Observable

Observable is a collaborative data notebook.

Here’s a short video guide on how to connect Observable to Cube.

Connect from Cube Cloud

Navigate to the Integrations page, click Connect to Cube, and choose Observable to get detailed instructions.

Connect from Cube Core

You can connect a Cube deployment to Observable using the SQL API or the REST (JSON) API.

In Cube Core, the SQL API is disabled by default. Enable it and configure the credentials to connect to Observable.

Connecting from Observable

Connecting via SQL API

Observable connects to Cube as to a Postgres database.

Querying data with SQL API

Your cubes will be exposed as tables, where both your measures and dimensions are columns.

Make sure to add a database to your notebook, and select Database query when adding a new block.

You can write SQL in Observable that will be executed in Cube. Learn more about Cube SQL syntax on the reference page.

You can also create a visualization of the executed SQL query.

Connecting via REST (JSON) API

For a Cube instance publicly available at a specific HOST, the REST (JSON) API URL would be HOST/cubejs-api/v1. Please refer to the REST (JSON) API page for details.

You will also need to generate a JSON Web Token that would be used to authenticate requests to Cube.

Please check the Security page to learn how to generate a token. We suggest generating a long-lived JWT that won’t expire soon.

Querying data with REST (JSON) API

First, add two generic JavaScript cells:

Next, copy Cube’s REST (JSON) API URL and the Authorization token and paste them into their respective cells.

cubeRestApi =
  "https://thirsty-raccoon.aws-eu-central-1.cubecloudapp.dev/cubejs-api/v1/load";

Because the Cube REST (JSON) API has the format of HOST/cubejs-api/v1, don’t forget to add the /load endpoint to the end of the data source API.

cubeRestApiJwtToken =
  "Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpYXQiOjE2NTgzMzM3OTZ9.gUOoDgo_RJka_ZANdwSw3v8GkM4ZzH9LjxrxKxkGAk0";

Also make sure to add the token next to the Bearer part of the Authorization header.

Get your Cube query in the JSON query format ready. You can copy it from Cube’s Playground or compose manually.

Paste the JSON query in another JavaScript cell as an object literal and give it a name, I chose jsonBody for simplicity. Make sure to add a query parameter for your JSON query.

jsonQuery = {
  query: {
    measures: ["orders.count"],
    timeDimensions: [
      {
        dimension: "orders.created_at",
        granularity: "month"
      }
    ],
    order: {
      "orders.created_at": "asc"
    }
  }
}

Next, create another JavaScript cell with a POST request. Paste this POST request in the cell. Don’t forget to put the jsonBody object inside the JSON.stringify call.

orders_over_time = fetch(cubeRestApi, {
  method: "POST",
  headers: {
    Authorization: cubeRestApiJwtToken,
    "Content-Type": "application/json",
  },
  body: JSON.stringify(jsonQuery),
})
  .then((response) => response.json())
  .then((json) => json.data);

Next, click the play button on the top right of the cell.

You can also create a visualization of the executed REST (JSON) API request.