📦 ReqScan - Dynamic requirements.txt Builder

"I got tired of manually tracking every import across every project. So I automated the suffering away."

---

😤 The Problem

You know the drill. You've been smacking away on your keyboard for weeks. It's beautiful. It works. You're proud. Then someone says:

"Hey, can you send me the requirements?"

And suddenly you're doing archaeology on your own code. Scrolling through 30 files. Squinting at imports. Trying to remember if you actually used requests or just thought about it. Manually checking versions. Forgetting one. Breaking their environment. Apologizing.

No more. ReqScan scans your entire project, figures out what you actually imported, cross-references it against what's installed, and spits out a clean requirements.txt all in seconds.

---

✨ Features

• 🔍 Recursive project scanning - walks your entire directory tree, skipping junk like venv, node_modules, __pycache__, and other things we don't talk about
• 🧠 Smart import classification - separates stdlib modules, local modules, and actual third-party dependencies (no more os==3.11 nonsense)
• ⚡ Parallel parsing - uses ProcessPoolExecutor to parse files fast
• 🔄 Merge mode - keeps your manually-added entries and updates detected versions instead of nuking everything
• 📍 Flexible output - save the file inside your project or wherever you're running the script
• 🪟 Windows compatible - ANSI color support for those of us living in cmd.exe purgatory

---

⚡ Quick Install (pip)

Want the no-fuss version? Install it globally and use it anywhere:

pip install reqscan

Then just cd into any project and run:

reqscan

That's it. It scans the current directory and generates your requirements.txt on the spot — no prompts, no setup.

---

🚀 Full Version Usage

For the interactive experience with more options, clone this repo and run it directly:

python main.py

Then follow the prompts:

1. Choose your project path - current directory or a custom one
2. Choose where to save requirements.txt - inside the project or here
3. Watch it go 🎉

---

📋 Example Output

requirements.txt
────────────────
fastapi==0.111.0
httpx==0.27.0
pydantic==2.7.1
rich==13.7.1
uvicorn==0.29.0

Clean. Sorted. Versioned. Done.

---

🛠 Requirements

• Python 3.10+ (uses sys.stdlib_module_names, introduced in 3.10)
• No third-party dependencies, ReqScan eats its own food and requires nothing outside the stdlib

---

⚠️ Caveats & Known Quirks

• Dynamic imports (importlib.import_module("something"), __import__) won't be detected, ReqScan uses static AST analysis
• Version detection only works for packages installed in your current Python environment, run it from inside your virtualenv for best results
• If a package is imported under a different name than its PyPI name (e.g., import cv2 → opencv-python), ReqScan uses top_level.txt metadata to resolve it, but some packages don't ship that file

---

🤝 Contributing

Found a bug? Have a wild idea? PRs welcome. This started as a personal itch-scratcher, so there's plenty of room to grow.

---

📜 License

Do whatever you want with it. Just don't manually maintain a requirements.txt ever again. That's the only rule.