Dmitry Timofeev

> we'll have to omit the --no-module-directories option Do I understand correctly that this issue is only to run tests on 17, but not to migrate our release scripts to...

One more bit of work to support 17: fix Javadoc generation there 🙃 Turns out it won’t work on 17 just yet, because, apparently, javadoc tool switched a warning on...

> Is there any interest in this functionality? Unfortunately, I won't be able to add much beyond discussion in #289, as I no longer work on a project that uses...

I see, thanks! The thing is 1. Not everything could be easily inferred from the code, hence that many questions. 2. Some things may be intentionally unspecified (implementation defined) —...

> I didn't found proof in the official documentation I think this could work: https://docs.oracle.com/en/java/javase/11/docs/specs/jni/design.html#accessing-fields-and-methods > A field or method ID does not prevent the VM from unloading the class...

Makes sense, understanding which Java threads correspond to which native threads could be useful. [Java 15 spec](https://docs.oracle.com/en/java/javase/15/docs/specs/jni/invocation.html#attachcurrentthread) has more details on the structure (but it also must be in the...

I remember when we implemented the `Executor`, we considered specifying the thread names. If we add this functionality, I'd also suggest to use it in the `Executor`

I assume the macros are now a part of the public API of the crate: https://github.com/jni-rs/jni-rs/blob/d4ed1b367c0bd0abfcef21b1bea6fd78f1941051/src/lib.rs#L208-L209 Shan't we make it private?

> I see no reason to make these specific macros public. Yeah, that's rather an implementation detail — and if some JNI methods are missing, it probably makes sense to...

Updated the description