Working with Turso

Turso is an edge-hosted, distributed database based on libSQL, an open-source and open-contribution fork of SQLite. It was designed to minimize query latency for applications where queries come from anywhere in the world.

Engineers can use Atlas to manage their Turso databases using the SQLite driver by using the libsql+ws:// (for local environments) or libsql+wss:// schemas in their connection URLs.

This guide will walk you through the process of setting up Turso and using Atlas to manage your Turso databases.

Set up Turso on your local machine

1. Start by installing Turso on your machine:

curl -sSfL https://get.tur.so/install.sh | bash

2. Verify that Turso is installed by running the following command:

turso --version

3. Sign up for a free Turso account using this command:

   turso auth signup
   The CLI launches your default browser and asks to log in with GitHub. The first time you log in, you are asked to grant the GitHub Turso app some permissions to your account. Accept this in order to continue.

Create your first Turso edge-database

1. Create a new Turso database using the following command:

   turso db create atlas

   The Turso system will provision a new edge database for you and print something similar to this:

   Created database atlas in Frankfurt, Germany (fra) in 9 seconds.

2. Next, run the following command to find the URL of your newly created database:

   turso db show atlas --url

   Make note of this address as we will use it in the next section.

3. Next, create an access token to use with Atlas:

   turso db tokens create atlas

   The CLI will print the access token to the console. Make note of this token
   as we will use it in the next section as well.

Create a configuration file for Atlas

Next, create an Atlas configuration file named atlas.hcl with the following contents:

variable "token" {
  type    = string
  default = getenv("TURSO_TOKEN")
}

env "turso" {
  url     = "libsql+wss://<REPLACE WITH YOUR SUBDOMAIN>.turso.io?authToken=${var.token}"
  exclude = ["_litestream*"]
}

Let's break down this configuration file:

1. We define an input variable named token which will be used to store the Turso access token. The variable is of the string type and takes the value of the TURSO_TOKEN environment variable by default.
2. We define an environment named turso which configures our interactions with our new Turso database:

• We use the libsql+wss:// schema to connect to the Turso database.
• The url parameter is set to the URL of the Turso database which we created earlier.
• The authToken parameter is set to the value of the token variable which we defined earlier.
• The exclude parameter is set to exclude the _litestream* tables which are created by the Litestream replication engine and are usually not relevant to the application.

Use Atlas to manage your Turso database

With this configuration file in place, we can now use Atlas to manage our Turso database.

Let's show some common operations we can perform with Atlas:

Declarative migrations (HCL)

We can use the atlas schema apply command to manage the schema of our Turso database.

Start by defining a schema file named schema.hcl with the following contents:

schema "main" {
}

table "users" {
  schema = schema.main
  column "id" {
    type = int
  }
  column "name" {
    type = varchar(255)
  }
  column "manager_id" {
    type = int
  }
  primary_key {
    columns = [
      column.id
    ]
  }
  index "idx_name" {
    columns = [
      column.name
    ]
    unique = true
  }
  foreign_key "manager_fk" {
    columns = [column.manager_id]
    ref_columns = [column.id]
  }
}

Next, set the TURSO_TOKEN environment variable to the value of the Turso access token which we created earlier:

export TURSO_TOKEN=<your jwt token>

Finally, run the following command to apply the schema to the Turso database:

atlas schema apply --env turso --to file://schema.hcl

Atlas will prompt you to approve the changes before applying them to the database:

-- Planned Changes:
-- Create "users" table
CREATE TABLE `users` (`id` int NOT NULL, `name` varchar NOT NULL, `manager_id` int NOT NULL, PRIMARY KEY (`id`), CONSTRAINT `manager_fk` FOREIGN KEY (`manager_id`) REFERENCES `users` (`id`) ON UPDATE NO ACTION ON DELETE CASCADE);
-- Create index "idx_name" to table: "users"
CREATE UNIQUE INDEX `idx_name` ON `users` (`name`);
Use the arrow keys to navigate: ↓ ↑ → ←
? Are you sure?:
  ▸ Apply
    Abort

After approving the changes, Atlas will apply the schema to the Turso database.

Inspection

We can use the atlas schema inspect command to inspect the schema of our Turso database. After applying the changes we mentioned above, we can run the following command to inspect the schema of our Turso database:

atlas schema inspect --env turso

Atlas will print the schema of the Turso database:

HCL Output

table "users" {
   schema = schema.main
   column "id" {
      null = false
      type = int
   }
   column "name" {
      null = false
      type = varchar
   }
   column "manager_id" {
      null = false
      type = int
   }
   primary_key {
      columns = [column.id]
   }
   foreign_key "manager_fk" {
      columns     = [column.manager_id]
      ref_columns = [table.users.column.id]
      on_update   = NO_ACTION
      on_delete   = CASCADE
   }
   index "idx_name" {
      unique  = true
      columns = [column.name]
   }
}
schema "main" {
}

You can also use the --format flag to change the output format. For example, you can use the --format flag to print the schema in SQL format:

atlas schema inspect --env turso --format '{{ sql . "  " }}'

SQL Output

Output:

-- Create "users" table
CREATE TABLE `users` (
  `id` int NOT NULL,
  `name` varchar NOT NULL,
  `manager_id` int NOT NULL,
  PRIMARY KEY (`id`),
  CONSTRAINT `manager_fk` FOREIGN KEY (`manager_id`) REFERENCES `users` (`id`) ON UPDATE NO ACTION ON DELETE CASCADE
);
-- Create index "idx_name" to table: "users"
CREATE UNIQUE INDEX `idx_name` ON `users` (`name`);