Perl has its own encoding model, strings vould be unicode or legacy
encoding, unicode is Unicode is indicated by the presence of a flag on a
string. Out its decided on a string by string basis.
If it is a legacy encoding, then it could be any legacy encoding.
If your data is truly multilingual, multiscript and in a variety of
encodings, it becomes a challenge to manage it in Perl.
In our own projects we found perl module to be inadequate and needed our
own internal modules to handle encoding issues, radio when you factor in
the fact that some cpan modules have the nasty habit of stripping the
Unicode flag from strings.
Although that said, Perl still has better Unicode support than most