TekTrack — Internal Photomask Production Operations Tracker

A full-stack internal operations tool for tracking photomask production orders through their lifecycle — from initial receipt through mask writing, quality check, approval, and delivery.

Built as a portfolio project demonstrating Python, UNIX/Linux Shell scripting, SQL databases, XML data ingestion, FastAPI REST API design, and lightweight frontend development.

Tech Stack

Layer	Technology
Backend API	Python · FastAPI · Pydantic
Database	SQLite · SQL · XML ingestion
Frontend	Streamlit
Testing	pytest (30 tests)
CI/CD	GitHub Actions (Ubuntu Linux, Python 3.10/3.11/3.12)
Scripting	Bash Shell (setup, ingest, backup)
Version Control	Git

Production Order Lifecycle

ORDER_RECEIVED → DESIGN_REVIEW → MASK_WRITING → QUALITY_CHECK → APPROVED → SHIPPED → DELIVERED

State transitions are enforced on both the API layer and database level — invalid transitions return HTTP 400.

Quick Start

# 1. Clone and set up
git clone https://github.com/YOUR_USERNAME/tektrack.git
cd tektrack
bash scripts/setup_env.sh

# 2. Start the API
source .venv/bin/activate
uvicorn main:app --reload

# 3. Start the dashboard (new terminal)
source .venv/bin/activate
streamlit run frontend/app.py

# 4. Run tests
pytest tests/ -v

Project Structure

tektrack/
├── .github/workflows/ci.yml     # GitHub Actions CI — Python 3.10/3.11/3.12 on Ubuntu
├── backend/
│   ├── database.py              # SQLite setup, XML ingestion, SQL queries
│   ├── models.py                # Pydantic validation models
│   ├── routes/
│   │   ├── orders.py            # CRUD + state machine endpoints
│   │   └── analytics.py        # Summary, pipeline, throughput endpoints
│   └── data/
│       └── sample_orders.xml    # Seed data (8 photomask orders)
├── frontend/
│   └── app.py                   # Streamlit dashboard
├── scripts/
│   ├── setup_env.sh             # Environment setup (venv, deps, DB init)
│   ├── ingest.sh                # XML → SQLite ingestion script
│   └── backup_db.sh             # Timestamped DB backup (cron-ready)
├── tests/
│   ├── conftest.py              # pytest fixtures (tmp DB, seeded client)
│   ├── test_database.py         # Unit tests: XML ingestion, SQL, state machine
│   ├── test_orders_api.py       # Integration tests: CRUD, transitions
│   └── test_analytics_api.py    # Integration tests: analytics endpoints
├── main.py                      # FastAPI app entry point
├── requirements.txt
└── .gitignore

API Endpoints

Method	Endpoint	Description
GET	/orders/	List all orders (filterable by status, priority, customer)
GET	/orders/{id}	Get single order
POST	/orders/	Create new order
PATCH	/orders/{id}/status	Advance status (state machine enforced)
GET	/orders/{id}/log	Full status change history
DELETE	/orders/{id}	Delete (ORDER_RECEIVED only)
GET	/analytics/summary	Aggregate counts by status/priority/customer
GET	/analytics/pipeline	Orders grouped by status
GET	/analytics/throughput	Delivered orders by week

Shell Scripts

# Set up environment from scratch
bash scripts/setup_env.sh

# Ingest a new XML orders file
bash scripts/ingest.sh path/to/new_orders.xml

# Backup the database (add to cron for daily backups)
bash scripts/backup_db.sh
# Cron: 0 2 * * * bash /path/to/tektrack/scripts/backup_db.sh

Deployment

• Backend: Render.com — connect GitHub repo, set start command to uvicorn main:app --host 0.0.0.0 --port $PORT
• Frontend: Streamlit Cloud — connect GitHub repo, set main file to frontend/app.py, set API_BASE env var

Author

Jattin Shah — MSc Applied AI, Technische Universität Dresden