jnext socket (like the other plugins) was written more as a proof of
concept than a production quality plugin, so unicode support was not
one of the objectives - however, it should be relatively easy to port
the code and make sure no ASCII assumptions are made in all the string
handling code (if I remember correctly most of the code uses
std::string so there shouldn't be too much to do).