Skip to main content

Add tables from a source

Goal​

You'll see every table and view in a connected data source, pick the ones you care about, and add them as data assets in one click. Adding tables is part of managing a source. It happens in the source's Manage source window, not on a separate screen.

Listing tables works for Postgres and the warehouse connectors: Snowflake, BigQuery, Redshift, Databricks SQL, Microsoft SQL Server, Trino / Presto, ClickHouse, SAP HANA and Oracle. It also lists objects under an object-storage prefix. See Connect object storage.

Prereqs​

  • A connected direct-mode data source. Agent-mode sources add their tables on-premises through the agent, so there's nothing to tick in the cloud.
  • The editor role or higher in the workspace. Listing a source's tables and adding them both need the permission to create data assets.

Steps​

  1. Open the source. On Data Assets, click the source's name in its group header, or its Manage source button. The Manage source window opens. If you're adding a brand-new source, click Add source in the page header instead: after step 1 (Connect the source) succeeds, step 2 (Add tables) opens in the same window with the same table list.
  2. Find the tables. The Tables section lists every table and view the source can see, with its schema, kind and an estimated row count. Tables you've already added read Added; click one's row to open its data asset. Sort or filter the list from its column headers: the funnel on Schema or Name narrows it to names containing some text, and Kind filters by table or view.
  3. Tick the tables and views you want Measured Cloud to manage.
  4. Click "Add N tables" in the toolbar at the top of the window. The button counts what you ticked. Every ticked table is added as a data asset in one atomic call, so the whole batch lands together or not at all, and a transient failure can't leave you half-registered. A table whose asset already exists is skipped (it's matched on source + name), so running it again over the same tables is safe. A toast confirms the outcome.

Verify​

The new assets appear under the source's group on the Data Assets page, and read Added in the Manage-source window's table list. Each one is ready for profiling and rules. An added table's row in the window also has a Profile control, so you can profile it without leaving.

Caveats​

  • System schemas are excluded: pg_catalog, information_schema and pg_toast on Postgres, and the equivalent schemas in other warehouses.
  • The table list is cached for 15 minutes per source. If you've just created a table and it isn't listed, click Refresh tables in the toolbar.
  • The row count shown is the warehouse's statistics estimate, not a fresh COUNT(*). Use it to decide which tables are worth validating, not as an exact figure.
  • If a table's name matches a data asset in the recycle bin, it's skipped rather than duplicated. The window shows an inline Restore for it.
  • If the source's type can't list tables, or the database doesn't answer, the window says so and shows the tables already added instead. Check the connection with Test connection, then click Refresh tables.

Profiling after you add​

Adding tables does not profile them automatically. The daily profile sweep picks them up, or you can profile one straight away from its Profile control in the Manage-source window, or from the play button on its row on Data Assets.