DataTongs — Reference Data Sync Scripts

In schema-as-code workflows, SchemaTongs and SchemaQuench handle structure: extraction and deployment of tables, views, procedures, and functions. But database schemas don't live in isolation — they travel with reference data. Lookup tables, configuration rows, seed data, and status enumerations need the same version control and deployment lifecycle as the schema itself. DataTongs fills that gap.

DataTongs connects to a live database, reads your configured reference tables, and writes one synchronization script per table to your output path. On SQL Server and PostgreSQL, these scripts use MERGE statements — conditional inserts and updates in a single atomic operation. On MySQL, DataTongs emits INSERT ... ON DUPLICATE KEY UPDATE scripts, MySQL's native syntax for idempotent upserts. The output is self-contained: drop a script into a schema package's Table Data folder and SchemaQuench will execute it during deployment, or run it directly against any compatible target to synchronize reference data in isolation.

One DataTongs executable runs on three platforms: SQL Server, PostgreSQL, and MySQL. The tool works identically on all three; what changes is the adapter underneath. Set your Source:Platform value to SqlServer, PostgreSQL, or MySQL and DataTongs picks the right connection protocol, platform metadata APIs, and sync-script dialect.

DataTongs also configures how extracted data flows into your deployment pipeline. Pass the --ConfigureDataDelivery flag and DataTongs writes a DataDelivery block into each table's JSON schema file alongside the extracted .tabledata content file. The DataDelivery block captures delivery settings — which columns to match on, merge type, any filters, and platform-specific trigger or rule handling — so SchemaQuench knows exactly how to synchronize that data during deployment. This turns the workflow into a single integrated pipeline: SchemaTongs extracts your schema structure, DataTongs extracts your reference data and wires it into the delivery configuration in one pass, and SchemaQuench deploys both together. See SchemaQuench for how the deployment side reads and executes these configured data deliveries.