I've been trying to figure out a way to allow dead_code for non-release builds. So far, I've got this:
// main.rs
#![allow(dead_code)] // disable it at the crate-level
This is great for non-release builds, but how do I disable it for --release builds? Is there a way to do this?
Compilation profiles don't really give you that level of control, they're just a set of optimization settings. You can't detect optimization settings very easily in code.
What you might look at is build config. You could maybe arrange a build alias or to pass in a rustc-cfg flag that you can test against with #cgf(...). I have never used it, so I can't say if there's a combination that works.
A better option is probably to use debug_assertions as a standin-for non-release builds, so you can do this:
#![cfg_attr(debug_assertions, allow(dead_code))]
#![cfg_attr(not(debug_assertions), deny(dead_code))]
You could also use debug here, but neither is perfect.
Ah-hah, cfg_attr() was exactly what I was looking for. I was using debug_assertions in some of my attempts since that seems to be a common indicator between dev/prod. I couldn't figure out how to use that to conditionally enable things, and it looks like cfg_attr() is what does just that. I was also hoping that there'd be a way to also undefined certain flags, and it appears that there is a way with deny().
Neat.
Deny doesn't really undefine cfg flags, what it does is convert a lint that would warn into a lint that errors. You can do allow, warn or deny on lints to choose the behavior.
This topic was automatically closed 90 days after the last reply. We invite you to open a new topic if you have further questions or comments.