Your progress is only saved in this browser. to protect it and pick up on any device.

Connect a Postman API key to check your work

You do each step in Postman, then LiftOff uses your key to look at your workspace and confirm it's done. Connect once and it works for every module. Your key is saved in this browser and is only used to check your work in Postman.

0 of 2 steps0%

A view is a saved SQL query, and it's the heart of a dataset. Views use SQLite-compatible SQL for files and standard databases, and because the view stays fixed even if the files behind it change, everything built on top of it keeps working.

4

Filter and Compute With a View

Your first views answer two simple questions: which accounts are on the enterprise plan, and what does each invoice look like in one line?

  1. If billing-data is still open in the active tab from the previous step, stay there. Otherwise open it from the sidebar. Then click Create view.

  2. In the Create view dialog, enter the View name:

    text
    enterprise_accounts
  3. Open the Sources dropdown, select source_accounts, then click Create view.

  4. The view opens with its data table at the top and a Query editor underneath, pre-filled with SELECT * FROM "source_accounts". The new view also appears under Datasets in the left sidebar.

  5. Replace the query in the editor with this one, click Run to see the result, then click Save:

    view · enterprise_accounts
    SELECT accountId, company, region
    FROM source_accounts
    WHERE plan = 'enterprise';
  6. Create a second view the same way. Click Create view again and enter the View name:

    text
    invoice_summary
  7. Select source_invoices in Sources and click Create view.

  8. Replace the query in the editor with this one, then Run and Save. It builds a computed column with ||, SQLite's string concatenation:

    view · invoice_summary
    SELECT invoiceId, product || ' (' || status || ')' AS summary, amount
    FROM source_invoices;
  9. Check that each view's data table matches the table under its query above.

Tip: The query editor autocompletes keywords, clauses and source names. Press ⌘+I on Mac or Ctrl+Space on Windows and Linux. To tweak a view without losing the original, use the menu next to Save to save your edit as a new view instead of saving over it.

Troubleshooting: If you see no such table, the query is using a source name your dataset doesn't have. The tables are source_accounts and source_invoices. If you renamed a source, open its auto-generated view and copy the name from its query.

5

Join the Two Sources

This is where a dataset earns its keep. Accounts and invoices came from different file types, yet one query can combine them into the table you actually want to test with.

  1. Click Create view and enter the View name:

    text
    account_invoices
  2. Open the Sources dropdown and select both source_accounts and source_invoices, then click Create view. A view that joins two sources needs both selected.

  3. Replace the query in the editor with this one, click Run, then Save. It joins the sources on accountId and keeps only the invoices that were paid:

    view · account_invoices
    SELECT i.invoiceId, a.company, a.plan, i.product, i.amount
    FROM source_accounts AS a
    JOIN source_invoices AS i ON a.accountId = i.accountId
    WHERE i.status = 'paid';
  4. Check that the data table shows four rows, one per paid invoice, each with the company name and plan.

A view can combine file sources and standard databases (MySQL, Postgres, SQL Server) freely, but a custom JDBC source can't be combined with other source types: its views use that database's own SQL dialect and must work against that source alone.

Troubleshooting: If you get no rows, check that both sources store accountId the same way. In invoices.json the ids are strings ("1"), matching the CSV.