Data Quality
Schema Validator
Checks that the columns you expect are there, that their values can be read as the right type, and that columns that must be filled have no blanks — and stops the run (or warns) when they do not.
When to use it
Put it straight after an input whose source can change without warning — a supplier file, a shared sheet, an API — so a renamed column or a text value in a number column stops the flow before it writes bad data.
In a pipeline
File InputSchema ValidatorSQL Server Output
Settings
| Setting | Type | Required | Default or hint |
|---|---|---|---|
| Schema (JSON) | textarea | Yes | [{"column":"Amount","type":"decimal","nullable":false}] |
| On mismatch | select | — | fail |
Example
Amount must stay a filled-in decimal
- Schema (JSON)=[{"column":"Amount","type":"decimal","nullable":false}]
- On mismatch=Fail the run
| Order | Amount |
|---|---|
| A-1 | 120.50 |
| A-2 | N/A |
| A-3 | — |
The run stops with: The data does not match the schema: "Amount" should be decimal, but 1 of 3 value(s) are not, e.g. "N/A" (row 2). "Amount" must not be blank, but 1 row(s) are (first at row 3).
Worth knowing
- Types: text, integer, decimal (number), date, boolean and any. A type only checks filled-in cells; set "nullable": false to forbid blanks too.
- Columns that are not in the schema are allowed — list only the ones the rest of the flow depends on.
- On mismatch: Warn lists every problem in the node's note and passes the rows on unchanged.