Skip to main content

DataQuery Module

The DataQuery Module (apps/backend/modules/dataQuery) allows users to write and execute queries (SQL, HTTP) against registered Datasources.

Architecture​

The core of execution is the QueryEngine.

Variable Interpolation​

Queries support dynamic arguments using {{variable}} syntax.

  1. Frontend sends argValues (e.g., { "limit": 10 }).
  2. Parser scans the query string for mapped arguments.
  3. Engine performs parameter substitution (using proper prepared statements for SQL to prevent Injection).

AI Generation​

The module includes an AI integration (generateAIPromptBasedQuery) that:

  1. Retrieves database schema.
  2. Sends schema + user prompt to LLM (Gemini).
  3. Returns a generated SQL query.

Query Engine (queryEngine/engine.js)​

A framework-agnostic runner that:

  1. Accepts a queryID and executionParams.
  2. Loads the correct driver based on datasourceType.
  3. Standardizes the output format.

Supported Drivers​

  • PostgreSQL: Uses pg pool.
  • REST API: Uses axios (implied).
  • RunJS: Executes sandboxed JavaScript.

Service API​

runDataQueryByID​

Executes a saved query.

await dataQueryService.runDataQueryByID({
  userID: 1,
  tenantID: 1,
  dataQueryID: 100,
  argValues: { "id": 5 }
});

runDataQueryByData​

Executes an ad-hoc ("dirty") query. Used during the editing/testing phase before saving.

Warning: This creates a temporary queryID internally to satisfy the engine interface.