Shopify ERP/WMS (Lightweight, Open Source)

This project is a lightweight ERP/WMS prototype built with Django.

We previously had a Shopify DB backend focused mainly on data storage. Now we’re rebuilding it with Django, connecting backend (data + logic) and frontend (user workflows) in a single system so different user roles (Admin and Manager) can use it easily. The goal is a lightweight, open-source WMS that integrates with Shopify first and can extend to other platforms (e.g., TikTok Shop).

---

Project Structure

• accounts — Custom user model; Tenant & User management
• erp/settings — Environment-based configuration (base/dev/test/prod)
• check_env.py — Environment variable validation script
• tests/ — Pytest tests (global or per-app)
• requirements.txt — Python dependencies

---

Environment Setup

1) Create virtual environment

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

2) Configure PostgreSQL

Local / CI (default)

By default, the project connects to a dummy test database:

DB_NAME=shopify_erp_test
DB_USER=postgres
DB_PASSWORD=postgres
DB_HOST=127.0.0.1
DB_PORT=5432

This is the configuration you should commit in .env.sample (safe for GitHub). If you clone the repo without additional setup, the system will fall back to this shopify_erp_test database.

AWS (real deployment)

Create the development database:

createdb -U postgres -h localhost shopify_erp_dev

Example .env.development:

DJANGO_ENV=development
DJANGO_SECRET_KEY=change-me
DEBUG=True

DB_NAME=shopify_erp_dev
DB_USER=postgres
DB_PASSWORD=postgres
DB_HOST=127.0.0.1
DB_PORT=5432

ALLOWED_HOSTS=127.0.0.1,localhost
CSRF_TRUSTED_ORIGINS=http://127.0.0.1:8000

Use .env.test for CI/tests and .env.production for deployment. Commit only .env.sample, never real .env.*.

3) Run migrations

python manage.py makemigrations
python manage.py migrate

4) Start development server

python manage.py runserver

---

Development Guidelines

• Branch strategy

  • develop — active development
  • main — production-ready code

• Commit messages

  • feat: ... (feature)
  • fix: ... (bug fix)
  • chore: ... (config/infra/cleanup)

• Environment isolation

  • All secrets/configs must come from .env.* files (validated by check_env.py).

---

TODO Roadmap

✅ Completed

• Django project initialization
• Multi-environment configuration (dotenv + dev/test/prod)
• Minimal pytest setup
• .gitignore and .env.sample
• check_env.py script
• Created accounts app
• Defined Tenant and custom User model (roles: admin/manager)

🚧 In Progress

• Apply migrations on the target DB and verify accounts_* tables
• Create initial Tenant + User via Django admin
• Verify Tenant/User via Django shell

📌 Upcoming

• Admin invites Manager (token + expiration, status=invited → active)
• Role-based access decorator @require_admin
• Shopify connector (OAuth + Webhooks with HMAC validation)
• Ingest module (EventLog + idempotency + normalization)
• Datahub module (Products, Orders, Inventory; CSV exports for Manager)
• Notify module (instant alerts + daily digest via email)
• Deployment (CI/CD; container/Docker optional later)

---

Architecture Overview

flowchart LR
    subgraph Shopify
        A[Shopify Store]
    end

    subgraph Connector
        B[Shopify Connector - OAuth and Webhooks]
    end

    subgraph Ingest
        C[EventLog - Idempotency - Normalizer]
    end

    subgraph Datahub
        D[Products / Orders / Inventory]
    end

    subgraph Notify
        E[Email Alerts / Digests]
    end

    subgraph Users
        U1[Admin]
        U2[Manager]
    end

    A -- Webhooks --> B
    B --> C
    C --> D
    C --> E
    D -- CSV Exports --> U2
    U1 -- Invite/Manage --> D

Loading

---

Project Goal

This repository is the Django-based rewrite of our original Shopify DB backend. We’re evolving from cross-platform data-only storage to a full-stack lightweight ERP/WMS that bridges backend data with a user-facing workflow layer. By open-sourcing it, we aim to help small e-commerce teams manage inventory, sync Shopify data, and scale to multiple platforms with minimal overhead.