dyolink/backend

Dyolink — Backend (NestJS)

Prerequisites

• Node.js 20+ and npm
• PostgreSQL reachable from your machine — either installed locally or run via Docker (see below)

First-time setup

1. Clone the monorepo and go to the backend app:

   git clone <repository-url> dyolink
   cd dyolink/backend
   

2. Install dependencies

   npm install
   

3. Environment

   Copy .env.example to .env and set at least:

   • DATABASE_URL — PostgreSQL connection string for your dev database
   • JWT_SECRET — strong secret for signing tokens
     Do not commit .env.

4. Database for local dev

   Option A — Postgres in Docker (no local install, e.g. Mac)
   From backend/, with .env present (copy from .env.example first):

   • Ensure DATABASE_URL uses localhost as the host (not postgres). Match user, password, and DB name to POSTGRES_USER, POSTGRES_PASSWORD, and POSTGRES_DB in the same file.

   docker compose -f docker-compose.postgres.yml up -d
   

   Wait until Postgres is healthy (docker compose -f docker-compose.postgres.yml ps). The container creates the database on first start.

   To stop Postgres (data is kept in the named volume): docker compose -f docker-compose.postgres.yml down

   Option B — Postgres installed on the machine
   Create an empty database, then point DATABASE_URL at it.

5. Generate Prisma Client

   npm run prisma:generate
   

6. Apply migrations (creates/updates tables to match prisma/schema.prisma)

   npm run prisma:migrate
   

   This runs prisma migrate dev. Use it during development when the schema changes.

7. Seed (optional — reference data only)

   npm run prisma:seed
   

   This does not wipe your database. It only upserts lookup data: organization types (CLINIC, LAB), subscription plans, and tab permissions. Existing users, organizations, memberships, patients, appointments, and links are left unchanged.

   To start from an empty database with fresh tables and reference data, see Reset database (clean slate) below.

Reset database (clean slate)

Use this when you want to delete all application data (users, organizations, patients, sessions, etc.) and rebuild the schema from migrations, then run the seed.

From backend/:

npx prisma migrate reset

Prisma will prompt for confirmation, drop the database, re-apply all migrations, and run prisma/seed.ts automatically.

What gets removed: everything in the database, including organizations and all related rows.

What the seed adds back: only reference data (types, plans, permissions) — not demo users or organizations. Register again or use your own test data after a reset.

Docker Postgres dev: if you also want to wipe the Docker volume (not only tables), stop the container and remove the volume:

docker compose -f docker-compose.postgres.yml down -v
docker compose -f docker-compose.postgres.yml up -d
npm run prisma:migrate
npm run prisma:seed

Do not run migrate reset against production or shared staging databases.

Run (development)

npm run start:dev

API listens on http://localhost:3000 by default (PORT in .env).

If the frontend runs on another origin (e.g. http://localhost:3001), set FRONTEND_URL in .env to that URL (CORS and invite links use it).

After pulling latest main

git pull
npm install
npm run prisma:generate
npm run prisma:migrate

If teammates added migrations, the migrate step above applies them. Resolve migration conflicts locally before pushing.

Useful commands

Command	Purpose
npm run prisma:generate	Regenerate client after schema.prisma changes
npm run prisma:migrate	Dev migrations (migrate dev)
npm run prisma:deploy	Production-style apply (migrate deploy) — e.g. CI/containers
npm run prisma:seed	Upsert reference data only (does not clear existing rows)
npx prisma migrate reset	Drop DB, re-migrate, run seed — dev clean slate
npm run build	Compile Nest app
npm run start:prod	Run compiled app (node dist/main)

Docker

File	Purpose
Dockerfile	Production API image
docker-compose.postgres.yml	Local dev Postgres only (port mapped to host)

For full-stack deployment and CI, see the repository root README.md.