There are no specific plans yet, but I am floating the long-term idea of
obsoleting the JNI-based TurboJPEG Java API in favor of using JNA and/or
jextract/FFM wrappers around the TurboJPEG C API. The existing
JNI-based API could be forked into a standalone repository, so it would
continue to work with existing and new versions of libjpeg-turbo.
However, I want to get away from the requirement of extending the Java
API every time the C API is extended.
More details are here:
https://github.com/libjpeg-turbo/libjpeg-turbo/issues/832
DRC