Input
Databricks Input
Reads a whole table or view from a Databricks SQL warehouse (Unity Catalog or hive_metastore) and turns it into rows.
When to use it
Use it when curated data lives in the lakehouse — a gold table, a feature table, or a view your data engineers publish.
In a pipeline
Databricks InputFilter RowsExport Excel
Settings
| Setting | Type | Required | Default or hint |
|---|---|---|---|
| Connection | select | Yes | Add a connection below |
| Table | select | Yes | Select a table |
Example
Weekly sales from the gold layer
- Connection=adb-1234567890123456.7.azuredatabricks.net
- Table=gold.weekly_sales
| week | region | revenue |
|---|---|---|
| 2026-W42 | EMEA | 182400 |
| 2026-W42 | APAC | 96120 |
Worth knowing
- The connection is a SQL warehouse: its Server hostname and HTTP path (Connection details tab) and an access token. Catalog picks the catalog; tables are listed as schema.table.
- The warehouse starts by itself if it is stopped; the first read can take a minute while it does.
- Runs through the server's Python bridge (databricks-sql-connector); your administrator installs it once.