Live Query Charts

This is the official documentation of the @forestadmin/agent Node.js agent.

Due to popular demand, Live Query Charts are planned to be reintroduced in a future version of the agent, but we don't have a timeline for it yet.

Live Query Charts ↗ allowed the creation of charts from SQL queries from the UI.

Because the new agent no longer exposes the nature of the underlying database, this feature was not ported into the new agent. Building a chart from SQL queries is only possible using API charts for now.

You can find the full documentation of chart customization here.

Steps

Step 1: Retrieve the SQL query from the UI

Live Query Chart configuration screen

You can retrieve the SQL query of a Live query chart by clicking on the Cog icon of the chart when using the Edit Layout mode.

Step 2: Create a new API chart

The next step will be to create a new API chart using the SQL query you retrieved in the previous step.

agent.addChart('appointments', async (context, resultBuilder) => {
  const rows = await context.dataSource.getCollection('appointments').nativeDriver
    .rawQuery(`
      SELECT current.count AS value, previous.count AS previous
      FROM (
        SELECT COUNT(*)
        FROM appointments
        WHERE start_date BETWEEN '2018-01-01' AND '2018-02-01'
      ) as current, (
        SELECT COUNT(*)
        FROM appointments
        WHERE start_date BETWEEN '2017-12-01' AND '2018-01-01'
      ) as previous;`);

  return resultBuilder.value(rows[0].value, rows[0].previous);
});

Step 3: Retrieve the URL of the generated chart

When you create a new API chart, you will need to retrieve the URL of the generated chart.

It is printed to the console when the agent starts (in development mode).

sandro@forestadmin $ yarn start
yarn run v1.22.19
info: Successfully mounted on Standalone server (http://0.0.0.0:3351)
info: Schema was updated, sending new version
info: Chart 'appointments' was mounted at '/forest/_charts/appointments'

Step 4: Change the old chart configuration

If you are following this guide step by step, you are probably testing your agent on a temporary project.

You may want to postpone this step until you have replaced your old agent with the new one.

Instead of Query, select Smart in the data source selector, and enter the path that was printed to the console in the previous step.