fit2csv

A simple open-source tool for batch converting .FIT files (Garmin, Polar, Wahoo, etc.) into .CSV files.

This project contains:

• A Python script (fit2csv_batch.py) that performs the actual FIT → CSV conversion
• A macOS Automator application (optional) that provides a simple GUI and wraps the Python script
• Documentation on how to build the macOS app

The goal is to provide a minimal, reliable, and flexible tool for anyone who wants to work with their own training data.

---

✨ Features

• Convert entire folders of .fit files to .csv
• Preserve original filenames
• Output folder selection (or automatic csv_out folder)
• macOS Automator app:
  • Choose input/output folders via GUI
  • Shows a “Conversion finished” notification
  • Opens output folder automatically
  • Embeds the Python script inside the .app bundle

---

📦 Repository Structure

fit2csv/
│
├── src/
│   └── fit2csv_batch.py          # Main Python conversion script
│
├── mac-app/
│   └── README_APP.md             # Instructions to build the Automator app
│
├── .gitignore
├── LICENSE
└── README.md

---

🐍 Python Script

The main script lives in:

src/fit2csv_batch.py

It expects:

• an input folder containing .fit files
• an output folder (created automatically if missing)

Example usage:

python3 src/fit2csv_batch.py /path/to/fit-files --out /path/to/output

Requirements

Install dependencies:

pip install fitdecode

---

🍏 macOS Application (optional)

The Automator application wraps the Python script and allows you to:

• Select input folder
• Select output folder
• Run the conversion without using Terminal

Building the macOS app yourself

See:

mac-app/README_APP.md

This explains how to:

• embed fit2csv_batch.py inside the .app
• use an AppleScript wrapper to call the Python script
• build your own standalone fit2csv.app

The Automator app is not stored directly in this repository (best practice).
Instead, you can download prebuilt versions via GitHub Releases.

---

🛠 Planned improvements

• Windows and Linux packaging
• Optional progress output
• Native Python GUI version (Tkinter / PyQt)
• FIT → Parquet support
• Better error handling and logging

Pull requests are welcome!

---

📝 License

This project is released under the MIT License.
See LICENSE for full text.

---

👤 Author

Ole Fredrik Lie
GitHub: https://github.com/olefredrik

---

💬 Feedback / Contributions

Pull requests and issues are welcome! If you build something cool on top of this tool — GUI, CLI enhancements, better FIT parsing — please consider contributing it back.