Lightweight Payroll and Attendance System

This repository demonstrates a minimal yet modular payroll system implemented in pure Python. It is designed for factories with fewer than 50 employees and is intentionally simple so that even newcomers can experiment with the code.

Quick Start

1. Install requirements. The only external packages are SQLAlchemy, pandas, cryptography, and scikit‑learn. Install them with:

   pip install sqlalchemy pandas cryptography scikit-learn

2. Launch the GUI to add employees or attendance records:

   python -m payroll_system.main --gui

   Default credentials are:
   • master / master
   • viewer / viewer
   • att / att
3. Command line tools. Run python -m payroll_system.main --help to see options such as creating a ZIP backup or exporting attendance to Excel/CSV/JSON files.

Project Layout

• payroll_system/db.py – database models and helper utilities.
• payroll_system/gui.py – tiny Tkinter interface with role based login.
• payroll_system/export.py – export helpers for Excel/CSV/JSON.
• payroll_system/festival.py – Bengali festival calendar helpers.
• payroll_system/ml_utils.py – lightweight machine learning helpers.
• tests/ – small unit tests to show expected behaviour.
• Payroll_Attendance_System.ipynb – Jupyter notebook walkthrough.

Troubleshooting

• Missing packages – install the requirements shown above. If you see an error like ModuleNotFoundError: No module named 'sklearn', make sure scikit-learn was installed successfully.
• Database locked – close other programs using employee_db_2025.sqlite before launching the GUI or running exports.
• Corrupt database – you can restore from a ZIP backup created with python -m payroll_system.main --backup backup.zip.

The code is intentionally concise but provides a starting point for a more complete system with role based access, encrypted data, and export capabilities.

Using the GUI

After logging in you will see three tabs:

1. Employee – enter name, Aadhar, PAN, contact number and hire date.
2. Attendance – record daily salary and role with optional leave type.
3. Festivals – view important Bengali holidays loaded from test_data/festivals.csv.

These examples are deliberately minimal so that even beginners can follow the flow. Feel free to inspect the Jupyter notebook for a step by step explanation.

License

This project is licensed under the MIT License. See the LICENSE file for details. The code is intentionally concise but provides a starting point for a more complete system with role based access, encrypted data, and export capabilities. Feel free to experiment! See the notebook for examples.