Skip to content


Write your own complete SQL SELECT statement.

This component can run queries on any tables in the project. You may link one or more inputs into the SQL component to indicate which data sources are used within the query. You can also use the SQL component as a source (for example, the first component in the transformation pipeline workflow), and don't strictly need any input links from other components.

To query a table from another component in the job, your SELECT statement must use an expression that references the component by name. This expression has the form $T{<component name>}. For example, to select all from a table you have read with a Table Input component called "Table Input 1", use the following statement:

SELECT * FROM $T{Table Input 1}

If you require non-SELECT statements, the SQL Script orchestration component allows for custom non-SELECT statements where the output can be used as part of the orchestration flow.


This component supports the use of pipeline and project variables. For more information, read Variables.


Name = string

A human-readable name for the component.

Query = code editor

The SQL query to execute.

Snowflake Databricks Amazon Redshift (preview)