Robert Bragg
Robert Bragg
Not only are these really unsafe APIs to access directly, they are also basically just tiny wrappers around the raw JNI function call and so there's really very little value...
The `Executor` API is an elaborate wrapper over: ```rust let mut jni_env = self.vm.attach_current_thread_as_daemon()?; jni_env.with_local_frame(capacity, |jni_env| f(jni_env)) ``` Considering https://github.com/jni-rs/jni-rs/issues/469 it seems like we should at least change this to...
The low-level JNI API supports the idea that you can attach a thread to a JVM in such a way that it won't automatically get detached from the JVM when...
This issue is really just a reminder to take some care to consider how safe it is to have multiple versions of the `jni` crate linked in a single application....
> I think that it makes sense to put `#![deny(missing_debug_implementations)]` at the top level and add a `Debug` implementation for every public type. Perhaps it isn't that useful for some...
If code has decided to `.discard()` the current pointer to the elements of an array and that code doesn't need to keep any changes (in case the elements were copied)...
Considering the build failure that was recently fixed with https://github.com/jni-rs/jni-rs/pull/419 if would be good to add at least one Android target that is known to have a differing ABI for...
# Motivation To make it convenient / practical to write independent crates that may rely on JNI usage it's helpful to have a standard crate with a stable semver API...
With CI currently building with latest stable Rust then whenever there are new Clippy lints the CI is liable to start failing which interferes with landing other pull requests. For...
If I've cross-compiled a binary via clang + ld64 (without using XCode) then the code path for signing from scratch doesn't seem to be working currently. The first issue I...