Jonathan Schwender

There seem to be 2 main aspects to this issue. 1. Discovering which source files belong to a crate. ~~According to this [rust-internals thread](https://internals.rust-lang.org/t/getting-all-sources-for-a-particular-target/9678), you would need to actually build...

> If you try to run `sudo cargo test --no-run` it will fail on most Linux systems with something like this: rustup could not choose a version of cargo to...

> Do you know of a programmatic way to identify the real cargo for cargo stable or cargo nightly for any specific platform (e.g. stable-x86_64-unknown-linux-gnu), rather than using $HOME/.cargo/bin/cargo +stable?...

> However, one of our library dependencies is the jpeg-decoder crate, and that crate has a rust-toolchain file that forces the toolchain back to version 1.56. The rust-toolchain file is...

Hi, Thanks for the report. > Is it possible to append a -bin suffix to the executable target and/or a -lib suffix to the library target? This would be a...

> How about a better error message? Maybe emitting a warning when an existing target with the same same exists? I guess emitting a warning and skipping the target is...

Blocked until: - preferably: https://github.com/rust-lang/cargo/issues/11228 - alternatively, parsing of the manifest ourselves.

@Be-ing I opened a new issue to discuss your request. As far as I can see you wish for the following: 1. Corrosion should set an environment variable with a...

> I am only concerned with adding include paths for the generated headers I still believe this is hard to do without the user specifying which targets should get the...

> I don't understand what the issue is with the user specifying which targets use the code generator. > Users may as well add target_include_directories after corrosion_import_crates as they can...