# io


11/25/2023, 8:10 AM
Any updates? I'm very excited to wait for the merge

Filipp Zhinkin

11/27/2023, 10:27 AM
From the feedback I received privately, it became clear that the lack of publicly available API for interaction with kotlinx-io Buffer's internals and the lack of ByteBuffers support on JVM makes implementing AsyncSink/Source interfaces almost unfeasible (in terms of development's ease and implementation's efficiency). The mentioned problems applies to blocking/sync API (RawSink/Source) too, but there are at least some implementations of these interfaces. For async API there will be no such implementations, so it does not make much sense to merge an API that's hard to implement. So the current plan is to integrate first and then continue working on async-io.
@andreypfau I'd appreciate it if you spill some beans on what you're willing to use async-io API for and what your expectations are from it.