nucleo000/planilla
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Refactor: Centralize Prisma schema and restrict DB access

Browse Source 
This commit refactors the project to use a shared Prisma schema and restricts direct database access to the API service.

Key changes:

- Created a new shared package `core/prisma` containing the Prisma schema, generated client, and types.
- Configured the monorepo to use NPM workspaces, including `core/prisma` and all services.
- Updated all services (`api`, `ui`, `mcp`, `agent`, and the background processing service) to depend on `@empresa/prisma-schema`.
- The API service now imports `PrismaClient` from `@empresa/prisma-schema/client`.
- Other services import only types from `@empresa/prisma-schema`.
- Removed redundant Prisma configurations from `api` and the background processing service.
- Updated the background processing service's `sync-empleados.js` to fetch data via an API call instead of direct database access.
- Updated TypeScript configurations (`tsconfig.base.json` and service-specific ones) to support the new structure and path aliases.
- Updated `README.md` to reflect the new architecture and added convenience scripts for Prisma operations.

This change promotes a single source of truth for data models, reduces code duplication, and improves the overall architecture by centralizing database operations within the API service.
This commit is contained in:
google-labs-jules[bot]
2025-05-30 23:40:00 +00:00
parent 4f1ec58a99
commit 242dc66983
26 changed files with 381 additions and 243 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
worker/jsconfig.json Normal file
View File

@@ -0,0 +1,14 @@
{
"extends": "../tsconfig.base.json",
"compilerOptions": {
"checkJs": false,
"resolveJsonModule": true,
"moduleResolution": "node",
"target": "esnext", // Or appropriate target for your Node.js version
"module": "esnext", // Since package.json has "type": "module"
// Paths are inherited from tsconfig.base.json
"baseUrl": "." // baseUrl is still needed if there are other local paths
},
"include": ["**/*.js"],
"exclude": ["node_modules", "prisma"] // Excluding worker/prisma as it's not used anymore
}

5
worker/package.json
View File

@@ -7,10 +7,9 @@
"start": "node server.js"
},
"dependencies": {
"@prisma/client": "^6.7.0",
"@empresa/prisma-schema": "1.0.0",
"express": "^4.18.2",
"node-cron": "^4.0.5",
"pg": "^8.8.0",
"prisma": "^6.7.0"
"pg": "^8.8.0"
}
}

15
worker/prisma/schema.prisma
View File

@@ -1,15 +0,0 @@
// This is your Prisma schema file,
// learn more about it in the docs: https://pris.ly/d/prisma-schema
// Looking for ways to speed up your queries, or scale easily with your serverless or edge functions?
// Try Prisma Accelerate: https://pris.ly/cli/accelerate-init
generator client {
provider = "prisma-client-js"
output = "../generated/prisma"
}
datasource db {
provider = "postgresql"
url = env("DATABASE_URL")
}

54
worker/sync-empleados.js
View File

@@ -17,33 +17,31 @@ const EXTERNAL_DB_NAME = process.env.EXTERNAL_DB_NAME || 'YOUR_EXTERNAL_DB_NAME'
const EXTERNAL_DB_EMPLEADOS_TABLE = process.env.EXTERNAL_DB_EMPLEADOS_TABLE || 'empleados_externos'; // User must set this.
// --- End of External Database Configuration ---
import { PrismaClient } from '../api/prisma/generated/client';
const prisma = new PrismaClient();
// Import types from the shared Prisma package.
// We are not importing PrismaClient here as the worker will fetch data from the API.
import type { Cliente } from '@empresa/prisma-schema'; // Assuming Cliente is the relevant type for employees
// Define a type for the employee data we expect from the API.
// This might be identical to Cliente or a subset, depending on the API endpoint.
type EmployeeDataFromAPI = Pick<Cliente, 'id' | 'name' | 'cedula' | 'ubicacion' | 'telefono'>;
async function syncEmpleadosToExternalDB() {
console.log('[SyncEmpleados] Starting synchronization process...');
// Core logic for synchronization, wrapped in try/catch/finally to ensure
// resources like the Prisma client are properly managed (e.g., disconnected).
try {
// Fetch all 'Cliente' records from the local Prisma database that are marked as 'empleado'.
const localEmpleados = await prisma.cliente.findMany({
where: { empleado: true }, // Filters for clients who are also employees
select: {
id: true, // Local ID, might be useful for logging/tracing
name: true,
cedula: true,
ubicacion: true,
telefono: true,
// avatar_url: true, // Add other relevant fields
// idciat: true, // Add other relevant fields
},
});
// TODO: Fetch employee data from the API (e.g., GET /api/empleados)
// This is a placeholder for the API call logic.
console.log('[SyncEmpleados] Fetching employee data from API...');
const response = await fetch('http://localhost:4000/api/empleados'); // Replace with your actual API endpoint
if (!response.ok) {
throw new Error(`API request failed with status ${response.status}`);
}
const localEmpleados: EmployeeDataFromAPI[] = await response.json();
if (!localEmpleados.length) {
console.log('[SyncEmpleados] No employees found in local database. Nothing to sync.');
if (!localEmpleados || !localEmpleados.length) {
console.log('[SyncEmpleados] No employees found from API. Nothing to sync.');
return;
}
console.log(`[SyncEmpleados] Found ${localEmpleados.length} employees to sync.`);
console.log(`[SyncEmpleados] Found ${localEmpleados.length} employees from API to sync.`);
// --- External DB Connection Logic (User to implement based on EXTERNAL_DB_TYPE) ---
// User must implement the actual database connection logic here based on EXTERNAL_DB_TYPE
@@ -119,9 +117,7 @@ async function syncEmpleadosToExternalDB() {
// Catch any errors that occur during the synchronization process.
console.error('[SyncEmpleados] Error during synchronization:', error);
} finally {
// Ensure the Prisma client is always disconnected, even if an error occurs.
await prisma.$disconnect();
console.log('[SyncEmpleados] Prisma client disconnected.');
// No Prisma client to disconnect as we are using API.
}
console.log('[SyncEmpleados] Synchronization process finished.');
}
@@ -136,19 +132,17 @@ export { syncEmpleadosToExternalDB };
// 2. **Manual Testing:**
// * Once configured, you can test the script by running it directly: `node worker/sync-empleados.js`
// * Observe the console logs for any errors or successful completion messages.
// * Verify that data from your local 'Cliente' table (where empleado=true) appears correctly
// in your designated external database table (`EXTERNAL_DB_EMPLEADOS_TABLE`).
// * Check that subsequent runs correctly update existing records and insert new ones.
// * The cron job in `worker/cron-worker.js` is scheduled to run this script daily at midnight.
// * Verify that data fetched from the API appears correctly (or is logged as intended for now)
// if you have implemented the external DB sync part.
// * The cron job in `worker/cron-worker.js` is scheduled to run this script.
// You can monitor logs after this time to see its execution.
//
// 3. **Automated Testing (Recommendations for future development):**
// * **Unit Tests:**
// - Mock the Prisma client (`../api/prisma/generated/client`) to return predefined employee data
// and spy on its methods.
// - Mock the API call (`fetch`) to return predefined employee data.
// - Mock the external database client (e.g., `pg`, `mysql2`) to simulate connection,
// query execution (select, insert, update), and disconnections. This allows testing
// the sync logic without actual database dependencies.
// the sync logic without actual API or database dependencies.
// * **Integration Tests:**
// - If possible, set up a dedicated test instance of your external database.
// - Write tests that populate the local Prisma test database with sample employee data,