Skip to main content
The Translation agent converts SQL from a source database dialect to a target dialect. You organise translations into projects, each scoped to a source and target data source. The agent translates each query, seeds data into the target, and runs a comparison to verify the result is correct.

​
What it does

  • Translates SQL queries, views, and stored procedures from source to target dialect
  • Verifies translations by running the translated SQL against the target database and comparing results
  • Automatically detects and fixes translation errors in a feedback loop
  • Generates PySpark scripts as an alternative output for stored procedures

​
Create a translation project

A translation project groups related translations under a single source and target data source pair.
  1. Navigate to Translation in the left sidebar.
  2. Click New Project.
  3. Enter a Project Name.
  4. Select the Source Datasource — the database you are migrating from.
  5. Select the Target Datasource — the database you are migrating to.
  6. Click Create.

​
Add a translation

Once a project is created, add individual SQL objects to translate.
  1. Open the translation project.
  2. Click New Translation.
  3. Select the Translation Type — SQL Query, View, or Stored Procedure.
  4. Paste or enter the source SQL.
  5. For stored procedures, choose the Output Format:
    • Target SQL Script — Generates the equivalent stored procedure in the target dialect.
    • PySpark Script — Generates a PySpark equivalent for platforms such as Databricks.
  6. Click Translate. The agent runs the translation and verification workflow.

​
Supported translation types

TypeDescription
SQL QueryTranslates a SQL SELECT or DML statement to the target dialect.
ViewTranslates a view definition to the target dialect.
Stored ProcedureTranslates a stored procedure. Choose between a target SQL script or a PySpark script as the output format.