Git和混淆代码

Question

I've been working on an idea, and needed to run it past someone to see if I'd gone off the deep end.

I'm working on a decently complex PHP project, I am the only developer. We're looking to hire some one else to help with the programming effort, but don't want them to see everything on day one. There is a pretty clear line between what would be "core architecture" and "non-core". I'd like the new person to have be able to read and modify the "non-core" section. The core section is required for the non-core to work.

My plan id this:

1. Make the Core and Non-Core sections separate git repositories
2. Make the Core section a submodule of the Non-Core
3. Add a post-commit hook to the Core repository, to run code obfuscation

Then when the submodule is updated, the files will execute, but will be very hard to read.

#1 and #2 are pretty easy. But #3... I don't know if it can even be done. I was looking into the git internals, to see how the files are actually stored, but it seems like they are just stored in the file system. So there wouldn't be any way to edit them and add obfuscation without totally trashing the "good" copy. That being said, I could always daisy chain another repo on there.

Good Repo -> Core Repo (with Obfuscation) -> Non-Core

I'm just wondering if there's a cleaner way to do this.

---

Just another thought, but I could not use a repository for the obfuscated code. Instead have the git commit hook call a script to copy and obfuscate the Core code and place it in the non-core repo. Add a .getignore so that folder isn't part of the non-core codebase.

Answer 1

Expose your core functionality as web services and access them over HTTP. Keep the non-core bits in the repository.