I've got an executable crate that uses OpenGL to render images to the screen as textures. I've got a few boilerplate modules that abstract a lot of the OpenGL setup, and also wraps the
unsafe OpenGL calls in safe code making OpenGL an implementation detail to the rest of the app. That all works fine, but I'm getting 30 warnings about unused methods and enum variants because I haven't needed to directly use some of that code yet. I will eventually need it, but not now.
The warnings make total sense, but I hate seeing them show up in my console every time I go to build something. I could allow the dead code, but that feels a bit messy as I'd have to declare the
allow() macro everywhere. Plus, once I finally use that code, I'll forget about it.
What kind of mindset should I have around this? Would it be best to allow the dead code? I'll eventually break all of this code out into its own crate since this will be useful across multiple crates. I only hesitate on abstracting my little graphics toolkit because I'm not completely sure how I'll want to setup my crates, workspaces, and repos yet.