of your site and your database.

  1. Open your static web app in the Azure portal.

  2. In the Settings section, select Database connection.

  3. Under the Production section, select the Link existing database link.

  4. In the Link existing database window, enter the following values:

    Property Value
    Database Type Select your database type from the dropdown list.
    Subscription Select your Azure subscription from the dropdown list.
    Resource Name Select the database server name that has your desired database.
    Database Name Select the name of the database you want to link to your static web app.
    Authentication Type Select Connection string, and enter the PostgreSQL user name and password. For PostgreSQL Single Server, don't include the @servername suffix.

  5. Select OK.

Verify that your database is connected to your Static Web Apps resource

Once you've connected your database to your static web app and the site is finished building, use the following steps to verify the database connection.

  1. Open your static web app in the Azure portal.

  2. In the Essentials section, select the URL of your Static Web Apps resource to navigate to your static web app.

  3. Select the List button to list all items.

    The output should resemble what's shown in this screenshot.

    Web browser showing results from listing records from the database in the developer tools console window.

Clean up resources

If you want to remove the resources created during this tutorial, you need to unlink the database and remove the sample data.

  1. Unlink database: Open your static web app in the Azure portal. Under the Settings section, select Database connection. Next to the linked database, select View details. In the Database connection details window, select the Unlink button.

  2. Remove sample data: In your database, delete the table named MyTestPersonTable.

Next steps

Add an API