Pascal Hertleif

Very interesting! Technically, this will require attibutes (or, this specific attribute) to produce diagnostic messages that contain suggestions -- which I think is already somewhat possible with proc macros?

So, I wrote https://github.com/killercup/rustfix/commit/9565a44dee0f96f19d899a76a566c1f84daa5d8c and ran it with `env RUST_LOG=cargo_fix=trace cargo test --release -- fix_perf --nocapture`. Output ``` Compiling cargo-fix v0.2.0 (file:///Users/pascal/Projekte/rustfix/cargo-fix) Finished release [optimized] target(s) in 7.00 secs Running...

Yeah, actual rustfixing should be fast, but I was surprised that buffering the compiler's output could actually have a perf impact (we could in theory switch to a buffered reader)....

Agreed. It's a good feature to have, especially if we can make it work smoothly (e.g. by checking that the path contains the expected crate and other things you'd not...

Interesting use case! I've seen a few multi-crate projects suggest this workflow and I think it's a pretty great feature of cargo (that is probably under-utilized). Please note that this...

Cool, I'm looking forward to hearing from you, @hobofan :) > Am 12.02.2016 um 17:22 schrieb Maximilian Goisser [email protected]: > > Please note that this config file need not be...

Absolutely. I'm not sure if this is a feature for rustfix or for the lint that generates these suggestions. rustfix could remove the newline if it clears the line when...

Yeah, good point. There's no rush, I'm okay with landing this in quicli itself (once I have a concrete use case) and switching to exporting the upstream version later on....

Hey, sorry for the late response! I'm stilling looking into refactoring the crate to get rid of the `main` macro. Keeping it easy to set up logging will be one...

Not sure if this - is any good - should be part of this crate (or be defined upstream somewhere) cc @steveklabnik