I am experiencing a strange panic that happens inside procedural macros when they are compiled as dynamic libraries, dynamically linked and then called explicitly using proc_macro::bridge::client. It's quite hard to reproduce in small am...
Using #![feature(proc_macro)] in combination with attributes enabled by #[proc_macro_derive(MyTrait, attributes(mycrate))] (e.g. #[mycrate(some_option = "something")]) fails to compile on the current nightly with error: macro undefined: ...