PotTrack

PotTrack is a hobby plant tracking application (per pot) from seeding to harvest.

This repository contains a Go/Gin backend and a SvelteKit + TailwindCSS frontend. The database is PostgreSQL using Indonesian table names and column names as specified in the PRD.

Repository Structure

backend/            # Go server, migrations, tests
frontend/           # SvelteKit application
PRD_POT_TRACK.txt   # Product requirements
README.md           # This file

Prerequisites

• Go 1.20+
• Node.js 18+
• Docker & docker-compose (for Postgres)

Database

The development database is pot_track_dev. A docker-compose file is provided below.

# docker-compose.yml
version: '3.8'
services:
  db:
    image: postgres:15
    environment:
      POSTGRES_USER: postgres
      POSTGRES_PASSWORD: password
      POSTGRES_DB: pot_track_dev
    ports:
      - "5432:5432"
    volumes:
      - ./backend/data:/var/lib/postgresql/data

Start the database:

```bash
cd backend
docker-compose up -d

The Go server will read DATABASE_URL from the environment (e.g. postgres://postgres:password@localhost:5432/pot_track_dev?sslmode=disable).

Migrations are run automatically on startup using github.com/golang-migrate/migrate and the SQL files located in backend/migrations.

Backend

1. Navigate to backend/.
2. Run go mod tidy to download dependencies.
3. Build or run:

   go run ./cmd/app

4. The API listens on port 8080 by default. Use the endpoints defined in the PRD (e.g. POST /api/auth/register).

To run unit tests:

go test ./...

Frontend

1. Navigate to frontend/.
2. Install dependencies:

   npm install

3. Run dev server:

   npm run dev -- --open

4. The app will open at http://localhost:5173 (or the port shown). Create a .env file in the frontend folder if you need to specify the backend base URL:

   VITE_API_BASE=http://localhost:8080

   The apiFetch helper will prefix requests with this value.

The UI is mobile-friendly and uses Indonesian navigation labels as per the requirements.

Development Notes

• Backend uses JWT authentication, bcrypt password hashing, and simple input validation with Gin's binding tags.
• Database schema uses Indonesian table/column names and corresponds to the PRD.
• Frontend currently has placeholders for each page; you can extend fetch logic to interact with the API.

Further Work

• Implement full business logic (stage computation, scheduling, reminders, photo upload to object storage, etc.).
• Add email notification support.
• Expand frontend with forms and real data binding.

Happy gardening! 🌱